TOP ACTIVATIONS
MAX = 4.375

. She used a m itt to take out the pan
When Daisy put the magn ifying glass down , the cater
She takes out a magn ifying glass . She looks at
the watch with the magn ifying glass . He saw the
He took out a magn ifying glass and a notebook and
the necklace and the magn ifying glass and started to examine
flashlight , and a magn ifying glass . As
, handing him the magn ifying glass . She looked around
put on an oven m itt and a a pron .
She picks up the re ind eer . She looks at
She took out a magn ifying glass and looked very carefully
took off the oven m itt s and opened the oven
said , holding the magn ifying glass over a black ant
a flashlight and a magn ifying glass to examine their finds
She grabbed oven m itt s and carefully pulled out
cried , dropping the magn ifying glass . "
and handed him the magn ifying glass . He
the necklace and the magn ifying glass and told him that
you to the brave l itt lest one ! <|endoftext|> M
They put on oven m itt s and took out the

MIN ACTIVATIONS
MIN = -2.624

to always keep his useful tool with him , just in
He realized that sharing his tool with Billy would have been
see and do . From time to time people visited the
No , it is my tool now ! I can blow
cleaning wasn 't her favorite thing to do , she knew
. Tim my had a w igg ly tooth that was
that sharing is a good thing and it made her feel
. It 's my favorite thing to play with ."
After their fun meeting , Lily went back
. You are the best thing in my life ." He
would never get his spicy thing , and his trust had
lying is never the right thing to do . From that
" It 's a good thing you know what you want
mean ! You ruined our fun !" Ben cries .
'm sorry for breaking your tool . I love you ,
" That 's the best thing to do . Then it
" It 's a new tool to open cans faster ."
could use her daddy 's tool box to fix anything that
do . From time to time people visited the Earth from
excited because today was his first day of school . He

INTERVAL 2.625 - 4.375
CONTAINS 0.002%

and hugged the rh in oc eros . He finally understood
try to grab the av oc ados . But the rock
," said the rh in oc eros , feeling discouraged .
examine it with the magn ifying glass ?" "
He brought his magn ifying glass and looked closely .

INTERVAL 0.876 - 2.625
CONTAINS 3.627%

and soon became very good at playing her fl ute .
time , a little girl named Lily went to a show
closely together . She looked closely at her fingers and tried
No , you 'll get too wet . Come on ,
again . He realized that sometimes struggling isn 't always a

INTERVAL -0.874 - 0.876
CONTAINS 92.939%

house and added them to the mountain . Anna saw what
come true . <|endoftext|> Once upon a time there was a
" Can I act with you , Ben ? Can
when it was raining outside . One day while he was
end of the trip , the queen returned home with a

INTERVAL -2.624 - -0.874
CONTAINS 3.432%

shield has a star on it . It is very bright
house ," her mom my said . L ily
My friend is in the pit . He jumped in and
day better . From that day on , Lily always remembered
one of the children slipped and fell . The others ran
Token.
Token ID13
Feature activation-3.53e-02

Pos logit contributions
isel+0.023
icker+0.023
oths+0.023
imming+0.022
apa+0.022
Neg logit contributions
rob-0.021
 l-0.018
 eighty-0.016
 Ske-0.016
 cand-0.015
Token She
Token ID1375
Feature activation+3.15e-01

Pos logit contributions
rob+0.001
 l+0.001
 Ske+0.001
 eighty+0.001
 Velvet+0.001
Neg logit contributions
isel-0.002
icker-0.002
imming-0.002
oths-0.002
ench-0.002
Token used
Token ID973
Feature activation+6.50e-02

Pos logit contributions
isel+0.015
icker+0.014
oths+0.014
apa+0.014
umped+0.014
Neg logit contributions
rob-0.014
 l-0.011
 eighty-0.011
 Ske-0.010
 cand-0.010
Token a
Token ID257
Feature activation+8.39e-02

Pos logit contributions
isel+0.003
icker+0.003
oths+0.003
apa+0.003
umped+0.003
Neg logit contributions
rob-0.003
 l-0.002
 eighty-0.002
 Ske-0.002
 cand-0.002
Token m
Token ID285
Feature activation+7.88e-01

Pos logit contributions
oths+0.004
isel+0.004
imming+0.004
aining+0.004
icker+0.004
Neg logit contributions
rob-0.003
 l-0.003
 eighty-0.002
 cand-0.002
 candy-0.002
Tokenitt
Token ID715
Feature activation+4.12e+00

Pos logit contributions
isel+0.017
icker+0.016
oths+0.016
imming+0.015
umped+0.015
Neg logit contributions
rob-0.053
 l-0.047
 eighty-0.045
 cand-0.045
 Ske-0.044
Token to
Token ID284
Feature activation-7.57e-02

Pos logit contributions
icker+0.186
isel+0.185
oths+0.184
imming+0.179
aining+0.177
Neg logit contributions
rob-0.177
 l-0.159
 cand-0.144
 eighty-0.138
 scar-0.132
Token take
Token ID1011
Feature activation-2.00e-01

Pos logit contributions
rob+0.003
 l+0.003
 cand+0.002
 eighty+0.002
 Ske+0.002
Neg logit contributions
isel-0.004
oths-0.004
imming-0.003
icker-0.003
umped-0.003
Token out
Token ID503
Feature activation+1.38e-01

Pos logit contributions
rob+0.007
 l+0.006
 cand+0.005
 Ske+0.005
 eighty+0.005
Neg logit contributions
isel-0.011
icker-0.011
oths-0.010
imming-0.010
leys-0.010
Token the
Token ID262
Feature activation+1.25e-01

Pos logit contributions
isel+0.008
icker+0.008
oths+0.008
imming+0.008
umped+0.007
Neg logit contributions
rob-0.004
 l-0.004
 cand-0.003
 eighty-0.003
 Ske-0.003
Token pan
Token ID3425
Feature activation+5.26e-01

Pos logit contributions
isel+0.006
icker+0.006
oths+0.006
apa+0.006
umped+0.006
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 Ske-0.004
 cand-0.004
Token When
Token ID1649
Feature activation+1.40e-02

Pos logit contributions
rob+0.001
 l+0.001
 Ske+0.001
 eighty+0.001
 Velvet+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.001
imming-0.001
umped-0.001
Token Daisy
Token ID40355
Feature activation+5.13e-01

Pos logit contributions
isel+0.001
icker+0.001
oths+0.001
apa+0.001
umped+0.001
Neg logit contributions
rob-0.000
 l-0.000
 eighty-0.000
 Ske-0.000
 cand-0.000
Token put
Token ID1234
Feature activation-9.71e-01

Pos logit contributions
isel+0.024
icker+0.023
oths+0.023
umped+0.022
apa+0.022
Neg logit contributions
rob-0.023
 l-0.019
 eighty-0.017
 Ske-0.017
 cand-0.017
Token the
Token ID262
Feature activation+4.98e-01

Pos logit contributions
rob+0.045
 l+0.035
 eighty+0.034
 Ske+0.033
 cand+0.033
Neg logit contributions
isel-0.043
oths-0.042
icker-0.041
apa-0.040
aining-0.039
Token magn
Token ID7842
Feature activation-1.64e-01

Pos logit contributions
isel+0.025
icker+0.024
oths+0.024
umped+0.023
apa+0.023
Neg logit contributions
rob-0.021
 l-0.017
 eighty-0.015
 cand-0.015
 Ske-0.015
Tokenifying
Token ID4035
Feature activation+3.94e+00

Pos logit contributions
rob+0.011
 eighty+0.010
 l+0.010
 seventy+0.010
 Ske+0.010
Neg logit contributions
isel-0.003
oths-0.003
icker-0.003
ooks-0.003
appiness-0.003
Token glass
Token ID5405
Feature activation-3.80e-01

Pos logit contributions
isel+0.188
icker+0.182
oths+0.182
umped+0.174
apa+0.174
Neg logit contributions
rob-0.172
 l-0.142
 eighty-0.129
 Ske-0.126
 cand-0.124
Token down
Token ID866
Feature activation-3.86e-02

Pos logit contributions
rob+0.017
 l+0.014
 eighty+0.012
 cand+0.012
 Ske+0.012
Neg logit contributions
isel-0.018
icker-0.018
oths-0.017
imming-0.017
umped-0.017
Token,
Token ID11
Feature activation-6.33e-01

Pos logit contributions
rob+0.001
 l+0.001
 Ske+0.001
 eighty+0.001
 cand+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
imming-0.002
umped-0.002
Token the
Token ID262
Feature activation-9.63e-02

Pos logit contributions
rob+0.014
 l+0.010
 Ske+0.008
 eighty+0.007
 cand+0.006
Neg logit contributions
isel-0.044
icker-0.044
oths-0.043
imming-0.043
umped-0.042
Token cater
Token ID20825
Feature activation+8.13e-01

Pos logit contributions
rob+0.004
 l+0.003
 eighty+0.003
 cand+0.003
 Ske+0.003
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
umped-0.004
imming-0.004
Token She
Token ID1375
Feature activation+3.89e-01

Pos logit contributions
isel+0.017
icker+0.016
imming+0.016
oths+0.016
ench+0.015
Neg logit contributions
rob-0.011
 l-0.009
 Ske-0.008
 Velvet-0.008
 eighty-0.008
Token takes
Token ID2753
Feature activation-4.34e-01

Pos logit contributions
isel+0.018
icker+0.018
oths+0.018
apa+0.017
umped+0.017
Neg logit contributions
rob-0.017
 l-0.014
 eighty-0.013
 Ske-0.012
 cand-0.012
Token out
Token ID503
Feature activation+1.66e-01

Pos logit contributions
rob+0.021
 l+0.017
 eighty+0.016
 Ske+0.016
 cand+0.015
Neg logit contributions
isel-0.019
icker-0.019
oths-0.018
umped-0.018
apa-0.017
Token a
Token ID257
Feature activation+1.10e-01

Pos logit contributions
isel+0.009
oths+0.009
icker+0.009
imming+0.009
umped+0.009
Neg logit contributions
rob-0.006
 l-0.005
 eighty-0.004
 Ske-0.004
 cand-0.004
Token magn
Token ID7842
Feature activation+9.21e-02

Pos logit contributions
isel+0.005
oths+0.005
icker+0.005
imming+0.005
umped+0.005
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 Ske-0.004
 cand-0.004
Tokenifying
Token ID4035
Feature activation+3.92e+00

Pos logit contributions
oths+0.001
isel+0.001
icker+0.001
ooks+0.001
ench+0.001
Neg logit contributions
rob-0.007
 eighty-0.006
 l-0.006
 Ske-0.006
 seventy-0.006
Token glass
Token ID5405
Feature activation-2.14e-01

Pos logit contributions
isel+0.188
icker+0.184
oths+0.182
apa+0.175
umped+0.175
Neg logit contributions
rob-0.171
 l-0.137
 eighty-0.126
 Ske-0.122
 cand-0.121
Token.
Token ID13
Feature activation+1.37e-01

Pos logit contributions
rob+0.010
 l+0.008
 eighty+0.008
 Ske+0.007
 cand+0.007
Neg logit contributions
isel-0.010
icker-0.009
oths-0.009
apa-0.009
umped-0.009
Token She
Token ID1375
Feature activation+4.35e-01

Pos logit contributions
isel+0.007
imming+0.006
icker+0.006
oths+0.006
ench+0.006
Neg logit contributions
rob-0.005
 l-0.005
 Ske-0.004
 Velvet-0.004
 eighty-0.004
Token looks
Token ID3073
Feature activation+6.53e-01

Pos logit contributions
isel+0.021
icker+0.020
oths+0.020
apa+0.019
umped+0.019
Neg logit contributions
rob-0.019
 l-0.015
 eighty-0.014
 Ske-0.014
 cand-0.014
Token at
Token ID379
Feature activation+7.66e-01

Pos logit contributions
isel+0.026
icker+0.026
oths+0.025
apa+0.024
umped+0.024
Neg logit contributions
rob-0.034
 l-0.028
 eighty-0.026
 Ske-0.026
 cand-0.025
Token the
Token ID262
Feature activation+1.04e-01

Pos logit contributions
isel+0.009
icker+0.008
oths+0.008
aining+0.008
amps+0.007
Neg logit contributions
rob-0.009
 eighty-0.007
 l-0.007
 cand-0.007
 seventy-0.007
Token watch
Token ID2342
Feature activation-2.98e-01

Pos logit contributions
isel+0.005
icker+0.005
oths+0.005
umped+0.005
apa+0.005
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.003
 Ske-0.003
 cand-0.003
Token with
Token ID351
Feature activation+3.00e-02

Pos logit contributions
rob+0.012
 l+0.011
 eighty+0.009
 cand+0.009
 Ske+0.009
Neg logit contributions
isel-0.015
oths-0.014
icker-0.014
imming-0.014
umped-0.014
Token the
Token ID262
Feature activation+3.65e-02

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
apa+0.002
umped+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.001
 Ske-0.001
 cand-0.001
Token magn
Token ID7842
Feature activation-1.72e-01

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
umped+0.002
apa+0.002
Neg logit contributions
rob-0.002
 l-0.001
 eighty-0.001
 Ske-0.001
 cand-0.001
Tokenifying
Token ID4035
Feature activation+3.49e+00

Pos logit contributions
rob+0.012
 l+0.011
 eighty+0.011
 seventy+0.010
 Ske+0.010
Neg logit contributions
isel-0.003
oths-0.003
ooks-0.003
appiness-0.003
icker-0.003
Token glass
Token ID5405
Feature activation-1.14e-01

Pos logit contributions
isel+0.168
icker+0.165
oths+0.163
apa+0.157
umped+0.157
Neg logit contributions
rob-0.151
 l-0.120
 eighty-0.112
 Ske-0.108
 cand-0.107
Token.
Token ID13
Feature activation+1.78e-01

Pos logit contributions
rob+0.005
 l+0.004
 cand+0.004
 eighty+0.004
 Ske+0.004
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
imming-0.005
umped-0.005
Token He
Token ID679
Feature activation+4.93e-01

Pos logit contributions
isel+0.009
icker+0.009
oths+0.009
imming+0.009
apa+0.009
Neg logit contributions
rob-0.006
 l-0.005
 Ske-0.005
 eighty-0.005
 cand-0.005
Token saw
Token ID2497
Feature activation+8.59e-01

Pos logit contributions
isel+0.020
icker+0.020
oths+0.020
apa+0.019
umped+0.019
Neg logit contributions
rob-0.025
 eighty-0.019
 l-0.019
 cand-0.018
 Ske-0.018
Token the
Token ID262
Feature activation+2.56e-01

Pos logit contributions
isel+0.039
icker+0.037
oths+0.037
apa+0.035
umped+0.035
Neg logit contributions
rob-0.040
 eighty-0.030
 l-0.030
 seventy-0.027
 Ske-0.027
Token He
Token ID679
Feature activation+7.52e-01

Pos logit contributions
isel+0.032
icker+0.032
oths+0.031
imming+0.031
apa+0.030
Neg logit contributions
rob-0.029
 l-0.024
 Ske-0.022
 eighty-0.021
 cand-0.020
Token took
Token ID1718
Feature activation-3.79e-01

Pos logit contributions
isel+0.028
umped+0.027
apa+0.027
icker+0.027
oths+0.026
Neg logit contributions
rob-0.040
 eighty-0.029
 Velvet-0.029
 cand-0.028
 l-0.028
Token out
Token ID503
Feature activation+1.07e-01

Pos logit contributions
rob+0.016
 l+0.013
 eighty+0.012
 Ske+0.012
 cand+0.011
Neg logit contributions
isel-0.018
icker-0.018
oths-0.018
umped-0.017
imming-0.017
Token a
Token ID257
Feature activation+1.47e-02

Pos logit contributions
isel+0.006
icker+0.006
oths+0.006
apa+0.006
umped+0.005
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 Ske-0.003
 cand-0.003
Token magn
Token ID7842
Feature activation-1.43e-01

Pos logit contributions
isel+0.001
icker+0.001
oths+0.001
umped+0.001
apa+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.000
 Ske-0.000
 cand-0.000
Tokenifying
Token ID4035
Feature activation+3.49e+00

Pos logit contributions
rob+0.011
 l+0.010
 eighty+0.010
 Ske+0.009
 seventy+0.009
Neg logit contributions
isel-0.002
oths-0.002
icker-0.002
apa-0.001
ooks-0.001
Token glass
Token ID5405
Feature activation-3.03e-01

Pos logit contributions
isel+0.163
icker+0.160
oths+0.157
apa+0.151
umped+0.151
Neg logit contributions
rob-0.156
 l-0.124
 eighty-0.117
 Ske-0.112
 cand-0.111
Token and
Token ID290
Feature activation+2.58e-02

Pos logit contributions
rob+0.014
 l+0.011
 eighty+0.011
 Ske+0.010
 cand+0.010
Neg logit contributions
isel-0.014
icker-0.013
oths-0.013
apa-0.013
umped-0.013
Token a
Token ID257
Feature activation-2.22e-01

Pos logit contributions
isel+0.001
icker+0.001
oths+0.001
umped+0.001
imming+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.001
 Ske-0.001
 cand-0.001
Token notebook
Token ID20922
Feature activation-5.47e-01

Pos logit contributions
rob+0.009
 l+0.008
 eighty+0.007
 cand+0.007
 Ske+0.007
Neg logit contributions
isel-0.011
oths-0.011
icker-0.011
imming-0.011
umped-0.010
Token and
Token ID290
Feature activation-3.83e-01

Pos logit contributions
rob+0.028
 l+0.021
 eighty+0.020
 Ske+0.020
 Velvet+0.020
Neg logit contributions
isel-0.021
icker-0.021
oths-0.020
umped-0.020
ap-0.020
Token the
Token ID262
Feature activation+3.10e-01

Pos logit contributions
isel+0.014
icker+0.013
oths+0.013
umped+0.013
imming+0.013
Neg logit contributions
rob-0.007
 l-0.005
 eighty-0.004
 Ske-0.004
 cand-0.004
Token necklace
Token ID38987
Feature activation-3.19e-01

Pos logit contributions
isel+0.015
icker+0.015
oths+0.015
apa+0.014
ap+0.014
Neg logit contributions
rob-0.014
 l-0.009
 Ske-0.009
 eighty-0.009
 ninety-0.008
Token and
Token ID290
Feature activation+1.17e-01

Pos logit contributions
rob+0.014
 l+0.012
 eighty+0.011
 Ske+0.011
 cand+0.011
Neg logit contributions
isel-0.015
icker-0.014
oths-0.014
apa-0.014
umped-0.014
Token the
Token ID262
Feature activation+4.95e-02

Pos logit contributions
isel+0.005
oths+0.005
icker+0.005
apa+0.005
umped+0.005
Neg logit contributions
rob-0.006
 l-0.004
 eighty-0.004
 Ske-0.004
 cand-0.004
Token magn
Token ID7842
Feature activation+9.99e-02

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
apa+0.002
ap+0.002
Neg logit contributions
rob-0.002
 l-0.002
 Ske-0.002
 eighty-0.001
 Velvet-0.001
Tokenifying
Token ID4035
Feature activation+3.46e+00

Pos logit contributions
isel+0.002
oths+0.002
icker+0.002
apa+0.001
imming+0.001
Neg logit contributions
rob-0.007
 l-0.007
 eighty-0.006
 Ske-0.006
 cand-0.006
Token glass
Token ID5405
Feature activation-2.46e-02

Pos logit contributions
isel+0.161
icker+0.158
oths+0.155
umped+0.149
apa+0.149
Neg logit contributions
rob-0.155
 l-0.122
 eighty-0.115
 Ske-0.111
 cand-0.109
Token and
Token ID290
Feature activation-2.51e-02

Pos logit contributions
rob+0.001
 l+0.001
 eighty+0.001
 Ske+0.001
 cand+0.001
Neg logit contributions
isel-0.001
icker-0.001
oths-0.001
apa-0.001
umped-0.001
Token started
Token ID2067
Feature activation+4.70e-01

Pos logit contributions
rob+0.001
 l+0.001
 eighty+0.001
 Ske+0.001
 cand+0.001
Neg logit contributions
isel-0.001
icker-0.001
oths-0.001
apa-0.001
umped-0.001
Token to
Token ID284
Feature activation+9.19e-02

Pos logit contributions
isel+0.022
icker+0.022
oths+0.022
apa+0.021
umped+0.021
Neg logit contributions
rob-0.021
 l-0.017
 eighty-0.015
 Ske-0.015
 cand-0.015
Token examine
Token ID10716
Feature activation+1.82e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
umped+0.004
apa+0.004
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 Ske-0.003
 cand-0.003
Token flashlight
Token ID38371
Feature activation+5.40e-01

Pos logit contributions
rob+0.017
 l+0.014
 eighty+0.013
 Ske+0.012
 cand+0.012
Neg logit contributions
isel-0.021
icker-0.021
oths-0.021
umped-0.020
imming-0.020
Token,
Token ID11
Feature activation+2.56e-01

Pos logit contributions
isel+0.025
icker+0.024
oths+0.024
umped+0.023
apa+0.023
Neg logit contributions
rob-0.024
 l-0.020
 eighty-0.018
 Ske-0.018
 cand-0.018
Token and
Token ID290
Feature activation-1.70e-01

Pos logit contributions
isel+0.012
icker+0.012
oths+0.012
apa+0.011
umped+0.011
Neg logit contributions
rob-0.011
 l-0.009
 eighty-0.008
 Ske-0.008
 cand-0.008
Token a
Token ID257
Feature activation-1.79e-01

Pos logit contributions
rob+0.007
 l+0.005
 eighty+0.005
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.008
oths-0.008
icker-0.008
apa-0.008
umped-0.008
Token magn
Token ID7842
Feature activation-8.06e-02

Pos logit contributions
rob+0.007
 l+0.006
 eighty+0.005
 cand+0.005
 Ske+0.005
Neg logit contributions
isel-0.009
icker-0.009
oths-0.009
umped-0.009
imming-0.009
Tokenifying
Token ID4035
Feature activation+3.44e+00

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.005
 Ske+0.005
 seventy+0.005
Neg logit contributions
isel-0.001
oths-0.001
icker-0.001
apa-0.001
ooks-0.001
Token glass
Token ID5405
Feature activation-2.79e-01

Pos logit contributions
isel+0.163
icker+0.159
oths+0.158
apa+0.152
umped+0.151
Neg logit contributions
rob-0.152
 l-0.125
 eighty-0.114
 Ske-0.111
 cand-0.110
Token.
Token ID13
Feature activation-1.05e-01

Pos logit contributions
rob+0.012
 l+0.011
 eighty+0.009
 cand+0.009
 Ske+0.009
Neg logit contributions
isel-0.013
icker-0.013
oths-0.013
imming-0.012
umped-0.012
Token
Token ID198
Feature activation-2.17e-01

Pos logit contributions
rob+0.004
 l+0.003
 eighty+0.003
 Ske+0.003
 cand+0.003
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
imming-0.005
umped-0.005
Token
Token ID198
Feature activation+3.94e-02

Pos logit contributions
rob+0.009
 l+0.007
 Ske+0.006
 eighty+0.006
 Velvet+0.006
Neg logit contributions
isel-0.011
icker-0.011
oths-0.010
imming-0.010
umped-0.010
TokenAs
Token ID1722
Feature activation-1.50e-01

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
ench+0.002
imming+0.002
Neg logit contributions
rob-0.002
 l-0.001
 eighty-0.001
 Velvet-0.001
 Ske-0.001
Token,
Token ID11
Feature activation+4.85e-01

Pos logit contributions
 l+0.004
rob+0.003
 eighty+0.003
 he+0.003
 cand+0.003
Neg logit contributions
isel-0.008
oths-0.008
imming-0.007
icker-0.007
amps-0.007
Token handing
Token ID22786
Feature activation-5.51e-01

Pos logit contributions
isel+0.025
icker+0.024
oths+0.024
amps+0.023
imming+0.023
Neg logit contributions
rob-0.019
 l-0.018
 eighty-0.016
 Ske-0.015
 seventy-0.015
Token him
Token ID683
Feature activation+7.00e-01

Pos logit contributions
rob+0.021
 l+0.016
 eighty+0.015
 cand+0.014
 Ske+0.014
Neg logit contributions
isel-0.028
icker-0.028
oths-0.028
apa-0.027
umped-0.026
Token the
Token ID262
Feature activation-2.28e-01

Pos logit contributions
isel+0.038
icker+0.037
oths+0.037
umped+0.036
imming+0.035
Neg logit contributions
rob-0.026
 l-0.021
 eighty-0.019
 Ske-0.018
 cand-0.018
Token magn
Token ID7842
Feature activation-1.70e-01

Pos logit contributions
rob+0.009
 l+0.008
 eighty+0.007
 cand+0.007
 candy+0.007
Neg logit contributions
isel-0.011
oths-0.011
umped-0.011
icker-0.011
aining-0.011
Tokenifying
Token ID4035
Feature activation+3.37e+00

Pos logit contributions
rob+0.012
 l+0.011
 eighty+0.011
 Ske+0.011
 seventy+0.011
Neg logit contributions
isel-0.003
oths-0.003
icker-0.002
ooks-0.002
appiness-0.002
Token glass
Token ID5405
Feature activation-3.31e-01

Pos logit contributions
isel+0.165
icker+0.161
oths+0.160
apa+0.154
umped+0.154
Neg logit contributions
rob-0.143
 l-0.115
 eighty-0.105
 Ske-0.102
 cand-0.101
Token.
Token ID13
Feature activation+2.43e-02

Pos logit contributions
rob+0.015
 l+0.013
 eighty+0.012
 Ske+0.011
 cand+0.011
Neg logit contributions
isel-0.015
icker-0.015
oths-0.015
umped-0.014
apa-0.014
Token She
Token ID1375
Feature activation+2.83e-01

Pos logit contributions
isel+0.001
icker+0.001
imming+0.001
ench+0.001
oths+0.001
Neg logit contributions
rob-0.001
 l-0.001
 Velvet-0.001
 Ske-0.001
 Star-0.001
Token looked
Token ID3114
Feature activation+6.98e-01

Pos logit contributions
isel+0.013
icker+0.013
oths+0.013
apa+0.012
umped+0.012
Neg logit contributions
rob-0.013
 eighty-0.009
 l-0.009
 Ske-0.009
 cand-0.008
Token around
Token ID1088
Feature activation+1.39e-01

Pos logit contributions
isel+0.029
icker+0.029
oths+0.028
apa+0.027
umped+0.027
Neg logit contributions
rob-0.034
 l-0.030
 eighty-0.027
 Ske-0.026
 cand-0.026
Token put
Token ID1234
Feature activation-1.03e+00

Pos logit contributions
rob+0.004
 l+0.004
 eighty+0.003
 Ske+0.003
 cand+0.003
Neg logit contributions
isel-0.005
icker-0.005
oths-0.004
imming-0.004
umped-0.004
Token on
Token ID319
Feature activation+3.49e-01

Pos logit contributions
rob+0.043
 l+0.034
 eighty+0.031
 Ske+0.030
 cand+0.030
Neg logit contributions
isel-0.051
icker-0.050
oths-0.050
apa-0.048
umped-0.048
Token an
Token ID281
Feature activation-7.80e-01

Pos logit contributions
isel+0.023
icker+0.022
oths+0.022
umped+0.021
imming+0.021
Neg logit contributions
rob-0.009
 l-0.007
 eighty-0.006
 cand-0.006
 Ske-0.005
Token oven
Token ID14361
Feature activation-3.40e-01

Pos logit contributions
rob+0.037
 l+0.029
 Ske+0.027
 eighty+0.027
 Velvet+0.027
Neg logit contributions
isel-0.035
icker-0.034
oths-0.033
apa-0.033
imming-0.032
Token m
Token ID285
Feature activation+6.25e-01

Pos logit contributions
rob+0.014
 l+0.012
 eighty+0.010
 cand+0.010
 Ske+0.010
Neg logit contributions
isel-0.017
icker-0.017
oths-0.017
umped-0.016
imming-0.016
Tokenitt
Token ID715
Feature activation+3.36e+00

Pos logit contributions
itating+0.017
isel+0.016
ench+0.016
eways+0.015
adoes+0.015
Neg logit contributions
rob-0.036
 l-0.030
 cand-0.030
 eighty-0.029
 Ske-0.028
Token and
Token ID290
Feature activation-3.54e-01

Pos logit contributions
isel+0.168
icker+0.165
oths+0.162
ench+0.157
umped+0.156
Neg logit contributions
rob-0.136
 l-0.117
 cand-0.104
 eighty-0.100
 scar-0.098
Token a
Token ID257
Feature activation-3.74e-01

Pos logit contributions
rob+0.012
 l+0.012
 cand+0.010
 eighty+0.009
 scar+0.009
Neg logit contributions
imming-0.020
isel-0.019
aining-0.019
itating-0.019
ench-0.019
Token a
Token ID257
Feature activation-4.65e-01

Pos logit contributions
rob+0.015
 l+0.015
 cand+0.012
 eighty+0.012
 candy+0.012
Neg logit contributions
isel-0.018
imming-0.018
oths-0.018
umped-0.017
aining-0.017
Tokenpron
Token ID31186
Feature activation-6.09e-01

Pos logit contributions
rob+0.019
 l+0.018
 eighty+0.015
 cand+0.015
 candy+0.015
Neg logit contributions
isel-0.023
oths-0.022
imming-0.022
umped-0.022
aining-0.022
Token.
Token ID13
Feature activation-2.62e-02

Pos logit contributions
rob+0.027
 l+0.023
 eighty+0.020
 cand+0.020
 Ske+0.020
Neg logit contributions
isel-0.029
icker-0.028
oths-0.028
apa-0.027
umped-0.027
Token She
Token ID1375
Feature activation+3.06e-01

Pos logit contributions
rob+0.002
 l+0.002
 Ske+0.002
 eighty+0.002
 Velvet+0.001
Neg logit contributions
isel-0.003
icker-0.002
imming-0.002
oths-0.002
umped-0.002
Token picks
Token ID11103
Feature activation-5.37e-01

Pos logit contributions
isel+0.016
icker+0.016
oths+0.016
imming+0.015
umped+0.015
Neg logit contributions
rob-0.012
 l-0.010
 eighty-0.008
 Ske-0.008
 cand-0.008
Token up
Token ID510
Feature activation+3.19e-01

Pos logit contributions
rob+0.030
 l+0.024
 eighty+0.024
 Ske+0.023
 cand+0.023
Neg logit contributions
isel-0.019
icker-0.018
oths-0.018
apa-0.018
umped-0.017
Token the
Token ID262
Feature activation+3.29e-01

Pos logit contributions
isel+0.017
icker+0.017
oths+0.017
apa+0.016
umped+0.016
Neg logit contributions
rob-0.012
 l-0.009
 eighty-0.008
 Ske-0.008
 cand-0.008
Token re
Token ID302
Feature activation+1.91e+00

Pos logit contributions
isel+0.016
icker+0.016
oths+0.016
umped+0.015
apa+0.015
Neg logit contributions
rob-0.014
 l-0.011
 eighty-0.010
 Ske-0.010
 cand-0.010
Tokenind
Token ID521
Feature activation+3.33e+00

Pos logit contributions
isel+0.063
icker+0.061
oths+0.061
apa+0.057
umped+0.057
Neg logit contributions
rob-0.112
 l-0.096
 eighty-0.091
 Ske-0.089
 cand-0.088
Tokeneer
Token ID28153
Feature activation+1.44e+00

Pos logit contributions
isel+0.150
icker+0.147
oths+0.145
apa+0.140
umped+0.138
Neg logit contributions
rob-0.153
 l-0.126
 eighty-0.116
 Ske-0.114
 cand-0.112
Token.
Token ID13
Feature activation-3.45e-01

Pos logit contributions
isel+0.082
icker+0.078
oths+0.077
ooks+0.075
umped+0.072
Neg logit contributions
rob-0.050
 l-0.049
 eighty-0.040
 cand-0.039
 Ske-0.038
Token She
Token ID1375
Feature activation+1.82e-01

Pos logit contributions
rob+0.014
 l+0.012
 eighty+0.011
 Ske+0.011
 Velvet+0.010
Neg logit contributions
isel-0.017
icker-0.017
oths-0.017
imming-0.016
umped-0.016
Token looks
Token ID3073
Feature activation+5.18e-01

Pos logit contributions
isel+0.009
oths+0.009
icker+0.008
apa+0.008
umped+0.008
Neg logit contributions
rob-0.008
 l-0.006
 eighty-0.006
 Ske-0.006
 cand-0.005
Token at
Token ID379
Feature activation+5.62e-01

Pos logit contributions
isel+0.020
icker+0.020
oths+0.019
apa+0.019
umped+0.018
Neg logit contributions
rob-0.027
 l-0.023
 eighty-0.021
 Ske-0.021
 cand-0.021
Token She
Token ID1375
Feature activation+5.89e-01

Pos logit contributions
isel+0.032
icker+0.031
oths+0.031
imming+0.030
leys+0.030
Neg logit contributions
rob-0.014
 l-0.014
 Ske-0.013
 Velvet-0.012
 Star-0.011
Token took
Token ID1718
Feature activation-3.13e-01

Pos logit contributions
icker+0.028
isel+0.028
oths+0.028
imming+0.027
aining+0.027
Neg logit contributions
rob-0.025
 l-0.022
 eighty-0.019
 Ske-0.019
 cand-0.018
Token out
Token ID503
Feature activation+3.73e-01

Pos logit contributions
rob+0.014
 l+0.012
 Ske+0.011
 eighty+0.011
 cand+0.010
Neg logit contributions
isel-0.014
icker-0.014
oths-0.014
imming-0.013
umped-0.013
Token a
Token ID257
Feature activation+3.21e-01

Pos logit contributions
isel+0.021
icker+0.020
oths+0.020
umped+0.020
apa+0.020
Neg logit contributions
rob-0.013
 l-0.010
 eighty-0.009
 Ske-0.009
 cand-0.009
Token magn
Token ID7842
Feature activation-1.58e-02

Pos logit contributions
oths+0.015
isel+0.015
icker+0.015
umped+0.015
imming+0.015
Neg logit contributions
rob-0.013
 l-0.012
 eighty-0.010
 Ske-0.010
 candy-0.010
Tokenifying
Token ID4035
Feature activation+3.32e+00

Pos logit contributions
rob+0.001
 eighty+0.001
 l+0.001
 seventy+0.001
 Ske+0.001
Neg logit contributions
oths-0.000
isel-0.000
icker-0.000
ooks-0.000
appiness-0.000
Token glass
Token ID5405
Feature activation+1.88e-02

Pos logit contributions
isel+0.160
icker+0.157
oths+0.155
apa+0.149
umped+0.149
Neg logit contributions
rob-0.144
 l-0.116
 eighty-0.106
 Ske-0.103
 cand-0.102
Token and
Token ID290
Feature activation+1.78e-01

Pos logit contributions
isel+0.001
oths+0.001
icker+0.001
imming+0.001
umped+0.001
Neg logit contributions
rob-0.001
 l-0.001
 cand-0.001
 eighty-0.001
 Ske-0.001
Token looked
Token ID3114
Feature activation+9.02e-01

Pos logit contributions
icker+0.009
isel+0.009
imming+0.009
oths+0.009
aining+0.008
Neg logit contributions
rob-0.007
 l-0.007
 Ske-0.005
 cand-0.005
 eighty-0.005
Token very
Token ID845
Feature activation+1.28e+00

Pos logit contributions
oths+0.044
icker+0.043
isel+0.043
umped+0.043
imming+0.042
Neg logit contributions
 l-0.034
rob-0.034
 eighty-0.028
 Ske-0.028
 scar-0.027
Token carefully
Token ID7773
Feature activation+1.57e+00

Pos logit contributions
oths+0.063
imming+0.062
isel+0.062
umped+0.061
arre+0.061
Neg logit contributions
 l-0.045
 Ske-0.036
rob-0.035
 he-0.035
 w-0.034
Token took
Token ID1718
Feature activation-5.86e-01

Pos logit contributions
isel+0.017
icker+0.017
oths+0.017
imming+0.016
aining+0.016
Neg logit contributions
rob-0.014
 l-0.012
 eighty-0.011
 Ske-0.011
 cand-0.010
Token off
Token ID572
Feature activation+4.51e-01

Pos logit contributions
rob+0.025
 l+0.022
 eighty+0.019
 Ske+0.019
 cand+0.019
Neg logit contributions
isel-0.028
icker-0.027
oths-0.027
imming-0.027
leys-0.026
Token the
Token ID262
Feature activation+2.11e-01

Pos logit contributions
isel+0.028
icker+0.027
oths+0.027
imming+0.027
apa+0.026
Neg logit contributions
rob-0.013
 l-0.011
 cand-0.009
 Ske-0.009
 eighty-0.009
Token oven
Token ID14361
Feature activation+1.13e-01

Pos logit contributions
isel+0.010
imming+0.009
umped+0.009
oths+0.009
icker+0.009
Neg logit contributions
rob-0.009
 l-0.009
 cand-0.008
 candy-0.008
 eighty-0.007
Token m
Token ID285
Feature activation+1.06e+00

Pos logit contributions
isel+0.006
oths+0.005
icker+0.005
imming+0.005
apa+0.005
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 Ske-0.003
 cand-0.003
Tokenitt
Token ID715
Feature activation+3.30e+00

Pos logit contributions
isel+0.030
imming+0.027
eways+0.026
leys+0.026
ifying+0.025
Neg logit contributions
rob-0.061
 cand-0.054
 l-0.052
 eighty-0.050
 Ske-0.048
Tokens
Token ID82
Feature activation+5.64e-01

Pos logit contributions
isel+0.168
icker+0.161
oths+0.160
ench+0.154
imming+0.154
Neg logit contributions
rob-0.127
 l-0.114
 cand-0.103
 eighty-0.097
 scar-0.092
Token and
Token ID290
Feature activation-8.92e-02

Pos logit contributions
isel+0.033
icker+0.032
oths+0.032
imming+0.031
ooks+0.031
Neg logit contributions
 l-0.017
rob-0.016
 cand-0.014
 candy-0.013
 Ske-0.012
Token opened
Token ID4721
Feature activation+1.46e-01

Pos logit contributions
 l+0.003
rob+0.003
 cand+0.002
 eighty+0.002
 candy+0.002
Neg logit contributions
imming-0.005
isel-0.005
icker-0.005
aining-0.005
oths-0.005
Token the
Token ID262
Feature activation+6.48e-01

Pos logit contributions
icker+0.008
isel+0.008
oths+0.008
imming+0.008
arre+0.008
Neg logit contributions
rob-0.004
 l-0.004
 eighty-0.003
 Ske-0.003
 Velvet-0.003
Token oven
Token ID14361
Feature activation+6.35e-01

Pos logit contributions
isel+0.028
imming+0.027
umped+0.027
oths+0.026
icker+0.026
Neg logit contributions
 l-0.029
rob-0.028
 cand-0.025
 candy-0.024
 bar-0.023
Token said
Token ID531
Feature activation-8.02e-01

Pos logit contributions
rob+0.022
 eighty+0.016
 ninety+0.014
 cand+0.014
 l+0.013
Neg logit contributions
umped-0.025
apa-0.025
isel-0.025
oths-0.024
icker-0.024
Token,
Token ID11
Feature activation+1.94e-01

Pos logit contributions
rob+0.024
 l+0.024
 Ske+0.020
 eighty+0.020
 cand+0.019
Neg logit contributions
isel-0.045
oths-0.043
imming-0.043
icker-0.041
amps-0.041
Token holding
Token ID4769
Feature activation-7.71e-01

Pos logit contributions
isel+0.010
oths+0.010
icker+0.010
apa+0.010
leys+0.010
Neg logit contributions
 l-0.007
rob-0.007
 eighty-0.006
 seventy-0.005
 Ske-0.005
Token the
Token ID262
Feature activation-4.20e-02

Pos logit contributions
rob+0.028
 l+0.026
 Ske+0.022
 Velvet+0.021
 eighty+0.021
Neg logit contributions
isel-0.041
oths-0.040
icker-0.040
leys-0.039
ooks-0.038
Token magn
Token ID7842
Feature activation-3.37e-01

Pos logit contributions
rob+0.002
 l+0.002
 eighty+0.002
 Ske+0.002
 cand+0.002
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
umped-0.002
ooks-0.002
Tokenifying
Token ID4035
Feature activation+3.29e+00

Pos logit contributions
rob+0.023
 eighty+0.021
 l+0.021
 seventy+0.021
 Ske+0.020
Neg logit contributions
oths-0.006
isel-0.006
icker-0.005
ooks-0.005
appiness-0.005
Token glass
Token ID5405
Feature activation-4.81e-01

Pos logit contributions
isel+0.164
icker+0.160
oths+0.159
apa+0.153
umped+0.153
Neg logit contributions
rob-0.137
 l-0.111
 eighty-0.100
 Ske-0.098
 cand-0.097
Token over
Token ID625
Feature activation-2.43e-01

Pos logit contributions
rob+0.018
 l+0.017
 eighty+0.015
 cand+0.015
 Ske+0.014
Neg logit contributions
isel-0.024
oths-0.024
icker-0.024
imming-0.023
apa-0.022
Token a
Token ID257
Feature activation-5.47e-01

Pos logit contributions
rob+0.007
 l+0.006
 Ske+0.005
 eighty+0.005
 Velvet+0.004
Neg logit contributions
isel-0.015
icker-0.015
ench-0.015
imming-0.015
oths-0.014
Token black
Token ID2042
Feature activation-2.11e-03

Pos logit contributions
rob+0.022
 l+0.021
 eighty+0.017
 cand+0.017
 candy+0.017
Neg logit contributions
isel-0.027
oths-0.027
icker-0.027
imming-0.026
umped-0.026
Token ant
Token ID1885
Feature activation-2.85e-01

Pos logit contributions
rob+0.000
 l+0.000
 eighty+0.000
 Ske+0.000
 cand+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
apa-0.000
umped-0.000
Token a
Token ID257
Feature activation-4.53e-01

Pos logit contributions
rob+0.001
 l+0.000
 eighty+0.000
 Ske+0.000
 cand+0.000
Neg logit contributions
isel-0.001
oths-0.001
icker-0.001
apa-0.001
umped-0.001
Token flashlight
Token ID38371
Feature activation+1.29e-01

Pos logit contributions
rob+0.021
 l+0.017
 eighty+0.016
 Ske+0.015
 cand+0.015
Neg logit contributions
isel-0.021
icker-0.020
oths-0.020
umped-0.019
apa-0.019
Token and
Token ID290
Feature activation-7.06e-01

Pos logit contributions
isel+0.006
icker+0.006
oths+0.006
apa+0.006
umped+0.006
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 Ske-0.004
 cand-0.004
Token a
Token ID257
Feature activation-2.53e-01

Pos logit contributions
rob+0.025
 l+0.021
 cand+0.017
 eighty+0.017
 Ske+0.017
Neg logit contributions
isel-0.039
icker-0.039
oths-0.038
imming-0.037
umped-0.037
Token magn
Token ID7842
Feature activation-1.09e-01

Pos logit contributions
rob+0.011
 l+0.009
 eighty+0.008
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.012
icker-0.012
oths-0.012
umped-0.012
imming-0.012
Tokenifying
Token ID4035
Feature activation+3.29e+00

Pos logit contributions
rob+0.008
 l+0.007
 eighty+0.007
 Ske+0.007
 cand+0.007
Neg logit contributions
isel-0.001
oths-0.001
icker-0.001
apa-0.001
umped-0.001
Token glass
Token ID5405
Feature activation-2.82e-01

Pos logit contributions
isel+0.141
icker+0.140
oths+0.134
ench+0.132
ap+0.129
Neg logit contributions
rob-0.160
 l-0.123
 eighty-0.121
 cand-0.115
 Ske-0.114
Token to
Token ID284
Feature activation-8.87e-01

Pos logit contributions
rob+0.013
 l+0.011
 eighty+0.010
 cand+0.010
 Ske+0.009
Neg logit contributions
isel-0.013
icker-0.013
oths-0.013
umped-0.012
apa-0.012
Token examine
Token ID10716
Feature activation+6.56e-02

Pos logit contributions
rob+0.039
 l+0.034
 Ske+0.030
 eighty+0.029
 cand+0.028
Neg logit contributions
isel-0.041
oths-0.040
icker-0.040
umped-0.039
imming-0.039
Token their
Token ID511
Feature activation+6.90e-01

Pos logit contributions
isel+0.003
icker+0.003
oths+0.003
umped+0.003
imming+0.003
Neg logit contributions
rob-0.003
 l-0.002
 Ske-0.002
 cand-0.002
 eighty-0.002
Token finds
Token ID7228
Feature activation-4.22e-01

Pos logit contributions
isel+0.030
icker+0.029
oths+0.028
umped+0.028
imming+0.027
Neg logit contributions
rob-0.032
 l-0.029
 Ske-0.026
 cand-0.026
 eighty-0.025
Token
Token ID220
Feature activation-1.85e-01

Pos logit contributions
isel+0.026
icker+0.026
imming+0.025
oths+0.025
ench+0.024
Neg logit contributions
rob-0.013
 l-0.011
 Ske-0.011
 eighty-0.010
 Velvet-0.010
Token She
Token ID1375
Feature activation+5.28e-01

Pos logit contributions
rob+0.006
 Ske+0.005
 l+0.005
 Velvet+0.005
 eighty+0.004
Neg logit contributions
isel-0.011
icker-0.010
oths-0.009
imming-0.009
ifying-0.009
Token grabbed
Token ID13176
Feature activation-6.90e-01

Pos logit contributions
icker+0.027
isel+0.027
oths+0.027
imming+0.025
ooks+0.025
Neg logit contributions
rob-0.021
 l-0.019
 eighty-0.016
 Ske-0.016
 cand-0.015
Token oven
Token ID14361
Feature activation+3.03e-01

Pos logit contributions
rob+0.021
 l+0.019
 Ske+0.015
 eighty+0.014
 cand+0.014
Neg logit contributions
icker-0.040
isel-0.040
oths-0.040
imming-0.039
ench-0.039
Token m
Token ID285
Feature activation+1.10e+00

Pos logit contributions
isel+0.016
oths+0.016
icker+0.015
imming+0.015
leys+0.015
Neg logit contributions
rob-0.011
 l-0.010
 cand-0.009
 eighty-0.008
 Ske-0.008
Tokenitt
Token ID715
Feature activation+3.28e+00

Pos logit contributions
isel+0.029
ench+0.026
ifying+0.026
itating+0.026
eways+0.025
Neg logit contributions
rob-0.064
 cand-0.057
 l-0.053
 eighty-0.052
 seventy-0.050
Tokens
Token ID82
Feature activation+1.09e+00

Pos logit contributions
isel+0.155
icker+0.155
oths+0.148
ench+0.146
appiness+0.142
Neg logit contributions
rob-0.135
 l-0.120
 cand-0.109
 eighty-0.104
 scar-0.102
Token and
Token ID290
Feature activation+2.06e-01

Pos logit contributions
isel+0.067
icker+0.065
oths+0.064
leys+0.063
ooks+0.063
Neg logit contributions
 l-0.029
 cand-0.022
rob-0.021
 candy-0.020
 eighty-0.019
Token carefully
Token ID7773
Feature activation+8.27e-01

Pos logit contributions
imming+0.012
icker+0.011
isel+0.011
oths+0.011
ench+0.011
Neg logit contributions
 l-0.007
rob-0.006
 cand-0.006
 eighty-0.005
 scar-0.005
Token pulled
Token ID5954
Feature activation-2.64e-01

Pos logit contributions
isel+0.042
icker+0.042
imming+0.042
oths+0.041
itating+0.041
Neg logit contributions
 l-0.029
rob-0.028
 cand-0.024
 eighty-0.023
 seventy-0.022
Token out
Token ID503
Feature activation+7.04e-01

Pos logit contributions
 l+0.009
rob+0.009
 Ske+0.008
 Velvet+0.007
 cand+0.007
Neg logit contributions
isel-0.013
icker-0.013
imming-0.013
oths-0.013
ench-0.013
Token cried
Token ID16896
Feature activation-9.19e-01

Pos logit contributions
rob+0.010
 l+0.008
 eighty+0.007
 Ske+0.007
 cand+0.007
Neg logit contributions
isel-0.013
icker-0.012
oths-0.012
apa-0.012
imming-0.012
Token,
Token ID11
Feature activation+5.09e-01

Pos logit contributions
rob+0.040
 l+0.034
 Ske+0.030
 eighty+0.030
 Velvet+0.029
Neg logit contributions
isel-0.043
oths-0.042
icker-0.041
imming-0.040
umped-0.040
Token dropping
Token ID12047
Feature activation-2.49e-01

Pos logit contributions
isel+0.029
icker+0.028
oths+0.028
apa+0.028
ooks+0.027
Neg logit contributions
 l-0.017
rob-0.015
 eighty-0.013
 w-0.012
 Velvet-0.012
Token the
Token ID262
Feature activation+3.55e-01

Pos logit contributions
rob+0.007
 l+0.006
 Ske+0.005
 Velvet+0.005
 candy+0.005
Neg logit contributions
isel-0.015
icker-0.015
oths-0.015
imming-0.015
leys-0.015
Token magn
Token ID7842
Feature activation-9.96e-03

Pos logit contributions
isel+0.016
icker+0.016
oths+0.016
umped+0.016
ooks+0.016
Neg logit contributions
rob-0.015
 l-0.015
 candy-0.012
 cand-0.012
 Ske-0.012
Tokenifying
Token ID4035
Feature activation+3.27e+00

Pos logit contributions
rob+0.001
 eighty+0.001
 l+0.001
 seventy+0.001
 Ske+0.001
Neg logit contributions
isel-0.000
oths-0.000
ooks-0.000
appiness-0.000
ealing-0.000
Token glass
Token ID5405
Feature activation-1.14e-01

Pos logit contributions
isel+0.166
icker+0.161
oths+0.161
imming+0.156
apa+0.155
Neg logit contributions
rob-0.133
 l-0.111
 Ske-0.097
 eighty-0.097
 cand-0.095
Token.
Token ID13
Feature activation+1.18e-01

Pos logit contributions
rob+0.004
 l+0.004
 cand+0.003
 candy+0.003
 Ske+0.003
Neg logit contributions
isel-0.006
oths-0.006
icker-0.006
imming-0.006
ooks-0.006
Token
Token ID198
Feature activation+1.42e-01

Pos logit contributions
isel+0.006
imming+0.006
icker+0.006
oths+0.006
ench+0.006
Neg logit contributions
rob-0.004
 l-0.004
 Ske-0.004
 Velvet-0.003
 Star-0.003
Token
Token ID198
Feature activation+2.97e-01

Pos logit contributions
isel+0.008
imming+0.008
icker+0.008
ench+0.008
oths+0.008
Neg logit contributions
rob-0.005
 l-0.004
 Ske-0.003
 Velvet-0.003
 Star-0.003
Token"
Token ID1
Feature activation-1.92e-01

Pos logit contributions
isel+0.017
ench+0.016
icker+0.015
ains+0.015
eways+0.015
Neg logit contributions
rob-0.009
 l-0.008
 Velvet-0.008
 he-0.007
 Ske-0.007
Token and
Token ID290
Feature activation-3.10e-01

Pos logit contributions
ap+0.012
isel+0.012
apa+0.011
oths+0.011
icker+0.011
Neg logit contributions
rob-0.022
 eighty-0.017
 l-0.016
 Ske-0.016
 seventy-0.015
Token handed
Token ID10158
Feature activation-7.27e-02

Pos logit contributions
rob+0.014
 l+0.010
 eighty+0.010
 Ske+0.010
 cand+0.009
Neg logit contributions
isel-0.014
icker-0.014
apa-0.014
oths-0.014
umped-0.013
Token him
Token ID683
Feature activation+6.23e-01

Pos logit contributions
rob+0.004
 l+0.003
 eighty+0.003
 cand+0.002
 Ske+0.002
Neg logit contributions
apa-0.003
isel-0.003
oths-0.003
icker-0.003
amps-0.003
Token the
Token ID262
Feature activation+2.88e-02

Pos logit contributions
isel+0.035
icker+0.034
oths+0.033
umped+0.032
aining+0.032
Neg logit contributions
rob-0.022
 l-0.019
 eighty-0.016
 Ske-0.016
 cand-0.016
Token magn
Token ID7842
Feature activation-4.51e-02

Pos logit contributions
isel+0.001
oths+0.001
aining+0.001
upid+0.001
icker+0.001
Neg logit contributions
rob-0.001
 Ske-0.001
 l-0.001
 Dr-0.001
 Velvet-0.001
Tokenifying
Token ID4035
Feature activation+3.24e+00

Pos logit contributions
rob+0.003
 l+0.003
 eighty+0.003
 Ske+0.003
 cand+0.003
Neg logit contributions
isel-0.001
icker-0.001
oths-0.001
apa-0.001
imming-0.001
Token glass
Token ID5405
Feature activation-2.91e-01

Pos logit contributions
isel+0.155
icker+0.153
oths+0.151
apa+0.145
umped+0.145
Neg logit contributions
rob-0.141
 l-0.112
 eighty-0.104
 Ske-0.100
 cand-0.099
Token.
Token ID13
Feature activation-1.02e-01

Pos logit contributions
rob+0.014
 l+0.012
 eighty+0.011
 Ske+0.011
 cand+0.010
Neg logit contributions
isel-0.012
icker-0.012
oths-0.012
apa-0.011
umped-0.011
Token
Token ID198
Feature activation-1.98e-01

Pos logit contributions
rob+0.005
 l+0.004
 eighty+0.004
 cand+0.003
 Ske+0.003
Neg logit contributions
icker-0.005
oths-0.005
isel-0.005
apa-0.004
umped-0.004
Token
Token ID198
Feature activation-7.11e-02

Pos logit contributions
rob+0.011
 l+0.009
 eighty+0.009
 cand+0.009
 ninety+0.008
Neg logit contributions
icker-0.007
oths-0.007
apa-0.007
isel-0.007
umped-0.007
TokenHe
Token ID1544
Feature activation-4.35e-02

Pos logit contributions
rob+0.004
 l+0.003
 eighty+0.003
 Ske+0.003
 cand+0.003
Neg logit contributions
icker-0.003
isel-0.003
oths-0.003
apa-0.003
umped-0.003
Token the
Token ID262
Feature activation+3.45e-02

Pos logit contributions
isel+0.046
icker+0.045
oths+0.044
apa+0.043
umped+0.043
Neg logit contributions
rob-0.037
 l-0.029
 eighty-0.027
 Ske-0.025
 cand-0.025
Token necklace
Token ID38987
Feature activation-4.91e-01

Pos logit contributions
isel+0.002
oths+0.001
aining+0.001
icker+0.001
ap+0.001
Neg logit contributions
rob-0.002
 Ske-0.001
 l-0.001
 Ping-0.001
 Velvet-0.001
Token and
Token ID290
Feature activation-7.19e-02

Pos logit contributions
rob+0.026
 l+0.020
 eighty+0.019
 Ske+0.019
 cand+0.018
Neg logit contributions
isel-0.019
icker-0.019
oths-0.018
umped-0.018
apa-0.018
Token the
Token ID262
Feature activation-1.06e-01

Pos logit contributions
rob+0.003
 eighty+0.002
 Ske+0.002
 l+0.002
 ninety+0.002
Neg logit contributions
isel-0.003
apa-0.003
oths-0.003
umped-0.003
icker-0.003
Token magn
Token ID7842
Feature activation+2.31e-02

Pos logit contributions
rob+0.005
 Ske+0.003
 l+0.003
 eighty+0.003
 Velvet+0.003
Neg logit contributions
isel-0.005
oths-0.005
icker-0.005
ap-0.005
apa-0.005
Tokenifying
Token ID4035
Feature activation+3.23e+00

Pos logit contributions
isel+0.000
icker+0.000
oths+0.000
apa+0.000
umped+0.000
Neg logit contributions
rob-0.002
 l-0.002
 eighty-0.002
 Ske-0.001
 cand-0.001
Token glass
Token ID5405
Feature activation-1.18e-01

Pos logit contributions
isel+0.144
icker+0.143
oths+0.140
umped+0.134
apa+0.134
Neg logit contributions
rob-0.151
 l-0.115
 eighty-0.112
 Ske-0.107
 cand-0.104
Token and
Token ID290
Feature activation+1.71e-02

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.005
 Ske+0.004
 cand+0.004
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
apa-0.004
umped-0.004
Token told
Token ID1297
Feature activation+1.88e-01

Pos logit contributions
isel+0.001
oths+0.001
apa+0.001
icker+0.001
umped+0.001
Neg logit contributions
rob-0.001
 eighty-0.001
 l-0.001
 Ske-0.001
 Velvet-0.001
Token him
Token ID683
Feature activation+7.89e-01

Pos logit contributions
isel+0.008
apa+0.008
icker+0.008
oths+0.008
amps+0.008
Neg logit contributions
rob-0.009
 eighty-0.007
 l-0.007
 cand-0.006
 seventy-0.006
Token that
Token ID326
Feature activation+8.49e-01

Pos logit contributions
isel+0.042
icker+0.041
oths+0.041
umped+0.039
imming+0.039
Neg logit contributions
rob-0.029
 l-0.025
 Ske-0.022
 eighty-0.022
 cand-0.022
Token you
Token ID345
Feature activation+2.39e-01

Pos logit contributions
rob+0.003
 l+0.003
 Ske+0.002
 cand+0.002
 Velvet+0.002
Neg logit contributions
isel-0.009
ooks-0.008
umped-0.008
oths-0.008
icker-0.008
Token to
Token ID284
Feature activation-3.82e-01

Pos logit contributions
isel+0.016
oths+0.015
icker+0.015
umped+0.014
imming+0.014
Neg logit contributions
 l-0.006
rob-0.005
 eighty-0.004
 Ske-0.004
 candy-0.004
Token the
Token ID262
Feature activation-1.37e-01

Pos logit contributions
rob+0.014
 l+0.012
 Ske+0.010
 eighty+0.010
 Velvet+0.010
Neg logit contributions
isel-0.021
oths-0.020
icker-0.020
umped-0.020
leys-0.020
Token brave
Token ID14802
Feature activation+7.79e-02

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.004
 Ske+0.004
 cand+0.004
Neg logit contributions
isel-0.007
icker-0.006
oths-0.006
umped-0.006
apa-0.006
Token l
Token ID300
Feature activation-5.74e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
apa+0.003
umped+0.003
Neg logit contributions
rob-0.003
 l-0.003
 eighty-0.003
 Ske-0.002
 cand-0.002
Tokenitt
Token ID715
Feature activation+3.17e+00

Pos logit contributions
rob+0.022
 l+0.018
 Ske+0.016
 eighty+0.016
 Velvet+0.014
Neg logit contributions
isel-0.030
oths-0.029
icker-0.028
umped-0.028
aining-0.028
Tokenlest
Token ID32712
Feature activation-7.13e-02

Pos logit contributions
isel+0.155
oths+0.150
icker+0.149
umped+0.144
imming+0.142
Neg logit contributions
rob-0.133
 l-0.119
 cand-0.106
 eighty-0.100
 candy-0.095
Token one
Token ID530
Feature activation-6.80e-01

Pos logit contributions
rob+0.003
 l+0.003
 eighty+0.003
 Ske+0.002
 cand+0.002
Neg logit contributions
isel-0.003
icker-0.003
oths-0.003
apa-0.003
umped-0.003
Token!
Token ID0
Feature activation-2.55e-01

Pos logit contributions
rob+0.029
 l+0.024
 eighty+0.021
 Ske+0.021
 cand+0.020
Neg logit contributions
isel-0.034
icker-0.033
oths-0.033
umped-0.031
apa-0.031
Token<|endoftext|>
Token ID50256
Feature activation+2.32e-01

Pos logit contributions
rob+0.011
 l+0.009
 Ske+0.008
 eighty+0.008
 Velvet+0.008
Neg logit contributions
isel-0.013
icker-0.012
oths-0.012
imming-0.012
leys-0.011
TokenM
Token ID44
Feature activation-1.83e-01

Pos logit contributions
isel+0.012
icker+0.011
oths+0.011
apa+0.011
umped+0.011
Neg logit contributions
rob-0.010
 l-0.008
 eighty-0.007
 Ske-0.007
 cand-0.007
Token They
Token ID1119
Feature activation-9.61e-02

Pos logit contributions
rob+0.024
 l+0.023
 Velvet+0.022
 Ske+0.022
 cand+0.021
Neg logit contributions
isel-0.027
imming-0.025
icker-0.024
leys-0.024
oths-0.024
Token put
Token ID1234
Feature activation-1.11e+00

Pos logit contributions
rob+0.004
 l+0.003
 eighty+0.003
 cand+0.003
 Ske+0.003
Neg logit contributions
isel-0.005
oths-0.005
icker-0.005
aining-0.005
imming-0.005
Token on
Token ID319
Feature activation+2.04e-01

Pos logit contributions
rob+0.048
 l+0.039
 eighty+0.036
 cand+0.034
 Ske+0.034
Neg logit contributions
isel-0.054
icker-0.053
oths-0.053
umped-0.051
apa-0.050
Token oven
Token ID14361
Feature activation+1.25e-01

Pos logit contributions
isel+0.011
icker+0.011
oths+0.011
aining+0.010
imming+0.010
Neg logit contributions
rob-0.008
 l-0.007
 cand-0.006
 eighty-0.006
 candy-0.005
Token m
Token ID285
Feature activation+7.60e-01

Pos logit contributions
isel+0.006
icker+0.006
oths+0.006
imming+0.006
umped+0.006
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 cand-0.004
 Ske-0.004
Tokenitt
Token ID715
Feature activation+3.12e+00

Pos logit contributions
isel+0.021
itating+0.019
ench+0.019
icker+0.019
aining+0.019
Neg logit contributions
rob-0.046
 cand-0.038
 l-0.037
 eighty-0.037
 Ske-0.037
Tokens
Token ID82
Feature activation+6.90e-01

Pos logit contributions
isel+0.145
icker+0.142
oths+0.138
ench+0.133
umped+0.133
Neg logit contributions
rob-0.140
 l-0.119
 cand-0.107
 eighty-0.105
 scar-0.101
Token and
Token ID290
Feature activation-3.38e-01

Pos logit contributions
isel+0.044
imming+0.041
icker+0.041
itating+0.040
aining+0.040
Neg logit contributions
 l-0.019
 cand-0.016
rob-0.015
 candy-0.014
 ice-0.013
Token took
Token ID1718
Feature activation-5.32e-01

Pos logit contributions
 l+0.012
rob+0.011
 cand+0.010
 candy+0.009
 eighty+0.009
Neg logit contributions
imming-0.018
aining-0.018
isel-0.018
icker-0.018
umped-0.018
Token out
Token ID503
Feature activation+3.34e-01

Pos logit contributions
rob+0.021
 l+0.019
 cand+0.017
 eighty+0.017
 Ske+0.016
Neg logit contributions
isel-0.026
icker-0.025
oths-0.025
imming-0.025
aining-0.025
Token the
Token ID262
Feature activation+2.38e-01

Pos logit contributions
isel+0.021
imming+0.020
icker+0.020
oths+0.020
umped+0.020
Neg logit contributions
 l-0.009
rob-0.009
 cand-0.007
 candy-0.007
 eighty-0.007
Token to
Token ID284
Feature activation-2.50e-01

Pos logit contributions
rob+0.002
 l+0.001
 eighty+0.001
 Ske+0.001
 cand+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
imming-0.002
umped-0.002
Token always
Token ID1464
Feature activation+2.27e-01

Pos logit contributions
rob+0.009
 l+0.009
 eighty+0.007
 Ske+0.007
 cand+0.007
Neg logit contributions
oths-0.013
icker-0.013
leys-0.012
umped-0.012
isel-0.012
Token keep
Token ID1394
Feature activation-3.76e-01

Pos logit contributions
oths+0.012
icker+0.012
umped+0.012
eways+0.012
isel+0.012
Neg logit contributions
rob-0.008
 l-0.008
 eighty-0.007
 candy-0.006
 cand-0.006
Token his
Token ID465
Feature activation-1.13e-01

Pos logit contributions
rob+0.010
 l+0.007
 cand+0.005
 Ske+0.005
 candy+0.005
Neg logit contributions
icker-0.025
isel-0.025
oths-0.024
apa-0.024
umped-0.024
Token useful
Token ID4465
Feature activation-9.67e-01

Pos logit contributions
 l+0.003
rob+0.003
 candy+0.003
 cand+0.003
 scar+0.003
Neg logit contributions
umped-0.006
icker-0.006
itating-0.006
ench-0.006
leys-0.006
Token tool
Token ID2891
Feature activation-2.62e+00

Pos logit contributions
rob+0.035
 l+0.034
 candy+0.030
 cand+0.029
 eighty+0.028
Neg logit contributions
icker-0.049
umped-0.047
itating-0.047
leys-0.047
oths-0.046
Token with
Token ID351
Feature activation+3.66e-01

Pos logit contributions
rob+0.094
 l+0.089
 cand+0.079
 candy+0.077
 eighty+0.077
Neg logit contributions
isel-0.133
oths-0.132
icker-0.129
imming-0.124
umped-0.122
Token him
Token ID683
Feature activation+1.49e-01

Pos logit contributions
icker+0.025
isel+0.024
oths+0.024
umped+0.024
itating+0.024
Neg logit contributions
rob-0.007
 l-0.007
 candy-0.005
 cand-0.005
 Ske-0.004
Token,
Token ID11
Feature activation+5.61e-01

Pos logit contributions
isel+0.008
icker+0.008
oths+0.008
ooks+0.008
umped+0.007
Neg logit contributions
rob-0.005
 l-0.005
 cand-0.004
 Ske-0.004
 candy-0.004
Token just
Token ID655
Feature activation+8.84e-01

Pos logit contributions
icker+0.030
isel+0.030
oths+0.029
imming+0.028
apa+0.028
Neg logit contributions
rob-0.021
 l-0.018
 eighty-0.016
 Ske-0.015
 cand-0.015
Token in
Token ID287
Feature activation+7.19e-02

Pos logit contributions
isel+0.047
icker+0.046
oths+0.046
umped+0.045
apa+0.044
Neg logit contributions
rob-0.032
 l-0.028
 eighty-0.025
 Ske-0.024
 cand-0.023
Token He
Token ID679
Feature activation-1.33e-01

Pos logit contributions
isel+0.009
icker+0.009
oths+0.009
imming+0.009
apa+0.009
Neg logit contributions
rob-0.005
 l-0.004
 Ske-0.004
 eighty-0.004
 cand-0.004
Token realized
Token ID6939
Feature activation+1.23e-01

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.005
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.006
icker-0.006
oths-0.006
apa-0.005
umped-0.005
Token that
Token ID326
Feature activation+5.83e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
apa+0.007
umped+0.007
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 Ske-0.003
 cand-0.003
Token sharing
Token ID7373
Feature activation-5.69e-01

Pos logit contributions
isel+0.031
icker+0.030
oths+0.030
apa+0.029
umped+0.029
Neg logit contributions
rob-0.022
 l-0.017
 eighty-0.016
 Ske-0.015
 cand-0.015
Token his
Token ID465
Feature activation-6.05e-01

Pos logit contributions
rob+0.023
 l+0.019
 cand+0.016
 eighty+0.016
 Ske+0.016
Neg logit contributions
isel-0.030
icker-0.029
oths-0.029
apa-0.028
umped-0.028
Token tool
Token ID2891
Feature activation-2.54e+00

Pos logit contributions
rob+0.026
 l+0.025
 candy+0.022
 Ske+0.021
 cand+0.021
Neg logit contributions
umped-0.027
itating-0.027
icker-0.026
isel-0.026
oths-0.026
Token with
Token ID351
Feature activation-5.08e-01

Pos logit contributions
rob+0.104
 l+0.088
 cand+0.077
 Ske+0.076
 eighty+0.076
Neg logit contributions
isel-0.127
icker-0.124
oths-0.122
imming-0.118
umped-0.116
Token Billy
Token ID15890
Feature activation-2.23e-01

Pos logit contributions
rob+0.017
 l+0.014
 cand+0.012
 candy+0.012
 Ske+0.012
Neg logit contributions
isel-0.029
icker-0.029
oths-0.028
imming-0.028
umped-0.028
Token would
Token ID561
Feature activation-4.90e-02

Pos logit contributions
rob+0.011
 l+0.009
 eighty+0.008
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.009
icker-0.009
oths-0.009
apa-0.009
umped-0.009
Token have
Token ID423
Feature activation-3.67e-01

Pos logit contributions
rob+0.002
 l+0.002
 eighty+0.002
 Ske+0.002
 cand+0.002
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
apa-0.002
umped-0.002
Token been
Token ID587
Feature activation+7.52e-02

Pos logit contributions
rob+0.014
 l+0.012
 cand+0.011
 eighty+0.010
 Ske+0.010
Neg logit contributions
isel-0.019
icker-0.019
oths-0.019
imming-0.018
apa-0.018
Token see
Token ID766
Feature activation+4.65e-01

Pos logit contributions
rob+0.012
 l+0.010
 eighty+0.009
 Ske+0.009
 cand+0.009
Neg logit contributions
isel-0.014
icker-0.013
oths-0.013
umped-0.013
apa-0.013
Token and
Token ID290
Feature activation-2.00e-01

Pos logit contributions
isel+0.027
icker+0.026
imming+0.024
umped+0.024
oths+0.024
Neg logit contributions
rob-0.017
 l-0.015
 cand-0.013
 candy-0.013
 Ske-0.012
Token do
Token ID466
Feature activation+1.50e-01

Pos logit contributions
rob+0.008
 l+0.007
 cand+0.006
 eighty+0.006
 Ske+0.006
Neg logit contributions
isel-0.010
icker-0.010
oths-0.010
umped-0.010
imming-0.010
Token.
Token ID13
Feature activation-1.24e-01

Pos logit contributions
isel+0.008
icker+0.007
oths+0.007
umped+0.007
apa+0.007
Neg logit contributions
rob-0.006
 l-0.005
 cand-0.005
 eighty-0.004
 Ske-0.004
Token From
Token ID3574
Feature activation+1.43e-01

Pos logit contributions
rob+0.006
 l+0.005
 Ske+0.005
 eighty+0.005
 Velvet+0.005
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
umped-0.005
imming-0.005
Token time
Token ID640
Feature activation-2.52e+00

Pos logit contributions
isel+0.008
icker+0.008
oths+0.007
apa+0.007
umped+0.007
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 cand-0.004
 Ske-0.004
Token to
Token ID284
Feature activation-6.71e-01

Pos logit contributions
rob+0.075
 l+0.066
 eighty+0.065
 Star+0.062
 Ske+0.062
Neg logit contributions
isel-0.148
icker-0.141
ench-0.137
oths-0.135
imming-0.134
Token time
Token ID640
Feature activation-2.25e+00

Pos logit contributions
rob+0.028
 l+0.024
 eighty+0.021
 Ske+0.021
 cand+0.020
Neg logit contributions
isel-0.033
icker-0.033
oths-0.032
umped-0.031
apa-0.031
Token people
Token ID661
Feature activation+3.77e-01

Pos logit contributions
rob+0.076
 l+0.067
 eighty+0.061
 Ske+0.059
 Star+0.058
Neg logit contributions
isel-0.127
icker-0.121
oths-0.116
imming-0.114
ooks-0.113
Token visited
Token ID8672
Feature activation-1.41e-01

Pos logit contributions
isel+0.020
icker+0.019
oths+0.019
aining+0.019
ooks+0.018
Neg logit contributions
rob-0.015
 l-0.013
 eighty-0.011
 cand-0.011
 Ske-0.011
Token the
Token ID262
Feature activation-1.91e-01

Pos logit contributions
rob+0.005
 l+0.004
 Ske+0.004
 cand+0.004
 Star+0.004
Neg logit contributions
isel-0.008
icker-0.008
oths-0.008
imming-0.007
umped-0.007
TokenNo
Token ID2949
Feature activation-6.94e-01

Pos logit contributions
rob+0.016
 eighty+0.013
 l+0.013
 cand+0.012
 Ske+0.012
Neg logit contributions
isel-0.011
icker-0.011
oths-0.011
umped-0.011
aining-0.010
Token,
Token ID11
Feature activation-2.37e-01

Pos logit contributions
rob+0.038
 eighty+0.030
 l+0.030
 cand+0.028
 Ske+0.028
Neg logit contributions
isel-0.027
oths-0.025
ap-0.024
icker-0.024
apa-0.024
Token it
Token ID340
Feature activation-1.56e+00

Pos logit contributions
rob+0.010
 eighty+0.006
 l+0.006
 ninety+0.006
 Ske+0.006
Neg logit contributions
isel-0.012
oths-0.012
apa-0.012
icker-0.011
ap-0.011
Token is
Token ID318
Feature activation-5.62e-01

Pos logit contributions
rob+0.070
 l+0.058
 eighty+0.052
 Ske+0.051
 cand+0.050
Neg logit contributions
isel-0.074
icker-0.072
oths-0.072
apa-0.068
umped-0.068
Token my
Token ID616
Feature activation-1.34e+00

Pos logit contributions
rob+0.025
 l+0.018
 eighty+0.017
 cand+0.017
 Ske+0.016
Neg logit contributions
isel-0.027
oths-0.026
icker-0.026
inging-0.025
ap-0.025
Token tool
Token ID2891
Feature activation-2.49e+00

Pos logit contributions
rob+0.065
 l+0.053
 eighty+0.049
 Ske+0.048
 cand+0.047
Neg logit contributions
isel-0.059
icker-0.058
oths-0.057
apa-0.055
umped-0.054
Token now
Token ID783
Feature activation-1.28e+00

Pos logit contributions
rob+0.133
 l+0.106
 eighty+0.102
 Ske+0.098
 cand+0.098
Neg logit contributions
isel-0.092
icker-0.091
ap-0.088
oths-0.088
apa-0.088
Token!
Token ID0
Feature activation+1.10e-01

Pos logit contributions
rob+0.061
 l+0.049
 eighty+0.046
 Ske+0.045
 cand+0.045
Neg logit contributions
isel-0.057
icker-0.055
oths-0.054
apa-0.052
umped-0.052
Token I
Token ID314
Feature activation-3.59e-01

Pos logit contributions
isel+0.005
icker+0.005
oths+0.005
imming+0.005
umped+0.005
Neg logit contributions
rob-0.004
 l-0.004
 eighty-0.003
 Ske-0.003
 Velvet-0.003
Token can
Token ID460
Feature activation+3.84e-01

Pos logit contributions
rob+0.018
 l+0.015
 eighty+0.014
 Ske+0.013
 cand+0.013
Neg logit contributions
isel-0.015
icker-0.015
oths-0.015
apa-0.014
umped-0.014
Token blow
Token ID6611
Feature activation+2.74e-01

Pos logit contributions
isel+0.016
icker+0.016
oths+0.015
umped+0.015
ap+0.015
Neg logit contributions
rob-0.020
 l-0.016
 eighty-0.015
 cand-0.015
 Ske-0.014
Token cleaning
Token ID12724
Feature activation-7.65e-01

Pos logit contributions
rob+0.007
 l+0.005
 Ske+0.004
 cand+0.004
 eighty+0.004
Neg logit contributions
isel-0.016
icker-0.015
oths-0.015
imming-0.015
umped-0.015
Token wasn
Token ID2492
Feature activation-1.57e-01

Pos logit contributions
rob+0.030
 l+0.025
 cand+0.022
 Ske+0.021
 Velvet+0.021
Neg logit contributions
isel-0.041
icker-0.038
oths-0.038
imming-0.038
leys-0.037
Token't
Token ID470
Feature activation-5.36e-01

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.005
 Ske+0.005
 Velvet+0.004
Neg logit contributions
isel-0.008
oths-0.008
icker-0.008
ooks-0.008
imming-0.008
Token her
Token ID607
Feature activation-7.63e-01

Pos logit contributions
rob+0.020
 l+0.018
 eighty+0.016
 cand+0.015
 Ske+0.015
Neg logit contributions
isel-0.028
oths-0.027
icker-0.027
apa-0.026
imming-0.026
Token favorite
Token ID4004
Feature activation-4.34e-01

Pos logit contributions
 l+0.026
 cand+0.022
 candy+0.022
rob+0.021
 scar+0.020
Neg logit contributions
isel-0.042
oths-0.039
imming-0.039
leys-0.039
icker-0.039
Token thing
Token ID1517
Feature activation-2.46e+00

Pos logit contributions
rob+0.019
 l+0.016
 cand+0.014
 eighty+0.014
 Ske+0.014
Neg logit contributions
isel-0.021
icker-0.021
oths-0.020
imming-0.020
umped-0.020
Token to
Token ID284
Feature activation-8.49e-01

Pos logit contributions
rob+0.100
 l+0.081
 eighty+0.073
 Ske+0.071
 cand+0.070
Neg logit contributions
isel-0.124
icker-0.121
oths-0.120
umped-0.115
apa-0.115
Token do
Token ID466
Feature activation-2.81e-01

Pos logit contributions
rob+0.032
 l+0.028
 Ske+0.024
 eighty+0.024
 cand+0.024
Neg logit contributions
isel-0.044
oths-0.043
icker-0.042
umped-0.041
leys-0.041
Token,
Token ID11
Feature activation-4.10e-01

Pos logit contributions
rob+0.011
 l+0.009
 cand+0.008
 Ske+0.008
 candy+0.008
Neg logit contributions
isel-0.015
icker-0.014
oths-0.014
apa-0.014
imming-0.013
Token she
Token ID673
Feature activation-2.67e-01

Pos logit contributions
rob+0.012
 l+0.009
 eighty+0.008
 Ske+0.007
 cand+0.007
Neg logit contributions
isel-0.025
icker-0.025
oths-0.025
apa-0.024
umped-0.024
Token knew
Token ID2993
Feature activation-3.76e-01

Pos logit contributions
 l+0.010
rob+0.010
 eighty+0.008
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.014
icker-0.013
oths-0.013
imming-0.013
eways-0.013
Token.
Token ID13
Feature activation-1.80e-01

Pos logit contributions
isel+0.016
icker+0.015
oths+0.015
apa+0.015
umped+0.015
Neg logit contributions
rob-0.018
 l-0.015
 eighty-0.014
 Ske-0.013
 cand-0.013
Token Tim
Token ID5045
Feature activation-1.64e-01

Pos logit contributions
rob+0.007
 l+0.006
 eighty+0.005
 cand+0.005
 Ske+0.005
Neg logit contributions
isel-0.009
icker-0.009
oths-0.009
umped-0.009
apa-0.008
Tokenmy
Token ID1820
Feature activation+3.08e-01

Pos logit contributions
rob+0.007
 l+0.005
 eighty+0.005
 cand+0.005
 Ske+0.005
Neg logit contributions
isel-0.008
icker-0.008
oths-0.008
umped-0.007
imming-0.007
Token had
Token ID550
Feature activation-4.01e-01

Pos logit contributions
isel+0.011
icker+0.011
oths+0.011
umped+0.011
apa+0.010
Neg logit contributions
rob-0.017
 l-0.014
 eighty-0.013
 Ske-0.013
 cand-0.013
Token a
Token ID257
Feature activation-8.67e-01

Pos logit contributions
rob+0.021
 eighty+0.016
 l+0.015
 cand+0.015
 seventy+0.015
Neg logit contributions
icker-0.016
oths-0.015
umped-0.014
isel-0.014
ap-0.014
Token w
Token ID266
Feature activation-2.41e+00

Pos logit contributions
rob+0.042
 eighty+0.031
 l+0.030
 Ske+0.029
 cand+0.029
Neg logit contributions
icker-0.039
isel-0.038
oths-0.036
ap-0.035
apa-0.035
Tokenigg
Token ID6950
Feature activation-1.45e+00

Pos logit contributions
rob+0.113
 l+0.092
 eighty+0.086
 Ske+0.083
 cand+0.082
Neg logit contributions
isel-0.110
icker-0.108
oths-0.105
apa-0.100
imming-0.100
Tokenly
Token ID306
Feature activation-1.07e+00

Pos logit contributions
rob+0.054
 l+0.043
 eighty+0.038
 Ske+0.037
 cand+0.037
Neg logit contributions
isel-0.078
oths-0.077
icker-0.076
umped-0.074
apa-0.074
Token tooth
Token ID16162
Feature activation-7.13e-01

Pos logit contributions
rob+0.045
 eighty+0.033
 l+0.032
 Ske+0.031
 cand+0.030
Neg logit contributions
isel-0.052
icker-0.051
oths-0.050
ap-0.048
apa-0.048
Token that
Token ID326
Feature activation+2.08e-01

Pos logit contributions
rob+0.035
 eighty+0.026
 l+0.025
 Ske+0.025
 ninety+0.024
Neg logit contributions
isel-0.030
icker-0.029
oths-0.028
umped-0.027
apa-0.027
Token was
Token ID373
Feature activation-2.81e-01

Pos logit contributions
icker+0.010
apa+0.010
isel+0.010
oths+0.009
ooks+0.009
Neg logit contributions
rob-0.009
 eighty-0.006
 seventy-0.006
 ninety-0.006
 scar-0.005
Token that
Token ID326
Feature activation+3.56e-01

Pos logit contributions
rob+0.014
 eighty+0.011
 l+0.010
 seventy+0.009
 ninety+0.009
Neg logit contributions
icker-0.016
isel-0.016
oths-0.015
aining-0.015
apa-0.015
Token sharing
Token ID7373
Feature activation-7.32e-01

Pos logit contributions
oths+0.016
isel+0.016
icker+0.015
ap+0.015
ooks+0.014
Neg logit contributions
rob-0.017
 l-0.013
 eighty-0.012
 Ske-0.011
 cand-0.011
Token is
Token ID318
Feature activation-1.12e+00

Pos logit contributions
rob+0.032
 l+0.026
 eighty+0.024
 Ske+0.023
 cand+0.023
Neg logit contributions
isel-0.035
icker-0.034
oths-0.034
apa-0.032
umped-0.032
Token a
Token ID257
Feature activation-9.49e-01

Pos logit contributions
rob+0.043
 l+0.039
 cand+0.033
 Ske+0.033
 eighty+0.033
Neg logit contributions
isel-0.057
icker-0.057
oths-0.056
umped-0.054
apa-0.054
Token good
Token ID922
Feature activation-2.98e-01

Pos logit contributions
rob+0.037
 l+0.035
 cand+0.029
 Ske+0.029
 candy+0.029
Neg logit contributions
isel-0.048
oths-0.048
icker-0.046
umped-0.046
imming-0.046
Token thing
Token ID1517
Feature activation-2.34e+00

Pos logit contributions
rob+0.017
 eighty+0.014
 l+0.013
 seventy+0.012
 Velvet+0.012
Neg logit contributions
isel-0.011
ap-0.010
icker-0.010
oths-0.010
ench-0.010
Token and
Token ID290
Feature activation-4.51e-03

Pos logit contributions
rob+0.103
 l+0.083
 eighty+0.077
 Ske+0.074
 cand+0.074
Neg logit contributions
isel-0.111
icker-0.108
oths-0.107
apa-0.103
umped-0.103
Token it
Token ID340
Feature activation-1.38e+00

Pos logit contributions
rob+0.000
 l+0.000
 cand+0.000
 eighty+0.000
 Ske+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
apa-0.000
imming-0.000
Token made
Token ID925
Feature activation-5.82e-01

Pos logit contributions
rob+0.058
 l+0.051
 cand+0.044
 eighty+0.043
 Ske+0.043
Neg logit contributions
isel-0.068
icker-0.066
oths-0.065
apa-0.064
ooks-0.062
Token her
Token ID607
Feature activation-6.55e-01

Pos logit contributions
rob+0.019
 l+0.015
 eighty+0.013
 cand+0.013
 Ske+0.013
Neg logit contributions
isel-0.034
icker-0.033
oths-0.032
umped-0.032
apa-0.032
Token feel
Token ID1254
Feature activation-4.99e-01

Pos logit contributions
rob+0.033
 l+0.027
 eighty+0.025
 Ske+0.025
 cand+0.024
Neg logit contributions
isel-0.027
icker-0.027
oths-0.026
apa-0.025
umped-0.025
Token.
Token ID13
Feature activation+2.18e-01

Pos logit contributions
rob+0.025
 l+0.020
 eighty+0.018
 Ske+0.017
 cand+0.017
Neg logit contributions
isel-0.034
icker-0.033
oths-0.033
apa-0.032
umped-0.031
Token It
Token ID632
Feature activation-5.98e-01

Pos logit contributions
isel+0.012
icker+0.012
imming+0.011
leys+0.011
umped+0.011
Neg logit contributions
rob-0.008
 l-0.007
 Ske-0.006
 Velvet-0.006
 eighty-0.006
Token's
Token ID338
Feature activation-1.02e+00

Pos logit contributions
rob+0.030
 l+0.025
 eighty+0.023
 Ske+0.022
 cand+0.022
Neg logit contributions
isel-0.025
icker-0.025
oths-0.024
apa-0.023
umped-0.023
Token my
Token ID616
Feature activation-1.49e+00

Pos logit contributions
rob+0.052
 l+0.041
 eighty+0.039
 Ske+0.038
 cand+0.038
Neg logit contributions
isel-0.043
icker-0.041
oths-0.041
imming-0.039
aining-0.039
Token favorite
Token ID4004
Feature activation-8.58e-01

Pos logit contributions
rob+0.068
 l+0.053
 eighty+0.050
 Ske+0.048
 cand+0.047
Neg logit contributions
isel-0.070
icker-0.068
oths-0.068
umped-0.065
apa-0.065
Token thing
Token ID1517
Feature activation-2.33e+00

Pos logit contributions
rob+0.040
 l+0.031
 eighty+0.031
 Ske+0.029
 cand+0.028
Neg logit contributions
isel-0.038
oths-0.037
icker-0.037
aining-0.035
ench-0.035
Token to
Token ID284
Feature activation-1.12e+00

Pos logit contributions
rob+0.118
 l+0.091
 eighty+0.089
 Ske+0.084
 cand+0.083
Neg logit contributions
icker-0.092
isel-0.092
oths-0.091
apa-0.088
ap-0.088
Token play
Token ID711
Feature activation-1.05e-01

Pos logit contributions
rob+0.054
 l+0.041
 eighty+0.041
 cand+0.039
 Ske+0.038
Neg logit contributions
isel-0.050
icker-0.049
oths-0.048
ap-0.046
apa-0.046
Token with
Token ID351
Feature activation-3.48e-01

Pos logit contributions
rob+0.005
 l+0.004
 eighty+0.003
 Ske+0.003
 cand+0.003
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
apa-0.005
umped-0.005
Token."
Token ID526
Feature activation+4.05e-01

Pos logit contributions
rob+0.014
 l+0.012
 cand+0.011
 eighty+0.010
 Ske+0.010
Neg logit contributions
isel-0.018
icker-0.017
apa-0.017
oths-0.017
umped-0.016
Token
Token ID198
Feature activation-7.98e-02

Pos logit contributions
isel+0.019
icker+0.019
oths+0.019
imming+0.018
apa+0.018
Neg logit contributions
rob-0.018
 l-0.015
 eighty-0.013
 Ske-0.013
 cand-0.013
Token
Token ID220
Feature activation-3.95e-01

Pos logit contributions
rob+0.013
 l+0.011
 Ske+0.010
 Velvet+0.010
 cand+0.009
Neg logit contributions
isel-0.017
icker-0.016
imming-0.016
oths-0.016
umped-0.015
Token
Token ID198
Feature activation-4.75e-01

Pos logit contributions
rob+0.011
 Velvet+0.008
 Ske+0.008
 l+0.008
 Robo+0.007
Neg logit contributions
isel-0.024
icker-0.022
imming-0.022
oths-0.021
appiness-0.021
Token
Token ID198
Feature activation-2.61e-01

Pos logit contributions
rob+0.019
 l+0.015
 Ske+0.014
 eighty+0.014
 Velvet+0.014
Neg logit contributions
isel-0.025
icker-0.023
oths-0.023
imming-0.022
apa-0.022
TokenAfter
Token ID3260
Feature activation-9.38e-01

Pos logit contributions
rob+0.011
 l+0.009
 Velvet+0.009
 Ske+0.008
 eighty+0.008
Neg logit contributions
isel-0.013
icker-0.012
apa-0.012
oths-0.012
ench-0.011
Token their
Token ID511
Feature activation+3.72e-01

Pos logit contributions
rob+0.039
 l+0.031
 eighty+0.028
 Ske+0.027
 cand+0.027
Neg logit contributions
isel-0.047
icker-0.047
oths-0.046
apa-0.044
umped-0.044
Token fun
Token ID1257
Feature activation-2.32e+00

Pos logit contributions
isel+0.019
icker+0.018
oths+0.018
umped+0.017
aining+0.017
Neg logit contributions
rob-0.015
 l-0.013
 eighty-0.011
 cand-0.011
 Ske-0.011
Token meeting
Token ID3249
Feature activation-9.43e-01

Pos logit contributions
rob+0.099
 l+0.077
 eighty+0.073
 Ske+0.069
 cand+0.067
Neg logit contributions
isel-0.115
icker-0.113
oths-0.112
apa-0.107
umped-0.107
Token,
Token ID11
Feature activation-1.12e+00

Pos logit contributions
rob+0.041
 l+0.034
 eighty+0.031
 Ske+0.031
 cand+0.031
Neg logit contributions
isel-0.045
icker-0.044
oths-0.043
umped-0.041
apa-0.041
Token Lily
Token ID20037
Feature activation-3.50e-01

Pos logit contributions
rob+0.034
 l+0.027
 Ske+0.023
 cand+0.022
 eighty+0.022
Neg logit contributions
isel-0.069
icker-0.066
oths-0.065
imming-0.064
umped-0.064
Token went
Token ID1816
Feature activation-5.96e-01

Pos logit contributions
rob+0.016
 l+0.014
 eighty+0.012
 Ske+0.012
 cand+0.012
Neg logit contributions
isel-0.017
icker-0.016
oths-0.016
imming-0.015
aining-0.015
Token back
Token ID736
Feature activation+3.91e-01

Pos logit contributions
rob+0.031
 l+0.025
 eighty+0.024
 Ske+0.024
 cand+0.024
Neg logit contributions
isel-0.023
icker-0.023
oths-0.023
umped-0.021
imming-0.021
Token.
Token ID13
Feature activation-2.30e-01

Pos logit contributions
rob+0.007
 l+0.006
 eighty+0.005
 cand+0.005
 Ske+0.005
Neg logit contributions
isel-0.011
icker-0.011
oths-0.011
apa-0.010
umped-0.010
Token You
Token ID921
Feature activation-4.01e-01

Pos logit contributions
rob+0.009
 l+0.008
 eighty+0.007
 Ske+0.007
 Velvet+0.007
Neg logit contributions
isel-0.012
icker-0.012
umped-0.011
oths-0.011
imming-0.011
Token are
Token ID389
Feature activation-1.37e-01

Pos logit contributions
rob+0.017
 l+0.015
 eighty+0.014
 Ske+0.013
 cand+0.013
Neg logit contributions
oths-0.020
icker-0.019
isel-0.019
ooks-0.019
aining-0.019
Token the
Token ID262
Feature activation-8.93e-01

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.004
 Ske+0.004
 cand+0.004
Neg logit contributions
isel-0.006
icker-0.006
oths-0.006
apa-0.006
umped-0.006
Token best
Token ID1266
Feature activation-8.44e-01

Pos logit contributions
rob+0.044
 l+0.036
 eighty+0.033
 Ske+0.033
 cand+0.032
Neg logit contributions
isel-0.038
icker-0.037
oths-0.036
apa-0.035
umped-0.035
Token thing
Token ID1517
Feature activation-2.32e+00

Pos logit contributions
rob+0.038
 l+0.032
 eighty+0.029
 Ske+0.028
 cand+0.028
Neg logit contributions
isel-0.040
icker-0.039
oths-0.039
umped-0.038
imming-0.037
Token in
Token ID287
Feature activation-7.38e-01

Pos logit contributions
rob+0.098
 l+0.078
 eighty+0.072
 Ske+0.069
 cand+0.069
Neg logit contributions
isel-0.114
icker-0.111
oths-0.110
apa-0.106
umped-0.106
Token my
Token ID616
Feature activation-1.22e+00

Pos logit contributions
rob+0.029
 l+0.023
 eighty+0.021
 Ske+0.021
 cand+0.020
Neg logit contributions
isel-0.039
icker-0.038
oths-0.037
umped-0.036
apa-0.036
Token life
Token ID1204
Feature activation-7.61e-01

Pos logit contributions
rob+0.058
 l+0.055
 cand+0.047
 Ske+0.047
 eighty+0.046
Neg logit contributions
isel-0.054
icker-0.049
umped-0.048
oths-0.048
aining-0.048
Token."
Token ID526
Feature activation-3.38e-01

Pos logit contributions
rob+0.030
 l+0.026
 cand+0.022
 candy+0.021
 eighty+0.021
Neg logit contributions
isel-0.042
icker-0.039
oths-0.038
apa-0.038
ooks-0.037
Token He
Token ID679
Feature activation-3.25e-02

Pos logit contributions
rob+0.018
 l+0.015
 eighty+0.014
 cand+0.013
 ninety+0.013
Neg logit contributions
oths-0.013
isel-0.013
icker-0.013
apa-0.013
ooks-0.012
Token would
Token ID561
Feature activation+5.28e-01

Pos logit contributions
rob+0.007
 l+0.006
 eighty+0.005
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.007
icker-0.007
oths-0.007
apa-0.007
umped-0.007
Token never
Token ID1239
Feature activation+3.91e-01

Pos logit contributions
oths+0.028
isel+0.028
icker+0.027
leys+0.026
eways+0.026
Neg logit contributions
 l-0.018
rob-0.018
 Ske-0.015
 eighty-0.015
 cand-0.014
Token get
Token ID651
Feature activation+2.50e-01

Pos logit contributions
isel+0.018
icker+0.018
oths+0.018
apa+0.017
umped+0.017
Neg logit contributions
rob-0.017
 l-0.015
 eighty-0.013
 Ske-0.013
 cand-0.013
Token his
Token ID465
Feature activation+1.43e-02

Pos logit contributions
isel+0.013
icker+0.012
oths+0.012
umped+0.012
apa+0.012
Neg logit contributions
rob-0.010
 l-0.009
 eighty-0.008
 cand-0.008
 Ske-0.008
Token spicy
Token ID26880
Feature activation-7.69e-01

Pos logit contributions
isel+0.001
umped+0.001
oths+0.001
aining+0.001
icker+0.001
Neg logit contributions
 l-0.001
rob-0.001
 Ske-0.000
 cand-0.000
 candy-0.000
Token thing
Token ID1517
Feature activation-2.31e+00

Pos logit contributions
rob+0.031
 l+0.027
 cand+0.023
 Ske+0.023
 eighty+0.022
Neg logit contributions
isel-0.039
icker-0.038
oths-0.038
umped-0.037
apa-0.037
Token,
Token ID11
Feature activation+7.03e-01

Pos logit contributions
rob+0.098
 l+0.082
 eighty+0.074
 Ske+0.072
 cand+0.071
Neg logit contributions
isel-0.114
icker-0.110
oths-0.110
umped-0.105
apa-0.104
Token and
Token ID290
Feature activation+3.71e-01

Pos logit contributions
isel+0.038
icker+0.038
oths+0.037
imming+0.036
umped+0.036
Neg logit contributions
rob-0.025
 l-0.021
 eighty-0.019
 Ske-0.018
 cand-0.018
Token his
Token ID465
Feature activation+1.54e-01

Pos logit contributions
icker+0.025
isel+0.025
oths+0.025
imming+0.024
umped+0.024
Neg logit contributions
rob-0.008
 l-0.007
 Ske-0.005
 eighty-0.005
 cand-0.005
Token trust
Token ID3774
Feature activation-5.18e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
umped+0.007
aining+0.007
Neg logit contributions
rob-0.007
 l-0.006
 Ske-0.005
 cand-0.005
 eighty-0.005
Token had
Token ID550
Feature activation-4.94e-02

Pos logit contributions
rob+0.022
 l+0.019
 Ske+0.016
 eighty+0.016
 cand+0.016
Neg logit contributions
isel-0.025
icker-0.024
oths-0.024
imming-0.023
umped-0.023
Token lying
Token ID9105
Feature activation+5.14e-03

Pos logit contributions
isel+0.030
icker+0.029
oths+0.029
umped+0.028
apa+0.028
Neg logit contributions
rob-0.018
 l-0.014
 eighty-0.012
 Ske-0.012
 cand-0.011
Token is
Token ID318
Feature activation-8.64e-01

Pos logit contributions
isel+0.000
icker+0.000
oths+0.000
ooks+0.000
umped+0.000
Neg logit contributions
rob-0.000
 l-0.000
 candy-0.000
 cand-0.000
 Ske-0.000
Token never
Token ID1239
Feature activation-3.51e-03

Pos logit contributions
 l+0.028
rob+0.026
 candy+0.023
 eighty+0.022
 cand+0.021
Neg logit contributions
oths-0.049
icker-0.048
isel-0.048
umped-0.046
apa-0.045
Token the
Token ID262
Feature activation-1.21e+00

Pos logit contributions
rob+0.000
 l+0.000
 eighty+0.000
 Ske+0.000
 cand+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
apa-0.000
umped-0.000
Token right
Token ID826
Feature activation-6.39e-01

Pos logit contributions
rob+0.047
 l+0.041
 eighty+0.036
 Ske+0.035
 cand+0.035
Neg logit contributions
isel-0.063
icker-0.062
oths-0.062
umped-0.060
apa-0.059
Token thing
Token ID1517
Feature activation-2.31e+00

Pos logit contributions
rob+0.028
 l+0.025
 eighty+0.022
 Ske+0.022
 cand+0.022
Neg logit contributions
icker-0.030
isel-0.029
oths-0.028
umped-0.028
apa-0.028
Token to
Token ID284
Feature activation-6.28e-01

Pos logit contributions
rob+0.091
 l+0.076
 eighty+0.067
 Ske+0.067
 cand+0.064
Neg logit contributions
isel-0.120
icker-0.117
oths-0.116
umped-0.112
imming-0.111
Token do
Token ID466
Feature activation-2.94e-01

Pos logit contributions
rob+0.029
 l+0.023
 eighty+0.022
 cand+0.021
 Ske+0.021
Neg logit contributions
isel-0.029
icker-0.028
oths-0.028
umped-0.027
ap-0.027
Token.
Token ID13
Feature activation-4.16e-01

Pos logit contributions
rob+0.012
 l+0.010
 Ske+0.009
 eighty+0.009
 cand+0.009
Neg logit contributions
isel-0.015
icker-0.014
oths-0.014
apa-0.014
umped-0.013
Token From
Token ID3574
Feature activation-2.29e-01

Pos logit contributions
rob+0.021
 l+0.017
 eighty+0.016
 cand+0.015
 Ske+0.015
Neg logit contributions
isel-0.017
oths-0.017
icker-0.017
apa-0.016
umped-0.016
Token that
Token ID326
Feature activation+3.40e-01

Pos logit contributions
rob+0.010
 l+0.008
 eighty+0.007
 cand+0.007
 Ske+0.007
Neg logit contributions
isel-0.011
icker-0.011
oths-0.010
apa-0.010
imming-0.010
Token "
Token ID366
Feature activation+5.00e-01

Pos logit contributions
isel+0.011
icker+0.011
oths+0.011
imming+0.010
umped+0.010
Neg logit contributions
rob-0.008
 l-0.007
 eighty-0.007
 Ske-0.006
 cand-0.006
TokenIt
Token ID1026
Feature activation-4.67e-01

Pos logit contributions
isel+0.022
icker+0.022
oths+0.021
apa+0.020
imming+0.020
Neg logit contributions
rob-0.024
 l-0.020
 eighty-0.018
 Ske-0.018
 cand-0.017
Token's
Token ID338
Feature activation-9.26e-01

Pos logit contributions
 l+0.017
rob+0.017
 Ske+0.013
 cand+0.013
 eighty+0.013
Neg logit contributions
icker-0.025
isel-0.025
oths-0.025
ooks-0.023
ifying-0.023
Token a
Token ID257
Feature activation-1.18e+00

Pos logit contributions
rob+0.047
 l+0.044
 eighty+0.041
 Ske+0.040
 cand+0.039
Neg logit contributions
isel-0.035
icker-0.035
oths-0.034
umped-0.033
imming-0.031
Token good
Token ID922
Feature activation-4.87e-01

Pos logit contributions
rob+0.057
 l+0.046
 eighty+0.043
 Ske+0.042
 cand+0.041
Neg logit contributions
isel-0.052
icker-0.051
oths-0.050
apa-0.048
umped-0.048
Token thing
Token ID1517
Feature activation-2.30e+00

Pos logit contributions
rob+0.027
 eighty+0.020
 l+0.019
 Velvet+0.019
 Ske+0.019
Neg logit contributions
isel-0.019
ench-0.018
ap-0.018
icker-0.018
oths-0.017
Token you
Token ID345
Feature activation-1.69e-01

Pos logit contributions
rob+0.097
 l+0.078
 eighty+0.071
 Ske+0.069
 cand+0.068
Neg logit contributions
isel-0.113
icker-0.110
oths-0.109
umped-0.105
apa-0.105
Token know
Token ID760
Feature activation+2.46e-01

Pos logit contributions
 l+0.006
rob+0.005
 Ske+0.005
 candy+0.005
 w+0.004
Neg logit contributions
icker-0.009
oths-0.009
isel-0.009
umped-0.009
aining-0.009
Token what
Token ID644
Feature activation+6.18e-02

Pos logit contributions
isel+0.014
icker+0.014
umped+0.013
oths+0.013
imming+0.013
Neg logit contributions
rob-0.008
 l-0.008
 cand-0.007
 candy-0.007
 eighty-0.006
Token you
Token ID345
Feature activation+4.72e-01

Pos logit contributions
isel+0.004
icker+0.003
oths+0.003
imming+0.003
umped+0.003
Neg logit contributions
rob-0.002
 l-0.002
 cand-0.002
 Ske-0.002
 eighty-0.002
Token want
Token ID765
Feature activation+2.44e-01

Pos logit contributions
icker+0.024
oths+0.024
isel+0.024
ooks+0.023
aining+0.023
Neg logit contributions
rob-0.017
 l-0.017
 Ske-0.014
 candy-0.013
 eighty-0.013
Token mean
Token ID1612
Feature activation-1.96e-01

Pos logit contributions
isel+0.000
icker+0.000
oths+0.000
umped+0.000
imming+0.000
Neg logit contributions
rob-0.001
 l-0.000
 eighty-0.000
 Ske-0.000
 cand-0.000
Token!
Token ID0
Feature activation+7.29e-02

Pos logit contributions
rob+0.009
 l+0.008
 eighty+0.007
 Ske+0.007
 cand+0.007
Neg logit contributions
isel-0.009
icker-0.009
oths-0.008
umped-0.008
apa-0.008
Token You
Token ID921
Feature activation-5.57e-01

Pos logit contributions
isel+0.004
icker+0.004
leys+0.004
ench+0.004
imming+0.004
Neg logit contributions
rob-0.003
 l-0.002
 Ske-0.002
 eighty-0.002
 Velvet-0.002
Token ruined
Token ID20484
Feature activation+7.86e-02

Pos logit contributions
rob+0.026
 l+0.022
 eighty+0.019
 Ske+0.019
 cand+0.019
Neg logit contributions
isel-0.025
icker-0.025
oths-0.025
apa-0.024
imming-0.024
Token our
Token ID674
Feature activation+4.49e-02

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
umped+0.004
apa+0.004
Neg logit contributions
rob-0.003
 l-0.002
 eighty-0.002
 Ske-0.002
 cand-0.002
Token fun
Token ID1257
Feature activation-2.30e+00

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
apa+0.002
inging+0.002
Neg logit contributions
rob-0.002
 eighty-0.002
 l-0.002
 Ske-0.002
 seventy-0.001
Token!"
Token ID2474
Feature activation+3.50e-01

Pos logit contributions
rob+0.110
 l+0.090
 eighty+0.085
 Ske+0.083
 cand+0.080
Neg logit contributions
isel-0.099
icker-0.096
oths-0.096
ap-0.091
apa-0.091
Token Ben
Token ID3932
Feature activation-3.53e-01

Pos logit contributions
isel+0.019
icker+0.018
oths+0.018
apa+0.018
umped+0.017
Neg logit contributions
rob-0.014
 l-0.010
 eighty-0.010
 cand-0.009
 Ske-0.009
Token cries
Token ID24691
Feature activation-4.57e-01

Pos logit contributions
rob+0.014
 eighty+0.010
 l+0.010
 cand+0.009
 Ske+0.009
Neg logit contributions
isel-0.018
oths-0.018
icker-0.018
ap-0.017
umped-0.017
Token.
Token ID13
Feature activation+1.65e-01

Pos logit contributions
rob+0.023
 l+0.019
 eighty+0.018
 Ske+0.017
 cand+0.017
Neg logit contributions
isel-0.019
icker-0.019
oths-0.018
apa-0.018
umped-0.018
Token
Token ID198
Feature activation+1.44e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
apa+0.006
umped+0.006
Neg logit contributions
rob-0.008
 l-0.007
 eighty-0.006
 cand-0.006
 Ske-0.006
Token'm
Token ID1101
Feature activation-6.36e-02

Pos logit contributions
isel+0.014
icker+0.013
oths+0.013
umped+0.013
apa+0.013
Neg logit contributions
rob-0.017
 l-0.015
 eighty-0.014
 Ske-0.013
 cand-0.013
Token sorry
Token ID7926
Feature activation+5.04e-01

Pos logit contributions
rob+0.003
 l+0.003
 eighty+0.002
 cand+0.002
 Ske+0.002
Neg logit contributions
isel-0.003
icker-0.002
oths-0.002
apa-0.002
umped-0.002
Token for
Token ID329
Feature activation-3.15e-01

Pos logit contributions
isel+0.026
icker+0.026
oths+0.025
umped+0.025
imming+0.024
Neg logit contributions
rob-0.020
 l-0.016
 Ske-0.015
 eighty-0.015
 cand-0.014
Token breaking
Token ID7163
Feature activation-1.85e-01

Pos logit contributions
rob+0.012
 l+0.010
 eighty+0.009
 Ske+0.009
 cand+0.009
Neg logit contributions
isel-0.016
icker-0.016
oths-0.016
apa-0.015
umped-0.015
Token your
Token ID534
Feature activation-2.68e-01

Pos logit contributions
rob+0.005
 l+0.005
 candy+0.004
 cand+0.004
 eighty+0.004
Neg logit contributions
icker-0.011
isel-0.011
umped-0.011
ench-0.011
oths-0.011
Token tool
Token ID2891
Feature activation-2.29e+00

Pos logit contributions
rob+0.010
 l+0.009
 cand+0.008
 candy+0.008
 scar+0.007
Neg logit contributions
isel-0.014
icker-0.013
ooks-0.013
umped-0.013
ench-0.013
Token.
Token ID13
Feature activation+5.07e-01

Pos logit contributions
rob+0.094
 l+0.080
 cand+0.072
 eighty+0.071
 Ske+0.067
Neg logit contributions
isel-0.116
icker-0.113
oths-0.110
imming-0.106
apa-0.106
Token I
Token ID314
Feature activation+7.84e-02

Pos logit contributions
isel+0.029
icker+0.029
imming+0.028
oths+0.028
leys+0.028
Neg logit contributions
rob-0.016
 l-0.013
 Ske-0.013
 eighty-0.013
 Velvet-0.012
Token love
Token ID1842
Feature activation+4.82e-01

Pos logit contributions
oths+0.003
isel+0.003
imming+0.003
icker+0.003
eways+0.003
Neg logit contributions
rob-0.003
 l-0.003
 Ske-0.003
 eighty-0.003
 cand-0.003
Token you
Token ID345
Feature activation-2.03e-01

Pos logit contributions
oths+0.022
isel+0.022
icker+0.021
apa+0.021
ap+0.020
Neg logit contributions
rob-0.022
 l-0.016
 eighty-0.015
 Ske-0.015
 Velvet-0.014
Token,
Token ID11
Feature activation+7.60e-01

Pos logit contributions
rob+0.007
 l+0.007
 eighty+0.006
 cand+0.006
 candy+0.005
Neg logit contributions
isel-0.011
oths-0.011
icker-0.011
apa-0.010
umped-0.010
Token "
Token ID366
Feature activation+1.14e+00

Pos logit contributions
isel+0.013
icker+0.013
oths+0.013
apa+0.012
umped+0.012
Neg logit contributions
rob-0.015
 l-0.012
 eighty-0.011
 cand-0.011
 Ske-0.010
TokenThat
Token ID2504
Feature activation-2.58e-01

Pos logit contributions
isel+0.046
icker+0.044
oths+0.043
apa+0.042
umped+0.041
Neg logit contributions
rob-0.059
 l-0.050
 eighty-0.047
 Ske-0.045
 cand-0.045
Token's
Token ID338
Feature activation-5.76e-01

Pos logit contributions
rob+0.011
 l+0.009
 cand+0.008
 eighty+0.007
 candy+0.007
Neg logit contributions
icker-0.014
isel-0.014
oths-0.013
ooks-0.013
umped-0.013
Token the
Token ID262
Feature activation-8.69e-01

Pos logit contributions
rob+0.030
 l+0.022
 eighty+0.021
 cand+0.021
 Ske+0.021
Neg logit contributions
isel-0.024
icker-0.023
oths-0.023
imming-0.022
ap-0.022
Token best
Token ID1266
Feature activation-3.16e-01

Pos logit contributions
rob+0.036
 l+0.033
 eighty+0.030
 Ske+0.028
 cand+0.028
Neg logit contributions
oths-0.042
icker-0.041
isel-0.041
umped-0.041
apa-0.040
Token thing
Token ID1517
Feature activation-2.27e+00

Pos logit contributions
rob+0.014
 l+0.012
 Ske+0.011
 cand+0.011
 eighty+0.011
Neg logit contributions
icker-0.014
isel-0.014
umped-0.014
oths-0.014
apa-0.014
Token to
Token ID284
Feature activation-6.37e-01

Pos logit contributions
rob+0.109
 l+0.084
 eighty+0.081
 Ske+0.076
 cand+0.076
Neg logit contributions
oths-0.094
isel-0.094
icker-0.094
apa-0.092
ap-0.090
Token do
Token ID466
Feature activation+4.15e-02

Pos logit contributions
rob+0.029
 l+0.023
 eighty+0.022
 Ske+0.021
 cand+0.021
Neg logit contributions
isel-0.030
icker-0.029
oths-0.029
apa-0.028
umped-0.028
Token.
Token ID13
Feature activation+3.39e-01

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
apa+0.002
umped+0.002
Neg logit contributions
rob-0.002
 l-0.002
 eighty-0.001
 cand-0.001
 Ske-0.001
Token Then
Token ID3244
Feature activation-7.42e-01

Pos logit contributions
icker+0.016
isel+0.016
oths+0.015
umped+0.015
leys+0.015
Neg logit contributions
rob-0.014
 l-0.012
 Ske-0.012
 eighty-0.011
 Velvet-0.011
Token it
Token ID340
Feature activation-7.67e-01

Pos logit contributions
rob+0.026
 l+0.021
 eighty+0.018
 Ske+0.018
 cand+0.018
Neg logit contributions
isel-0.042
icker-0.041
oths-0.040
umped-0.040
imming-0.039
Token "
Token ID366
Feature activation+4.70e-02

Pos logit contributions
rob+0.019
 l+0.016
 eighty+0.014
 Ske+0.014
 cand+0.014
Neg logit contributions
isel-0.016
icker-0.016
oths-0.016
apa-0.015
umped-0.015
TokenIt
Token ID1026
Feature activation-5.39e-01

Pos logit contributions
isel+0.002
icker+0.002
oths+0.002
umped+0.002
aining+0.002
Neg logit contributions
rob-0.003
 l-0.002
 eighty-0.002
 cand-0.002
 Ske-0.002
Token's
Token ID338
Feature activation-9.47e-01

Pos logit contributions
rob+0.024
 l+0.021
 Ske+0.019
 eighty+0.018
 cand+0.018
Neg logit contributions
isel-0.025
icker-0.025
oths-0.025
umped-0.023
imming-0.023
Token a
Token ID257
Feature activation-1.21e+00

Pos logit contributions
rob+0.039
 l+0.030
 eighty+0.028
 Ske+0.027
 cand+0.026
Neg logit contributions
isel-0.048
icker-0.047
oths-0.047
apa-0.045
umped-0.045
Token new
Token ID649
Feature activation-7.36e-01

Pos logit contributions
rob+0.060
 eighty+0.044
 l+0.042
 cand+0.040
 Ske+0.040
Neg logit contributions
isel-0.051
icker-0.051
ench-0.048
oths-0.048
apa-0.048
Token tool
Token ID2891
Feature activation-2.26e+00

Pos logit contributions
rob+0.037
 eighty+0.027
 cand+0.024
 l+0.024
 seventy+0.024
Neg logit contributions
isel-0.030
ench-0.030
ap-0.030
ooks-0.029
icker-0.029
Token to
Token ID284
Feature activation-1.06e+00

Pos logit contributions
rob+0.107
 l+0.085
 eighty+0.082
 Ske+0.077
 cand+0.076
Neg logit contributions
isel-0.096
icker-0.094
oths-0.094
apa-0.091
ap-0.090
Token open
Token ID1280
Feature activation-3.64e-01

Pos logit contributions
rob+0.052
 l+0.041
 eighty+0.040
 Ske+0.037
 cand+0.037
Neg logit contributions
isel-0.046
icker-0.044
oths-0.043
apa-0.042
ap-0.041
Token cans
Token ID23916
Feature activation+1.82e-01

Pos logit contributions
rob+0.015
 l+0.012
 eighty+0.011
 Ske+0.011
 cand+0.011
Neg logit contributions
isel-0.018
icker-0.018
oths-0.017
umped-0.017
apa-0.017
Token faster
Token ID5443
Feature activation+9.55e-02

Pos logit contributions
isel+0.010
icker+0.009
oths+0.009
umped+0.009
apa+0.009
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.005
 cand-0.005
 Ske-0.005
Token."
Token ID526
Feature activation-4.95e-01

Pos logit contributions
isel+0.005
icker+0.005
oths+0.005
apa+0.005
umped+0.005
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 cand-0.003
 Ske-0.003
Token could
Token ID714
Feature activation-2.26e-01

Pos logit contributions
rob+0.037
 l+0.030
 eighty+0.028
 Ske+0.027
 cand+0.027
Neg logit contributions
isel-0.030
icker-0.030
oths-0.029
umped-0.028
apa-0.028
Token use
Token ID779
Feature activation-5.09e-01

Pos logit contributions
rob+0.010
 l+0.009
 Ske+0.008
 eighty+0.008
 cand+0.007
Neg logit contributions
isel-0.010
oths-0.010
icker-0.010
apa-0.010
umped-0.010
Token her
Token ID607
Feature activation-3.73e-01

Pos logit contributions
rob+0.018
 l+0.013
 eighty+0.012
 Ske+0.012
 cand+0.011
Neg logit contributions
isel-0.029
icker-0.028
oths-0.028
apa-0.027
umped-0.027
Token daddy
Token ID35822
Feature activation+2.71e-01

Pos logit contributions
rob+0.015
 l+0.015
 candy+0.013
 cand+0.013
 Ske+0.013
Neg logit contributions
ench-0.016
icker-0.016
umped-0.016
itating-0.016
imming-0.016
Token's
Token ID338
Feature activation-3.24e-01

Pos logit contributions
isel+0.012
icker+0.011
oths+0.011
ooks+0.011
umped+0.011
Neg logit contributions
rob-0.013
 l-0.011
 eighty-0.010
 cand-0.010
 Ske-0.010
Token tool
Token ID2891
Feature activation-2.26e+00

Pos logit contributions
 l+0.012
rob+0.012
 candy+0.011
 cand+0.011
 scar+0.010
Neg logit contributions
itating-0.015
icker-0.014
umped-0.014
ench-0.014
oths-0.014
Tokenbox
Token ID3524
Feature activation-1.16e+00

Pos logit contributions
rob+0.098
 l+0.081
 eighty+0.073
 Ske+0.072
 cand+0.072
Neg logit contributions
isel-0.108
icker-0.105
oths-0.104
umped-0.100
imming-0.099
Token to
Token ID284
Feature activation-5.65e-01

Pos logit contributions
rob+0.049
 l+0.041
 eighty+0.037
 cand+0.036
 Ske+0.035
Neg logit contributions
isel-0.057
icker-0.055
oths-0.055
imming-0.052
apa-0.052
Token fix
Token ID4259
Feature activation-4.13e-01

Pos logit contributions
rob+0.023
 l+0.021
 Ske+0.018
 cand+0.018
 eighty+0.018
Neg logit contributions
oths-0.027
isel-0.026
umped-0.026
icker-0.026
imming-0.026
Token anything
Token ID1997
Feature activation-3.93e-01

Pos logit contributions
rob+0.016
 l+0.013
 eighty+0.011
 Ske+0.011
 cand+0.011
Neg logit contributions
isel-0.022
icker-0.021
oths-0.021
umped-0.020
imming-0.020
Token that
Token ID326
Feature activation+8.72e-01

Pos logit contributions
rob+0.019
 l+0.014
 eighty+0.014
 Ske+0.013
 cand+0.013
Neg logit contributions
isel-0.017
icker-0.017
oths-0.017
apa-0.016
umped-0.016
Token do
Token ID466
Feature activation+1.50e-01

Pos logit contributions
rob+0.008
 l+0.007
 cand+0.006
 eighty+0.006
 Ske+0.006
Neg logit contributions
isel-0.010
icker-0.010
oths-0.010
umped-0.010
imming-0.010
Token.
Token ID13
Feature activation-1.24e-01

Pos logit contributions
isel+0.008
icker+0.007
oths+0.007
umped+0.007
apa+0.007
Neg logit contributions
rob-0.006
 l-0.005
 cand-0.005
 eighty-0.004
 Ske-0.004
Token From
Token ID3574
Feature activation+1.43e-01

Pos logit contributions
rob+0.006
 l+0.005
 Ske+0.005
 eighty+0.005
 Velvet+0.005
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
umped-0.005
imming-0.005
Token time
Token ID640
Feature activation-2.52e+00

Pos logit contributions
isel+0.008
icker+0.008
oths+0.007
apa+0.007
umped+0.007
Neg logit contributions
rob-0.005
 l-0.004
 eighty-0.004
 cand-0.004
 Ske-0.004
Token to
Token ID284
Feature activation-6.71e-01

Pos logit contributions
rob+0.075
 l+0.066
 eighty+0.065
 Star+0.062
 Ske+0.062
Neg logit contributions
isel-0.148
icker-0.141
ench-0.137
oths-0.135
imming-0.134
Token time
Token ID640
Feature activation-2.25e+00

Pos logit contributions
rob+0.028
 l+0.024
 eighty+0.021
 Ske+0.021
 cand+0.020
Neg logit contributions
isel-0.033
icker-0.033
oths-0.032
umped-0.031
apa-0.031
Token people
Token ID661
Feature activation+3.77e-01

Pos logit contributions
rob+0.076
 l+0.067
 eighty+0.061
 Ske+0.059
 Star+0.058
Neg logit contributions
isel-0.127
icker-0.121
oths-0.116
imming-0.114
ooks-0.113
Token visited
Token ID8672
Feature activation-1.41e-01

Pos logit contributions
isel+0.020
icker+0.019
oths+0.019
aining+0.019
ooks+0.018
Neg logit contributions
rob-0.015
 l-0.013
 eighty-0.011
 cand-0.011
 Ske-0.011
Token the
Token ID262
Feature activation-1.91e-01

Pos logit contributions
rob+0.005
 l+0.004
 Ske+0.004
 cand+0.004
 Star+0.004
Neg logit contributions
isel-0.008
icker-0.008
oths-0.008
imming-0.007
umped-0.007
Token Earth
Token ID3668
Feature activation-4.07e-01

Pos logit contributions
rob+0.009
 l+0.009
 eighty+0.008
 cand+0.007
 Ske+0.007
Neg logit contributions
isel-0.008
apa-0.008
oths-0.008
icker-0.008
umped-0.007
Token from
Token ID422
Feature activation+2.43e-01

Pos logit contributions
rob+0.016
 l+0.014
 Ske+0.012
 eighty+0.012
 cand+0.012
Neg logit contributions
isel-0.022
icker-0.020
oths-0.020
imming-0.019
apa-0.019
Token excited
Token ID6568
Feature activation-7.22e-01

Pos logit contributions
rob+0.020
 l+0.019
 eighty+0.018
 Ske+0.017
 he+0.017
Neg logit contributions
isel-0.028
aining-0.027
oths-0.027
imming-0.026
leys-0.026
Token because
Token ID780
Feature activation-3.41e-01

Pos logit contributions
rob+0.034
 l+0.028
 eighty+0.026
 Ske+0.025
 cand+0.025
Neg logit contributions
isel-0.032
icker-0.031
oths-0.031
apa-0.030
umped-0.029
Token today
Token ID1909
Feature activation-5.16e-02

Pos logit contributions
rob+0.009
 l+0.007
 eighty+0.005
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.022
icker-0.022
oths-0.021
umped-0.021
apa-0.021
Token was
Token ID373
Feature activation-4.47e-01

Pos logit contributions
rob+0.002
 l+0.001
 Ske+0.001
 cand+0.001
 candy+0.001
Neg logit contributions
isel-0.003
icker-0.003
oths-0.003
imming-0.003
umped-0.003
Token his
Token ID465
Feature activation-4.88e-01

Pos logit contributions
rob+0.018
 l+0.014
 eighty+0.013
 Ske+0.013
 cand+0.012
Neg logit contributions
isel-0.023
icker-0.022
oths-0.022
apa-0.021
umped-0.021
Token first
Token ID717
Feature activation-2.25e+00

Pos logit contributions
rob+0.023
 l+0.018
 eighty+0.017
 cand+0.016
 Ske+0.016
Neg logit contributions
icker-0.022
isel-0.022
oths-0.022
apa-0.021
ap-0.021
Token day
Token ID1110
Feature activation-9.07e-01

Pos logit contributions
rob+0.090
 l+0.077
 eighty+0.067
 Ske+0.066
 cand+0.064
Neg logit contributions
isel-0.114
icker-0.110
oths-0.110
umped-0.106
imming-0.105
Token of
Token ID286
Feature activation+4.75e-02

Pos logit contributions
rob+0.030
 l+0.027
 Ske+0.023
 candy+0.023
 cand+0.022
Neg logit contributions
isel-0.052
icker-0.050
oths-0.049
umped-0.048
imming-0.048
Token school
Token ID1524
Feature activation-1.20e+00

Pos logit contributions
isel+0.002
oths+0.002
icker+0.002
ap+0.002
imming+0.002
Neg logit contributions
rob-0.002
 l-0.002
 eighty-0.002
 cand-0.001
 Ske-0.001
Token.
Token ID13
Feature activation-3.87e-02

Pos logit contributions
rob+0.063
 l+0.050
 eighty+0.050
 Ske+0.047
 cand+0.047
Neg logit contributions
icker-0.044
oths-0.044
isel-0.043
ap-0.043
umped-0.041
Token He
Token ID679
Feature activation-3.68e-02

Pos logit contributions
rob+0.001
 l+0.001
 Ske+0.001
 eighty+0.001
 he+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
inging-0.002
apa-0.002
Token and
Token ID290
Feature activation+5.13e-01

Pos logit contributions
isel+0.003
oths+0.003
icker+0.003
imming+0.003
apa+0.003
Neg logit contributions
rob-0.002
 l-0.002
 eighty-0.002
 Ske-0.002
 cand-0.002
Token hugged
Token ID41130
Feature activation+1.22e-01

Pos logit contributions
isel+0.030
icker+0.030
oths+0.029
imming+0.028
ooks+0.028
Neg logit contributions
rob-0.017
 l-0.015
 eighty-0.012
 Ske-0.012
 cand-0.011
Token the
Token ID262
Feature activation+3.50e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
imming+0.007
umped+0.007
Neg logit contributions
rob-0.004
 l-0.003
 Ske-0.003
 eighty-0.003
 cand-0.003
Token rh
Token ID9529
Feature activation+1.20e+00

Pos logit contributions
isel+0.017
icker+0.017
oths+0.017
apa+0.016
amps+0.016
Neg logit contributions
rob-0.014
 l-0.009
 eighty-0.009
 Velvet-0.009
 Ske-0.009
Tokenin
Token ID259
Feature activation+5.87e-01

Pos logit contributions
isel+0.047
oths+0.045
icker+0.045
apa+0.043
umped+0.043
Neg logit contributions
rob-0.064
 l-0.054
 eighty-0.050
 cand-0.049
 Ske-0.049
Tokenoc
Token ID420
Feature activation+3.05e+00

Pos logit contributions
isel+0.019
icker+0.018
oths+0.018
apa+0.017
umped+0.017
Neg logit contributions
rob-0.034
 l-0.029
 eighty-0.028
 Ske-0.027
 cand-0.027
Tokeneros
Token ID27498
Feature activation+4.86e-01

Pos logit contributions
isel+0.148
icker+0.145
oths+0.143
apa+0.137
umped+0.137
Neg logit contributions
rob-0.133
 l-0.107
 eighty-0.098
 cand-0.096
 Ske-0.095
Token.
Token ID13
Feature activation+3.85e-01

Pos logit contributions
isel+0.026
oths+0.025
icker+0.024
umped+0.023
imming+0.023
Neg logit contributions
rob-0.018
 l-0.018
 eighty-0.015
 Ske-0.014
 cand-0.014
Token He
Token ID679
Feature activation+5.91e-01

Pos logit contributions
isel+0.020
icker+0.020
oths+0.019
leys+0.019
imming+0.019
Neg logit contributions
rob-0.015
 l-0.013
 Ske-0.012
 eighty-0.012
 Star-0.011
Token finally
Token ID3443
Feature activation-3.56e-01

Pos logit contributions
isel+0.028
icker+0.028
oths+0.027
umped+0.026
apa+0.026
Neg logit contributions
rob-0.026
 l-0.021
 eighty-0.019
 Ske-0.019
 cand-0.019
Token understood
Token ID7247
Feature activation+7.96e-01

Pos logit contributions
rob+0.015
 l+0.013
 eighty+0.012
 Ske+0.011
 cand+0.011
Neg logit contributions
oths-0.017
icker-0.017
isel-0.017
imming-0.016
ooks-0.016
Token try
Token ID1949
Feature activation+9.08e-01

Pos logit contributions
rob+0.013
 l+0.010
 eighty+0.010
 Ske+0.009
 cand+0.009
Neg logit contributions
isel-0.011
icker-0.010
oths-0.010
apa-0.010
umped-0.010
Token to
Token ID284
Feature activation+1.47e-01

Pos logit contributions
isel+0.049
icker+0.048
oths+0.047
apa+0.047
umped+0.046
Neg logit contributions
rob-0.034
 l-0.030
 Ske-0.025
 eighty-0.025
 Velvet-0.024
Token grab
Token ID5552
Feature activation-2.29e-01

Pos logit contributions
isel+0.006
icker+0.006
oths+0.006
apa+0.006
umped+0.006
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.006
 Ske-0.006
 cand-0.006
Token the
Token ID262
Feature activation+2.16e-01

Pos logit contributions
rob+0.008
 l+0.007
 Ske+0.006
 eighty+0.006
 cand+0.006
Neg logit contributions
icker-0.013
isel-0.012
oths-0.012
umped-0.012
leys-0.012
Token av
Token ID1196
Feature activation+2.74e+00

Pos logit contributions
icker+0.010
isel+0.010
oths+0.010
umped+0.009
apa+0.009
Neg logit contributions
rob-0.010
 l-0.009
 cand-0.008
 eighty-0.008
 candy-0.007
Tokenoc
Token ID420
Feature activation+2.96e+00

Pos logit contributions
isel+0.095
icker+0.092
oths+0.091
apa+0.086
umped+0.086
Neg logit contributions
rob-0.155
 l-0.132
 eighty-0.125
 Ske-0.122
 cand-0.121
Tokenados
Token ID22484
Feature activation+2.82e-01

Pos logit contributions
icker+0.143
isel+0.139
leys+0.134
oths+0.132
eways+0.131
Neg logit contributions
rob-0.130
 l-0.104
 cand-0.103
 eighty-0.096
 ninety-0.091
Token.
Token ID13
Feature activation-9.85e-02

Pos logit contributions
isel+0.014
oths+0.014
icker+0.014
umped+0.013
apa+0.013
Neg logit contributions
rob-0.012
 l-0.010
 eighty-0.009
 Ske-0.009
 cand-0.009
Token But
Token ID887
Feature activation-7.79e-01

Pos logit contributions
rob+0.004
 l+0.004
 Ske+0.004
 Velvet+0.003
 Star+0.003
Neg logit contributions
isel-0.005
imming-0.004
icker-0.004
leys-0.004
umped-0.004
Token the
Token ID262
Feature activation-1.59e-01

Pos logit contributions
rob+0.032
 l+0.026
 eighty+0.024
 Ske+0.023
 cand+0.023
Neg logit contributions
isel-0.039
icker-0.039
oths-0.038
umped-0.037
apa-0.036
Token rock
Token ID3881
Feature activation-7.88e-01

Pos logit contributions
rob+0.007
 l+0.006
 eighty+0.005
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.007
icker-0.007
oths-0.007
apa-0.007
umped-0.007
Token,"
Token ID553
Feature activation+4.26e-01

Pos logit contributions
isel+0.005
apa+0.005
icker+0.005
ooks+0.005
oths+0.005
Neg logit contributions
rob-0.008
 l-0.006
 eighty-0.006
 Velvet-0.006
 cand-0.006
Token said
Token ID531
Feature activation-7.81e-01

Pos logit contributions
isel+0.020
apa+0.020
icker+0.019
ap+0.019
oths+0.019
Neg logit contributions
rob-0.018
 eighty-0.013
 ninety-0.012
 seventy-0.012
 Velvet-0.012
Token the
Token ID262
Feature activation+4.74e-01

Pos logit contributions
rob+0.039
 l+0.029
 eighty+0.029
 cand+0.027
 ninety+0.026
Neg logit contributions
isel-0.033
icker-0.033
apa-0.032
ap-0.031
oths-0.031
Token rh
Token ID9529
Feature activation+1.20e+00

Pos logit contributions
inging+0.022
amps+0.021
isel+0.021
umped+0.021
inct+0.021
Neg logit contributions
rob-0.018
 Velvet-0.012
 l-0.011
 Ske-0.011
 eighty-0.011
Tokenin
Token ID259
Feature activation+4.24e-01

Pos logit contributions
isel+0.038
icker+0.038
oths+0.037
umped+0.035
apa+0.034
Neg logit contributions
rob-0.069
 l-0.060
 eighty-0.057
 Ske-0.056
 cand-0.055
Tokenoc
Token ID420
Feature activation+2.63e+00

Pos logit contributions
isel+0.019
oths+0.019
icker+0.019
imming+0.018
umped+0.018
Neg logit contributions
rob-0.019
 l-0.017
 Ske-0.015
 eighty-0.015
 cand-0.014
Tokeneros
Token ID27498
Feature activation+3.53e-01

Pos logit contributions
isel+0.105
oths+0.102
icker+0.100
aining+0.098
umped+0.098
Neg logit contributions
rob-0.127
 l-0.107
 Velvet-0.098
 eighty-0.098
 scar-0.098
Token,
Token ID11
Feature activation+1.93e-01

Pos logit contributions
isel+0.019
oths+0.019
icker+0.019
imming+0.018
umped+0.018
Neg logit contributions
 l-0.012
rob-0.012
 Ske-0.010
 eighty-0.009
 cand-0.009
Token feeling
Token ID4203
Feature activation-4.21e-02

Pos logit contributions
isel+0.009
icker+0.008
oths+0.008
umped+0.008
imming+0.008
Neg logit contributions
rob-0.009
 l-0.008
 eighty-0.007
 Ske-0.007
 cand-0.007
Token discouraged
Token ID30170
Feature activation-4.43e-02

Pos logit contributions
rob+0.002
 l+0.002
 eighty+0.001
 Ske+0.001
 cand+0.001
Neg logit contributions
oths-0.002
icker-0.002
isel-0.002
aining-0.002
imming-0.002
Token.
Token ID13
Feature activation+2.98e-01

Pos logit contributions
rob+0.002
 l+0.002
 eighty+0.002
 Ske+0.002
 cand+0.002
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
apa-0.002
umped-0.002
Token examine
Token ID10716
Feature activation-1.73e-01

Pos logit contributions
isel+0.002
oths+0.002
icker+0.002
umped+0.002
imming+0.002
Neg logit contributions
rob-0.002
 l-0.001
 Ske-0.001
 eighty-0.001
 Velvet-0.001
Token it
Token ID340
Feature activation-5.48e-01

Pos logit contributions
rob+0.008
 l+0.006
 eighty+0.006
 Ske+0.005
 cand+0.005
Neg logit contributions
isel-0.008
icker-0.008
oths-0.008
apa-0.008
umped-0.008
Token with
Token ID351
Feature activation+3.76e-01

Pos logit contributions
rob+0.028
 l+0.022
 eighty+0.020
 cand+0.020
 Ske+0.020
Neg logit contributions
icker-0.022
isel-0.022
oths-0.022
ap-0.021
ench-0.021
Token the
Token ID262
Feature activation-5.38e-02

Pos logit contributions
isel+0.019
icker+0.018
oths+0.018
apa+0.018
umped+0.017
Neg logit contributions
rob-0.016
 l-0.012
 eighty-0.011
 Ske-0.011
 cand-0.011
Token magn
Token ID7842
Feature activation-1.91e-01

Pos logit contributions
rob+0.003
 l+0.002
 eighty+0.002
 Ske+0.002
 cand+0.002
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
apa-0.002
ap-0.002
Tokenifying
Token ID4035
Feature activation+2.86e+00

Pos logit contributions
rob+0.014
 l+0.013
 eighty+0.013
 Ske+0.012
 cand+0.012
Neg logit contributions
isel-0.003
icker-0.003
oths-0.003
ooks-0.002
apa-0.002
Token glass
Token ID5405
Feature activation-3.49e-01

Pos logit contributions
isel+0.133
icker+0.131
oths+0.129
apa+0.124
umped+0.124
Neg logit contributions
rob-0.128
 l-0.100
 eighty-0.095
 Ske-0.091
 cand-0.089
Token?"
Token ID1701
Feature activation+6.55e-01

Pos logit contributions
rob+0.016
 l+0.013
 eighty+0.012
 Ske+0.012
 cand+0.012
Neg logit contributions
isel-0.015
icker-0.015
oths-0.015
apa-0.014
ap-0.014
Token
Token ID198
Feature activation+4.87e-01

Pos logit contributions
oths+0.028
isel+0.027
icker+0.027
apa+0.026
umped+0.025
Neg logit contributions
rob-0.033
 eighty-0.026
 l-0.025
 cand-0.025
 ninety-0.024
Token
Token ID198
Feature activation+5.67e-01

Pos logit contributions
isel+0.025
icker+0.024
oths+0.023
imming+0.023
umped+0.023
Neg logit contributions
rob-0.020
 l-0.016
 Ske-0.015
 eighty-0.014
 Velvet-0.014
Token"
Token ID1
Feature activation-7.74e-02

Pos logit contributions
isel+0.031
icker+0.029
imming+0.029
oths+0.028
amps+0.028
Neg logit contributions
rob-0.020
 l-0.019
 Ske-0.016
 eighty-0.015
 Velvet-0.015
Token
Token ID198
Feature activation+2.95e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
apa+0.004
umped+0.004
Neg logit contributions
rob-0.004
 l-0.004
 eighty-0.003
 cand-0.003
 Ske-0.003
TokenHe
Token ID1544
Feature activation+2.38e-01

Pos logit contributions
isel+0.015
oths+0.015
icker+0.015
ains+0.014
apa+0.014
Neg logit contributions
rob-0.011
 l-0.010
 Ske-0.009
 eighty-0.008
 he-0.008
Token brought
Token ID3181
Feature activation+3.06e-01

Pos logit contributions
isel+0.010
icker+0.009
oths+0.009
apa+0.009
umped+0.009
Neg logit contributions
rob-0.012
 l-0.010
 eighty-0.009
 cand-0.009
 Ske-0.009
Token his
Token ID465
Feature activation+8.23e-01

Pos logit contributions
icker+0.018
isel+0.018
oths+0.018
imming+0.017
aining+0.017
Neg logit contributions
rob-0.010
 l-0.008
 eighty-0.007
 Ske-0.007
 cand-0.007
Token magn
Token ID7842
Feature activation-2.57e-01

Pos logit contributions
isel+0.037
icker+0.035
oths+0.035
ooks+0.034
aining+0.034
Neg logit contributions
rob-0.038
 l-0.035
 cand-0.030
 Ske-0.030
 eighty-0.030
Tokenifying
Token ID4035
Feature activation+3.08e+00

Pos logit contributions
rob+0.017
 eighty+0.016
 l+0.016
 seventy+0.015
 Ske+0.015
Neg logit contributions
isel-0.005
oths-0.005
icker-0.004
appiness-0.004
ooks-0.004
Token glass
Token ID5405
Feature activation-2.83e-01

Pos logit contributions
isel+0.152
icker+0.149
oths+0.148
apa+0.142
imming+0.142
Neg logit contributions
rob-0.129
 l-0.106
 eighty-0.095
 Ske-0.093
 cand-0.092
Token and
Token ID290
Feature activation-5.45e-03

Pos logit contributions
rob+0.012
 l+0.010
 eighty+0.009
 cand+0.009
 Ske+0.009
Neg logit contributions
isel-0.014
oths-0.013
icker-0.013
imming-0.013
umped-0.013
Token looked
Token ID3114
Feature activation+8.36e-01

Pos logit contributions
rob+0.000
 l+0.000
 cand+0.000
 Ske+0.000
 eighty+0.000
Neg logit contributions
icker-0.000
isel-0.000
oths-0.000
imming-0.000
aining-0.000
Token closely
Token ID7173
Feature activation+1.31e+00

Pos logit contributions
oths+0.041
isel+0.041
icker+0.041
umped+0.040
apa+0.039
Neg logit contributions
rob-0.033
 l-0.030
 eighty-0.026
 Ske-0.025
 scar-0.024
Token.
Token ID13
Feature activation+5.38e-01

Pos logit contributions
isel+0.065
icker+0.063
oths+0.063
umped+0.060
imming+0.060
Neg logit contributions
rob-0.053
 l-0.045
 eighty-0.041
 Ske-0.039
 cand-0.038
Token and
Token ID290
Feature activation+1.61e-01

Pos logit contributions
isel+0.014
icker+0.014
oths+0.013
apa+0.013
imming+0.013
Neg logit contributions
rob-0.012
 l-0.010
 Ske-0.009
 eighty-0.009
 cand-0.009
Token soon
Token ID2582
Feature activation+2.76e-01

Pos logit contributions
isel+0.008
icker+0.007
oths+0.007
apa+0.007
imming+0.007
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.005
 Ske-0.005
 cand-0.005
Token became
Token ID2627
Feature activation+1.65e-02

Pos logit contributions
isel+0.016
icker+0.016
oths+0.015
apa+0.015
umped+0.015
Neg logit contributions
rob-0.009
 l-0.007
 eighty-0.006
 Ske-0.006
 cand-0.006
Token very
Token ID845
Feature activation+2.17e-01

Pos logit contributions
isel+0.001
icker+0.001
oths+0.001
imming+0.001
apa+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.000
 Ske-0.000
 cand-0.000
Token good
Token ID922
Feature activation-2.28e-01

Pos logit contributions
isel+0.010
oths+0.010
apa+0.010
imming+0.010
icker+0.010
Neg logit contributions
 l-0.008
rob-0.008
 eighty-0.007
 Ske-0.007
 ninety-0.007
Token at
Token ID379
Feature activation+1.11e+00

Pos logit contributions
rob+0.011
 l+0.009
 Ske+0.008
 eighty+0.008
 cand+0.008
Neg logit contributions
isel-0.010
icker-0.010
oths-0.009
apa-0.009
imming-0.009
Token playing
Token ID2712
Feature activation-1.78e-01

Pos logit contributions
isel+0.055
icker+0.054
oths+0.053
apa+0.052
umped+0.051
Neg logit contributions
rob-0.047
 l-0.038
 eighty-0.034
 Ske-0.033
 cand-0.033
Token her
Token ID607
Feature activation-3.38e-01

Pos logit contributions
rob+0.008
 l+0.007
 eighty+0.006
 Ske+0.006
 cand+0.006
Neg logit contributions
isel-0.008
icker-0.008
oths-0.008
umped-0.007
apa-0.007
Token fl
Token ID781
Feature activation-6.83e-01

Pos logit contributions
rob+0.013
 l+0.010
 eighty+0.010
 Ske+0.009
 cand+0.009
Neg logit contributions
icker-0.017
isel-0.017
oths-0.017
imming-0.016
umped-0.016
Tokenute
Token ID1133
Feature activation-3.22e-01

Pos logit contributions
rob+0.040
 l+0.035
 eighty+0.033
 Ske+0.033
 Velvet+0.032
Neg logit contributions
isel-0.020
oths-0.019
leys-0.018
apa-0.018
umped-0.017
Token.
Token ID13
Feature activation+9.97e-02

Pos logit contributions
rob+0.013
 l+0.011
 Ske+0.009
 eighty+0.009
 cand+0.009
Neg logit contributions
isel-0.017
oths-0.016
icker-0.016
apa-0.016
umped-0.015
Token time
Token ID640
Feature activation-3.21e-01

Pos logit contributions
isel+0.011
icker+0.010
oths+0.010
apa+0.010
umped+0.010
Neg logit contributions
rob-0.014
 l-0.012
 eighty-0.011
 Ske-0.011
 cand-0.011
Token,
Token ID11
Feature activation-3.95e-04

Pos logit contributions
rob+0.017
 l+0.014
 eighty+0.013
 cand+0.012
 Ske+0.012
Neg logit contributions
isel-0.013
oths-0.013
icker-0.013
apa-0.012
ap-0.012
Token a
Token ID257
Feature activation+7.20e-02

Pos logit contributions
rob+0.000
 l+0.000
 eighty+0.000
 Ske+0.000
 cand+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
apa-0.000
umped-0.000
Token little
Token ID1310
Feature activation+1.23e+00

Pos logit contributions
isel+0.003
oths+0.003
icker+0.003
aining+0.003
umped+0.003
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 Ske-0.003
 Velvet-0.003
Token girl
Token ID2576
Feature activation+4.98e-01

Pos logit contributions
icker+0.062
isel+0.062
oths+0.060
ap+0.058
apa+0.058
Neg logit contributions
rob-0.049
 l-0.038
 eighty-0.036
 Ske-0.034
 cand-0.034
Token named
Token ID3706
Feature activation+9.27e-01

Pos logit contributions
isel+0.023
icker+0.022
imming+0.022
aining+0.022
umped+0.021
Neg logit contributions
rob-0.022
 l-0.020
 eighty-0.017
 Ske-0.017
 cand-0.016
Token Lily
Token ID20037
Feature activation+2.51e-01

Pos logit contributions
isel+0.061
icker+0.060
oths+0.059
apa+0.058
umped+0.058
Neg logit contributions
rob-0.024
 l-0.017
 eighty-0.014
 Ske-0.013
 cand-0.013
Token went
Token ID1816
Feature activation-7.71e-01

Pos logit contributions
isel+0.012
icker+0.012
oths+0.012
imming+0.011
apa+0.011
Neg logit contributions
rob-0.011
 l-0.009
 eighty-0.008
 Ske-0.008
 cand-0.008
Token to
Token ID284
Feature activation-6.23e-01

Pos logit contributions
rob+0.043
 eighty+0.034
 cand+0.033
 l+0.033
 ninety+0.033
Neg logit contributions
icker-0.026
ap-0.025
oths-0.025
isel-0.024
leys-0.023
Token a
Token ID257
Feature activation+1.95e-01

Pos logit contributions
rob+0.039
 cand+0.030
 eighty+0.029
 ninety+0.028
 seventy+0.027
Neg logit contributions
icker-0.018
oths-0.017
ap-0.017
isel-0.016
 forces-0.016
Token show
Token ID905
Feature activation+4.51e-02

Pos logit contributions
icker+0.008
oths+0.007
isel+0.007
ap+0.007
ench+0.007
Neg logit contributions
rob-0.010
 eighty-0.008
 l-0.007
 cand-0.007
 Ske-0.007
Token closely
Token ID7173
Feature activation+3.62e-01

Pos logit contributions
isel+0.007
oths+0.006
icker+0.006
ooks+0.006
apa+0.006
Neg logit contributions
 l-0.005
rob-0.005
 cand-0.004
 Ske-0.004
 eighty-0.004
Token together
Token ID1978
Feature activation+1.17e+00

Pos logit contributions
isel+0.018
icker+0.017
oths+0.017
umped+0.016
imming+0.016
Neg logit contributions
rob-0.015
 l-0.013
 eighty-0.011
 cand-0.011
 Ske-0.011
Token.
Token ID13
Feature activation+1.75e-01

Pos logit contributions
isel+0.065
oths+0.063
icker+0.062
imming+0.060
ench+0.060
Neg logit contributions
 l-0.036
rob-0.035
 Ske-0.029
 cand-0.029
 eighty-0.028
Token She
Token ID1375
Feature activation+1.75e-01

Pos logit contributions
isel+0.010
icker+0.009
imming+0.009
oths+0.009
apa+0.009
Neg logit contributions
 l-0.005
rob-0.005
 Ske-0.004
 eighty-0.004
 candy-0.004
Token looked
Token ID3114
Feature activation+6.64e-01

Pos logit contributions
isel+0.009
icker+0.008
oths+0.008
imming+0.008
aining+0.008
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.005
 Ske-0.005
 cand-0.005
Token closely
Token ID7173
Feature activation+1.09e+00

Pos logit contributions
isel+0.029
icker+0.029
oths+0.029
apa+0.028
umped+0.028
Neg logit contributions
rob-0.030
 l-0.027
 eighty-0.024
 Ske-0.023
 cand-0.023
Token at
Token ID379
Feature activation+8.09e-01

Pos logit contributions
isel+0.055
icker+0.052
oths+0.052
imming+0.050
umped+0.050
Neg logit contributions
rob-0.041
 l-0.039
 eighty-0.033
 cand-0.032
 Ske-0.032
Token her
Token ID607
Feature activation+5.84e-01

Pos logit contributions
isel+0.044
icker+0.043
oths+0.043
umped+0.042
apa+0.042
Neg logit contributions
rob-0.029
 l-0.024
 eighty-0.021
 Ske-0.020
 cand-0.020
Token fingers
Token ID9353
Feature activation+7.20e-01

Pos logit contributions
isel+0.029
oths+0.027
icker+0.027
leys+0.026
ooks+0.026
Neg logit contributions
 l-0.023
rob-0.022
 Ske-0.019
 cand-0.019
 candy-0.018
Token and
Token ID290
Feature activation-3.14e-01

Pos logit contributions
isel+0.039
oths+0.038
icker+0.037
apa+0.036
ooks+0.036
Neg logit contributions
rob-0.025
 l-0.023
 cand-0.019
 eighty-0.019
 Ske-0.019
Token tried
Token ID3088
Feature activation+9.82e-01

Pos logit contributions
rob+0.013
 l+0.011
 eighty+0.009
 Ske+0.009
 cand+0.009
Neg logit contributions
isel-0.016
icker-0.016
oths-0.015
imming-0.015
apa-0.015
TokenNo
Token ID2949
Feature activation-6.29e-01

Pos logit contributions
isel+0.019
icker+0.019
oths+0.019
apa+0.018
umped+0.018
Neg logit contributions
rob-0.021
 l-0.018
 eighty-0.016
 Ske-0.016
 cand-0.016
Token,
Token ID11
Feature activation-6.41e-01

Pos logit contributions
rob+0.030
 l+0.024
 eighty+0.023
 Ske+0.022
 cand+0.021
Neg logit contributions
isel-0.028
icker-0.027
oths-0.027
apa-0.026
umped-0.026
Token you
Token ID345
Feature activation-1.54e-01

Pos logit contributions
rob+0.020
 l+0.018
 eighty+0.015
 cand+0.015
 Ske+0.015
Neg logit contributions
icker-0.037
isel-0.037
oths-0.036
umped-0.036
ench-0.035
Token'll
Token ID1183
Feature activation+4.61e-01

Pos logit contributions
 l+0.004
 Ske+0.004
 candy+0.003
 nap+0.003
 scar+0.003
Neg logit contributions
eways-0.009
oths-0.008
icker-0.008
ments-0.008
ooks-0.008
Token get
Token ID651
Feature activation+1.56e-01

Pos logit contributions
oths+0.026
isel+0.026
leys+0.025
icker+0.025
eways+0.025
Neg logit contributions
 l-0.014
rob-0.014
 Ske-0.012
 Velvet-0.012
 Star-0.011
Token too
Token ID1165
Feature activation+9.16e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
umped+0.007
imming+0.007
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.005
 cand-0.005
 Ske-0.005
Token wet
Token ID9583
Feature activation+6.34e-01

Pos logit contributions
isel+0.047
imming+0.045
umped+0.044
arre+0.044
icker+0.044
Neg logit contributions
 l-0.030
 candy-0.026
rob-0.026
 eighty-0.025
 he-0.024
Token.
Token ID13
Feature activation+1.72e-01

Pos logit contributions
isel+0.032
icker+0.031
oths+0.030
apa+0.030
umped+0.029
Neg logit contributions
rob-0.027
 l-0.021
 eighty-0.020
 Ske-0.019
 cand-0.019
Token Come
Token ID7911
Feature activation+9.87e-02

Pos logit contributions
leys+0.010
isel+0.009
eways+0.009
icker+0.009
imming+0.009
Neg logit contributions
rob-0.005
 Velvet-0.005
 Ske-0.005
 Star-0.005
 l-0.004
Token on
Token ID319
Feature activation-6.62e-02

Pos logit contributions
isel+0.005
icker+0.005
leys+0.005
apa+0.005
oths+0.005
Neg logit contributions
rob-0.003
 l-0.003
 Ske-0.003
 Velvet-0.003
 eighty-0.003
Token,
Token ID11
Feature activation-2.25e-01

Pos logit contributions
rob+0.002
 l+0.002
 Velvet+0.002
 Star+0.002
 Ske+0.001
Neg logit contributions
isel-0.004
icker-0.004
leys-0.004
imming-0.004
oths-0.004
Token again
Token ID757
Feature activation+2.62e-01

Pos logit contributions
rob+0.000
 l+0.000
 eighty+0.000
 Ske+0.000
 cand+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
umped-0.000
apa-0.000
Token.
Token ID13
Feature activation+5.28e-01

Pos logit contributions
isel+0.014
icker+0.013
oths+0.013
imming+0.013
apa+0.013
Neg logit contributions
rob-0.010
 l-0.009
 Ske-0.008
 eighty-0.007
 cand-0.007
Token He
Token ID679
Feature activation+4.93e-01

Pos logit contributions
isel+0.029
icker+0.029
oths+0.028
imming+0.027
leys+0.027
Neg logit contributions
rob-0.018
 l-0.015
 Ske-0.014
 eighty-0.014
 Velvet-0.012
Token realized
Token ID6939
Feature activation+5.24e-01

Pos logit contributions
isel+0.023
icker+0.023
oths+0.022
imming+0.021
umped+0.021
Neg logit contributions
rob-0.022
 l-0.019
 eighty-0.017
 Ske-0.016
 cand-0.016
Token that
Token ID326
Feature activation+1.30e+00

Pos logit contributions
icker+0.033
isel+0.033
oths+0.032
umped+0.031
imming+0.031
Neg logit contributions
rob-0.015
 l-0.012
 Ske-0.010
 eighty-0.010
 cand-0.010
Token sometimes
Token ID3360
Feature activation+8.92e-01

Pos logit contributions
isel+0.070
icker+0.069
oths+0.068
umped+0.066
apa+0.066
Neg logit contributions
rob-0.048
 l-0.039
 eighty-0.035
 Ske-0.034
 cand-0.033
Token struggling
Token ID9648
Feature activation+7.60e-02

Pos logit contributions
isel+0.047
icker+0.047
oths+0.045
imming+0.044
umped+0.044
Neg logit contributions
rob-0.033
 l-0.029
 cand-0.025
 eighty-0.025
 Ske-0.025
Token isn
Token ID2125
Feature activation-5.48e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
umped+0.004
imming+0.004
Neg logit contributions
rob-0.002
 l-0.002
 Ske-0.002
 candy-0.002
 cand-0.002
Token't
Token ID470
Feature activation-1.58e-01

Pos logit contributions
rob+0.022
 l+0.018
 eighty+0.017
 Ske+0.016
 Velvet+0.015
Neg logit contributions
isel-0.028
oths-0.027
icker-0.027
ooks-0.026
apa-0.026
Token always
Token ID1464
Feature activation+9.94e-02

Pos logit contributions
rob+0.006
 l+0.005
 eighty+0.004
 cand+0.004
 Ske+0.004
Neg logit contributions
icker-0.009
isel-0.009
oths-0.008
apa-0.008
imming-0.008
Token a
Token ID257
Feature activation-1.36e-03

Pos logit contributions
icker+0.006
isel+0.005
oths+0.005
apa+0.005
umped+0.005
Neg logit contributions
rob-0.003
 l-0.003
 candy-0.003
 eighty-0.003
 cand-0.003
Token house
Token ID2156
Feature activation-8.15e-03

Pos logit contributions
rob+0.014
 l+0.013
 eighty+0.012
 cand+0.011
 Ske+0.011
Neg logit contributions
isel-0.013
icker-0.013
oths-0.012
umped-0.012
imming-0.012
Token and
Token ID290
Feature activation-2.92e-01

Pos logit contributions
rob+0.000
 l+0.000
 cand+0.000
 eighty+0.000
 Ske+0.000
Neg logit contributions
isel-0.000
icker-0.000
oths-0.000
imming-0.000
apa-0.000
Token added
Token ID2087
Feature activation+6.04e-01

Pos logit contributions
rob+0.014
 l+0.011
 eighty+0.010
 Ske+0.010
 cand+0.009
Neg logit contributions
isel-0.013
oths-0.013
icker-0.013
apa-0.012
umped-0.012
Token them
Token ID606
Feature activation+5.42e-01

Pos logit contributions
isel+0.029
icker+0.029
oths+0.028
umped+0.027
apa+0.027
Neg logit contributions
rob-0.026
 l-0.021
 eighty-0.019
 cand-0.019
 Ske-0.019
Token to
Token ID284
Feature activation-2.74e-01

Pos logit contributions
isel+0.030
oths+0.029
icker+0.028
umped+0.027
ench+0.027
Neg logit contributions
rob-0.019
 l-0.018
 Ske-0.015
 cand-0.015
 eighty-0.015
Token the
Token ID262
Feature activation-2.18e-01

Pos logit contributions
rob+0.012
 l+0.009
 eighty+0.008
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.013
icker-0.013
oths-0.013
apa-0.013
ap-0.012
Token mountain
Token ID8598
Feature activation+1.85e-01

Pos logit contributions
rob+0.010
 l+0.009
 eighty+0.008
 cand+0.008
 Ske+0.008
Neg logit contributions
isel-0.010
oths-0.010
icker-0.010
umped-0.009
imming-0.009
Token.
Token ID13
Feature activation+2.13e-01

Pos logit contributions
isel+0.009
oths+0.009
icker+0.009
imming+0.009
umped+0.009
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.006
 cand-0.006
 Ske-0.006
Token Anna
Token ID11735
Feature activation+1.47e-01

Pos logit contributions
isel+0.010
icker+0.010
oths+0.010
apa+0.010
umped+0.009
Neg logit contributions
rob-0.009
 l-0.007
 eighty-0.007
 cand-0.007
 Ske-0.006
Token saw
Token ID2497
Feature activation+5.37e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
apa+0.007
imming+0.007
Neg logit contributions
rob-0.006
 l-0.005
 eighty-0.005
 Ske-0.005
 cand-0.005
Token what
Token ID644
Feature activation+4.97e-01

Pos logit contributions
isel+0.024
icker+0.024
oths+0.023
apa+0.023
umped+0.022
Neg logit contributions
rob-0.025
 l-0.018
 eighty-0.018
 Ske-0.017
 cand-0.016
Token come
Token ID1282
Feature activation-2.45e-01

Pos logit contributions
isel+0.036
oths+0.035
icker+0.034
imming+0.033
apa+0.033
Neg logit contributions
rob-0.028
 l-0.025
 eighty-0.022
 cand-0.021
 Ske-0.021
Token true
Token ID2081
Feature activation+6.45e-01

Pos logit contributions
 l+0.008
rob+0.008
 eighty+0.007
 scar+0.007
 Star+0.007
Neg logit contributions
isel-0.014
icker-0.013
oths-0.012
itt-0.012
apa-0.012
Token.
Token ID13
Feature activation+3.71e-01

Pos logit contributions
isel+0.034
icker+0.033
oths+0.032
imming+0.031
umped+0.031
Neg logit contributions
rob-0.025
 l-0.022
 Ske-0.019
 cand-0.018
 eighty-0.018
Token<|endoftext|>
Token ID50256
Feature activation+3.71e-01

Pos logit contributions
isel+0.018
icker+0.017
oths+0.017
imming+0.016
apa+0.016
Neg logit contributions
rob-0.016
 l-0.013
 eighty-0.012
 Ske-0.012
 cand-0.012
TokenOnce
Token ID7454
Feature activation-5.88e-01

Pos logit contributions
isel+0.018
icker+0.017
oths+0.017
apa+0.017
umped+0.017
Neg logit contributions
rob-0.016
 l-0.013
 eighty-0.012
 Ske-0.012
 cand-0.012
Token upon
Token ID2402
Feature activation+7.79e-02

Pos logit contributions
rob+0.033
 l+0.027
 eighty+0.026
 cand+0.025
 Ske+0.024
Neg logit contributions
icker-0.022
isel-0.022
oths-0.021
apa-0.021
leys-0.020
Token a
Token ID257
Feature activation+2.30e-01

Pos logit contributions
icker+0.003
isel+0.003
oths+0.003
ench+0.003
umped+0.003
Neg logit contributions
rob-0.004
 l-0.003
 cand-0.003
 eighty-0.003
 Ske-0.003
Token time
Token ID640
Feature activation-1.56e-01

Pos logit contributions
icker+0.008
isel+0.008
oths+0.007
apa+0.007
ap+0.007
Neg logit contributions
rob-0.013
 l-0.011
 eighty-0.011
 cand-0.011
 Ske-0.011
Token there
Token ID612
Feature activation+2.81e-01

Pos logit contributions
rob+0.009
 l+0.007
 eighty+0.006
 cand+0.006
 scar+0.006
Neg logit contributions
oths-0.006
isel-0.006
icker-0.006
apa-0.006
ap-0.005
Token was
Token ID373
Feature activation-2.15e-01

Pos logit contributions
isel+0.011
oths+0.011
icker+0.011
umped+0.011
apa+0.011
Neg logit contributions
rob-0.014
 l-0.011
 eighty-0.011
 cand-0.010
 scar-0.010
Token a
Token ID257
Feature activation-1.02e-01

Pos logit contributions
rob+0.010
 l+0.008
 eighty+0.007
 cand+0.007
 Ske+0.007
Neg logit contributions
icker-0.010
oths-0.009
isel-0.009
umped-0.009
ap-0.009
Token
Token ID198
Feature activation+1.11e-01

Pos logit contributions
rob+0.005
 l+0.004
 eighty+0.004
 cand+0.004
 Ske+0.003
Neg logit contributions
isel-0.004
icker-0.004
oths-0.004
apa-0.004
umped-0.004
Token"
Token ID1
Feature activation-1.25e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
apa+0.004
ench+0.004
Neg logit contributions
rob-0.006
 l-0.005
 eighty-0.005
 Ske-0.005
 Velvet-0.005
TokenCan
Token ID6090
Feature activation-1.72e-01

Pos logit contributions
rob+0.006
 l+0.005
 cand+0.005
 Ske+0.005
 eighty+0.005
Neg logit contributions
isel-0.004
icker-0.004
oths-0.004
aining-0.004
umped-0.004
Token I
Token ID314
Feature activation+1.66e-02

Pos logit contributions
rob+0.009
 l+0.007
 eighty+0.006
 cand+0.006
 scar+0.006
Neg logit contributions
oths-0.007
ap-0.007
icker-0.007
isel-0.007
apa-0.006
Token act
Token ID719
Feature activation+8.33e-02

Pos logit contributions
isel+0.001
oths+0.001
icker+0.001
umped+0.001
apa+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.001
 Ske-0.001
 cand-0.001
Token with
Token ID351
Feature activation+2.56e-01

Pos logit contributions
isel+0.004
icker+0.004
ap+0.004
oths+0.003
aining+0.003
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 cand-0.003
 Ske-0.003
Token you
Token ID345
Feature activation-6.02e-01

Pos logit contributions
isel+0.012
icker+0.012
oths+0.012
apa+0.011
ap+0.011
Neg logit contributions
rob-0.012
 eighty-0.008
 l-0.008
 Ske-0.008
 cand-0.008
Token,
Token ID11
Feature activation+2.99e-01

Pos logit contributions
rob+0.033
 l+0.026
 cand+0.025
 eighty+0.024
 Ske+0.024
Neg logit contributions
icker-0.022
ap-0.022
isel-0.021
oths-0.021
leys-0.020
Token Ben
Token ID3932
Feature activation+6.62e-02

Pos logit contributions
ap+0.012
isel+0.012
icker+0.012
apa+0.012
oths+0.011
Neg logit contributions
rob-0.016
 eighty-0.011
 cand-0.011
 l-0.011
 ninety-0.010
Token?
Token ID30
Feature activation+2.45e-01

Pos logit contributions
isel+0.003
icker+0.003
oths+0.003
ap+0.003
apa+0.003
Neg logit contributions
rob-0.003
 l-0.003
 eighty-0.002
 Ske-0.002
 cand-0.002
Token Can
Token ID1680
Feature activation-2.35e-01

Pos logit contributions
isel+0.010
icker+0.010
oths+0.010
ap+0.010
apa+0.009
Neg logit contributions
rob-0.013
 l-0.010
 cand-0.009
 eighty-0.009
 Ske-0.008
Token when
Token ID618
Feature activation+8.03e-01

Pos logit contributions
rob+0.010
 l+0.008
 eighty+0.008
 cand+0.007
 Velvet+0.007
Neg logit contributions
icker-0.009
isel-0.009
oths-0.009
ap-0.009
aining-0.009
Token it
Token ID340
Feature activation-2.37e-01

Pos logit contributions
isel+0.053
icker+0.052
oths+0.051
umped+0.051
apa+0.051
Neg logit contributions
rob-0.020
 l-0.015
 eighty-0.012
 Ske-0.012
 cand-0.011
Token was
Token ID373
Feature activation-3.11e-01

Pos logit contributions
rob+0.011
 l+0.009
 eighty+0.008
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.011
icker-0.010
oths-0.010
umped-0.010
apa-0.010
Token raining
Token ID43079
Feature activation+2.14e-01

Pos logit contributions
rob+0.013
 l+0.011
 eighty+0.010
 Ske+0.010
 cand+0.010
Neg logit contributions
isel-0.015
oths-0.015
icker-0.015
apa-0.014
umped-0.014
Token outside
Token ID2354
Feature activation-3.30e-01

Pos logit contributions
isel+0.009
icker+0.009
oths+0.008
umped+0.008
imming+0.008
Neg logit contributions
rob-0.011
 l-0.009
 eighty-0.008
 Ske-0.008
 cand-0.008
Token.
Token ID13
Feature activation+3.83e-01

Pos logit contributions
rob+0.011
 l+0.010
 Ske+0.008
 cand+0.008
 candy+0.008
Neg logit contributions
isel-0.019
icker-0.018
oths-0.017
imming-0.017
apa-0.017
Token One
Token ID1881
Feature activation+2.33e-01

Pos logit contributions
isel+0.017
oths+0.016
icker+0.016
apa+0.015
umped+0.015
Neg logit contributions
rob-0.018
 l-0.015
 eighty-0.014
 cand-0.014
 Ske-0.013
Token day
Token ID1110
Feature activation+5.91e-01

Pos logit contributions
isel+0.014
oths+0.013
icker+0.013
umped+0.013
aining+0.013
Neg logit contributions
rob-0.008
 l-0.006
 Ske-0.005
 eighty-0.005
 cand-0.005
Token while
Token ID981
Feature activation+1.45e-01

Pos logit contributions
isel+0.027
icker+0.027
oths+0.027
apa+0.025
umped+0.025
Neg logit contributions
rob-0.027
 l-0.022
 eighty-0.020
 Ske-0.020
 cand-0.020
Token he
Token ID339
Feature activation-1.10e-01

Pos logit contributions
isel+0.009
icker+0.009
oths+0.008
apa+0.008
umped+0.008
Neg logit contributions
rob-0.005
 l-0.003
 eighty-0.003
 Ske-0.003
 cand-0.003
Token was
Token ID373
Feature activation+3.34e-01

Pos logit contributions
rob+0.005
 l+0.004
 eighty+0.004
 cand+0.004
 Ske+0.004
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
umped-0.005
apa-0.005
Token end
Token ID886
Feature activation-7.01e-01

Pos logit contributions
rob+0.009
 l+0.007
 eighty+0.007
 Ske+0.007
 Velvet+0.007
Neg logit contributions
icker-0.005
ap-0.005
imming-0.004
apa-0.004
ench-0.004
Token of
Token ID286
Feature activation+2.87e-01

Pos logit contributions
rob+0.025
 l+0.024
 Ske+0.021
 cand+0.021
 eighty+0.020
Neg logit contributions
isel-0.036
imming-0.035
icker-0.034
oths-0.034
umped-0.033
Token the
Token ID262
Feature activation-4.19e-02

Pos logit contributions
ap+0.009
icker+0.009
isel+0.009
oths+0.009
 forces+0.009
Neg logit contributions
rob-0.015
 eighty-0.013
 cand-0.013
 l-0.012
 ninety-0.012
Token trip
Token ID5296
Feature activation+2.67e-02

Pos logit contributions
rob+0.002
 Velvet+0.002
 l+0.002
 eighty+0.002
 Ske+0.002
Neg logit contributions
ap-0.002
oths-0.001
icker-0.001
appiness-0.001
 forces-0.001
Token,
Token ID11
Feature activation-3.46e-01

Pos logit contributions
isel+0.001
icker+0.001
oths+0.001
imming+0.001
apa+0.001
Neg logit contributions
rob-0.001
 l-0.001
 eighty-0.001
 Ske-0.001
 cand-0.001
Token the
Token ID262
Feature activation+8.94e-02

Pos logit contributions
rob+0.010
 l+0.009
 Ske+0.008
 cand+0.008
 eighty+0.007
Neg logit contributions
isel-0.021
imming-0.020
oths-0.020
icker-0.020
umped-0.020
Token queen
Token ID16599
Feature activation+7.12e-02

Pos logit contributions
isel+0.005
icker+0.005
oths+0.005
umped+0.004
apa+0.004
Neg logit contributions
rob-0.004
 l-0.003
 eighty-0.003
 cand-0.002
 Ske-0.002
Token returned
Token ID4504
Feature activation+2.95e-01

Pos logit contributions
isel+0.004
icker+0.003
oths+0.003
imming+0.003
aining+0.003
Neg logit contributions
rob-0.003
 l-0.003
 eighty-0.002
 cand-0.002
 Ske-0.002
Token home
Token ID1363
Feature activation-5.56e-01

Pos logit contributions
isel+0.016
icker+0.016
oths+0.015
imming+0.015
apa+0.014
Neg logit contributions
rob-0.011
 l-0.010
 cand-0.009
 eighty-0.009
 Ske-0.008
Token with
Token ID351
Feature activation-8.36e-02

Pos logit contributions
rob+0.021
 l+0.019
 eighty+0.016
 cand+0.016
 Ske+0.016
Neg logit contributions
isel-0.030
oths-0.028
icker-0.028
imming-0.027
apa-0.027
Token a
Token ID257
Feature activation-3.19e-01

Pos logit contributions
rob+0.004
 eighty+0.003
 l+0.003
 Ske+0.003
 ninety+0.003
Neg logit contributions
isel-0.004
icker-0.004
apa-0.004
oths-0.004
umped-0.003
Token shield
Token ID7614
Feature activation-7.35e-01

Pos logit contributions
isel+0.012
icker+0.012
oths+0.012
umped+0.012
imming+0.012
Neg logit contributions
rob-0.009
 l-0.008
 eighty-0.007
 Ske-0.007
 scar-0.007
Token has
Token ID468
Feature activation-2.42e-01

Pos logit contributions
rob+0.038
 l+0.030
 eighty+0.029
 Ske+0.029
 Velvet+0.028
Neg logit contributions
isel-0.029
icker-0.027
oths-0.027
umped-0.026
ap-0.026
Token a
Token ID257
Feature activation-6.04e-01

Pos logit contributions
rob+0.011
 l+0.008
 eighty+0.008
 Ske+0.008
 cand+0.007
Neg logit contributions
isel-0.012
oths-0.011
icker-0.011
apa-0.011
umped-0.011
Token star
Token ID3491
Feature activation-4.05e-01

Pos logit contributions
rob+0.024
 l+0.021
 eighty+0.019
 cand+0.018
 Ske+0.018
Neg logit contributions
icker-0.030
isel-0.030
oths-0.030
imming-0.029
umped-0.029
Token on
Token ID319
Feature activation+4.93e-01

Pos logit contributions
rob+0.019
 l+0.015
 eighty+0.014
 Ske+0.014
 cand+0.014
Neg logit contributions
isel-0.018
icker-0.018
oths-0.018
apa-0.017
umped-0.017
Token it
Token ID340
Feature activation-9.32e-01

Pos logit contributions
icker+0.025
isel+0.024
oths+0.024
apa+0.024
umped+0.024
Neg logit contributions
rob-0.019
 l-0.016
 eighty-0.015
 Ske-0.014
 Velvet-0.014
Token.
Token ID13
Feature activation+2.29e-01

Pos logit contributions
rob+0.043
 l+0.034
 eighty+0.032
 Ske+0.031
 cand+0.030
Neg logit contributions
isel-0.042
icker-0.041
oths-0.041
apa-0.039
umped-0.039
Token It
Token ID632
Feature activation+1.69e-01

Pos logit contributions
icker+0.011
isel+0.011
oths+0.010
umped+0.010
imming+0.010
Neg logit contributions
rob-0.010
 l-0.008
 eighty-0.008
 Ske-0.008
 Velvet-0.008
Token is
Token ID318
Feature activation+1.69e-01

Pos logit contributions
isel+0.007
icker+0.007
oths+0.007
apa+0.006
umped+0.006
Neg logit contributions
rob-0.008
 l-0.007
 eighty-0.006
 Ske-0.006
 cand-0.006
Token very
Token ID845
Feature activation+8.71e-02

Pos logit contributions
isel+0.008
icker+0.008
oths+0.008
apa+0.007
umped+0.007
Neg logit contributions
rob-0.007
 l-0.006
 eighty-0.005
 Ske-0.005
 cand-0.005
Token bright
Token ID6016
Feature activation+4.77e-01

Pos logit contributions
isel+0.004
icker+0.004
oths+0.004
umped+0.004
imming+0.004
Neg logit contributions
rob-0.003
 l-0.003
 eighty-0.003
 Ske-0.003
 cand-0.003
Token house
Token ID2156
Feature activation-6.88e-01

Pos logit contributions
rob+0.028
 l+0.023
 eighty+0.021
 Ske+0.021
 cand+0.020
Neg logit contributions
isel-0.026
icker-0.025
oths-0.025
apa-0.024
umped-0.024
Token,"
Token ID553
Feature activation+3.01e-01

Pos logit contributions
rob+0.029
 l+0.023
 eighty+0.021
 Ske+0.020
 cand+0.020
Neg logit contributions
isel-0.034
icker-0.033
oths-0.033
apa-0.032
umped-0.032
Token her
Token ID607
Feature activation+1.61e-01

Pos logit contributions
apa+0.012
ap+0.012
oths+0.012
icker+0.012
isel+0.012
Neg logit contributions
rob-0.015
 eighty-0.011
 ninety-0.010
 Ske-0.010
 carn-0.010
Token mom
Token ID1995
Feature activation-4.88e-01

Pos logit contributions
icker+0.004
oths+0.004
apa+0.004
isel+0.004
ap+0.004
Neg logit contributions
rob-0.011
 eighty-0.008
 l-0.008
 ninety-0.008
 cand-0.008
Tokenmy
Token ID1820
Feature activation-3.20e-01

Pos logit contributions
rob+0.026
 Ske+0.019
 eighty+0.019
 l+0.019
 Velvet+0.018
Neg logit contributions
isel-0.019
ap-0.018
icker-0.018
apa-0.018
leys-0.018
Token said
Token ID531
Feature activation-1.08e+00

Pos logit contributions
rob+0.013
 l+0.009
 eighty+0.009
 Ske+0.008
 cand+0.008
Neg logit contributions
isel-0.017
icker-0.016
apa-0.016
oths-0.016
umped-0.016
Token.
Token ID13
Feature activation-9.84e-01

Pos logit contributions
 l+0.033
rob+0.033
 cand+0.028
 eighty+0.028
 Ske+0.026
Neg logit contributions
isel-0.060
oths-0.058
imming-0.057
icker-0.055
amps-0.054
Token
Token ID198
Feature activation-9.12e-01

Pos logit contributions
rob+0.044
 l+0.036
 eighty+0.033
 Ske+0.033
 cand+0.032
Neg logit contributions
isel-0.046
icker-0.045
oths-0.044
imming-0.043
apa-0.042
Token
Token ID198
Feature activation-7.73e-01

Pos logit contributions
rob+0.045
 l+0.037
 eighty+0.035
 cand+0.034
 Ske+0.034
Neg logit contributions
icker-0.038
isel-0.038
oths-0.038
apa-0.036
umped-0.036
TokenL
Token ID43
Feature activation-8.67e-01

Pos logit contributions
rob+0.026
 l+0.023
 eighty+0.019
 cand+0.019
 Velvet+0.019
Neg logit contributions
isel-0.045
icker-0.042
oths-0.042
apa-0.041
imming-0.040
Tokenily
Token ID813
Feature activation-5.06e-01

Pos logit contributions
rob+0.045
 l+0.037
 eighty+0.034
 Ske+0.033
 scar+0.032
Neg logit contributions
icker-0.035
isel-0.035
oths-0.035
aining-0.033
ooks-0.033
TokenMy
Token ID3666
Feature activation+4.09e-01

Pos logit contributions
isel+0.044
icker+0.042
oths+0.042
apa+0.040
ooks+0.040
Neg logit contributions
rob-0.043
 l-0.035
 eighty-0.033
 Ske-0.032
 cand-0.031
Token friend
Token ID1545
Feature activation-6.42e-01

Pos logit contributions
isel+0.019
icker+0.018
leys+0.018
umped+0.018
oths+0.018
Neg logit contributions
 l-0.017
rob-0.017
 candy-0.015
 eighty-0.014
 Ske-0.014
Token is
Token ID318
Feature activation+4.70e-01

Pos logit contributions
rob+0.026
 l+0.022
 eighty+0.019
 cand+0.019
 Ske+0.019
Neg logit contributions
isel-0.033
icker-0.032
oths-0.032
imming-0.031
umped-0.030
Token in
Token ID287
Feature activation-4.13e-01

Pos logit contributions
isel+0.024
icker+0.023
oths+0.023
umped+0.022
apa+0.022
Neg logit contributions
rob-0.019
 l-0.017
 eighty-0.015
 Ske-0.014
 cand-0.014
Token the
Token ID262
Feature activation-5.57e-01

Pos logit contributions
rob+0.018
 l+0.014
 eighty+0.013
 Ske+0.013
 cand+0.013
Neg logit contributions
isel-0.020
icker-0.019
oths-0.019
apa-0.018
umped-0.018
Token pit
Token ID6028
Feature activation-1.23e+00

Pos logit contributions
rob+0.027
 l+0.026
 candy+0.022
 cand+0.022
 eighty+0.022
Neg logit contributions
isel-0.023
icker-0.022
umped-0.022
oths-0.022
imming-0.022
Token.
Token ID13
Feature activation+1.97e-01

Pos logit contributions
rob+0.035
 l+0.034
 he+0.025
 eighty+0.025
 cand+0.024
Neg logit contributions
isel-0.074
icker-0.069
oths-0.069
imming-0.068
umped-0.068
Token He
Token ID679
Feature activation+1.55e-01

Pos logit contributions
isel+0.010
icker+0.010
oths+0.009
umped+0.009
imming+0.009
Neg logit contributions
rob-0.008
 l-0.006
 Ske-0.006
 eighty-0.006
 Velvet-0.006
Token jumped
Token ID11687
Feature activation-6.76e-02

Pos logit contributions
isel+0.008
icker+0.008
oths+0.008
umped+0.007
aining+0.007
Neg logit contributions
rob-0.006
 l-0.005
 eighty-0.005
 Ske-0.005
 cand-0.005
Token in
Token ID287
Feature activation+2.22e-01

Pos logit contributions
rob+0.003
 l+0.002
 Ske+0.002
 eighty+0.002
 cand+0.002
Neg logit contributions
isel-0.003
icker-0.003
oths-0.003
apa-0.003
umped-0.003
Token and
Token ID290
Feature activation-3.84e-02

Pos logit contributions
icker+0.012
isel+0.012
oths+0.012
apa+0.012
imming+0.012
Neg logit contributions
rob-0.008
 l-0.007
 eighty-0.006
 Ske-0.006
 cand-0.006
Token day
Token ID1110
Feature activation-3.07e-02

Pos logit contributions
isel+0.023
icker+0.022
oths+0.021
umped+0.021
leys+0.021
Neg logit contributions
 l-0.018
rob-0.018
 cand-0.015
 candy-0.015
 eighty-0.015
Token better
Token ID1365
Feature activation+1.22e+00

Pos logit contributions
rob+0.001
 l+0.001
 eighty+0.001
 candy+0.001
 cand+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
umped-0.002
apa-0.002
Token.
Token ID13
Feature activation+6.00e-02

Pos logit contributions
isel+0.066
icker+0.064
oths+0.062
apa+0.061
imming+0.061
Neg logit contributions
rob-0.043
 l-0.037
 eighty-0.032
 Ske-0.031
 cand-0.031
Token From
Token ID3574
Feature activation-4.87e-02

Pos logit contributions
isel+0.003
icker+0.003
oths+0.003
umped+0.003
imming+0.003
Neg logit contributions
rob-0.003
 l-0.002
 eighty-0.002
 Ske-0.002
 cand-0.002
Token that
Token ID326
Feature activation+3.17e-01

Pos logit contributions
rob+0.002
 l+0.001
 eighty+0.001
 cand+0.001
 Ske+0.001
Neg logit contributions
isel-0.002
icker-0.002
oths-0.002
apa-0.002
imming-0.002
Token day
Token ID1110
Feature activation-1.05e+00

Pos logit contributions
isel+0.017
icker+0.016
oths+0.016
apa+0.016
umped+0.016
Neg logit contributions
rob-0.012
 l-0.010
 eighty-0.009
 Ske-0.008
 cand-0.008
Token on
Token ID319
Feature activation-2.37e-01

Pos logit contributions
rob+0.029
 l+0.023
 eighty+0.021
 Ske+0.020
 Velvet+0.020
Neg logit contributions
icker-0.064
isel-0.064
ooks-0.062
oths-0.062
ench-0.061
Token,
Token ID11
Feature activation-8.09e-01

Pos logit contributions
rob+0.009
 l+0.008
 Ske+0.007
 eighty+0.007
 cand+0.007
Neg logit contributions
isel-0.012
icker-0.012
oths-0.012
imming-0.011
umped-0.011
Token Lily
Token ID20037
Feature activation-3.88e-01

Pos logit contributions
rob+0.027
 l+0.019
 eighty+0.017
 Ske+0.016
 cand+0.016
Neg logit contributions
isel-0.047
icker-0.046
oths-0.046
apa-0.045
umped-0.045
Token always
Token ID1464
Feature activation-7.28e-01

Pos logit contributions
rob+0.016
 l+0.014
 eighty+0.012
 Ske+0.012
 cand+0.011
Neg logit contributions
isel-0.020
icker-0.019
oths-0.019
apa-0.018
imming-0.018
Token remembered
Token ID12086
Feature activation-7.91e-01

Pos logit contributions
rob+0.033
 l+0.028
 eighty+0.026
 Ske+0.024
 cand+0.024
Neg logit contributions
isel-0.033
icker-0.033
oths-0.033
apa-0.031
imming-0.031
Token one
Token ID530
Feature activation-5.78e-01

Pos logit contributions
 l+0.006
rob+0.005
 he+0.005
 Ske+0.004
 candy+0.004
Neg logit contributions
isel-0.017
imming-0.017
umped-0.017
aining-0.017
amps-0.017
Token of
Token ID286
Feature activation+4.42e-01

Pos logit contributions
rob+0.024
 l+0.020
 eighty+0.018
 Ske+0.018
 cand+0.017
Neg logit contributions
isel-0.028
oths-0.028
icker-0.028
umped-0.027
imming-0.027
Token the
Token ID262
Feature activation+2.41e-01

Pos logit contributions
isel+0.025
icker+0.025
oths+0.024
ossom+0.024
ench+0.024
Neg logit contributions
rob-0.015
 l-0.012
 Ske-0.010
 eighty-0.010
 Velvet-0.010
Token children
Token ID1751
Feature activation+4.95e-01

Pos logit contributions
isel+0.013
icker+0.012
oths+0.012
aining+0.012
umped+0.012
Neg logit contributions
rob-0.010
 l-0.008
 cand-0.007
 eighty-0.007
 Ske-0.007
Token slipped
Token ID18859
Feature activation-2.23e-01

Pos logit contributions
isel+0.029
aining+0.028
icker+0.028
ooks+0.027
appiness+0.027
Neg logit contributions
 l-0.014
rob-0.014
 he-0.011
 cand-0.011
 Ske-0.011
Token and
Token ID290
Feature activation-9.20e-01

Pos logit contributions
rob+0.009
 l+0.008
 eighty+0.007
 Ske+0.007
 Velvet+0.007
Neg logit contributions
isel-0.011
icker-0.011
oths-0.011
apa-0.011
imming-0.010
Token fell
Token ID3214
Feature activation-1.06e-01

Pos logit contributions
rob+0.046
 l+0.037
 eighty+0.035
 cand+0.034
 Ske+0.034
Neg logit contributions
isel-0.039
icker-0.038
oths-0.037
apa-0.036
umped-0.036
Token.
Token ID13
Feature activation+2.08e-01

Pos logit contributions
rob+0.005
 l+0.004
 eighty+0.004
 Ske+0.004
 cand+0.004
Neg logit contributions
isel-0.005
icker-0.005
oths-0.005
apa-0.004
imming-0.004
Token The
Token ID383
Feature activation+4.73e-01

Pos logit contributions
isel+0.011
imming+0.010
leys+0.010
icker+0.010
wing+0.010
Neg logit contributions
 l-0.007
 Ske-0.006
rob-0.006
 Star-0.006
 cand-0.006
Token others
Token ID1854
Feature activation+1.46e-01

Pos logit contributions
isel+0.027
aining+0.025
icker+0.025
oths+0.025
ooks+0.025
Neg logit contributions
rob-0.017
 l-0.015
 cand-0.013
 eighty-0.012
 candy-0.012
Token ran
Token ID4966
Feature activation-1.98e-01

Pos logit contributions
isel+0.008
imming+0.008
oths+0.008
ooks+0.008
apa+0.008
Neg logit contributions
rob-0.005
 l-0.005
 cand-0.003
 Ske-0.003
 he-0.003