TOP ACTIVATIONS
MAX = 0.104

With a mis chie vous gl int in her
always acted silly and mis chie vous . One
old lady with a mis chie vous tw inkle in her
ier ? You are y umm ier . You are my
at him with a mis chie vous grin . She said
and replied with a mis chie vous gl int in his
" Y um , y um , these nuts are delicious
. " Please mum , I want a spoon
there lived two sisters - Mon a and T oni .
, y um , y um ," the k ang aroo
- d um - d um !" " OK
, y um , y um !" Tom says . "
, y um , y um ," Anna and Ben say
" We are sorry , Mum . We are sorry ,
, y um , y um ," they say .
that day . <|endoftext|> M um woke up early . She
He kept saying " Please mum , can I go to
" Y um , y um , y um ," she
You just have to str um the strings and sing along
to paint on paper or canvas . And you have to

MIN ACTIVATIONS
MIN = -0.122

All the things that the fle a recorded made a beautiful
she noticed that the books he lf was gone . It
It became the happiest books he lf in the world !
He noticed that the books he lf was very smooth and
around to all his feat hered friends . Everyone felt so
was so strong that the fle a could hardly keep driving
something more about his feat hered friends . John
so beautiful with its feat hered wings gl iding against the
my never saw the books he lf again . But they
birds stop singing ! The rac ers dashed around the town
and upset that her books he lf was a mess .
books on the new books he lf and it was full
to see the rare books he lf . They looked at
He also picked some lettuce from the garden .
Jake was jealous because his k ite wasn 't flying as
scary that the boss y k itty ran away and hid
a fle a . The fle a bit her nose and
was full . The books he lf looked great in their
accidentally knocked over the books he lf .
peek ed around the books he lf again , she noticed

INTERVAL 0.048 - 0.104
CONTAINS 1.156%

. Do you want to see it ?" L
at them and smiles . She sees their tent and says
again . M um said , " The earthquake
it sounded like singing . They slowly reached
, can you see the sun up there ? It looks

INTERVAL -0.009 - 0.048
CONTAINS 64.892%

" Yes , please !" Lily and Ben said
The girl was so happy to be back in
does that sound ?" S ara and Ben smiled
water from the tap . They felt better , but their
needed a break . He found a big rock to sit

INTERVAL -0.066 - -0.009
CONTAINS 33.843%

had a basket with a picnic inside . He saw his
p uddle ! Sam was scared and ran back to his
. Jimmy ran to the truck to help
moved his toys around the room but he kept them in
Her mom smiled and said , " That 's a great

INTERVAL -0.122 - -0.066
CONTAINS 0.109%

The elderly man and his dog would swing back and forth
Jenny was relieved that the monster she pretended to be scared
smelled the y ummy food from the grill and their tum
was too full . The bath robe was too big .
gifts . He told his re ind eer to stay put
Token
Token ID198
Feature activation-1.42e-02

Pos logit contributions
clock+0.076
ups+0.072
rays+0.069
up+0.069
 grows+0.066
Neg logit contributions
rily-0.136
stack-0.133
ishly-0.132
 afloat-0.122
bled-0.117
Token
Token ID198
Feature activation-1.27e-02

Pos logit contributions
rily+0.028
stack+0.028
ishly+0.027
 afloat+0.024
 themselves+0.024
Neg logit contributions
clock-0.021
ups-0.020
up-0.019
rays-0.019
 cares-0.019
TokenWith
Token ID3152
Feature activation-8.50e-03

Pos logit contributions
stack+0.019
rily+0.019
ishly+0.018
 themselves+0.018
oned+0.017
Neg logit contributions
 cares-0.023
clock-0.022
 grows-0.022
rays-0.021
 dreams-0.021
Token a
Token ID257
Feature activation+1.58e-02

Pos logit contributions
rily+0.017
stack+0.017
ishly+0.016
 afloat+0.015
bled+0.014
Neg logit contributions
clock-0.012
 cares-0.012
rays-0.011
 grows-0.011
ups-0.011
Token mis
Token ID2984
Feature activation+2.50e-02

Pos logit contributions
clock+0.027
ups+0.025
rays+0.024
up+0.024
 cares+0.024
Neg logit contributions
stack-0.027
rily-0.027
ishly-0.027
 afloat-0.025
 towel-0.024
Tokenchie
Token ID3043
Feature activation+1.04e-01

Pos logit contributions
clock+0.048
 listens+0.047
 dreams+0.047
 grows+0.046
rays+0.046
Neg logit contributions
rily-0.037
stack-0.034
ishly-0.033
 afloat-0.031
 themselves-0.029
Tokenvous
Token ID31222
Feature activation+2.63e-02

Pos logit contributions
clock+0.118
 grows+0.115
 mysteries+0.114
rays+0.113
 dreams+0.111
Neg logit contributions
stack-0.209
rily-0.206
ishly-0.199
 themselves-0.184
ged-0.183
Token gl
Token ID1278
Feature activation+4.42e-02

Pos logit contributions
clock+0.045
ups+0.043
up+0.042
rays+0.042
 grows+0.041
Neg logit contributions
rily-0.046
stack-0.045
ishly-0.044
 afloat-0.039
 themselves-0.038
Tokenint
Token ID600
Feature activation+6.24e-02

Pos logit contributions
clock+0.088
ups+0.081
 grows+0.080
rays+0.080
 cares+0.079
Neg logit contributions
rily-0.069
stack-0.065
ishly-0.064
bled-0.057
 afloat-0.055
Token in
Token ID287
Feature activation-5.60e-03

Pos logit contributions
clock+0.098
 cares+0.085
ups+0.084
 mysteries+0.084
 grows+0.080
Neg logit contributions
rily-0.121
stack-0.116
ishly-0.115
bled-0.104
ged-0.103
Token her
Token ID607
Feature activation+1.26e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.009
 afloat+0.008
 themselves+0.008
Neg logit contributions
clock-0.010
ups-0.009
up-0.009
 cares-0.009
rays-0.009
Token always
Token ID1464
Feature activation+3.12e-02

Pos logit contributions
clock+0.018
rays+0.017
ups+0.017
up+0.016
 cares+0.016
Neg logit contributions
rily-0.020
stack-0.020
ishly-0.019
bled-0.017
 themselves-0.017
Token acted
Token ID13134
Feature activation+1.50e-02

Pos logit contributions
clock+0.047
ups+0.043
up+0.043
 cares+0.043
rays+0.042
Neg logit contributions
rily-0.062
stack-0.060
ishly-0.058
 afloat-0.053
 themselves-0.053
Token silly
Token ID14397
Feature activation+4.90e-06

Pos logit contributions
clock+0.025
rays+0.023
ups+0.022
 cares+0.022
 grows+0.022
Neg logit contributions
rily-0.028
stack-0.026
ishly-0.026
 afloat-0.024
 themselves-0.024
Token and
Token ID290
Feature activation+1.76e-02

Feature not active on prev token;
no predictions were affected.
Token mis
Token ID2984
Feature activation+1.10e-02

Pos logit contributions
clock+0.026
ups+0.024
up+0.024
rays+0.023
 cares+0.023
Neg logit contributions
rily-0.035
stack-0.034
ishly-0.033
 afloat-0.030
 themselves-0.030
Tokenchie
Token ID3043
Feature activation+1.04e-01

Pos logit contributions
clock+0.018
 listens+0.018
 dreams+0.018
 grows+0.018
 cares+0.018
Neg logit contributions
rily-0.019
ishly-0.017
stack-0.017
 afloat-0.017
 themselves-0.016
Tokenvous
Token ID31222
Feature activation+2.07e-02

Pos logit contributions
clock+0.121
rays+0.118
 mysteries+0.117
 grows+0.115
 dreams+0.114
Neg logit contributions
rily-0.195
stack-0.192
ishly-0.191
 themselves-0.185
tered-0.181
Token.
Token ID13
Feature activation-1.49e-02

Pos logit contributions
clock+0.032
rays+0.030
up+0.030
 grows+0.029
ups+0.029
Neg logit contributions
rily-0.038
ishly-0.037
stack-0.036
 themselves-0.033
 afloat-0.032
Token
Token ID198
Feature activation-1.65e-02

Pos logit contributions
rily+0.029
stack+0.029
ishly+0.027
 afloat+0.025
 themselves+0.025
Neg logit contributions
clock-0.023
ups-0.021
up-0.021
 cares-0.021
rays-0.021
Token
Token ID198
Feature activation-1.39e-02

Pos logit contributions
rily+0.037
ishly+0.036
stack+0.036
 afloat+0.032
bled+0.032
Neg logit contributions
clock-0.021
ups-0.019
up-0.019
rays-0.018
 grows-0.018
TokenOne
Token ID3198
Feature activation+3.65e-02

Pos logit contributions
stack+0.033
rily+0.032
ishly+0.031
 themselves+0.030
 afloat+0.029
Neg logit contributions
clock-0.015
 cares-0.014
ups-0.013
rays-0.013
up-0.013
Token old
Token ID1468
Feature activation-2.34e-02

Pos logit contributions
clock+0.006
ups+0.006
rays+0.006
 cares+0.006
up+0.006
Neg logit contributions
stack-0.008
rily-0.008
ishly-0.008
 afloat-0.007
 themselves-0.007
Token lady
Token ID10846
Feature activation+9.15e-03

Pos logit contributions
stack+0.037
rily+0.034
ishly+0.033
 themselves+0.030
 towel+0.028
Neg logit contributions
clock-0.042
rays-0.040
ups-0.040
 grows-0.039
 cares-0.038
Token with
Token ID351
Feature activation-4.37e-02

Pos logit contributions
clock+0.014
ups+0.013
up+0.013
 cares+0.012
rays+0.012
Neg logit contributions
rily-0.018
stack-0.018
ishly-0.017
 themselves-0.015
 afloat-0.015
Token a
Token ID257
Feature activation+7.67e-03

Pos logit contributions
rily+0.093
stack+0.091
ishly+0.087
 afloat+0.081
bled+0.080
Neg logit contributions
clock-0.059
 cares-0.055
ups-0.054
up-0.054
rays-0.053
Token mis
Token ID2984
Feature activation+1.46e-02

Pos logit contributions
clock+0.012
ups+0.011
rays+0.011
up+0.011
 cares+0.010
Neg logit contributions
rily-0.015
stack-0.015
ishly-0.014
 afloat-0.013
 themselves-0.013
Tokenchie
Token ID3043
Feature activation+1.02e-01

Pos logit contributions
 listens+0.029
clock+0.029
 grows+0.029
rays+0.029
 dreams+0.029
Neg logit contributions
rily-0.020
ishly-0.018
stack-0.017
 afloat-0.017
 themselves-0.015
Tokenvous
Token ID31222
Feature activation+2.43e-02

Pos logit contributions
clock+0.119
rays+0.117
 grows+0.115
 mysteries+0.112
ups+0.110
Neg logit contributions
rily-0.200
stack-0.196
ishly-0.187
 themselves-0.176
 afloat-0.176
Token tw
Token ID665
Feature activation+2.98e-02

Pos logit contributions
clock+0.042
ups+0.040
rays+0.039
up+0.039
 grows+0.038
Neg logit contributions
rily-0.042
stack-0.041
ishly-0.040
 afloat-0.036
 themselves-0.035
Tokeninkle
Token ID19894
Feature activation+1.49e-02

Pos logit contributions
clock+0.053
rays+0.048
 listens+0.045
 cares+0.045
ups+0.045
Neg logit contributions
rily-0.058
stack-0.048
bled-0.047
ishly-0.047
 afloat-0.047
Token in
Token ID287
Feature activation-4.60e-03

Pos logit contributions
clock+0.024
ups+0.022
up+0.022
rays+0.022
 grows+0.022
Neg logit contributions
stack-0.027
rily-0.027
ishly-0.026
 themselves-0.024
 afloat-0.024
Token her
Token ID607
Feature activation+1.94e-03

Pos logit contributions
rily+0.009
stack+0.009
ishly+0.009
bled+0.008
 afloat+0.008
Neg logit contributions
clock-0.007
 cares-0.006
up-0.006
 grows-0.006
ups-0.006
Tokenier
Token ID959
Feature activation+2.81e-02

Pos logit contributions
clock+0.156
ups+0.146
up+0.144
rays+0.144
 cares+0.143
Neg logit contributions
rily-0.140
stack-0.134
ishly-0.127
bled-0.118
 themselves-0.116
Token?
Token ID30
Feature activation+3.51e-02

Pos logit contributions
clock+0.039
ups+0.036
up+0.035
 cares+0.035
rays+0.035
Neg logit contributions
rily-0.059
stack-0.058
ishly-0.056
 afloat-0.051
 themselves-0.051
Token You
Token ID921
Feature activation+1.72e-03

Pos logit contributions
ups+0.057
up+0.057
clock+0.056
 mysteries+0.053
 cares+0.050
Neg logit contributions
stack-0.061
rily-0.060
ishly-0.057
 themselves-0.054
 Nem-0.048
Token are
Token ID389
Feature activation-1.11e-02

Pos logit contributions
clock+0.003
ups+0.003
up+0.003
rays+0.003
 cares+0.002
Neg logit contributions
stack-0.003
rily-0.003
ishly-0.003
 themselves-0.003
 afloat-0.003
Token y
Token ID331
Feature activation+2.49e-02

Pos logit contributions
rily+0.019
stack+0.019
ishly+0.017
 themselves+0.017
bled+0.016
Neg logit contributions
clock-0.019
up-0.018
ups-0.018
rays-0.018
 cares-0.017
Tokenumm
Token ID13929
Feature activation+1.00e-01

Pos logit contributions
rays+0.046
clock+0.045
 mysteries+0.045
 listens+0.045
 cares+0.044
Neg logit contributions
rily-0.038
ishly-0.034
stack-0.032
 themselves-0.031
bled-0.030
Tokenier
Token ID959
Feature activation+3.91e-02

Pos logit contributions
clock+0.187
ups+0.177
rays+0.175
up+0.175
 grows+0.174
Neg logit contributions
rily-0.158
stack-0.147
ishly-0.143
bled-0.132
 themselves-0.127
Token.
Token ID13
Feature activation+3.91e-02

Pos logit contributions
clock+0.050
ups+0.046
up+0.045
 cares+0.045
rays+0.044
Neg logit contributions
rily-0.085
stack-0.084
ishly-0.082
 afloat-0.075
bled-0.074
Token You
Token ID921
Feature activation+3.39e-03

Pos logit contributions
ups+0.062
up+0.062
clock+0.059
 mysteries+0.056
rays+0.055
Neg logit contributions
stack-0.070
rily-0.067
ishly-0.065
 themselves-0.063
ter-0.056
Token are
Token ID389
Feature activation-9.58e-03

Pos logit contributions
clock+0.006
ups+0.006
up+0.005
rays+0.005
 mysteries+0.005
Neg logit contributions
stack-0.006
rily-0.006
ishly-0.005
 themselves-0.005
 afloat-0.005
Token my
Token ID616
Feature activation-4.24e-03

Pos logit contributions
rily+0.016
stack+0.016
ishly+0.015
 themselves+0.015
bled+0.014
Neg logit contributions
clock-0.016
up-0.016
rays-0.016
ups-0.016
 cares-0.015
Token at
Token ID379
Feature activation+1.04e-02

Pos logit contributions
clock+0.029
ups+0.027
up+0.027
 cares+0.027
rays+0.027
Neg logit contributions
rily-0.035
stack-0.035
ishly-0.033
 afloat-0.030
 themselves-0.030
Token him
Token ID683
Feature activation-1.44e-02

Pos logit contributions
rays+0.014
clock+0.014
ups+0.014
up+0.013
 mysteries+0.013
Neg logit contributions
rily-0.021
stack-0.020
ishly-0.019
 afloat-0.019
bled-0.018
Token with
Token ID351
Feature activation-4.18e-02

Pos logit contributions
stack+0.032
rily+0.031
ishly+0.029
bled+0.028
 afloat+0.028
Neg logit contributions
clock-0.017
ups-0.015
 grows-0.015
 looks-0.015
 listens-0.014
Token a
Token ID257
Feature activation-3.36e-03

Pos logit contributions
rily+0.087
stack+0.086
ishly+0.081
 afloat+0.075
bled+0.075
Neg logit contributions
clock-0.058
 cares-0.054
 grows-0.053
up-0.053
ups-0.053
Token mis
Token ID2984
Feature activation+1.94e-02

Pos logit contributions
rily+0.007
stack+0.006
ishly+0.006
 afloat+0.006
 themselves+0.006
Neg logit contributions
clock-0.005
ups-0.005
up-0.005
rays-0.005
 cares-0.005
Tokenchie
Token ID3043
Feature activation+9.98e-02

Pos logit contributions
clock+0.035
rays+0.033
 listens+0.033
 cares+0.033
 grows+0.033
Neg logit contributions
rily-0.032
stack-0.030
ishly-0.029
 afloat-0.027
 themselves-0.026
Tokenvous
Token ID31222
Feature activation+2.01e-02

Pos logit contributions
clock+0.112
rays+0.109
 mysteries+0.103
ups+0.101
 grows+0.100
Neg logit contributions
rily-0.214
stack-0.206
ishly-0.197
 themselves-0.184
 afloat-0.180
Token grin
Token ID16770
Feature activation+2.88e-02

Pos logit contributions
clock+0.036
ups+0.034
up+0.033
rays+0.033
 cares+0.033
Neg logit contributions
rily-0.034
stack-0.033
ishly-0.032
 afloat-0.029
 themselves-0.028
Token.
Token ID13
Feature activation-6.89e-03

Pos logit contributions
 grows+0.025
clock+0.024
rays+0.024
up+0.023
 happens+0.022
Neg logit contributions
stack-0.069
rily-0.066
ishly-0.066
 afloat-0.063
 themselves-0.063
Token She
Token ID1375
Feature activation+2.31e-02

Pos logit contributions
stack+0.012
ishly+0.011
rily+0.011
 themselves+0.010
 Ph+0.009
Neg logit contributions
up-0.012
 cares-0.012
ups-0.011
clock-0.011
 grows-0.011
Token said
Token ID531
Feature activation-2.80e-04

Pos logit contributions
clock+0.043
ups+0.040
rays+0.039
up+0.039
 cares+0.039
Neg logit contributions
rily-0.037
stack-0.036
ishly-0.035
bled-0.031
 themselves-0.030
Token and
Token ID290
Feature activation+4.46e-03

Pos logit contributions
clock+0.041
ups+0.037
up+0.037
rays+0.036
 grows+0.035
Neg logit contributions
stack-0.051
rily-0.048
ishly-0.046
bled-0.042
 afloat-0.042
Token replied
Token ID8712
Feature activation-3.80e-03

Pos logit contributions
clock+0.007
up+0.007
ups+0.007
 cares+0.007
rays+0.006
Neg logit contributions
rily-0.008
stack-0.008
ishly-0.008
 afloat-0.007
 themselves-0.007
Token with
Token ID351
Feature activation-4.52e-02

Pos logit contributions
rily+0.008
stack+0.008
ishly+0.007
 afloat+0.007
 themselves+0.007
Neg logit contributions
clock-0.005
ups-0.005
up-0.005
rays-0.005
 cares-0.005
Token a
Token ID257
Feature activation+2.29e-03

Pos logit contributions
rily+0.098
stack+0.098
ishly+0.092
 afloat+0.085
bled+0.085
Neg logit contributions
clock-0.058
 cares-0.055
up-0.053
 grows-0.053
rays-0.051
Token mis
Token ID2984
Feature activation+1.21e-02

Pos logit contributions
clock+0.004
ups+0.003
rays+0.003
up+0.003
 cares+0.003
Neg logit contributions
rily-0.004
stack-0.004
ishly-0.004
 afloat-0.004
 themselves-0.004
Tokenchie
Token ID3043
Feature activation+9.88e-02

Pos logit contributions
clock+0.024
 listens+0.024
 dreams+0.024
rays+0.024
 grows+0.023
Neg logit contributions
rily-0.017
stack-0.016
ishly-0.015
 afloat-0.014
 themselves-0.013
Tokenvous
Token ID31222
Feature activation+1.63e-02

Pos logit contributions
clock+0.109
rays+0.104
 grows+0.101
ups+0.098
 mysteries+0.098
Neg logit contributions
stack-0.211
rily-0.211
ishly-0.202
 themselves-0.189
 afloat-0.185
Token gl
Token ID1278
Feature activation+2.98e-02

Pos logit contributions
clock+0.027
ups+0.025
up+0.025
rays+0.025
 cares+0.025
Neg logit contributions
rily-0.029
stack-0.029
ishly-0.028
 afloat-0.025
 themselves-0.025
Tokenint
Token ID600
Feature activation+5.18e-02

Pos logit contributions
clock+0.056
ups+0.051
 grows+0.051
 cares+0.051
rays+0.051
Neg logit contributions
rily-0.050
stack-0.047
ishly-0.046
bled-0.041
 afloat-0.041
Token in
Token ID287
Feature activation-1.91e-02

Pos logit contributions
clock+0.068
ups+0.062
 cares+0.061
up+0.061
rays+0.060
Neg logit contributions
rily-0.112
stack-0.109
ishly-0.106
 afloat-0.098
 themselves-0.097
Token his
Token ID465
Feature activation-1.95e-02

Pos logit contributions
rily+0.040
stack+0.039
ishly+0.038
bled+0.036
 afloat+0.034
Neg logit contributions
clock-0.026
 grows-0.024
 cares-0.023
up-0.023
 mysteries-0.023
Token"
Token ID1
Feature activation+2.07e-02

Pos logit contributions
clock+0.009
ups+0.008
up+0.008
 cares+0.008
rays+0.008
Neg logit contributions
rily-0.016
stack-0.016
ishly-0.016
 afloat-0.014
bled-0.014
TokenY
Token ID56
Feature activation+4.56e-02

Pos logit contributions
clock+0.027
up+0.024
 cares+0.024
ups+0.024
rays+0.023
Neg logit contributions
rily-0.047
ishly-0.045
stack-0.045
 afloat-0.042
bled-0.040
Tokenum
Token ID388
Feature activation+8.97e-02

Pos logit contributions
clock+0.061
 mysteries+0.060
 cares+0.057
rays+0.056
ups+0.056
Neg logit contributions
rily-0.094
ishly-0.091
stack-0.089
bled-0.083
 themselves-0.083
Token,
Token ID11
Feature activation+1.61e-02

Pos logit contributions
clock+0.152
rays+0.138
ups+0.137
 mysteries+0.136
 grows+0.134
Neg logit contributions
rily-0.156
stack-0.150
ishly-0.150
 themselves-0.133
bled-0.129
Token y
Token ID331
Feature activation+1.17e-02

Pos logit contributions
 cares+0.019
clock+0.018
 grows+0.018
 looks+0.018
up+0.018
Neg logit contributions
rily-0.036
stack-0.036
ishly-0.035
 afloat-0.034
bled-0.033
Tokenum
Token ID388
Feature activation+9.75e-02

Pos logit contributions
clock+0.016
rays+0.015
 cares+0.014
ups+0.014
up+0.014
Neg logit contributions
rily-0.025
stack-0.024
ishly-0.023
 themselves-0.022
 afloat-0.022
Token,
Token ID11
Feature activation+2.05e-02

Pos logit contributions
clock+0.156
rays+0.143
ups+0.141
 cares+0.141
 mysteries+0.140
Neg logit contributions
rily-0.179
stack-0.171
ishly-0.166
 themselves-0.156
 afloat-0.149
Token these
Token ID777
Feature activation-6.69e-03

Pos logit contributions
clock+0.027
 grows+0.026
 cares+0.026
up+0.025
 looks+0.025
Neg logit contributions
rily-0.042
stack-0.042
ishly-0.040
 afloat-0.040
bled-0.038
Token nuts
Token ID14380
Feature activation-2.93e-02

Pos logit contributions
rily+0.011
stack+0.011
ishly+0.010
 themselves+0.010
 afloat+0.009
Neg logit contributions
clock-0.012
ups-0.011
rays-0.011
up-0.011
 cares-0.011
Token are
Token ID389
Feature activation-2.44e-02

Pos logit contributions
rily+0.062
stack+0.061
ishly+0.059
 afloat+0.054
bled+0.054
Neg logit contributions
clock-0.039
ups-0.036
up-0.036
 cares-0.036
 grows-0.036
Token delicious
Token ID12625
Feature activation-2.21e-03

Pos logit contributions
stack+0.050
rily+0.049
ishly+0.049
 afloat+0.043
bled+0.043
Neg logit contributions
clock-0.034
 grows-0.032
ups-0.032
up-0.031
 cares-0.031
Token.
Token ID13
Feature activation-7.27e-03

Pos logit contributions
clock+0.041
up+0.039
ups+0.039
rays+0.035
 cares+0.035
Neg logit contributions
rily-0.086
stack-0.085
ishly-0.081
bled-0.079
 afloat-0.076
Token
Token ID198
Feature activation-1.45e-02

Pos logit contributions
stack+0.012
ishly+0.011
rily+0.011
 themselves+0.010
 Betsy+0.009
Neg logit contributions
ups-0.013
clock-0.013
up-0.012
 cares-0.012
 mysteries-0.012
Token
Token ID198
Feature activation-1.33e-02

Pos logit contributions
rily+0.029
stack+0.028
ishly+0.027
 afloat+0.025
 themselves+0.024
Neg logit contributions
clock-0.022
ups-0.020
up-0.020
 cares-0.020
rays-0.020
Token"
Token ID1
Feature activation+1.87e-03

Pos logit contributions
stack+0.021
rily+0.020
ishly+0.020
 themselves+0.019
mine+0.017
Neg logit contributions
clock-0.024
 cares-0.022
rays-0.022
 grows-0.022
 happens-0.021
TokenPlease
Token ID5492
Feature activation-3.98e-03

Pos logit contributions
clock+0.003
ups+0.003
 cares+0.003
up+0.003
 grows+0.003
Neg logit contributions
stack-0.004
rily-0.004
ishly-0.003
 themselves-0.003
bled-0.003
Token mum
Token ID25682
Feature activation+9.74e-02

Pos logit contributions
stack+0.007
ishly+0.006
rily+0.006
 afloat+0.005
ter+0.005
Neg logit contributions
clock-0.007
ups-0.007
up-0.007
 mysteries-0.007
 cares-0.006
Token,
Token ID11
Feature activation-7.68e-03

Pos logit contributions
clock+0.154
ups+0.143
rays+0.143
 cares+0.142
 mysteries+0.141
Neg logit contributions
rily-0.177
stack-0.174
ishly-0.170
 themselves-0.153
 afloat-0.151
Token I
Token ID314
Feature activation+8.68e-03

Pos logit contributions
stack+0.016
rily+0.016
ishly+0.016
 themselves+0.014
 afloat+0.014
Neg logit contributions
clock-0.010
ups-0.010
up-0.009
rays-0.009
 cares-0.009
Token want
Token ID765
Feature activation-2.57e-02

Pos logit contributions
clock+0.016
ups+0.015
rays+0.015
 mysteries+0.014
 facts+0.014
Neg logit contributions
stack-0.013
rily-0.013
ishly-0.013
 afloat-0.012
 towel-0.011
Token a
Token ID257
Feature activation+2.18e-02

Pos logit contributions
rily+0.036
 themselves+0.035
 towel+0.035
 Nem+0.034
stack+0.032
Neg logit contributions
clock-0.047
 cares-0.044
neys-0.043
ups-0.042
up-0.042
Token spoon
Token ID24556
Feature activation+3.94e-02

Pos logit contributions
clock+0.052
ups+0.051
rays+0.050
neys+0.050
 cares+0.048
Neg logit contributions
rily-0.025
 towel-0.021
stack-0.020
ishly-0.018
ula-0.017
Token there
Token ID612
Feature activation-2.14e-02

Pos logit contributions
ishly+0.016
rily+0.016
stack+0.015
 stret+0.014
ured+0.014
Neg logit contributions
 grows-0.007
 looks-0.007
 is-0.006
 hears-0.006
?-0.006
Token lived
Token ID5615
Feature activation-1.71e-02

Pos logit contributions
rily+0.057
ishly+0.054
stack+0.054
 afloat+0.051
ured+0.050
Neg logit contributions
 grows-0.018
 looks-0.017
clock-0.016
 is-0.015
 dreams-0.015
Token two
Token ID734
Feature activation+6.23e-03

Pos logit contributions
rily+0.043
stack+0.043
ishly+0.043
 afloat+0.037
pan+0.036
Neg logit contributions
clock-0.016
up-0.015
 grows-0.014
ups-0.014
 mysteries-0.013
Token sisters
Token ID15153
Feature activation+2.96e-02

Pos logit contributions
clock+0.010
rays+0.009
ups+0.009
up+0.009
 cares+0.009
Neg logit contributions
rily-0.012
stack-0.011
ishly-0.011
 themselves-0.010
 afloat-0.010
Token -
Token ID532
Feature activation+1.15e-02

Pos logit contributions
clock+0.051
ups+0.048
up+0.047
 cares+0.047
rays+0.047
Neg logit contributions
rily-0.052
stack-0.050
ishly-0.047
 afloat-0.044
bled-0.043
Token Mon
Token ID2892
Feature activation+9.68e-02

Pos logit contributions
clock+0.024
ups+0.023
up+0.022
rays+0.022
 cares+0.022
Neg logit contributions
rily-0.016
stack-0.016
ishly-0.015
 themselves-0.013
 afloat-0.013
Tokena
Token ID64
Feature activation+4.94e-02

Pos logit contributions
clock+0.146
rays+0.133
ups+0.133
up+0.131
 cares+0.131
Neg logit contributions
rily-0.190
stack-0.181
ishly-0.176
 themselves-0.163
 afloat-0.161
Token and
Token ID290
Feature activation+2.01e-03

Pos logit contributions
clock+0.080
ups+0.075
up+0.074
 cares+0.073
rays+0.073
Neg logit contributions
rily-0.092
stack-0.090
ishly-0.086
 afloat-0.079
bled-0.077
Token T
Token ID309
Feature activation+4.08e-02

Pos logit contributions
 cares+0.003
clock+0.003
 listens+0.003
 dreams+0.003
 grows+0.003
Neg logit contributions
rily-0.004
stack-0.003
 afloat-0.003
ishly-0.003
bled-0.003
Tokenoni
Token ID14651
Feature activation+3.76e-02

Pos logit contributions
clock+0.064
rays+0.058
ups+0.058
 cares+0.057
 grows+0.057
Neg logit contributions
rily-0.079
stack-0.075
ishly-0.073
 themselves-0.067
bled-0.067
Token.
Token ID13
Feature activation-5.05e-03

Pos logit contributions
clock+0.057
ups+0.053
up+0.053
 cares+0.052
rays+0.052
Neg logit contributions
rily-0.073
stack-0.072
ishly-0.069
 afloat-0.063
 themselves-0.062
Token,
Token ID11
Feature activation+4.88e-03

Pos logit contributions
clock+0.138
ups+0.127
rays+0.124
up+0.123
 cares+0.122
Neg logit contributions
rily-0.145
stack-0.145
ishly-0.142
 themselves-0.126
 afloat-0.121
Token y
Token ID331
Feature activation+9.65e-03

Pos logit contributions
 grows+0.006
 looks+0.006
 cares+0.006
 listens+0.006
 knows+0.005
Neg logit contributions
rily-0.011
stack-0.011
ishly-0.010
bled-0.010
 afloat-0.010
Tokenum
Token ID388
Feature activation+9.24e-02

Pos logit contributions
clock+0.014
rays+0.013
 cares+0.013
ups+0.013
 listens+0.012
Neg logit contributions
rily-0.020
stack-0.018
ishly-0.018
 themselves-0.017
 afloat-0.017
Token,
Token ID11
Feature activation+1.11e-02

Pos logit contributions
clock+0.147
ups+0.135
rays+0.133
 cares+0.132
up+0.132
Neg logit contributions
rily-0.172
stack-0.170
ishly-0.164
 themselves-0.151
 afloat-0.146
Token y
Token ID331
Feature activation+1.30e-02

Pos logit contributions
 grows+0.014
 looks+0.014
 cares+0.014
 listens+0.013
 knows+0.013
Neg logit contributions
rily-0.024
stack-0.023
bled-0.022
 afloat-0.022
ishly-0.022
Tokenum
Token ID388
Feature activation+9.65e-02

Pos logit contributions
clock+0.019
rays+0.018
 cares+0.017
ups+0.017
 listens+0.017
Neg logit contributions
rily-0.027
stack-0.025
ishly-0.025
 themselves-0.023
 afloat-0.023
Token,"
Token ID553
Feature activation+8.07e-03

Pos logit contributions
clock+0.150
ups+0.137
 cares+0.136
up+0.135
rays+0.135
Neg logit contributions
rily-0.184
stack-0.180
ishly-0.173
 themselves-0.161
 afloat-0.157
Token the
Token ID262
Feature activation+7.05e-03

Pos logit contributions
clock+0.015
 cares+0.014
 grows+0.014
ups+0.014
up+0.014
Neg logit contributions
rily-0.013
stack-0.013
ishly-0.012
 afloat-0.011
bled-0.011
Token k
Token ID479
Feature activation-6.67e-02

Pos logit contributions
clock+0.012
up+0.011
ups+0.011
 grows+0.011
 knows+0.010
Neg logit contributions
rily-0.013
stack-0.012
ishly-0.012
 afloat-0.011
bled-0.011
Tokenang
Token ID648
Feature activation-3.59e-02

Pos logit contributions
rily+0.097
stack+0.096
ishly+0.092
 afloat+0.080
 themselves+0.080
Neg logit contributions
clock-0.134
ups-0.126
 cares-0.125
up-0.125
rays-0.124
Tokenaroo
Token ID38049
Feature activation-2.85e-02

Pos logit contributions
rily+0.048
stack+0.044
ishly+0.042
 afloat+0.041
 themselves+0.037
Neg logit contributions
clock-0.073
up-0.072
ups-0.072
rays-0.071
 grows-0.071
Token-
Token ID12
Feature activation+1.37e-02

Pos logit contributions
clock+0.148
ups+0.139
up+0.134
rays+0.134
 cares+0.131
Neg logit contributions
stack-0.157
rily-0.156
ishly-0.153
 themselves-0.136
 afloat-0.136
Tokend
Token ID67
Feature activation+1.50e-02

Pos logit contributions
clock+0.021
 dreams+0.020
 cares+0.020
 visits+0.019
 grows+0.019
Neg logit contributions
stack-0.025
rily-0.025
ishly-0.024
 themselves-0.022
ter-0.022
Tokenum
Token ID388
Feature activation+9.33e-02

Pos logit contributions
clock+0.024
ups+0.022
rays+0.021
up+0.021
 cares+0.021
Neg logit contributions
rily-0.028
stack-0.027
ishly-0.027
 themselves-0.025
 afloat-0.024
Token-
Token ID12
Feature activation+1.47e-02

Pos logit contributions
clock+0.143
ups+0.134
up+0.131
rays+0.129
 cares+0.128
Neg logit contributions
rily-0.176
stack-0.174
ishly-0.169
 themselves-0.153
 afloat-0.151
Tokend
Token ID67
Feature activation+1.60e-02

Pos logit contributions
clock+0.023
 cares+0.021
 dreams+0.021
 grows+0.020
 visits+0.020
Neg logit contributions
stack-0.028
rily-0.027
ishly-0.026
 themselves-0.024
bled-0.024
Tokenum
Token ID388
Feature activation+9.62e-02

Pos logit contributions
clock+0.025
ups+0.023
rays+0.022
up+0.022
 cares+0.022
Neg logit contributions
rily-0.030
stack-0.029
ishly-0.028
 themselves-0.026
 afloat-0.026
Token!"
Token ID2474
Feature activation-2.40e-03

Pos logit contributions
clock+0.142
ups+0.133
up+0.131
 cares+0.129
rays+0.129
Neg logit contributions
rily-0.187
stack-0.183
ishly-0.178
 themselves-0.162
 afloat-0.159
Token
Token ID198
Feature activation-8.92e-03

Pos logit contributions
rily+0.004
stack+0.004
ishly+0.004
 themselves+0.004
 afloat+0.003
Neg logit contributions
clock-0.004
rays-0.004
ups-0.004
 cares-0.004
up-0.004
Token
Token ID198
Feature activation-1.13e-02

Pos logit contributions
rily+0.015
stack+0.015
ishly+0.014
 themselves+0.013
bled+0.013
Neg logit contributions
clock-0.015
 cares-0.014
rays-0.014
ups-0.014
 grows-0.014
Token"
Token ID1
Feature activation+6.41e-03

Pos logit contributions
rily+0.022
stack+0.022
 themselves+0.020
ishly+0.020
bled+0.019
Neg logit contributions
clock-0.016
rays-0.016
 cares-0.015
 mysteries-0.015
 dreams-0.015
TokenOK
Token ID11380
Feature activation+2.45e-04

Pos logit contributions
clock+0.009
ups+0.008
up+0.008
 cares+0.008
 grows+0.008
Neg logit contributions
rily-0.013
stack-0.013
ishly-0.013
 afloat-0.012
 themselves-0.011
Token,
Token ID11
Feature activation+5.06e-03

Pos logit contributions
clock+0.143
rays+0.133
ups+0.132
 mysteries+0.130
up+0.127
Neg logit contributions
rily-0.138
stack-0.134
ishly-0.131
 themselves-0.117
ter-0.113
Token y
Token ID331
Feature activation+1.05e-02

Pos logit contributions
clock+0.007
 cares+0.007
 grows+0.007
up+0.007
ups+0.006
Neg logit contributions
rily-0.010
stack-0.010
ishly-0.010
 afloat-0.010
bled-0.009
Tokenum
Token ID388
Feature activation+9.27e-02

Pos logit contributions
clock+0.016
rays+0.015
 cares+0.015
 listens+0.014
 mysteries+0.014
Neg logit contributions
rily-0.021
ishly-0.019
stack-0.019
 themselves-0.018
bled-0.018
Token,
Token ID11
Feature activation+9.96e-03

Pos logit contributions
clock+0.156
rays+0.147
ups+0.144
 mysteries+0.143
 cares+0.142
Neg logit contributions
rily-0.158
stack-0.151
ishly-0.143
 themselves-0.137
ter-0.125
Token y
Token ID331
Feature activation+1.36e-02

Pos logit contributions
clock+0.014
 grows+0.013
 cares+0.013
up+0.013
ups+0.013
Neg logit contributions
stack-0.020
rily-0.020
ishly-0.020
 afloat-0.019
bled-0.018
Tokenum
Token ID388
Feature activation+9.54e-02

Pos logit contributions
clock+0.020
rays+0.020
 cares+0.019
 listens+0.018
 mysteries+0.018
Neg logit contributions
rily-0.027
stack-0.025
ishly-0.025
 themselves-0.023
bled-0.023
Token!"
Token ID2474
Feature activation+1.15e-02

Pos logit contributions
clock+0.159
rays+0.149
ups+0.146
 cares+0.146
 grows+0.145
Neg logit contributions
rily-0.166
stack-0.156
ishly-0.148
 themselves-0.144
 afloat-0.131
Token Tom
Token ID4186
Feature activation-1.77e-02

Pos logit contributions
clock+0.021
 cares+0.019
ups+0.019
up+0.019
 grows+0.019
Neg logit contributions
stack-0.020
rily-0.020
ishly-0.019
 afloat-0.018
bled-0.017
Token says
Token ID1139
Feature activation+9.81e-03

Pos logit contributions
rily+0.035
ishly+0.033
stack+0.033
 themselves+0.031
 afloat+0.029
Neg logit contributions
clock-0.025
rays-0.024
ups-0.023
 mysteries-0.023
up-0.023
Token.
Token ID13
Feature activation+4.61e-03

Pos logit contributions
clock+0.013
ups+0.012
up+0.012
rays+0.011
 cares+0.011
Neg logit contributions
stack-0.022
rily-0.021
ishly-0.020
 afloat-0.019
bled-0.018
Token "
Token ID366
Feature activation+1.49e-02

Pos logit contributions
clock+0.007
ups+0.006
up+0.006
 grows+0.006
rays+0.006
Neg logit contributions
stack-0.009
rily-0.009
ishly-0.009
 afloat-0.008
bled-0.008
Token,
Token ID11
Feature activation+8.56e-03

Pos logit contributions
clock+0.137
ups+0.126
rays+0.125
up+0.122
 mysteries+0.122
Neg logit contributions
stack-0.133
rily-0.133
ishly-0.131
 themselves-0.117
ter-0.110
Token y
Token ID331
Feature activation+1.07e-02

Pos logit contributions
 grows+0.010
 looks+0.010
 cares+0.010
 knows+0.009
 listens+0.009
Neg logit contributions
rily-0.019
stack-0.019
 afloat-0.018
ishly-0.018
bled-0.018
Tokenum
Token ID388
Feature activation+9.07e-02

Pos logit contributions
clock+0.015
rays+0.014
 cares+0.014
ups+0.014
 mysteries+0.013
Neg logit contributions
rily-0.022
stack-0.021
ishly-0.021
 themselves-0.020
 afloat-0.019
Token,
Token ID11
Feature activation+1.61e-02

Pos logit contributions
clock+0.149
ups+0.138
rays+0.136
 cares+0.135
up+0.135
Neg logit contributions
rily-0.163
stack-0.161
ishly-0.154
 themselves-0.143
 afloat-0.137
Token y
Token ID331
Feature activation+1.50e-02

Pos logit contributions
 grows+0.020
 looks+0.020
clock+0.019
 cares+0.019
 knows+0.019
Neg logit contributions
rily-0.033
stack-0.033
 afloat-0.032
ishly-0.032
bled-0.031
Tokenum
Token ID388
Feature activation+9.54e-02

Pos logit contributions
clock+0.021
rays+0.020
 cares+0.019
ups+0.019
 listens+0.018
Neg logit contributions
rily-0.031
stack-0.029
ishly-0.029
 themselves-0.027
 afloat-0.027
Token,"
Token ID553
Feature activation+7.03e-03

Pos logit contributions
clock+0.151
ups+0.139
 cares+0.138
up+0.137
rays+0.137
Neg logit contributions
rily-0.178
stack-0.174
ishly-0.168
 themselves-0.156
 afloat-0.152
Token Anna
Token ID11735
Feature activation-2.98e-03

Pos logit contributions
clock+0.014
ups+0.013
up+0.013
 cares+0.013
 grows+0.013
Neg logit contributions
rily-0.011
stack-0.010
ishly-0.010
 afloat-0.009
bled-0.009
Token and
Token ID290
Feature activation-4.55e-03

Pos logit contributions
rily+0.006
stack+0.006
ishly+0.006
 themselves+0.005
 afloat+0.005
Neg logit contributions
clock-0.004
ups-0.004
rays-0.004
up-0.004
 cares-0.004
Token Ben
Token ID3932
Feature activation-1.41e-02

Pos logit contributions
rily+0.008
stack+0.007
ishly+0.007
 afloat+0.007
bled+0.006
Neg logit contributions
clock-0.008
 cares-0.008
 grows-0.008
 listens-0.008
up-0.008
Token say
Token ID910
Feature activation+9.28e-03

Pos logit contributions
rily+0.026
stack+0.025
ishly+0.024
bled+0.023
 afloat+0.022
Neg logit contributions
clock-0.023
up-0.022
 cares-0.022
 grows-0.021
ups-0.021
Token"
Token ID1
Feature activation+1.50e-02

Pos logit contributions
clock+0.003
rays+0.003
ups+0.003
up+0.003
 cares+0.003
Neg logit contributions
rily-0.004
stack-0.004
ishly-0.003
 themselves-0.003
 afloat-0.003
TokenWe
Token ID1135
Feature activation-2.18e-02

Pos logit contributions
clock+0.019
ups+0.018
 cares+0.017
up+0.017
 grows+0.017
Neg logit contributions
rily-0.034
stack-0.033
ishly-0.033
 afloat-0.030
bled-0.029
Token are
Token ID389
Feature activation-8.23e-03

Pos logit contributions
rily+0.047
stack+0.045
ishly+0.043
bled+0.040
 afloat+0.039
Neg logit contributions
clock-0.030
 cares-0.029
up-0.029
 grows-0.028
ups-0.028
Token sorry
Token ID7926
Feature activation-6.74e-03

Pos logit contributions
rily+0.014
stack+0.014
ishly+0.013
 afloat+0.012
bled+0.011
Neg logit contributions
clock-0.015
ups-0.014
 grows-0.013
 cares-0.013
up-0.013
Token,
Token ID11
Feature activation+2.13e-02

Pos logit contributions
rily+0.013
stack+0.012
ishly+0.012
 themselves+0.011
 afloat+0.011
Neg logit contributions
clock-0.011
rays-0.010
ups-0.010
up-0.010
 cares-0.010
Token Mum
Token ID38367
Feature activation+9.52e-02

Pos logit contributions
clock+0.029
 cares+0.028
 looks+0.028
 grows+0.027
 dreams+0.027
Neg logit contributions
rily-0.044
stack-0.043
ishly-0.043
ged-0.041
 afloat-0.040
Token.
Token ID13
Feature activation+3.41e-02

Pos logit contributions
clock+0.153
ups+0.142
up+0.141
 cares+0.140
rays+0.139
Neg logit contributions
rily-0.176
stack-0.172
ishly-0.166
 afloat-0.151
 themselves-0.150
Token We
Token ID775
Feature activation-1.03e-02

Pos logit contributions
clock+0.047
ups+0.045
up+0.045
rays+0.042
 cares+0.042
Neg logit contributions
stack-0.070
rily-0.070
ishly-0.067
 themselves-0.062
 afloat-0.061
Token are
Token ID389
Feature activation-1.57e-03

Pos logit contributions
rily+0.022
stack+0.021
ishly+0.019
bled+0.018
 afloat+0.018
Neg logit contributions
clock-0.015
 cares-0.015
up-0.014
ups-0.014
 grows-0.014
Token sorry
Token ID7926
Feature activation-7.01e-03

Pos logit contributions
rily+0.003
stack+0.003
ishly+0.003
 afloat+0.002
bled+0.002
Neg logit contributions
clock-0.003
ups-0.003
 cares-0.003
up-0.003
 grows-0.003
Token,
Token ID11
Feature activation+2.02e-02

Pos logit contributions
rily+0.012
stack+0.011
ishly+0.011
 themselves+0.011
 afloat+0.010
Neg logit contributions
clock-0.012
rays-0.011
up-0.011
 cares-0.011
ups-0.011
Token,
Token ID11
Feature activation+9.72e-03

Pos logit contributions
clock+0.146
rays+0.135
 mysteries+0.134
ups+0.134
 grows+0.129
Neg logit contributions
rily-0.129
ishly-0.123
stack-0.121
 themselves-0.109
ter-0.105
Token y
Token ID331
Feature activation+1.13e-02

Pos logit contributions
 cares+0.012
clock+0.012
 grows+0.012
 listens+0.012
 looks+0.012
Neg logit contributions
rily-0.021
stack-0.021
ishly-0.020
 afloat-0.020
bled-0.019
Tokenum
Token ID388
Feature activation+9.10e-02

Pos logit contributions
clock+0.018
rays+0.017
 cares+0.016
 mysteries+0.016
ups+0.016
Neg logit contributions
rily-0.022
ishly-0.020
stack-0.020
 themselves-0.018
bled-0.018
Token,
Token ID11
Feature activation+1.37e-02

Pos logit contributions
clock+0.158
rays+0.149
ups+0.147
 mysteries+0.145
 grows+0.143
Neg logit contributions
rily-0.152
stack-0.142
ishly-0.138
 themselves-0.132
ter-0.119
Token y
Token ID331
Feature activation+1.41e-02

Pos logit contributions
clock+0.018
 cares+0.018
 grows+0.018
 listens+0.018
 looks+0.017
Neg logit contributions
stack-0.029
rily-0.028
ishly-0.027
 afloat-0.027
bled-0.025
Tokenum
Token ID388
Feature activation+9.51e-02

Pos logit contributions
clock+0.022
rays+0.021
 cares+0.020
 mysteries+0.020
ups+0.020
Neg logit contributions
rily-0.027
ishly-0.025
stack-0.025
 themselves-0.023
 afloat-0.023
Token,"
Token ID553
Feature activation+1.11e-02

Pos logit contributions
clock+0.161
rays+0.151
ups+0.148
 mysteries+0.147
 cares+0.147
Neg logit contributions
rily-0.165
stack-0.153
ishly-0.148
 themselves-0.143
 afloat-0.131
Token they
Token ID484
Feature activation-1.06e-02

Pos logit contributions
clock+0.020
 cares+0.020
 grows+0.019
ups+0.019
up+0.019
Neg logit contributions
rily-0.018
stack-0.018
ishly-0.017
 afloat-0.016
bled-0.015
Token say
Token ID910
Feature activation+2.12e-02

Pos logit contributions
rily+0.023
stack+0.023
bled+0.021
ishly+0.021
 afloat+0.020
Neg logit contributions
up-0.014
clock-0.014
 cares-0.014
 dreams-0.013
 listens-0.013
Token.
Token ID13
Feature activation+2.99e-02

Pos logit contributions
clock+0.028
ups+0.026
up+0.026
rays+0.025
 cares+0.025
Neg logit contributions
stack-0.045
rily-0.044
ishly-0.043
 afloat-0.040
bled-0.038
Token
Token ID198
Feature activation+1.22e-02

Pos logit contributions
clock+0.034
 grows+0.030
ups+0.029
 cares+0.029
 listens+0.029
Neg logit contributions
stack-0.069
rily-0.069
ishly-0.068
 afloat-0.064
bled-0.061
Token that
Token ID326
Feature activation-3.96e-02

Pos logit contributions
rily+0.020
stack+0.020
ishly+0.019
 afloat+0.017
 themselves+0.017
Neg logit contributions
clock-0.019
ups-0.018
up-0.018
 cares-0.018
rays-0.018
Token day
Token ID1110
Feature activation-2.04e-02

Pos logit contributions
rily+0.076
stack+0.074
ishly+0.071
 afloat+0.065
 themselves+0.065
Neg logit contributions
clock-0.061
ups-0.057
up-0.056
rays-0.055
 cares-0.055
Token.
Token ID13
Feature activation-5.87e-03

Pos logit contributions
rily+0.044
stack+0.043
ishly+0.041
 afloat+0.038
bled+0.038
Neg logit contributions
clock-0.027
up-0.025
ups-0.025
rays-0.024
 cares-0.024
Token<|endoftext|>
Token ID50256
Feature activation+3.08e-02

Pos logit contributions
rily+0.012
stack+0.012
ishly+0.012
 afloat+0.011
bled+0.011
Neg logit contributions
clock-0.008
ups-0.007
rays-0.007
 cares-0.007
 grows-0.007
TokenM
Token ID44
Feature activation+6.62e-02

Pos logit contributions
clock+0.041
ups+0.037
up+0.037
rays+0.036
 cares+0.036
Neg logit contributions
rily-0.066
stack-0.065
ishly-0.063
 afloat-0.058
 themselves-0.057
Tokenum
Token ID388
Feature activation+9.50e-02

Pos logit contributions
clock+0.101
rays+0.097
ups+0.096
 cares+0.094
up+0.092
Neg logit contributions
rily-0.125
stack-0.123
ishly-0.117
 themselves-0.107
bled-0.106
Token woke
Token ID19092
Feature activation+5.68e-03

Pos logit contributions
clock+0.162
ups+0.149
up+0.146
 cares+0.146
rays+0.144
Neg logit contributions
rily-0.167
stack-0.164
ishly-0.158
 themselves-0.141
bled-0.139
Token up
Token ID510
Feature activation+2.08e-02

Pos logit contributions
clock+0.008
up+0.008
ups+0.008
rays+0.008
 grows+0.008
Neg logit contributions
stack-0.011
rily-0.010
ishly-0.010
 afloat-0.010
 themselves-0.009
Token early
Token ID1903
Feature activation+1.55e-02

Pos logit contributions
clock+0.030
 grows+0.027
rays+0.026
ups+0.026
up+0.026
Neg logit contributions
stack-0.041
rily-0.041
ishly-0.040
bled-0.036
 afloat-0.036
Token.
Token ID13
Feature activation-1.46e-03

Pos logit contributions
clock+0.021
 grows+0.018
rays+0.018
up+0.018
ups+0.017
Neg logit contributions
rily-0.032
stack-0.032
ishly-0.031
 themselves-0.029
bled-0.029
Token She
Token ID1375
Feature activation+3.49e-02

Pos logit contributions
rily+0.003
stack+0.003
ishly+0.003
 afloat+0.003
bled+0.003
Neg logit contributions
clock-0.002
rays-0.002
ups-0.002
 grows-0.002
up-0.002
Token He
Token ID679
Feature activation+1.58e-02

Pos logit contributions
rily+0.009
stack+0.009
ishly+0.008
bled+0.008
 afloat+0.008
Neg logit contributions
clock-0.007
rays-0.006
 grows-0.006
ups-0.006
 cares-0.006
Token kept
Token ID4030
Feature activation+1.69e-02

Pos logit contributions
clock+0.027
ups+0.026
rays+0.025
up+0.025
 cares+0.025
Neg logit contributions
rily-0.027
stack-0.026
ishly-0.026
 afloat-0.023
 themselves-0.023
Token saying
Token ID2282
Feature activation+1.87e-02

Pos logit contributions
clock+0.026
rays+0.023
ups+0.023
 grows+0.023
 listens+0.023
Neg logit contributions
rily-0.034
stack-0.033
ishly-0.031
 afloat-0.028
bled-0.028
Token "
Token ID366
Feature activation-2.85e-03

Pos logit contributions
clock+0.033
ups+0.031
up+0.030
 cares+0.030
rays+0.030
Neg logit contributions
rily-0.032
stack-0.031
ishly-0.030
 themselves-0.027
 afloat-0.027
TokenPlease
Token ID5492
Feature activation-1.17e-02

Pos logit contributions
rily+0.006
stack+0.006
ishly+0.006
 afloat+0.005
 themselves+0.005
Neg logit contributions
clock-0.004
ups-0.004
up-0.004
rays-0.004
 cares-0.004
Token mum
Token ID25682
Feature activation+9.50e-02

Pos logit contributions
rily+0.021
stack+0.021
ishly+0.020
 afloat+0.018
 themselves+0.018
Neg logit contributions
clock-0.019
ups-0.018
up-0.018
 cares-0.017
rays-0.017
Token,
Token ID11
Feature activation-1.00e-02

Pos logit contributions
clock+0.140
ups+0.129
up+0.128
 cares+0.128
rays+0.127
Neg logit contributions
rily-0.186
stack-0.183
ishly-0.177
 themselves-0.161
 afloat-0.160
Token can
Token ID460
Feature activation-2.99e-03

Pos logit contributions
rily+0.022
stack+0.021
ishly+0.021
bled+0.020
 afloat+0.020
Neg logit contributions
clock-0.012
 cares-0.012
 grows-0.011
ups-0.011
up-0.011
Token I
Token ID314
Feature activation+1.39e-02

Pos logit contributions
rily+0.007
stack+0.007
ishly+0.007
 afloat+0.006
 themselves+0.006
Neg logit contributions
clock-0.003
ups-0.003
up-0.003
 cares-0.003
rays-0.003
Token go
Token ID467
Feature activation+1.12e-02

Pos logit contributions
clock+0.022
ups+0.021
rays+0.020
up+0.020
 cares+0.020
Neg logit contributions
rily-0.025
stack-0.025
ishly-0.024
 afloat-0.022
 themselves-0.022
Token to
Token ID284
Feature activation+3.89e-02

Pos logit contributions
clock+0.016
ups+0.015
up+0.015
 cares+0.014
rays+0.014
Neg logit contributions
rily-0.023
stack-0.022
ishly-0.022
 afloat-0.020
 themselves-0.020
Token"
Token ID1
Feature activation+1.80e-02

Pos logit contributions
clock+0.003
ups+0.003
up+0.003
 cares+0.003
rays+0.003
Neg logit contributions
rily-0.006
stack-0.006
ishly-0.005
 afloat-0.005
bled-0.005
TokenY
Token ID56
Feature activation+4.54e-02

Pos logit contributions
clock+0.021
 cares+0.020
 grows+0.019
up+0.019
rays+0.019
Neg logit contributions
rily-0.042
ishly-0.040
stack-0.040
 afloat-0.037
bled-0.035
Tokenum
Token ID388
Feature activation+8.72e-02

Pos logit contributions
 mysteries+0.068
clock+0.066
 cares+0.061
 visits+0.060
rays+0.060
Neg logit contributions
rily-0.088
ishly-0.087
stack-0.082
bled-0.079
 themselves-0.075
Token,
Token ID11
Feature activation+1.22e-02

Pos logit contributions
clock+0.149
rays+0.135
ups+0.134
 mysteries+0.134
 cares+0.130
Neg logit contributions
rily-0.151
stack-0.147
ishly-0.146
 themselves-0.126
ter-0.124
Token y
Token ID331
Feature activation+1.31e-02

Pos logit contributions
 cares+0.014
 grows+0.013
 looks+0.013
 knows+0.013
up+0.013
Neg logit contributions
rily-0.028
stack-0.027
 afloat-0.027
ishly-0.026
bled-0.025
Tokenum
Token ID388
Feature activation+9.48e-02

Pos logit contributions
clock+0.018
rays+0.017
 cares+0.016
ups+0.016
 mysteries+0.016
Neg logit contributions
rily-0.028
stack-0.026
ishly-0.026
 themselves-0.024
bled-0.024
Token,
Token ID11
Feature activation+1.81e-02

Pos logit contributions
clock+0.152
rays+0.139
ups+0.137
 cares+0.136
 mysteries+0.136
Neg logit contributions
rily-0.175
stack-0.168
ishly-0.163
 themselves-0.151
 afloat-0.144
Token y
Token ID331
Feature activation+1.45e-02

Pos logit contributions
 grows+0.023
clock+0.023
 cares+0.023
 looks+0.022
up+0.022
Neg logit contributions
rily-0.037
stack-0.037
 afloat-0.036
ishly-0.036
bled-0.034
Tokenum
Token ID388
Feature activation+9.38e-02

Pos logit contributions
clock+0.021
rays+0.019
 cares+0.019
ups+0.018
 mysteries+0.018
Neg logit contributions
rily-0.030
stack-0.028
ishly-0.028
 themselves-0.026
 afloat-0.026
Token,"
Token ID553
Feature activation+4.69e-03

Pos logit contributions
clock+0.146
rays+0.134
ups+0.131
 cares+0.131
 grows+0.130
Neg logit contributions
rily-0.178
stack-0.170
ishly-0.164
 themselves-0.154
 afloat-0.149
Token she
Token ID673
Feature activation+3.48e-02

Pos logit contributions
clock+0.007
 cares+0.007
 grows+0.006
ups+0.006
 dreams+0.006
Neg logit contributions
rily-0.009
stack-0.009
 afloat-0.009
ishly-0.008
bled-0.008
Token You
Token ID921
Feature activation+5.93e-04

Pos logit contributions
ups+0.054
up+0.052
clock+0.050
 mysteries+0.048
rays+0.048
Neg logit contributions
stack-0.057
ishly-0.055
rily-0.053
 themselves-0.053
 Woo-0.046
Token just
Token ID655
Feature activation+3.25e-02

Pos logit contributions
clock+0.001
ups+0.001
rays+0.001
up+0.001
 cares+0.001
Neg logit contributions
stack-0.001
rily-0.001
ishly-0.001
 themselves-0.001
 afloat-0.001
Token have
Token ID423
Feature activation+1.09e-02

Pos logit contributions
clock+0.055
ups+0.055
rays+0.052
up+0.051
neys+0.045
Neg logit contributions
stack-0.054
ishly-0.052
rily-0.051
 themselves-0.049
 afloat-0.047
Token to
Token ID284
Feature activation+3.43e-02

Pos logit contributions
rays+0.016
clock+0.015
up+0.015
ups+0.014
ray+0.014
Neg logit contributions
 themselves-0.019
rily-0.018
 towel-0.017
stack-0.016
ishly-0.016
Token str
Token ID965
Feature activation+3.60e-02

Pos logit contributions
clock+0.046
up+0.043
 cares+0.042
ups+0.041
 grows+0.041
Neg logit contributions
rily-0.076
stack-0.073
ishly-0.070
bled-0.065
 afloat-0.064
Tokenum
Token ID388
Feature activation+9.47e-02

Pos logit contributions
clock+0.051
 cares+0.048
rays+0.048
 mysteries+0.047
ups+0.046
Neg logit contributions
rily-0.075
ishly-0.067
 themselves-0.066
bled-0.066
stack-0.065
Token the
Token ID262
Feature activation+1.18e-02

Pos logit contributions
clock+0.158
 cares+0.149
up+0.145
 grows+0.144
 mysteries+0.144
Neg logit contributions
rily-0.149
 themselves-0.142
ishly-0.131
stack-0.129
bled-0.117
Token strings
Token ID13042
Feature activation-8.80e-03

Pos logit contributions
clock+0.020
ups+0.019
up+0.019
 cares+0.018
rays+0.018
Neg logit contributions
stack-0.021
rily-0.021
ishly-0.020
 afloat-0.018
 themselves-0.017
Token and
Token ID290
Feature activation+3.31e-02

Pos logit contributions
rily+0.018
stack+0.018
ishly+0.018
bled+0.016
 afloat+0.016
Neg logit contributions
clock-0.012
ups-0.011
up-0.011
 grows-0.011
 cares-0.011
Token sing
Token ID1702
Feature activation+2.61e-02

Pos logit contributions
clock+0.050
ups+0.047
up+0.047
rays+0.046
 cares+0.046
Neg logit contributions
rily-0.063
stack-0.062
ishly-0.060
 themselves-0.055
 afloat-0.055
Token along
Token ID1863
Feature activation+7.36e-03

Pos logit contributions
clock+0.044
ups+0.041
 cares+0.041
up+0.040
rays+0.040
Neg logit contributions
rily-0.046
stack-0.043
ishly-0.042
 themselves-0.041
bled-0.038
Token to
Token ID284
Feature activation+2.84e-02

Pos logit contributions
rays+0.003
clock+0.003
up+0.003
 listens+0.003
 grows+0.003
Neg logit contributions
 themselves-0.005
stack-0.004
rily-0.004
 towel-0.004
ishly-0.004
Token paint
Token ID7521
Feature activation+5.96e-02

Pos logit contributions
clock+0.040
up+0.037
ups+0.036
 cares+0.036
 grows+0.035
Neg logit contributions
rily-0.061
stack-0.059
ishly-0.056
 afloat-0.052
bled-0.052
Token on
Token ID319
Feature activation-1.85e-02

Pos logit contributions
clock+0.095
 cares+0.090
ups+0.089
up+0.089
rays+0.087
Neg logit contributions
rily-0.106
stack-0.100
 themselves-0.098
ishly-0.097
 afloat-0.091
Token paper
Token ID3348
Feature activation+4.67e-02

Pos logit contributions
rily+0.036
stack+0.035
ishly+0.033
 themselves+0.031
 afloat+0.031
Neg logit contributions
clock-0.028
ups-0.026
up-0.026
 cares-0.026
rays-0.026
Token or
Token ID393
Feature activation-1.02e-04

Pos logit contributions
clock+0.074
ups+0.068
up+0.068
 cares+0.067
rays+0.067
Neg logit contributions
rily-0.088
stack-0.085
ishly-0.082
 themselves-0.075
 afloat-0.075
Token canvas
Token ID21978
Feature activation+9.41e-02

Pos logit contributions
rily+0.000
stack+0.000
ishly+0.000
 themselves+0.000
 afloat+0.000
Neg logit contributions
clock-0.000
ups-0.000
up-0.000
rays-0.000
 cares-0.000
Token.
Token ID13
Feature activation+3.15e-02

Pos logit contributions
clock+0.146
 cares+0.136
up+0.135
ups+0.134
rays+0.133
Neg logit contributions
rily-0.176
stack-0.169
ishly-0.164
 themselves-0.154
 afloat-0.150
Token And
Token ID843
Feature activation+1.01e-02

Pos logit contributions
clock+0.046
ups+0.045
up+0.045
rays+0.043
 cares+0.042
Neg logit contributions
stack-0.059
rily-0.059
ishly-0.057
 themselves-0.054
 afloat-0.050
Token you
Token ID345
Feature activation+2.33e-03

Pos logit contributions
clock+0.013
 cares+0.012
ups+0.012
 grows+0.012
up+0.012
Neg logit contributions
rily-0.022
stack-0.021
ishly-0.021
 afloat-0.019
bled-0.019
Token have
Token ID423
Feature activation+5.69e-03

Pos logit contributions
clock+0.004
ups+0.003
rays+0.003
up+0.003
 cares+0.003
Neg logit contributions
stack-0.004
rily-0.004
ishly-0.004
 themselves-0.004
 afloat-0.004
Token to
Token ID284
Feature activation+2.86e-02

Pos logit contributions
rays+0.008
clock+0.007
up+0.007
 writer+0.007
 listens+0.007
Neg logit contributions
 themselves-0.010
 towel-0.010
stack-0.009
rily-0.009
ishly-0.009
Token All
Token ID1439
Feature activation-1.90e-02

Pos logit contributions
stack+0.036
rily+0.036
ishly+0.034
 themselves+0.031
 afloat+0.029
Neg logit contributions
clock-0.033
ups-0.032
up-0.032
 cares-0.031
rays-0.031
Token the
Token ID262
Feature activation-1.63e-02

Pos logit contributions
rily+0.026
stack+0.025
ishly+0.023
 themselves+0.023
 afloat+0.021
Neg logit contributions
clock-0.040
rays-0.036
ups-0.036
up-0.036
 cares-0.035
Token things
Token ID1243
Feature activation-2.31e-02

Pos logit contributions
rily+0.032
stack+0.031
ishly+0.030
 afloat+0.028
 themselves+0.027
Neg logit contributions
clock-0.025
ups-0.023
up-0.023
 cares-0.023
rays-0.023
Token that
Token ID326
Feature activation-3.48e-02

Pos logit contributions
rily+0.039
stack+0.038
ishly+0.036
 afloat+0.032
 themselves+0.032
Neg logit contributions
clock-0.040
ups-0.039
 grows-0.038
up-0.038
 cares-0.038
Token the
Token ID262
Feature activation-2.95e-02

Pos logit contributions
rily+0.065
stack+0.062
ishly+0.061
 afloat+0.054
bled+0.054
Neg logit contributions
clock-0.057
 grows-0.053
ups-0.053
 cares-0.052
rays-0.052
Token fle
Token ID5104
Feature activation-9.97e-02

Pos logit contributions
rily+0.062
stack+0.061
ishly+0.060
bled+0.053
 afloat+0.051
Neg logit contributions
clock-0.042
 grows-0.040
up-0.038
 dreams-0.038
ups-0.038
Tokena
Token ID64
Feature activation-3.88e-02

Pos logit contributions
rily+0.203
stack+0.198
ishly+0.189
 themselves+0.177
 afloat+0.177
Neg logit contributions
clock-0.143
rays-0.129
ups-0.128
up-0.127
 cares-0.127
Token recorded
Token ID6264
Feature activation-1.12e-02

Pos logit contributions
rily+0.073
stack+0.071
ishly+0.069
 themselves+0.063
 afloat+0.063
Neg logit contributions
clock-0.060
ups-0.055
rays-0.055
up-0.055
 cares-0.054
Token made
Token ID925
Feature activation-1.30e-02

Pos logit contributions
rily+0.022
stack+0.022
ishly+0.021
 afloat+0.019
bled+0.019
Neg logit contributions
clock-0.017
ups-0.016
 grows-0.016
 cares-0.016
up-0.016
Token a
Token ID257
Feature activation-9.83e-03

Pos logit contributions
rily+0.024
stack+0.024
ishly+0.023
 afloat+0.021
bled+0.021
Neg logit contributions
clock-0.021
ups-0.020
up-0.020
rays-0.019
 grows-0.019
Token beautiful
Token ID4950
Feature activation+9.80e-03

Pos logit contributions
rily+0.020
stack+0.020
ishly+0.019
 afloat+0.018
 themselves+0.018
Neg logit contributions
clock-0.014
ups-0.013
up-0.012
 cares-0.012
rays-0.012
Token she
Token ID673
Feature activation+5.03e-03

Pos logit contributions
stack+0.038
rily+0.037
ishly+0.036
 afloat+0.035
 stret+0.034
Neg logit contributions
 grows-0.019
ups-0.019
clock-0.019
 cares-0.019
 looks-0.018
Token noticed
Token ID6810
Feature activation+1.52e-02

Pos logit contributions
clock+0.008
ups+0.007
rays+0.007
up+0.007
 cares+0.007
Neg logit contributions
rily-0.010
stack-0.009
ishly-0.009
 afloat-0.008
bled-0.008
Token that
Token ID326
Feature activation-5.89e-03

Pos logit contributions
clock+0.018
ups+0.018
up+0.018
rays+0.017
 grows+0.016
Neg logit contributions
rily-0.034
stack-0.033
ishly-0.032
 afloat-0.031
bled-0.029
Token the
Token ID262
Feature activation-1.69e-02

Pos logit contributions
rily+0.012
stack+0.012
ishly+0.012
 afloat+0.011
bled+0.011
Neg logit contributions
clock-0.008
 grows-0.008
 mysteries-0.008
 cares-0.008
ups-0.008
Token books
Token ID3835
Feature activation-2.64e-02

Pos logit contributions
stack+0.034
rily+0.034
ishly+0.032
 afloat+0.029
 themselves+0.029
Neg logit contributions
clock-0.022
ups-0.022
up-0.021
 grows-0.021
 cares-0.021
Tokenhe
Token ID258
Feature activation-9.95e-02

Pos logit contributions
rily+0.043
stack+0.040
ishly+0.039
 afloat+0.037
 themselves+0.035
Neg logit contributions
clock-0.047
 grows-0.047
up-0.045
ups-0.045
rays-0.045
Tokenlf
Token ID1652
Feature activation-4.08e-02

Pos logit contributions
rily+0.164
ishly+0.162
stack+0.158
 afloat+0.150
 themselves+0.137
Neg logit contributions
clock-0.179
ups-0.169
rays-0.166
up-0.166
 grows-0.164
Token was
Token ID373
Feature activation-5.25e-02

Pos logit contributions
rily+0.082
stack+0.080
ishly+0.077
 afloat+0.074
 themselves+0.070
Neg logit contributions
clock-0.058
ups-0.055
up-0.054
 grows-0.053
rays-0.052
Token gone
Token ID3750
Feature activation-5.39e-02

Pos logit contributions
rily+0.098
stack+0.096
ishly+0.095
 themselves+0.083
 afloat+0.083
Neg logit contributions
clock-0.083
ups-0.077
 grows-0.076
up-0.075
rays-0.073
Token.
Token ID13
Feature activation-1.48e-02

Pos logit contributions
stack+0.121
rily+0.118
ishly+0.117
bled+0.106
 themselves+0.106
Neg logit contributions
clock-0.065
ups-0.058
up-0.058
 grows-0.058
 looks-0.055
Token It
Token ID632
Feature activation-2.29e-02

Pos logit contributions
rily+0.028
stack+0.027
ishly+0.026
 afloat+0.024
 themselves+0.024
Neg logit contributions
clock-0.023
ups-0.022
up-0.021
 cares-0.021
rays-0.021
Token It
Token ID632
Feature activation-9.63e-03

Pos logit contributions
stack+0.004
rily+0.004
ishly+0.004
 themselves+0.004
 afloat+0.003
Neg logit contributions
clock-0.004
up-0.004
ups-0.004
 cares-0.004
rays-0.004
Token became
Token ID2627
Feature activation-1.22e-02

Pos logit contributions
rily+0.018
stack+0.018
ishly+0.017
 themselves+0.015
 afloat+0.015
Neg logit contributions
clock-0.015
ups-0.014
up-0.014
rays-0.014
 cares-0.014
Token the
Token ID262
Feature activation-3.11e-03

Pos logit contributions
rily+0.023
stack+0.022
ishly+0.021
 afloat+0.020
 themselves+0.020
Neg logit contributions
clock-0.020
ups-0.018
up-0.018
 cares-0.018
rays-0.018
Token happiest
Token ID49414
Feature activation-2.74e-03

Pos logit contributions
stack+0.006
rily+0.006
ishly+0.006
 afloat+0.005
 themselves+0.005
Neg logit contributions
clock-0.005
ups-0.005
up-0.004
rays-0.004
 grows-0.004
Token books
Token ID3835
Feature activation-2.52e-02

Pos logit contributions
rily+0.005
stack+0.005
ishly+0.004
 themselves+0.004
 afloat+0.004
Neg logit contributions
clock-0.005
ups-0.005
up-0.004
 cares-0.004
rays-0.004
Tokenhe
Token ID258
Feature activation-9.76e-02

Pos logit contributions
rily+0.051
ishly+0.050
stack+0.049
 afloat+0.047
 themselves+0.046
Neg logit contributions
clock-0.032
 grows-0.032
rays-0.030
 looks-0.029
up-0.029
Tokenlf
Token ID1652
Feature activation-4.96e-02

Pos logit contributions
ishly+0.198
 afloat+0.187
rily+0.180
stack+0.172
 Woo+0.163
Neg logit contributions
clock-0.127
 grows-0.123
ups-0.119
 math-0.118
 writer-0.118
Token in
Token ID287
Feature activation-3.20e-02

Pos logit contributions
rily+0.095
ishly+0.093
stack+0.093
 afloat+0.090
 themselves+0.083
Neg logit contributions
clock-0.067
ups-0.064
rays-0.063
 grows-0.062
up-0.062
Token the
Token ID262
Feature activation-1.15e-02

Pos logit contributions
ishly+0.068
rily+0.068
stack+0.065
bled+0.059
 afloat+0.059
Neg logit contributions
clock-0.041
 grows-0.041
 mysteries-0.039
 dreams-0.037
 visits-0.037
Token world
Token ID995
Feature activation+7.20e-03

Pos logit contributions
stack+0.020
rily+0.018
ishly+0.018
 afloat+0.016
 themselves+0.016
Neg logit contributions
ups-0.018
up-0.017
 cares-0.017
rays-0.016
 happens-0.016
Token!
Token ID0
Feature activation-1.58e-03

Pos logit contributions
clock+0.010
ups+0.009
up+0.009
 grows+0.009
 cares+0.009
Neg logit contributions
rily-0.015
stack-0.014
ishly-0.014
 afloat-0.013
bled-0.013
Token He
Token ID679
Feature activation+3.34e-03

Pos logit contributions
rily+0.017
stack+0.017
ishly+0.016
 afloat+0.015
 themselves+0.015
Neg logit contributions
clock-0.012
ups-0.011
up-0.011
 cares-0.011
rays-0.011
Token noticed
Token ID6810
Feature activation+1.47e-02

Pos logit contributions
clock+0.006
ups+0.005
rays+0.005
up+0.005
 cares+0.005
Neg logit contributions
rily-0.006
stack-0.006
ishly-0.006
 themselves-0.005
 afloat-0.005
Token that
Token ID326
Feature activation-1.01e-02

Pos logit contributions
clock+0.021
ups+0.020
up+0.020
rays+0.019
 cares+0.019
Neg logit contributions
rily-0.030
stack-0.029
ishly-0.028
 afloat-0.026
bled-0.025
Token the
Token ID262
Feature activation-1.22e-02

Pos logit contributions
rily+0.019
stack+0.019
ishly+0.018
 afloat+0.016
bled+0.016
Neg logit contributions
clock-0.016
ups-0.015
up-0.015
rays-0.015
 cares-0.015
Token books
Token ID3835
Feature activation-2.41e-02

Pos logit contributions
stack+0.023
rily+0.022
ishly+0.021
 afloat+0.019
 themselves+0.019
Neg logit contributions
ups-0.017
 cares-0.017
clock-0.017
 grows-0.017
up-0.016
Tokenhe
Token ID258
Feature activation-9.68e-02

Pos logit contributions
rily+0.037
stack+0.034
ishly+0.034
 afloat+0.031
 themselves+0.030
Neg logit contributions
clock-0.046
 grows-0.045
rays-0.044
up-0.044
ups-0.044
Tokenlf
Token ID1652
Feature activation-3.78e-02

Pos logit contributions
rily+0.147
stack+0.143
ishly+0.141
 afloat+0.127
 themselves+0.121
Neg logit contributions
clock-0.189
ups-0.178
up-0.176
rays-0.175
 grows-0.173
Token was
Token ID373
Feature activation-4.50e-02

Pos logit contributions
rily+0.078
stack+0.076
ishly+0.074
 afloat+0.070
 themselves+0.067
Neg logit contributions
clock-0.052
ups-0.049
up-0.048
 grows-0.047
rays-0.047
Token very
Token ID845
Feature activation-1.46e-03

Pos logit contributions
rily+0.087
stack+0.085
ishly+0.084
 themselves+0.074
 afloat+0.074
Neg logit contributions
clock-0.070
ups-0.064
 grows-0.062
up-0.062
rays-0.061
Token smooth
Token ID7209
Feature activation-7.79e-03

Pos logit contributions
rily+0.003
stack+0.003
ishly+0.003
 afloat+0.003
 themselves+0.003
Neg logit contributions
clock-0.002
rays-0.002
 cares-0.002
ups-0.002
up-0.002
Token and
Token ID290
Feature activation-1.32e-02

Pos logit contributions
rily+0.015
stack+0.015
ishly+0.014
 afloat+0.013
 themselves+0.013
Neg logit contributions
clock-0.012
ups-0.011
up-0.011
rays-0.011
 cares-0.011
Token around
Token ID1088
Feature activation-2.01e-02

Pos logit contributions
stack+0.026
rily+0.025
ishly+0.025
 themselves+0.022
 afloat+0.022
Neg logit contributions
clock-0.023
up-0.022
ups-0.022
 grows-0.022
rays-0.021
Token to
Token ID284
Feature activation-1.02e-02

Pos logit contributions
rily+0.042
stack+0.040
ishly+0.040
bled+0.038
 afloat+0.037
Neg logit contributions
clock-0.025
 grows-0.023
 dreams-0.023
rays-0.023
 looks-0.022
Token all
Token ID477
Feature activation+1.35e-02

Pos logit contributions
rily+0.023
ishly+0.022
stack+0.021
bled+0.020
 afloat+0.020
Neg logit contributions
 dreams-0.012
 grows-0.012
 looks-0.012
clock-0.011
 listens-0.011
Token his
Token ID465
Feature activation-3.75e-02

Pos logit contributions
clock+0.019
ups+0.018
 grows+0.018
up+0.018
 cares+0.017
Neg logit contributions
rily-0.027
stack-0.027
ishly-0.026
bled-0.024
 afloat-0.024
Token feat
Token ID2218
Feature activation-1.77e-02

Pos logit contributions
rily+0.078
stack+0.077
ishly+0.075
 themselves+0.069
 afloat+0.069
Neg logit contributions
clock-0.051
ups-0.047
rays-0.047
up-0.046
 cares-0.045
Tokenhered
Token ID6083
Feature activation-9.51e-02

Pos logit contributions
rily+0.018
stack+0.017
ishly+0.015
 afloat+0.013
bled+0.013
Neg logit contributions
clock-0.044
up-0.042
ups-0.042
 cares-0.042
rays-0.041
Token friends
Token ID2460
Feature activation-2.02e-02

Pos logit contributions
stack+0.203
rily+0.203
ishly+0.194
bled+0.178
 afloat+0.175
Neg logit contributions
clock-0.126
up-0.118
ups-0.118
 cares-0.115
 grows-0.112
Token.
Token ID13
Feature activation-1.50e-02

Pos logit contributions
rily+0.047
stack+0.046
bled+0.044
ishly+0.043
 afloat+0.041
Neg logit contributions
clock-0.019
 looks-0.018
 grows-0.018
 visits-0.018
 dreams-0.018
Token Everyone
Token ID11075
Feature activation-3.95e-03

Pos logit contributions
rily+0.032
stack+0.031
ishly+0.030
 afloat+0.028
bled+0.028
Neg logit contributions
clock-0.020
rays-0.018
ups-0.018
 grows-0.018
 cares-0.017
Token felt
Token ID2936
Feature activation+4.00e-03

Pos logit contributions
rily+0.007
stack+0.007
ishly+0.007
 afloat+0.006
bled+0.006
Neg logit contributions
clock-0.006
ups-0.006
up-0.006
 cares-0.006
 grows-0.006
Token so
Token ID523
Feature activation+1.11e-02

Pos logit contributions
clock+0.006
up+0.006
 grows+0.006
 cares+0.006
ups+0.006
Neg logit contributions
stack-0.008
rily-0.008
ishly-0.007
bled-0.006
 afloat-0.006
Token was
Token ID373
Feature activation-2.65e-02

Pos logit contributions
rily+0.021
stack+0.021
ishly+0.020
 themselves+0.018
 afloat+0.018
Neg logit contributions
clock-0.015
up-0.014
ups-0.014
 grows-0.014
rays-0.014
Token so
Token ID523
Feature activation+5.36e-03

Pos logit contributions
rily+0.056
stack+0.055
ishly+0.052
 tripod+0.045
mine+0.045
Neg logit contributions
clock-0.035
 grows-0.032
rays-0.030
ups-0.030
 tests-0.030
Token strong
Token ID1913
Feature activation+4.10e-03

Pos logit contributions
clock+0.007
ups+0.006
up+0.006
 grows+0.006
rays+0.006
Neg logit contributions
rily-0.012
stack-0.012
ishly-0.012
 themselves-0.011
 afloat-0.011
Token that
Token ID326
Feature activation-2.74e-02

Pos logit contributions
clock+0.006
ups+0.005
up+0.005
rays+0.005
 grows+0.005
Neg logit contributions
rily-0.008
stack-0.008
ishly-0.008
 afloat-0.007
bled-0.007
Token the
Token ID262
Feature activation-2.30e-02

Pos logit contributions
rily+0.047
stack+0.046
ishly+0.044
 afloat+0.039
 themselves+0.039
Neg logit contributions
clock-0.048
ups-0.045
 grows-0.045
up-0.045
rays-0.045
Token fle
Token ID5104
Feature activation-9.44e-02

Pos logit contributions
rily+0.050
stack+0.049
ishly+0.048
bled+0.043
 themselves+0.042
Neg logit contributions
clock-0.031
 grows-0.029
ups-0.028
up-0.028
 cares-0.027
Tokena
Token ID64
Feature activation-3.67e-02

Pos logit contributions
rily+0.207
stack+0.205
ishly+0.201
 afloat+0.183
 themselves+0.181
Neg logit contributions
clock-0.118
ups-0.114
up-0.112
 cares-0.107
 grows-0.106
Token could
Token ID714
Feature activation-2.22e-02

Pos logit contributions
rily+0.073
stack+0.073
ishly+0.069
 themselves+0.064
 afloat+0.064
Neg logit contributions
clock-0.053
ups-0.052
up-0.051
 grows-0.049
 cares-0.049
Token hardly
Token ID8941
Feature activation+4.38e-04

Pos logit contributions
rily+0.048
ishly+0.044
stack+0.043
bled+0.040
 afloat+0.040
Neg logit contributions
clock-0.027
up-0.026
 grows-0.026
rays-0.026
ups-0.025
Token keep
Token ID1394
Feature activation-6.66e-03

Pos logit contributions
clock+0.001
 grows+0.001
rays+0.001
up+0.001
 cares+0.001
Neg logit contributions
rily-0.001
ishly-0.001
stack-0.001
bled-0.001
 themselves-0.001
Token driving
Token ID5059
Feature activation-1.16e-02

Pos logit contributions
stack+0.014
rily+0.013
ishly+0.013
bled+0.011
kin+0.011
Neg logit contributions
clock-0.009
rays-0.008
 grows-0.008
ups-0.008
 mysteries-0.008
Token something
Token ID1223
Feature activation-7.71e-03

Pos logit contributions
clock+0.023
ups+0.022
up+0.022
 cares+0.021
rays+0.021
Neg logit contributions
rily-0.031
stack-0.030
ishly-0.029
 afloat-0.027
 themselves-0.027
Token more
Token ID517
Feature activation+5.25e-03

Pos logit contributions
rily+0.016
stack+0.016
ishly+0.016
 afloat+0.014
bled+0.014
Neg logit contributions
clock-0.010
ups-0.010
up-0.009
 cares-0.009
 grows-0.009
Token about
Token ID546
Feature activation-2.31e-02

Pos logit contributions
clock+0.007
rays+0.007
ups+0.006
up+0.006
 cares+0.006
Neg logit contributions
rily-0.011
stack-0.011
ishly-0.010
 themselves-0.010
 afloat-0.010
Token his
Token ID465
Feature activation-3.37e-02

Pos logit contributions
rily+0.041
ishly+0.040
stack+0.040
 afloat+0.036
bled+0.035
Neg logit contributions
clock-0.038
 grows-0.036
ups-0.036
 mysteries-0.036
rays-0.035
Token feat
Token ID2218
Feature activation-1.69e-02

Pos logit contributions
rily+0.068
stack+0.066
ishly+0.064
bled+0.058
 afloat+0.058
Neg logit contributions
clock-0.050
ups-0.045
 cares-0.045
up-0.045
 grows-0.044
Tokenhered
Token ID6083
Feature activation-9.41e-02

Pos logit contributions
rily+0.022
stack+0.021
ishly+0.019
bled+0.017
 themselves+0.017
Neg logit contributions
clock-0.037
up-0.035
 cares-0.035
ups-0.035
rays-0.035
Token friends
Token ID2460
Feature activation-1.70e-02

Pos logit contributions
rily+0.207
stack+0.204
ishly+0.198
bled+0.185
 afloat+0.182
Neg logit contributions
clock-0.117
ups-0.109
up-0.108
 cares-0.107
 grows-0.105
Token.
Token ID13
Feature activation-8.59e-03

Pos logit contributions
rily+0.037
stack+0.036
ishly+0.034
bled+0.033
 afloat+0.032
Neg logit contributions
clock-0.021
ups-0.020
 grows-0.020
up-0.019
rays-0.019
Token
Token ID198
Feature activation-1.08e-02

Pos logit contributions
rily+0.015
stack+0.015
ishly+0.014
 themselves+0.013
 afloat+0.013
Neg logit contributions
clock-0.014
ups-0.014
up-0.014
 cares-0.013
 grows-0.013
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
rily+0.019
stack+0.018
ishly+0.018
 afloat+0.015
bled+0.015
Neg logit contributions
clock-0.019
ups-0.018
up-0.017
rays-0.017
 cares-0.017
TokenJohn
Token ID7554
Feature activation-8.02e-03

Pos logit contributions
rily+0.019
stack+0.019
ishly+0.018
 themselves+0.017
 afloat+0.017
Neg logit contributions
clock-0.017
 cares-0.016
ups-0.016
up-0.016
 grows-0.016
Token so
Token ID523
Feature activation+5.92e-03

Pos logit contributions
rily+0.069
stack+0.068
ishly+0.068
 afloat+0.059
bled+0.059
Neg logit contributions
clock-0.043
 grows-0.038
 history-0.037
 mysteries-0.036
 dreams-0.036
Token beautiful
Token ID4950
Feature activation+6.09e-03

Pos logit contributions
clock+0.007
ups+0.007
up+0.007
 cares+0.007
rays+0.007
Neg logit contributions
rily-0.013
stack-0.013
ishly-0.013
 afloat-0.012
 themselves-0.012
Token with
Token ID351
Feature activation-3.84e-02

Pos logit contributions
clock+0.010
ups+0.009
 cares+0.009
up+0.009
rays+0.009
Neg logit contributions
rily-0.012
stack-0.011
ishly-0.011
 themselves-0.010
 afloat-0.010
Token its
Token ID663
Feature activation-3.11e-02

Pos logit contributions
stack+0.077
rily+0.076
ishly+0.073
 afloat+0.067
bled+0.067
Neg logit contributions
clock-0.055
rays-0.051
 grows-0.050
 mysteries-0.050
 cares-0.050
Token feat
Token ID2218
Feature activation-1.98e-02

Pos logit contributions
rily+0.064
stack+0.063
ishly+0.061
 afloat+0.056
bled+0.056
Neg logit contributions
clock-0.043
ups-0.040
up-0.039
 cares-0.039
rays-0.039
Tokenhered
Token ID6083
Feature activation-9.35e-02

Pos logit contributions
rily+0.016
stack+0.016
ishly+0.015
 afloat+0.012
 themselves+0.011
Neg logit contributions
clock-0.052
ups-0.050
up-0.049
rays-0.049
 grows-0.049
Token wings
Token ID12098
Feature activation-3.35e-03

Pos logit contributions
rily+0.182
stack+0.177
ishly+0.171
 afloat+0.157
 themselves+0.156
Neg logit contributions
clock-0.144
ups-0.132
up-0.130
rays-0.130
 cares-0.129
Token gl
Token ID1278
Feature activation+7.13e-03

Pos logit contributions
rily+0.007
stack+0.007
ishly+0.007
bled+0.006
 themselves+0.006
Neg logit contributions
clock-0.004
up-0.004
ups-0.004
 grows-0.004
rays-0.004
Tokeniding
Token ID2530
Feature activation-3.80e-02

Pos logit contributions
clock+0.015
ups+0.014
 cares+0.014
 grows+0.014
rays+0.014
Neg logit contributions
rily-0.010
stack-0.010
ishly-0.009
bled-0.008
 afloat-0.008
Token against
Token ID1028
Feature activation-3.40e-02

Pos logit contributions
rily+0.068
stack+0.067
ishly+0.064
 afloat+0.059
bled+0.058
Neg logit contributions
clock-0.060
ups-0.055
rays-0.055
 grows-0.055
up-0.055
Token the
Token ID262
Feature activation-1.14e-02

Pos logit contributions
stack+0.073
ishly+0.070
rily+0.069
bled+0.065
 afloat+0.063
Neg logit contributions
clock-0.044
 mysteries-0.041
rays-0.040
 dreams-0.039
 history-0.039
Tokenmy
Token ID1820
Feature activation+2.35e-03

Pos logit contributions
clock+0.001
ups+0.001
up+0.001
rays+0.001
 cares+0.001
Neg logit contributions
rily-0.002
stack-0.002
ishly-0.002
 afloat-0.002
bled-0.002
Token never
Token ID1239
Feature activation-8.72e-03

Pos logit contributions
up+0.003
clock+0.003
ups+0.003
 cares+0.003
rays+0.003
Neg logit contributions
rily-0.005
stack-0.005
 afloat-0.004
bled-0.004
ishly-0.004
Token saw
Token ID2497
Feature activation+2.38e-02

Pos logit contributions
ishly+0.013
rily+0.013
stack+0.013
 afloat+0.012
ged+0.012
Neg logit contributions
clock-0.015
ups-0.014
rays-0.014
up-0.013
neys-0.013
Token the
Token ID262
Feature activation-6.90e-03

Pos logit contributions
clock+0.035
ups+0.033
 cares+0.033
 grows+0.033
rays+0.032
Neg logit contributions
stack-0.048
rily-0.046
ishly-0.045
 afloat-0.042
bled-0.041
Token books
Token ID3835
Feature activation-5.84e-03

Pos logit contributions
stack+0.013
rily+0.013
ishly+0.012
 themselves+0.011
 spilling+0.011
Neg logit contributions
clock-0.010
ups-0.010
up-0.009
 grows-0.009
rays-0.009
Tokenhe
Token ID258
Feature activation-9.33e-02

Pos logit contributions
stack+0.009
ishly+0.009
rily+0.009
 themselves+0.008
 afloat+0.008
Neg logit contributions
clock-0.010
 grows-0.009
up-0.009
 looks-0.009
ups-0.009
Tokenlf
Token ID1652
Feature activation-2.50e-02

Pos logit contributions
rily+0.153
ishly+0.150
stack+0.149
 afloat+0.140
 themselves+0.130
Neg logit contributions
clock-0.169
ups-0.157
rays-0.156
up-0.154
 grows-0.154
Token again
Token ID757
Feature activation+2.70e-03

Pos logit contributions
stack+0.050
rily+0.049
ishly+0.048
 afloat+0.047
bled+0.046
Neg logit contributions
clock-0.033
ups-0.031
up-0.030
 grows-0.029
rays-0.028
Token.
Token ID13
Feature activation-8.62e-03

Pos logit contributions
clock+0.003
 grows+0.003
ups+0.003
 cares+0.002
up+0.002
Neg logit contributions
stack-0.007
rily-0.007
ishly-0.006
bled-0.006
 afloat-0.006
Token But
Token ID887
Feature activation-1.48e-02

Pos logit contributions
rily+0.017
stack+0.017
ishly+0.016
 afloat+0.015
bled+0.015
Neg logit contributions
clock-0.013
 cares-0.012
ups-0.012
 grows-0.012
up-0.012
Token they
Token ID484
Feature activation-4.41e-02

Pos logit contributions
rily+0.028
stack+0.028
 stret+0.027
 afloat+0.026
ishly+0.025
Neg logit contributions
 grows-0.022
 dreams-0.022
 cares-0.022
 looks-0.021
clock-0.021
Token birds
Token ID10087
Feature activation-5.53e-03

Pos logit contributions
clock+0.001
ups+0.001
up+0.001
 cares+0.001
rays+0.001
Neg logit contributions
rily-0.001
stack-0.001
ishly-0.001
bled-0.001
 afloat-0.001
Token stop
Token ID2245
Feature activation-1.56e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.010
bled+0.010
 themselves+0.009
Neg logit contributions
clock-0.008
up-0.008
 grows-0.008
ups-0.008
rays-0.007
Token singing
Token ID13777
Feature activation-8.12e-04

Pos logit contributions
rily+0.030
stack+0.030
ishly+0.028
bled+0.026
 afloat+0.026
Neg logit contributions
clock-0.024
ups-0.023
 grows-0.022
rays-0.022
up-0.022
Token!
Token ID0
Feature activation-7.91e-03

Pos logit contributions
stack+0.002
rily+0.002
ishly+0.002
 afloat+0.001
bled+0.001
Neg logit contributions
clock-0.001
ups-0.001
up-0.001
rays-0.001
 grows-0.001
Token The
Token ID383
Feature activation-2.46e-03

Pos logit contributions
rily+0.016
stack+0.016
ishly+0.015
 afloat+0.015
bled+0.014
Neg logit contributions
clock-0.011
ups-0.010
rays-0.010
 grows-0.010
 cares-0.010
Token rac
Token ID3444
Feature activation-9.31e-02

Pos logit contributions
rily+0.005
stack+0.005
ishly+0.005
 afloat+0.004
 themselves+0.004
Neg logit contributions
clock-0.004
ups-0.003
up-0.003
 cares-0.003
rays-0.003
Tokeners
Token ID364
Feature activation-4.08e-02

Pos logit contributions
rily+0.161
stack+0.147
ishly+0.137
 afloat+0.132
 themselves+0.128
Neg logit contributions
ups-0.167
up-0.167
clock-0.163
 cares-0.156
rays-0.155
Token dashed
Token ID37901
Feature activation-4.52e-02

Pos logit contributions
rily+0.083
stack+0.079
ishly+0.075
 afloat+0.072
bled+0.070
Neg logit contributions
clock-0.059
ups-0.056
up-0.055
 grows-0.055
 cares-0.055
Token around
Token ID1088
Feature activation-1.22e-02

Pos logit contributions
rily+0.085
stack+0.084
ishly+0.077
 afloat+0.074
bled+0.073
Neg logit contributions
clock-0.068
up-0.065
ups-0.064
rays-0.060
 looks-0.060
Token the
Token ID262
Feature activation-1.48e-02

Pos logit contributions
rily+0.027
stack+0.024
ishly+0.024
bled+0.024
 afloat+0.023
Neg logit contributions
clock-0.015
 grows-0.014
 tests-0.014
 mysteries-0.014
 visits-0.014
Token town
Token ID3240
Feature activation-1.39e-02

Pos logit contributions
rily+0.027
stack+0.027
ishly+0.026
 afloat+0.023
 themselves+0.023
Neg logit contributions
clock-0.024
ups-0.022
up-0.022
 cares-0.022
rays-0.022
Token and
Token ID290
Feature activation-1.18e-02

Pos logit contributions
stack+0.065
rily+0.064
ishly+0.060
bled+0.056
 afloat+0.056
Neg logit contributions
clock-0.045
ups-0.042
 cares-0.041
up-0.041
 grows-0.040
Token upset
Token ID9247
Feature activation-2.86e-02

Pos logit contributions
rily+0.025
stack+0.023
 afloat+0.022
ishly+0.021
bled+0.020
Neg logit contributions
clock-0.017
ups-0.016
 cares-0.016
 grows-0.016
 listens-0.016
Token that
Token ID326
Feature activation-1.87e-02

Pos logit contributions
stack+0.064
rily+0.064
ishly+0.058
 afloat+0.058
bled+0.055
Neg logit contributions
clock-0.036
ups-0.034
up-0.033
 cares-0.032
rays-0.032
Token her
Token ID607
Feature activation-2.15e-02

Pos logit contributions
rily+0.044
stack+0.041
ishly+0.040
 afloat+0.038
bled+0.036
Neg logit contributions
clock-0.023
 grows-0.022
ups-0.022
rays-0.021
 dreams-0.021
Token books
Token ID3835
Feature activation-2.58e-02

Pos logit contributions
rily+0.037
stack+0.037
ishly+0.036
 themselves+0.031
 afloat+0.031
Neg logit contributions
clock-0.035
ups-0.033
rays-0.033
 cares-0.032
up-0.032
Tokenhe
Token ID258
Feature activation-9.30e-02

Pos logit contributions
rily+0.048
stack+0.045
ishly+0.043
 afloat+0.041
 themselves+0.039
Neg logit contributions
clock-0.042
 grows-0.041
up-0.041
ups-0.040
 cares-0.040
Tokenlf
Token ID1652
Feature activation-4.33e-02

Pos logit contributions
rily+0.139
stack+0.135
ishly+0.130
 afloat+0.118
 themselves+0.113
Neg logit contributions
clock-0.184
ups-0.174
up-0.172
rays-0.172
 cares-0.171
Token was
Token ID373
Feature activation-4.82e-02

Pos logit contributions
rily+0.088
stack+0.086
ishly+0.082
 afloat+0.079
 themselves+0.075
Neg logit contributions
clock-0.061
ups-0.058
up-0.057
 cares-0.056
 grows-0.056
Token a
Token ID257
Feature activation-1.72e-02

Pos logit contributions
stack+0.103
rily+0.103
ishly+0.100
 afloat+0.088
 themselves+0.087
Neg logit contributions
clock-0.065
ups-0.059
 grows-0.058
up-0.057
 mysteries-0.056
Token mess
Token ID2085
Feature activation-3.68e-02

Pos logit contributions
rily+0.034
stack+0.034
ishly+0.033
 afloat+0.030
 themselves+0.030
Neg logit contributions
clock-0.025
ups-0.023
up-0.023
 cares-0.023
rays-0.023
Token.
Token ID13
Feature activation-1.13e-02

Pos logit contributions
stack+0.087
rily+0.085
ishly+0.081
 afloat+0.078
 themselves+0.073
Neg logit contributions
clock-0.041
up-0.039
ups-0.039
 grows-0.038
rays-0.038
Token books
Token ID3835
Feature activation+6.66e-03

Pos logit contributions
rily+0.028
stack+0.026
ishly+0.026
 themselves+0.025
 afloat+0.024
Neg logit contributions
clock-0.029
ups-0.025
 cares-0.025
 grows-0.025
rays-0.025
Token on
Token ID319
Feature activation-5.16e-02

Pos logit contributions
clock+0.009
up+0.008
ups+0.008
rays+0.008
 grows+0.008
Neg logit contributions
stack-0.014
rily-0.014
ishly-0.013
 afloat-0.012
bled-0.012
Token the
Token ID262
Feature activation-1.16e-02

Pos logit contributions
rily+0.098
ishly+0.097
stack+0.097
 afloat+0.089
bled+0.087
Neg logit contributions
clock-0.078
rays-0.073
ups-0.073
 grows-0.072
 mysteries-0.070
Token new
Token ID649
Feature activation-4.49e-02

Pos logit contributions
rily+0.021
stack+0.021
ishly+0.020
 afloat+0.018
bled+0.018
Neg logit contributions
clock-0.019
ups-0.018
up-0.017
rays-0.017
 cares-0.017
Token books
Token ID3835
Feature activation-1.45e-02

Pos logit contributions
rily+0.099
stack+0.098
ishly+0.095
 themselves+0.087
 afloat+0.086
Neg logit contributions
clock-0.054
ups-0.050
 cares-0.049
up-0.049
 grows-0.049
Tokenhe
Token ID258
Feature activation-9.28e-02

Pos logit contributions
ishly+0.031
rily+0.031
stack+0.031
 afloat+0.030
bled+0.029
Neg logit contributions
clock-0.017
ups-0.016
up-0.015
 grows-0.015
rays-0.015
Tokenlf
Token ID1652
Feature activation-2.87e-02

Pos logit contributions
rily+0.152
ishly+0.152
stack+0.148
 afloat+0.140
bled+0.128
Neg logit contributions
clock-0.167
ups-0.158
rays-0.153
up-0.152
 grows-0.151
Token and
Token ID290
Feature activation-1.75e-02

Pos logit contributions
rily+0.056
stack+0.056
ishly+0.055
 afloat+0.053
bled+0.049
Neg logit contributions
clock-0.042
ups-0.039
up-0.038
rays-0.036
 grows-0.035
Token it
Token ID340
Feature activation-2.74e-02

Pos logit contributions
rily+0.034
stack+0.033
ishly+0.031
 afloat+0.029
bled+0.028
Neg logit contributions
clock-0.028
ups-0.026
rays-0.026
up-0.025
 cares-0.025
Token was
Token ID373
Feature activation-4.05e-02

Pos logit contributions
rily+0.056
stack+0.055
ishly+0.053
 afloat+0.049
 themselves+0.049
Neg logit contributions
clock-0.039
ups-0.036
up-0.035
rays-0.035
 cares-0.035
Token full
Token ID1336
Feature activation-3.79e-05

Pos logit contributions
rily+0.080
stack+0.080
ishly+0.078
 afloat+0.070
bled+0.069
Neg logit contributions
clock-0.061
ups-0.055
up-0.054
 grows-0.054
rays-0.053
Token to
Token ID284
Feature activation-2.93e-03

Pos logit contributions
clock+0.013
ups+0.011
up+0.011
 grows+0.011
rays+0.011
Neg logit contributions
rily-0.020
stack-0.019
ishly-0.018
bled-0.018
 afloat-0.017
Token see
Token ID766
Feature activation+2.74e-02

Pos logit contributions
rily+0.007
bled+0.007
ishly+0.007
stack+0.007
ged+0.006
Neg logit contributions
 study-0.003
 history-0.003
clock-0.003
 grows-0.003
 dreams-0.003
Token the
Token ID262
Feature activation-7.17e-03

Pos logit contributions
clock+0.039
ups+0.038
rays+0.038
 cares+0.038
 grows+0.037
Neg logit contributions
rily-0.053
stack-0.053
ishly-0.050
 afloat-0.048
bled-0.048
Token rare
Token ID4071
Feature activation-3.17e-02

Pos logit contributions
stack+0.015
rily+0.015
ishly+0.014
 afloat+0.013
 themselves+0.013
Neg logit contributions
clock-0.010
ups-0.009
up-0.009
 cares-0.009
rays-0.009
Token books
Token ID3835
Feature activation-1.08e-02

Pos logit contributions
stack+0.062
rily+0.060
ishly+0.057
ter+0.050
 themselves+0.049
Neg logit contributions
ups-0.047
clock-0.046
up-0.041
 cares-0.040
rays-0.040
Tokenhe
Token ID258
Feature activation-9.26e-02

Pos logit contributions
rily+0.024
ishly+0.023
stack+0.023
 afloat+0.022
bled+0.022
Neg logit contributions
clock-0.013
 looks-0.011
 grows-0.011
up-0.011
ups-0.011
Tokenlf
Token ID1652
Feature activation-3.00e-02

Pos logit contributions
ishly+0.181
rily+0.172
 afloat+0.171
stack+0.164
bled+0.153
Neg logit contributions
clock-0.134
ups-0.127
 mysteries-0.125
 grows-0.124
rays-0.122
Token.
Token ID13
Feature activation-1.03e-02

Pos logit contributions
stack+0.067
rily+0.066
ishly+0.065
 afloat+0.063
bled+0.062
Neg logit contributions
clock-0.035
ups-0.031
up-0.029
rays-0.028
 grows-0.028
Token They
Token ID1119
Feature activation-3.59e-02

Pos logit contributions
rily+0.023
stack+0.022
ishly+0.021
bled+0.021
 afloat+0.020
Neg logit contributions
clock-0.013
rays-0.012
ups-0.012
 cares-0.011
 grows-0.011
Token looked
Token ID3114
Feature activation+2.07e-02

Pos logit contributions
rily+0.070
 afloat+0.061
stack+0.061
ula+0.059
bled+0.057
Neg logit contributions
 knows-0.054
 grows-0.052
 cares-0.051
 listens-0.051
 looks-0.050
Token at
Token ID379
Feature activation+7.92e-03

Pos logit contributions
clock+0.037
ups+0.035
up+0.034
 cares+0.034
rays+0.034
Neg logit contributions
rily-0.035
stack-0.034
ishly-0.033
 afloat-0.030
 themselves-0.029
Token He
Token ID679
Feature activation+4.92e-03

Pos logit contributions
rily+0.028
stack+0.028
ishly+0.026
 themselves+0.025
 afloat+0.024
Neg logit contributions
clock-0.023
 cares-0.022
ups-0.022
up-0.022
 mysteries-0.021
Token also
Token ID635
Feature activation+2.97e-02

Pos logit contributions
clock+0.007
ups+0.007
up+0.007
 grows+0.007
rays+0.007
Neg logit contributions
rily-0.009
stack-0.009
ishly-0.009
 afloat-0.008
 themselves-0.008
Token picked
Token ID6497
Feature activation+6.49e-03

Pos logit contributions
clock+0.045
ups+0.042
up+0.042
 grows+0.041
 cares+0.041
Neg logit contributions
rily-0.058
stack-0.057
ishly-0.055
 afloat-0.050
 themselves-0.050
Token some
Token ID617
Feature activation-2.18e-02

Pos logit contributions
clock+0.010
up+0.010
ups+0.010
rays+0.010
 grows+0.009
Neg logit contributions
stack-0.012
rily-0.012
ishly-0.011
 afloat-0.010
bled-0.010
Token lettuce
Token ID39406
Feature activation+3.22e-03

Pos logit contributions
rily+0.041
stack+0.039
ishly+0.038
 afloat+0.035
 themselves+0.035
Neg logit contributions
clock-0.035
ups-0.032
up-0.032
 cares-0.032
rays-0.032
Token from
Token ID422
Feature activation-9.20e-02

Pos logit contributions
 grows+0.003
ups+0.003
up+0.003
rays+0.003
clock+0.003
Neg logit contributions
stack-0.007
ishly-0.007
rily-0.007
 stret-0.007
 Ph-0.006
Token the
Token ID262
Feature activation-3.30e-03

Pos logit contributions
ishly+0.199
rily+0.199
stack+0.195
 afloat+0.185
bled+0.179
Neg logit contributions
clock-0.099
rays-0.093
 grows-0.091
 mysteries-0.089
 math-0.088
Token garden
Token ID11376
Feature activation+2.36e-02

Pos logit contributions
rily+0.006
stack+0.006
ishly+0.005
 afloat+0.005
bled+0.005
Neg logit contributions
clock-0.006
ups-0.005
up-0.005
 grows-0.005
rays-0.005
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
clock+0.033
ups+0.031
up+0.030
 grows+0.029
rays+0.029
Neg logit contributions
rily-0.048
stack-0.047
ishly-0.046
 afloat-0.043
bled-0.043
Token
Token ID220
Feature activation+5.87e-02

Pos logit contributions
rily+0.023
stack+0.022
ishly+0.021
 themselves+0.020
 afloat+0.019
Neg logit contributions
clock-0.019
ups-0.018
 cares-0.018
up-0.018
rays-0.017
Token
Token ID198
Feature activation-1.78e-02

Pos logit contributions
clock+0.075
ups+0.071
rays+0.068
up+0.067
 grows+0.065
Neg logit contributions
rily-0.128
stack-0.127
ishly-0.126
 afloat-0.113
bled-0.110
Token Jake
Token ID14757
Feature activation-2.94e-02

Pos logit contributions
rily+0.024
stack+0.024
ishly+0.023
 afloat+0.022
bled+0.021
Neg logit contributions
clock-0.017
ups-0.015
rays-0.015
 grows-0.015
 cares-0.014
Token was
Token ID373
Feature activation-1.80e-02

Pos logit contributions
rily+0.056
stack+0.054
ishly+0.052
 afloat+0.048
 themselves+0.047
Neg logit contributions
clock-0.046
ups-0.043
up-0.042
 cares-0.042
rays-0.042
Token jealous
Token ID19354
Feature activation-2.65e-02

Pos logit contributions
rily+0.033
stack+0.032
ishly+0.030
 afloat+0.027
bled+0.026
Neg logit contributions
clock-0.030
ups-0.028
up-0.027
 grows-0.027
 cares-0.027
Token because
Token ID780
Feature activation-1.18e-02

Pos logit contributions
stack+0.058
rily+0.058
ishly+0.054
 afloat+0.051
bled+0.051
Neg logit contributions
clock-0.034
ups-0.033
up-0.032
 cares-0.031
 grows-0.030
Token his
Token ID465
Feature activation-4.45e-02

Pos logit contributions
rily+0.024
stack+0.024
ishly+0.023
 afloat+0.021
bled+0.021
Neg logit contributions
clock-0.017
ups-0.015
up-0.015
rays-0.015
 cares-0.015
Token k
Token ID479
Feature activation-9.11e-02

Pos logit contributions
rily+0.079
stack+0.074
ishly+0.072
 afloat+0.065
bled+0.065
Neg logit contributions
clock-0.077
 grows-0.070
ups-0.069
 cares-0.069
up-0.069
Tokenite
Token ID578
Feature activation-3.75e-02

Pos logit contributions
rily+0.221
stack+0.220
ishly+0.216
 afloat+0.200
 themselves+0.199
Neg logit contributions
clock-0.094
ups-0.082
up-0.081
rays-0.079
 cares-0.079
Token wasn
Token ID2492
Feature activation-3.82e-02

Pos logit contributions
rily+0.077
stack+0.075
ishly+0.072
 afloat+0.067
 themselves+0.066
Neg logit contributions
clock-0.053
ups-0.049
up-0.048
 cares-0.048
rays-0.048
Token't
Token ID470
Feature activation-4.06e-02

Pos logit contributions
rily+0.081
stack+0.080
ishly+0.076
 afloat+0.069
 themselves+0.069
Neg logit contributions
clock-0.050
ups-0.047
up-0.047
 grows-0.046
rays-0.044
Token flying
Token ID7348
Feature activation-4.21e-03

Pos logit contributions
stack+0.084
ishly+0.082
rily+0.080
bled+0.069
 themselves+0.068
Neg logit contributions
clock-0.058
up-0.053
ups-0.053
 grows-0.052
 mysteries-0.051
Token as
Token ID355
Feature activation-1.80e-02

Pos logit contributions
stack+0.008
rily+0.008
ishly+0.008
bled+0.007
 afloat+0.007
Neg logit contributions
clock-0.006
ups-0.006
 grows-0.006
up-0.006
 mysteries-0.006
Token scary
Token ID14343
Feature activation-2.86e-02

Pos logit contributions
rily+0.024
stack+0.024
ishly+0.023
 themselves+0.021
 afloat+0.021
Neg logit contributions
clock-0.016
up-0.015
ups-0.015
 cares-0.015
rays-0.015
Token that
Token ID326
Feature activation-3.28e-02

Pos logit contributions
rily+0.053
stack+0.052
ishly+0.050
 themselves+0.046
 afloat+0.045
Neg logit contributions
clock-0.045
ups-0.042
up-0.042
 cares-0.042
rays-0.041
Token the
Token ID262
Feature activation-3.34e-02

Pos logit contributions
rily+0.060
stack+0.058
ishly+0.057
 afloat+0.052
bled+0.051
Neg logit contributions
clock-0.054
ups-0.050
up-0.050
rays-0.049
 grows-0.049
Token boss
Token ID6478
Feature activation-1.54e-02

Pos logit contributions
rily+0.069
stack+0.068
ishly+0.065
 afloat+0.060
 themselves+0.060
Neg logit contributions
clock-0.047
ups-0.043
up-0.042
 cares-0.042
rays-0.042
Tokeny
Token ID88
Feature activation-5.45e-02

Pos logit contributions
rily+0.030
stack+0.029
ishly+0.028
 afloat+0.026
 themselves+0.025
Neg logit contributions
clock-0.024
ups-0.023
up-0.022
rays-0.022
 cares-0.022
Token k
Token ID479
Feature activation-9.11e-02

Pos logit contributions
rily+0.104
stack+0.102
ishly+0.098
 afloat+0.090
 themselves+0.089
Neg logit contributions
clock-0.085
ups-0.079
up-0.078
 cares-0.077
rays-0.077
Tokenitty
Token ID9760
Feature activation-5.21e-02

Pos logit contributions
stack+0.169
ishly+0.162
rily+0.158
ter+0.141
bled+0.140
Neg logit contributions
clock-0.147
 listens-0.136
 cares-0.135
 dreams-0.131
 mysteries-0.131
Token ran
Token ID4966
Feature activation-5.49e-02

Pos logit contributions
rily+0.105
stack+0.103
ishly+0.099
 afloat+0.091
 themselves+0.090
Neg logit contributions
clock-0.076
ups-0.070
up-0.069
 cares-0.068
rays-0.068
Token away
Token ID1497
Feature activation-4.83e-02

Pos logit contributions
rily+0.097
stack+0.097
ishly+0.089
 afloat+0.082
bled+0.082
Neg logit contributions
clock-0.094
up-0.089
ups-0.089
rays-0.083
 grows-0.082
Token and
Token ID290
Feature activation-1.38e-02

Pos logit contributions
stack+0.109
rily+0.109
ishly+0.106
bled+0.099
 afloat+0.095
Neg logit contributions
clock-0.056
ups-0.052
 grows-0.049
up-0.049
rays-0.047
Token hid
Token ID24519
Feature activation-5.08e-02

Pos logit contributions
rily+0.027
stack+0.024
ishly+0.024
 afloat+0.022
bled+0.022
Neg logit contributions
clock-0.022
ups-0.021
 grows-0.020
rays-0.020
 cares-0.020
Token a
Token ID257
Feature activation-9.57e-03

Pos logit contributions
rily+0.060
stack+0.059
ishly+0.058
 afloat+0.052
 themselves+0.051
Neg logit contributions
clock-0.037
 grows-0.034
ups-0.033
rays-0.033
up-0.032
Token fle
Token ID5104
Feature activation-7.64e-02

Pos logit contributions
rily+0.020
stack+0.020
ishly+0.019
 themselves+0.018
bled+0.018
Neg logit contributions
clock-0.013
ups-0.012
up-0.012
 grows-0.011
 cares-0.011
Tokena
Token ID64
Feature activation-1.43e-02

Pos logit contributions
rily+0.160
stack+0.157
ishly+0.152
 afloat+0.139
 themselves+0.139
Neg logit contributions
clock-0.105
ups-0.098
up-0.096
 cares-0.095
rays-0.095
Token.
Token ID13
Feature activation-6.72e-03

Pos logit contributions
rily+0.029
stack+0.029
ishly+0.027
bled+0.025
 afloat+0.025
Neg logit contributions
clock-0.020
up-0.019
 cares-0.019
 grows-0.018
ups-0.018
Token The
Token ID383
Feature activation+1.30e-03

Pos logit contributions
rily+0.011
stack+0.010
ishly+0.010
 themselves+0.009
 afloat+0.009
Neg logit contributions
clock-0.013
ups-0.012
up-0.012
 cares-0.012
rays-0.012
Token fle
Token ID5104
Feature activation-9.11e-02

Pos logit contributions
clock+0.002
ups+0.002
up+0.002
 cares+0.002
rays+0.002
Neg logit contributions
rily-0.002
stack-0.002
ishly-0.002
 afloat-0.002
 themselves-0.002
Tokena
Token ID64
Feature activation-3.12e-02

Pos logit contributions
rily+0.180
stack+0.175
ishly+0.168
bled+0.156
 themselves+0.155
Neg logit contributions
clock-0.139
rays-0.124
ups-0.124
 cares-0.121
up-0.121
Token bit
Token ID1643
Feature activation-2.25e-02

Pos logit contributions
rily+0.053
stack+0.051
ishly+0.050
 themselves+0.044
 afloat+0.044
Neg logit contributions
clock-0.055
ups-0.051
rays-0.051
up-0.050
 cares-0.050
Token her
Token ID607
Feature activation-7.25e-03

Pos logit contributions
stack+0.041
rily+0.039
ishly+0.038
 afloat+0.034
bled+0.034
Neg logit contributions
clock-0.037
ups-0.036
up-0.035
 grows-0.035
rays-0.034
Token nose
Token ID9686
Feature activation-1.29e-02

Pos logit contributions
stack+0.013
rily+0.013
ishly+0.012
bled+0.011
 themselves+0.011
Neg logit contributions
clock-0.012
 cares-0.011
 dreams-0.011
up-0.011
rays-0.011
Token and
Token ID290
Feature activation-1.76e-02

Pos logit contributions
stack+0.030
rily+0.029
ishly+0.026
 afloat+0.026
 themselves+0.026
Neg logit contributions
clock-0.014
 grows-0.014
rays-0.014
up-0.014
 looks-0.013
Token was
Token ID373
Feature activation-4.05e-02

Pos logit contributions
rily+0.056
stack+0.055
ishly+0.053
 afloat+0.049
 themselves+0.049
Neg logit contributions
clock-0.039
ups-0.036
up-0.035
rays-0.035
 cares-0.035
Token full
Token ID1336
Feature activation-3.79e-05

Pos logit contributions
rily+0.080
stack+0.080
ishly+0.078
 afloat+0.070
bled+0.069
Neg logit contributions
clock-0.061
ups-0.055
up-0.054
 grows-0.054
rays-0.053
Token.
Token ID13
Feature activation-1.71e-02

Pos logit contributions
rily+0.000
stack+0.000
ishly+0.000
 afloat+0.000
 themselves+0.000
Neg logit contributions
clock-0.000
ups-0.000
up-0.000
 cares-0.000
rays-0.000
Token The
Token ID383
Feature activation+2.49e-03

Pos logit contributions
rily+0.034
stack+0.032
ishly+0.032
 afloat+0.031
bled+0.031
Neg logit contributions
clock-0.026
ups-0.023
rays-0.022
 grows-0.022
 cares-0.022
Token books
Token ID3835
Feature activation-2.03e-02

Pos logit contributions
clock+0.003
ups+0.003
up+0.003
 cares+0.003
rays+0.003
Neg logit contributions
stack-0.005
rily-0.005
ishly-0.005
 themselves-0.004
 afloat-0.004
Tokenhe
Token ID258
Feature activation-9.08e-02

Pos logit contributions
rily+0.033
stack+0.031
ishly+0.031
 afloat+0.030
 themselves+0.028
Neg logit contributions
clock-0.036
 grows-0.035
up-0.034
 looks-0.033
ups-0.033
Tokenlf
Token ID1652
Feature activation-3.30e-02

Pos logit contributions
rily+0.140
stack+0.135
ishly+0.133
 afloat+0.122
 themselves+0.114
Neg logit contributions
clock-0.175
ups-0.166
up-0.164
rays-0.163
 cares-0.161
Token looked
Token ID3114
Feature activation-1.56e-03

Pos logit contributions
rily+0.067
stack+0.066
ishly+0.063
 afloat+0.061
 themselves+0.057
Neg logit contributions
clock-0.047
ups-0.044
up-0.044
 cares-0.042
 grows-0.042
Token great
Token ID1049
Feature activation-2.15e-02

Pos logit contributions
rily+0.003
stack+0.003
ishly+0.003
 afloat+0.002
 themselves+0.002
Neg logit contributions
clock-0.003
ups-0.002
rays-0.002
 cares-0.002
up-0.002
Token in
Token ID287
Feature activation-2.43e-02

Pos logit contributions
rily+0.042
stack+0.042
ishly+0.040
 afloat+0.037
 themselves+0.037
Neg logit contributions
clock-0.032
ups-0.030
up-0.029
 cares-0.029
rays-0.029
Token their
Token ID511
Feature activation-3.46e-02

Pos logit contributions
ishly+0.053
rily+0.051
stack+0.051
bled+0.046
 afloat+0.045
Neg logit contributions
clock-0.032
 mysteries-0.030
 history-0.030
 grows-0.030
 visits-0.028
Token accidentally
Token ID14716
Feature activation+5.01e-03

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.010
 themselves+0.009
 afloat+0.009
Neg logit contributions
clock-0.010
ups-0.009
up-0.009
rays-0.009
 cares-0.009
Token knocked
Token ID13642
Feature activation-1.49e-03

Pos logit contributions
clock+0.010
ups+0.009
 cares+0.009
up+0.009
 grows+0.009
Neg logit contributions
rily-0.008
stack-0.007
ishly-0.007
bled-0.006
 themselves-0.006
Token over
Token ID625
Feature activation+7.04e-03

Pos logit contributions
rily+0.003
stack+0.003
ishly+0.003
 afloat+0.002
bled+0.002
Neg logit contributions
clock-0.002
ups-0.002
rays-0.002
up-0.002
 grows-0.002
Token the
Token ID262
Feature activation-3.46e-03

Pos logit contributions
clock+0.010
ups+0.010
up+0.010
rays+0.009
 cares+0.009
Neg logit contributions
rily-0.014
stack-0.014
ishly-0.014
 afloat-0.012
bled-0.012
Token books
Token ID3835
Feature activation-1.69e-03

Pos logit contributions
stack+0.006
rily+0.006
ishly+0.005
 stret+0.005
 towel+0.005
Neg logit contributions
ups-0.005
up-0.005
clock-0.005
 cares-0.005
 grows-0.005
Tokenhe
Token ID258
Feature activation-9.07e-02

Pos logit contributions
rily+0.003
stack+0.002
ishly+0.002
 afloat+0.002
bled+0.002
Neg logit contributions
clock-0.003
rays-0.003
up-0.003
ups-0.003
 grows-0.003
Tokenlf
Token ID1652
Feature activation-2.45e-02

Pos logit contributions
rily+0.144
ishly+0.142
stack+0.141
 afloat+0.127
 themselves+0.121
Neg logit contributions
clock-0.171
ups-0.159
rays-0.159
up-0.157
 grows-0.156
Token.
Token ID13
Feature activation-2.56e-02

Pos logit contributions
stack+0.057
rily+0.057
ishly+0.056
bled+0.055
 afloat+0.055
Neg logit contributions
clock-0.027
ups-0.023
rays-0.022
up-0.022
 looks-0.020
Token
Token ID220
Feature activation+5.06e-02

Pos logit contributions
rily+0.041
stack+0.038
ishly+0.037
 afloat+0.035
bled+0.033
Neg logit contributions
clock-0.049
ups-0.045
rays-0.045
 grows-0.045
up-0.045
Token
Token ID198
Feature activation-2.19e-02

Pos logit contributions
clock+0.055
ups+0.053
rays+0.050
up+0.050
 grows+0.046
Neg logit contributions
stack-0.121
ishly-0.120
rily-0.120
 afloat-0.107
bled-0.103
Token
Token ID198
Feature activation-1.93e-02

Pos logit contributions
rily+0.047
stack+0.046
ishly+0.045
 afloat+0.041
bled+0.041
Neg logit contributions
clock-0.029
ups-0.027
up-0.027
rays-0.026
 grows-0.025
Token peek
Token ID27185
Feature activation+1.70e-02

Pos logit contributions
rily+0.031
stack+0.030
ishly+0.029
 afloat+0.027
 themselves+0.026
Neg logit contributions
clock-0.023
ups-0.022
up-0.022
 grows-0.021
 cares-0.021
Tokened
Token ID276
Feature activation-2.01e-02

Pos logit contributions
clock+0.027
up+0.026
ups+0.026
rays+0.025
 grows+0.025
Neg logit contributions
stack-0.031
rily-0.031
ishly-0.029
 afloat-0.027
 themselves-0.026
Token around
Token ID1088
Feature activation-1.33e-02

Pos logit contributions
rily+0.031
stack+0.030
ishly+0.029
 afloat+0.026
 themselves+0.026
Neg logit contributions
clock-0.038
ups-0.036
up-0.036
rays-0.035
 cares-0.035
Token the
Token ID262
Feature activation-1.50e-02

Pos logit contributions
rily+0.030
stack+0.029
ishly+0.028
bled+0.027
 afloat+0.027
Neg logit contributions
 grows-0.015
clock-0.014
 looks-0.014
 cares-0.014
 dreams-0.014
Token books
Token ID3835
Feature activation-1.74e-02

Pos logit contributions
stack+0.020
rily+0.017
 towel+0.016
 stret+0.015
ishly+0.015
Neg logit contributions
 cares-0.030
clock-0.029
 happens-0.029
ups-0.028
 listens-0.028
Tokenhe
Token ID258
Feature activation-9.07e-02

Pos logit contributions
rily+0.028
 afloat+0.024
ishly+0.024
bled+0.024
stack+0.024
Neg logit contributions
clock-0.030
 grows-0.030
 looks-0.030
up-0.030
rays-0.029
Tokenlf
Token ID1652
Feature activation-3.74e-02

Pos logit contributions
rily+0.157
ishly+0.156
stack+0.150
 afloat+0.146
bled+0.135
Neg logit contributions
clock-0.154
ups-0.144
 grows-0.142
up-0.140
rays-0.140
Token again
Token ID757
Feature activation-6.50e-03

Pos logit contributions
rily+0.086
stack+0.081
ishly+0.079
 afloat+0.079
bled+0.077
Neg logit contributions
clock-0.041
ups-0.038
up-0.038
 grows-0.036
 looks-0.035
Token,
Token ID11
Feature activation-1.75e-02

Pos logit contributions
rily+0.015
stack+0.014
ishly+0.014
bled+0.013
 afloat+0.013
Neg logit contributions
clock-0.008
 grows-0.007
up-0.007
ups-0.007
 cares-0.007
Token she
Token ID673
Feature activation+5.03e-03

Pos logit contributions
stack+0.038
rily+0.037
ishly+0.036
 afloat+0.035
 stret+0.034
Neg logit contributions
 grows-0.019
ups-0.019
clock-0.019
 cares-0.019
 looks-0.018
Token noticed
Token ID6810
Feature activation+1.52e-02

Pos logit contributions
clock+0.008
ups+0.007
rays+0.007
up+0.007
 cares+0.007
Neg logit contributions
rily-0.010
stack-0.009
ishly-0.009
 afloat-0.008
bled-0.008
Token.
Token ID13
Feature activation+3.15e-02

Pos logit contributions
clock+0.006
ups+0.005
 grows+0.005
up+0.005
 cares+0.005
Neg logit contributions
rily-0.010
stack-0.010
ishly-0.010
bled-0.009
 afloat-0.009
Token Do
Token ID2141
Feature activation+4.02e-03

Pos logit contributions
clock+0.043
ups+0.041
up+0.040
 cares+0.039
rays+0.039
Neg logit contributions
rily-0.065
stack-0.064
ishly-0.062
 themselves-0.058
 afloat-0.056
Token you
Token ID345
Feature activation+1.80e-02

Pos logit contributions
clock+0.006
rays+0.006
ups+0.005
up+0.005
 cares+0.005
Neg logit contributions
stack-0.007
 themselves-0.007
ishly-0.007
rily-0.007
bled-0.006
Token want
Token ID765
Feature activation-2.82e-02

Pos logit contributions
clock+0.034
ups+0.034
rays+0.033
up+0.031
 mysteries+0.030
Neg logit contributions
rily-0.026
stack-0.025
ishly-0.023
 afloat-0.023
 themselves-0.023
Token to
Token ID284
Feature activation+3.86e-02

Pos logit contributions
 themselves+0.046
rily+0.044
stack+0.044
 Nem+0.044
 towel+0.041
Neg logit contributions
 cares-0.042
clock-0.042
rays-0.042
ups-0.041
neys-0.039
Token see
Token ID766
Feature activation+5.13e-02

Pos logit contributions
clock+0.064
ups+0.063
rays+0.062
 cares+0.060
up+0.058
Neg logit contributions
stack-0.066
rily-0.066
ishly-0.061
 themselves-0.060
 afloat-0.058
Token it
Token ID340
Feature activation+1.98e-02

Pos logit contributions
clock+0.097
ups+0.091
 cares+0.090
up+0.089
rays+0.087
Neg logit contributions
rily-0.083
stack-0.076
ishly-0.072
 themselves-0.069
bled-0.065
Token?"
Token ID1701
Feature activation+2.18e-02

Pos logit contributions
clock+0.017
ups+0.014
 grows+0.014
up+0.014
rays+0.014
Neg logit contributions
rily-0.052
stack-0.051
ishly-0.050
 afloat-0.047
 themselves-0.046
Token
Token ID198
Feature activation+1.32e-03

Pos logit contributions
clock+0.038
ups+0.036
 cares+0.036
rays+0.035
up+0.035
Neg logit contributions
rily-0.036
stack-0.036
ishly-0.034
 themselves-0.032
 afloat-0.030
Token
Token ID198
Feature activation-4.67e-03

Pos logit contributions
clock+0.002
ups+0.002
 cares+0.002
rays+0.002
up+0.002
Neg logit contributions
rily-0.002
stack-0.002
ishly-0.002
 themselves-0.002
 afloat-0.002
TokenL
Token ID43
Feature activation+4.40e-03

Pos logit contributions
rily+0.007
stack+0.007
 themselves+0.006
ishly+0.006
bled+0.006
Neg logit contributions
clock-0.009
 cares-0.009
rays-0.008
 grows-0.008
 listens-0.008
Token at
Token ID379
Feature activation+2.15e-02

Pos logit contributions
clock+0.048
ups+0.045
up+0.044
 cares+0.044
rays+0.043
Neg logit contributions
rily-0.059
stack-0.058
ishly-0.056
 afloat-0.051
 themselves-0.050
Token them
Token ID606
Feature activation+5.89e-03

Pos logit contributions
 mysteries+0.025
 dreams+0.024
 grows+0.024
clock+0.023
 tests+0.023
Neg logit contributions
rily-0.048
stack-0.046
 afloat-0.046
ishly-0.045
oned-0.041
Token and
Token ID290
Feature activation+1.70e-02

Pos logit contributions
clock+0.006
ups+0.006
 grows+0.005
 listens+0.005
 dreams+0.005
Neg logit contributions
stack-0.014
rily-0.014
ishly-0.013
bled-0.013
 afloat-0.013
Token smiles
Token ID21845
Feature activation+3.31e-02

Pos logit contributions
clock+0.021
ups+0.020
up+0.019
rays+0.019
ray+0.017
Neg logit contributions
stack-0.037
rily-0.037
ishly-0.037
 themselves-0.034
 afloat-0.033
Token.
Token ID13
Feature activation+3.79e-02

Pos logit contributions
clock+0.049
ups+0.045
up+0.045
 cares+0.044
rays+0.044
Neg logit contributions
rily-0.066
stack-0.065
ishly-0.062
 afloat-0.057
 themselves-0.056
Token She
Token ID1375
Feature activation+5.46e-02

Pos logit contributions
clock+0.053
ups+0.049
up+0.049
rays+0.048
 cares+0.048
Neg logit contributions
rily-0.078
stack-0.077
ishly-0.074
 themselves-0.069
 afloat-0.067
Token sees
Token ID7224
Feature activation+2.00e-02

Pos logit contributions
clock+0.079
rays+0.074
ups+0.072
neys+0.067
up+0.066
Neg logit contributions
ishly-0.104
stack-0.097
rily-0.096
 themselves-0.093
 Wally-0.088
Token their
Token ID511
Feature activation-2.55e-02

Pos logit contributions
ups+0.028
up+0.028
clock+0.028
 cares+0.027
 grows+0.026
Neg logit contributions
stack-0.042
rily-0.041
ishly-0.038
 afloat-0.037
bled-0.034
Token tent
Token ID11105
Feature activation-2.10e-02

Pos logit contributions
rily+0.053
stack+0.051
ishly+0.049
bled+0.046
 afloat+0.046
Neg logit contributions
clock-0.037
ups-0.034
 cares-0.034
up-0.033
 grows-0.033
Token and
Token ID290
Feature activation+1.46e-02

Pos logit contributions
rily+0.042
stack+0.041
ishly+0.040
 afloat+0.038
bled+0.037
Neg logit contributions
clock-0.030
 grows-0.028
ups-0.028
up-0.027
 cares-0.027
Token says
Token ID1139
Feature activation+8.74e-03

Pos logit contributions
clock+0.021
ups+0.020
up+0.019
rays+0.019
 cares+0.019
Neg logit contributions
rily-0.029
stack-0.029
ishly-0.028
 themselves-0.026
 afloat-0.025
Token again
Token ID757
Feature activation+1.75e-02

Pos logit contributions
stack+0.058
rily+0.055
ishly+0.052
bled+0.050
 afloat+0.050
Neg logit contributions
clock-0.027
ups-0.024
rays-0.023
up-0.023
 grows-0.022
Token.
Token ID13
Feature activation-6.07e-03

Pos logit contributions
clock+0.021
ups+0.018
 grows+0.018
 cares+0.017
 visits+0.017
Neg logit contributions
rily-0.041
stack-0.040
ishly-0.039
bled-0.038
 afloat-0.035
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.009
 themselves+0.008
 afloat+0.008
Neg logit contributions
clock-0.011
ups-0.011
up-0.010
rays-0.010
 cares-0.010
Token
Token ID198
Feature activation-1.12e-02

Pos logit contributions
rily+0.024
stack+0.022
ishly+0.022
 afloat+0.020
ter+0.019
Neg logit contributions
clock-0.013
ups-0.013
up-0.013
rays-0.012
 grows-0.012
TokenM
Token ID44
Feature activation+3.90e-02

Pos logit contributions
rily+0.023
stack+0.022
ishly+0.021
 afloat+0.020
 themselves+0.020
Neg logit contributions
clock-0.016
ups-0.015
up-0.015
 cares-0.015
rays-0.015
Tokenum
Token ID388
Feature activation+5.79e-02

Pos logit contributions
rays+0.061
clock+0.060
 mysteries+0.058
 cares+0.057
 grows+0.055
Neg logit contributions
rily-0.069
stack-0.067
ishly-0.067
bled-0.064
 themselves-0.060
Token said
Token ID531
Feature activation-4.11e-03

Pos logit contributions
clock+0.083
up+0.083
ups+0.083
rays+0.078
 cares+0.078
Neg logit contributions
rily-0.115
stack-0.112
ishly-0.103
 afloat-0.103
 themselves-0.099
Token,
Token ID11
Feature activation-3.62e-02

Pos logit contributions
rily+0.009
stack+0.009
 afloat+0.009
ishly+0.009
 themselves+0.008
Neg logit contributions
clock-0.005
ups-0.004
up-0.004
 grows-0.004
 cares-0.004
Token "
Token ID366
Feature activation-7.11e-03

Pos logit contributions
rily+0.077
 afloat+0.073
stack+0.073
ishly+0.070
 themselves+0.067
Neg logit contributions
clock-0.046
rays-0.043
 grows-0.041
 listens-0.041
ups-0.040
TokenThe
Token ID464
Feature activation+1.35e-02

Pos logit contributions
rily+0.013
stack+0.013
ishly+0.012
 themselves+0.011
 afloat+0.011
Neg logit contributions
clock-0.011
ups-0.010
up-0.010
 cares-0.010
 grows-0.010
Token earthquake
Token ID16295
Feature activation-6.16e-03

Pos logit contributions
clock+0.020
ups+0.019
up+0.019
 cares+0.018
rays+0.018
Neg logit contributions
rily-0.027
stack-0.026
ishly-0.025
 afloat-0.023
 themselves-0.023
Token it
Token ID340
Feature activation-1.23e-02

Pos logit contributions
rily+0.025
stack+0.023
ishly+0.023
 afloat+0.021
bled+0.021
Neg logit contributions
clock-0.017
 grows-0.016
rays-0.016
 cares-0.016
 listens-0.015
Token sounded
Token ID14846
Feature activation-2.08e-02

Pos logit contributions
rily+0.028
stack+0.027
ishly+0.026
 afloat+0.024
 themselves+0.024
Neg logit contributions
clock-0.015
ups-0.014
up-0.014
 grows-0.014
rays-0.014
Token like
Token ID588
Feature activation-1.78e-02

Pos logit contributions
rily+0.042
stack+0.042
ishly+0.041
 afloat+0.037
bled+0.036
Neg logit contributions
clock-0.030
ups-0.027
up-0.027
 cares-0.027
 grows-0.026
Token singing
Token ID13777
Feature activation+1.13e-03

Pos logit contributions
rily+0.037
stack+0.036
ishly+0.035
 afloat+0.032
bled+0.032
Neg logit contributions
clock-0.025
ups-0.023
up-0.022
 cares-0.022
 grows-0.022
Token.
Token ID13
Feature activation-2.07e-03

Pos logit contributions
clock+0.002
ups+0.002
up+0.002
rays+0.002
 cares+0.002
Neg logit contributions
rily-0.002
stack-0.002
ishly-0.002
 afloat-0.002
bled-0.002
Token
Token ID220
Feature activation+6.12e-02

Pos logit contributions
rily+0.004
stack+0.004
 afloat+0.004
ishly+0.004
bled+0.004
Neg logit contributions
clock-0.003
ups-0.003
 cares-0.003
up-0.003
 looks-0.003
Token
Token ID198
Feature activation-4.62e-03

Pos logit contributions
clock+0.079
ups+0.074
rays+0.071
up+0.070
 grows+0.069
Neg logit contributions
rily-0.134
stack-0.131
ishly-0.129
 afloat-0.118
bled-0.116
Token
Token ID198
Feature activation-2.87e-03

Pos logit contributions
rily+0.011
stack+0.011
ishly+0.011
 afloat+0.010
bled+0.010
Neg logit contributions
clock-0.005
ups-0.005
up-0.005
rays-0.004
 cares-0.004
TokenThey
Token ID2990
Feature activation-3.63e-02

Pos logit contributions
rily+0.007
stack+0.006
ishly+0.006
 afloat+0.006
bled+0.006
Neg logit contributions
ups-0.004
clock-0.004
up-0.004
 cares-0.003
rays-0.003
Token slowly
Token ID6364
Feature activation+3.10e-02

Pos logit contributions
rily+0.081
stack+0.078
ishly+0.072
 afloat+0.070
bled+0.068
Neg logit contributions
clock-0.045
up-0.044
ups-0.043
 grows-0.042
 cares-0.042
Token reached
Token ID4251
Feature activation+2.72e-02

Pos logit contributions
clock+0.047
ups+0.044
up+0.043
 cares+0.043
rays+0.043
Neg logit contributions
rily-0.060
stack-0.059
ishly-0.057
 afloat-0.052
 themselves-0.051
Token,
Token ID11
Feature activation+8.31e-03

Pos logit contributions
clock+0.007
up+0.007
ups+0.007
 cares+0.006
rays+0.006
Neg logit contributions
stack-0.011
rily-0.011
ishly-0.010
 afloat-0.010
 themselves-0.009
Token can
Token ID460
Feature activation-4.71e-03

Pos logit contributions
clock+0.013
up+0.012
ups+0.012
 cares+0.012
 grows+0.012
Neg logit contributions
rily-0.016
stack-0.015
ishly-0.015
 afloat-0.014
bled-0.013
Token you
Token ID345
Feature activation+1.82e-02

Pos logit contributions
rily+0.009
stack+0.009
ishly+0.009
 themselves+0.008
 afloat+0.008
Neg logit contributions
clock-0.007
ups-0.006
up-0.006
rays-0.006
 cares-0.006
Token see
Token ID766
Feature activation+4.95e-02

Pos logit contributions
clock+0.029
ups+0.027
 cares+0.026
rays+0.026
up+0.026
Neg logit contributions
stack-0.033
rily-0.033
ishly-0.032
 afloat-0.030
 themselves-0.029
Token the
Token ID262
Feature activation+1.51e-03

Pos logit contributions
clock+0.083
ups+0.079
up+0.078
 cares+0.078
rays+0.076
Neg logit contributions
rily-0.088
stack-0.085
ishly-0.081
 themselves-0.075
 afloat-0.074
Token sun
Token ID4252
Feature activation+4.92e-02

Pos logit contributions
ups+0.002
up+0.002
clock+0.002
 cares+0.002
rays+0.002
Neg logit contributions
stack-0.003
rily-0.003
ishly-0.003
 afloat-0.003
 themselves-0.003
Token up
Token ID510
Feature activation-3.03e-02

Pos logit contributions
clock+0.075
ups+0.068
up+0.067
 grows+0.067
rays+0.067
Neg logit contributions
rily-0.096
stack-0.094
ishly-0.091
 afloat-0.084
 themselves-0.083
Token there
Token ID612
Feature activation-2.04e-02

Pos logit contributions
rily+0.074
stack+0.073
ishly+0.071
 afloat+0.066
 themselves+0.065
Neg logit contributions
clock-0.031
ups-0.028
up-0.027
 cares-0.027
rays-0.027
Token?
Token ID30
Feature activation+1.79e-02

Pos logit contributions
rily+0.054
stack+0.053
ishly+0.052
 afloat+0.048
bled+0.048
Neg logit contributions
clock-0.017
 grows-0.014
ups-0.014
up-0.014
rays-0.014
Token It
Token ID632
Feature activation+9.19e-04

Pos logit contributions
ups+0.026
clock+0.026
up+0.025
 cares+0.024
 mysteries+0.023
Neg logit contributions
stack-0.035
rily-0.034
ishly-0.033
 themselves-0.030
 afloat-0.029
Token looks
Token ID3073
Feature activation+1.52e-02

Pos logit contributions
clock+0.001
ups+0.001
up+0.001
 exams+0.001
rays+0.001
Neg logit contributions
stack-0.002
rily-0.002
ishly-0.002
 themselves-0.002
 afloat-0.002
Token
Token ID198
Feature activation-9.75e-03

Pos logit contributions
clock+0.012
ups+0.011
up+0.011
 cares+0.011
rays+0.011
Neg logit contributions
rily-0.017
stack-0.016
ishly-0.016
 afloat-0.014
 themselves-0.014
Token
Token ID198
Feature activation-1.15e-02

Pos logit contributions
rily+0.020
stack+0.020
ishly+0.019
 afloat+0.017
bled+0.017
Neg logit contributions
clock-0.014
ups-0.013
up-0.013
rays-0.013
 grows-0.013
Token"
Token ID1
Feature activation+3.70e-03

Pos logit contributions
rily+0.023
stack+0.022
ishly+0.021
 themselves+0.021
 afloat+0.021
Neg logit contributions
clock-0.016
rays-0.015
 cares-0.015
 listens-0.014
 grows-0.014
TokenYes
Token ID5297
Feature activation-1.29e-02

Pos logit contributions
clock+0.006
ups+0.006
up+0.005
 cares+0.005
rays+0.005
Neg logit contributions
rily-0.007
stack-0.007
ishly-0.006
 themselves-0.006
 afloat-0.006
Token,
Token ID11
Feature activation+1.19e-02

Pos logit contributions
stack+0.031
rily+0.031
ishly+0.029
 afloat+0.029
bled+0.027
Neg logit contributions
clock-0.012
up-0.012
ups-0.011
 grows-0.011
 cares-0.011
Token please
Token ID3387
Feature activation+1.07e-02

Pos logit contributions
clock+0.014
up+0.014
 cares+0.014
 grows+0.013
ups+0.013
Neg logit contributions
stack-0.027
rily-0.027
ishly-0.026
 afloat-0.024
bled-0.023
Token!"
Token ID2474
Feature activation+8.84e-03

Pos logit contributions
clock+0.018
ups+0.017
rays+0.017
up+0.016
 cares+0.016
Neg logit contributions
stack-0.019
rily-0.018
ishly-0.018
 themselves-0.016
 afloat-0.016
Token Lily
Token ID20037
Feature activation+3.62e-03

Pos logit contributions
clock+0.018
ups+0.017
up+0.017
 cares+0.017
 grows+0.016
Neg logit contributions
rily-0.013
stack-0.013
ishly-0.012
 afloat-0.011
bled-0.011
Token and
Token ID290
Feature activation-2.08e-02

Pos logit contributions
clock+0.006
ups+0.006
up+0.005
 cares+0.005
rays+0.005
Neg logit contributions
rily-0.007
stack-0.007
ishly-0.006
 afloat-0.006
bled-0.006
Token Ben
Token ID3932
Feature activation-1.79e-02

Pos logit contributions
rily+0.039
stack+0.037
ishly+0.034
 afloat+0.033
bled+0.032
Neg logit contributions
clock-0.035
 cares-0.033
up-0.032
 grows-0.032
ups-0.032
Token said
Token ID531
Feature activation+5.81e-03

Pos logit contributions
rily+0.040
stack+0.039
bled+0.035
 afloat+0.034
ishly+0.033
Neg logit contributions
clock-0.025
up-0.024
 cares-0.023
 listens-0.022
ups-0.022
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
clock+0.076
ups+0.071
up+0.069
rays+0.069
 grows+0.067
Neg logit contributions
rily-0.120
stack-0.117
ishly-0.115
 afloat-0.105
bled-0.102
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
rily+0.035
stack+0.034
ishly+0.033
 afloat+0.030
 themselves+0.030
Neg logit contributions
clock-0.027
ups-0.025
up-0.025
rays-0.024
 cares-0.024
TokenThe
Token ID464
Feature activation+3.04e-03

Pos logit contributions
rily+0.032
stack+0.031
ishly+0.029
 themselves+0.028
bled+0.027
Neg logit contributions
clock-0.030
 cares-0.028
rays-0.028
ups-0.027
 grows-0.027
Token girl
Token ID2576
Feature activation-1.37e-02

Pos logit contributions
clock+0.005
ups+0.005
up+0.005
 cares+0.005
rays+0.005
Neg logit contributions
rily-0.006
stack-0.005
ishly-0.005
 afloat-0.005
 themselves-0.005
Token was
Token ID373
Feature activation-1.68e-02

Pos logit contributions
stack+0.023
rily+0.022
ishly+0.022
 themselves+0.019
 afloat+0.019
Neg logit contributions
clock-0.024
ups-0.022
rays-0.022
 cares-0.021
up-0.021
Token so
Token ID523
Feature activation+6.74e-03

Pos logit contributions
rily+0.032
stack+0.031
ishly+0.029
bled+0.026
 themselves+0.026
Neg logit contributions
clock-0.027
ups-0.025
up-0.024
 grows-0.024
 cares-0.024
Token happy
Token ID3772
Feature activation-5.66e-04

Pos logit contributions
clock+0.013
ups+0.012
up+0.012
 cares+0.012
 grows+0.012
Neg logit contributions
rily-0.011
stack-0.010
ishly-0.010
 afloat-0.009
 themselves-0.009
Token to
Token ID284
Feature activation+6.41e-03

Pos logit contributions
rily+0.001
stack+0.001
ishly+0.001
 themselves+0.001
 afloat+0.001
Neg logit contributions
clock-0.001
ups-0.001
up-0.001
rays-0.001
 cares-0.001
Token be
Token ID307
Feature activation+2.00e-03

Pos logit contributions
clock+0.009
ups+0.008
up+0.008
 cares+0.008
rays+0.008
Neg logit contributions
rily-0.014
stack-0.014
ishly-0.013
 afloat-0.012
 themselves-0.012
Token back
Token ID736
Feature activation+2.20e-03

Pos logit contributions
clock+0.004
ups+0.004
up+0.004
rays+0.004
 cares+0.003
Neg logit contributions
rily-0.003
stack-0.003
ishly-0.003
 afloat-0.003
 themselves-0.003
Token in
Token ID287
Feature activation-7.08e-03

Pos logit contributions
clock+0.003
ups+0.003
up+0.003
 cares+0.003
rays+0.003
Neg logit contributions
rily-0.004
stack-0.004
ishly-0.004
 afloat-0.004
 themselves-0.004
Token does
Token ID857
Feature activation+8.71e-03

Pos logit contributions
rily+0.012
 themselves+0.011
stack+0.010
tered+0.010
kin+0.009
Neg logit contributions
rays-0.012
clock-0.012
ups-0.011
up-0.011
 grows-0.010
Token that
Token ID326
Feature activation+2.34e-02

Pos logit contributions
clock+0.016
ups+0.015
 cares+0.014
up+0.014
rays+0.014
Neg logit contributions
rily-0.015
stack-0.014
ishly-0.014
 themselves-0.012
 afloat-0.012
Token sound
Token ID2128
Feature activation-4.20e-03

Pos logit contributions
clock+0.040
ups+0.038
up+0.037
 cares+0.037
rays+0.037
Neg logit contributions
rily-0.040
stack-0.039
ishly-0.037
 afloat-0.034
 themselves-0.034
Token?"
Token ID1701
Feature activation+2.36e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.010
 afloat+0.009
 themselves+0.009
Neg logit contributions
clock-0.004
ups-0.004
up-0.004
 cares-0.004
rays-0.004
Token
Token ID198
Feature activation+5.54e-04

Pos logit contributions
clock+0.033
ups+0.029
 grows+0.029
rays+0.029
 cares+0.029
Neg logit contributions
rily-0.050
stack-0.049
ishly-0.048
 afloat-0.044
bled-0.044
Token
Token ID198
Feature activation-4.72e-03

Pos logit contributions
clock+0.001
ups+0.001
up+0.001
 grows+0.001
rays+0.001
Neg logit contributions
rily-0.001
ishly-0.001
stack-0.001
 afloat-0.001
bled-0.001
TokenS
Token ID50
Feature activation+2.24e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.009
 afloat+0.009
bled+0.008
Neg logit contributions
clock-0.007
ups-0.006
up-0.006
 cares-0.006
rays-0.006
Tokenara
Token ID3301
Feature activation+1.90e-02

Pos logit contributions
 mysteries+0.036
clock+0.035
 tests+0.034
 exams+0.034
 cares+0.033
Neg logit contributions
rily-0.040
stack-0.035
ishly-0.034
bled-0.032
 themselves-0.031
Token and
Token ID290
Feature activation-2.06e-02

Pos logit contributions
clock+0.031
ups+0.029
up+0.028
 cares+0.028
rays+0.028
Neg logit contributions
rily-0.035
stack-0.035
ishly-0.033
 afloat-0.031
 themselves-0.030
Token Ben
Token ID3932
Feature activation-2.26e-02

Pos logit contributions
rily+0.039
stack+0.037
ishly+0.034
 afloat+0.034
bled+0.032
Neg logit contributions
clock-0.034
rays-0.032
 cares-0.031
 listens-0.031
 grows-0.031
Token smiled
Token ID13541
Feature activation+2.11e-02

Pos logit contributions
rily+0.055
stack+0.053
 afloat+0.049
ula+0.047
bled+0.047
Neg logit contributions
 listens-0.025
 cares-0.025
 looks-0.024
up-0.023
 knows-0.023
Token water
Token ID1660
Feature activation+3.77e-03

Pos logit contributions
rily+0.042
stack+0.042
ishly+0.040
 afloat+0.036
bled+0.036
Neg logit contributions
clock-0.036
ups-0.034
up-0.033
 cares-0.033
rays-0.033
Token from
Token ID422
Feature activation-5.42e-02

Pos logit contributions
ups+0.004
clock+0.004
 grows+0.004
up+0.004
 looks+0.004
Neg logit contributions
stack-0.008
rily-0.008
ishly-0.008
 afloat-0.007
bled-0.007
Token the
Token ID262
Feature activation+4.86e-03

Pos logit contributions
rily+0.117
ishly+0.111
stack+0.111
 afloat+0.103
bled+0.102
Neg logit contributions
clock-0.067
 grows-0.064
 dreams-0.062
ups-0.060
 mysteries-0.060
Token tap
Token ID9814
Feature activation+1.78e-02

Pos logit contributions
clock+0.008
ups+0.007
up+0.007
 grows+0.007
rays+0.007
Neg logit contributions
rily-0.009
stack-0.008
ishly-0.008
bled-0.008
 themselves-0.008
Token.
Token ID13
Feature activation-3.80e-03

Pos logit contributions
clock+0.023
ups+0.022
up+0.021
 grows+0.021
rays+0.020
Neg logit contributions
rily-0.037
stack-0.037
ishly-0.035
bled-0.034
 afloat-0.034
Token They
Token ID1119
Feature activation-5.79e-03

Pos logit contributions
rily+0.008
stack+0.008
ishly+0.007
 afloat+0.007
bled+0.007
Neg logit contributions
clock-0.005
 grows-0.005
rays-0.005
ups-0.005
 cares-0.005
Token felt
Token ID2936
Feature activation+1.72e-02

Pos logit contributions
rily+0.012
stack+0.011
bled+0.010
ishly+0.010
 afloat+0.010
Neg logit contributions
 cares-0.008
 grows-0.008
 knows-0.008
 listens-0.008
up-0.008
Token better
Token ID1365
Feature activation+1.49e-02

Pos logit contributions
clock+0.024
 grows+0.024
up+0.024
ups+0.023
 cares+0.023
Neg logit contributions
stack-0.035
rily-0.034
ishly-0.032
bled-0.028
 afloat-0.028
Token,
Token ID11
Feature activation+1.92e-02

Pos logit contributions
clock+0.017
ups+0.016
 cares+0.015
 grows+0.014
 listens+0.014
Neg logit contributions
rily-0.035
stack-0.035
ishly-0.034
 trigger-0.031
bled-0.031
Token but
Token ID475
Feature activation-1.31e-02

Pos logit contributions
clock+0.028
 dreams+0.026
 cares+0.025
ups+0.025
 looks+0.025
Neg logit contributions
stack-0.036
rily-0.036
ishly-0.034
bled-0.033
 trigger-0.031
Token their
Token ID511
Feature activation-2.95e-02

Pos logit contributions
rily+0.027
 stret+0.024
ishly+0.023
 trigger+0.023
oned+0.022
Neg logit contributions
 grows-0.020
clock-0.019
 listens-0.018
 dreams-0.018
 cares-0.018
Token needed
Token ID2622
Feature activation-1.07e-02

Pos logit contributions
rily+0.002
stack+0.002
ishly+0.001
 afloat+0.001
bled+0.001
Neg logit contributions
clock-0.001
ups-0.001
up-0.001
 grows-0.001
 cares-0.001
Token a
Token ID257
Feature activation+7.69e-03

Pos logit contributions
rily+0.023
stack+0.023
ishly+0.022
 themselves+0.021
 afloat+0.020
Neg logit contributions
clock-0.014
ups-0.013
 cares-0.013
up-0.013
rays-0.013
Token break
Token ID2270
Feature activation+2.35e-03

Pos logit contributions
clock+0.013
ups+0.013
up+0.013
 cares+0.012
rays+0.012
Neg logit contributions
rily-0.013
stack-0.013
ishly-0.012
 afloat-0.011
 themselves-0.011
Token.
Token ID13
Feature activation-2.05e-02

Pos logit contributions
clock+0.003
ups+0.003
up+0.003
rays+0.003
 cares+0.003
Neg logit contributions
rily-0.005
stack-0.005
ishly-0.005
bled-0.004
 afloat-0.004
Token He
Token ID679
Feature activation-8.18e-04

Pos logit contributions
rily+0.043
stack+0.042
ishly+0.041
 themselves+0.037
 afloat+0.037
Neg logit contributions
clock-0.027
ups-0.026
 cares-0.026
up-0.026
 grows-0.025
Token found
Token ID1043
Feature activation+3.49e-03

Pos logit contributions
rily+0.002
stack+0.002
ishly+0.001
 afloat+0.001
 themselves+0.001
Neg logit contributions
clock-0.001
ups-0.001
up-0.001
 cares-0.001
rays-0.001
Token a
Token ID257
Feature activation-1.33e-02

Pos logit contributions
clock+0.004
up+0.004
rays+0.004
ups+0.004
 cares+0.004
Neg logit contributions
rily-0.008
stack-0.008
ishly-0.007
 afloat-0.007
bled-0.007
Token big
Token ID1263
Feature activation+5.50e-03

Pos logit contributions
rily+0.027
stack+0.027
ishly+0.026
 themselves+0.024
bled+0.024
Neg logit contributions
clock-0.019
ups-0.017
up-0.017
 grows-0.017
 cares-0.017
Token rock
Token ID3881
Feature activation+3.44e-03

Pos logit contributions
clock+0.009
 grows+0.008
 cares+0.008
up+0.008
ups+0.008
Neg logit contributions
rily-0.010
stack-0.010
ishly-0.010
 themselves-0.009
bled-0.009
Token to
Token ID284
Feature activation-1.63e-02

Pos logit contributions
clock+0.004
up+0.004
 grows+0.004
 looks+0.004
ups+0.004
Neg logit contributions
rily-0.007
stack-0.007
ishly-0.007
 themselves-0.007
bled-0.007
Token sit
Token ID1650
Feature activation-4.11e-03

Pos logit contributions
rily+0.034
ishly+0.032
stack+0.031
bled+0.028
 themselves+0.028
Neg logit contributions
clock-0.022
up-0.022
 grows-0.022
 cares-0.021
rays-0.021
Token had
Token ID550
Feature activation-1.40e-02

Pos logit contributions
rily+0.010
stack+0.010
ishly+0.009
 afloat+0.009
 themselves+0.009
Neg logit contributions
clock-0.007
ups-0.007
up-0.007
 cares-0.007
rays-0.007
Token a
Token ID257
Feature activation-2.50e-03

Pos logit contributions
rily+0.023
stack+0.023
 themselves+0.022
ishly+0.021
bled+0.019
Neg logit contributions
clock-0.022
ups-0.022
rays-0.022
up-0.022
 cares-0.021
Token basket
Token ID7988
Feature activation+1.24e-02

Pos logit contributions
rily+0.005
stack+0.005
ishly+0.005
 afloat+0.005
 themselves+0.005
Neg logit contributions
clock-0.003
ups-0.003
up-0.003
 cares-0.003
rays-0.003
Token with
Token ID351
Feature activation-4.42e-02

Pos logit contributions
clock+0.018
 grows+0.017
up+0.017
ups+0.017
rays+0.016
Neg logit contributions
rily-0.024
stack-0.024
ishly-0.023
 themselves-0.021
 afloat-0.021
Token a
Token ID257
Feature activation-1.08e-02

Pos logit contributions
rily+0.090
stack+0.087
ishly+0.084
 themselves+0.078
 afloat+0.077
Neg logit contributions
clock-0.063
ups-0.059
up-0.058
 cares-0.058
rays-0.057
Token picnic
Token ID35715
Feature activation-3.12e-02

Pos logit contributions
rily+0.022
stack+0.021
ishly+0.021
 themselves+0.019
bled+0.019
Neg logit contributions
clock-0.016
ups-0.014
up-0.014
 grows-0.014
 cares-0.014
Token inside
Token ID2641
Feature activation-1.88e-02

Pos logit contributions
ishly+0.061
rily+0.059
stack+0.056
bled+0.054
 themselves+0.053
Neg logit contributions
clock-0.043
 grows-0.041
 looks-0.039
 dreams-0.039
 mysteries-0.039
Token.
Token ID13
Feature activation+1.46e-02

Pos logit contributions
rily+0.038
stack+0.037
ishly+0.036
 afloat+0.033
bled+0.032
Neg logit contributions
clock-0.027
ups-0.025
 grows-0.025
rays-0.024
up-0.024
Token He
Token ID679
Feature activation+1.57e-02

Pos logit contributions
clock+0.019
rays+0.017
ups+0.017
 cares+0.017
 grows+0.017
Neg logit contributions
rily-0.032
stack-0.031
ishly-0.031
bled-0.028
 afloat-0.028
Token saw
Token ID2497
Feature activation+1.24e-02

Pos logit contributions
clock+0.026
ups+0.024
up+0.023
rays+0.023
 cares+0.023
Neg logit contributions
rily-0.029
stack-0.028
ishly-0.027
 afloat-0.025
 themselves-0.025
Token his
Token ID465
Feature activation-2.45e-02

Pos logit contributions
clock+0.017
ups+0.016
up+0.016
rays+0.015
 grows+0.015
Neg logit contributions
rily-0.026
stack-0.026
ishly-0.025
 afloat-0.023
bled-0.022
Token p
Token ID279
Feature activation+2.07e-02

Pos logit contributions
clock+0.021
ups+0.019
up+0.019
 cares+0.019
 grows+0.019
Neg logit contributions
rily-0.025
stack-0.025
ishly-0.024
bled-0.022
 themselves-0.022
Tokenuddle
Token ID24500
Feature activation+1.20e-02

Pos logit contributions
clock+0.036
rays+0.035
 mysteries+0.033
 cares+0.033
 grows+0.033
Neg logit contributions
rily-0.035
ishly-0.033
stack-0.033
 afloat-0.030
bled-0.028
Token!
Token ID0
Feature activation+6.14e-03

Pos logit contributions
clock+0.013
ups+0.012
up+0.012
 grows+0.011
 looks+0.011
Neg logit contributions
rily-0.027
stack-0.025
ishly-0.025
bled-0.024
 afloat-0.024
Token Sam
Token ID3409
Feature activation-2.91e-02

Pos logit contributions
clock+0.010
ups+0.009
up+0.009
 cares+0.009
rays+0.009
Neg logit contributions
stack-0.011
rily-0.011
ishly-0.011
 themselves-0.010
bled-0.010
Token was
Token ID373
Feature activation-1.53e-02

Pos logit contributions
rily+0.061
stack+0.060
ishly+0.055
 afloat+0.053
 themselves+0.052
Neg logit contributions
clock-0.040
ups-0.039
up-0.039
 cares-0.037
 grows-0.037
Token scared
Token ID12008
Feature activation-2.25e-02

Pos logit contributions
rily+0.034
stack+0.034
ishly+0.031
 trigger+0.028
bled+0.028
Neg logit contributions
clock-0.020
ups-0.018
up-0.017
 tests-0.016
 grows-0.016
Token and
Token ID290
Feature activation-1.34e-02

Pos logit contributions
stack+0.051
rily+0.049
ishly+0.046
bled+0.045
 afloat+0.042
Neg logit contributions
clock-0.028
ups-0.027
up-0.026
 cares-0.025
rays-0.024
Token ran
Token ID4966
Feature activation-3.23e-02

Pos logit contributions
rily+0.028
stack+0.026
ishly+0.025
 afloat+0.024
ula+0.023
Neg logit contributions
clock-0.018
 grows-0.017
ups-0.017
 cares-0.017
 looks-0.017
Token back
Token ID736
Feature activation+6.08e-03

Pos logit contributions
rily+0.063
stack+0.061
ishly+0.057
bled+0.053
 afloat+0.051
Neg logit contributions
clock-0.050
up-0.050
ups-0.049
 cares-0.044
rays-0.044
Token to
Token ID284
Feature activation-5.13e-03

Pos logit contributions
clock+0.006
ups+0.006
up+0.006
rays+0.006
 cares+0.005
Neg logit contributions
rily-0.015
stack-0.014
ishly-0.014
bled-0.013
 afloat-0.013
Token his
Token ID465
Feature activation-2.14e-02

Pos logit contributions
rily+0.013
stack+0.011
ishly+0.011
bled+0.011
 afloat+0.010
Neg logit contributions
clock-0.005
 dreams-0.005
 listens-0.005
 visits-0.005
 looks-0.005
Token.
Token ID13
Feature activation-1.51e-02

Pos logit contributions
rily+0.131
stack+0.126
ishly+0.120
bled+0.115
 afloat+0.110
Neg logit contributions
clock-0.055
ups-0.051
up-0.048
 grows-0.048
 cares-0.046
Token
Token ID220
Feature activation+5.53e-02

Pos logit contributions
rily+0.029
stack+0.029
ishly+0.028
 afloat+0.026
bled+0.025
Neg logit contributions
clock-0.023
ups-0.021
 cares-0.021
rays-0.021
up-0.021
Token
Token ID198
Feature activation-1.67e-02

Pos logit contributions
clock+0.077
ups+0.072
rays+0.070
up+0.070
 cares+0.068
Neg logit contributions
rily-0.115
stack-0.112
ishly-0.110
 afloat-0.099
bled-0.098
Token
Token ID198
Feature activation-1.61e-02

Pos logit contributions
rily+0.037
stack+0.037
ishly+0.036
 afloat+0.032
bled+0.031
Neg logit contributions
clock-0.021
ups-0.020
up-0.020
rays-0.019
 grows-0.018
TokenJimmy
Token ID40335
Feature activation-2.38e-02

Pos logit contributions
stack+0.030
rily+0.030
ishly+0.029
 afloat+0.026
 themselves+0.025
Neg logit contributions
clock-0.026
ups-0.025
up-0.025
 cares-0.023
rays-0.023
Token ran
Token ID4966
Feature activation-3.26e-02

Pos logit contributions
rily+0.047
stack+0.046
ishly+0.044
 afloat+0.041
 themselves+0.040
Neg logit contributions
clock-0.036
ups-0.033
up-0.033
 cares-0.032
rays-0.032
Token to
Token ID284
Feature activation-8.08e-04

Pos logit contributions
rily+0.068
stack+0.065
ishly+0.062
bled+0.057
 afloat+0.056
Neg logit contributions
clock-0.047
up-0.045
ups-0.044
rays-0.040
 tests-0.039
Token the
Token ID262
Feature activation+9.83e-03

Pos logit contributions
rily+0.002
ishly+0.002
stack+0.002
bled+0.002
kin+0.002
Neg logit contributions
 study-0.001
 tests-0.001
 answers-0.001
 looks-0.001
 math-0.001
Token truck
Token ID7779
Feature activation-5.61e-03

Pos logit contributions
clock+0.015
ups+0.014
up+0.014
rays+0.013
 grows+0.013
Neg logit contributions
rily-0.021
stack-0.018
ishly-0.018
bled-0.017
 afloat-0.017
Token to
Token ID284
Feature activation-1.02e-02

Pos logit contributions
rily+0.013
bled+0.012
stack+0.011
ishly+0.011
 afloat+0.011
Neg logit contributions
clock-0.007
rays-0.006
ups-0.006
up-0.006
 looks-0.006
Token help
Token ID1037
Feature activation-2.29e-03

Pos logit contributions
rily+0.025
bled+0.022
ishly+0.022
stack+0.021
ged+0.021
Neg logit contributions
 study-0.012
 tests-0.011
 looks-0.011
clock-0.011
 answers-0.010
Token moved
Token ID3888
Feature activation+5.93e-03

Pos logit contributions
rily+0.023
stack+0.023
ishly+0.022
 afloat+0.020
 themselves+0.020
Neg logit contributions
clock-0.020
ups-0.019
 cares-0.018
rays-0.018
up-0.018
Token his
Token ID465
Feature activation-1.46e-02

Pos logit contributions
clock+0.011
ups+0.010
 cares+0.010
up+0.010
 grows+0.010
Neg logit contributions
rily-0.010
stack-0.010
ishly-0.009
 themselves-0.009
 afloat-0.009
Token toys
Token ID14958
Feature activation-3.39e-02

Pos logit contributions
rily+0.024
stack+0.023
ishly+0.022
 afloat+0.020
 themselves+0.020
Neg logit contributions
clock-0.027
ups-0.025
up-0.025
rays-0.025
 cares-0.025
Token around
Token ID1088
Feature activation-1.58e-02

Pos logit contributions
rily+0.075
ishly+0.070
stack+0.070
bled+0.067
 Ph+0.066
Neg logit contributions
clock-0.040
ups-0.039
up-0.037
rays-0.036
 math-0.035
Token the
Token ID262
Feature activation-1.18e-02

Pos logit contributions
rily+0.034
ishly+0.033
stack+0.033
bled+0.031
 afloat+0.029
Neg logit contributions
clock-0.020
 mysteries-0.019
 math-0.018
 tests-0.018
 grows-0.018
Token room
Token ID2119
Feature activation-1.09e-02

Pos logit contributions
rily+0.021
stack+0.021
ishly+0.020
 afloat+0.019
 themselves+0.019
Neg logit contributions
clock-0.019
 cares-0.018
ups-0.018
up-0.018
rays-0.018
Token but
Token ID475
Feature activation-3.63e-02

Pos logit contributions
rily+0.024
stack+0.023
ishly+0.023
bled+0.021
 afloat+0.021
Neg logit contributions
clock-0.015
ups-0.013
up-0.013
rays-0.012
 grows-0.012
Token he
Token ID339
Feature activation-1.91e-02

Pos logit contributions
rily+0.077
stack+0.073
ishly+0.072
 afloat+0.063
 stret+0.063
Neg logit contributions
clock-0.051
ups-0.049
 grows-0.047
up-0.047
rays-0.046
Token kept
Token ID4030
Feature activation+3.94e-03

Pos logit contributions
rily+0.036
stack+0.035
ishly+0.034
 afloat+0.031
 themselves+0.031
Neg logit contributions
clock-0.030
ups-0.029
up-0.028
 cares-0.028
 grows-0.028
Token them
Token ID606
Feature activation-3.53e-03

Pos logit contributions
clock+0.007
ups+0.006
rays+0.006
 grows+0.006
up+0.006
Neg logit contributions
stack-0.007
rily-0.007
ishly-0.007
 afloat-0.006
 themselves-0.006
Token in
Token ID287
Feature activation-5.78e-03

Pos logit contributions
stack+0.007
rily+0.007
ishly+0.006
 Ph+0.006
bled+0.006
Neg logit contributions
clock-0.005
ups-0.005
up-0.005
 dreams-0.004
rays-0.004
TokenHer
Token ID9360
Feature activation+1.68e-02

Pos logit contributions
rily+0.024
stack+0.024
ishly+0.023
 themselves+0.022
bled+0.021
Neg logit contributions
clock-0.016
 cares-0.015
rays-0.015
 grows-0.014
ups-0.014
Token mom
Token ID1995
Feature activation+2.11e-02

Pos logit contributions
clock+0.033
ups+0.031
rays+0.031
up+0.031
 cares+0.031
Neg logit contributions
rily-0.025
stack-0.024
ishly-0.023
 afloat-0.021
 themselves-0.021
Token smiled
Token ID13541
Feature activation+2.79e-02

Pos logit contributions
clock+0.034
ups+0.032
up+0.032
rays+0.031
 cares+0.031
Neg logit contributions
rily-0.039
stack-0.038
ishly-0.036
 afloat-0.034
 themselves-0.033
Token and
Token ID290
Feature activation+1.41e-03

Pos logit contributions
clock+0.042
up+0.039
ups+0.038
rays+0.038
 grows+0.038
Neg logit contributions
stack-0.055
rily-0.053
ishly-0.051
 afloat-0.047
 themselves-0.046
Token said
Token ID531
Feature activation-2.49e-04

Pos logit contributions
up+0.002
clock+0.002
ups+0.002
 grows+0.002
 listens+0.002
Neg logit contributions
rily-0.003
stack-0.003
ishly-0.003
 afloat-0.003
 themselves-0.002
Token,
Token ID11
Feature activation-2.47e-02

Pos logit contributions
rily+0.000
stack+0.000
ishly+0.000
 themselves+0.000
 afloat+0.000
Neg logit contributions
clock-0.000
ups-0.000
up-0.000
rays-0.000
 cares-0.000
Token "
Token ID366
Feature activation-5.25e-03

Pos logit contributions
rily+0.054
 afloat+0.054
stack+0.054
ishly+0.051
 themselves+0.049
Neg logit contributions
clock-0.028
rays-0.025
 grows-0.025
 listens-0.025
 looks-0.024
TokenThat
Token ID2504
Feature activation-1.68e-02

Pos logit contributions
stack+0.008
rily+0.008
ishly+0.007
 themselves+0.007
bled+0.007
Neg logit contributions
clock-0.010
 cares-0.009
 grows-0.009
rays-0.009
 listens-0.009
Token's
Token ID338
Feature activation+1.51e-02

Pos logit contributions
rily+0.020
ishly+0.019
stack+0.019
 towel+0.019
 themselves+0.017
Neg logit contributions
 cares-0.032
clock-0.031
smart-0.031
neys-0.031
ups-0.030
Token a
Token ID257
Feature activation-3.23e-03

Pos logit contributions
ups+0.025
clock+0.025
up+0.025
rays+0.025
 cares+0.023
Neg logit contributions
stack-0.025
rily-0.025
 themselves-0.024
ishly-0.023
bled-0.022
Token great
Token ID1049
Feature activation+6.41e-03

Pos logit contributions
stack+0.004
rily+0.004
 towel+0.004
ishly+0.004
 afloat+0.004
Neg logit contributions
ups-0.006
rays-0.006
clock-0.006
 cares-0.006
up-0.006
TokenThe
Token ID464
Feature activation+1.10e-02

Pos logit contributions
rily+0.012
stack+0.012
ishly+0.012
 afloat+0.011
bled+0.011
Neg logit contributions
clock-0.007
ups-0.007
up-0.007
 cares-0.006
rays-0.006
Token elderly
Token ID13830
Feature activation-2.11e-02

Pos logit contributions
clock+0.016
ups+0.014
up+0.014
 cares+0.014
 grows+0.014
Neg logit contributions
rily-0.023
stack-0.021
ishly-0.021
bled-0.019
 afloat-0.019
Token man
Token ID582
Feature activation+5.81e-03

Pos logit contributions
rily+0.042
stack+0.040
ishly+0.039
 afloat+0.037
bled+0.036
Neg logit contributions
clock-0.031
ups-0.029
up-0.029
 cares-0.028
rays-0.028
Token and
Token ID290
Feature activation-3.58e-02

Pos logit contributions
clock+0.009
ups+0.009
rays+0.008
up+0.008
 mysteries+0.008
Neg logit contributions
rily-0.011
stack-0.010
ishly-0.010
 themselves-0.009
bled-0.009
Token his
Token ID465
Feature activation-4.05e-02

Pos logit contributions
rily+0.073
stack+0.066
ishly+0.066
 afloat+0.064
bled+0.063
Neg logit contributions
clock-0.049
 cares-0.049
 grows-0.047
 mysteries-0.047
 dreams-0.047
Token dog
Token ID3290
Feature activation-6.65e-02

Pos logit contributions
rily+0.078
stack+0.074
ishly+0.071
 afloat+0.066
bled+0.065
Neg logit contributions
clock-0.064
ups-0.059
 cares-0.059
up-0.058
rays-0.057
Token would
Token ID561
Feature activation-1.69e-02

Pos logit contributions
rily+0.129
stack+0.120
ishly+0.112
bled+0.109
 afloat+0.107
Neg logit contributions
clock-0.105
 cares-0.099
 grows-0.098
ups-0.098
up-0.096
Token swing
Token ID9628
Feature activation+2.32e-02

Pos logit contributions
rily+0.038
bled+0.034
ishly+0.033
stack+0.033
ged+0.032
Neg logit contributions
 grows-0.020
 study-0.020
clock-0.020
 looks-0.019
 visits-0.019
Token back
Token ID736
Feature activation-6.67e-03

Pos logit contributions
clock+0.038
ups+0.035
up+0.034
 grows+0.034
rays+0.034
Neg logit contributions
rily-0.043
stack-0.041
ishly-0.040
bled-0.035
 afloat-0.035
Token and
Token ID290
Feature activation-6.73e-03

Pos logit contributions
rily+0.013
stack+0.013
ishly+0.012
 afloat+0.011
bled+0.011
Neg logit contributions
clock-0.010
ups-0.009
 grows-0.009
rays-0.009
up-0.009
Token forth
Token ID6071
Feature activation+8.23e-03

Pos logit contributions
rily+0.014
ishly+0.012
stack+0.012
bled+0.012
ula+0.011
Neg logit contributions
clock-0.010
 grows-0.009
 listens-0.009
 looks-0.009
 cares-0.009
Token Jenny
Token ID28437
Feature activation-2.37e-04

Pos logit contributions
rily+0.001
stack+0.001
ishly+0.001
 themselves+0.001
bled+0.001
Neg logit contributions
clock-0.001
ups-0.001
up-0.001
rays-0.001
 cares-0.001
Token was
Token ID373
Feature activation-1.84e-02

Pos logit contributions
rily+0.000
stack+0.000
ishly+0.000
 afloat+0.000
 themselves+0.000
Neg logit contributions
clock-0.000
ups-0.000
up-0.000
 cares-0.000
rays-0.000
Token relieved
Token ID22598
Feature activation-4.30e-03

Pos logit contributions
rily+0.039
stack+0.038
ishly+0.035
bled+0.032
 afloat+0.031
Neg logit contributions
clock-0.026
ups-0.024
up-0.024
 cares-0.023
 visits-0.023
Token that
Token ID326
Feature activation-1.59e-02

Pos logit contributions
rily+0.010
stack+0.009
ishly+0.009
 afloat+0.008
bled+0.008
Neg logit contributions
clock-0.005
ups-0.005
 cares-0.005
up-0.005
 grows-0.005
Token the
Token ID262
Feature activation-1.66e-02

Pos logit contributions
rily+0.034
stack+0.032
ishly+0.031
 afloat+0.029
bled+0.028
Neg logit contributions
clock-0.023
ups-0.021
 cares-0.021
 grows-0.021
rays-0.020
Token monster
Token ID9234
Feature activation-6.64e-02

Pos logit contributions
rily+0.034
stack+0.032
ishly+0.031
 themselves+0.028
 afloat+0.028
Neg logit contributions
clock-0.025
ups-0.023
up-0.023
 cares-0.023
rays-0.022
Token she
Token ID673
Feature activation+7.18e-04

Pos logit contributions
rily+0.129
stack+0.127
ishly+0.123
 afloat+0.111
 themselves+0.111
Neg logit contributions
clock-0.100
ups-0.093
up-0.091
rays-0.090
 cares-0.090
Token pretended
Token ID36766
Feature activation-4.37e-02

Pos logit contributions
clock+0.001
ups+0.001
up+0.001
rays+0.001
 cares+0.001
Neg logit contributions
rily-0.001
stack-0.001
ishly-0.001
 afloat-0.001
 themselves-0.001
Token to
Token ID284
Feature activation+1.15e-02

Pos logit contributions
rily+0.098
stack+0.097
ishly+0.094
 themselves+0.087
 afloat+0.086
Neg logit contributions
clock-0.052
ups-0.046
up-0.046
rays-0.046
 cares-0.046
Token be
Token ID307
Feature activation+1.58e-03

Pos logit contributions
clock+0.015
 cares+0.014
up+0.014
 grows+0.014
ups+0.014
Neg logit contributions
rily-0.026
stack-0.023
ishly-0.023
bled-0.022
 afloat-0.021
Token scared
Token ID12008
Feature activation-2.18e-02

Pos logit contributions
clock+0.002
ups+0.002
up+0.002
rays+0.002
 cares+0.002
Neg logit contributions
rily-0.003
stack-0.003
ishly-0.003
 afloat-0.003
bled-0.003
Token smelled
Token ID37609
Feature activation+4.97e-02

Pos logit contributions
rily+0.007
stack+0.006
bled+0.006
ishly+0.006
 afloat+0.006
Neg logit contributions
 cares-0.006
clock-0.005
 grows-0.005
 listens-0.005
up-0.005
Token the
Token ID262
Feature activation+1.76e-03

Pos logit contributions
clock+0.063
up+0.059
ups+0.059
 cares+0.058
 grows+0.057
Neg logit contributions
stack-0.111
rily-0.106
ishly-0.103
 afloat-0.092
bled-0.092
Token y
Token ID331
Feature activation-1.14e-02

Pos logit contributions
clock+0.003
ups+0.002
up+0.002
 cares+0.002
rays+0.002
Neg logit contributions
rily-0.003
stack-0.003
ishly-0.003
 themselves-0.003
bled-0.003
Tokenummy
Token ID13513
Feature activation+2.07e-02

Pos logit contributions
rily+0.024
stack+0.023
ishly+0.022
 afloat+0.021
 themselves+0.021
Neg logit contributions
clock-0.016
ups-0.015
up-0.014
 cares-0.014
rays-0.014
Token food
Token ID2057
Feature activation-3.23e-02

Pos logit contributions
clock+0.030
ups+0.027
up+0.027
 cares+0.027
rays+0.027
Neg logit contributions
rily-0.042
stack-0.042
ishly-0.040
bled-0.037
 afloat-0.037
Token from
Token ID422
Feature activation-6.82e-02

Pos logit contributions
rily+0.069
ishly+0.065
stack+0.064
bled+0.063
 afloat+0.061
Neg logit contributions
 looks-0.037
clock-0.036
ups-0.036
 grows-0.035
 smells-0.033
Token the
Token ID262
Feature activation-5.83e-03

Pos logit contributions
rily+0.145
ishly+0.137
stack+0.133
bled+0.126
 afloat+0.126
Neg logit contributions
clock-0.088
 history-0.083
 mysteries-0.083
 math-0.080
 dreams-0.079
Token grill
Token ID29901
Feature activation-1.57e-02

Pos logit contributions
rily+0.011
stack+0.011
ishly+0.010
bled+0.009
 themselves+0.009
Neg logit contributions
clock-0.009
ups-0.008
up-0.008
 cares-0.008
rays-0.008
Token and
Token ID290
Feature activation+1.11e-02

Pos logit contributions
rily+0.036
bled+0.034
stack+0.033
 afloat+0.033
ishly+0.033
Neg logit contributions
clock-0.018
ups-0.016
 grows-0.015
 looks-0.015
rays-0.015
Token their
Token ID511
Feature activation-3.84e-02

Pos logit contributions
clock+0.016
 cares+0.016
ups+0.016
 grows+0.015
 listens+0.015
Neg logit contributions
rily-0.022
stack-0.021
ishly-0.021
 afloat-0.020
bled-0.019
Token tum
Token ID11814
Feature activation-2.67e-03

Pos logit contributions
rily+0.073
stack+0.070
ishly+0.067
 afloat+0.062
bled+0.062
Neg logit contributions
clock-0.062
ups-0.057
 cares-0.056
up-0.055
 grows-0.055
Token was
Token ID373
Feature activation-2.41e-02

Pos logit contributions
rily+0.057
stack+0.055
ishly+0.053
 afloat+0.049
 themselves+0.049
Neg logit contributions
clock-0.037
ups-0.035
up-0.034
 cares-0.034
rays-0.033
Token too
Token ID1165
Feature activation+2.03e-02

Pos logit contributions
rily+0.046
stack+0.045
ishly+0.044
 themselves+0.040
 afloat+0.039
Neg logit contributions
clock-0.038
ups-0.035
up-0.034
 cares-0.034
 grows-0.034
Token full
Token ID1336
Feature activation+3.62e-03

Pos logit contributions
clock+0.030
 cares+0.028
rays+0.028
ups+0.028
up+0.027
Neg logit contributions
rily-0.041
stack-0.039
ishly-0.037
 afloat-0.036
 themselves-0.035
Token.
Token ID13
Feature activation-6.38e-03

Pos logit contributions
clock+0.005
ups+0.004
up+0.004
rays+0.004
 grows+0.004
Neg logit contributions
rily-0.008
stack-0.008
ishly-0.007
 afloat-0.007
bled-0.007
Token The
Token ID383
Feature activation+6.32e-03

Pos logit contributions
rily+0.013
stack+0.013
ishly+0.013
 afloat+0.012
 themselves+0.011
Neg logit contributions
clock-0.009
ups-0.008
up-0.008
rays-0.008
 cares-0.008
Token bath
Token ID7837
Feature activation-7.38e-02

Pos logit contributions
clock+0.010
ups+0.009
up+0.009
 cares+0.009
rays+0.009
Neg logit contributions
rily-0.012
stack-0.012
ishly-0.011
 themselves-0.010
 afloat-0.010
Tokenrobe
Token ID25481
Feature activation-5.22e-02

Pos logit contributions
rily+0.130
stack+0.123
ishly+0.119
 afloat+0.109
 themselves+0.105
Neg logit contributions
up-0.125
clock-0.125
ups-0.124
 grows-0.119
rays-0.118
Token was
Token ID373
Feature activation-2.67e-02

Pos logit contributions
rily+0.111
stack+0.108
ishly+0.103
 afloat+0.098
 themselves+0.097
Neg logit contributions
clock-0.069
up-0.066
ups-0.065
 grows-0.064
 cares-0.063
Token too
Token ID1165
Feature activation+1.70e-02

Pos logit contributions
stack+0.055
rily+0.055
ishly+0.054
 themselves+0.047
 afloat+0.046
Neg logit contributions
clock-0.037
ups-0.033
 grows-0.033
up-0.032
 mysteries-0.030
Token big
Token ID1263
Feature activation-2.45e-03

Pos logit contributions
clock+0.023
ups+0.021
up+0.021
 cares+0.021
rays+0.021
Neg logit contributions
rily-0.036
stack-0.035
ishly-0.034
 afloat-0.031
 themselves-0.031
Token.
Token ID13
Feature activation-1.34e-02

Pos logit contributions
rily+0.005
stack+0.005
ishly+0.005
 afloat+0.004
bled+0.004
Neg logit contributions
clock-0.004
ups-0.003
up-0.003
rays-0.003
 cares-0.003
Token gifts
Token ID13201
Feature activation-2.34e-03

Pos logit contributions
rily+0.070
stack+0.069
ishly+0.067
 afloat+0.061
bled+0.061
Neg logit contributions
clock-0.052
ups-0.048
up-0.048
rays-0.048
 cares-0.047
Token.
Token ID13
Feature activation+1.17e-02

Pos logit contributions
rily+0.005
ishly+0.005
stack+0.005
bled+0.005
 afloat+0.005
Neg logit contributions
clock-0.002
rays-0.002
ups-0.002
up-0.002
 grows-0.002
Token He
Token ID679
Feature activation+1.27e-02

Pos logit contributions
clock+0.015
ups+0.014
up+0.014
rays+0.014
 cares+0.014
Neg logit contributions
rily-0.025
stack-0.025
ishly-0.024
 afloat-0.022
bled-0.022
Token told
Token ID1297
Feature activation-2.16e-02

Pos logit contributions
clock+0.021
ups+0.019
up+0.019
 cares+0.019
rays+0.019
Neg logit contributions
rily-0.023
stack-0.023
ishly-0.022
 afloat-0.020
 themselves-0.020
Token his
Token ID465
Feature activation-1.24e-02

Pos logit contributions
stack+0.039
rily+0.039
ishly+0.036
 afloat+0.035
bled+0.032
Neg logit contributions
clock-0.036
up-0.036
ups-0.035
rays-0.034
 cares-0.033
Token re
Token ID302
Feature activation-6.66e-02

Pos logit contributions
rily+0.018
stack+0.017
ishly+0.016
 themselves+0.015
 afloat+0.014
Neg logit contributions
clock-0.024
ups-0.023
 cares-0.023
up-0.023
rays-0.023
Tokenind
Token ID521
Feature activation-6.54e-03

Pos logit contributions
rily+0.130
stack+0.123
ishly+0.119
 themselves+0.110
 afloat+0.109
Neg logit contributions
clock-0.101
 grows-0.094
up-0.091
 cares-0.091
rays-0.091
Tokeneer
Token ID28153
Feature activation-3.74e-02

Pos logit contributions
rily+0.009
stack+0.008
 themselves+0.008
ishly+0.007
fully+0.007
Neg logit contributions
clock-0.013
ups-0.012
 grows-0.012
up-0.012
 cares-0.011
Token to
Token ID284
Feature activation-9.40e-03

Pos logit contributions
stack+0.073
rily+0.072
ishly+0.072
 themselves+0.066
 afloat+0.062
Neg logit contributions
clock-0.057
ups-0.050
up-0.049
 grows-0.049
rays-0.049
Token stay
Token ID2652
Feature activation-1.98e-02

Pos logit contributions
rily+0.021
ishly+0.019
bled+0.019
stack+0.018
ged+0.018
Neg logit contributions
clock-0.012
up-0.011
 grows-0.011
 cares-0.011
 study-0.011
Token put
Token ID1234
Feature activation-1.39e-02

Pos logit contributions
stack+0.040
ishly+0.038
rily+0.036
bled+0.032
ula+0.032
Neg logit contributions
clock-0.030
up-0.027
 grows-0.027
 visits-0.027
ups-0.027