TOP ACTIVATIONS
MAX = 8.626

Lily . She loved l oll ip ops , especially the
Lily . She loved l oll ip ops and would always
little girl who loved l oll ip ops . One day
Lily . She loved l oll ip ops and always had
Lily . She loved l oll ip ops and always had
there was a great l oll ipop . It was pink
John . John loved l oll ip ops ! One day
Lily . She loved l oll ip ops and ate them
named Lily who loved l oll ip ops . She would
to walk and give l oll ip ops to children ."
Lily . She loved l oll ip ops and always had
I love Car - D oll . And I love our
friends and shared their l oll ip ops every time they
Tess . She loved l oll ip ops . One day
as she held the l oll ipop in her hand .
and they enjoyed the l oll ipop together . The guard
happy . And giving l oll ip ops to children makes
generous and share her l oll ip ops with others ,
because she loved her l oll ip ops so much .
Lily happily took the l oll ipop and started licking it

MIN ACTIVATIONS
MIN = -10.188

the sky , the dragon ruled over the land with might
grass around the hill was wit hered and yellow . He
, feeling the peace and tranqu ility of the night ,
the garden . A subtle tranqu ility filled the air and
, it kept on flying err atically with inexplicable speed .
saw that the land was ruled by a mean monster .
big , scary lion who ruled over his nation of animals
and mysterious land that was ruled by a powerful witch .
he was in a kingdom ruled by a wicked wizard .
They said , " L adies and gentlemen , welcome to
night , the wild cat lur ked through the woods ,
. He says " L adies and gentlemen , boys and
a sign that says " tem ple ". They ask mom
. Luckily Sarah ret uned home safe and sound
a patch of butter c ups near the hedge . Daisy
't out run the enormous wolves . Stru gg ling to
and then a loud " CR ASH !" echoed around the
table , making the candle topple over . Everyone quickly rushed
became good friends . They ruled the land together . They
admired the way it flew grace fully in the sky and

INTERVAL 3.923 - 8.626
CONTAINS 0.327%

They say , " We do not like the monster card
said , " Hi , what are you doing ?" The
. From that day on , Jack and
Lily . " Here you go . Tell me
nodded and said , " That sounds good , grandma .

INTERVAL -0.781 - 3.923
CONTAINS 73.170%

ed with delight . At the end , Chloe
a few minutes , the car stopped . They looked in
the ornament if you can have it ?" So , Twe
little creature dressed in green . The creature said , "
help the dog learn to balance . So , the wise

INTERVAL -5.484 - -0.781
CONTAINS 26.482%

quite dark out . His owner was waiting patiently and walked
on the table . She wanted to give her grandma a
at Fl uffy and said , " Sure ! My name
replied , " I have never seen a bald man before
. When he got there , he saw

INTERVAL -10.188 - -5.484
CONTAINS 0.021%

. Suddenly , Ham ster tri pped and the bone flew
The fire work show lasted for what seemed like hours
the seams of her sweater * po pped * ! Everyone
to his friend . He scam pered around trying to find
and saw there were no monsters . She felt proud of
Token Lily
Token ID20037
Feature activation+1.74e+00

Pos logit contributions
iflower+0.007
pron+0.006
wl+0.005
live+0.005
tto+0.005
Neg logit contributions
 seen-0.006
hole-0.005
 towards-0.005
keeping-0.005
 stations-0.005
Token.
Token ID13
Feature activation+3.40e-02

Pos logit contributions
iflower+0.062
pron+0.056
live+0.051
wl+0.049
innie+0.048
Neg logit contributions
 seen-0.038
keeping-0.037
 stations-0.037
hole-0.037
 contain-0.037
Token She
Token ID1375
Feature activation+1.18e+00

Pos logit contributions
iflower+0.001
pron+0.001
live+0.001
wl+0.001
ama+0.001
Neg logit contributions
 seen-0.001
hole-0.001
 towards-0.001
keeping-0.001
 stations-0.001
Token loved
Token ID6151
Feature activation+9.64e-01

Pos logit contributions
iflower+0.041
ama+0.039
pron+0.038
omm+0.037
innie+0.036
Neg logit contributions
 seen-0.026
 want-0.021
 like-0.019
ets-0.018
 towards-0.018
Token l
Token ID300
Feature activation+2.77e-01

Pos logit contributions
iflower+0.038
pron+0.036
live+0.035
ama+0.034
wl+0.033
Neg logit contributions
 seen-0.018
keeping-0.016
 umb-0.015
 crashing-0.015
 stations-0.015
Tokenoll
Token ID692
Feature activation+8.63e+00

Pos logit contributions
iflower+0.009
pron+0.009
lf+0.007
very+0.007
wl+0.006
Neg logit contributions
 seen-0.008
 contain-0.007
hole-0.007
ales-0.007
vious-0.007
Tokenip
Token ID541
Feature activation+5.68e+00

Pos logit contributions
pron+0.165
iflower+0.150
Mon+0.127
live+0.121
Me+0.121
Neg logit contributions
 seen-0.293
ales-0.285
 permission-0.273
owing-0.271
ously-0.271
Tokenops
Token ID2840
Feature activation+1.67e+00

Pos logit contributions
pron+0.160
iflower+0.151
innie+0.118
ama+0.117
very+0.114
Neg logit contributions
ales-0.149
ets-0.147
 seen-0.145
 officers-0.142
 argued-0.141
Token,
Token ID11
Feature activation+2.62e-01

Pos logit contributions
iflower+0.064
pron+0.061
live+0.055
innie+0.055
wl+0.054
Neg logit contributions
 seen-0.031
hole-0.028
keeping-0.028
 contain-0.027
 stations-0.027
Token especially
Token ID2592
Feature activation+1.41e+00

Pos logit contributions
iflower+0.010
pron+0.009
live+0.009
ama+0.008
wl+0.008
Neg logit contributions
 seen-0.005
hole-0.005
 stations-0.004
 crashing-0.004
 contain-0.004
Token the
Token ID262
Feature activation-1.19e-01

Pos logit contributions
iflower+0.048
pron+0.047
ama+0.045
wl+0.043
oppy+0.041
Neg logit contributions
 seen-0.029
 towards-0.029
 like-0.028
 crashing-0.027
ked-0.026
Token Lily
Token ID20037
Feature activation+1.83e+00

Pos logit contributions
iflower+0.003
pron+0.003
wl+0.003
live+0.003
tto+0.003
Neg logit contributions
 seen-0.003
hole-0.003
keeping-0.003
 towards-0.003
 contain-0.003
Token.
Token ID13
Feature activation+4.11e-02

Pos logit contributions
iflower+0.064
pron+0.058
live+0.053
wl+0.051
innie+0.051
Neg logit contributions
 seen-0.040
keeping-0.040
 stations-0.039
hole-0.039
 officers-0.039
Token She
Token ID1375
Feature activation+1.28e+00

Pos logit contributions
iflower+0.002
pron+0.001
live+0.001
wl+0.001
ama+0.001
Neg logit contributions
 seen-0.001
hole-0.001
 towards-0.001
 stations-0.001
keeping-0.001
Token loved
Token ID6151
Feature activation+1.01e+00

Pos logit contributions
iflower+0.044
ama+0.041
pron+0.041
omm+0.038
innie+0.037
Neg logit contributions
 seen-0.029
 want-0.024
 like-0.022
 towards-0.021
keeping-0.021
Token l
Token ID300
Feature activation+2.39e-01

Pos logit contributions
iflower+0.041
pron+0.038
live+0.037
ama+0.035
wl+0.035
Neg logit contributions
 seen-0.019
keeping-0.017
 umb-0.016
 stations-0.016
 crashing-0.016
Tokenoll
Token ID692
Feature activation+8.56e+00

Pos logit contributions
iflower+0.008
pron+0.008
lf+0.006
very+0.006
wl+0.006
Neg logit contributions
 seen-0.006
 contain-0.006
hole-0.006
ets-0.006
vious-0.006
Tokenip
Token ID541
Feature activation+5.55e+00

Pos logit contributions
pron+0.159
iflower+0.144
Mon+0.125
Me+0.119
live+0.116
Neg logit contributions
 seen-0.292
ales-0.284
ets-0.273
 permission-0.272
owing-0.272
Tokenops
Token ID2840
Feature activation+1.51e+00

Pos logit contributions
pron+0.159
iflower+0.150
ama+0.117
innie+0.116
very+0.114
Neg logit contributions
ales-0.145
ets-0.143
 seen-0.140
 officers-0.138
keeping-0.135
Token and
Token ID290
Feature activation+1.86e+00

Pos logit contributions
iflower+0.058
pron+0.056
innie+0.050
live+0.049
wl+0.049
Neg logit contributions
 seen-0.029
hole-0.026
keeping-0.025
 contain-0.024
ales-0.024
Token would
Token ID561
Feature activation-3.04e-01

Pos logit contributions
iflower+0.057
pron+0.049
ama+0.047
live+0.046
wl+0.045
Neg logit contributions
 seen-0.052
 argued-0.046
 crashing-0.045
 stations-0.045
 officers-0.044
Token always
Token ID1464
Feature activation+1.07e+00

Pos logit contributions
 seen+0.008
 contain+0.007
 crashing+0.007
 stations+0.007
hole+0.007
Neg logit contributions
iflower-0.010
pron-0.009
live-0.008
ama-0.008
innie-0.008
Token little
Token ID1310
Feature activation+8.25e-01

Pos logit contributions
iflower+0.032
live+0.031
wl+0.030
innie+0.028
tto+0.028
Neg logit contributions
 crashing-0.016
 want-0.016
 brake-0.014
 officers-0.014
 steering-0.014
Token girl
Token ID2576
Feature activation+1.07e+00

Pos logit contributions
iflower+0.028
pron+0.025
live+0.025
wl+0.023
innie+0.023
Neg logit contributions
 seen-0.019
 officers-0.018
 crashing-0.017
hole-0.017
 stations-0.017
Token who
Token ID508
Feature activation+1.96e+00

Pos logit contributions
iflower+0.035
pron+0.032
live+0.029
innie+0.027
ama+0.027
Neg logit contributions
 seen-0.028
hole-0.026
keeping-0.025
 stations-0.025
 crashing-0.024
Token loved
Token ID6151
Feature activation+6.00e-01

Pos logit contributions
iflower+0.060
pron+0.054
ama+0.049
innie+0.046
live+0.046
Neg logit contributions
 seen-0.059
 want-0.052
keeping-0.050
 towards-0.048
hole-0.047
Token l
Token ID300
Feature activation+1.93e-01

Pos logit contributions
iflower+0.022
pron+0.021
live+0.019
wl+0.019
ama+0.018
Neg logit contributions
 seen-0.013
keeping-0.012
 stations-0.012
 crashing-0.011
 officers-0.011
Tokenoll
Token ID692
Feature activation+8.51e+00

Pos logit contributions
iflower+0.006
pron+0.005
wl+0.004
live+0.004
innie+0.004
Neg logit contributions
 seen-0.006
hole-0.005
 contain-0.005
keeping-0.005
ales-0.005
Tokenip
Token ID541
Feature activation+5.50e+00

Pos logit contributions
iflower+0.144
pron+0.140
live+0.109
wl+0.096
innie+0.095
Neg logit contributions
 seen-0.313
ales-0.297
hole-0.294
 officers-0.291
 permission-0.290
Tokenops
Token ID2840
Feature activation+1.54e+00

Pos logit contributions
iflower+0.141
pron+0.140
innie+0.107
live+0.104
ama+0.102
Neg logit contributions
 seen-0.160
 officers-0.158
ales-0.157
keeping-0.155
hole-0.152
Token.
Token ID13
Feature activation-7.00e-02

Pos logit contributions
iflower+0.058
pron+0.054
live+0.048
ama+0.046
innie+0.046
Neg logit contributions
 seen-0.032
hole-0.030
keeping-0.030
 stations-0.029
 brake-0.028
Token One
Token ID1881
Feature activation-8.06e-02

Pos logit contributions
 seen+0.002
keeping+0.001
hole+0.001
 towards+0.001
 stations+0.001
Neg logit contributions
iflower-0.003
pron-0.002
live-0.002
wl-0.002
tto-0.002
Token day
Token ID1110
Feature activation-4.91e-01

Pos logit contributions
 seen+0.002
 want+0.002
 towards+0.002
 eye+0.002
 contain+0.002
Neg logit contributions
iflower-0.002
pron-0.002
wl-0.002
innie-0.002
live-0.002
Token Lily
Token ID20037
Feature activation+1.77e+00

Pos logit contributions
iflower+0.004
pron+0.004
wl+0.003
tto+0.003
live+0.003
Neg logit contributions
 seen-0.004
hole-0.003
keeping-0.003
 towards-0.003
 stations-0.003
Token.
Token ID13
Feature activation-1.62e-01

Pos logit contributions
iflower+0.062
pron+0.056
live+0.051
wl+0.049
innie+0.049
Neg logit contributions
 seen-0.039
keeping-0.038
 stations-0.038
 contain-0.037
hole-0.037
Token She
Token ID1375
Feature activation+1.17e+00

Pos logit contributions
 seen+0.004
hole+0.004
 stations+0.003
keeping+0.003
 towards+0.003
Neg logit contributions
iflower-0.006
pron-0.005
live-0.005
wl-0.005
ama-0.005
Token loved
Token ID6151
Feature activation+1.01e+00

Pos logit contributions
iflower+0.041
ama+0.039
pron+0.039
omm+0.036
innie+0.035
Neg logit contributions
 seen-0.025
 want-0.021
 like-0.019
keeping-0.018
 towards-0.018
Token l
Token ID300
Feature activation+2.00e-01

Pos logit contributions
iflower+0.040
pron+0.038
live+0.036
ama+0.035
wl+0.035
Neg logit contributions
 seen-0.019
keeping-0.017
 umb-0.016
 crashing-0.016
 stations-0.016
Tokenoll
Token ID692
Feature activation+8.51e+00

Pos logit contributions
iflower+0.007
pron+0.006
lf+0.005
very+0.005
wl+0.005
Neg logit contributions
 seen-0.005
 contain-0.005
ets-0.005
hole-0.005
vious-0.005
Tokenip
Token ID541
Feature activation+5.44e+00

Pos logit contributions
pron+0.163
iflower+0.150
Mon+0.132
live+0.122
Me+0.121
Neg logit contributions
 seen-0.285
ales-0.276
 permission-0.265
owing-0.265
ets-0.264
Tokenops
Token ID2840
Feature activation+1.38e+00

Pos logit contributions
pron+0.154
iflower+0.145
innie+0.114
ama+0.113
very+0.109
Neg logit contributions
ales-0.145
ets-0.142
 seen-0.138
 officers-0.138
keeping-0.137
Token and
Token ID290
Feature activation+1.87e+00

Pos logit contributions
iflower+0.053
pron+0.051
live+0.045
innie+0.045
wl+0.045
Neg logit contributions
 seen-0.027
hole-0.024
keeping-0.024
 stations-0.023
 contain-0.023
Token always
Token ID1464
Feature activation+1.25e+00

Pos logit contributions
iflower+0.055
pron+0.048
ama+0.045
live+0.044
wl+0.043
Neg logit contributions
 seen-0.055
 crashing-0.048
 stations-0.048
 argued-0.047
keeping-0.047
Token had
Token ID550
Feature activation+2.44e+00

Pos logit contributions
iflower+0.040
pron+0.036
live+0.033
ama+0.032
innie+0.031
Neg logit contributions
 seen-0.034
hole-0.031
keeping-0.031
 stations-0.029
 contain-0.029
Token Lily
Token ID20037
Feature activation+1.49e+00

Pos logit contributions
 seen+0.001
keeping+0.001
hole+0.001
 towards+0.001
 contain+0.001
Neg logit contributions
iflower-0.001
pron-0.001
wl-0.001
live-0.001
tto-0.001
Token.
Token ID13
Feature activation-1.93e-01

Pos logit contributions
iflower+0.051
pron+0.046
live+0.042
wl+0.040
innie+0.040
Neg logit contributions
 seen-0.033
keeping-0.033
 contain-0.033
hole-0.033
 officers-0.033
Token She
Token ID1375
Feature activation+1.10e+00

Pos logit contributions
 seen+0.005
hole+0.004
 towards+0.004
 stations+0.004
keeping+0.004
Neg logit contributions
iflower-0.007
pron-0.006
live-0.006
wl-0.006
ama-0.005
Token loved
Token ID6151
Feature activation+8.25e-01

Pos logit contributions
iflower+0.038
pron+0.036
ama+0.035
omm+0.033
innie+0.032
Neg logit contributions
 seen-0.024
 want-0.021
 like-0.018
keeping-0.018
 towards-0.018
Token l
Token ID300
Feature activation+1.27e-01

Pos logit contributions
iflower+0.033
pron+0.031
live+0.030
ama+0.029
wl+0.028
Neg logit contributions
 seen-0.015
keeping-0.014
 umb-0.013
 stations-0.013
 crashing-0.013
Tokenoll
Token ID692
Feature activation+8.46e+00

Pos logit contributions
iflower+0.004
pron+0.004
wl+0.003
lf+0.003
live+0.003
Neg logit contributions
 seen-0.004
hole-0.003
 contain-0.003
ales-0.003
keeping-0.003
Tokenip
Token ID541
Feature activation+5.53e+00

Pos logit contributions
pron+0.153
iflower+0.144
live+0.113
Mon+0.109
Me+0.103
Neg logit contributions
 seen-0.303
ales-0.295
 permission-0.283
hole-0.282
ets-0.282
Tokenops
Token ID2840
Feature activation+1.49e+00

Pos logit contributions
pron+0.156
iflower+0.148
innie+0.114
ama+0.112
very+0.110
Neg logit contributions
ales-0.147
ets-0.145
 seen-0.143
 officers-0.141
keeping-0.140
Token and
Token ID290
Feature activation+1.97e+00

Pos logit contributions
iflower+0.058
pron+0.055
live+0.049
innie+0.049
wl+0.048
Neg logit contributions
 seen-0.029
hole-0.026
keeping-0.026
 stations-0.025
 contain-0.024
Token always
Token ID1464
Feature activation+1.10e+00

Pos logit contributions
iflower+0.057
pron+0.050
live+0.045
ama+0.045
wl+0.044
Neg logit contributions
 seen-0.058
 stations-0.052
 crashing-0.051
keeping-0.051
 argued-0.051
Token had
Token ID550
Feature activation+2.59e+00

Pos logit contributions
iflower+0.036
pron+0.033
live+0.029
ama+0.028
innie+0.028
Neg logit contributions
 seen-0.029
hole-0.026
keeping-0.026
 stations-0.026
 contain-0.025
Token there
Token ID612
Feature activation+1.21e+00

Pos logit contributions
iflower+0.058
pron+0.054
live+0.045
innie+0.045
wl+0.043
Neg logit contributions
 seen-0.035
hole-0.032
 towards-0.030
 ago-0.030
 crashing-0.030
Token was
Token ID373
Feature activation+1.26e+00

Pos logit contributions
iflower+0.041
pron+0.039
innie+0.032
wl+0.031
live+0.030
Neg logit contributions
 seen-0.034
hole-0.029
 gasp-0.027
 crashing-0.027
 towards-0.027
Token a
Token ID257
Feature activation+1.35e+00

Pos logit contributions
iflower+0.044
pron+0.041
live+0.036
wl+0.035
ama+0.035
Neg logit contributions
 seen-0.031
keeping-0.027
hole-0.026
 stations-0.026
 crashing-0.025
Token great
Token ID1049
Feature activation+2.74e+00

Pos logit contributions
iflower+0.049
live+0.047
wl+0.045
innie+0.042
tto+0.042
Neg logit contributions
 want-0.022
 crashing-0.022
 brake-0.019
Run-0.018
 seen-0.018
Token l
Token ID300
Feature activation-2.31e-01

Pos logit contributions
iflower+0.092
pron+0.080
live+0.079
wl+0.077
innie+0.074
Neg logit contributions
 want-0.059
 seen-0.059
 crashing-0.057
 officers-0.053
 eye-0.053
Tokenoll
Token ID692
Feature activation+8.45e+00

Pos logit contributions
 seen+0.005
hole+0.005
ets+0.004
 contain+0.004
vious+0.004
Neg logit contributions
pron-0.008
iflower-0.008
innie-0.007
tto-0.006
very-0.006
Tokenipop
Token ID42800
Feature activation+3.64e+00

Pos logit contributions
pron+0.180
iflower+0.180
Mon+0.166
Me+0.164
Rich+0.160
Neg logit contributions
 seen-0.251
ales-0.223
hole-0.219
owing-0.219
ously-0.217
Token.
Token ID13
Feature activation-4.56e-02

Pos logit contributions
iflower+0.140
pron+0.134
live+0.122
innie+0.121
wl+0.118
Neg logit contributions
 seen-0.061
hole-0.058
 contain-0.053
 crashing-0.053
 officers-0.052
Token It
Token ID632
Feature activation+2.81e+00

Pos logit contributions
 seen+0.001
keeping+0.001
hole+0.001
 towards+0.001
 crashing+0.001
Neg logit contributions
iflower-0.002
pron-0.002
live-0.001
wl-0.001
tto-0.001
Token was
Token ID373
Feature activation-4.95e-01

Pos logit contributions
ama+0.100
iflower+0.097
pron+0.096
innie+0.086
irl+0.085
Neg logit contributions
 seen-0.051
 like-0.048
 want-0.042
 towards-0.041
 hid-0.035
Token pink
Token ID11398
Feature activation+2.34e+00

Pos logit contributions
 seen+0.012
 crashing+0.009
hole+0.009
 towards+0.009
 contain+0.009
Neg logit contributions
iflower-0.018
pron-0.016
wl-0.015
live-0.015
ama-0.015
Token John
Token ID1757
Feature activation+1.88e+00

Pos logit contributions
 seen+0.036
hole+0.035
 stations+0.034
keeping+0.034
 contain+0.032
Neg logit contributions
iflower-0.043
pron-0.040
live-0.036
innie-0.035
ama-0.035
Token.
Token ID13
Feature activation-3.74e-01

Pos logit contributions
iflower+0.066
pron+0.061
live+0.054
innie+0.053
wl+0.052
Neg logit contributions
keeping-0.041
 seen-0.041
 stations-0.041
hole-0.041
 contain-0.040
Token John
Token ID1757
Feature activation+2.23e+00

Pos logit contributions
 seen+0.008
hole+0.008
keeping+0.008
 stations+0.008
 contain+0.007
Neg logit contributions
iflower-0.013
pron-0.012
live-0.011
innie-0.011
ama-0.011
Token loved
Token ID6151
Feature activation+8.05e-01

Pos logit contributions
iflower+0.066
pron+0.063
ama+0.055
live+0.054
innie+0.053
Neg logit contributions
 seen-0.066
hole-0.056
keeping-0.055
 want-0.054
 stations-0.054
Token l
Token ID300
Feature activation+1.40e-01

Pos logit contributions
iflower+0.030
pron+0.027
live+0.026
ama+0.025
wl+0.025
Neg logit contributions
 seen-0.018
keeping-0.016
 stations-0.016
 crashing-0.015
 officers-0.015
Tokenoll
Token ID692
Feature activation+8.34e+00

Pos logit contributions
iflower+0.004
pron+0.004
live+0.003
wl+0.003
innie+0.003
Neg logit contributions
 seen-0.004
hole-0.004
keeping-0.004
 contain-0.004
ales-0.004
Tokenip
Token ID541
Feature activation+5.45e+00

Pos logit contributions
pron+0.134
iflower+0.132
live+0.102
Mon+0.089
wl+0.088
Neg logit contributions
 seen-0.313
ales-0.302
hole-0.293
ets-0.292
 officers-0.290
Tokenops
Token ID2840
Feature activation+1.43e+00

Pos logit contributions
pron+0.139
iflower+0.137
innie+0.105
ama+0.103
live+0.102
Neg logit contributions
 seen-0.159
ales-0.157
keeping-0.155
 officers-0.155
ets-0.153
Token!
Token ID0
Feature activation-4.05e-01

Pos logit contributions
iflower+0.055
pron+0.051
live+0.047
innie+0.045
wl+0.045
Neg logit contributions
 seen-0.028
hole-0.026
keeping-0.026
 stations-0.025
 contain-0.024
Token One
Token ID1881
Feature activation-1.47e-01

Pos logit contributions
 seen+0.010
hole+0.010
keeping+0.010
 stations+0.010
 crashing+0.009
Neg logit contributions
iflower-0.013
pron-0.012
live-0.011
wl-0.010
ama-0.010
Token day
Token ID1110
Feature activation-5.72e-01

Pos logit contributions
 seen+0.004
 want+0.004
 towards+0.004
 eye+0.004
 contain+0.004
Neg logit contributions
iflower-0.004
pron-0.004
wl-0.003
innie-0.003
live-0.003
Token Lily
Token ID20037
Feature activation+1.52e+00

Pos logit contributions
 seen+0.001
keeping+0.001
 towards+0.001
hole+0.001
 contain+0.001
Neg logit contributions
iflower-0.002
pron-0.001
wl-0.001
tto-0.001
live-0.001
Token.
Token ID13
Feature activation-2.27e-01

Pos logit contributions
iflower+0.053
pron+0.048
live+0.044
wl+0.042
innie+0.042
Neg logit contributions
 seen-0.033
keeping-0.033
 stations-0.033
hole-0.033
 officers-0.032
Token She
Token ID1375
Feature activation+1.12e+00

Pos logit contributions
 seen+0.005
hole+0.005
 stations+0.005
 towards+0.005
keeping+0.005
Neg logit contributions
iflower-0.008
pron-0.007
live-0.007
wl-0.007
ama-0.006
Token loved
Token ID6151
Feature activation+8.79e-01

Pos logit contributions
iflower+0.039
ama+0.037
pron+0.036
omm+0.035
innie+0.033
Neg logit contributions
 seen-0.025
 want-0.020
 like-0.018
 towards-0.018
ets-0.017
Token l
Token ID300
Feature activation+3.81e-02

Pos logit contributions
iflower+0.035
pron+0.033
live+0.032
ama+0.031
wl+0.031
Neg logit contributions
 seen-0.016
keeping-0.014
 stations-0.014
 crashing-0.014
 umb-0.014
Tokenoll
Token ID692
Feature activation+8.32e+00

Pos logit contributions
iflower+0.001
pron+0.001
lf+0.001
wl+0.001
very+0.001
Neg logit contributions
 seen-0.001
hole-0.001
 contain-0.001
vious-0.001
ales-0.001
Tokenip
Token ID541
Feature activation+5.33e+00

Pos logit contributions
pron+0.147
iflower+0.144
live+0.112
Mon+0.106
omm+0.101
Neg logit contributions
 seen-0.294
ales-0.279
ked-0.276
ously-0.274
hole-0.273
Tokenops
Token ID2840
Feature activation+1.34e+00

Pos logit contributions
pron+0.154
iflower+0.147
innie+0.113
ama+0.111
omm+0.110
Neg logit contributions
ales-0.135
ets-0.134
 officers-0.132
 argued-0.131
 seen-0.130
Token and
Token ID290
Feature activation+1.83e+00

Pos logit contributions
iflower+0.052
pron+0.050
live+0.045
innie+0.044
wl+0.044
Neg logit contributions
 seen-0.025
hole-0.023
keeping-0.022
 stations-0.022
 contain-0.022
Token ate
Token ID15063
Feature activation+2.09e+00

Pos logit contributions
iflower+0.052
pron+0.046
live+0.042
ama+0.041
wl+0.040
Neg logit contributions
 seen-0.055
 stations-0.050
keeping-0.049
 crashing-0.049
 officers-0.049
Token them
Token ID606
Feature activation-3.39e-01

Pos logit contributions
iflower+0.083
pron+0.080
wl+0.076
live+0.076
ama+0.074
Neg logit contributions
 towards-0.031
 crashing-0.031
 seen-0.030
 umb-0.029
 stations-0.028
Token named
Token ID3706
Feature activation-1.75e-01

Pos logit contributions
iflower+0.021
pron+0.019
live+0.017
innie+0.016
ama+0.016
Neg logit contributions
 seen-0.017
hole-0.015
keeping-0.015
 stations-0.015
 crashing-0.014
Token Lily
Token ID20037
Feature activation+1.35e+00

Pos logit contributions
 seen+0.005
keeping+0.005
hole+0.004
 towards+0.004
 contain+0.004
Neg logit contributions
iflower-0.006
pron-0.005
wl-0.004
live-0.004
tto-0.004
Token who
Token ID508
Feature activation+1.65e+00

Pos logit contributions
iflower+0.046
pron+0.041
live+0.038
wl+0.036
innie+0.036
Neg logit contributions
 seen-0.030
keeping-0.030
 contain-0.030
 stations-0.030
hole-0.030
Token loved
Token ID6151
Feature activation+5.91e-01

Pos logit contributions
iflower+0.057
pron+0.052
ama+0.050
omm+0.047
opus+0.046
Neg logit contributions
 seen-0.041
 want-0.038
 like-0.032
 towards-0.030
 hope-0.029
Token l
Token ID300
Feature activation+8.21e-03

Pos logit contributions
iflower+0.022
pron+0.021
live+0.019
ama+0.019
wl+0.019
Neg logit contributions
 seen-0.012
keeping-0.012
 stations-0.011
 crashing-0.011
 officers-0.011
Tokenoll
Token ID692
Feature activation+8.19e+00

Pos logit contributions
iflower+0.000
pron+0.000
live+0.000
innie+0.000
wl+0.000
Neg logit contributions
 seen-0.000
hole-0.000
keeping-0.000
 contain-0.000
ales-0.000
Tokenip
Token ID541
Feature activation+5.11e+00

Pos logit contributions
pron+0.129
iflower+0.125
live+0.095
innie+0.080
wl+0.079
Neg logit contributions
 seen-0.315
ales-0.302
hole-0.298
 permission-0.293
ously-0.292
Tokenops
Token ID2840
Feature activation+1.12e+00

Pos logit contributions
pron+0.132
iflower+0.130
innie+0.098
live+0.095
ama+0.094
Neg logit contributions
 seen-0.147
ales-0.145
 officers-0.144
keeping-0.142
ets-0.140
Token.
Token ID13
Feature activation-1.44e-01

Pos logit contributions
iflower+0.043
pron+0.040
live+0.036
innie+0.035
ama+0.035
Neg logit contributions
 seen-0.023
hole-0.021
keeping-0.021
 stations-0.020
 officers-0.019
Token She
Token ID1375
Feature activation+1.36e+00

Pos logit contributions
 seen+0.003
hole+0.003
 stations+0.003
keeping+0.003
 towards+0.003
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
wl-0.004
tto-0.004
Token would
Token ID561
Feature activation-1.21e+00

Pos logit contributions
iflower+0.053
ama+0.049
pron+0.045
omm+0.044
opus+0.044
Neg logit contributions
 seen-0.028
 argued-0.022
 want-0.021
 hid-0.018
ets-0.018
Token to
Token ID284
Feature activation+3.02e+00

Pos logit contributions
iflower+0.019
pron+0.019
live+0.017
wl+0.017
ama+0.016
Neg logit contributions
 seen-0.015
keeping-0.013
 towards-0.013
 crashing-0.013
hole-0.012
Token walk
Token ID2513
Feature activation+2.92e-01

Pos logit contributions
iflower+0.109
pron+0.100
live+0.090
wl+0.087
ama+0.087
Neg logit contributions
 seen-0.069
hole-0.062
keeping-0.062
 contain-0.060
 stations-0.060
Token and
Token ID290
Feature activation+1.34e+00

Pos logit contributions
iflower+0.009
pron+0.009
live+0.008
innie+0.007
ama+0.007
Neg logit contributions
 seen-0.008
hole-0.007
keeping-0.007
 stations-0.007
 contain-0.007
Token give
Token ID1577
Feature activation+2.94e+00

Pos logit contributions
iflower+0.045
pron+0.042
live+0.037
innie+0.036
ama+0.035
Neg logit contributions
 seen-0.032
hole-0.031
keeping-0.030
 stations-0.030
 officers-0.029
Token l
Token ID300
Feature activation-5.26e-01

Pos logit contributions
iflower+0.098
pron+0.097
live+0.084
wl+0.082
innie+0.081
Neg logit contributions
 seen-0.075
keeping-0.066
 towards-0.064
 officers-0.062
ales-0.060
Tokenoll
Token ID692
Feature activation+8.17e+00

Pos logit contributions
 seen+0.017
hole+0.017
keeping+0.016
 contain+0.016
 stations+0.016
Neg logit contributions
iflower-0.013
pron-0.012
live-0.010
wl-0.010
innie-0.010
Tokenip
Token ID541
Feature activation+5.33e+00

Pos logit contributions
iflower+0.154
pron+0.153
live+0.128
wl+0.119
innie+0.111
Neg logit contributions
 seen-0.283
ales-0.270
 permission-0.269
ets-0.265
ked-0.265
Tokenops
Token ID2840
Feature activation+1.06e+00

Pos logit contributions
iflower+0.122
pron+0.120
innie+0.094
live+0.093
wl+0.092
Neg logit contributions
 seen-0.167
keeping-0.166
ets-0.162
ales-0.162
 officers-0.160
Token to
Token ID284
Feature activation+1.06e+00

Pos logit contributions
iflower+0.039
pron+0.036
live+0.032
ama+0.031
innie+0.031
Neg logit contributions
 seen-0.023
hole-0.022
 stations-0.021
keeping-0.021
 crashing-0.020
Token children
Token ID1751
Feature activation-1.23e+00

Pos logit contributions
iflower+0.033
pron+0.030
live+0.027
wl+0.026
innie+0.026
Neg logit contributions
 seen-0.029
keeping-0.026
 contain-0.026
 stations-0.026
hole-0.025
Token."
Token ID526
Feature activation+7.73e-01

Pos logit contributions
 seen+0.027
hole+0.025
 stations+0.025
keeping+0.025
 brake+0.024
Neg logit contributions
iflower-0.044
pron-0.041
live-0.037
ama-0.035
innie-0.035
Token Lily
Token ID20037
Feature activation+1.49e+00

Pos logit contributions
 seen+0.002
 towards+0.001
keeping+0.001
hole+0.001
 contain+0.001
Neg logit contributions
iflower-0.002
pron-0.002
wl-0.001
live-0.001
tto-0.001
Token.
Token ID13
Feature activation-3.18e-01

Pos logit contributions
iflower+0.050
pron+0.045
live+0.041
innie+0.039
wl+0.039
Neg logit contributions
 seen-0.034
keeping-0.034
hole-0.033
 stations-0.033
 contain-0.033
Token She
Token ID1375
Feature activation+9.94e-01

Pos logit contributions
 seen+0.008
hole+0.007
 stations+0.007
 towards+0.007
keeping+0.007
Neg logit contributions
iflower-0.012
pron-0.010
live-0.009
wl-0.009
ama-0.009
Token loved
Token ID6151
Feature activation+7.63e-01

Pos logit contributions
iflower+0.035
ama+0.032
pron+0.032
omm+0.029
innie+0.029
Neg logit contributions
 seen-0.022
 want-0.019
 towards-0.016
 like-0.016
ets-0.016
Token l
Token ID300
Feature activation-5.60e-02

Pos logit contributions
iflower+0.030
pron+0.028
live+0.027
ama+0.026
wl+0.026
Neg logit contributions
 seen-0.015
keeping-0.013
 stations-0.013
 umb-0.012
 crashing-0.012
Tokenoll
Token ID692
Feature activation+8.16e+00

Pos logit contributions
 seen+0.002
hole+0.001
 contain+0.001
keeping+0.001
vious+0.001
Neg logit contributions
iflower-0.002
pron-0.002
wl-0.001
lf-0.001
live-0.001
Tokenip
Token ID541
Feature activation+5.22e+00

Pos logit contributions
pron+0.134
iflower+0.133
live+0.098
Mon+0.090
wl+0.087
Neg logit contributions
 seen-0.300
ales-0.291
hole-0.283
ets-0.283
 officers-0.283
Tokenops
Token ID2840
Feature activation+1.21e+00

Pos logit contributions
pron+0.145
iflower+0.141
ama+0.107
innie+0.105
wl+0.103
Neg logit contributions
ales-0.138
ets-0.138
 officers-0.136
 seen-0.135
 argued-0.134
Token and
Token ID290
Feature activation+1.71e+00

Pos logit contributions
iflower+0.046
pron+0.044
live+0.039
wl+0.039
innie+0.039
Neg logit contributions
 seen-0.024
hole-0.022
keeping-0.021
 stations-0.021
 contain-0.021
Token always
Token ID1464
Feature activation+7.56e-01

Pos logit contributions
iflower+0.049
pron+0.043
live+0.039
ama+0.038
wl+0.037
Neg logit contributions
 seen-0.051
 stations-0.046
keeping-0.046
hole-0.046
 crashing-0.046
Token had
Token ID550
Feature activation+2.35e+00

Pos logit contributions
iflower+0.025
pron+0.022
live+0.020
ama+0.020
innie+0.019
Neg logit contributions
 seen-0.020
hole-0.018
keeping-0.018
 contain-0.017
 stations-0.017
Token I
Token ID314
Feature activation-7.07e-01

Pos logit contributions
 seen+0.025
 contain+0.023
hole+0.023
 stations+0.023
keeping+0.022
Neg logit contributions
iflower-0.039
pron-0.036
live-0.032
wl-0.031
innie-0.031
Token love
Token ID1842
Feature activation+5.95e-01

Pos logit contributions
 seen+0.014
hole+0.013
keeping+0.013
 stations+0.012
 contain+0.012
Neg logit contributions
iflower-0.027
pron-0.026
live-0.023
innie-0.022
ama-0.022
Token Car
Token ID1879
Feature activation+1.97e+00

Pos logit contributions
iflower+0.020
pron+0.019
live+0.017
wl+0.016
innie+0.016
Neg logit contributions
 seen-0.015
keeping-0.014
 stations-0.014
 crashing-0.013
 umb-0.013
Token-
Token ID12
Feature activation+1.93e+00

Pos logit contributions
iflower+0.061
pron+0.056
live+0.049
wl+0.047
innie+0.046
Neg logit contributions
 seen-0.054
hole-0.051
 officers-0.050
keeping-0.050
 stations-0.050
TokenD
Token ID35
Feature activation+2.20e-01

Pos logit contributions
iflower+0.062
pron+0.056
live+0.050
ama+0.048
innie+0.048
Neg logit contributions
 seen-0.051
keeping-0.048
 stations-0.047
 officers-0.047
hole-0.046
Tokenoll
Token ID692
Feature activation+8.07e+00

Pos logit contributions
iflower+0.006
pron+0.006
live+0.005
innie+0.005
ama+0.005
Neg logit contributions
 seen-0.007
hole-0.006
keeping-0.006
 stations-0.006
 contain-0.006
Token.
Token ID13
Feature activation+6.74e-01

Pos logit contributions
iflower+0.157
pron+0.137
live+0.115
wl+0.104
innie+0.103
Neg logit contributions
 seen-0.303
hole-0.288
 officers-0.284
keeping-0.284
ales-0.280
Token And
Token ID843
Feature activation-5.22e-01

Pos logit contributions
iflower+0.022
pron+0.020
live+0.018
innie+0.017
ama+0.017
Neg logit contributions
 seen-0.017
hole-0.016
keeping-0.016
 stations-0.016
 contain-0.015
Token I
Token ID314
Feature activation-4.54e-01

Pos logit contributions
 seen+0.012
 contain+0.011
hole+0.011
keeping+0.011
 stations+0.011
Neg logit contributions
iflower-0.019
pron-0.017
live-0.016
wl-0.015
innie-0.015
Token love
Token ID1842
Feature activation+7.13e-01

Pos logit contributions
 seen+0.009
hole+0.008
keeping+0.008
 stations+0.008
 contain+0.007
Neg logit contributions
iflower-0.018
pron-0.017
live-0.015
innie-0.014
ama-0.014
Token our
Token ID674
Feature activation+1.48e+00

Pos logit contributions
iflower+0.024
pron+0.024
live+0.021
wl+0.020
ama+0.019
Neg logit contributions
 seen-0.017
keeping-0.016
 officers-0.016
 stations-0.015
 crashing-0.015
Token friends
Token ID2460
Feature activation-7.19e-01

Pos logit contributions
iflower+0.066
pron+0.059
live+0.049
innie+0.046
wl+0.045
Neg logit contributions
 seen-0.096
keeping-0.091
hole-0.091
 stations-0.090
 officers-0.088
Token and
Token ID290
Feature activation+1.20e+00

Pos logit contributions
 brake+0.019
hole+0.019
 seen+0.018
 brakes+0.018
ales+0.018
Neg logit contributions
iflower-0.021
pron-0.018
live-0.017
wl-0.016
ama-0.016
Token shared
Token ID4888
Feature activation+3.31e+00

Pos logit contributions
iflower+0.038
pron+0.035
live+0.031
ama+0.030
innie+0.030
Neg logit contributions
 seen-0.032
keeping-0.030
 stations-0.029
hole-0.029
 officers-0.028
Token their
Token ID511
Feature activation+7.84e-01

Pos logit contributions
iflower+0.124
pron+0.119
wl+0.108
ama+0.108
live+0.106
Neg logit contributions
 seen-0.074
keeping-0.064
 stations-0.063
 towards-0.060
 umb-0.060
Token l
Token ID300
Feature activation-1.03e+00

Pos logit contributions
iflower+0.027
pron+0.026
wl+0.024
ama+0.023
live+0.022
Neg logit contributions
 seen-0.018
keeping-0.015
 ways-0.015
 umb-0.015
 want-0.015
Tokenoll
Token ID692
Feature activation+8.05e+00

Pos logit contributions
 seen+0.029
ets+0.027
hole+0.027
 contain+0.026
keeping+0.026
Neg logit contributions
iflower-0.030
pron-0.029
wl-0.024
ama-0.023
innie-0.023
Tokenip
Token ID541
Feature activation+5.23e+00

Pos logit contributions
pron+0.184
iflower+0.168
Mon+0.149
wl+0.141
Rich+0.140
Neg logit contributions
 seen-0.228
ales-0.227
ets-0.217
keeping-0.217
ly-0.214
Tokenops
Token ID2840
Feature activation+1.48e+00

Pos logit contributions
pron+0.140
iflower+0.134
innie+0.111
ama+0.109
wl+0.106
Neg logit contributions
keeping-0.153
ets-0.143
ales-0.142
 seen-0.142
 officers-0.137
Token every
Token ID790
Feature activation+1.51e-01

Pos logit contributions
iflower+0.046
pron+0.043
live+0.038
innie+0.038
wl+0.037
Neg logit contributions
 seen-0.039
keeping-0.037
hole-0.037
 contain-0.036
 stations-0.035
Token time
Token ID640
Feature activation-7.23e-01

Pos logit contributions
iflower+0.003
pron+0.003
live+0.003
wl+0.002
ama+0.002
Neg logit contributions
 seen-0.006
keeping-0.005
hole-0.005
 towards-0.005
 contain-0.005
Token they
Token ID484
Feature activation-2.97e+00

Pos logit contributions
 seen+0.018
keeping+0.017
hole+0.017
 stations+0.017
 crashing+0.017
Neg logit contributions
iflower-0.024
pron-0.022
live-0.020
innie-0.019
wl-0.019
Token Tess
Token ID39412
Feature activation+8.76e-01

Pos logit contributions
 seen+0.001
hole+0.001
keeping+0.001
 stations+0.001
 contain+0.001
Neg logit contributions
iflower-0.001
pron-0.001
live-0.001
wl-0.001
innie-0.001
Token.
Token ID13
Feature activation-4.41e-01

Pos logit contributions
iflower+0.029
pron+0.026
live+0.023
innie+0.023
wl+0.022
Neg logit contributions
 brakes-0.021
 brake-0.021
 seen-0.021
keeping-0.020
 crashing-0.020
Token She
Token ID1375
Feature activation+9.07e-01

Pos logit contributions
 seen+0.013
hole+0.013
keeping+0.013
 stations+0.012
 contain+0.012
Neg logit contributions
iflower-0.012
pron-0.011
live-0.010
innie-0.009
ama-0.009
Token loved
Token ID6151
Feature activation+6.54e-01

Pos logit contributions
iflower+0.032
pron+0.030
ama+0.029
innie+0.026
omm+0.026
Neg logit contributions
 seen-0.021
 want-0.018
 like-0.016
 towards-0.015
 argued-0.015
Token l
Token ID300
Feature activation-1.99e-01

Pos logit contributions
iflower+0.025
pron+0.023
live+0.022
ama+0.021
wl+0.021
Neg logit contributions
 seen-0.013
keeping-0.012
 stations-0.012
 crashing-0.012
 umb-0.011
Tokenoll
Token ID692
Feature activation+7.98e+00

Pos logit contributions
 seen+0.006
hole+0.006
 contain+0.006
keeping+0.005
ets+0.005
Neg logit contributions
iflower-0.006
pron-0.006
live-0.004
wl-0.004
innie-0.004
Tokenip
Token ID541
Feature activation+5.18e+00

Pos logit contributions
pron+0.130
iflower+0.127
live+0.092
Mon+0.081
innie+0.077
Neg logit contributions
 seen-0.306
ales-0.293
hole-0.286
ets-0.284
 officers-0.284
Tokenops
Token ID2840
Feature activation+1.08e+00

Pos logit contributions
pron+0.136
iflower+0.134
innie+0.098
ama+0.097
live+0.095
Neg logit contributions
 seen-0.150
ets-0.145
ales-0.145
keeping-0.145
 officers-0.145
Token.
Token ID13
Feature activation-9.10e-01

Pos logit contributions
iflower+0.041
pron+0.038
live+0.034
innie+0.034
wl+0.033
Neg logit contributions
 seen-0.023
hole-0.021
keeping-0.020
 stations-0.020
 contain-0.019
Token One
Token ID1881
Feature activation-3.24e-01

Pos logit contributions
 seen+0.025
hole+0.024
 stations+0.024
keeping+0.024
 officers+0.023
Neg logit contributions
iflower-0.028
pron-0.026
live-0.022
wl-0.021
tto-0.021
Token day
Token ID1110
Feature activation-9.13e-01

Pos logit contributions
 seen+0.010
 want+0.009
 towards+0.009
 eye+0.008
 contain+0.008
Neg logit contributions
iflower-0.009
pron-0.008
wl-0.007
innie-0.007
live-0.007
Token as
Token ID355
Feature activation-3.55e-01

Pos logit contributions
 seen+0.024
keeping+0.022
hole+0.022
 stations+0.022
ales+0.021
Neg logit contributions
iflower-0.035
pron-0.033
live-0.029
ama-0.028
innie-0.028
Token she
Token ID673
Feature activation-1.60e+00

Pos logit contributions
 seen+0.012
keeping+0.011
 crashing+0.011
 stations+0.011
 contain+0.011
Neg logit contributions
iflower-0.009
pron-0.008
live-0.007
wl-0.007
innie-0.007
Token held
Token ID2714
Feature activation+2.52e+00

Pos logit contributions
 seen+0.039
hole+0.035
keeping+0.034
 stations+0.033
 officers+0.033
Neg logit contributions
iflower-0.055
pron-0.051
live-0.045
ama-0.044
innie-0.043
Token the
Token ID262
Feature activation+2.69e-01

Pos logit contributions
iflower+0.084
pron+0.076
live+0.072
wl+0.070
tto+0.065
Neg logit contributions
 seen-0.062
hole-0.059
keeping-0.058
 towards-0.058
 toward-0.056
Token l
Token ID300
Feature activation-7.59e-01

Pos logit contributions
iflower+0.008
pron+0.007
live+0.006
wl+0.006
innie+0.006
Neg logit contributions
 seen-0.008
hole-0.008
 brake-0.007
keeping-0.007
 officers-0.007
Tokenoll
Token ID692
Feature activation+7.91e+00

Pos logit contributions
 seen+0.023
hole+0.022
 contain+0.021
keeping+0.021
ets+0.020
Neg logit contributions
iflower-0.021
pron-0.019
live-0.016
wl-0.016
innie-0.015
Tokenipop
Token ID42800
Feature activation+3.72e+00

Pos logit contributions
iflower+0.208
pron+0.199
live+0.188
wl+0.178
irl+0.176
Neg logit contributions
 seen-0.199
hole-0.198
ly-0.192
 permission-0.192
ously-0.188
Token in
Token ID287
Feature activation-2.69e-01

Pos logit contributions
iflower+0.120
pron+0.111
live+0.099
innie+0.094
wl+0.093
Neg logit contributions
 seen-0.097
hole-0.092
keeping-0.091
 stations-0.089
 contain-0.086
Token her
Token ID607
Feature activation-6.38e-01

Pos logit contributions
hole+0.010
 stations+0.010
 seen+0.010
ales+0.010
keeping+0.009
Neg logit contributions
iflower-0.005
pron-0.005
innie-0.004
live-0.004
ama-0.004
Token hand
Token ID1021
Feature activation+7.55e-02

Pos logit contributions
 seen+0.011
hole+0.010
keeping+0.010
 stations+0.009
 officers+0.009
Neg logit contributions
iflower-0.026
pron-0.024
live-0.023
wl-0.022
innie-0.022
Token.
Token ID13
Feature activation-3.52e-01

Pos logit contributions
iflower+0.002
pron+0.002
ama+0.002
live+0.002
innie+0.002
Neg logit contributions
 stations-0.002
 seen-0.002
keeping-0.002
 brake-0.002
 brakes-0.002
Token and
Token ID290
Feature activation+5.95e-01

Pos logit contributions
 seen+0.024
hole+0.024
 stations+0.024
keeping+0.023
ales+0.023
Neg logit contributions
iflower-0.022
pron-0.020
live-0.017
ama-0.017
innie-0.016
Token they
Token ID484
Feature activation-2.00e+00

Pos logit contributions
iflower+0.019
pron+0.017
live+0.015
innie+0.015
wl+0.015
Neg logit contributions
 seen-0.016
keeping-0.015
 stations-0.015
 officers-0.015
hole-0.015
Token enjoyed
Token ID8359
Feature activation+1.94e+00

Pos logit contributions
 seen+0.050
keeping+0.046
 contain+0.045
 stations+0.045
 officers+0.045
Neg logit contributions
iflower-0.068
pron-0.062
live-0.055
ama-0.054
innie-0.054
Token the
Token ID262
Feature activation+8.02e-01

Pos logit contributions
iflower+0.067
pron+0.061
live+0.056
wl+0.055
ama+0.054
Neg logit contributions
 seen-0.048
keeping-0.046
 crashing-0.043
 stations-0.043
 contain-0.041
Token l
Token ID300
Feature activation-5.17e-01

Pos logit contributions
iflower+0.026
pron+0.023
wl+0.022
ama+0.021
live+0.020
Neg logit contributions
 seen-0.021
 contain-0.020
 want-0.019
 brake-0.018
 gasp-0.018
Tokenoll
Token ID692
Feature activation+7.87e+00

Pos logit contributions
 seen+0.016
hole+0.015
keeping+0.015
 contain+0.015
ets+0.014
Neg logit contributions
iflower-0.014
pron-0.013
innie-0.011
wl-0.010
live-0.010
Tokenipop
Token ID42800
Feature activation+3.88e+00

Pos logit contributions
iflower+0.208
pron+0.206
wl+0.180
Mon+0.178
live+0.175
Neg logit contributions
 seen-0.203
ales-0.189
ets-0.182
ly-0.182
ried-0.182
Token together
Token ID1978
Feature activation-1.42e+00

Pos logit contributions
iflower+0.126
pron+0.116
live+0.103
ama+0.098
innie+0.098
Neg logit contributions
 seen-0.101
hole-0.095
keeping-0.094
 stations-0.093
 crashing-0.089
Token.
Token ID13
Feature activation-6.59e-01

Pos logit contributions
 seen+0.033
 stations+0.032
hole+0.032
keeping+0.031
 brake+0.031
Neg logit contributions
iflower-0.048
pron-0.045
live-0.040
ama-0.037
innie-0.037
Token The
Token ID383
Feature activation+5.80e-01

Pos logit contributions
 seen+0.015
keeping+0.014
hole+0.014
 stations+0.014
 officers+0.013
Neg logit contributions
iflower-0.024
pron-0.022
live-0.020
wl-0.019
innie-0.019
Token guard
Token ID4860
Feature activation-3.21e-01

Pos logit contributions
iflower+0.020
pron+0.018
live+0.016
innie+0.016
ama+0.015
Neg logit contributions
 seen-0.014
hole-0.014
keeping-0.014
 stations-0.013
 contain-0.013
Token happy
Token ID3772
Feature activation+8.80e-01

Pos logit contributions
iflower+0.066
pron+0.061
live+0.051
ama+0.048
wl+0.047
Neg logit contributions
hole-0.069
 seen-0.068
 stations-0.067
keeping-0.067
ales-0.065
Token.
Token ID13
Feature activation+2.72e-01

Pos logit contributions
iflower+0.027
pron+0.024
live+0.021
ama+0.020
wl+0.019
Neg logit contributions
hole-0.024
 seen-0.023
 brake-0.022
 stations-0.022
 contain-0.022
Token And
Token ID843
Feature activation-1.43e+00

Pos logit contributions
iflower+0.010
pron+0.009
live+0.008
wl+0.007
innie+0.007
Neg logit contributions
 seen-0.006
hole-0.006
keeping-0.006
 stations-0.006
 crashing-0.006
Token giving
Token ID3501
Feature activation+1.30e+00

Pos logit contributions
 seen+0.039
hole+0.038
keeping+0.037
 stations+0.037
 officers+0.035
Neg logit contributions
iflower-0.044
pron-0.040
live-0.035
ama-0.034
innie-0.033
Token l
Token ID300
Feature activation-5.72e-01

Pos logit contributions
iflower+0.043
pron+0.041
live+0.035
wl+0.034
innie+0.034
Neg logit contributions
 seen-0.034
keeping-0.032
 officers-0.030
 towards-0.030
 stations-0.030
Tokenoll
Token ID692
Feature activation+7.87e+00

Pos logit contributions
 seen+0.018
hole+0.017
keeping+0.017
 contain+0.016
 crashing+0.016
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
innie-0.011
wl-0.011
Tokenip
Token ID541
Feature activation+5.15e+00

Pos logit contributions
iflower+0.155
pron+0.148
live+0.122
wl+0.113
innie+0.108
Neg logit contributions
 seen-0.277
ets-0.263
ales-0.262
 permission-0.259
ried-0.256
Tokenops
Token ID2840
Feature activation+7.21e-01

Pos logit contributions
iflower+0.125
pron+0.120
live+0.095
innie+0.094
wl+0.093
Neg logit contributions
 seen-0.162
keeping-0.160
ets-0.155
 officers-0.153
ales-0.152
Token to
Token ID284
Feature activation+1.14e+00

Pos logit contributions
iflower+0.026
pron+0.024
live+0.021
ama+0.020
innie+0.020
Neg logit contributions
 seen-0.016
hole-0.015
 stations-0.015
keeping-0.015
 brake-0.014
Token children
Token ID1751
Feature activation-1.21e+00

Pos logit contributions
iflower+0.034
pron+0.031
live+0.028
wl+0.026
innie+0.026
Neg logit contributions
 seen-0.033
 stations-0.030
keeping-0.030
hole-0.029
 contain-0.029
Token makes
Token ID1838
Feature activation+9.44e-01

Pos logit contributions
 seen+0.027
hole+0.025
keeping+0.025
 stations+0.025
 brake+0.024
Neg logit contributions
iflower-0.043
pron-0.040
live-0.036
ama-0.034
innie-0.034
Token generous
Token ID14431
Feature activation-2.02e-01

Pos logit contributions
iflower+0.046
pron+0.041
ama+0.037
wl+0.037
live+0.037
Neg logit contributions
 seen-0.037
 stations-0.031
 towards-0.030
 crashing-0.030
keeping-0.030
Token and
Token ID290
Feature activation+2.83e+00

Pos logit contributions
 seen+0.005
hole+0.005
 stations+0.005
keeping+0.005
 brake+0.005
Neg logit contributions
iflower-0.006
pron-0.006
live-0.005
ama-0.005
innie-0.005
Token share
Token ID2648
Feature activation+3.76e+00

Pos logit contributions
iflower+0.091
pron+0.083
live+0.074
innie+0.072
ama+0.071
Neg logit contributions
 seen-0.077
keeping-0.071
 contain-0.068
 stations-0.068
hole-0.067
Token her
Token ID607
Feature activation+3.96e-01

Pos logit contributions
iflower+0.117
pron+0.110
ama+0.098
live+0.098
wl+0.097
Neg logit contributions
 seen-0.104
keeping-0.097
 stations-0.093
 towards-0.092
 officers-0.091
Token l
Token ID300
Feature activation-6.96e-01

Pos logit contributions
iflower+0.012
pron+0.011
live+0.011
wl+0.011
ama+0.010
Neg logit contributions
 seen-0.010
keeping-0.009
 permission-0.009
 want-0.009
 ways-0.009
Tokenoll
Token ID692
Feature activation+7.83e+00

Pos logit contributions
 seen+0.020
hole+0.018
ets+0.018
keeping+0.018
 contain+0.018
Neg logit contributions
iflower-0.020
pron-0.019
wl-0.016
innie-0.015
live-0.015
Tokenip
Token ID541
Feature activation+5.10e+00

Pos logit contributions
pron+0.198
iflower+0.183
Mon+0.169
Ros+0.164
irl+0.162
Neg logit contributions
 seen-0.197
ales-0.197
ly-0.191
owing-0.191
hole-0.189
Tokenops
Token ID2840
Feature activation+1.01e+00

Pos logit contributions
pron+0.136
iflower+0.130
innie+0.105
ama+0.103
wl+0.103
Neg logit contributions
keeping-0.144
ets-0.137
ales-0.134
 officers-0.131
 seen-0.131
Token with
Token ID351
Feature activation+1.55e+00

Pos logit contributions
iflower+0.032
pron+0.030
live+0.027
innie+0.027
wl+0.026
Neg logit contributions
 seen-0.026
keeping-0.025
hole-0.025
 contain-0.024
 stations-0.023
Token others
Token ID1854
Feature activation-3.18e-01

Pos logit contributions
iflower+0.046
pron+0.043
live+0.037
ama+0.036
tto+0.036
Neg logit contributions
 seen-0.045
keeping-0.041
 stations-0.040
 officers-0.040
hole-0.039
Token,
Token ID11
Feature activation+3.65e-01

Pos logit contributions
 seen+0.006
hole+0.006
keeping+0.006
 stations+0.006
 officers+0.005
Neg logit contributions
iflower-0.012
pron-0.012
live-0.010
innie-0.010
ama-0.010
Token because
Token ID780
Feature activation-5.04e-01

Pos logit contributions
keeping+0.048
 seen+0.046
 stations+0.046
 contain+0.044
ales+0.044
Neg logit contributions
iflower-0.047
pron-0.041
ama-0.036
innie-0.035
live-0.035
Token she
Token ID673
Feature activation+5.53e-02

Pos logit contributions
 seen+0.015
keeping+0.013
 crashing+0.013
 stations+0.013
 footsteps+0.013
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
wl-0.012
innie-0.012
Token loved
Token ID6151
Feature activation+5.38e-01

Pos logit contributions
iflower+0.002
pron+0.002
ama+0.002
live+0.002
innie+0.001
Neg logit contributions
 seen-0.001
 argued-0.001
hole-0.001
 want-0.001
keeping-0.001
Token her
Token ID607
Feature activation+1.20e+00

Pos logit contributions
iflower+0.018
pron+0.016
live+0.016
ama+0.016
wl+0.016
Neg logit contributions
 seen-0.013
keeping-0.013
 stations-0.013
 umb-0.012
 crashing-0.012
Token l
Token ID300
Feature activation-4.04e-01

Pos logit contributions
iflower+0.037
live+0.031
wl+0.031
pron+0.030
tto+0.029
Neg logit contributions
 seen-0.035
hole-0.031
 officers-0.030
 umb-0.030
 stations-0.029
Tokenoll
Token ID692
Feature activation+7.82e+00

Pos logit contributions
 seen+0.012
hole+0.011
ets+0.010
vious+0.010
keeping+0.010
Neg logit contributions
iflower-0.011
pron-0.011
wl-0.009
live-0.009
innie-0.009
Tokenip
Token ID541
Feature activation+5.09e+00

Pos logit contributions
iflower+0.165
pron+0.161
Mon+0.146
Rich+0.143
irl+0.139
Neg logit contributions
 seen-0.233
hole-0.219
ly-0.212
owing-0.209
 permission-0.206
Tokenops
Token ID2840
Feature activation+1.26e+00

Pos logit contributions
iflower+0.135
pron+0.133
ama+0.109
wl+0.107
innie+0.105
Neg logit contributions
 seen-0.132
keeping-0.132
ets-0.130
hole-0.126
 argued-0.123
Token so
Token ID523
Feature activation+2.15e+00

Pos logit contributions
iflower+0.045
pron+0.042
live+0.037
ama+0.036
innie+0.035
Neg logit contributions
 seen-0.028
 stations-0.027
keeping-0.027
hole-0.026
 officers-0.025
Token much
Token ID881
Feature activation+2.37e-01

Pos logit contributions
iflower+0.061
pron+0.058
live+0.049
ama+0.046
wl+0.046
Neg logit contributions
 seen-0.064
keeping-0.057
hole-0.057
 stations-0.056
 crashing-0.056
Token.
Token ID13
Feature activation-2.32e-01

Pos logit contributions
iflower+0.008
pron+0.007
live+0.007
innie+0.006
ama+0.006
Neg logit contributions
 seen-0.006
 stations-0.006
hole-0.005
keeping-0.005
 contain-0.005
Token Lily
Token ID20037
Feature activation+3.22e-01

Pos logit contributions
iflower+0.074
pron+0.064
wl+0.059
live+0.057
tto+0.056
Neg logit contributions
 seen-0.056
 officers-0.052
keeping-0.052
 crashing-0.051
 umb-0.051
Token happily
Token ID18177
Feature activation-5.19e-01

Pos logit contributions
iflower+0.012
pron+0.010
wl+0.010
live+0.009
ama+0.009
Neg logit contributions
 seen-0.007
hole-0.007
keeping-0.006
 argued-0.006
 officers-0.006
Token took
Token ID1718
Feature activation+3.41e-01

Pos logit contributions
 seen+0.014
hole+0.013
keeping+0.013
 stations+0.013
 crashing+0.012
Neg logit contributions
iflower-0.017
pron-0.015
live-0.014
innie-0.013
ama-0.013
Token the
Token ID262
Feature activation+4.18e-01

Pos logit contributions
iflower+0.013
pron+0.012
wl+0.011
live+0.011
tto+0.010
Neg logit contributions
 seen-0.008
keeping-0.007
 towards-0.007
 officers-0.006
 permission-0.006
Token l
Token ID300
Feature activation-8.94e-01

Pos logit contributions
iflower+0.013
wl+0.011
ama+0.011
live+0.011
pron+0.011
Neg logit contributions
 seen-0.011
hole-0.010
 contain-0.010
 towards-0.010
 brake-0.010
Tokenoll
Token ID692
Feature activation+7.82e+00

Pos logit contributions
 seen+0.027
hole+0.025
ets+0.024
 contain+0.024
keeping+0.023
Neg logit contributions
iflower-0.025
pron-0.023
wl-0.020
live-0.019
ama-0.019
Tokenipop
Token ID42800
Feature activation+2.85e+00

Pos logit contributions
iflower+0.220
pron+0.209
live+0.193
wl+0.188
irl+0.184
Neg logit contributions
 seen-0.194
hole-0.182
ly-0.182
 permission-0.177
ales-0.173
Token and
Token ID290
Feature activation+7.69e-01

Pos logit contributions
iflower+0.103
pron+0.096
live+0.086
innie+0.082
ama+0.082
Neg logit contributions
 seen-0.064
hole-0.059
keeping-0.059
 stations-0.059
 crashing-0.055
Token started
Token ID2067
Feature activation+5.20e-02

Pos logit contributions
iflower+0.027
pron+0.024
wl+0.023
live+0.022
ama+0.022
Neg logit contributions
 seen-0.019
keeping-0.017
 officers-0.016
 argued-0.016
hole-0.016
Token licking
Token ID43390
Feature activation+1.07e+00

Pos logit contributions
iflower+0.002
pron+0.002
live+0.002
wl+0.002
ama+0.002
Neg logit contributions
 seen-0.001
keeping-0.001
 crashing-0.001
 towards-0.001
hole-0.001
Token it
Token ID340
Feature activation+1.83e+00

Pos logit contributions
iflower+0.040
pron+0.037
live+0.037
wl+0.036
tto+0.034
Neg logit contributions
 towards-0.021
 seen-0.021
 crashing-0.020
hole-0.020
ked-0.020
Token the
Token ID262
Feature activation-1.53e+00

Pos logit contributions
 seen+0.006
hole+0.006
keeping+0.006
 stations+0.006
 contain+0.006
Neg logit contributions
iflower-0.006
pron-0.005
live-0.005
innie-0.004
wl-0.004
Token sky
Token ID6766
Feature activation-1.53e+00

Pos logit contributions
 seen+0.044
hole+0.041
keeping+0.041
 stations+0.041
 crashing+0.039
Neg logit contributions
iflower-0.046
pron-0.042
live-0.037
innie-0.035
ama-0.035
Token,
Token ID11
Feature activation-4.75e-01

Pos logit contributions
keeping+0.045
 seen+0.044
 stations+0.043
hole+0.042
 officers+0.042
Neg logit contributions
iflower-0.044
pron-0.039
live-0.033
wl-0.032
ama-0.032
Token the
Token ID262
Feature activation-1.14e+00

Pos logit contributions
keeping+0.013
 seen+0.013
hole+0.013
 officers+0.013
ales+0.013
Neg logit contributions
iflower-0.014
pron-0.012
live-0.010
wl-0.010
ama-0.009
Token dragon
Token ID10441
Feature activation-5.88e-01

Pos logit contributions
 seen+0.034
hole+0.033
keeping+0.033
 stations+0.032
ales+0.031
Neg logit contributions
iflower-0.033
pron-0.030
live-0.024
ama-0.023
wl-0.023
Token ruled
Token ID8879
Feature activation-7.96e+00

Pos logit contributions
 seen+0.018
keeping+0.017
 stations+0.017
hole+0.017
 brake+0.016
Neg logit contributions
iflower-0.016
pron-0.015
live-0.013
wl-0.012
ama-0.012
Token over
Token ID625
Feature activation+4.55e-01

Pos logit contributions
 seen+0.201
hole+0.189
keeping+0.188
 stations+0.186
 contain+0.178
Neg logit contributions
iflower-0.263
pron-0.242
live-0.216
innie-0.206
ama-0.206
Token the
Token ID262
Feature activation-2.01e+00

Pos logit contributions
iflower+0.012
pron+0.010
live+0.009
ama+0.008
wl+0.008
Neg logit contributions
hole-0.014
 seen-0.013
 contain-0.013
keeping-0.013
 stations-0.013
Token land
Token ID1956
Feature activation-1.53e-01

Pos logit contributions
 seen+0.065
hole+0.062
keeping+0.062
 stations+0.061
ales+0.059
Neg logit contributions
iflower-0.053
pron-0.047
live-0.039
ama-0.037
wl-0.037
Token with
Token ID351
Feature activation-1.45e-02

Pos logit contributions
 seen+0.004
keeping+0.004
 stations+0.004
hole+0.004
 brake+0.004
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
ama-0.004
wl-0.004
Token might
Token ID1244
Feature activation-3.73e+00

Pos logit contributions
hole+0.000
 seen+0.000
keeping+0.000
 stations+0.000
 contain+0.000
Neg logit contributions
iflower-0.000
pron-0.000
live-0.000
wl-0.000
ama-0.000
Token grass
Token ID8701
Feature activation-6.16e-01

Pos logit contributions
 seen+0.032
keeping+0.030
hole+0.030
 stations+0.029
 contain+0.029
Neg logit contributions
iflower-0.032
pron-0.029
live-0.025
ama-0.024
innie-0.024
Token around
Token ID1088
Feature activation+1.19e-01

Pos logit contributions
 seen+0.022
keeping+0.021
hole+0.021
 stations+0.021
 contain+0.021
Neg logit contributions
iflower-0.014
pron-0.012
live-0.010
ama-0.009
wl-0.009
Token the
Token ID262
Feature activation-2.22e+00

Pos logit contributions
iflower+0.003
pron+0.003
live+0.003
ama+0.003
wl+0.002
Neg logit contributions
hole-0.003
 seen-0.003
 stations-0.003
keeping-0.003
 contain-0.003
Token hill
Token ID12788
Feature activation-2.24e-01

Pos logit contributions
 seen+0.060
keeping+0.056
hole+0.055
 stations+0.054
 contain+0.054
Neg logit contributions
iflower-0.072
pron-0.065
live-0.056
ama-0.055
innie-0.054
Token was
Token ID373
Feature activation-1.82e+00

Pos logit contributions
 seen+0.008
keeping+0.007
 stations+0.007
hole+0.007
 brake+0.007
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
wl-0.004
ama-0.004
Token wit
Token ID20868
Feature activation-7.76e+00

Pos logit contributions
 seen+0.053
keeping+0.051
hole+0.051
 stations+0.050
 officers+0.049
Neg logit contributions
iflower-0.053
pron-0.049
live-0.042
ama-0.039
innie-0.039
Tokenhered
Token ID6083
Feature activation-6.98e-02

Pos logit contributions
 seen+0.212
keeping+0.207
 umb+0.194
 contain+0.192
 stations+0.188
Neg logit contributions
iflower-0.232
pron-0.214
live-0.187
wl-0.180
ama-0.180
Token and
Token ID290
Feature activation+2.27e-01

Pos logit contributions
 stations+0.002
 brakes+0.002
 unison+0.002
hole+0.002
 footsteps+0.002
Neg logit contributions
iflower-0.001
tto-0.001
live-0.001
P-0.001
in-0.001
Token yellow
Token ID7872
Feature activation+6.56e-01

Pos logit contributions
iflower+0.005
pron+0.005
live+0.004
wl+0.004
lf+0.004
Neg logit contributions
hole-0.007
 stations-0.007
keeping-0.007
ales-0.007
ets-0.007
Token.
Token ID13
Feature activation-1.23e+00

Pos logit contributions
iflower+0.014
time+0.011
pron+0.011
 mint+0.010
ama+0.010
Neg logit contributions
 seen-0.021
 contain-0.020
 announcements-0.020
hole-0.020
keeping-0.020
Token He
Token ID679
Feature activation+4.23e-01

Pos logit contributions
 seen+0.033
hole+0.031
keeping+0.031
 stations+0.031
 officers+0.030
Neg logit contributions
iflower-0.039
pron-0.036
live-0.032
wl-0.029
innie-0.029
Token,
Token ID11
Feature activation-6.73e-01

Pos logit contributions
keeping+0.057
 seen+0.056
 stations+0.055
 brake+0.053
 officers+0.053
Neg logit contributions
iflower-0.066
pron-0.062
ama-0.052
live-0.051
innie-0.050
Token feeling
Token ID4203
Feature activation-4.97e-01

Pos logit contributions
keeping+0.018
 seen+0.018
 stations+0.018
hole+0.017
 officers+0.017
Neg logit contributions
iflower-0.020
pron-0.019
live-0.016
ama-0.016
innie-0.015
Token the
Token ID262
Feature activation+2.95e-01

Pos logit contributions
 seen+0.014
keeping+0.014
hole+0.014
 stations+0.014
ales+0.014
Neg logit contributions
iflower-0.014
pron-0.013
live-0.011
ama-0.010
wl-0.010
Token peace
Token ID4167
Feature activation-5.11e-01

Pos logit contributions
iflower+0.008
pron+0.008
ama+0.006
live+0.006
wl+0.006
Neg logit contributions
keeping-0.009
 seen-0.008
ales-0.008
hole-0.008
 stations-0.008
Token and
Token ID290
Feature activation+1.32e+00

Pos logit contributions
 seen+0.014
keeping+0.014
hole+0.013
 stations+0.013
ales+0.013
Neg logit contributions
iflower-0.016
pron-0.014
live-0.012
ama-0.012
wl-0.011
Token tranqu
Token ID31882
Feature activation-7.71e+00

Pos logit contributions
iflower+0.038
pron+0.034
live+0.028
ama+0.027
wl+0.026
Neg logit contributions
keeping-0.037
ales-0.037
hole-0.036
 seen-0.036
 officers-0.036
Tokenility
Token ID879
Feature activation+1.46e+00

Pos logit contributions
 brakes+0.162
 officers+0.152
 buses+0.151
 stations+0.150
 levers+0.149
Neg logit contributions
pron-0.260
iflower-0.259
live-0.233
ama-0.231
innie-0.225
Token of
Token ID286
Feature activation+3.58e-02

Pos logit contributions
iflower+0.046
pron+0.042
live+0.038
ama+0.036
innie+0.035
Neg logit contributions
 seen-0.039
keeping-0.037
hole-0.036
 stations-0.036
 brake-0.035
Token the
Token ID262
Feature activation-5.51e-01

Pos logit contributions
iflower+0.001
pron+0.001
live+0.001
ama+0.001
innie+0.001
Neg logit contributions
 seen-0.001
hole-0.001
keeping-0.001
 officers-0.001
 stations-0.001
Token night
Token ID1755
Feature activation-8.22e-01

Pos logit contributions
 seen+0.017
keeping+0.016
hole+0.016
 stations+0.016
 officers+0.015
Neg logit contributions
iflower-0.016
pron-0.014
live-0.012
ama-0.011
wl-0.011
Token,
Token ID11
Feature activation-3.45e-01

Pos logit contributions
 seen+0.022
keeping+0.022
hole+0.021
 stations+0.021
 brake+0.021
Neg logit contributions
iflower-0.025
pron-0.023
live-0.020
ama-0.019
wl-0.019
Token the
Token ID262
Feature activation-1.50e+00

Pos logit contributions
hole+0.042
ets+0.041
 seen+0.040
 officers+0.040
 stations+0.039
Neg logit contributions
iflower-0.030
pron-0.026
live-0.020
innie-0.019
lf-0.018
Token garden
Token ID11376
Feature activation+2.77e-01

Pos logit contributions
 seen+0.044
keeping+0.042
hole+0.042
 stations+0.041
ales+0.040
Neg logit contributions
iflower-0.044
pron-0.040
live-0.034
ama-0.033
innie-0.032
Token.
Token ID13
Feature activation-4.54e-01

Pos logit contributions
iflower+0.008
pron+0.007
ama+0.006
live+0.006
on+0.005
Neg logit contributions
 brake-0.008
 brakes-0.008
 stations-0.007
 officers-0.007
Run-0.007
Token A
Token ID317
Feature activation-1.22e+00

Pos logit contributions
 seen+0.014
keeping+0.013
hole+0.013
 stations+0.013
 officers+0.012
Neg logit contributions
iflower-0.013
pron-0.012
live-0.010
innie-0.010
wl-0.010
Token subtle
Token ID11800
Feature activation-1.59e+00

Pos logit contributions
 seen+0.038
hole+0.036
 crashing+0.036
 stations+0.035
 officers+0.035
Neg logit contributions
iflower-0.032
pron-0.028
live-0.026
innie-0.025
wl-0.024
Token tranqu
Token ID31882
Feature activation-7.69e+00

Pos logit contributions
 seen+0.049
keeping+0.048
hole+0.048
 stations+0.047
ales+0.045
Neg logit contributions
iflower-0.043
pron-0.040
live-0.033
ama-0.032
wl-0.031
Tokenility
Token ID879
Feature activation-6.20e-02

Pos logit contributions
 brakes+0.162
 officers+0.156
keeping+0.153
 levers+0.152
 stations+0.152
Neg logit contributions
pron-0.261
iflower-0.259
live-0.235
ama-0.220
innie-0.219
Token filled
Token ID5901
Feature activation-7.07e-02

Pos logit contributions
 seen+0.002
keeping+0.002
 brake+0.002
hole+0.002
 stations+0.002
Neg logit contributions
iflower-0.002
pron-0.002
live-0.001
ama-0.001
wl-0.001
Token the
Token ID262
Feature activation-9.70e-01

Pos logit contributions
 seen+0.002
keeping+0.002
 stations+0.002
hole+0.002
 contain+0.002
Neg logit contributions
iflower-0.002
pron-0.002
live-0.002
innie-0.002
ama-0.002
Token air
Token ID1633
Feature activation-5.23e-01

Pos logit contributions
 seen+0.030
hole+0.029
keeping+0.028
ales+0.028
 contain+0.028
Neg logit contributions
iflower-0.027
pron-0.024
live-0.019
ama-0.019
wl-0.018
Token and
Token ID290
Feature activation-8.62e-02

Pos logit contributions
 seen+0.014
keeping+0.014
 stations+0.013
 brake+0.013
 officers+0.013
Neg logit contributions
iflower-0.016
pron-0.015
live-0.012
ama-0.012
wl-0.012
Token,
Token ID11
Feature activation+2.95e-02

Pos logit contributions
iflower+0.010
pron+0.009
live+0.008
ama+0.008
innie+0.007
Neg logit contributions
 seen-0.011
keeping-0.011
 stations-0.011
hole-0.011
 contain-0.010
Token it
Token ID340
Feature activation+1.32e+00

Pos logit contributions
iflower+0.001
pron+0.001
live+0.000
ama+0.000
innie+0.000
Neg logit contributions
 seen-0.001
hole-0.001
keeping-0.001
 stations-0.001
 contain-0.001
Token kept
Token ID4030
Feature activation+8.24e-01

Pos logit contributions
iflower+0.046
pron+0.043
live+0.038
innie+0.037
ama+0.037
Neg logit contributions
 seen-0.032
hole-0.029
keeping-0.028
 stations-0.028
 officers-0.028
Token on
Token ID319
Feature activation+6.92e-01

Pos logit contributions
iflower+0.028
pron+0.026
live+0.024
innie+0.023
ama+0.023
Neg logit contributions
 seen-0.020
keeping-0.019
hole-0.019
 crashing-0.018
 stations-0.018
Token flying
Token ID7348
Feature activation-1.73e+00

Pos logit contributions
iflower+0.022
pron+0.020
live+0.018
innie+0.017
ama+0.017
Neg logit contributions
 seen-0.018
hole-0.017
keeping-0.017
 stations-0.017
 contain-0.016
Token err
Token ID11454
Feature activation-7.60e+00

Pos logit contributions
 seen+0.050
 supervision+0.050
 officers+0.049
keeping+0.048
 brake+0.048
Neg logit contributions
iflower-0.042
pron-0.035
live-0.033
wl-0.032
P-0.030
Tokenatically
Token ID4142
Feature activation-1.65e+00

Pos logit contributions
 seen+0.159
hole+0.149
 stations+0.143
keeping+0.142
 crashing+0.140
Neg logit contributions
iflower-0.281
pron-0.256
live-0.239
innie-0.231
wl-0.227
Token with
Token ID351
Feature activation-5.25e-01

Pos logit contributions
Run+0.046
 seen+0.046
 brake+0.045
 stations+0.044
ales+0.044
Neg logit contributions
iflower-0.042
pron-0.035
ama-0.028
ip-0.028
 best-0.027
Token inexplicable
Token ID49602
Feature activation-9.67e-01

Pos logit contributions
hole+0.017
 stations+0.016
keeping+0.016
 contain+0.016
 brake+0.016
Neg logit contributions
iflower-0.012
pron-0.010
live-0.008
wl-0.008
ama-0.008
Token speed
Token ID2866
Feature activation-2.02e+00

Pos logit contributions
 stations+0.027
keeping+0.026
Run+0.026
ales+0.025
rified+0.025
Neg logit contributions
iflower-0.028
pron-0.024
 story-0.020
 special-0.020
 important-0.020
Token.
Token ID13
Feature activation-9.76e-01

Pos logit contributions
 seen+0.056
 horror+0.055
 footsteps+0.055
 stations+0.054
Run+0.053
Neg logit contributions
iflower-0.051
pron-0.046
live-0.040
ip-0.040
wl-0.039
Token saw
Token ID2497
Feature activation+7.29e-01

Pos logit contributions
iflower+0.039
pron+0.037
innie+0.034
live+0.033
ama+0.032
Neg logit contributions
 seen-0.024
hole-0.021
 towards-0.020
 stations-0.020
 crashing-0.019
Token that
Token ID326
Feature activation+2.41e+00

Pos logit contributions
iflower+0.021
pron+0.018
live+0.016
wl+0.015
ama+0.015
Neg logit contributions
 seen-0.022
hole-0.021
keeping-0.021
 stations-0.021
 contain-0.021
Token the
Token ID262
Feature activation-1.22e+00

Pos logit contributions
iflower+0.072
pron+0.067
live+0.060
innie+0.058
ama+0.056
Neg logit contributions
 seen-0.070
hole-0.065
 stations-0.064
keeping-0.064
 crashing-0.062
Token land
Token ID1956
Feature activation-4.18e-01

Pos logit contributions
keeping+0.037
 seen+0.037
hole+0.035
ales+0.035
ously+0.035
Neg logit contributions
iflower-0.033
pron-0.029
live-0.024
ama-0.024
wl-0.023
Token was
Token ID373
Feature activation-1.19e+00

Pos logit contributions
 seen+0.014
keeping+0.014
 stations+0.014
hole+0.013
ales+0.013
Neg logit contributions
iflower-0.010
pron-0.009
live-0.008
ama-0.007
wl-0.007
Token ruled
Token ID8879
Feature activation-7.46e+00

Pos logit contributions
 seen+0.036
keeping+0.035
hole+0.035
 stations+0.035
ales+0.034
Neg logit contributions
iflower-0.033
pron-0.030
live-0.025
ama-0.023
wl-0.023
Token by
Token ID416
Feature activation-1.98e+00

Pos logit contributions
 seen+0.184
keeping+0.175
 stations+0.174
hole+0.173
 officers+0.168
Neg logit contributions
iflower-0.247
pron-0.228
live-0.203
ama-0.192
wl-0.191
Token a
Token ID257
Feature activation-2.20e+00

Pos logit contributions
hole+0.065
 officers+0.064
 contain+0.063
 brake+0.063
ales+0.061
Neg logit contributions
iflower-0.046
pron-0.033
live-0.030
wl-0.029
 important-0.028
Token mean
Token ID1612
Feature activation-1.46e+00

Pos logit contributions
 seen+0.067
hole+0.063
 crashing+0.063
 stations+0.062
keeping+0.060
Neg logit contributions
iflower-0.061
pron-0.055
live-0.051
innie-0.049
ama-0.047
Token monster
Token ID9234
Feature activation-2.17e+00

Pos logit contributions
 seen+0.044
keeping+0.042
hole+0.042
 stations+0.042
 contain+0.041
Neg logit contributions
iflower-0.041
pron-0.036
live-0.031
ama-0.030
wl-0.030
Token.
Token ID13
Feature activation-5.99e-01

Pos logit contributions
 stations+0.056
 seen+0.054
keeping+0.054
ales+0.054
 contain+0.053
Neg logit contributions
iflower-0.067
pron-0.058
live-0.052
lf-0.051
wl-0.050
Token big
Token ID1263
Feature activation-5.52e-01

Pos logit contributions
 seen+0.029
 crashing+0.028
 officers+0.027
 stations+0.027
hole+0.027
Neg logit contributions
iflower-0.026
pron-0.023
live-0.022
innie-0.021
tto-0.020
Token,
Token ID11
Feature activation-8.68e-01

Pos logit contributions
 seen+0.016
keeping+0.016
hole+0.016
 stations+0.015
 contain+0.015
Neg logit contributions
iflower-0.016
pron-0.015
live-0.012
ama-0.012
innie-0.012
Token scary
Token ID14343
Feature activation-1.34e+00

Pos logit contributions
 seen+0.029
keeping+0.028
hole+0.028
 stations+0.027
ales+0.026
Neg logit contributions
iflower-0.022
pron-0.020
live-0.016
innie-0.015
wl-0.015
Token lion
Token ID18744
Feature activation-1.66e+00

Pos logit contributions
 seen+0.039
hole+0.037
keeping+0.037
 stations+0.036
 contain+0.036
Neg logit contributions
iflower-0.040
pron-0.036
live-0.031
wl-0.030
ama-0.030
Token who
Token ID508
Feature activation+6.06e-01

Pos logit contributions
 seen+0.043
 contain+0.043
keeping+0.043
 stations+0.042
ales+0.042
Neg logit contributions
iflower-0.052
pron-0.045
live-0.041
wl-0.039
lf-0.038
Token ruled
Token ID8879
Feature activation-7.27e+00

Pos logit contributions
iflower+0.018
pron+0.017
live+0.014
innie+0.014
ama+0.014
Neg logit contributions
 seen-0.017
hole-0.016
keeping-0.016
 stations-0.016
 officers-0.016
Token over
Token ID625
Feature activation+8.17e-01

Pos logit contributions
 seen+0.185
hole+0.176
 stations+0.173
keeping+0.171
 contain+0.170
Neg logit contributions
iflower-0.233
pron-0.215
live-0.194
innie-0.183
wl-0.183
Token his
Token ID465
Feature activation-1.42e+00

Pos logit contributions
iflower+0.019
pron+0.016
live+0.014
wl+0.014
P+0.013
Neg logit contributions
hole-0.026
 contain-0.025
 spine-0.025
 brake-0.024
 stations-0.023
Token nation
Token ID3277
Feature activation-1.70e+00

Pos logit contributions
 seen+0.044
hole+0.042
keeping+0.042
 stations+0.041
 contain+0.040
Neg logit contributions
iflower-0.039
pron-0.034
live-0.030
ama-0.029
wl-0.028
Token of
Token ID286
Feature activation-1.17e+00

Pos logit contributions
 seen+0.041
 stations+0.040
keeping+0.040
 brake+0.039
ales+0.039
Neg logit contributions
iflower-0.054
pron-0.049
live-0.043
ama-0.041
wl-0.041
Token animals
Token ID4695
Feature activation-2.94e+00

Pos logit contributions
hole+0.033
 seen+0.033
 contain+0.032
keeping+0.032
 stations+0.031
Neg logit contributions
iflower-0.034
pron-0.029
live-0.025
wl-0.024
innie-0.024
Token and
Token ID290
Feature activation+1.02e+00

Pos logit contributions
ked+0.039
ales+0.038
 argued+0.038
 officers+0.037
 stations+0.037
Neg logit contributions
iflower-0.036
pron-0.030
live-0.027
ama-0.027
tto-0.025
Token mysterious
Token ID11428
Feature activation-7.31e-02

Pos logit contributions
iflower+0.023
pron+0.020
live+0.018
ama+0.016
tto+0.016
Neg logit contributions
hole-0.034
 stations-0.033
ales-0.033
keeping-0.032
 officers-0.032
Token land
Token ID1956
Feature activation-1.14e-01

Pos logit contributions
ales+0.002
hole+0.002
 seen+0.002
 stations+0.002
ked+0.002
Neg logit contributions
iflower-0.002
pron-0.002
ama-0.001
live-0.001
 story-0.001
Token that
Token ID326
Feature activation+2.30e+00

Pos logit contributions
 stations+0.003
 seen+0.003
hole+0.003
keeping+0.003
 brake+0.003
Neg logit contributions
iflower-0.003
pron-0.003
live-0.003
ama-0.003
wl-0.003
Token was
Token ID373
Feature activation-5.70e-01

Pos logit contributions
iflower+0.062
pron+0.056
live+0.048
innie+0.046
ama+0.045
Neg logit contributions
 seen-0.072
hole-0.069
keeping-0.069
 stations-0.068
 contain-0.066
Token ruled
Token ID8879
Feature activation-7.24e+00

Pos logit contributions
keeping+0.016
hole+0.016
 seen+0.016
 stations+0.016
ales+0.015
Neg logit contributions
iflower-0.016
pron-0.015
live-0.013
ama-0.012
tto-0.012
Token by
Token ID416
Feature activation-1.48e+00

Pos logit contributions
 seen+0.184
 stations+0.174
hole+0.173
keeping+0.173
 contain+0.168
Neg logit contributions
iflower-0.233
pron-0.216
live-0.192
ama-0.183
wl-0.181
Token a
Token ID257
Feature activation-1.51e+00

Pos logit contributions
hole+0.046
 seen+0.044
ales+0.043
 officers+0.043
 brake+0.042
Neg logit contributions
iflower-0.039
pron-0.031
live-0.028
ama-0.027
wl-0.026
Token powerful
Token ID3665
Feature activation-1.29e+00

Pos logit contributions
 seen+0.047
hole+0.045
keeping+0.045
 stations+0.044
 contain+0.043
Neg logit contributions
iflower-0.041
pron-0.038
live-0.032
ama-0.030
innie-0.030
Token witch
Token ID16365
Feature activation-2.63e+00

Pos logit contributions
 seen+0.038
keeping+0.038
hole+0.038
 stations+0.037
ales+0.036
Neg logit contributions
iflower-0.036
pron-0.033
ama-0.027
live-0.026
innie-0.025
Token.
Token ID13
Feature activation-3.84e-01

Pos logit contributions
 stations+0.063
 seen+0.063
ales+0.063
keeping+0.062
 brake+0.061
Neg logit contributions
iflower-0.085
pron-0.077
live-0.067
wl-0.065
ama-0.063
Token he
Token ID339
Feature activation-7.54e-01

Pos logit contributions
iflower+0.064
pron+0.054
live+0.045
ama+0.044
wl+0.043
Neg logit contributions
 seen-0.084
hole-0.083
keeping-0.081
ales-0.080
ets-0.079
Token was
Token ID373
Feature activation+1.30e-01

Pos logit contributions
keeping+0.025
 seen+0.025
 stations+0.024
hole+0.024
 crashing+0.024
Neg logit contributions
iflower-0.018
pron-0.016
live-0.014
wl-0.013
tto-0.013
Token in
Token ID287
Feature activation+3.74e-01

Pos logit contributions
iflower+0.004
pron+0.003
live+0.003
tto+0.003
ama+0.003
Neg logit contributions
keeping-0.004
hole-0.004
 stations-0.004
ales-0.004
 seen-0.003
Token a
Token ID257
Feature activation-7.08e-01

Pos logit contributions
iflower+0.010
pron+0.009
live+0.007
innie+0.007
ama+0.007
Neg logit contributions
ales-0.011
hole-0.011
ked-0.011
ets-0.011
ried-0.011
Token kingdom
Token ID13239
Feature activation-2.03e-01

Pos logit contributions
keeping+0.022
 seen+0.022
hole+0.021
ales+0.021
 stations+0.021
Neg logit contributions
iflower-0.019
pron-0.018
live-0.014
ama-0.013
wl-0.013
Token ruled
Token ID8879
Feature activation-7.23e+00

Pos logit contributions
 stations+0.005
ales+0.005
 umb+0.005
keeping+0.005
ked+0.005
Neg logit contributions
iflower-0.006
pron-0.005
live-0.004
ama-0.004
lf-0.004
Token by
Token ID416
Feature activation-1.32e+00

Pos logit contributions
 seen+0.185
hole+0.174
 stations+0.174
keeping+0.173
 contain+0.169
Neg logit contributions
iflower-0.233
pron-0.214
live-0.191
innie-0.181
ama-0.181
Token a
Token ID257
Feature activation-1.03e+00

Pos logit contributions
hole+0.041
 seen+0.040
ales+0.039
ked+0.039
 officers+0.039
Neg logit contributions
iflower-0.034
pron-0.027
live-0.024
wl-0.023
ama-0.023
Token wicked
Token ID20589
Feature activation-2.87e+00

Pos logit contributions
 seen+0.032
hole+0.031
keeping+0.031
 stations+0.030
 contain+0.029
Neg logit contributions
iflower-0.028
pron-0.025
live-0.021
ama-0.020
wl-0.020
Token wizard
Token ID18731
Feature activation-1.50e+00

Pos logit contributions
 seen+0.085
keeping+0.083
hole+0.082
 stations+0.079
 contain+0.077
Neg logit contributions
iflower-0.082
pron-0.073
ama-0.063
live-0.061
wl-0.061
Token.
Token ID13
Feature activation-1.26e-01

Pos logit contributions
 stations+0.039
 umb+0.038
 brake+0.037
ales+0.037
 footsteps+0.037
Neg logit contributions
iflower-0.045
pron-0.041
live-0.034
wl-0.034
ama-0.034
Token They
Token ID1119
Feature activation-1.10e+00

Pos logit contributions
 seen+0.010
hole+0.009
keeping+0.009
 stations+0.009
 contain+0.009
Neg logit contributions
iflower-0.015
pron-0.014
live-0.013
innie-0.012
ama-0.012
Token said
Token ID531
Feature activation+8.89e-01

Pos logit contributions
 seen+0.028
hole+0.026
keeping+0.026
 stations+0.026
 contain+0.025
Neg logit contributions
iflower-0.037
pron-0.033
live-0.030
innie-0.029
ama-0.029
Token,
Token ID11
Feature activation-5.01e-01

Pos logit contributions
iflower+0.031
pron+0.028
live+0.026
wl+0.025
innie+0.024
Neg logit contributions
 seen-0.022
 stations-0.020
 crashing-0.020
 contain-0.020
 towards-0.020
Token "
Token ID366
Feature activation-6.90e-01

Pos logit contributions
 crashing+0.013
 towards+0.013
 seen+0.012
 stations+0.012
 toward+0.012
Neg logit contributions
iflower-0.017
live-0.014
pron-0.014
tto-0.013
wl-0.013
TokenL
Token ID43
Feature activation-1.09e+00

Pos logit contributions
 seen+0.020
 stations+0.020
keeping+0.019
hole+0.019
 contain+0.018
Neg logit contributions
iflower-0.020
pron-0.018
live-0.016
ama-0.016
innie-0.015
Tokenadies
Token ID50192
Feature activation-7.08e+00

Pos logit contributions
 seen+0.033
keeping+0.032
hole+0.031
 contain+0.031
 crashing+0.031
Neg logit contributions
iflower-0.030
pron-0.027
live-0.024
innie-0.023
ama-0.022
Token and
Token ID290
Feature activation-1.72e+00

Pos logit contributions
 brake+0.226
Run+0.225
 contain+0.223
 rubble+0.222
hole+0.217
Neg logit contributions
iflower-0.126
ama-0.126
pron-0.125
omm-0.113
 best-0.104
Token gentlemen
Token ID28527
Feature activation-2.67e+00

Pos logit contributions
 seen+0.032
 stations+0.029
hole+0.028
 officers+0.028
keeping+0.027
Neg logit contributions
iflower-0.070
pron-0.065
live-0.061
tto-0.058
innie-0.058
Token,
Token ID11
Feature activation-1.06e+00

Pos logit contributions
 seen+0.084
keeping+0.081
hole+0.080
 stations+0.080
 contain+0.077
Neg logit contributions
iflower-0.069
pron-0.063
live-0.054
ama-0.052
innie-0.051
Token welcome
Token ID7062
Feature activation+1.56e+00

Pos logit contributions
 seen+0.027
hole+0.026
keeping+0.026
 stations+0.025
ales+0.024
Neg logit contributions
iflower-0.034
pron-0.032
live-0.027
innie-0.027
ama-0.027
Token to
Token ID284
Feature activation+1.33e+00

Pos logit contributions
iflower+0.042
pron+0.037
ama+0.034
live+0.032
P+0.032
Neg logit contributions
hole-0.047
 seen-0.045
 brake-0.044
 stations-0.044
 contain-0.043
Token night
Token ID1755
Feature activation-1.08e+00

Pos logit contributions
iflower+0.005
pron+0.004
live+0.004
innie+0.004
wl+0.004
Neg logit contributions
 seen-0.006
 crashing-0.005
 stations-0.005
 towards-0.005
 zone-0.005
Token,
Token ID11
Feature activation+4.62e-01

Pos logit contributions
 seen+0.028
 towards+0.025
 crashing+0.024
keeping+0.024
hole+0.024
Neg logit contributions
iflower-0.037
pron-0.033
live-0.031
innie-0.030
wl-0.030
Token the
Token ID262
Feature activation-5.82e-02

Pos logit contributions
iflower+0.016
pron+0.015
live+0.014
innie+0.013
tto+0.013
Neg logit contributions
 seen-0.011
 crashing-0.010
 stations-0.010
 towards-0.010
 officers-0.010
Token wild
Token ID4295
Feature activation-3.86e-01

Pos logit contributions
 brake+0.001
 footsteps+0.001
 officers+0.001
 brakes+0.001
 seen+0.001
Neg logit contributions
iflower-0.002
pron-0.002
live-0.002
innie-0.002
tto-0.002
Token cat
Token ID3797
Feature activation-4.99e-01

Pos logit contributions
 seen+0.010
 towards+0.009
hole+0.009
 stations+0.009
 officers+0.009
Neg logit contributions
iflower-0.013
pron-0.012
live-0.010
innie-0.010
P-0.010
Token lur
Token ID20605
Feature activation-7.05e+00

Pos logit contributions
 seen+0.014
 stations+0.013
hole+0.013
keeping+0.012
 officers+0.012
Neg logit contributions
iflower-0.015
pron-0.014
innie-0.012
live-0.012
ama-0.012
Tokenked
Token ID9091
Feature activation-6.53e-01

Pos logit contributions
keeping+0.113
 stations+0.103
hole+0.103
 rubble+0.101
 umb+0.101
Neg logit contributions
iflower-0.243
pron-0.235
P-0.226
ip-0.221
wl-0.217
Token through
Token ID832
Feature activation-8.85e-01

Pos logit contributions
 seen+0.016
hole+0.016
keeping+0.016
 stations+0.015
 contain+0.015
Neg logit contributions
iflower-0.022
pron-0.020
live-0.017
ama-0.017
wl-0.017
Token the
Token ID262
Feature activation-1.66e+00

Pos logit contributions
 seen+0.029
hole+0.029
keeping+0.028
 stations+0.027
 contain+0.027
Neg logit contributions
iflower-0.022
pron-0.020
live-0.016
innie-0.016
ama-0.015
Token woods
Token ID16479
Feature activation-1.17e+00

Pos logit contributions
 seen+0.043
hole+0.041
keeping+0.041
 stations+0.040
 contain+0.039
Neg logit contributions
iflower-0.054
pron-0.050
live-0.044
innie-0.042
ama-0.042
Token,
Token ID11
Feature activation-1.25e+00

Pos logit contributions
 seen+0.033
keeping+0.033
 brake+0.032
 umb+0.031
hole+0.031
Neg logit contributions
iflower-0.032
pron-0.028
wl-0.022
live-0.022
ama-0.022
Token.
Token ID13
Feature activation-4.43e-01

Pos logit contributions
 seen+0.035
keeping+0.032
 umb+0.032
 risks+0.032
 wanna+0.031
Neg logit contributions
iflower-0.032
pron-0.030
ama-0.026
on-0.026
lf-0.024
Token He
Token ID679
Feature activation+3.64e-01

Pos logit contributions
 seen+0.012
hole+0.011
keeping+0.011
 stations+0.011
 officers+0.010
Neg logit contributions
iflower-0.014
pron-0.013
live-0.012
innie-0.011
ama-0.011
Token says
Token ID1139
Feature activation+1.34e+00

Pos logit contributions
iflower+0.013
pron+0.012
live+0.011
innie+0.010
ama+0.010
Neg logit contributions
 seen-0.008
hole-0.008
keeping-0.008
 stations-0.008
 contain-0.007
Token "
Token ID366
Feature activation-9.98e-01

Pos logit contributions
iflower+0.043
pron+0.039
live+0.036
innie+0.034
wl+0.034
Neg logit contributions
 crashing-0.035
 seen-0.035
 towards-0.033
 stations-0.032
 officers-0.032
TokenL
Token ID43
Feature activation-1.58e+00

Pos logit contributions
 seen+0.026
hole+0.025
keeping+0.025
 stations+0.025
 contain+0.023
Neg logit contributions
iflower-0.033
pron-0.030
live-0.027
ama-0.026
innie-0.026
Tokenadies
Token ID50192
Feature activation-7.05e+00

Pos logit contributions
keeping+0.053
hole+0.053
 seen+0.053
 stations+0.052
 brakes+0.052
Neg logit contributions
iflower-0.035
live-0.030
pron-0.030
innie-0.026
wl-0.026
Token and
Token ID290
Feature activation-1.52e+00

Pos logit contributions
 rubble+0.247
Run+0.239
 contain+0.238
 brake+0.234
 levers+0.230
Neg logit contributions
ama-0.106
iflower-0.097
omm-0.090
 mom-0.090
pron-0.090
Token gentlemen
Token ID28527
Feature activation-2.61e+00

Pos logit contributions
 seen+0.032
hole+0.029
keeping+0.029
 stations+0.029
 officers+0.027
Neg logit contributions
iflower-0.057
pron-0.053
live-0.048
innie-0.046
ama-0.046
Token,
Token ID11
Feature activation-1.01e+00

Pos logit contributions
 seen+0.083
keeping+0.080
hole+0.078
 stations+0.078
 contain+0.076
Neg logit contributions
iflower-0.068
pron-0.062
live-0.053
ama-0.052
innie-0.050
Token boys
Token ID6510
Feature activation-4.47e-01

Pos logit contributions
 seen+0.028
keeping+0.027
hole+0.027
ales+0.026
 stations+0.025
Neg logit contributions
iflower-0.030
pron-0.028
live-0.024
innie-0.024
ama-0.023
Token and
Token ID290
Feature activation-8.79e-01

Pos logit contributions
 seen+0.012
keeping+0.012
 stations+0.011
hole+0.011
 contain+0.011
Neg logit contributions
iflower-0.013
pron-0.012
live-0.011
ama-0.011
innie-0.010
Token a
Token ID257
Feature activation-1.20e+00

Pos logit contributions
iflower+0.038
pron+0.031
ama+0.025
 best+0.025
lf+0.024
Neg logit contributions
keeping-0.066
 seen-0.063
hole-0.063
 spine-0.062
 contain-0.062
Token sign
Token ID1051
Feature activation-2.10e+00

Pos logit contributions
 seen+0.035
hole+0.033
 stations+0.033
 crashing+0.033
 officers+0.032
Neg logit contributions
iflower-0.035
pron-0.031
live-0.029
innie-0.027
tto-0.027
Token that
Token ID326
Feature activation+1.10e+00

Pos logit contributions
 seen+0.066
keeping+0.065
hole+0.062
 stations+0.062
ales+0.061
Neg logit contributions
iflower-0.055
pron-0.050
live-0.042
ama-0.040
wl-0.039
Token says
Token ID1139
Feature activation+5.90e-02

Pos logit contributions
iflower+0.043
pron+0.040
live+0.037
innie+0.036
tto+0.035
Neg logit contributions
 seen-0.022
hole-0.020
 stations-0.020
keeping-0.019
 officers-0.019
Token "
Token ID366
Feature activation-1.83e+00

Pos logit contributions
iflower+0.001
pron+0.001
live+0.001
innie+0.001
wl+0.001
Neg logit contributions
 seen-0.002
hole-0.002
 stations-0.002
keeping-0.002
 crashing-0.002
Tokentem
Token ID11498
Feature activation-6.94e+00

Pos logit contributions
 seen+0.052
keeping+0.049
hole+0.049
 stations+0.048
 contain+0.046
Neg logit contributions
iflower-0.057
pron-0.052
live-0.046
ama-0.044
innie-0.044
Tokenple
Token ID1154
Feature activation+6.66e-01

Pos logit contributions
 seen+0.164
keeping+0.150
hole+0.150
 stations+0.146
 contain+0.144
Neg logit contributions
iflower-0.238
pron-0.217
live-0.198
innie-0.192
wl-0.191
Token".
Token ID1911
Feature activation-4.63e-01

Pos logit contributions
iflower+0.022
pron+0.021
live+0.018
innie+0.017
wl+0.017
Neg logit contributions
 seen-0.016
hole-0.016
keeping-0.015
 stations-0.015
 officers-0.015
Token They
Token ID1119
Feature activation-1.36e+00

Pos logit contributions
 seen+0.013
hole+0.012
keeping+0.012
 stations+0.012
 contain+0.012
Neg logit contributions
iflower-0.015
pron-0.013
live-0.012
wl-0.011
innie-0.011
Token ask
Token ID1265
Feature activation-1.15e+00

Pos logit contributions
keeping+0.036
 seen+0.032
hole+0.030
ously+0.030
rying+0.029
Neg logit contributions
iflower-0.039
pron-0.037
ama-0.033
wl-0.033
irl-0.033
Token mom
Token ID1995
Feature activation-2.24e+00

Pos logit contributions
 seen+0.031
 stations+0.029
hole+0.029
keeping+0.029
 officers+0.028
Neg logit contributions
iflower-0.037
pron-0.034
live-0.030
wl-0.029
innie-0.028
Token.
Token ID13
Feature activation-5.78e-01

Pos logit contributions
 stations+0.004
 seen+0.004
hole+0.004
ales+0.004
keeping+0.004
Neg logit contributions
iflower-0.003
pron-0.003
live-0.002
ama-0.002
innie-0.002
Token
Token ID198
Feature activation+2.22e-01

Pos logit contributions
 seen+0.017
keeping+0.016
hole+0.016
 towards+0.015
 stations+0.015
Neg logit contributions
iflower-0.017
pron-0.015
live-0.014
wl-0.013
tto-0.012
Token
Token ID198
Feature activation+3.86e-02

Pos logit contributions
iflower+0.007
pron+0.006
live+0.005
wl+0.005
ama+0.005
Neg logit contributions
 seen-0.006
keeping-0.006
 contain-0.005
 towards-0.005
hole-0.005
TokenLuckily
Token ID42332
Feature activation-2.71e-01

Pos logit contributions
iflower+0.001
pron+0.001
ama+0.001
live+0.001
oppy+0.001
Neg logit contributions
 seen-0.001
 stations-0.001
keeping-0.001
 contain-0.001
 crashing-0.001
Token Sarah
Token ID10490
Feature activation-4.85e-01

Pos logit contributions
 seen+0.007
hole+0.007
 stations+0.006
 officers+0.006
keeping+0.006
Neg logit contributions
iflower-0.009
pron-0.008
live-0.007
wl-0.007
innie-0.007
Token ret
Token ID1005
Feature activation-6.93e+00

Pos logit contributions
 seen+0.014
hole+0.013
 officers+0.013
keeping+0.013
 stations+0.012
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
innie-0.011
ama-0.011
Tokenuned
Token ID40881
Feature activation-1.04e+00

Pos logit contributions
 seen+0.129
 stations+0.128
 levers+0.125
 brakes+0.125
 crashing+0.124
Neg logit contributions
iflower-0.237
live-0.216
pron-0.213
innie-0.212
wl-0.212
Token home
Token ID1363
Feature activation+1.77e-01

Pos logit contributions
 seen+0.032
hole+0.031
keeping+0.030
 stations+0.030
 contain+0.030
Neg logit contributions
iflower-0.028
pron-0.025
live-0.022
ama-0.021
innie-0.021
Token safe
Token ID3338
Feature activation-4.84e-01

Pos logit contributions
iflower+0.005
pron+0.005
live+0.004
ama+0.004
innie+0.004
Neg logit contributions
 seen-0.005
hole-0.005
 stations-0.005
keeping-0.005
ales-0.005
Token and
Token ID290
Feature activation+1.13e+00

Pos logit contributions
 seen+0.013
hole+0.013
 stations+0.012
 contain+0.012
 brake+0.012
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
ama-0.012
innie-0.011
Token sound
Token ID2128
Feature activation-1.55e+00

Pos logit contributions
iflower+0.033
pron+0.030
live+0.026
innie+0.025
ama+0.025
Neg logit contributions
 seen-0.032
hole-0.032
keeping-0.031
 stations-0.031
ales-0.030
Token a
Token ID257
Feature activation-1.47e+00

Pos logit contributions
iflower+0.010
pron+0.009
live+0.008
innie+0.007
ama+0.007
Neg logit contributions
 seen-0.011
hole-0.010
keeping-0.010
 stations-0.010
 contain-0.010
Token patch
Token ID8529
Feature activation-1.48e-01

Pos logit contributions
 seen+0.046
hole+0.044
 crashing+0.043
 stations+0.043
keeping+0.043
Neg logit contributions
iflower-0.038
pron-0.034
live-0.031
innie-0.030
wl-0.029
Token of
Token ID286
Feature activation-1.03e+00

Pos logit contributions
 seen+0.004
 stations+0.004
keeping+0.004
 brake+0.004
hole+0.003
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
ama-0.004
wl-0.004
Token butter
Token ID9215
Feature activation+2.54e-01

Pos logit contributions
 seen+0.030
hole+0.030
keeping+0.029
 stations+0.029
 brake+0.028
Neg logit contributions
iflower-0.030
pron-0.027
live-0.023
innie-0.021
ama-0.021
Tokenc
Token ID66
Feature activation-8.66e-01

Pos logit contributions
iflower+0.008
pron+0.007
live+0.006
wl+0.006
ama+0.006
Neg logit contributions
 seen-0.007
 stations-0.007
keeping-0.007
hole-0.007
 brake-0.006
Tokenups
Token ID4739
Feature activation-6.90e+00

Pos logit contributions
 seen+0.026
 stations+0.025
hole+0.025
 brake+0.025
keeping+0.024
Neg logit contributions
iflower-0.025
pron-0.022
live-0.020
ama-0.019
wl-0.019
Token near
Token ID1474
Feature activation-2.28e+00

Pos logit contributions
 stations+0.187
 brake+0.186
 brakes+0.184
 footsteps+0.180
keeping+0.179
Neg logit contributions
iflower-0.195
pron-0.185
live-0.153
on-0.146
wl-0.142
Token the
Token ID262
Feature activation-2.84e+00

Pos logit contributions
 seen+0.070
hole+0.068
 stations+0.067
keeping+0.066
 contain+0.065
Neg logit contributions
iflower-0.061
pron-0.054
live-0.047
innie-0.045
ama-0.045
Token hedge
Token ID19859
Feature activation-2.28e+00

Pos logit contributions
 seen+0.082
hole+0.078
keeping+0.076
 stations+0.075
 contain+0.074
Neg logit contributions
iflower-0.085
pron-0.077
live-0.067
ama-0.063
wl-0.063
Token.
Token ID13
Feature activation-2.88e-01

Pos logit contributions
 seen+0.062
 brake+0.061
 brakes+0.059
 stations+0.057
 footsteps+0.056
Neg logit contributions
iflower-0.068
pron-0.061
live-0.056
wl-0.053
ama-0.051
Token Daisy
Token ID40355
Feature activation+2.39e-01

Pos logit contributions
 seen+0.008
keeping+0.008
hole+0.007
 stations+0.007
 officers+0.007
Neg logit contributions
iflower-0.009
pron-0.008
live-0.007
wl-0.007
tto-0.007
Token't
Token ID470
Feature activation-1.82e+00

Pos logit contributions
 seen+0.062
hole+0.061
 crashing+0.060
 brake+0.060
keeping+0.060
Neg logit contributions
iflower-0.050
pron-0.048
live-0.037
innie-0.036
wl-0.034
Token out
Token ID503
Feature activation-2.89e-01

Pos logit contributions
 stations+0.050
keeping+0.050
hole+0.048
 umb+0.047
 buses+0.046
Neg logit contributions
iflower-0.051
pron-0.043
live-0.040
ama-0.039
wl-0.039
Tokenrun
Token ID5143
Feature activation+3.76e-01

Pos logit contributions
ales+0.009
 stations+0.009
Run+0.008
keeping+0.008
 officers+0.008
Neg logit contributions
iflower-0.006
pron-0.005
 story-0.005
lf-0.005
on-0.005
Token the
Token ID262
Feature activation-1.42e+00

Pos logit contributions
iflower+0.011
pron+0.010
live+0.009
wl+0.009
innie+0.009
Neg logit contributions
 seen-0.010
keeping-0.010
hole-0.010
 stations-0.010
 contain-0.009
Token enormous
Token ID9812
Feature activation-1.98e-01

Pos logit contributions
keeping+0.046
hole+0.045
 seen+0.044
 contain+0.043
ried+0.043
Neg logit contributions
iflower-0.036
pron-0.031
 story-0.026
wl-0.026
 best-0.025
Token wolves
Token ID23214
Feature activation-6.86e+00

Pos logit contributions
keeping+0.007
 contain+0.007
hole+0.007
 seen+0.007
ried+0.006
Neg logit contributions
iflower-0.005
pron-0.004
wl-0.003
ama-0.003
live-0.003
Token.
Token ID13
Feature activation-7.74e-01

Pos logit contributions
 stations+0.197
ales+0.192
keeping+0.189
 umb+0.186
awks+0.186
Neg logit contributions
iflower-0.167
pron-0.145
on-0.128
 story-0.123
lf-0.123
Token Stru
Token ID34887
Feature activation-1.68e+00

Pos logit contributions
 seen+0.022
hole+0.021
 stations+0.021
keeping+0.021
 contain+0.021
Neg logit contributions
iflower-0.022
pron-0.020
live-0.018
ama-0.017
innie-0.017
Tokengg
Token ID1130
Feature activation+3.11e+00

Pos logit contributions
 seen+0.056
hole+0.054
keeping+0.053
 stations+0.053
 officers+0.051
Neg logit contributions
iflower-0.043
pron-0.037
live-0.033
wl-0.031
innie-0.031
Tokenling
Token ID1359
Feature activation-2.82e-01

Pos logit contributions
iflower+0.090
pron+0.079
live+0.071
ama+0.068
wl+0.066
Neg logit contributions
 seen-0.093
 stations-0.088
hole-0.088
keeping-0.087
 contain-0.084
Token to
Token ID284
Feature activation+5.54e-01

Pos logit contributions
 seen+0.008
hole+0.008
keeping+0.008
 stations+0.008
 contain+0.007
Neg logit contributions
iflower-0.008
pron-0.007
live-0.006
ama-0.006
wl-0.006
Token and
Token ID290
Feature activation-6.07e-01

Pos logit contributions
 seen+0.062
 stations+0.058
keeping+0.057
 brake+0.057
hole+0.056
Neg logit contributions
iflower-0.054
pron-0.051
ama-0.044
live-0.043
wl-0.040
Token then
Token ID788
Feature activation+2.09e-01

Pos logit contributions
 seen+0.015
hole+0.014
keeping+0.014
 stations+0.013
 officers+0.013
Neg logit contributions
iflower-0.021
pron-0.019
live-0.017
innie-0.017
ama-0.017
Token a
Token ID257
Feature activation-2.18e+00

Pos logit contributions
iflower+0.007
pron+0.006
live+0.006
ama+0.005
innie+0.005
Neg logit contributions
 seen-0.005
hole-0.005
keeping-0.005
 stations-0.005
 contain-0.005
Token loud
Token ID7812
Feature activation-3.42e+00

Pos logit contributions
 seen+0.068
hole+0.065
keeping+0.065
 stations+0.064
 contain+0.061
Neg logit contributions
iflower-0.059
pron-0.054
live-0.046
innie-0.044
ama-0.044
Token "
Token ID366
Feature activation-1.85e+00

Pos logit contributions
 seen+0.088
keeping+0.082
hole+0.082
 stations+0.078
ales+0.076
Neg logit contributions
iflower-0.111
pron-0.107
live-0.088
ama-0.087
lf-0.087
TokenCR
Token ID9419
Feature activation-6.84e+00

Pos logit contributions
 seen+0.051
hole+0.049
keeping+0.049
 stations+0.048
 contain+0.046
Neg logit contributions
iflower-0.055
pron-0.051
live-0.045
innie-0.042
wl-0.042
TokenASH
Token ID11211
Feature activation-6.36e-01

Pos logit contributions
 contain+0.152
ales+0.133
 seen+0.133
 stations+0.132
ets+0.131
Neg logit contributions
iflower-0.225
pron-0.216
ama-0.207
innie-0.199
live-0.195
Token!"
Token ID2474
Feature activation+4.25e-01

Pos logit contributions
 seen+0.015
 stations+0.015
 contain+0.014
keeping+0.013
hole+0.013
Neg logit contributions
iflower-0.021
pron-0.020
ama-0.019
live-0.018
ip-0.017
Token echoed
Token ID22211
Feature activation-2.56e+00

Pos logit contributions
iflower+0.010
pron+0.009
live+0.007
ama+0.007
innie+0.007
Neg logit contributions
 seen-0.015
 stations-0.014
hole-0.014
keeping-0.014
ales-0.014
Token around
Token ID1088
Feature activation-3.94e-01

Pos logit contributions
 seen+0.069
keeping+0.068
hole+0.068
 stations+0.065
 supervision+0.064
Neg logit contributions
iflower-0.065
pron-0.064
ama-0.057
 mummy-0.052
lf-0.052
Token the
Token ID262
Feature activation-1.69e+00

Pos logit contributions
hole+0.014
 seen+0.014
ales+0.013
keeping+0.013
 contain+0.013
Neg logit contributions
iflower-0.008
pron-0.007
lf-0.005
ama-0.005
live-0.005
Token table
Token ID3084
Feature activation+6.47e-01

Pos logit contributions
 seen+0.057
keeping+0.054
hole+0.054
ales+0.053
 contain+0.052
Neg logit contributions
iflower-0.052
pron-0.049
ama-0.038
live-0.037
innie-0.036
Token,
Token ID11
Feature activation-1.13e+00

Pos logit contributions
iflower+0.018
pron+0.018
ama+0.015
live+0.014
lf+0.014
Neg logit contributions
 seen-0.017
 stations-0.017
 brake-0.016
keeping-0.016
 brakes-0.016
Token making
Token ID1642
Feature activation+5.02e-01

Pos logit contributions
keeping+0.030
 seen+0.029
ales+0.029
 stations+0.029
hole+0.029
Neg logit contributions
iflower-0.033
pron-0.033
live-0.025
ama-0.025
wl-0.024
Token the
Token ID262
Feature activation-7.36e-01

Pos logit contributions
iflower+0.014
pron+0.013
live+0.011
ama+0.011
innie+0.010
Neg logit contributions
 seen-0.015
hole-0.015
keeping-0.014
 contain-0.014
 stations-0.014
Token candle
Token ID26839
Feature activation+1.08e+00

Pos logit contributions
 seen+0.021
hole+0.020
keeping+0.020
ales+0.019
 stations+0.019
Neg logit contributions
iflower-0.022
pron-0.021
ama-0.016
live-0.016
innie-0.016
Token topple
Token ID49377
Feature activation-6.84e+00

Pos logit contributions
iflower+0.034
pron+0.032
ama+0.026
live+0.025
wl+0.024
Neg logit contributions
keeping-0.028
 seen-0.028
 stations-0.027
hole-0.026
 officers-0.026
Token over
Token ID625
Feature activation-1.21e+00

Pos logit contributions
 contain+0.205
 seen+0.203
awks+0.202
 unison+0.201
 supervision+0.201
Neg logit contributions
iflower-0.151
pron-0.140
live-0.118
ama-0.114
M-0.111
Token.
Token ID13
Feature activation-1.12e+00

Pos logit contributions
 contain+0.033
hole+0.032
ales+0.032
 officers+0.031
 seen+0.031
Neg logit contributions
iflower-0.035
pron-0.030
ama-0.026
live-0.024
innie-0.024
Token Everyone
Token ID11075
Feature activation-3.01e+00

Pos logit contributions
 seen+0.032
keeping+0.030
hole+0.030
 stations+0.029
 officers+0.028
Neg logit contributions
iflower-0.034
pron-0.031
live-0.027
wl-0.026
innie-0.025
Token quickly
Token ID2952
Feature activation-2.66e+00

Pos logit contributions
keeping+0.099
 seen+0.098
hole+0.096
ales+0.096
 stations+0.096
Neg logit contributions
iflower-0.072
pron-0.064
live-0.054
ama-0.051
wl-0.050
Token rushed
Token ID13999
Feature activation-4.51e+00

Pos logit contributions
hole+0.074
keeping+0.071
 spine+0.069
 stations+0.068
ales+0.068
Neg logit contributions
iflower-0.073
pron-0.068
live-0.059
lf-0.055
lly-0.054
Token became
Token ID2627
Feature activation+4.77e-01

Pos logit contributions
iflower+0.013
pron+0.012
live+0.010
wl+0.010
ama+0.010
Neg logit contributions
 seen-0.011
keeping-0.011
hole-0.011
 stations-0.011
 brake-0.011
Token good
Token ID922
Feature activation+9.07e-01

Pos logit contributions
iflower+0.013
pron+0.012
live+0.011
innie+0.010
wl+0.010
Neg logit contributions
 seen-0.015
 stations-0.015
 crashing-0.014
 officers-0.014
keeping-0.014
Token friends
Token ID2460
Feature activation-2.83e-01

Pos logit contributions
iflower+0.015
pron+0.011
ama+0.009
live+0.008
ip+0.007
Neg logit contributions
hole-0.038
 brake-0.037
 seen-0.036
 contain-0.035
 stations-0.035
Token.
Token ID13
Feature activation-5.76e-01

Pos logit contributions
 brake+0.009
ales+0.008
 stations+0.008
 seen+0.008
 brakes+0.008
Neg logit contributions
iflower-0.007
pron-0.006
live-0.005
ama-0.005
ip-0.005
Token They
Token ID1119
Feature activation-8.58e-01

Pos logit contributions
 seen+0.014
hole+0.013
keeping+0.013
 stations+0.013
 contain+0.012
Neg logit contributions
iflower-0.019
pron-0.018
live-0.016
innie-0.015
ama-0.015
Token ruled
Token ID8879
Feature activation-6.83e+00

Pos logit contributions
 seen+0.025
hole+0.024
keeping+0.024
 stations+0.023
 contain+0.022
Neg logit contributions
iflower-0.025
pron-0.023
live-0.020
innie-0.019
ama-0.019
Token the
Token ID262
Feature activation-1.07e+00

Pos logit contributions
 seen+0.177
hole+0.166
keeping+0.166
 stations+0.163
 contain+0.156
Neg logit contributions
iflower-0.223
pron-0.205
live-0.182
innie-0.174
ama-0.173
Token land
Token ID1956
Feature activation+3.81e-01

Pos logit contributions
 seen+0.032
 stations+0.030
hole+0.029
keeping+0.029
 officers+0.029
Neg logit contributions
iflower-0.031
pron-0.028
live-0.026
innie-0.024
wl-0.023
Token together
Token ID1978
Feature activation-1.19e+00

Pos logit contributions
iflower+0.012
pron+0.011
live+0.010
innie+0.009
ama+0.009
Neg logit contributions
 seen-0.010
hole-0.010
keeping-0.010
 stations-0.010
 contain-0.009
Token.
Token ID13
Feature activation-6.17e-01

Pos logit contributions
 seen+0.032
hole+0.030
keeping+0.030
 stations+0.029
 contain+0.028
Neg logit contributions
iflower-0.038
pron-0.035
live-0.031
innie-0.030
ama-0.029
Token They
Token ID1119
Feature activation-9.11e-01

Pos logit contributions
 seen+0.016
keeping+0.015
hole+0.015
 stations+0.015
 officers+0.014
Neg logit contributions
iflower-0.020
pron-0.019
live-0.016
innie-0.016
wl-0.016
Token admired
Token ID29382
Feature activation+1.14e+00

Pos logit contributions
 seen+0.002
keeping+0.002
hole+0.002
 stations+0.002
ales+0.002
Neg logit contributions
iflower-0.002
pron-0.002
live-0.001
ama-0.001
innie-0.001
Token the
Token ID262
Feature activation+3.39e-01

Pos logit contributions
iflower+0.030
pron+0.027
live+0.023
innie+0.022
ama+0.022
Neg logit contributions
 seen-0.036
hole-0.035
keeping-0.034
 stations-0.034
 contain-0.033
Token way
Token ID835
Feature activation-3.78e-01

Pos logit contributions
iflower+0.009
pron+0.008
live+0.006
ama+0.006
wl+0.006
Neg logit contributions
 seen-0.011
keeping-0.011
hole-0.011
 stations-0.010
 contain-0.010
Token it
Token ID340
Feature activation+1.22e+00

Pos logit contributions
keeping+0.011
 seen+0.011
 stations+0.011
ales+0.011
 brake+0.011
Neg logit contributions
iflower-0.009
pron-0.009
ama-0.008
live-0.007
wl-0.007
Token flew
Token ID13112
Feature activation-3.02e+00

Pos logit contributions
iflower+0.035
pron+0.032
live+0.028
wl+0.026
ama+0.026
Neg logit contributions
 seen-0.035
keeping-0.035
 stations-0.034
hole-0.034
 officers-0.033
Token grace
Token ID11542
Feature activation-6.82e+00

Pos logit contributions
 seen+0.080
keeping+0.080
 stations+0.078
 supervision+0.075
 brake+0.074
Neg logit contributions
iflower-0.085
pron-0.081
live-0.070
ama-0.068
P-0.068
Tokenfully
Token ID2759
Feature activation-1.75e+00

Pos logit contributions
 seen+0.175
 stations+0.170
 brakes+0.169
keeping+0.168
 brake+0.167
Neg logit contributions
iflower-0.203
pron-0.200
ama-0.167
live-0.164
ip-0.155
Token in
Token ID287
Feature activation-1.68e+00

Pos logit contributions
 stations+0.049
 seen+0.048
keeping+0.047
ales+0.046
 brake+0.046
Neg logit contributions
iflower-0.049
pron-0.044
ama-0.036
live-0.035
ip-0.034
Token the
Token ID262
Feature activation-1.51e+00

Pos logit contributions
ales+0.060
ked+0.060
 officers+0.059
ets+0.058
ried+0.058
Neg logit contributions
iflower-0.029
pron-0.027
M-0.021
P-0.020
innie-0.020
Token sky
Token ID6766
Feature activation-1.03e+00

Pos logit contributions
 seen+0.043
keeping+0.043
hole+0.041
ales+0.041
 contain+0.040
Neg logit contributions
iflower-0.043
pron-0.039
P-0.033
ama-0.032
live-0.032
Token and
Token ID290
Feature activation+1.53e-01

Pos logit contributions
 seen+0.028
keeping+0.027
 brake+0.027
 stations+0.027
 officers+0.027
Neg logit contributions
iflower-0.030
pron-0.028
live-0.024
ama-0.024
wl-0.022
Token They
Token ID1119
Feature activation-1.26e+00

Pos logit contributions
 seen+0.002
keeping+0.002
hole+0.002
 stations+0.002
 crashing+0.002
Neg logit contributions
iflower-0.003
pron-0.003
live-0.002
innie-0.002
wl-0.002
Token say
Token ID910
Feature activation+1.11e+00

Pos logit contributions
keeping+0.037
hole+0.033
 stations+0.033
 seen+0.031
ales+0.031
Neg logit contributions
iflower-0.033
pron-0.030
live-0.029
irl-0.028
oppy-0.027
Token,
Token ID11
Feature activation-7.33e-01

Pos logit contributions
iflower+0.039
pron+0.036
live+0.034
innie+0.032
wl+0.031
Neg logit contributions
 seen-0.026
 crashing-0.024
 officers-0.024
 stations-0.023
 footsteps-0.023
Token "
Token ID366
Feature activation-7.21e-01

Pos logit contributions
 seen+0.022
 crashing+0.022
hole+0.021
 stations+0.021
 towards+0.021
Neg logit contributions
iflower-0.021
pron-0.017
live-0.017
innie-0.015
tto-0.015
TokenWe
Token ID1135
Feature activation-2.51e+00

Pos logit contributions
 seen+0.020
keeping+0.018
hole+0.018
 stations+0.018
 brake+0.017
Neg logit contributions
iflower-0.023
pron-0.021
live-0.019
innie-0.018
ama-0.018
Token do
Token ID466
Feature activation+4.03e+00

Pos logit contributions
keeping+0.060
hole+0.058
 stations+0.057
 crashing+0.054
 spine+0.053
Neg logit contributions
iflower-0.081
pron-0.076
live-0.073
wl-0.067
tto-0.067
Token not
Token ID407
Feature activation-1.47e+00

Pos logit contributions
iflower+0.092
pron+0.082
live+0.068
innie+0.064
ama+0.063
Neg logit contributions
 seen-0.145
hole-0.138
keeping-0.138
 stations-0.136
 crashing-0.132
Token like
Token ID588
Feature activation+1.24e-01

Pos logit contributions
hole+0.034
ously+0.031
 stations+0.031
keeping+0.030
ked+0.029
Neg logit contributions
iflower-0.048
pron-0.045
live-0.042
P-0.041
wl-0.039
Token the
Token ID262
Feature activation-5.13e-02

Pos logit contributions
iflower+0.004
pron+0.003
live+0.003
innie+0.003
ama+0.003
Neg logit contributions
 seen-0.004
hole-0.003
keeping-0.003
 stations-0.003
 contain-0.003
Token monster
Token ID9234
Feature activation-9.00e-02

Pos logit contributions
 seen+0.002
keeping+0.002
hole+0.002
ales+0.002
 stations+0.002
Neg logit contributions
iflower-0.001
pron-0.001
live-0.001
wl-0.001
ama-0.001
Token card
Token ID2657
Feature activation+1.95e+00

Pos logit contributions
 seen+0.002
keeping+0.002
hole+0.002
 stations+0.002
 contain+0.002
Neg logit contributions
iflower-0.003
pron-0.003
live-0.003
ama-0.003
wl-0.003
Token said
Token ID531
Feature activation+7.20e-01

Pos logit contributions
iflower+0.016
pron+0.014
live+0.013
innie+0.013
ama+0.013
Neg logit contributions
 seen-0.012
 officers-0.010
hole-0.010
 towards-0.010
keeping-0.010
Token,
Token ID11
Feature activation-8.57e-01

Pos logit contributions
iflower+0.026
pron+0.024
live+0.022
innie+0.021
wl+0.021
Neg logit contributions
 seen-0.016
 towards-0.016
 crashing-0.015
 contain-0.015
ously-0.015
Token "
Token ID366
Feature activation-8.04e-01

Pos logit contributions
 crashing+0.023
 towards+0.022
hole+0.021
 seen+0.021
 toward+0.021
Neg logit contributions
iflower-0.027
live-0.023
pron-0.023
wl-0.021
innie-0.021
TokenHi
Token ID17250
Feature activation+1.91e+00

Pos logit contributions
 seen+0.021
keeping+0.020
hole+0.020
 stations+0.020
 officers+0.019
Neg logit contributions
iflower-0.026
pron-0.024
live-0.021
ama-0.020
innie-0.020
Token,
Token ID11
Feature activation+8.20e-01

Pos logit contributions
iflower+0.063
pron+0.059
live+0.054
wl+0.051
innie+0.050
Neg logit contributions
 seen-0.049
hole-0.046
 officers-0.044
 crashing-0.043
 towards-0.043
Token what
Token ID644
Feature activation+4.49e+00

Pos logit contributions
iflower+0.024
pron+0.022
live+0.018
ama+0.018
innie+0.018
Neg logit contributions
 seen-0.024
keeping-0.023
hole-0.023
 stations-0.023
 contain-0.022
Token are
Token ID389
Feature activation+1.52e+00

Pos logit contributions
iflower+0.166
pron+0.155
live+0.140
innie+0.134
ama+0.134
Neg logit contributions
 seen-0.096
hole-0.089
keeping-0.089
 stations-0.087
 contain-0.083
Token you
Token ID345
Feature activation-4.01e-01

Pos logit contributions
iflower+0.050
pron+0.047
live+0.041
innie+0.039
ama+0.039
Neg logit contributions
 seen-0.039
keeping-0.035
 officers-0.035
hole-0.035
 stations-0.035
Token doing
Token ID1804
Feature activation+2.47e+00

Pos logit contributions
 seen+0.010
 crashing+0.009
hole+0.009
 want+0.008
 officers+0.008
Neg logit contributions
iflower-0.014
pron-0.013
live-0.012
wl-0.011
innie-0.011
Token?"
Token ID1701
Feature activation+2.04e+00

Pos logit contributions
iflower+0.084
pron+0.076
live+0.072
innie+0.069
wl+0.069
Neg logit contributions
 seen-0.059
keeping-0.053
 towards-0.053
hole-0.052
 crashing-0.051
Token The
Token ID383
Feature activation+7.82e-01

Pos logit contributions
iflower+0.069
pron+0.062
live+0.058
tto+0.053
wl+0.053
Neg logit contributions
 seen-0.053
 officers-0.048
 towards-0.048
keeping-0.047
hole-0.047
Token.
Token ID13
Feature activation+4.96e-01

Pos logit contributions
 contain+0.013
hole+0.013
 umb+0.013
 brake+0.012
 stations+0.012
Neg logit contributions
iflower-0.012
pron-0.010
live-0.009
ama-0.009
tto-0.008
Token
Token ID220
Feature activation+2.99e+00

Pos logit contributions
iflower+0.015
pron+0.014
live+0.012
wl+0.011
innie+0.011
Neg logit contributions
 seen-0.014
keeping-0.013
hole-0.013
 towards-0.013
 stations-0.012
Token
Token ID198
Feature activation+7.28e-01

Pos logit contributions
iflower+0.089
pron+0.084
live+0.066
wl+0.063
innie+0.062
Neg logit contributions
 seen-0.087
hole-0.084
keeping-0.083
 crashing-0.078
 towards-0.077
Token
Token ID198
Feature activation+4.54e-01

Pos logit contributions
iflower+0.023
pron+0.023
ama+0.019
omm+0.019
innie+0.018
Neg logit contributions
 seen-0.017
keeping-0.016
 contain-0.016
 towards-0.016
hole-0.015
TokenFrom
Token ID4863
Feature activation-5.38e-01

Pos logit contributions
iflower+0.012
pron+0.011
ama+0.011
oppy+0.010
innie+0.010
Neg logit contributions
 seen-0.012
 contain-0.012
 stations-0.012
 crashing-0.011
ets-0.011
Token that
Token ID326
Feature activation+5.15e+00

Pos logit contributions
 seen+0.014
hole+0.012
 officers+0.012
 stations+0.012
keeping+0.012
Neg logit contributions
iflower-0.018
pron-0.017
live-0.014
innie-0.014
ama-0.014
Token day
Token ID1110
Feature activation+2.00e+00

Pos logit contributions
iflower+0.166
pron+0.146
live+0.140
lly+0.129
oppy+0.128
Neg logit contributions
 seen-0.121
 towards-0.119
 crashing-0.115
 want-0.114
hole-0.108
Token on
Token ID319
Feature activation+2.35e+00

Pos logit contributions
iflower+0.069
pron+0.065
live+0.056
innie+0.053
wl+0.052
Neg logit contributions
 seen-0.047
 crashing-0.045
hole-0.044
 towards-0.043
 brake-0.042
Token,
Token ID11
Feature activation+1.22e+00

Pos logit contributions
iflower+0.096
pron+0.094
live+0.082
innie+0.078
wl+0.078
Neg logit contributions
 seen-0.034
 towards-0.033
 crashing-0.031
ously-0.030
 horror-0.030
Token Jack
Token ID3619
Feature activation+2.64e+00

Pos logit contributions
iflower+0.044
pron+0.042
live+0.040
innie+0.038
wl+0.038
Neg logit contributions
 seen-0.026
 towards-0.024
 contain-0.023
 crashing-0.023
 want-0.023
Token and
Token ID290
Feature activation+2.07e-01

Pos logit contributions
pron+0.092
iflower+0.092
live+0.080
innie+0.080
ama+0.078
Neg logit contributions
 seen-0.058
 stations-0.050
ets-0.050
keeping-0.048
hole-0.048
Token Lily
Token ID20037
Feature activation+1.49e-01

Pos logit contributions
iflower+0.013
pron+0.011
live+0.010
wl+0.010
tto+0.010
Neg logit contributions
 seen-0.011
 footsteps-0.010
 crashing-0.010
 officers-0.010
keeping-0.010
Token.
Token ID13
Feature activation+1.19e-03

Pos logit contributions
iflower+0.006
pron+0.005
live+0.005
wl+0.005
innie+0.005
Neg logit contributions
 seen-0.003
hole-0.003
keeping-0.003
 stations-0.002
 towards-0.002
Token
Token ID198
Feature activation+1.34e+00

Pos logit contributions
iflower+0.000
pron+0.000
live+0.000
wl+0.000
tto+0.000
Neg logit contributions
 seen-0.000
 towards-0.000
 crashing-0.000
hole-0.000
 footsteps-0.000
Token
Token ID198
Feature activation+1.14e+00

Pos logit contributions
iflower+0.047
pron+0.044
live+0.037
wl+0.036
tto+0.034
Neg logit contributions
 seen-0.030
hole-0.028
keeping-0.027
 towards-0.027
 crashing-0.026
Token"
Token ID1
Feature activation-2.18e+00

Pos logit contributions
iflower+0.030
pron+0.025
ama+0.022
live+0.022
innie+0.022
Neg logit contributions
 seen-0.036
 stations-0.035
hole-0.034
keeping-0.034
 towards-0.034
TokenHere
Token ID4342
Feature activation+4.12e+00

Pos logit contributions
Run+0.053
 seen+0.052
 stations+0.050
 brake+0.048
 contain+0.047
Neg logit contributions
iflower-0.070
ama-0.061
pron-0.058
oppy-0.058
wl-0.055
Token you
Token ID345
Feature activation-1.36e+00

Pos logit contributions
iflower+0.138
pron+0.138
live+0.118
innie+0.116
tto+0.114
Neg logit contributions
 seen-0.096
hole-0.086
 officers-0.083
keeping-0.082
 want-0.081
Token go
Token ID467
Feature activation+7.90e-01

Pos logit contributions
 seen+0.028
keeping+0.027
hole+0.026
 stations+0.026
 officers+0.024
Neg logit contributions
iflower-0.051
pron-0.047
live-0.043
ama-0.042
innie-0.041
Token.
Token ID13
Feature activation+6.50e-01

Pos logit contributions
iflower+0.022
pron+0.020
live+0.018
ama+0.017
innie+0.017
Neg logit contributions
 seen-0.024
hole-0.023
 stations-0.023
keeping-0.023
 contain-0.022
Token Tell
Token ID14026
Feature activation-5.13e-01

Pos logit contributions
iflower+0.022
pron+0.021
live+0.019
wl+0.018
innie+0.017
Neg logit contributions
 seen-0.015
hole-0.014
keeping-0.014
 stations-0.014
 crashing-0.014
Token me
Token ID502
Feature activation+2.59e-02

Pos logit contributions
 seen+0.012
 permission+0.010
ales+0.010
 officers+0.010
 stations+0.010
Neg logit contributions
iflower-0.018
pron-0.018
live-0.016
wl-0.015
tto-0.015
Token nodded
Token ID14464
Feature activation+4.81e-02

Pos logit contributions
 seen+0.009
 argued+0.008
hole+0.008
 officers+0.007
 gasp+0.007
Neg logit contributions
iflower-0.018
pron-0.016
live-0.015
wl-0.015
ama-0.015
Token and
Token ID290
Feature activation-1.80e-01

Pos logit contributions
iflower+0.002
pron+0.002
live+0.001
innie+0.001
ama+0.001
Neg logit contributions
 seen-0.001
hole-0.001
keeping-0.001
 stations-0.001
 contain-0.001
Token said
Token ID531
Feature activation+3.54e-01

Pos logit contributions
keeping+0.005
 seen+0.005
 stations+0.005
hole+0.005
 crashing+0.005
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
ama-0.004
innie-0.004
Token,
Token ID11
Feature activation-8.07e-01

Pos logit contributions
iflower+0.011
pron+0.010
live+0.010
wl+0.009
innie+0.009
Neg logit contributions
 seen-0.009
hole-0.009
 crashing-0.008
keeping-0.008
 stations-0.008
Token "
Token ID366
Feature activation-5.88e-01

Pos logit contributions
 seen+0.024
 crashing+0.024
hole+0.023
 towards+0.023
 contain+0.022
Neg logit contributions
iflower-0.023
pron-0.019
live-0.019
wl-0.017
innie-0.017
TokenThat
Token ID2504
Feature activation+4.07e+00

Pos logit contributions
 stations+0.014
 seen+0.014
hole+0.013
keeping+0.013
 brake+0.013
Neg logit contributions
iflower-0.019
pron-0.017
live-0.016
innie-0.015
ama-0.015
Token sounds
Token ID5238
Feature activation-1.89e+00

Pos logit contributions
iflower+0.138
pron+0.129
live+0.117
innie+0.113
wl+0.110
Neg logit contributions
 seen-0.096
hole-0.092
 stations-0.091
keeping-0.088
 officers-0.085
Token good
Token ID922
Feature activation+9.48e-01

Pos logit contributions
 seen+0.060
hole+0.058
keeping+0.058
 stations+0.057
 officers+0.055
Neg logit contributions
iflower-0.050
pron-0.045
live-0.039
innie-0.037
ama-0.036
Token,
Token ID11
Feature activation+7.34e-01

Pos logit contributions
iflower+0.029
pron+0.026
live+0.023
ama+0.022
innie+0.022
Neg logit contributions
 seen-0.027
hole-0.026
keeping-0.026
 stations-0.025
 contain-0.025
Token grandma
Token ID49890
Feature activation+1.16e+00

Pos logit contributions
iflower+0.019
pron+0.017
live+0.015
wl+0.014
tto+0.014
Neg logit contributions
 seen-0.024
hole-0.023
 stations-0.022
keeping-0.022
 officers-0.022
Token.
Token ID13
Feature activation+3.08e-01

Pos logit contributions
iflower+0.047
pron+0.044
live+0.041
innie+0.040
wl+0.039
Neg logit contributions
 seen-0.020
hole-0.018
keeping-0.018
 stations-0.017
 officers-0.016
Tokened
Token ID276
Feature activation+1.02e+00

Pos logit contributions
iflower+0.044
pron+0.040
live+0.037
ama+0.035
innie+0.034
Neg logit contributions
 seen-0.038
hole-0.037
 brake-0.036
 stations-0.034
ales-0.034
Token with
Token ID351
Feature activation+6.40e-01

Pos logit contributions
iflower+0.026
pron+0.024
live+0.020
ama+0.020
innie+0.019
Neg logit contributions
 seen-0.032
hole-0.031
keeping-0.030
 stations-0.030
ales-0.030
Token delight
Token ID10974
Feature activation+1.56e+00

Pos logit contributions
iflower+0.020
pron+0.019
innie+0.017
live+0.017
ama+0.016
Neg logit contributions
hole-0.016
 seen-0.015
 spine-0.015
ales-0.014
keeping-0.014
Token.
Token ID13
Feature activation-1.61e-01

Pos logit contributions
iflower+0.051
pron+0.046
ama+0.042
live+0.040
innie+0.039
Neg logit contributions
 seen-0.042
 brake-0.039
hole-0.039
 stations-0.038
 brakes-0.038
Token
Token ID198
Feature activation+2.60e-01

Pos logit contributions
hole+0.004
 stations+0.004
 crashing+0.004
 seen+0.004
keeping+0.004
Neg logit contributions
iflower-0.005
pron-0.005
live-0.004
P-0.004
tto-0.004
Token
Token ID198
Feature activation+6.04e-02

Pos logit contributions
iflower+0.009
pron+0.008
live+0.007
heddar+0.006
wl+0.006
Neg logit contributions
 crashing-0.006
 stations-0.006
 seen-0.006
hole-0.006
 footsteps-0.005
TokenAt
Token ID2953
Feature activation-6.97e-01

Pos logit contributions
iflower+0.002
pron+0.002
ama+0.001
oppy+0.001
live+0.001
Neg logit contributions
 stations-0.002
 crashing-0.002
hole-0.001
 brake-0.001
 seen-0.001
Token the
Token ID262
Feature activation+3.46e-01

Pos logit contributions
 seen+0.018
keeping+0.016
 stations+0.016
hole+0.016
 contain+0.016
Neg logit contributions
iflower-0.023
pron-0.022
live-0.019
wl-0.019
innie-0.018
Token end
Token ID886
Feature activation+4.50e-01

Pos logit contributions
iflower+0.011
pron+0.010
live+0.009
innie+0.008
ama+0.008
Neg logit contributions
 seen-0.009
hole-0.009
keeping-0.009
 stations-0.009
 contain-0.008
Token,
Token ID11
Feature activation+3.92e-01

Pos logit contributions
iflower+0.019
pron+0.018
live+0.016
wl+0.016
tto+0.015
Neg logit contributions
 seen-0.008
hole-0.007
 stations-0.007
 crashing-0.007
 towards-0.006
Token Chloe
Token ID29476
Feature activation+6.93e-01

Pos logit contributions
iflower+0.013
pron+0.011
live+0.010
wl+0.010
tto+0.010
Neg logit contributions
 seen-0.010
 stations-0.010
hole-0.010
 crashing-0.009
keeping-0.009
Token a
Token ID257
Feature activation-9.54e-01

Pos logit contributions
 seen+0.059
hole+0.058
keeping+0.057
 contain+0.056
 stations+0.055
Neg logit contributions
iflower-0.052
pron-0.046
live-0.040
innie-0.039
ama-0.038
Token few
Token ID1178
Feature activation-6.58e-01

Pos logit contributions
 seen+0.026
hole+0.024
 stations+0.024
 crashing+0.023
 contain+0.023
Neg logit contributions
iflower-0.030
pron-0.027
live-0.025
innie-0.023
ama-0.023
Token minutes
Token ID2431
Feature activation-3.01e+00

Pos logit contributions
keeping+0.018
hole+0.017
 seen+0.016
ales+0.016
 contain+0.015
Neg logit contributions
iflower-0.020
pron-0.018
lf-0.016
live-0.016
innie-0.015
Token,
Token ID11
Feature activation-2.88e-01

Pos logit contributions
keeping+0.103
 stations+0.100
 seen+0.099
hole+0.098
 brake+0.097
Neg logit contributions
iflower-0.062
pron-0.058
ama-0.051
live-0.051
innie-0.047
Token the
Token ID262
Feature activation-1.14e+00

Pos logit contributions
 seen+0.010
hole+0.009
keeping+0.009
 stations+0.009
 contain+0.009
Neg logit contributions
iflower-0.007
pron-0.006
live-0.005
ama-0.005
innie-0.005
Token car
Token ID1097
Feature activation-2.30e-01

Pos logit contributions
 seen+0.039
hole+0.039
keeping+0.038
 contain+0.037
ales+0.036
Neg logit contributions
iflower-0.027
pron-0.026
ama-0.022
innie-0.019
live-0.018
Token stopped
Token ID5025
Feature activation-1.02e+00

Pos logit contributions
keeping+0.007
 seen+0.006
hole+0.006
 stations+0.006
 contain+0.006
Neg logit contributions
iflower-0.006
pron-0.006
live-0.005
ama-0.005
wl-0.005
Token.
Token ID13
Feature activation-1.15e+00

Pos logit contributions
keeping+0.030
 seen+0.029
 contain+0.028
ets+0.028
 stations+0.028
Neg logit contributions
iflower-0.029
pron-0.026
ama-0.023
wl-0.021
innie-0.021
Token They
Token ID1119
Feature activation-1.61e+00

Pos logit contributions
 seen+0.028
keeping+0.027
hole+0.027
 stations+0.027
 officers+0.026
Neg logit contributions
iflower-0.039
pron-0.036
live-0.032
wl-0.031
innie-0.030
Token looked
Token ID3114
Feature activation+5.08e-01

Pos logit contributions
keeping+0.045
hole+0.042
 seen+0.042
 stations+0.041
ales+0.040
Neg logit contributions
iflower-0.045
pron-0.042
live-0.037
wl-0.035
ama-0.035
Token in
Token ID287
Feature activation+3.81e-02

Pos logit contributions
iflower+0.012
pron+0.011
 important+0.008
 best+0.008
ip+0.008
Neg logit contributions
keeping-0.018
 stations-0.016
hole-0.015
 seen-0.015
ales-0.015
Token the
Token ID262
Feature activation-6.27e-01

Pos logit contributions
iflower+0.023
pron+0.021
live+0.019
innie+0.018
ama+0.018
Neg logit contributions
 seen-0.023
keeping-0.022
hole-0.022
 stations-0.021
 officers-0.021
Token ornament
Token ID45618
Feature activation+2.92e+00

Pos logit contributions
 seen+0.018
hole+0.016
keeping+0.016
 stations+0.016
 brake+0.016
Neg logit contributions
iflower-0.019
pron-0.018
live-0.016
innie-0.015
wl-0.015
Token if
Token ID611
Feature activation-2.25e-02

Pos logit contributions
iflower+0.085
pron+0.077
live+0.069
innie+0.066
tto+0.063
Neg logit contributions
 seen-0.086
hole-0.081
keeping-0.080
 stations-0.078
 contain-0.077
Token you
Token ID345
Feature activation-4.81e+00

Pos logit contributions
 seen+0.001
 officers+0.001
keeping+0.000
hole+0.000
 towards+0.000
Neg logit contributions
iflower-0.001
pron-0.001
live-0.001
innie-0.001
tto-0.001
Token can
Token ID460
Feature activation-9.68e-01

Pos logit contributions
 stations+0.095
keeping+0.094
hole+0.091
 seen+0.091
 crashing+0.086
Neg logit contributions
iflower-0.179
pron-0.167
live-0.150
ama-0.146
wl-0.145
Token have
Token ID423
Feature activation+1.37e+00

Pos logit contributions
 seen+0.022
hole+0.020
 contain+0.019
keeping+0.019
 brake+0.019
Neg logit contributions
iflower-0.035
pron-0.033
live-0.030
innie-0.028
wl-0.028
Token it
Token ID340
Feature activation+1.22e+00

Pos logit contributions
iflower+0.045
pron+0.042
live+0.039
ama+0.037
wl+0.037
Neg logit contributions
 seen-0.037
hole-0.032
keeping-0.031
 stations-0.031
 officers-0.030
Token?"
Token ID1701
Feature activation+1.64e+00

Pos logit contributions
iflower+0.037
pron+0.033
live+0.031
innie+0.030
wl+0.029
Neg logit contributions
 seen-0.035
hole-0.032
keeping-0.031
 contain-0.030
 towards-0.030
Token So
Token ID1406
Feature activation-5.18e-02

Pos logit contributions
iflower+0.056
pron+0.052
live+0.045
tto+0.043
wl+0.042
Neg logit contributions
 seen-0.041
keeping-0.040
 officers-0.040
 towards-0.038
hole-0.038
Token,
Token ID11
Feature activation+2.65e-01

Pos logit contributions
 seen+0.001
keeping+0.001
hole+0.001
 officers+0.001
 stations+0.001
Neg logit contributions
iflower-0.002
pron-0.002
live-0.001
innie-0.001
wl-0.001
Token Twe
Token ID24205
Feature activation-1.73e-02

Pos logit contributions
iflower+0.009
pron+0.008
live+0.007
wl+0.007
innie+0.007
Neg logit contributions
 seen-0.007
keeping-0.007
 officers-0.007
hole-0.007
 stations-0.006
Token little
Token ID1310
Feature activation-6.92e-01

Pos logit contributions
keeping+0.029
 seen+0.029
 stations+0.028
ales+0.028
 contain+0.027
Neg logit contributions
iflower-0.024
pron-0.022
ama-0.017
innie-0.017
live-0.016
Token creature
Token ID7185
Feature activation-2.10e+00

Pos logit contributions
ales+0.023
keeping+0.023
 seen+0.023
 stations+0.023
ets+0.023
Neg logit contributions
iflower-0.015
pron-0.014
ama-0.011
very-0.010
 mummy-0.010
Token dressed
Token ID12049
Feature activation+1.24e+00

Pos logit contributions
 stations+0.060
ales+0.058
keeping+0.057
 brake+0.056
 brakes+0.055
Neg logit contributions
iflower-0.059
pron-0.053
live-0.045
on-0.044
lf-0.042
Token in
Token ID287
Feature activation-5.92e-01

Pos logit contributions
iflower+0.032
pron+0.031
live+0.022
tto+0.022
ama+0.022
Neg logit contributions
 stations-0.039
 seen-0.039
ales-0.038
 brake-0.038
ets-0.038
Token green
Token ID4077
Feature activation+2.04e-01

Pos logit contributions
ets+0.021
ales+0.020
ried+0.020
hole+0.020
 stations+0.019
Neg logit contributions
iflower-0.013
pron-0.012
M-0.008
innie-0.008
lf-0.008
Token.
Token ID13
Feature activation-9.68e-02

Pos logit contributions
iflower+0.006
pron+0.005
P+0.004
 mint+0.004
 bean+0.004
Neg logit contributions
 brake-0.006
ried-0.006
 contain-0.006
 seen-0.006
 wanna-0.005
Token The
Token ID383
Feature activation+4.55e-01

Pos logit contributions
 seen+0.003
hole+0.003
keeping+0.003
 officers+0.003
 stations+0.003
Neg logit contributions
iflower-0.003
pron-0.003
live-0.002
wl-0.002
innie-0.002
Token creature
Token ID7185
Feature activation+9.26e-02

Pos logit contributions
iflower+0.014
pron+0.013
live+0.011
ama+0.011
innie+0.011
Neg logit contributions
 seen-0.013
hole-0.012
keeping-0.012
 stations-0.012
 crashing-0.011
Token said
Token ID531
Feature activation+3.66e-02

Pos logit contributions
iflower+0.003
pron+0.002
live+0.002
innie+0.002
ama+0.002
Neg logit contributions
 seen-0.003
hole-0.003
keeping-0.002
 contain-0.002
 stations-0.002
Token,
Token ID11
Feature activation-1.07e+00

Pos logit contributions
iflower+0.001
pron+0.001
live+0.001
innie+0.001
wl+0.001
Neg logit contributions
 seen-0.001
hole-0.001
keeping-0.001
 crashing-0.001
 stations-0.001
Token "
Token ID366
Feature activation-1.06e+00

Pos logit contributions
 seen+0.033
hole+0.032
 crashing+0.032
keeping+0.031
 stations+0.031
Neg logit contributions
iflower-0.029
pron-0.025
live-0.023
innie-0.021
wl-0.021
Token help
Token ID1037
Feature activation+4.63e-01

Pos logit contributions
iflower+0.029
pron+0.027
live+0.024
ama+0.023
innie+0.023
Neg logit contributions
hole-0.020
 seen-0.020
 stations-0.019
keeping-0.019
ets-0.018
Token the
Token ID262
Feature activation+1.22e-01

Pos logit contributions
iflower+0.015
pron+0.014
live+0.012
innie+0.012
wl+0.012
Neg logit contributions
 seen-0.012
hole-0.011
keeping-0.011
 stations-0.011
 contain-0.011
Token dog
Token ID3290
Feature activation-1.42e+00

Pos logit contributions
iflower+0.004
pron+0.003
live+0.003
ama+0.003
innie+0.003
Neg logit contributions
 seen-0.004
hole-0.003
keeping-0.003
 stations-0.003
 contain-0.003
Token learn
Token ID2193
Feature activation-5.05e-01

Pos logit contributions
 seen+0.034
hole+0.034
 stations+0.034
keeping+0.033
 crashing+0.032
Neg logit contributions
iflower-0.046
pron-0.042
live-0.037
ama-0.036
wl-0.034
Token to
Token ID284
Feature activation+4.44e-01

Pos logit contributions
 seen+0.014
keeping+0.013
hole+0.013
 spine+0.013
ked+0.013
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
innie-0.011
ama-0.011
Token balance
Token ID5236
Feature activation+1.90e+00

Pos logit contributions
iflower+0.015
pron+0.014
live+0.013
innie+0.012
ama+0.012
Neg logit contributions
 seen-0.011
hole-0.010
keeping-0.010
 stations-0.010
 officers-0.009
Token.
Token ID13
Feature activation+2.49e-04

Pos logit contributions
iflower+0.063
pron+0.057
live+0.049
innie+0.047
wl+0.047
Neg logit contributions
 seen-0.049
hole-0.046
keeping-0.045
 stations-0.045
 contain-0.044
Token So
Token ID1406
Feature activation-4.86e-02

Pos logit contributions
iflower+0.000
pron+0.000
live+0.000
wl+0.000
innie+0.000
Neg logit contributions
 seen-0.000
hole-0.000
keeping-0.000
 stations-0.000
 officers-0.000
Token,
Token ID11
Feature activation+8.29e-01

Pos logit contributions
 seen+0.001
hole+0.001
keeping+0.001
 stations+0.001
 officers+0.001
Neg logit contributions
iflower-0.002
pron-0.001
live-0.001
innie-0.001
wl-0.001
Token the
Token ID262
Feature activation+5.96e-01

Pos logit contributions
iflower+0.024
pron+0.021
live+0.019
innie+0.018
wl+0.017
Neg logit contributions
 seen-0.025
keeping-0.024
hole-0.024
 stations-0.023
 officers-0.023
Token wise
Token ID10787
Feature activation-1.35e+00

Pos logit contributions
iflower+0.020
pron+0.019
live+0.017
innie+0.016
wl+0.016
Neg logit contributions
 seen-0.014
hole-0.013
 stations-0.013
keeping-0.013
 officers-0.013
Token quite
Token ID2407
Feature activation-1.48e+00

Pos logit contributions
 seen+0.038
keeping+0.038
hole+0.038
 stations+0.037
 officers+0.036
Neg logit contributions
iflower-0.026
pron-0.024
live-0.020
ama-0.018
innie-0.018
Token dark
Token ID3223
Feature activation-3.28e+00

Pos logit contributions
hole+0.046
keeping+0.046
 stations+0.045
 seen+0.045
ales+0.043
Neg logit contributions
iflower-0.038
pron-0.036
live-0.030
tto-0.028
ama-0.027
Token out
Token ID503
Feature activation-7.10e-01

Pos logit contributions
 contain+0.105
 brake+0.104
Run+0.103
awks+0.103
 brakes+0.102
Neg logit contributions
iflower-0.069
pron-0.062
time-0.051
lf-0.047
live-0.047
Token.
Token ID13
Feature activation-4.15e-01

Pos logit contributions
keeping+0.022
 stations+0.022
 dashboard+0.022
 brake+0.022
 brakes+0.022
Neg logit contributions
iflower-0.016
pron-0.013
ama-0.013
on-0.011
 mummy-0.011
Token His
Token ID2399
Feature activation+7.48e-01

Pos logit contributions
 seen+0.011
hole+0.011
keeping+0.011
 crashing+0.010
 officers+0.010
Neg logit contributions
iflower-0.013
pron-0.012
live-0.010
wl-0.010
innie-0.010
Token owner
Token ID4870
Feature activation-1.86e+00

Pos logit contributions
iflower+0.021
pron+0.019
live+0.016
innie+0.016
wl+0.015
Neg logit contributions
 seen-0.023
keeping-0.021
 officers-0.021
hole-0.021
 stations-0.021
Token was
Token ID373
Feature activation-5.95e-01

Pos logit contributions
 seen+0.054
keeping+0.054
hole+0.053
 stations+0.053
ales+0.051
Neg logit contributions
iflower-0.051
pron-0.047
live-0.042
ama-0.039
innie-0.039
Token waiting
Token ID4953
Feature activation+2.52e+00

Pos logit contributions
 seen+0.016
hole+0.015
keeping+0.015
 stations+0.015
 contain+0.014
Neg logit contributions
iflower-0.019
pron-0.018
live-0.015
innie-0.015
ama-0.015
Token patiently
Token ID37220
Feature activation-2.57e+00

Pos logit contributions
pron+0.060
iflower+0.057
ama+0.045
 best+0.043
 important+0.040
Neg logit contributions
keeping-0.079
 stations-0.079
ales-0.077
Run-0.074
hole-0.074
Token and
Token ID290
Feature activation+2.44e-01

Pos logit contributions
 seen+0.073
 stations+0.072
ales+0.072
keeping+0.071
angs+0.070
Neg logit contributions
iflower-0.067
pron-0.064
ama-0.051
live-0.048
very-0.048
Token walked
Token ID6807
Feature activation-2.64e+00

Pos logit contributions
iflower+0.007
pron+0.007
live+0.006
ama+0.005
innie+0.005
Neg logit contributions
 seen-0.007
hole-0.007
keeping-0.007
 stations-0.007
ales-0.006
Token on
Token ID319
Feature activation-1.69e+00

Pos logit contributions
iflower+0.066
pron+0.063
live+0.056
ama+0.052
wl+0.051
Neg logit contributions
 seen-0.049
keeping-0.048
 stations-0.048
hole-0.046
 crashing-0.046
Token the
Token ID262
Feature activation-2.57e+00

Pos logit contributions
 seen+0.050
hole+0.047
keeping+0.047
 stations+0.047
 contain+0.045
Neg logit contributions
iflower-0.049
pron-0.045
live-0.039
innie-0.037
ama-0.037
Token table
Token ID3084
Feature activation+1.25e-02

Pos logit contributions
 seen+0.074
hole+0.069
keeping+0.067
 stations+0.065
 crashing+0.065
Neg logit contributions
iflower-0.077
pron-0.071
live-0.061
innie-0.058
ama-0.057
Token.
Token ID13
Feature activation-4.80e-01

Pos logit contributions
pron+0.000
iflower+0.000
live+0.000
lf+0.000
ama+0.000
Neg logit contributions
 brake-0.000
 brakes-0.000
keeping-0.000
ales-0.000
 levers-0.000
Token She
Token ID1375
Feature activation+2.94e-02

Pos logit contributions
 seen+0.011
keeping+0.011
hole+0.011
 officers+0.011
 stations+0.010
Neg logit contributions
iflower-0.017
pron-0.015
live-0.014
wl-0.014
tto-0.013
Token wanted
Token ID2227
Feature activation-8.99e-01

Pos logit contributions
iflower+0.001
pron+0.001
wl+0.001
live+0.001
ama+0.001
Neg logit contributions
 seen-0.001
 argued-0.001
 officers-0.001
 backed-0.000
ked-0.000
Token to
Token ID284
Feature activation+7.45e-01

Pos logit contributions
 seen+0.019
keeping+0.015
 officers+0.015
 stations+0.014
 towards+0.014
Neg logit contributions
iflower-0.037
pron-0.034
live-0.032
wl-0.031
ama-0.030
Token give
Token ID1577
Feature activation+1.94e+00

Pos logit contributions
iflower+0.022
pron+0.022
live+0.019
innie+0.018
ama+0.018
Neg logit contributions
hole-0.020
 stations-0.019
 seen-0.019
keeping-0.018
ales-0.018
Token her
Token ID607
Feature activation+3.99e-01

Pos logit contributions
iflower+0.072
pron+0.066
live+0.062
wl+0.060
innie+0.058
Neg logit contributions
 seen-0.045
keeping-0.041
 towards-0.039
 officers-0.039
hole-0.038
Token grandma
Token ID49890
Feature activation-2.98e-02

Pos logit contributions
iflower+0.012
pron+0.011
live+0.009
innie+0.009
ama+0.009
Neg logit contributions
 seen-0.012
hole-0.011
keeping-0.011
 stations-0.011
 crashing-0.011
Token a
Token ID257
Feature activation-2.54e-01

Pos logit contributions
 seen+0.001
hole+0.001
keeping+0.001
 stations+0.001
 officers+0.001
Neg logit contributions
iflower-0.001
pron-0.001
live-0.001
innie-0.001
wl-0.001
Token at
Token ID379
Feature activation+8.25e-01

Pos logit contributions
iflower+0.025
pron+0.023
live+0.021
innie+0.020
ama+0.020
Neg logit contributions
 seen-0.019
keeping-0.018
hole-0.018
 stations-0.018
 contain-0.017
Token Fl
Token ID1610
Feature activation+1.65e+00

Pos logit contributions
iflower+0.026
pron+0.025
ama+0.020
live+0.020
wl+0.019
Neg logit contributions
 seen-0.023
 stations-0.022
 towards-0.021
keeping-0.021
hole-0.021
Tokenuffy
Token ID15352
Feature activation+8.47e-01

Pos logit contributions
iflower+0.032
pron+0.031
innie+0.021
lf+0.020
wl+0.020
Neg logit contributions
hole-0.058
 seen-0.057
ales-0.057
ked-0.055
owing-0.054
Token and
Token ID290
Feature activation+6.24e-01

Pos logit contributions
iflower+0.029
pron+0.027
live+0.024
innie+0.023
ama+0.023
Neg logit contributions
 seen-0.020
hole-0.019
keeping-0.019
 stations-0.019
 officers-0.018
Token said
Token ID531
Feature activation+5.95e-01

Pos logit contributions
iflower+0.024
pron+0.022
innie+0.021
ama+0.021
live+0.021
Neg logit contributions
 seen-0.013
 argued-0.011
 towards-0.011
ked-0.011
 officers-0.011
Token,
Token ID11
Feature activation-8.96e-01

Pos logit contributions
iflower+0.021
pron+0.019
live+0.017
innie+0.017
wl+0.016
Neg logit contributions
 seen-0.014
 towards-0.013
 crashing-0.013
hole-0.013
ked-0.013
Token "
Token ID366
Feature activation-8.51e-01

Pos logit contributions
 crashing+0.026
 seen+0.025
 towards+0.025
hole+0.024
 officers+0.024
Neg logit contributions
iflower-0.026
pron-0.023
live-0.022
innie-0.020
wl-0.020
TokenSure
Token ID19457
Feature activation+1.73e+00

Pos logit contributions
 seen+0.023
 towards+0.021
 contain+0.021
hole+0.021
 stations+0.020
Neg logit contributions
iflower-0.028
pron-0.025
ama-0.022
innie-0.022
live-0.022
Token!
Token ID0
Feature activation-1.80e-01

Pos logit contributions
iflower+0.041
pron+0.036
live+0.032
ama+0.031
innie+0.029
Neg logit contributions
 stations-0.059
keeping-0.058
 seen-0.058
hole-0.058
 contain-0.055
Token My
Token ID2011
Feature activation+8.29e-01

Pos logit contributions
 seen+0.004
hole+0.004
keeping+0.004
 stations+0.004
 crashing+0.003
Neg logit contributions
iflower-0.007
pron-0.006
live-0.005
innie-0.005
wl-0.005
Token name
Token ID1438
Feature activation+1.13e+00

Pos logit contributions
iflower+0.027
pron+0.025
live+0.022
innie+0.021
wl+0.021
Neg logit contributions
 seen-0.022
hole-0.021
keeping-0.020
 stations-0.020
 brake-0.019
Token replied
Token ID8712
Feature activation+1.07e-01

Pos logit contributions
iflower+0.067
pron+0.063
live+0.056
innie+0.054
P+0.053
Neg logit contributions
 argued-0.033
 seen-0.033
ked-0.028
 towards-0.027
 backed-0.027
Token,
Token ID11
Feature activation-1.37e+00

Pos logit contributions
iflower+0.004
pron+0.004
live+0.004
innie+0.004
wl+0.004
Neg logit contributions
 crashing-0.002
 towards-0.002
ked-0.002
 seen-0.002
 footsteps-0.002
Token "
Token ID366
Feature activation-1.01e+00

Pos logit contributions
 crashing+0.037
 towards+0.034
 seen+0.034
ked+0.034
 officers+0.033
Neg logit contributions
iflower-0.043
pron-0.037
live-0.037
innie-0.035
wl-0.035
TokenI
Token ID40
Feature activation-9.90e-01

Pos logit contributions
 seen+0.030
 towards+0.028
hole+0.028
keeping+0.028
 crashing+0.028
Neg logit contributions
iflower-0.028
pron-0.027
innie-0.023
ama-0.022
live-0.022
Token have
Token ID423
Feature activation+1.53e+00

Pos logit contributions
 seen+0.025
hole+0.023
keeping+0.022
 stations+0.022
 contain+0.021
Neg logit contributions
iflower-0.034
pron-0.031
live-0.028
innie-0.027
ama-0.027
Token never
Token ID1239
Feature activation-1.30e+00

Pos logit contributions
iflower+0.045
pron+0.041
live+0.036
innie+0.034
wl+0.033
Neg logit contributions
hole-0.043
 seen-0.042
keeping-0.041
 stations-0.041
 contain-0.040
Token seen
Token ID1775
Feature activation+1.44e+00

Pos logit contributions
 stations+0.015
hole+0.014
keeping+0.013
 levers+0.012
 crashing+0.012
Neg logit contributions
iflower-0.059
pron-0.056
live-0.052
P-0.050
tto-0.049
Token a
Token ID257
Feature activation-8.05e-01

Pos logit contributions
iflower+0.047
pron+0.043
live+0.039
innie+0.037
wl+0.037
Neg logit contributions
 seen-0.038
hole-0.035
keeping-0.034
 stations-0.034
 officers-0.033
Token bald
Token ID28273
Feature activation-8.84e-01

Pos logit contributions
 seen+0.023
hole+0.021
 crashing+0.021
 officers+0.021
 brake+0.021
Neg logit contributions
iflower-0.024
pron-0.021
live-0.021
innie-0.020
wl-0.019
Token man
Token ID582
Feature activation-2.63e+00

Pos logit contributions
 seen+0.031
hole+0.030
 stations+0.030
keeping+0.030
 contain+0.029
Neg logit contributions
iflower-0.021
pron-0.019
live-0.015
ama-0.014
innie-0.013
Token before
Token ID878
Feature activation-1.48e+00

Pos logit contributions
 seen+0.063
 stations+0.062
keeping+0.061
hole+0.060
 officers+0.057
Neg logit contributions
iflower-0.088
pron-0.082
live-0.071
ama-0.068
innie-0.067
Token.
Token ID13
Feature activation-8.97e-01

Pos logit contributions
 seen+0.052
Run+0.052
 stations+0.051
keeping+0.051
ked+0.051
Neg logit contributions
iflower-0.047
pron-0.045
lf-0.035
ama-0.034
live-0.033
Token
Token ID220
Feature activation+1.82e+00

Pos logit contributions
 seen+0.025
hole+0.023
keeping+0.023
 stations+0.023
 crashing+0.022
Neg logit contributions
iflower-0.028
pron-0.025
live-0.023
wl-0.021
innie-0.021
Token
Token ID198
Feature activation-3.12e-01

Pos logit contributions
iflower+0.045
pron+0.041
live+0.033
innie+0.031
wl+0.031
Neg logit contributions
 seen-0.062
hole-0.060
keeping-0.059
 crashing-0.057
 stations-0.057
Token
Token ID198
Feature activation-3.83e-01

Pos logit contributions
 seen+0.010
keeping+0.009
hole+0.009
 crashing+0.009
 towards+0.009
Neg logit contributions
iflower-0.009
pron-0.008
live-0.007
wl-0.006
ama-0.006
TokenWhen
Token ID2215
Feature activation-2.04e+00

Pos logit contributions
 seen+0.012
keeping+0.011
 stations+0.011
 crashing+0.011
hole+0.011
Neg logit contributions
iflower-0.011
pron-0.009
live-0.008
ama-0.008
wl-0.008
Token he
Token ID339
Feature activation-8.51e-01

Pos logit contributions
 seen+0.062
keeping+0.059
hole+0.058
 stations+0.058
 crashing+0.056
Neg logit contributions
iflower-0.058
pron-0.052
live-0.046
innie-0.043
wl-0.043
Token got
Token ID1392
Feature activation+1.53e+00

Pos logit contributions
 seen+0.021
hole+0.019
keeping+0.018
 contain+0.018
 stations+0.018
Neg logit contributions
iflower-0.030
pron-0.027
live-0.025
innie-0.024
wl-0.024
Token there
Token ID612
Feature activation-4.34e-01

Pos logit contributions
iflower+0.051
pron+0.047
live+0.042
ama+0.041
innie+0.040
Neg logit contributions
 seen-0.037
hole-0.036
keeping-0.036
 stations-0.036
 contain-0.033
Token,
Token ID11
Feature activation+6.71e-01

Pos logit contributions
 seen+0.013
hole+0.013
keeping+0.012
 stations+0.012
 contain+0.012
Neg logit contributions
iflower-0.012
pron-0.011
live-0.010
innie-0.009
ama-0.009
Token he
Token ID339
Feature activation+2.86e-01

Pos logit contributions
iflower+0.018
pron+0.016
live+0.014
innie+0.013
ama+0.013
Neg logit contributions
 seen-0.021
hole-0.020
keeping-0.020
 stations-0.020
 officers-0.019
Token saw
Token ID2497
Feature activation+1.02e+00

Pos logit contributions
iflower+0.010
pron+0.009
live+0.008
innie+0.008
ama+0.008
Neg logit contributions
 seen-0.007
hole-0.006
 stations-0.006
keeping-0.006
 officers-0.006
Token.
Token ID13
Feature activation-2.07e-01

Pos logit contributions
 seen+0.050
 stations+0.049
Run+0.047
ried+0.047
rying+0.046
Neg logit contributions
iflower-0.048
pron-0.045
live-0.038
ama-0.037
lf-0.037
Token Suddenly
Token ID24975
Feature activation-1.09e-02

Pos logit contributions
 crashing+0.005
keeping+0.005
 officers+0.005
hole+0.005
 seen+0.005
Neg logit contributions
iflower-0.007
pron-0.007
live-0.006
lf-0.005
wl-0.005
Token,
Token ID11
Feature activation-3.26e-01

Pos logit contributions
 seen+0.000
hole+0.000
keeping+0.000
 stations+0.000
 officers+0.000
Neg logit contributions
iflower-0.000
pron-0.000
live-0.000
innie-0.000
wl-0.000
Token Ham
Token ID4345
Feature activation+8.52e-01

Pos logit contributions
 seen+0.009
keeping+0.008
 officers+0.008
 stations+0.008
hole+0.008
Neg logit contributions
iflower-0.010
pron-0.009
live-0.008
innie-0.008
wl-0.008
Tokenster
Token ID1706
Feature activation+7.31e-01

Pos logit contributions
iflower+0.026
pron+0.025
live+0.021
innie+0.020
ama+0.020
Neg logit contributions
 seen-0.023
hole-0.022
 officers-0.020
ales-0.020
keeping-0.020
Token tri
Token ID1333
Feature activation-5.74e+00

Pos logit contributions
iflower+0.024
pron+0.022
live+0.020
innie+0.019
ama+0.019
Neg logit contributions
 seen-0.018
hole-0.017
keeping-0.017
 stations-0.017
 contain-0.016
Tokenpped
Token ID1496
Feature activation-1.40e+00

Pos logit contributions
 stations+0.138
 seen+0.137
keeping+0.135
 officers+0.133
hole+0.131
Neg logit contributions
iflower-0.185
pron-0.172
live-0.159
tto-0.151
P-0.150
Token and
Token ID290
Feature activation-9.52e-01

Pos logit contributions
 seen+0.037
 stations+0.035
keeping+0.035
 contain+0.035
hole+0.034
Neg logit contributions
iflower-0.043
pron-0.040
live-0.035
tto-0.033
ama-0.033
Token the
Token ID262
Feature activation-1.92e+00

Pos logit contributions
 seen+0.028
 stations+0.027
hole+0.027
keeping+0.027
ales+0.026
Neg logit contributions
iflower-0.027
pron-0.024
live-0.022
tto-0.020
ama-0.020
Token bone
Token ID9970
Feature activation+1.25e+00

Pos logit contributions
 seen+0.060
 stations+0.058
keeping+0.058
hole+0.058
 contain+0.056
Neg logit contributions
iflower-0.052
pron-0.048
live-0.040
ama-0.038
innie-0.038
Token flew
Token ID13112
Feature activation-3.25e+00

Pos logit contributions
iflower+0.034
pron+0.031
live+0.026
tto+0.024
lf+0.024
Neg logit contributions
keeping-0.038
 stations-0.037
 seen-0.037
ales-0.037
hole-0.035
Token
Token ID198
Feature activation-3.21e-01

Pos logit contributions
 seen+0.003
 towards+0.003
hole+0.003
 crashing+0.002
 toward+0.002
Neg logit contributions
iflower-0.004
pron-0.004
live-0.003
wl-0.003
innie-0.003
TokenThe
Token ID464
Feature activation+5.81e-01

Pos logit contributions
 seen+0.009
 contain+0.009
 towards+0.009
 crashing+0.009
 toward+0.008
Neg logit contributions
iflower-0.009
pron-0.008
live-0.007
innie-0.007
wl-0.007
Token fire
Token ID2046
Feature activation-4.82e-01

Pos logit contributions
iflower+0.017
pron+0.016
live+0.014
ama+0.013
innie+0.013
Neg logit contributions
 seen-0.017
hole-0.016
keeping-0.016
 contain-0.015
ales-0.015
Tokenwork
Token ID1818
Feature activation+1.95e+00

Pos logit contributions
 seen+0.016
keeping+0.016
hole+0.016
 stations+0.015
 contain+0.015
Neg logit contributions
iflower-0.012
pron-0.010
live-0.009
ama-0.009
wl-0.008
Token show
Token ID905
Feature activation-8.34e-01

Pos logit contributions
iflower+0.059
pron+0.054
live+0.048
ama+0.045
innie+0.045
Neg logit contributions
 seen-0.054
keeping-0.052
hole-0.051
 stations-0.050
 contain-0.048
Token lasted
Token ID15436
Feature activation-6.24e+00

Pos logit contributions
 seen+0.023
hole+0.023
keeping+0.023
 stations+0.022
 brake+0.022
Neg logit contributions
iflower-0.023
pron-0.021
live-0.019
ama-0.018
wl-0.018
Token for
Token ID329
Feature activation-1.48e+00

Pos logit contributions
keeping+0.212
 officers+0.197
ales+0.196
hole+0.195
 brakes+0.188
Neg logit contributions
pron-0.130
iflower-0.130
 important-0.101
innie-0.099
 mummy-0.097
Token what
Token ID644
Feature activation+3.21e+00

Pos logit contributions
hole+0.044
 seen+0.043
keeping+0.042
ried+0.040
 stations+0.040
Neg logit contributions
iflower-0.041
pron-0.037
live-0.032
innie-0.031
ama-0.029
Token seemed
Token ID3947
Feature activation-3.82e+00

Pos logit contributions
iflower+0.090
pron+0.082
live+0.071
innie+0.069
ama+0.067
Neg logit contributions
 seen-0.099
keeping-0.092
hole-0.092
 stations-0.091
 officers-0.088
Token like
Token ID588
Feature activation-2.30e+00

Pos logit contributions
keeping+0.092
hole+0.084
 stations+0.080
ales+0.080
 brake+0.074
Neg logit contributions
iflower-0.120
pron-0.115
 important-0.109
 special-0.105
 famous-0.101
Token hours
Token ID2250
Feature activation-2.11e+00

Pos logit contributions
hole+0.068
 seen+0.066
keeping+0.065
ales+0.064
 stations+0.063
Neg logit contributions
iflower-0.064
pron-0.060
live-0.049
ama-0.047
innie-0.046
Token the
Token ID262
Feature activation-9.58e-01

Pos logit contributions
iflower+0.002
pron+0.001
live+0.001
innie+0.001
wl+0.001
Neg logit contributions
 seen-0.002
hole-0.001
 stations-0.001
keeping-0.001
 crashing-0.001
Token seams
Token ID41280
Feature activation-1.88e+00

Pos logit contributions
 seen+0.030
hole+0.028
keeping+0.028
 stations+0.028
 contain+0.027
Neg logit contributions
iflower-0.027
pron-0.025
live-0.021
ama-0.020
innie-0.020
Token of
Token ID286
Feature activation-8.19e-01

Pos logit contributions
keeping+0.053
 seen+0.052
 stations+0.051
ales+0.049
hole+0.049
Neg logit contributions
iflower-0.053
pron-0.052
live-0.043
ama-0.042
wl-0.039
Token her
Token ID607
Feature activation-7.11e-01

Pos logit contributions
 seen+0.025
hole+0.024
keeping+0.024
 contain+0.023
 stations+0.023
Neg logit contributions
iflower-0.022
pron-0.020
live-0.017
innie-0.016
ama-0.016
Token sweater
Token ID35685
Feature activation-3.30e-01

Pos logit contributions
 seen+0.020
hole+0.019
keeping+0.019
 stations+0.018
 officers+0.018
Neg logit contributions
iflower-0.022
pron-0.020
live-0.017
innie-0.017
wl-0.017
Token *
Token ID1635
Feature activation-6.35e+00

Pos logit contributions
 seen+0.010
 stations+0.009
keeping+0.009
hole+0.009
ales+0.009
Neg logit contributions
iflower-0.009
pron-0.009
live-0.007
ama-0.007
wl-0.007
Tokenpo
Token ID7501
Feature activation-3.40e+00

Pos logit contributions
 seen+0.167
hole+0.158
 stations+0.156
keeping+0.156
 contain+0.149
Neg logit contributions
iflower-0.201
pron-0.186
live-0.164
innie-0.157
ama-0.155
Tokenpped
Token ID1496
Feature activation-2.87e-01

Pos logit contributions
 seen+0.092
 contain+0.092
 stations+0.090
keeping+0.089
 footsteps+0.089
Neg logit contributions
iflower-0.103
pron-0.097
live-0.084
innie-0.083
wl-0.079
Token*
Token ID9
Feature activation-2.04e+00

Pos logit contributions
 seen+0.008
 stations+0.007
 contain+0.007
keeping+0.007
hole+0.007
Neg logit contributions
iflower-0.009
pron-0.009
live-0.007
ama-0.007
wl-0.007
Token!
Token ID0
Feature activation-7.81e-01

Pos logit contributions
 seen+0.049
hole+0.045
keeping+0.045
 stations+0.045
 contain+0.043
Neg logit contributions
iflower-0.071
pron-0.065
live-0.059
innie-0.056
ama-0.056
Token Everyone
Token ID11075
Feature activation-2.58e+00

Pos logit contributions
 seen+0.020
hole+0.019
keeping+0.018
 crashing+0.018
 towards+0.018
Neg logit contributions
iflower-0.026
pron-0.024
live-0.022
wl-0.021
tto-0.020
Token to
Token ID284
Feature activation+1.86e-01

Pos logit contributions
 seen+0.022
keeping+0.021
 stations+0.021
 crashing+0.021
hole+0.020
Neg logit contributions
iflower-0.024
pron-0.023
live-0.019
ama-0.018
innie-0.018
Token his
Token ID465
Feature activation-7.21e-01

Pos logit contributions
iflower+0.006
pron+0.005
live+0.005
innie+0.004
ama+0.004
Neg logit contributions
 seen-0.005
hole-0.005
keeping-0.005
 stations-0.005
 crashing-0.005
Token friend
Token ID1545
Feature activation-2.30e+00

Pos logit contributions
 seen+0.019
 footsteps+0.019
 officers+0.018
 ways+0.018
 stations+0.018
Neg logit contributions
iflower-0.021
pron-0.020
live-0.018
wl-0.017
tto-0.017
Token.
Token ID13
Feature activation-8.14e-01

Pos logit contributions
 seen+0.063
hole+0.059
keeping+0.059
 stations+0.058
 officers+0.056
Neg logit contributions
iflower-0.072
pron-0.066
live-0.058
innie-0.056
ama-0.055
Token He
Token ID679
Feature activation+8.67e-01

Pos logit contributions
 seen+0.021
hole+0.021
 towards+0.021
keeping+0.020
 officers+0.020
Neg logit contributions
iflower-0.027
pron-0.025
live-0.022
wl-0.020
innie-0.020
Token scam
Token ID19126
Feature activation-6.26e+00

Pos logit contributions
iflower+0.031
pron+0.028
live+0.026
ama+0.025
innie+0.024
Neg logit contributions
 seen-0.021
hole-0.019
 officers-0.018
 stations-0.018
keeping-0.018
Tokenpered
Token ID13653
Feature activation-3.01e+00

Pos logit contributions
 seen+0.069
hole+0.066
 stations+0.062
keeping+0.059
 officers+0.055
Neg logit contributions
iflower-0.287
pron-0.272
live-0.259
innie-0.253
ama-0.250
Token around
Token ID1088
Feature activation-4.20e-02

Pos logit contributions
 seen+0.083
hole+0.080
keeping+0.079
 stations+0.078
 contain+0.075
Neg logit contributions
iflower-0.088
pron-0.081
live-0.071
innie-0.069
ama-0.067
Token trying
Token ID2111
Feature activation+3.77e-01

Pos logit contributions
 seen+0.001
hole+0.001
keeping+0.001
 stations+0.001
ales+0.001
Neg logit contributions
iflower-0.001
pron-0.001
lf-0.001
innie-0.001
live-0.001
Token to
Token ID284
Feature activation-1.12e+00

Pos logit contributions
iflower+0.012
pron+0.011
live+0.010
innie+0.009
ama+0.009
Neg logit contributions
 seen-0.010
keeping-0.009
hole-0.009
 stations-0.009
 officers-0.009
Token find
Token ID1064
Feature activation+1.22e+00

Pos logit contributions
 seen+0.027
hole+0.027
keeping+0.026
 stations+0.026
ously+0.026
Neg logit contributions
iflower-0.035
pron-0.033
live-0.028
innie-0.027
lf-0.027
Token and
Token ID290
Feature activation+5.61e-02

Pos logit contributions
iflower+0.016
pron+0.016
live+0.013
ama+0.013
innie+0.013
Neg logit contributions
 stations-0.015
 brake-0.015
keeping-0.015
 seen-0.015
 crashing-0.015
Token saw
Token ID2497
Feature activation+1.31e+00

Pos logit contributions
iflower+0.002
pron+0.002
live+0.002
innie+0.001
ama+0.001
Neg logit contributions
 seen-0.001
hole-0.001
keeping-0.001
 stations-0.001
 crashing-0.001
Token there
Token ID612
Feature activation-5.71e-01

Pos logit contributions
iflower+0.036
pron+0.033
live+0.028
innie+0.027
ama+0.027
Neg logit contributions
 seen-0.040
hole-0.039
keeping-0.038
 stations-0.037
 contain-0.037
Token were
Token ID547
Feature activation-5.21e-01

Pos logit contributions
 seen+0.018
keeping+0.018
hole+0.017
 stations+0.017
 brake+0.017
Neg logit contributions
iflower-0.014
pron-0.013
live-0.012
ama-0.011
innie-0.010
Token no
Token ID645
Feature activation-2.40e-01

Pos logit contributions
 seen+0.015
hole+0.015
keeping+0.015
ales+0.014
 contain+0.014
Neg logit contributions
iflower-0.015
pron-0.013
live-0.012
innie-0.011
ama-0.011
Token monsters
Token ID10963
Feature activation-5.76e+00

Pos logit contributions
keeping+0.007
ked+0.007
ried+0.007
hole+0.007
ales+0.007
Neg logit contributions
iflower-0.006
pron-0.006
 best-0.005
 important-0.005
 story-0.005
Token.
Token ID13
Feature activation-4.59e-01

Pos logit contributions
ales+0.163
awks+0.157
 brakes+0.155
keeping+0.154
 anticipation+0.152
Neg logit contributions
iflower-0.138
pron-0.120
lf-0.110
live-0.105
on-0.103
Token She
Token ID1375
Feature activation-2.62e-01

Pos logit contributions
 seen+0.012
keeping+0.012
hole+0.012
 stations+0.011
 officers+0.011
Neg logit contributions
iflower-0.015
pron-0.014
live-0.012
wl-0.011
innie-0.011
Token felt
Token ID2936
Feature activation+9.49e-02

Pos logit contributions
 seen+0.007
hole+0.007
keeping+0.007
 stations+0.007
 officers+0.006
Neg logit contributions
iflower-0.008
pron-0.008
live-0.007
innie-0.006
ama-0.006
Token proud
Token ID6613
Feature activation-1.82e+00

Pos logit contributions
iflower+0.003
pron+0.002
live+0.002
tto+0.002
ama+0.002
Neg logit contributions
hole-0.003
keeping-0.003
ales-0.003
 seen-0.003
 stations-0.003
Token of
Token ID286
Feature activation+1.50e+00

Pos logit contributions
keeping+0.063
 brake+0.061
ales+0.059
 stations+0.059
ried+0.059
Neg logit contributions
iflower-0.041
pron-0.037
ama-0.030
 story-0.028
 mummy-0.028