TOP ACTIVATIONS
MAX = 0.059

is not the only red cr ay on . There are
, scissors , glue and cr ay ons . They make
to draw pictures with her cr ay ons , but she
draw and color with her cr ay ons . One day
draw and color with her cr ay ons . One day
wrote and wrote on the soft wood . Now she could
a special color . Purple is a fun color . I
. She picks a red cr ay on and starts to
M oral of the story : Regular practice makes perfect
can draw on paper with cr ay ons and stickers .
a different color . Purple is a special color . Purple
M oral of the story : If you are sorry
She is using the red cr ay on . He says
liked to draw with his cr ay ons and markers .
. Tom has a red cr ay on and Lily has
with his drum . It was big and heavy , but
M oral of the story : Sharing is caring !
his sc ooter . It was red and had two wheels
ay ons . She writes words and draws pictures on paper
in her bedroom . It was a very messy place ,

MIN ACTIVATIONS
MIN = -0.065

Tom felt angry and sad . He tried to
burned his hand and made him cry . His mom heard
so disappointed and he never got the special gift from the
. Tim and Sam were both hurt and scared .
What â s wrong , Jane ?
The two became insepar able and each day they
scissors slipped , giving the boy a bad cut on the
got hurt . You made us worry ." "
't leave our homes for anyone , not even a mighty
it , so Jill got very mad . Her
then Lily 's grandma got very tired and fell asleep .
so sad that she never got her skirt back . Lily
the toy away and made him go back to the store
day , the puppy got very sick and had to go
Unfortunately , the infant never got to see his friend again
and gave back the shiny things . They hugged
me ! Aren â t we both so excited
. It hurt and made him sad . D ucky started
but Buddy just loved the mud too much .
she decided to give the boy one of her own .

INTERVAL 0.028 - 0.059
CONTAINS 0.811%

!" " This is the best gift ever !"
From that day on , Lily loved playing in the empty
, the teacher asked Lily 's mom to send a note
. From then on , Lily always brushed Max softly and
in the jungle . One day , he heard a

INTERVAL -0.003 - 0.028
CONTAINS 60.793%

When it started getting dark , Mild and her
with hair . The bunny and the k
but warned her that the toy was very special and she
back to the house , but she slipped and fell .
magic wand again !" They went to dry themselves and find

INTERVAL -0.034 - -0.003
CONTAINS 38.153%

But then something happened that changed the race . The
not want to play with the baby sheep . They wanted
to explore . Tom started to explore the camp , discovering
sour candy to share . Tim and Jake both loved the
. They are glad they have float ies . They are

INTERVAL -0.065 - -0.034
CONTAINS 0.242%

hairy and bent . Lily got upset and told Max he
for a while , he started to get tired . He
a big rock . Jane got scared and started to cry
were not ashamed anymore . <|endoftext|> L ily and Tom were
girl was scared and began to back away slowly . She
Token is
Token ID318
Feature activation+3.49e-02

Pos logit contributions
ine+0.022
 birthday+0.021
hip+0.020
 Office+0.019
 Jupiter+0.019
Neg logit contributions
ls-0.025
Run-0.025
mun-0.024
P-0.022
Hey-0.022
Token not
Token ID407
Feature activation+3.41e-02

Pos logit contributions
hip+0.061
ine+0.061
 Office+0.055
rap+0.055
 birthday+0.054
Neg logit contributions
ls-0.053
Run-0.051
 bitten-0.047
P-0.047
mun-0.045
Token the
Token ID262
Feature activation+2.53e-03

Pos logit contributions
ine+0.057
hip+0.053
 Office+0.051
 birthday+0.051
rap+0.051
Neg logit contributions
ls-0.054
Run-0.052
 bitten-0.049
mun-0.047
P-0.047
Token only
Token ID691
Feature activation+2.19e-02

Pos logit contributions
ine+0.004
 birthday+0.003
hip+0.003
 Jupiter+0.003
 Office+0.003
Neg logit contributions
Run-0.005
ls-0.005
mun-0.004
P-0.004
Hey-0.004
Token red
Token ID2266
Feature activation+1.96e-02

Pos logit contributions
ine+0.031
hip+0.029
rap+0.028
 birthday+0.028
 Jupiter+0.028
Neg logit contributions
ls-0.040
Run-0.039
mun-0.037
 bitten-0.035
Hey-0.034
Token cr
Token ID1067
Feature activation+5.92e-02

Pos logit contributions
ine+0.030
 Jupiter+0.029
 Office+0.028
 birthday+0.028
hip+0.027
Neg logit contributions
ls-0.033
Run-0.032
mun-0.030
Hey-0.028
 bitten-0.027
Tokenay
Token ID323
Feature activation+2.54e-02

Pos logit contributions
 birthday+0.074
 Office+0.074
sing+0.071
 experiment+0.069
 Jupiter+0.068
Neg logit contributions
ls-0.111
P-0.100
awed-0.092
led-0.091
Hey-0.090
Tokenon
Token ID261
Feature activation+1.48e-02

Pos logit contributions
ine+0.054
 birthday+0.054
sing+0.052
 Office+0.052
 Jupiter+0.051
Neg logit contributions
ls-0.029
Run-0.024
mun-0.023
led-0.021
P-0.021
Token.
Token ID13
Feature activation+2.24e-02

Pos logit contributions
ine+0.029
hip+0.028
 Jupiter+0.027
 Office+0.027
 birthday+0.027
Neg logit contributions
ls-0.020
Run-0.015
led-0.015
mun-0.015
 bitten-0.014
Token There
Token ID1318
Feature activation+9.96e-03

Pos logit contributions
ine+0.036
 birthday+0.031
rap+0.031
 desk+0.030
 end+0.030
Neg logit contributions
Run-0.039
ls-0.034
Hey-0.034
P-0.033
Sorry-0.032
Token are
Token ID389
Feature activation+2.21e-02

Pos logit contributions
ine+0.017
 birthday+0.016
hip+0.015
 Office+0.015
rap+0.015
Neg logit contributions
ls-0.017
Run-0.016
mun-0.016
Hey-0.014
P-0.014
Token,
Token ID11
Feature activation+1.74e-02

Pos logit contributions
hip+0.035
ine+0.035
 Office+0.034
 birthday+0.033
uries+0.033
Neg logit contributions
ls-0.035
led-0.026
 bitten-0.025
Run-0.024
mun-0.024
Token scissors
Token ID41786
Feature activation+2.47e-02

Pos logit contributions
ine+0.027
hip+0.027
 Office+0.026
rap+0.026
 birthday+0.025
Neg logit contributions
ls-0.030
Run-0.029
mun-0.027
 bitten-0.025
P-0.025
Token,
Token ID11
Feature activation+2.25e-02

Pos logit contributions
 birthday+0.044
 Office+0.043
hip+0.041
 Jupiter+0.041
ine+0.040
Neg logit contributions
ls-0.039
 bitten-0.032
led-0.031
mun-0.031
Run-0.031
Token glue
Token ID22749
Feature activation+2.69e-02

Pos logit contributions
ine+0.036
hip+0.035
 Office+0.035
 birthday+0.034
rap+0.033
Neg logit contributions
ls-0.038
Run-0.036
mun-0.033
P-0.033
 bitten-0.032
Token and
Token ID290
Feature activation+3.02e-02

Pos logit contributions
 birthday+0.045
hip+0.044
 Office+0.044
ine+0.043
 Jupiter+0.043
Neg logit contributions
ls-0.045
mun-0.036
Run-0.035
led-0.035
 bitten-0.035
Token cr
Token ID1067
Feature activation+5.84e-02

Pos logit contributions
ine+0.049
hip+0.046
 birthday+0.045
 Office+0.045
rap+0.044
Neg logit contributions
ls-0.051
Run-0.050
mun-0.047
P-0.045
 bitten-0.045
Tokenay
Token ID323
Feature activation+2.83e-02

Pos logit contributions
 Office+0.077
 birthday+0.074
 experiment+0.072
sing+0.072
 reminder+0.071
Neg logit contributions
ls-0.117
P-0.100
led-0.099
awed-0.097
 bitten-0.095
Tokenons
Token ID684
Feature activation+2.11e-02

Pos logit contributions
 birthday+0.073
ine+0.071
 Office+0.071
hip+0.070
 Jupiter+0.069
Neg logit contributions
ls-0.022
mun-0.016
Run-0.016
P-0.015
led-0.015
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
ine+0.041
 birthday+0.040
 Office+0.040
 Jupiter+0.039
hip+0.039
Neg logit contributions
ls-0.029
Run-0.023
mun-0.022
led-0.022
 bitten-0.022
Token They
Token ID1119
Feature activation+2.45e-02

Pos logit contributions
ine+0.025
 birthday+0.022
hip+0.022
urger+0.022
 Office+0.022
Neg logit contributions
Run-0.020
ls-0.019
mun-0.018
Hey-0.017
P-0.017
Token make
Token ID787
Feature activation+3.33e-02

Pos logit contributions
ine+0.034
 birthday+0.032
hip+0.032
 Office+0.031
rap+0.030
Neg logit contributions
ls-0.045
Run-0.043
mun-0.043
 distracted-0.040
P-0.040
Token to
Token ID284
Feature activation+1.06e-02

Pos logit contributions
ls+0.001
Run+0.001
mun+0.001
P+0.001
 bitten+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
hip-0.001
 Office-0.001
 X-0.001
Token draw
Token ID3197
Feature activation+2.02e-02

Pos logit contributions
ine+0.014
 Office+0.014
 birthday+0.013
 Jupiter+0.013
hip+0.013
Neg logit contributions
Run-0.019
ls-0.018
mun-0.018
P-0.017
 bitten-0.017
Token pictures
Token ID5986
Feature activation+3.83e-02

Pos logit contributions
ine+0.038
sing+0.037
hip+0.036
 Office+0.036
uries+0.035
Neg logit contributions
ls-0.032
mun-0.026
led-0.025
P-0.024
Run-0.024
Token with
Token ID351
Feature activation+2.01e-02

Pos logit contributions
ine+0.080
sing+0.077
 Office+0.077
urger+0.076
artments+0.074
Neg logit contributions
ls-0.049
mun-0.039
Run-0.035
P-0.033
led-0.032
Token her
Token ID607
Feature activation+2.17e-02

Pos logit contributions
ine+0.034
 Office+0.033
hip+0.033
 birthday+0.032
sing+0.031
Neg logit contributions
ls-0.033
mun-0.029
Run-0.028
P-0.027
 bitten-0.026
Token cr
Token ID1067
Feature activation+5.64e-02

Pos logit contributions
ine+0.034
hip+0.033
 Office+0.031
rap+0.031
 birthday+0.030
Neg logit contributions
ls-0.038
mun-0.034
Run-0.033
P-0.031
 bitten-0.030
Tokenay
Token ID323
Feature activation+2.87e-02

Pos logit contributions
 Office+0.073
 birthday+0.067
sing+0.065
 experiment+0.065
 believe+0.064
Neg logit contributions
ls-0.110
P-0.099
awed-0.095
led-0.091
 bitten-0.088
Tokenons
Token ID684
Feature activation+1.25e-02

Pos logit contributions
 birthday+0.076
 Office+0.074
ine+0.074
sing+0.073
hip+0.072
Neg logit contributions
ls-0.019
mun-0.012
P-0.011
led-0.010
Run-0.009
Token,
Token ID11
Feature activation+1.51e-02

Pos logit contributions
ine+0.023
 Office+0.022
 birthday+0.021
 X+0.021
 Jupiter+0.021
Neg logit contributions
ls-0.018
mun-0.015
Run-0.014
P-0.014
led-0.014
Token but
Token ID475
Feature activation+1.83e-02

Pos logit contributions
ine+0.024
 Office+0.023
hip+0.022
rap+0.022
 birthday+0.021
Neg logit contributions
ls-0.025
Run-0.025
mun-0.025
Hey-0.021
P-0.021
Token she
Token ID673
Feature activation+1.06e-02

Pos logit contributions
ine+0.038
hip+0.037
 Office+0.036
 birthday+0.036
rap+0.035
Neg logit contributions
ls-0.021
Run-0.020
mun-0.020
 distracted-0.017
P-0.016
Token draw
Token ID3197
Feature activation+2.46e-02

Pos logit contributions
ine+0.022
 Office+0.022
uries+0.021
rap+0.021
hip+0.021
Neg logit contributions
Run-0.021
ls-0.020
P-0.020
mun-0.020
 bitten-0.019
Token and
Token ID290
Feature activation+2.29e-02

Pos logit contributions
ine+0.047
sing+0.046
 Office+0.044
uries+0.044
hip+0.044
Neg logit contributions
ls-0.039
mun-0.031
P-0.031
led-0.030
 distracted-0.027
Token color
Token ID3124
Feature activation+3.03e-02

Pos logit contributions
ine+0.030
 Office+0.027
hip+0.026
urger+0.025
uries+0.025
Neg logit contributions
mun-0.044
ls-0.044
Run-0.042
 av-0.040
P-0.039
Token with
Token ID351
Feature activation+2.13e-02

Pos logit contributions
sing+0.068
uries+0.065
 Office+0.065
ine+0.064
hip+0.063
Neg logit contributions
ls-0.042
mun-0.028
ries-0.027
 av-0.026
P-0.025
Token her
Token ID607
Feature activation+1.74e-02

Pos logit contributions
 Office+0.044
ine+0.042
sing+0.042
hip+0.042
uries+0.041
Neg logit contributions
ls-0.027
P-0.023
mun-0.022
 av-0.021
 bitten-0.021
Token cr
Token ID1067
Feature activation+5.62e-02

Pos logit contributions
ine+0.030
hip+0.029
 Office+0.028
rap+0.027
uries+0.027
Neg logit contributions
ls-0.029
mun-0.025
P-0.024
Run-0.024
 bitten-0.023
Tokenay
Token ID323
Feature activation+2.81e-02

Pos logit contributions
 Office+0.072
sing+0.065
 birthday+0.064
 experiment+0.062
 believe+0.062
Neg logit contributions
ls-0.112
P-0.103
awed-0.097
led-0.094
 bitten-0.091
Tokenons
Token ID684
Feature activation+1.14e-02

Pos logit contributions
 birthday+0.075
 Office+0.074
ine+0.073
hip+0.073
sing+0.072
Neg logit contributions
ls-0.019
P-0.013
mun-0.012
led-0.010
Run-0.010
Token.
Token ID13
Feature activation+2.12e-02

Pos logit contributions
ine+0.023
 Office+0.021
 birthday+0.021
 X+0.021
hip+0.021
Neg logit contributions
ls-0.015
mun-0.013
Run-0.012
P-0.012
led-0.012
Token One
Token ID1881
Feature activation+3.28e-02

Pos logit contributions
ine+0.035
urger+0.032
rap+0.031
 Office+0.031
 believe+0.031
Neg logit contributions
mun-0.034
Run-0.032
ls-0.032
Hey-0.030
P-0.030
Token day
Token ID1110
Feature activation-6.20e-05

Pos logit contributions
uries+0.062
artments+0.061
hip+0.059
urger+0.059
rap+0.057
Neg logit contributions
P-0.044
ls-0.043
Run-0.038
mun-0.037
led-0.035
Token draw
Token ID3197
Feature activation+2.46e-02

Pos logit contributions
ine+0.022
 Office+0.022
uries+0.021
rap+0.021
hip+0.021
Neg logit contributions
Run-0.021
ls-0.020
P-0.020
mun-0.020
 bitten-0.019
Token and
Token ID290
Feature activation+2.29e-02

Pos logit contributions
ine+0.047
sing+0.046
 Office+0.044
uries+0.044
hip+0.044
Neg logit contributions
ls-0.039
mun-0.031
P-0.031
led-0.030
 distracted-0.027
Token color
Token ID3124
Feature activation+3.03e-02

Pos logit contributions
ine+0.030
 Office+0.027
hip+0.026
urger+0.025
uries+0.025
Neg logit contributions
mun-0.044
ls-0.044
Run-0.042
 av-0.040
P-0.039
Token with
Token ID351
Feature activation+2.13e-02

Pos logit contributions
sing+0.068
uries+0.065
 Office+0.065
ine+0.064
hip+0.063
Neg logit contributions
ls-0.042
mun-0.028
ries-0.027
 av-0.026
P-0.025
Token her
Token ID607
Feature activation+1.74e-02

Pos logit contributions
 Office+0.044
ine+0.042
sing+0.042
hip+0.042
uries+0.041
Neg logit contributions
ls-0.027
P-0.023
mun-0.022
 av-0.021
 bitten-0.021
Token cr
Token ID1067
Feature activation+5.62e-02

Pos logit contributions
ine+0.030
hip+0.029
 Office+0.028
rap+0.027
uries+0.027
Neg logit contributions
ls-0.029
mun-0.025
P-0.024
Run-0.024
 bitten-0.023
Tokenay
Token ID323
Feature activation+2.81e-02

Pos logit contributions
 Office+0.072
sing+0.065
 birthday+0.064
 experiment+0.062
 believe+0.062
Neg logit contributions
ls-0.112
P-0.103
awed-0.097
led-0.094
 bitten-0.091
Tokenons
Token ID684
Feature activation+1.14e-02

Pos logit contributions
 birthday+0.075
 Office+0.074
ine+0.073
hip+0.073
sing+0.072
Neg logit contributions
ls-0.019
P-0.013
mun-0.012
led-0.010
Run-0.010
Token.
Token ID13
Feature activation+2.12e-02

Pos logit contributions
ine+0.023
 Office+0.021
 birthday+0.021
 X+0.021
hip+0.021
Neg logit contributions
ls-0.015
mun-0.013
Run-0.012
P-0.012
led-0.012
Token One
Token ID1881
Feature activation+3.28e-02

Pos logit contributions
ine+0.035
urger+0.032
rap+0.031
 Office+0.031
 believe+0.031
Neg logit contributions
mun-0.034
Run-0.032
ls-0.032
Hey-0.030
P-0.030
Token day
Token ID1110
Feature activation-6.20e-05

Pos logit contributions
uries+0.062
artments+0.061
hip+0.059
urger+0.059
rap+0.057
Neg logit contributions
P-0.044
ls-0.043
Run-0.038
mun-0.037
led-0.035
Token wrote
Token ID2630
Feature activation+2.78e-02

Pos logit contributions
ine+0.026
 birthday+0.025
hip+0.024
 Office+0.023
 Jupiter+0.023
Neg logit contributions
ls-0.018
Run-0.018
mun-0.017
P-0.015
 bitten-0.015
Token and
Token ID290
Feature activation+7.16e-03

Pos logit contributions
ine+0.054
sing+0.050
 birthday+0.050
uries+0.049
 Office+0.049
Neg logit contributions
ls-0.040
led-0.035
 distracted-0.033
mun-0.031
Run-0.031
Token wrote
Token ID2630
Feature activation+3.76e-02

Pos logit contributions
ine+0.012
 birthday+0.011
hip+0.010
 Office+0.010
 Jupiter+0.010
Neg logit contributions
Run-0.012
ls-0.012
mun-0.012
P-0.011
 bitten-0.011
Token on
Token ID319
Feature activation+2.43e-02

Pos logit contributions
ine+0.068
sing+0.067
 birthday+0.063
uries+0.062
 Office+0.062
Neg logit contributions
ls-0.055
led-0.049
mun-0.045
 distracted-0.044
Hey-0.042
Token the
Token ID262
Feature activation+3.33e-02

Pos logit contributions
 birthday+0.048
ine+0.048
hip+0.046
sing+0.046
 Office+0.046
Neg logit contributions
ls-0.030
mun-0.026
Run-0.025
Hey-0.022
 distracted-0.022
Token soft
Token ID2705
Feature activation+5.59e-02

Pos logit contributions
ine+0.056
uries+0.048
 birthday+0.048
hip+0.048
rap+0.047
Neg logit contributions
 av-0.047
ls-0.047
Run-0.046
 bitten-0.045
P-0.044
Token wood
Token ID4898
Feature activation+2.38e-02

Pos logit contributions
uries+0.089
 birthday+0.082
 anniversary+0.082
icity+0.080
sing+0.079
Neg logit contributions
 av-0.072
ls-0.070
 st-0.062
led-0.060
 bitten-0.060
Token.
Token ID13
Feature activation+1.74e-02

Pos logit contributions
ine+0.041
 birthday+0.039
rap+0.037
 Office+0.037
hip+0.037
Neg logit contributions
ls-0.036
mun-0.032
Run-0.032
led-0.029
 bitten-0.028
Token Now
Token ID2735
Feature activation+1.45e-02

Pos logit contributions
ine+0.033
urger+0.030
 birthday+0.030
 Office+0.029
hip+0.029
Neg logit contributions
Run-0.023
ls-0.022
mun-0.022
Hey-0.020
P-0.019
Token she
Token ID673
Feature activation+1.84e-02

Pos logit contributions
ine+0.031
sing+0.030
hip+0.030
 Office+0.029
 birthday+0.029
Neg logit contributions
ls-0.017
Run-0.015
mun-0.014
 distracted-0.013
Hey-0.012
Token could
Token ID714
Feature activation+5.60e-03

Pos logit contributions
ine+0.034
 birthday+0.033
hip+0.032
 Office+0.031
 Jupiter+0.030
Neg logit contributions
ls-0.027
mun-0.025
Run-0.025
led-0.022
 bitten-0.022
Token a
Token ID257
Feature activation+2.13e-02

Pos logit contributions
hip+0.101
 basis+0.097
 Office+0.091
sing+0.090
jo+0.090
Neg logit contributions
ls-0.067
 bitten-0.061
 st-0.056
P-0.056
 distracted-0.056
Token special
Token ID2041
Feature activation+2.64e-02

Pos logit contributions
hip+0.034
ine+0.033
rap+0.032
urg+0.029
ika+0.029
Neg logit contributions
P-0.033
ls-0.032
Run-0.031
mun-0.029
 st-0.029
Token color
Token ID3124
Feature activation+2.50e-02

Pos logit contributions
hip+0.037
ine+0.035
rap+0.035
 Jupiter+0.035
 Him+0.033
Neg logit contributions
ls-0.043
Run-0.040
mun-0.039
P-0.039
 bitten-0.037
Token.
Token ID13
Feature activation+2.59e-02

Pos logit contributions
sing+0.054
ine+0.053
 Office+0.053
 Jupiter+0.053
artments+0.052
Neg logit contributions
ls-0.025
P-0.016
guard-0.014
mun-0.013
Run-0.013
Token Purple
Token ID17265
Feature activation+2.24e-02

Pos logit contributions
ine+0.047
 birthday+0.043
 desk+0.043
sing+0.042
urger+0.042
Neg logit contributions
Run-0.037
Hey-0.034
mun-0.033
P-0.032
ls-0.032
Token is
Token ID318
Feature activation+5.57e-02

Pos logit contributions
ine+0.036
sing+0.036
 birthday+0.035
hip+0.035
 X+0.035
Neg logit contributions
ls-0.037
mun-0.032
Run-0.031
P-0.030
Hey-0.029
Token a
Token ID257
Feature activation+2.10e-02

Pos logit contributions
hip+0.102
 basis+0.098
 Office+0.092
otive+0.091
sing+0.091
Neg logit contributions
ls-0.069
 bitten-0.062
P-0.058
 st-0.058
 distracted-0.057
Token fun
Token ID1257
Feature activation+2.08e-02

Pos logit contributions
hip+0.033
ine+0.033
rap+0.031
artments+0.029
urg+0.029
Neg logit contributions
P-0.033
ls-0.031
Run-0.030
mun-0.029
 st-0.028
Token color
Token ID3124
Feature activation+2.34e-02

Pos logit contributions
hip+0.027
 Jupiter+0.027
ine+0.027
sing+0.026
 Him+0.026
Neg logit contributions
ls-0.033
P-0.032
Run-0.032
mun-0.030
Hey-0.029
Token.
Token ID13
Feature activation+2.77e-02

Pos logit contributions
ine+0.049
sing+0.049
 Office+0.049
 Jupiter+0.048
artments+0.048
Neg logit contributions
ls-0.024
P-0.016
mun-0.014
guard-0.014
Run-0.014
Token I
Token ID314
Feature activation+2.81e-02

Pos logit contributions
ine+0.050
 birthday+0.047
 desk+0.047
urger+0.045
rap+0.045
Neg logit contributions
Run-0.040
Hey-0.036
mun-0.036
P-0.035
ls-0.034
Token.
Token ID13
Feature activation-3.54e-03

Pos logit contributions
ine+0.052
sing+0.051
urger+0.050
hip+0.049
 Office+0.049
Neg logit contributions
ls-0.029
Run-0.021
Hey-0.019
 av-0.018
mun-0.017
Token She
Token ID1375
Feature activation+1.43e-02

Pos logit contributions
Run+0.006
ls+0.006
mun+0.006
P+0.005
Hey+0.005
Neg logit contributions
ine-0.006
 birthday-0.006
hip-0.005
 Office-0.005
 X-0.005
Token picks
Token ID11103
Feature activation+9.16e-03

Pos logit contributions
ine+0.023
 birthday+0.022
hip+0.021
 Office+0.020
 Jupiter+0.020
Neg logit contributions
ls-0.025
Run-0.024
mun-0.023
P-0.021
Hey-0.021
Token a
Token ID257
Feature activation+2.74e-02

Pos logit contributions
ine+0.020
 birthday+0.019
hip+0.019
sing+0.019
uries+0.019
Neg logit contributions
ls-0.012
Run-0.009
 distracted-0.009
led-0.008
Hey-0.008
Token red
Token ID2266
Feature activation+2.69e-02

Pos logit contributions
ine+0.040
hip+0.040
uries+0.039
 Office+0.036
artments+0.035
Neg logit contributions
ls-0.052
Run-0.043
P-0.041
mun-0.039
 bitten-0.039
Token cr
Token ID1067
Feature activation+5.55e-02

Pos logit contributions
uries+0.043
ine+0.040
 Office+0.037
artments+0.036
hip+0.036
Neg logit contributions
ls-0.045
Run-0.041
Hey-0.039
P-0.038
 av-0.035
Tokenay
Token ID323
Feature activation+2.34e-02

Pos logit contributions
 birthday+0.084
 Office+0.081
sing+0.080
 appearance+0.080
 experiment+0.077
Neg logit contributions
ls-0.090
P-0.085
awed-0.077
Hey-0.074
umbled-0.073
Tokenon
Token ID261
Feature activation+1.27e-02

Pos logit contributions
 birthday+0.055
uries+0.054
sing+0.053
 appearance+0.052
ine+0.052
Neg logit contributions
ls-0.024
Run-0.018
P-0.017
Hey-0.015
led-0.015
Token and
Token ID290
Feature activation+1.53e-02

Pos logit contributions
ine+0.024
 Office+0.024
uries+0.023
 perfection+0.023
hip+0.022
Neg logit contributions
ls-0.018
Run-0.013
led-0.013
 bitten-0.013
 distracted-0.012
Token starts
Token ID4940
Feature activation+1.58e-02

Pos logit contributions
ine+0.024
 birthday+0.023
hip+0.022
 Office+0.022
uries+0.021
Neg logit contributions
ls-0.027
Run-0.027
mun-0.025
P-0.023
 bitten-0.023
Token to
Token ID284
Feature activation+1.15e-02

Pos logit contributions
ine+0.033
hip+0.032
 Office+0.032
uries+0.032
sing+0.031
Neg logit contributions
ls-0.017
 bitten-0.015
 distracted-0.014
Run-0.014
mun-0.014
Token
Token ID198
Feature activation+1.26e-02

Pos logit contributions
ine+0.034
 birthday+0.033
 Office+0.031
 X+0.031
hip+0.031
Neg logit contributions
ls-0.024
Run-0.024
mun-0.023
P-0.021
Hey-0.020
TokenM
Token ID44
Feature activation+1.63e-02

Pos logit contributions
ine+0.023
 birthday+0.022
 X+0.020
 Office+0.020
 Jupiter+0.020
Neg logit contributions
Run-0.019
ls-0.019
mun-0.018
P-0.017
Hey-0.017
Tokenoral
Token ID6864
Feature activation+2.12e-02

Pos logit contributions
ine+0.026
 birthday+0.024
 Office+0.023
hip+0.023
 X+0.023
Neg logit contributions
ls-0.029
Run-0.028
mun-0.027
P-0.025
Hey-0.024
Token of
Token ID286
Feature activation+4.37e-02

Pos logit contributions
ine+0.037
 birthday+0.035
 X+0.032
hip+0.032
 Office+0.032
Neg logit contributions
ls-0.034
Run-0.033
mun-0.031
P-0.030
Hey-0.029
Token the
Token ID262
Feature activation+1.22e-02

Pos logit contributions
uries+0.099
ine+0.099
artments+0.098
hip+0.092
sing+0.092
Neg logit contributions
mun-0.038
Run-0.036
ls-0.035
P-0.034
Hey-0.030
Token story
Token ID1621
Feature activation+5.54e-02

Pos logit contributions
ine+0.014
hip+0.012
artments+0.012
 birthday+0.012
 Office+0.011
Neg logit contributions
ls-0.027
Run-0.027
mun-0.026
P-0.025
Hey-0.024
Token:
Token ID25
Feature activation+3.49e-02

Pos logit contributions
ine+0.069
 birthday+0.062
 X+0.060
urger+0.060
sing+0.060
Neg logit contributions
ls-0.112
Run-0.101
mun-0.099
Hey-0.094
led-0.093
Token Regular
Token ID23603
Feature activation+1.60e-02

Pos logit contributions
ine+0.057
 birthday+0.052
 X+0.050
 eye+0.050
 pattern+0.050
Neg logit contributions
Run-0.055
mun-0.053
ls-0.050
Hey-0.046
 distracted-0.045
Token practice
Token ID3357
Feature activation+1.45e-02

Pos logit contributions
ine+0.023
 birthday+0.021
 X+0.021
 Office+0.020
 Jupiter+0.020
Neg logit contributions
ls-0.029
Run-0.028
mun-0.028
led-0.026
Hey-0.025
Token makes
Token ID1838
Feature activation+1.72e-02

Pos logit contributions
ine+0.021
 birthday+0.021
 X+0.020
 Office+0.019
 Jupiter+0.019
Neg logit contributions
ls-0.026
mun-0.025
Run-0.025
P-0.022
led-0.022
Token perfect
Token ID2818
Feature activation+1.06e-02

Pos logit contributions
ine+0.028
 X+0.027
 birthday+0.026
 Office+0.026
hip+0.026
Neg logit contributions
ls-0.028
mun-0.027
Run-0.027
P-0.025
 bitten-0.023
Token can
Token ID460
Feature activation+1.77e-04

Pos logit contributions
ine+0.025
 birthday+0.023
hip+0.022
 Office+0.022
rap+0.022
Neg logit contributions
Run-0.029
ls-0.029
mun-0.028
P-0.025
Hey-0.025
Token draw
Token ID3197
Feature activation+3.40e-02

Pos logit contributions
ine+0.000
 birthday+0.000
 Office+0.000
hip+0.000
 X+0.000
Neg logit contributions
ls-0.000
Run-0.000
mun-0.000
P-0.000
Hey-0.000
Token on
Token ID319
Feature activation+2.76e-02

Pos logit contributions
ine+0.063
hip+0.059
sing+0.057
 Office+0.057
 birthday+0.056
Neg logit contributions
ls-0.057
Run-0.046
mun-0.046
led-0.043
P-0.041
Token paper
Token ID3348
Feature activation+3.32e-02

Pos logit contributions
ine+0.053
hip+0.051
 birthday+0.049
 Office+0.049
sing+0.049
Neg logit contributions
ls-0.039
Run-0.035
mun-0.034
P-0.030
Hey-0.030
Token with
Token ID351
Feature activation+1.99e-02

Pos logit contributions
 Office+0.058
hip+0.058
ine+0.057
sing+0.056
uries+0.055
Neg logit contributions
ls-0.052
mun-0.038
led-0.038
Run-0.034
P-0.034
Token cr
Token ID1067
Feature activation+5.47e-02

Pos logit contributions
ine+0.033
 Office+0.032
hip+0.031
rap+0.030
uries+0.029
Neg logit contributions
ls-0.033
Run-0.030
mun-0.028
P-0.028
 bitten-0.028
Tokenay
Token ID323
Feature activation+2.91e-02

Pos logit contributions
 Office+0.067
 birthday+0.064
sing+0.063
 reminder+0.060
 experiment+0.059
Neg logit contributions
ls-0.112
P-0.103
awed-0.095
led-0.094
mun-0.091
Tokenons
Token ID684
Feature activation+2.01e-02

Pos logit contributions
 birthday+0.076
ine+0.075
 Office+0.074
hip+0.074
sing+0.073
Neg logit contributions
ls-0.021
mun-0.015
Run-0.014
P-0.014
led-0.012
Token and
Token ID290
Feature activation+2.62e-02

Pos logit contributions
ine+0.035
 Office+0.034
 birthday+0.033
hip+0.033
rap+0.033
Neg logit contributions
ls-0.031
mun-0.025
Run-0.025
P-0.024
led-0.024
Token stickers
Token ID28568
Feature activation+3.75e-02

Pos logit contributions
ine+0.039
hip+0.037
rap+0.036
 Office+0.036
 birthday+0.034
Neg logit contributions
ls-0.047
Run-0.047
mun-0.044
P-0.042
 bitten-0.041
Token.
Token ID13
Feature activation+7.85e-03

Pos logit contributions
ine+0.076
 Office+0.074
 birthday+0.073
sing+0.072
 Jupiter+0.070
Neg logit contributions
ls-0.048
led-0.037
mun-0.033
Run-0.032
P-0.032
Token a
Token ID257
Feature activation+2.05e-02

Pos logit contributions
hip+0.095
 basis+0.090
sing+0.085
jo+0.085
ine+0.084
Neg logit contributions
ls-0.058
 bitten-0.053
 distracted-0.048
 st-0.047
P-0.047
Token different
Token ID1180
Feature activation+1.30e-02

Pos logit contributions
hip+0.034
ine+0.033
rap+0.031
 Him+0.029
urg+0.029
Neg logit contributions
P-0.031
ls-0.030
Run-0.029
 st-0.027
 bitten-0.027
Token color
Token ID3124
Feature activation+2.25e-02

Pos logit contributions
hip+0.017
ine+0.017
rap+0.017
 Jupiter+0.016
 Office+0.016
Neg logit contributions
ls-0.022
Run-0.021
mun-0.020
 bitten-0.019
Hey-0.018
Token.
Token ID13
Feature activation+2.27e-02

Pos logit contributions
ine+0.049
sing+0.048
 Office+0.048
 Jupiter+0.048
hip+0.047
Neg logit contributions
ls-0.023
P-0.013
guard-0.013
Run-0.012
 bitten-0.011
Token Purple
Token ID17265
Feature activation+1.98e-02

Pos logit contributions
ine+0.041
 birthday+0.038
 desk+0.037
rap+0.037
sing+0.037
Neg logit contributions
Run-0.033
Hey-0.029
ls-0.028
mun-0.028
P-0.028
Token is
Token ID318
Feature activation+5.47e-02

Pos logit contributions
ine+0.032
 birthday+0.031
hip+0.031
sing+0.031
 X+0.031
Neg logit contributions
ls-0.033
mun-0.028
Run-0.028
P-0.026
Hey-0.026
Token a
Token ID257
Feature activation+2.13e-02

Pos logit contributions
hip+0.101
 basis+0.097
 Office+0.091
sing+0.090
jo+0.090
Neg logit contributions
ls-0.067
 bitten-0.061
 st-0.056
P-0.056
 distracted-0.056
Token special
Token ID2041
Feature activation+2.64e-02

Pos logit contributions
hip+0.034
ine+0.033
rap+0.032
urg+0.029
ika+0.029
Neg logit contributions
P-0.033
ls-0.032
Run-0.031
mun-0.029
 st-0.029
Token color
Token ID3124
Feature activation+2.50e-02

Pos logit contributions
hip+0.037
ine+0.035
rap+0.035
 Jupiter+0.035
 Him+0.033
Neg logit contributions
ls-0.043
Run-0.040
mun-0.039
P-0.039
 bitten-0.037
Token.
Token ID13
Feature activation+2.59e-02

Pos logit contributions
sing+0.054
ine+0.053
 Office+0.053
 Jupiter+0.053
artments+0.052
Neg logit contributions
ls-0.025
P-0.016
guard-0.014
mun-0.013
Run-0.013
Token Purple
Token ID17265
Feature activation+2.24e-02

Pos logit contributions
ine+0.047
 birthday+0.043
 desk+0.043
sing+0.042
urger+0.042
Neg logit contributions
Run-0.037
Hey-0.034
mun-0.033
P-0.032
ls-0.032
Token
Token ID198
Feature activation+1.13e-02

Pos logit contributions
ine+0.033
 birthday+0.031
 X+0.031
 Office+0.030
hip+0.030
Neg logit contributions
ls-0.021
Run-0.020
mun-0.020
P-0.019
Hey-0.017
TokenM
Token ID44
Feature activation+1.97e-02

Pos logit contributions
ine+0.021
 birthday+0.020
 X+0.020
 Office+0.019
hip+0.019
Neg logit contributions
Run-0.016
ls-0.016
mun-0.016
P-0.015
Hey-0.014
Tokenoral
Token ID6864
Feature activation+1.52e-02

Pos logit contributions
ine+0.032
 birthday+0.029
 Office+0.029
 X+0.029
hip+0.029
Neg logit contributions
ls-0.035
Run-0.033
mun-0.031
P-0.029
led-0.028
Token of
Token ID286
Feature activation+4.73e-02

Pos logit contributions
ine+0.024
 birthday+0.023
hip+0.021
 X+0.021
 Office+0.021
Neg logit contributions
ls-0.027
Run-0.026
mun-0.025
P-0.024
Hey-0.023
Token the
Token ID262
Feature activation+1.12e-02

Pos logit contributions
ine+0.108
artments+0.106
uries+0.106
sing+0.099
urger+0.099
Neg logit contributions
mun-0.041
P-0.041
Run-0.040
ls-0.037
Hey-0.036
Token story
Token ID1621
Feature activation+5.42e-02

Pos logit contributions
ine+0.013
 birthday+0.011
hip+0.011
artments+0.011
 Office+0.011
Neg logit contributions
ls-0.025
Run-0.024
mun-0.024
P-0.023
Hey-0.022
Token:
Token ID25
Feature activation+3.14e-02

Pos logit contributions
ine+0.075
urger+0.073
 X+0.071
uries+0.071
artments+0.070
Neg logit contributions
ls-0.101
mun-0.084
Run-0.081
led-0.081
Hey-0.080
Token If
Token ID1002
Feature activation+7.54e-03

Pos logit contributions
ine+0.055
 X+0.049
 birthday+0.049
urger+0.048
 Office+0.047
Neg logit contributions
Run-0.046
mun-0.045
ls-0.044
 distracted-0.039
 bitten-0.039
Token you
Token ID345
Feature activation-9.88e-03

Pos logit contributions
ine+0.016
 birthday+0.015
hip+0.014
 Office+0.014
 X+0.014
Neg logit contributions
ls-0.009
Run-0.009
mun-0.009
P-0.008
Hey-0.008
Token are
Token ID389
Feature activation-9.32e-03

Pos logit contributions
ls+0.023
Run+0.022
mun+0.022
P+0.020
awed+0.020
Neg logit contributions
ine-0.011
 birthday-0.010
hip-0.010
 X-0.009
 eye-0.009
Token sorry
Token ID7926
Feature activation+9.00e-03

Pos logit contributions
Run+0.017
ls+0.017
mun+0.016
P+0.015
Hey+0.015
Neg logit contributions
ine-0.014
 birthday-0.013
hip-0.012
 Office-0.012
 X-0.012
Token She
Token ID1375
Feature activation-4.96e-03

Pos logit contributions
ls+0.009
Run+0.009
mun+0.009
P+0.008
led+0.008
Neg logit contributions
ine-0.008
 birthday-0.008
 Office-0.007
 X-0.007
hip-0.007
Token is
Token ID318
Feature activation+1.38e-02

Pos logit contributions
Run+0.008
ls+0.008
mun+0.008
P+0.008
 bitten+0.007
Neg logit contributions
ine-0.008
 birthday-0.007
 X-0.007
 Office-0.007
 Jupiter-0.007
Token using
Token ID1262
Feature activation+2.30e-03

Pos logit contributions
ine+0.026
 birthday+0.025
hip+0.024
 Office+0.024
 X+0.023
Neg logit contributions
ls-0.019
Run-0.019
mun-0.018
 bitten-0.017
P-0.016
Token the
Token ID262
Feature activation+2.04e-02

Pos logit contributions
 birthday+0.005
hip+0.005
 Office+0.005
ine+0.005
 Birthday+0.004
Neg logit contributions
ls-0.003
 bitten-0.002
Run-0.002
mun-0.002
 distracted-0.002
Token red
Token ID2266
Feature activation+2.19e-02

Pos logit contributions
ine+0.035
hip+0.033
 Office+0.032
rap+0.031
uries+0.031
Neg logit contributions
ls-0.032
Run-0.029
P-0.029
mun-0.028
 bitten-0.027
Token cr
Token ID1067
Feature activation+5.39e-02

Pos logit contributions
ine+0.040
uries+0.039
hip+0.039
 Jupiter+0.039
 Office+0.038
Neg logit contributions
ls-0.030
P-0.026
Run-0.026
Hey-0.024
mun-0.024
Tokenay
Token ID323
Feature activation+2.17e-02

Pos logit contributions
 Office+0.074
 birthday+0.074
sing+0.070
 appearance+0.069
 experiment+0.068
Neg logit contributions
ls-0.098
P-0.093
awed-0.083
led-0.081
itt-0.080
Tokenon
Token ID261
Feature activation+7.79e-03

Pos logit contributions
 birthday+0.048
 Office+0.047
sing+0.047
ine+0.047
uries+0.047
Neg logit contributions
ls-0.024
P-0.019
Run-0.017
led-0.017
 bitten-0.016
Token.
Token ID13
Feature activation+1.44e-03

Pos logit contributions
ine+0.015
 birthday+0.014
 Office+0.014
hip+0.014
 Jupiter+0.014
Neg logit contributions
ls-0.011
Run-0.010
mun-0.009
led-0.009
P-0.009
Token He
Token ID679
Feature activation+3.28e-03

Pos logit contributions
ine+0.002
 birthday+0.002
 Office+0.002
hip+0.002
 X+0.002
Neg logit contributions
ls-0.003
Run-0.003
mun-0.002
P-0.002
 bitten-0.002
Token says
Token ID1139
Feature activation+8.47e-03

Pos logit contributions
ine+0.005
 birthday+0.005
 Office+0.004
hip+0.004
 X+0.004
Neg logit contributions
Run-0.006
ls-0.006
mun-0.006
P-0.005
Hey-0.005
Token liked
Token ID8288
Feature activation-7.01e-03

Pos logit contributions
ine+0.009
 birthday+0.008
 Office+0.008
hip+0.008
 X+0.008
Neg logit contributions
Run-0.008
ls-0.008
mun-0.008
P-0.007
Hey-0.007
Token to
Token ID284
Feature activation+9.54e-03

Pos logit contributions
Run+0.012
ls+0.012
mun+0.012
P+0.011
led+0.010
Neg logit contributions
ine-0.011
 birthday-0.011
 X-0.010
 Office-0.010
 Jupiter-0.010
Token draw
Token ID3197
Feature activation+1.61e-02

Pos logit contributions
ine+0.013
 Office+0.012
 birthday+0.012
rap+0.012
 Jupiter+0.012
Neg logit contributions
Run-0.017
ls-0.017
mun-0.016
P-0.016
 bitten-0.015
Token with
Token ID351
Feature activation+6.35e-03

Pos logit contributions
ine+0.031
uries+0.029
hip+0.029
sing+0.029
 Office+0.029
Neg logit contributions
ls-0.025
mun-0.020
led-0.020
P-0.020
Run-0.020
Token his
Token ID465
Feature activation+1.53e-02

Pos logit contributions
ine+0.012
 birthday+0.011
 Office+0.011
hip+0.011
uries+0.010
Neg logit contributions
ls-0.009
Run-0.009
mun-0.009
P-0.008
 bitten-0.008
Token cr
Token ID1067
Feature activation+5.38e-02

Pos logit contributions
ine+0.025
hip+0.024
 Office+0.023
 birthday+0.022
rap+0.022
Neg logit contributions
ls-0.027
Run-0.024
mun-0.023
P-0.023
 bitten-0.022
Tokenay
Token ID323
Feature activation+2.69e-02

Pos logit contributions
 Office+0.068
 birthday+0.062
sing+0.059
ine+0.059
 experiment+0.058
Neg logit contributions
ls-0.110
P-0.100
awed-0.094
led-0.092
 bitten-0.089
Tokenons
Token ID684
Feature activation+8.13e-03

Pos logit contributions
 birthday+0.070
ine+0.069
 Office+0.069
uries+0.067
sing+0.067
Neg logit contributions
ls-0.019
mun-0.013
P-0.012
led-0.011
Run-0.011
Token and
Token ID290
Feature activation+2.18e-02

Pos logit contributions
ine+0.015
 Office+0.014
 birthday+0.014
 Jupiter+0.014
rap+0.014
Neg logit contributions
ls-0.012
mun-0.010
P-0.010
led-0.010
Run-0.010
Token markers
Token ID19736
Feature activation+2.60e-02

Pos logit contributions
ine+0.032
hip+0.028
 Office+0.028
 birthday+0.027
rap+0.027
Neg logit contributions
ls-0.041
mun-0.039
Run-0.039
 bitten-0.035
P-0.035
Token.
Token ID13
Feature activation+1.17e-02

Pos logit contributions
ine+0.052
 Office+0.051
 birthday+0.048
sing+0.048
ika+0.048
Neg logit contributions
ls-0.034
led-0.028
mun-0.026
 bitten-0.024
P-0.024
Token.
Token ID13
Feature activation+8.38e-03

Pos logit contributions
ine+0.073
 Office+0.069
sing+0.068
artments+0.066
urger+0.066
Neg logit contributions
ls-0.039
mun-0.028
Run-0.026
led-0.025
P-0.025
Token Tom
Token ID4186
Feature activation+1.49e-02

Pos logit contributions
ine+0.015
 birthday+0.014
hip+0.014
rap+0.013
 Office+0.013
Neg logit contributions
Run-0.013
ls-0.012
mun-0.012
Hey-0.011
P-0.011
Token has
Token ID468
Feature activation+1.33e-02

Pos logit contributions
 birthday+0.026
hip+0.024
ine+0.023
 perfection+0.022
sing+0.022
Neg logit contributions
mun-0.024
ls-0.024
Run-0.021
Hey-0.021
 teased-0.021
Token a
Token ID257
Feature activation+6.48e-03

Pos logit contributions
ine+0.024
hip+0.023
rap+0.022
 Office+0.022
 birthday+0.022
Neg logit contributions
Run-0.019
ls-0.018
P-0.018
 bitten-0.018
mun-0.018
Token red
Token ID2266
Feature activation+1.84e-02

Pos logit contributions
hip+0.011
ine+0.011
rap+0.010
uries+0.010
 Office+0.010
Neg logit contributions
ls-0.010
P-0.010
mun-0.009
Run-0.009
 bitten-0.008
Token cr
Token ID1067
Feature activation+5.36e-02

Pos logit contributions
uries+0.030
ine+0.028
hip+0.028
 Office+0.028
 Jupiter+0.028
Neg logit contributions
P-0.027
ls-0.027
Run-0.026
Hey-0.024
mun-0.024
Tokenay
Token ID323
Feature activation+2.50e-02

Pos logit contributions
 Office+0.072
 birthday+0.068
 experiment+0.066
sing+0.064
 Jupiter+0.063
Neg logit contributions
ls-0.097
P-0.095
awed-0.086
 bitten-0.083
led-0.083
Tokenon
Token ID261
Feature activation+1.61e-02

Pos logit contributions
 birthday+0.052
 Office+0.051
ine+0.051
uries+0.051
sing+0.049
Neg logit contributions
ls-0.029
P-0.026
Run-0.024
 bitten-0.023
mun-0.023
Token and
Token ID290
Feature activation+1.36e-02

Pos logit contributions
ine+0.030
 Office+0.029
 Jupiter+0.028
hip+0.028
rap+0.027
Neg logit contributions
ls-0.022
P-0.018
led-0.017
 bitten-0.017
Run-0.017
Token Lily
Token ID20037
Feature activation+1.40e-02

Pos logit contributions
ine+0.024
hip+0.023
 birthday+0.022
 Office+0.022
rap+0.021
Neg logit contributions
ls-0.022
Run-0.021
mun-0.020
P-0.019
 bitten-0.019
Token has
Token ID468
Feature activation+2.14e-02

Pos logit contributions
 birthday+0.025
hip+0.025
 Jupiter+0.024
rap+0.024
ine+0.024
Neg logit contributions
ls-0.019
mun-0.018
led-0.017
 cried-0.017
 bitten-0.017
Token with
Token ID351
Feature activation+1.06e-02

Pos logit contributions
ine+0.005
 birthday+0.005
 Office+0.004
hip+0.004
 X+0.004
Neg logit contributions
Run-0.006
ls-0.006
mun-0.005
P-0.005
Hey-0.005
Token his
Token ID465
Feature activation+9.31e-03

Pos logit contributions
ine+0.017
 birthday+0.016
 X+0.015
 moon+0.015
 Jupiter+0.015
Neg logit contributions
Run-0.018
ls-0.018
mun-0.018
P-0.016
Hey-0.016
Token drum
Token ID13026
Feature activation+1.71e-02

Pos logit contributions
ine+0.013
 birthday+0.013
 X+0.012
 Office+0.012
 Jupiter+0.011
Neg logit contributions
Run-0.018
ls-0.017
mun-0.017
Hey-0.016
P-0.015
Token.
Token ID13
Feature activation+2.83e-02

Pos logit contributions
ine+0.029
 birthday+0.026
 Office+0.026
 X+0.026
rap+0.025
Neg logit contributions
ls-0.028
mun-0.026
Run-0.025
P-0.024
led-0.023
Token It
Token ID632
Feature activation+2.28e-02

Pos logit contributions
ine+0.056
 believe+0.053
 birthday+0.052
uries+0.051
urger+0.051
Neg logit contributions
Run-0.030
mun-0.028
ls-0.027
Hey-0.027
-0.022
Token was
Token ID373
Feature activation+5.34e-02

Pos logit contributions
ine+0.037
 birthday+0.035
 Office+0.032
hip+0.032
 Jupiter+0.032
Neg logit contributions
Run-0.039
ls-0.039
mun-0.037
P-0.034
Hey-0.033
Token big
Token ID1263
Feature activation+1.32e-02

Pos logit contributions
 Office+0.098
hip+0.098
ine+0.095
uries+0.094
urger+0.092
Neg logit contributions
Run-0.067
P-0.066
ls-0.066
 bitten-0.066
 distracted-0.064
Token and
Token ID290
Feature activation+1.89e-02

Pos logit contributions
ine+0.022
hip+0.020
 birthday+0.020
 Office+0.020
uries+0.019
Neg logit contributions
ls-0.022
Run-0.021
mun-0.021
P-0.019
 bitten-0.019
Token heavy
Token ID4334
Feature activation+1.15e-02

Pos logit contributions
ine+0.032
 birthday+0.030
hip+0.029
 Office+0.029
rap+0.029
Neg logit contributions
Run-0.032
ls-0.031
mun-0.029
P-0.028
Hey-0.027
Token,
Token ID11
Feature activation+7.01e-03

Pos logit contributions
ine+0.022
rap+0.021
 Office+0.021
uries+0.020
urger+0.020
Neg logit contributions
ls-0.015
Run-0.014
mun-0.013
 bitten-0.012
led-0.012
Token but
Token ID475
Feature activation+1.91e-02

Pos logit contributions
ine+0.011
 birthday+0.010
 Office+0.009
hip+0.009
 X+0.009
Neg logit contributions
Run-0.013
ls-0.013
mun-0.012
P-0.011
Hey-0.011
Token
Token ID198
Feature activation+6.21e-03

Pos logit contributions
ine+0.028
 birthday+0.027
 Office+0.025
 X+0.025
hip+0.025
Neg logit contributions
ls-0.019
Run-0.019
mun-0.018
P-0.017
Hey-0.016
TokenM
Token ID44
Feature activation+1.01e-02

Pos logit contributions
ine+0.011
 birthday+0.010
hip+0.010
 Office+0.010
 X+0.009
Neg logit contributions
ls-0.010
Run-0.010
mun-0.010
P-0.008
 bitten-0.008
Tokenoral
Token ID6864
Feature activation+1.49e-02

Pos logit contributions
ine+0.016
 birthday+0.015
 Office+0.014
hip+0.014
 Jupiter+0.014
Neg logit contributions
ls-0.018
Run-0.017
mun-0.017
P-0.016
Hey-0.015
Token of
Token ID286
Feature activation+4.11e-02

Pos logit contributions
ine+0.022
 birthday+0.021
 X+0.020
sing+0.020
hip+0.020
Neg logit contributions
ls-0.027
Run-0.026
mun-0.025
P-0.024
Hey-0.024
Token the
Token ID262
Feature activation+7.40e-03

Pos logit contributions
ine+0.093
artments+0.093
uries+0.092
sing+0.089
urger+0.088
Neg logit contributions
mun-0.036
Run-0.034
P-0.034
ls-0.034
Hey-0.030
Token story
Token ID1621
Feature activation+5.33e-02

Pos logit contributions
ine+0.009
artments+0.008
sing+0.008
 birthday+0.008
hip+0.008
Neg logit contributions
ls-0.015
Run-0.015
mun-0.015
P-0.014
Hey-0.014
Token:
Token ID25
Feature activation+3.34e-02

Pos logit contributions
ine+0.070
urger+0.067
 birthday+0.066
sing+0.065
uries+0.063
Neg logit contributions
ls-0.103
Run-0.087
mun-0.086
Hey-0.084
led-0.081
Token Sharing
Token ID36644
Feature activation+1.23e-02

Pos logit contributions
ine+0.056
 birthday+0.051
urger+0.051
 X+0.050
 eye+0.050
Neg logit contributions
Run-0.049
mun-0.047
ls-0.046
Hey-0.042
 distracted-0.041
Token is
Token ID318
Feature activation+2.12e-02

Pos logit contributions
ine+0.019
 birthday+0.018
 X+0.017
 Office+0.017
hip+0.017
Neg logit contributions
ls-0.022
Run-0.021
mun-0.021
P-0.019
Hey-0.019
Token caring
Token ID18088
Feature activation+1.69e-02

Pos logit contributions
ine+0.030
 birthday+0.027
 X+0.027
hip+0.026
 Office+0.026
Neg logit contributions
ls-0.039
Run-0.039
mun-0.038
P-0.036
 distracted-0.035
Token!
Token ID0
Feature activation+1.06e-02

Pos logit contributions
ine+0.026
 X+0.024
 Jupiter+0.024
 Office+0.024
 birthday+0.024
Neg logit contributions
ls-0.029
mun-0.026
Run-0.026
led-0.024
Hey-0.023
Token his
Token ID465
Feature activation+1.33e-02

Pos logit contributions
ine+0.008
 birthday+0.007
hip+0.007
 Office+0.007
 X+0.007
Neg logit contributions
ls-0.005
Run-0.004
mun-0.004
P-0.004
Hey-0.004
Token sc
Token ID629
Feature activation+3.99e-03

Pos logit contributions
ine+0.019
 birthday+0.017
hip+0.016
 Office+0.016
 Jupiter+0.016
Neg logit contributions
ls-0.026
Run-0.025
mun-0.025
P-0.023
Hey-0.022
Tokenooter
Token ID25141
Feature activation+8.00e-03

Pos logit contributions
ine+0.007
 birthday+0.006
 Office+0.006
 Jupiter+0.006
 moon+0.006
Neg logit contributions
ls-0.006
Run-0.006
mun-0.006
P-0.005
 distracted-0.005
Token.
Token ID13
Feature activation+2.45e-02

Pos logit contributions
ine+0.013
 birthday+0.012
 Office+0.012
 Jupiter+0.012
 X+0.012
Neg logit contributions
ls-0.013
Run-0.013
mun-0.013
P-0.012
led-0.011
Token It
Token ID632
Feature activation+1.75e-02

Pos logit contributions
ine+0.051
 birthday+0.047
 believe+0.047
urger+0.046
 Birthday+0.045
Neg logit contributions
Run-0.025
mun-0.024
Hey-0.022
ls-0.022
P-0.019
Token was
Token ID373
Feature activation+5.31e-02

Pos logit contributions
ine+0.031
 birthday+0.029
hip+0.027
 Office+0.027
 X+0.027
Neg logit contributions
ls-0.028
Run-0.028
mun-0.027
P-0.024
Hey-0.024
Token red
Token ID2266
Feature activation+6.35e-03

Pos logit contributions
hip+0.101
 Office+0.100
ine+0.099
uries+0.096
urger+0.096
Neg logit contributions
Run-0.061
 distracted-0.060
P-0.060
 bitten-0.060
 tangled-0.059
Token and
Token ID290
Feature activation+2.30e-02

Pos logit contributions
ine+0.010
 Office+0.009
 birthday+0.009
 Jupiter+0.009
hip+0.009
Neg logit contributions
ls-0.011
Run-0.011
mun-0.010
P-0.009
Hey-0.009
Token had
Token ID550
Feature activation+3.86e-02

Pos logit contributions
ine+0.044
 birthday+0.043
hip+0.042
rap+0.042
 Office+0.041
Neg logit contributions
ls-0.033
Run-0.032
mun-0.029
P-0.028
Hey-0.027
Token two
Token ID734
Feature activation+5.35e-03

Pos logit contributions
ine+0.065
 birthday+0.060
 Office+0.057
hip+0.057
rap+0.057
Neg logit contributions
Run-0.063
ls-0.062
mun-0.059
 bitten-0.056
P-0.056
Token wheels
Token ID13666
Feature activation+1.58e-02

Pos logit contributions
ine+0.008
 birthday+0.007
hip+0.007
 Office+0.007
 Jupiter+0.007
Neg logit contributions
ls-0.010
Run-0.010
mun-0.010
P-0.009
 bitten-0.009
Tokenay
Token ID323
Feature activation+9.78e-03

Pos logit contributions
 Office+0.041
 birthday+0.038
ine+0.038
 X+0.037
 Jupiter+0.037
Neg logit contributions
ls-0.078
P-0.073
awed-0.069
mun-0.068
led-0.068
Tokenons
Token ID684
Feature activation+1.86e-03

Pos logit contributions
ine+0.024
 birthday+0.024
 Office+0.023
 X+0.023
hip+0.023
Neg logit contributions
ls-0.008
mun-0.007
Run-0.007
P-0.006
led-0.005
Token.
Token ID13
Feature activation+6.03e-03

Pos logit contributions
ine+0.003
 birthday+0.003
 Office+0.003
hip+0.003
 X+0.003
Neg logit contributions
ls-0.003
Run-0.003
mun-0.003
P-0.002
led-0.002
Token She
Token ID1375
Feature activation+1.82e-02

Pos logit contributions
ine+0.011
 Jupiter+0.009
 Office+0.009
 birthday+0.009
hip+0.009
Neg logit contributions
Run-0.009
mun-0.009
ls-0.009
Hey-0.008
P-0.008
Token writes
Token ID6797
Feature activation+3.28e-02

Pos logit contributions
hip+0.029
 birthday+0.029
ine+0.028
 Office+0.027
rap+0.026
Neg logit contributions
ls-0.031
mun-0.030
Run-0.028
P-0.027
 distracted-0.026
Token words
Token ID2456
Feature activation+5.31e-02

Pos logit contributions
ine+0.061
sing+0.059
 eye+0.058
 basis+0.057
hip+0.057
Neg logit contributions
ls-0.051
led-0.042
mun-0.041
Hey-0.037
 distracted-0.036
Token and
Token ID290
Feature activation+2.73e-02

Pos logit contributions
ine+0.100
sing+0.100
 Office+0.098
 birthday+0.098
urger+0.096
Neg logit contributions
ls-0.070
mun-0.057
led-0.056
Hey-0.053
 teased-0.052
Token draws
Token ID14293
Feature activation+3.11e-02

Pos logit contributions
ine+0.037
 Office+0.035
rap+0.035
hip+0.035
 birthday+0.033
Neg logit contributions
ls-0.049
mun-0.049
Run-0.049
 teased-0.044
 bitten-0.044
Token pictures
Token ID5986
Feature activation+3.97e-02

Pos logit contributions
sing+0.057
ine+0.055
hip+0.055
 Office+0.054
otive+0.053
Neg logit contributions
ls-0.047
mun-0.040
led-0.038
P-0.037
Run-0.033
Token on
Token ID319
Feature activation+2.71e-02

Pos logit contributions
ine+0.084
sing+0.081
 Office+0.080
urger+0.079
 birthday+0.077
Neg logit contributions
ls-0.049
mun-0.039
Run-0.034
led-0.033
Hey-0.032
Token paper
Token ID3348
Feature activation+3.05e-02

Pos logit contributions
sing+0.054
 birthday+0.053
 Office+0.052
ine+0.052
hip+0.052
Neg logit contributions
ls-0.034
mun-0.027
Run-0.026
P-0.026
Hey-0.026
Token in
Token ID287
Feature activation+1.85e-02

Pos logit contributions
ls+0.007
Run+0.007
mun+0.007
led+0.006
 bitten+0.006
Neg logit contributions
 birthday-0.005
ine-0.004
 pattern-0.004
 X-0.004
 Office-0.004
Token her
Token ID607
Feature activation+1.42e-02

Pos logit contributions
ine+0.033
 birthday+0.031
hip+0.030
 Office+0.030
 X+0.029
Neg logit contributions
Run-0.029
ls-0.028
mun-0.027
P-0.025
Hey-0.025
Token bedroom
Token ID14043
Feature activation+2.32e-02

Pos logit contributions
ine+0.018
hip+0.016
 Office+0.015
 birthday+0.015
 Jupiter+0.015
Neg logit contributions
ls-0.030
Run-0.028
mun-0.028
P-0.027
 bitten-0.025
Token.
Token ID13
Feature activation+2.47e-02

Pos logit contributions
ine+0.037
 birthday+0.035
 Office+0.032
hip+0.032
 X+0.031
Neg logit contributions
Run-0.041
ls-0.041
mun-0.040
P-0.036
Hey-0.035
Token It
Token ID632
Feature activation+3.08e-02

Pos logit contributions
ine+0.051
urger+0.046
rap+0.045
hip+0.045
 birthday+0.044
Neg logit contributions
Run-0.030
ls-0.027
mun-0.026
Hey-0.026
 bitten-0.024
Token was
Token ID373
Feature activation+5.31e-02

Pos logit contributions
ine+0.061
 birthday+0.059
hip+0.057
 Office+0.055
 X+0.055
Neg logit contributions
ls-0.042
Run-0.041
mun-0.040
P-0.036
Hey-0.035
Token a
Token ID257
Feature activation+1.82e-02

Pos logit contributions
hip+0.104
ine+0.102
urger+0.098
 Office+0.098
uries+0.097
Neg logit contributions
Run-0.065
 bitten-0.063
P-0.063
ls-0.061
 distracted-0.060
Token very
Token ID845
Feature activation+1.44e-02

Pos logit contributions
ine+0.036
hip+0.035
uries+0.033
rap+0.033
artments+0.031
Neg logit contributions
Run-0.026
ls-0.025
P-0.025
mun-0.022
Hey-0.021
Token messy
Token ID24097
Feature activation+1.88e-02

Pos logit contributions
ine+0.024
hip+0.022
 Office+0.021
 birthday+0.021
 Jupiter+0.020
Neg logit contributions
ls-0.026
Run-0.024
mun-0.023
P-0.022
 bitten-0.021
Token place
Token ID1295
Feature activation+1.34e-02

Pos logit contributions
ine+0.027
hip+0.026
uries+0.026
rap+0.025
 Jupiter+0.025
Neg logit contributions
ls-0.032
Run-0.032
mun-0.030
P-0.030
Hey-0.029
Token,
Token ID11
Feature activation+8.06e-03

Pos logit contributions
ine+0.023
hip+0.022
 Office+0.021
urger+0.021
 Jupiter+0.021
Neg logit contributions
ls-0.020
Run-0.019
mun-0.018
 bitten-0.017
P-0.017
Token
Token ID198
Feature activation-9.61e-03

Pos logit contributions
mun+0.026
ls+0.026
Run+0.025
P+0.024
led+0.024
Neg logit contributions
ine-0.018
 birthday-0.018
 Grand-0.017
 desk-0.017
 Today-0.016
Token
Token ID198
Feature activation-1.44e-02

Pos logit contributions
Run+0.017
ls+0.017
mun+0.016
led+0.015
P+0.015
Neg logit contributions
ine-0.015
 birthday-0.014
hip-0.013
 X-0.013
 Office-0.013
TokenTom
Token ID13787
Feature activation-9.87e-03

Pos logit contributions
ls+0.024
Run+0.023
mun+0.023
led+0.020
 bitten+0.020
Neg logit contributions
ine-0.025
 birthday-0.023
hip-0.022
 Office-0.022
 X-0.022
Token felt
Token ID2936
Feature activation-3.36e-02

Pos logit contributions
ls+0.013
Run+0.013
mun+0.012
P+0.011
Hey+0.011
Neg logit contributions
ine-0.020
 birthday-0.018
 Office-0.018
hip-0.018
 X-0.018
Token angry
Token ID7954
Feature activation-2.26e-02

Pos logit contributions
Run+0.050
mun+0.049
ls+0.049
P+0.047
Hey+0.042
Neg logit contributions
ine-0.058
 birthday-0.055
 X-0.053
 eye-0.053
 end-0.052
Token and
Token ID290
Feature activation-5.78e-02

Pos logit contributions
ls+0.036
Run+0.036
mun+0.034
P+0.030
 bitten+0.030
Neg logit contributions
ine-0.040
 birthday-0.037
 Office-0.036
hip-0.035
 Jupiter-0.035
Token sad
Token ID6507
Feature activation-1.53e-02

Pos logit contributions
P+0.108
mun+0.105
ries+0.101
ls+0.096
Run+0.095
Neg logit contributions
 for-0.072
 be-0.068
 eye-0.066
sing-0.065
 end-0.065
Token.
Token ID13
Feature activation-2.41e-02

Pos logit contributions
ls+0.023
Run+0.023
mun+0.022
P+0.019
led+0.019
Neg logit contributions
ine-0.029
 birthday-0.026
 Office-0.026
hip-0.025
 Jupiter-0.025
Token He
Token ID679
Feature activation-2.36e-02

Pos logit contributions
Run+0.037
ls+0.036
mun+0.035
P+0.031
Hey+0.031
Neg logit contributions
ine-0.045
 birthday-0.041
hip-0.039
 Office-0.039
 Jupiter-0.039
Token tried
Token ID3088
Feature activation-1.22e-02

Pos logit contributions
ls+0.037
mun+0.036
Run+0.036
P+0.032
aro+0.031
Neg logit contributions
ine-0.043
 X-0.036
 Office-0.035
 pattern-0.035
 birthday-0.034
Token to
Token ID284
Feature activation-3.20e-02

Pos logit contributions
Run+0.021
ls+0.021
mun+0.020
P+0.019
led+0.018
Neg logit contributions
ine-0.020
 birthday-0.019
hip-0.017
 Office-0.017
 X-0.017
Token burned
Token ID11544
Feature activation-1.75e-03

Pos logit contributions
ine+0.003
 birthday+0.003
 Office+0.003
 X+0.003
 eye+0.003
Neg logit contributions
Run-0.002
mun-0.002
ls-0.002
P-0.002
Hey-0.002
Token his
Token ID465
Feature activation+3.36e-04

Pos logit contributions
ls+0.003
Run+0.003
mun+0.002
Hey+0.002
led+0.002
Neg logit contributions
ine-0.003
 birthday-0.003
hip-0.003
 Office-0.003
 Jupiter-0.003
Token hand
Token ID1021
Feature activation-1.20e-02

Pos logit contributions
ine+0.000
hip+0.000
 Jupiter+0.000
rap+0.000
 Office+0.000
Neg logit contributions
ls-0.001
Run-0.001
mun-0.001
 bitten-0.001
led-0.001
Token and
Token ID290
Feature activation-2.71e-02

Pos logit contributions
ls+0.021
Run+0.020
mun+0.019
 bitten+0.018
led+0.018
Neg logit contributions
ine-0.019
 birthday-0.019
 Jupiter-0.017
 Office-0.017
hip-0.017
Token made
Token ID925
Feature activation+2.86e-03

Pos logit contributions
ls+0.041
Run+0.041
mun+0.041
P+0.037
Hey+0.035
Neg logit contributions
ine-0.049
 birthday-0.046
 X-0.043
 Office-0.043
 Jupiter-0.042
Token him
Token ID683
Feature activation-5.64e-02

Pos logit contributions
ine+0.005
hip+0.005
 Office+0.005
 birthday+0.005
uries+0.005
Neg logit contributions
ls-0.004
Run-0.004
 bitten-0.004
 distracted-0.003
led-0.003
Token cry
Token ID3960
Feature activation-4.00e-03

Pos logit contributions
Run+0.095
ls+0.093
mun+0.093
P+0.087
Hey+0.086
Neg logit contributions
ine-0.096
 birthday-0.093
 eye-0.085
 desk-0.082
hip-0.081
Token.
Token ID13
Feature activation-2.02e-02

Pos logit contributions
ls+0.004
Run+0.003
mun+0.003
led+0.003
 distracted+0.003
Neg logit contributions
ine-0.009
 X-0.009
 Office-0.009
rap-0.009
artments-0.009
Token His
Token ID2399
Feature activation-2.68e-03

Pos logit contributions
ls+0.036
Run+0.036
mun+0.034
P+0.031
Hey+0.031
Neg logit contributions
ine-0.032
 birthday-0.030
 Office-0.028
hip-0.028
 X-0.028
Token mom
Token ID1995
Feature activation-1.23e-02

Pos logit contributions
ls+0.005
Run+0.005
mun+0.005
 bitten+0.005
 distracted+0.005
Neg logit contributions
ine-0.003
hip-0.003
 Office-0.003
rap-0.003
 Jupiter-0.003
Token heard
Token ID2982
Feature activation+5.03e-04

Pos logit contributions
ls+0.018
Run+0.018
mun+0.017
P+0.015
Hey+0.015
Neg logit contributions
ine-0.023
 birthday-0.022
 Office-0.021
hip-0.020
 X-0.020
Token so
Token ID523
Feature activation-3.28e-02

Pos logit contributions
ls+0.013
Run+0.013
mun+0.013
P+0.012
Hey+0.011
Neg logit contributions
ine-0.011
 birthday-0.010
 Office-0.009
hip-0.009
 Jupiter-0.009
Token disappointed
Token ID11679
Feature activation-9.16e-03

Pos logit contributions
ls+0.063
Run+0.062
mun+0.062
P+0.057
Hey+0.055
Neg logit contributions
ine-0.045
 birthday-0.044
 believe-0.039
 eye-0.039
 end-0.038
Token and
Token ID290
Feature activation-4.09e-02

Pos logit contributions
ls+0.017
Run+0.016
mun+0.016
P+0.014
Hey+0.014
Neg logit contributions
ine-0.014
 birthday-0.013
 Office-0.012
hip-0.012
 X-0.012
Token he
Token ID339
Feature activation-1.44e-02

Pos logit contributions
mun+0.062
ls+0.062
P+0.059
Run+0.059
awed+0.052
Neg logit contributions
ine-0.073
 birthday-0.065
 Office-0.063
 sleep-0.063
 X-0.063
Token never
Token ID1239
Feature activation-3.73e-03

Pos logit contributions
ls+0.021
Run+0.020
mun+0.020
P+0.018
Hey+0.017
Neg logit contributions
ine-0.028
 birthday-0.024
 Office-0.024
 X-0.024
 Jupiter-0.023
Token got
Token ID1392
Feature activation-5.56e-02

Pos logit contributions
Run+0.005
ls+0.005
mun+0.005
 bitten+0.004
P+0.004
Neg logit contributions
ine-0.008
 birthday-0.007
hip-0.007
 Office-0.007
 Jupiter-0.007
Token the
Token ID262
Feature activation-1.03e-02

Pos logit contributions
Run+0.138
mun+0.130
ls+0.124
Sorry+0.118
Hey+0.118
Neg logit contributions
ine-0.050
 for-0.050
 birthday-0.048
 sleep-0.041
 three-0.040
Token special
Token ID2041
Feature activation+1.27e-02

Pos logit contributions
Run+0.024
ls+0.024
mun+0.024
P+0.022
Hey+0.022
Neg logit contributions
ine-0.010
 birthday-0.010
 Office-0.008
 X-0.008
hip-0.008
Token gift
Token ID6979
Feature activation-6.30e-03

Pos logit contributions
ine+0.013
 birthday+0.011
hip+0.010
 Office+0.010
 Jupiter+0.010
Neg logit contributions
ls-0.030
Run-0.030
mun-0.029
P-0.027
Hey-0.027
Token from
Token ID422
Feature activation-1.63e-02

Pos logit contributions
ls+0.010
Run+0.010
mun+0.009
 bitten+0.008
P+0.008
Neg logit contributions
ine-0.011
 birthday-0.010
 Office-0.010
 Jupiter-0.010
hip-0.010
Token the
Token ID262
Feature activation-2.77e-03

Pos logit contributions
ls+0.023
Run+0.022
mun+0.021
 bitten+0.018
Hey+0.018
Neg logit contributions
ine-0.032
hip-0.029
 birthday-0.029
 Office-0.028
 X-0.027
Token.
Token ID13
Feature activation-2.21e-02

Pos logit contributions
mun+0.048
Run+0.046
ls+0.044
P+0.044
Hey+0.044
Neg logit contributions
ine-0.027
 birthday-0.024
 sleep-0.024
 X-0.023
 end-0.023
Token Tim
Token ID5045
Feature activation-3.09e-02

Pos logit contributions
ls+0.045
mun+0.045
Run+0.044
led+0.040
P+0.040
Neg logit contributions
ine-0.028
 birthday-0.026
 Jupiter-0.025
 X-0.025
hip-0.024
Token and
Token ID290
Feature activation+3.25e-03

Pos logit contributions
ls+0.050
mun+0.049
Run+0.049
P+0.046
 av+0.045
Neg logit contributions
ine-0.052
 eye-0.045
 end-0.043
 X-0.041
 aunt-0.040
Token Sam
Token ID3409
Feature activation-3.71e-02

Pos logit contributions
ine+0.005
 birthday+0.004
 Office+0.004
hip+0.004
 X+0.004
Neg logit contributions
ls-0.006
Run-0.006
mun-0.006
P-0.006
Hey-0.006
Token were
Token ID547
Feature activation-2.61e-02

Pos logit contributions
ls+0.058
mun+0.055
P+0.055
Run+0.054
 av+0.053
Neg logit contributions
ine-0.058
 end-0.057
 eye-0.055
 sleep-0.051
 X-0.050
Token both
Token ID1111
Feature activation-5.51e-02

Pos logit contributions
mun+0.043
Run+0.043
ls+0.041
P+0.039
Hey+0.038
Neg logit contributions
ine-0.042
 birthday-0.040
 eye-0.038
 end-0.038
sing-0.037
Token hurt
Token ID5938
Feature activation-2.43e-02

Pos logit contributions
Run+0.092
ls+0.090
P+0.087
mun+0.085
Hey+0.085
Neg logit contributions
 birthday-0.091
 eye-0.087
 sleep-0.085
ine-0.084
sing-0.082
Token and
Token ID290
Feature activation-4.15e-02

Pos logit contributions
mun+0.054
ls+0.054
Run+0.054
P+0.051
Hey+0.049
Neg logit contributions
ine-0.026
 birthday-0.023
 eye-0.021
 end-0.021
hip-0.021
Token scared
Token ID12008
Feature activation-2.02e-02

Pos logit contributions
mun+0.061
P+0.060
Run+0.059
ls+0.059
Hey+0.055
Neg logit contributions
ine-0.072
 birthday-0.071
 eye-0.068
 sleep-0.067
 end-0.066
Token.
Token ID13
Feature activation-2.66e-02

Pos logit contributions
ls+0.037
Run+0.036
mun+0.036
P+0.034
 bitten+0.032
Neg logit contributions
ine-0.030
 birthday-0.029
hip-0.026
 X-0.026
 Jupiter-0.025
Token
Token ID198
Feature activation+1.83e-03

Pos logit contributions
ls+0.056
mun+0.055
Run+0.054
led+0.050
P+0.050
Neg logit contributions
ine-0.033
 birthday-0.030
 Jupiter-0.029
 Office-0.028
 Today-0.028
Token
Token ID129
Feature activation-9.10e-03

Pos logit contributions
Run+0.036
ls+0.035
mun+0.034
P+0.031
Hey+0.030
Neg logit contributions
ine-0.039
 birthday-0.037
 Office-0.034
hip-0.034
 X-0.033
Token
Token ID241
Feature activation-3.67e-02

Pos logit contributions
Run+0.015
ls+0.015
mun+0.014
 distracted+0.013
P+0.013
Neg logit contributions
 birthday-0.015
ine-0.013
 moon-0.013
 X-0.013
 astronaut-0.013
TokenWhat
Token ID2061
Feature activation-1.80e-02

Pos logit contributions
ls+0.056
mun+0.052
Run+0.049
led+0.046
P+0.044
Neg logit contributions
ine-0.070
 birthday-0.066
 Office-0.060
 Today-0.060
hip-0.060
Tokenâ
Token ID22940
Feature activation-2.12e-02

Pos logit contributions
Run+0.038
ls+0.037
mun+0.035
awed+0.034
P+0.034
Neg logit contributions
 birthday-0.023
ine-0.022
 desk-0.019
 Office-0.019
hip-0.018
Token
Token ID26391
Feature activation-2.60e-02

Pos logit contributions
ls+0.041
Run+0.039
mun+0.038
 bitten+0.035
 distracted+0.034
Neg logit contributions
ine-0.031
 birthday-0.029
 Office-0.028
hip-0.027
 X-0.026
Token
Token ID8151
Feature activation-5.47e-02

Pos logit contributions
Run+0.045
ls+0.044
mun+0.043
P+0.038
Hey+0.038
Neg logit contributions
ine-0.043
 birthday-0.040
 Office-0.037
hip-0.036
 Jupiter-0.036
Tokens
Token ID82
Feature activation-3.21e-03

Pos logit contributions
Run+0.096
mun+0.084
 bitten+0.079
Help+0.078
Sorry+0.078
Neg logit contributions
ine-0.086
 Today-0.078
 birthday-0.075
 for-0.071
 desk-0.070
Token wrong
Token ID2642
Feature activation-2.54e-03

Pos logit contributions
Run+0.005
mun+0.004
ls+0.004
Hey+0.004
P+0.004
Neg logit contributions
ine-0.006
 birthday-0.006
 Office-0.006
hip-0.006
 Jupiter-0.005
Token,
Token ID11
Feature activation-1.23e-02

Pos logit contributions
ls+0.004
Run+0.004
mun+0.004
P+0.003
Hey+0.003
Neg logit contributions
ine-0.005
 birthday-0.004
hip-0.004
 X-0.004
 Jupiter-0.004
Token Jane
Token ID12091
Feature activation-1.42e-02

Pos logit contributions
ls+0.023
Run+0.023
mun+0.023
P+0.021
 bitten+0.020
Neg logit contributions
ine-0.018
 birthday-0.017
 Office-0.015
 X-0.015
hip-0.015
Token?
Token ID30
Feature activation-1.79e-02

Pos logit contributions
Run+0.025
ls+0.024
mun+0.024
P+0.022
Hey+0.021
Neg logit contributions
ine-0.023
 birthday-0.021
 Office-0.020
hip-0.019
 Jupiter-0.019
Token
Token ID198
Feature activation+1.46e-02

Pos logit contributions
ine+0.003
 birthday+0.002
 Office+0.002
hip+0.002
 Jupiter+0.002
Neg logit contributions
ls-0.003
Run-0.003
mun-0.003
P-0.003
led-0.003
Token
Token ID198
Feature activation+4.50e-03

Pos logit contributions
ine+0.027
 birthday+0.025
hip+0.024
 Office+0.024
 X+0.024
Neg logit contributions
ls-0.022
Run-0.022
mun-0.021
P-0.019
Hey-0.018
TokenThe
Token ID464
Feature activation-2.38e-03

Pos logit contributions
ine+0.007
 birthday+0.007
 X+0.006
 eye+0.006
hip+0.006
Neg logit contributions
Run-0.008
ls-0.008
mun-0.008
P-0.007
Hey-0.007
Token two
Token ID734
Feature activation-1.93e-02

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
Hey+0.003
P+0.003
Neg logit contributions
ine-0.005
hip-0.005
artments-0.004
uries-0.004
rap-0.004
Token became
Token ID2627
Feature activation-1.11e-02

Pos logit contributions
Run+0.031
ls+0.030
mun+0.028
P+0.026
 bitten+0.026
Neg logit contributions
ine-0.033
 birthday-0.032
 X-0.030
hip-0.029
 Office-0.029
Token insepar
Token ID50140
Feature activation-5.46e-02

Pos logit contributions
Run+0.023
ls+0.022
mun+0.021
awed+0.021
led+0.020
Neg logit contributions
ine-0.014
 birthday-0.014
 Jupiter-0.012
 Office-0.012
 X-0.012
Tokenable
Token ID540
Feature activation-2.08e-02

Pos logit contributions
Run+0.070
 bitten+0.058
ls+0.055
mun+0.055
awed+0.053
Neg logit contributions
ine-0.109
 birthday-0.101
 end-0.101
 eye-0.097
hip-0.097
Token and
Token ID290
Feature activation-6.79e-03

Pos logit contributions
Run+0.038
ls+0.037
mun+0.036
P+0.034
Hey+0.033
Neg logit contributions
ine-0.032
 birthday-0.031
 X-0.028
 Office-0.028
hip-0.027
Token each
Token ID1123
Feature activation-1.77e-02

Pos logit contributions
Run+0.011
ls+0.011
mun+0.010
P+0.009
led+0.009
Neg logit contributions
ine-0.012
 birthday-0.011
hip-0.011
 Office-0.011
rap-0.011
Token day
Token ID1110
Feature activation-1.11e-02

Pos logit contributions
ls+0.030
Run+0.030
mun+0.029
P+0.026
awed+0.025
Neg logit contributions
ine-0.030
 birthday-0.027
 Office-0.026
hip-0.026
 X-0.026
Token they
Token ID484
Feature activation-2.53e-03

Pos logit contributions
Run+0.018
ls+0.018
mun+0.017
P+0.016
Hey+0.016
Neg logit contributions
ine-0.019
 birthday-0.018
 Office-0.017
 Jupiter-0.017
hip-0.016
Token scissors
Token ID41786
Feature activation-2.77e-02

Pos logit contributions
Run+0.005
ls+0.005
mun+0.004
P+0.004
Hey+0.004
Neg logit contributions
ine-0.004
 birthday-0.004
hip-0.004
 Office-0.004
 Jupiter-0.004
Token slipped
Token ID18859
Feature activation-8.58e-03

Pos logit contributions
Run+0.044
ls+0.042
mun+0.041
P+0.040
Hey+0.037
Neg logit contributions
ine-0.050
 birthday-0.044
hip-0.044
 desk-0.042
 pattern-0.042
Token,
Token ID11
Feature activation-2.44e-02

Pos logit contributions
ls+0.014
Run+0.014
mun+0.014
P+0.013
Hey+0.012
Neg logit contributions
ine-0.015
 birthday-0.013
hip-0.013
 X-0.013
 Office-0.012
Token giving
Token ID3501
Feature activation-1.31e-02

Pos logit contributions
ls+0.052
Run+0.050
mun+0.050
P+0.048
Hey+0.045
Neg logit contributions
ine-0.031
 birthday-0.026
 X-0.025
hip-0.025
 desk-0.024
Token the
Token ID262
Feature activation-7.05e-03

Pos logit contributions
Run+0.024
ls+0.024
mun+0.023
P+0.021
Hey+0.020
Neg logit contributions
ine-0.020
 birthday-0.018
 X-0.017
hip-0.017
 Office-0.017
Token boy
Token ID2933
Feature activation-5.46e-02

Pos logit contributions
Run+0.012
ls+0.012
mun+0.011
P+0.010
Hey+0.010
Neg logit contributions
ine-0.012
 birthday-0.011
hip-0.011
 Office-0.011
 Jupiter-0.010
Token a
Token ID257
Feature activation-2.49e-02

Pos logit contributions
Run+0.109
ls+0.108
mun+0.104
Hey+0.095
P+0.095
Neg logit contributions
ine-0.077
 birthday-0.070
 Office-0.064
hip-0.063
 X-0.062
Token bad
Token ID2089
Feature activation-1.02e-02

Pos logit contributions
Run+0.048
ls+0.048
mun+0.046
P+0.042
led+0.041
Neg logit contributions
ine-0.036
 birthday-0.034
 X-0.030
 Office-0.030
hip-0.030
Token cut
Token ID2005
Feature activation-1.40e-02

Pos logit contributions
Run+0.019
ls+0.019
mun+0.018
 bitten+0.017
P+0.017
Neg logit contributions
ine-0.014
 Jupiter-0.013
 Office-0.013
 birthday-0.013
hip-0.013
Token on
Token ID319
Feature activation-8.34e-03

Pos logit contributions
ls+0.021
Run+0.021
mun+0.020
P+0.018
 bitten+0.018
Neg logit contributions
ine-0.025
 birthday-0.024
 Office-0.023
 Jupiter-0.023
hip-0.022
Token the
Token ID262
Feature activation+3.81e-03

Pos logit contributions
Run+0.011
ls+0.010
mun+0.010
 bitten+0.009
Hey+0.009
Neg logit contributions
ine-0.016
 birthday-0.016
 Jupiter-0.015
 Office-0.015
hip-0.015
Token got
Token ID1392
Feature activation-4.12e-02

Pos logit contributions
ls+0.033
mun+0.031
Run+0.031
P+0.027
aro+0.026
Neg logit contributions
ine-0.033
 sleep-0.030
 X-0.030
 birthday-0.030
 eye-0.030
Token hurt
Token ID5938
Feature activation-7.16e-03

Pos logit contributions
mun+0.081
Run+0.076
ls+0.070
+0.069
awed+0.068
Neg logit contributions
 for-0.054
 birthday-0.052
ine-0.051
 sleep-0.049
 Grand-0.045
Token.
Token ID13
Feature activation-5.13e-03

Pos logit contributions
mun+0.014
ls+0.014
Run+0.014
P+0.012
awed+0.012
Neg logit contributions
ine-0.009
 birthday-0.009
hip-0.008
 end-0.008
 X-0.008
Token You
Token ID921
Feature activation-1.33e-02

Pos logit contributions
ls+0.009
mun+0.009
Run+0.009
led+0.008
P+0.008
Neg logit contributions
ine-0.008
 birthday-0.008
 Office-0.007
hip-0.007
 X-0.007
Token made
Token ID925
Feature activation+3.86e-03

Pos logit contributions
ls+0.028
Run+0.027
mun+0.027
awed+0.026
P+0.026
Neg logit contributions
ine-0.015
 eye-0.013
 believe-0.013
 sleep-0.013
 imagine-0.013
Token us
Token ID514
Feature activation-5.45e-02

Pos logit contributions
ine+0.006
 birthday+0.006
 Office+0.006
hip+0.006
 X+0.006
Neg logit contributions
Run-0.007
ls-0.007
mun-0.006
Hey-0.006
 bitten-0.006
Token worry
Token ID5490
Feature activation-1.03e-02

Pos logit contributions
Run+0.097
ls+0.094
mun+0.093
P+0.088
Hey+0.087
Neg logit contributions
 birthday-0.087
ine-0.085
hip-0.076
 eye-0.075
 end-0.073
Token."
Token ID526
Feature activation-1.45e-02

Pos logit contributions
ls+0.017
Run+0.016
mun+0.015
P+0.014
Hey+0.014
Neg logit contributions
ine-0.018
 birthday-0.017
 X-0.016
 Office-0.016
hip-0.016
Token
Token ID198
Feature activation+4.18e-04

Pos logit contributions
mun+0.028
ls+0.028
Run+0.028
led+0.027
 bitten+0.026
Neg logit contributions
ine-0.019
 birthday-0.018
 Today-0.017
 X-0.016
 Jupiter-0.016
Token
Token ID198
Feature activation-7.81e-03

Pos logit contributions
ine+0.001
 birthday+0.001
hip+0.001
 Office+0.001
 X+0.001
Neg logit contributions
Run-0.001
ls-0.001
mun-0.001
P-0.001
Hey-0.001
Token"
Token ID1
Feature activation-3.43e-02

Pos logit contributions
ls+0.013
mun+0.012
Run+0.012
led+0.011
 bitten+0.011
Neg logit contributions
ine-0.013
hip-0.012
 birthday-0.012
 Office-0.011
 X-0.011
Token't
Token ID470
Feature activation-1.06e-02

Pos logit contributions
Run+0.013
ls+0.013
mun+0.013
led+0.011
P+0.011
Neg logit contributions
ine-0.020
 birthday-0.018
hip-0.018
 Office-0.017
rap-0.017
Token leave
Token ID2666
Feature activation-1.72e-02

Pos logit contributions
ls+0.024
Run+0.023
mun+0.023
led+0.021
P+0.021
Neg logit contributions
ine-0.012
 birthday-0.011
hip-0.011
 X-0.010
 Office-0.010
Token our
Token ID674
Feature activation-2.60e-03

Pos logit contributions
Run+0.029
ls+0.029
mun+0.028
P+0.025
Hey+0.025
Neg logit contributions
ine-0.028
 birthday-0.027
hip-0.025
 Office-0.025
 X-0.025
Token homes
Token ID5682
Feature activation-1.77e-02

Pos logit contributions
ls+0.005
Run+0.005
mun+0.005
P+0.004
 bitten+0.004
Neg logit contributions
ine-0.004
 birthday-0.004
hip-0.004
 Office-0.003
 Jupiter-0.003
Token for
Token ID329
Feature activation-6.86e-03

Pos logit contributions
ls+0.031
Run+0.030
mun+0.029
P+0.027
 bitten+0.026
Neg logit contributions
ine-0.028
 birthday-0.026
 Office-0.025
hip-0.025
 X-0.025
Token anyone
Token ID2687
Feature activation-5.43e-02

Pos logit contributions
Run+0.011
ls+0.011
mun+0.010
P+0.009
 bitten+0.009
Neg logit contributions
ine-0.012
 birthday-0.011
 Office-0.011
hip-0.011
 X-0.011
Token,
Token ID11
Feature activation-1.24e-02

Pos logit contributions
Run+0.108
mun+0.101
P+0.098
ls+0.098
Hey+0.094
Neg logit contributions
 birthday-0.077
ine-0.073
hip-0.064
 sleep-0.063
 believe-0.063
Token not
Token ID407
Feature activation+6.88e-03

Pos logit contributions
ls+0.023
Run+0.023
mun+0.022
P+0.020
Hey+0.020
Neg logit contributions
ine-0.019
 birthday-0.017
hip-0.016
 Office-0.016
 X-0.016
Token even
Token ID772
Feature activation-2.28e-02

Pos logit contributions
ine+0.010
 birthday+0.009
hip+0.008
 Office+0.008
 X+0.008
Neg logit contributions
Run-0.013
ls-0.013
mun-0.013
P-0.012
Hey-0.012
Token a
Token ID257
Feature activation-2.32e-03

Pos logit contributions
Run+0.042
ls+0.041
mun+0.040
P+0.038
awed+0.037
Neg logit contributions
ine-0.033
 birthday-0.033
 X-0.029
hip-0.029
 Office-0.029
Token mighty
Token ID18680
Feature activation+1.11e-02

Pos logit contributions
Run+0.004
ls+0.004
mun+0.004
P+0.004
Hey+0.003
Neg logit contributions
ine-0.004
 birthday-0.003
hip-0.003
 Office-0.003
 X-0.003
Token it
Token ID340
Feature activation-1.51e-02

Pos logit contributions
ls+0.027
Run+0.026
mun+0.025
 bitten+0.023
P+0.023
Neg logit contributions
ine-0.030
 birthday-0.027
 Office-0.027
hip-0.026
 Jupiter-0.026
Token,
Token ID11
Feature activation-2.24e-02

Pos logit contributions
Run+0.027
ls+0.027
mun+0.026
P+0.024
Hey+0.024
Neg logit contributions
ine-0.023
 birthday-0.022
hip-0.020
 Office-0.020
 X-0.020
Token so
Token ID523
Feature activation-1.69e-02

Pos logit contributions
ls+0.041
Run+0.041
mun+0.040
P+0.036
Hey+0.035
Neg logit contributions
ine-0.034
 birthday-0.031
hip-0.029
 Office-0.029
 X-0.029
Token Jill
Token ID22772
Feature activation-1.77e-02

Pos logit contributions
Run+0.020
ls+0.020
mun+0.019
P+0.017
Hey+0.016
Neg logit contributions
ine-0.036
 birthday-0.035
hip-0.033
 X-0.033
 Office-0.032
Token got
Token ID1392
Feature activation-4.89e-02

Pos logit contributions
Run+0.034
ls+0.032
mun+0.032
P+0.028
Hey+0.028
Neg logit contributions
ine-0.029
 birthday-0.022
 Office-0.022
 desk-0.022
 eye-0.022
Token very
Token ID845
Feature activation-5.37e-02

Pos logit contributions
Run+0.103
mun+0.098
ls+0.092
awed+0.087
Hey+0.087
Neg logit contributions
 for-0.062
 birthday-0.058
ine-0.058
 end-0.055
 eye-0.053
Token mad
Token ID8805
Feature activation-2.71e-02

Pos logit contributions
Run+0.098
mun+0.094
Hey+0.092
P+0.081
rob+0.080
Neg logit contributions
 birthday-0.081
 eye-0.074
ine-0.074
 for-0.073
 end-0.070
Token.
Token ID13
Feature activation-2.17e-02

Pos logit contributions
Run+0.050
ls+0.050
mun+0.048
P+0.044
Hey+0.043
Neg logit contributions
ine-0.041
 birthday-0.038
hip-0.035
 Office-0.035
 X-0.035
Token
Token ID220
Feature activation-2.15e-02

Pos logit contributions
ls+0.040
Run+0.040
mun+0.038
P+0.035
 bitten+0.034
Neg logit contributions
ine-0.033
 birthday-0.031
hip-0.029
 Office-0.029
 Jupiter-0.029
Token
Token ID198
Feature activation-1.43e-02

Pos logit contributions
Run+0.041
ls+0.040
mun+0.038
P+0.035
led+0.035
Neg logit contributions
ine-0.032
 birthday-0.029
hip-0.028
 Office-0.027
rap-0.027
TokenHer
Token ID9360
Feature activation-4.30e-03

Pos logit contributions
ls+0.025
Run+0.024
mun+0.023
 bitten+0.021
led+0.021
Neg logit contributions
ine-0.024
 birthday-0.022
hip-0.021
 Office-0.020
 Jupiter-0.020
Token then
Token ID788
Feature activation-1.02e-02

Pos logit contributions
ine+0.024
hip+0.022
 X+0.022
 Office+0.022
 birthday+0.022
Neg logit contributions
Run-0.021
ls-0.021
mun-0.020
 distracted-0.018
Hey-0.018
Token Lily
Token ID20037
Feature activation+3.05e-03

Pos logit contributions
ls+0.016
Run+0.015
mun+0.015
 distracted+0.013
P+0.013
Neg logit contributions
ine-0.018
hip-0.018
 birthday-0.017
 Office-0.017
 X-0.017
Token's
Token ID338
Feature activation+2.20e-02

Pos logit contributions
ine+0.005
 birthday+0.005
hip+0.005
 Office+0.005
 X+0.005
Neg logit contributions
ls-0.005
Run-0.005
mun-0.004
P-0.004
 bitten-0.004
Token grandma
Token ID49890
Feature activation+9.26e-04

Pos logit contributions
ine+0.028
 Jupiter+0.024
rap+0.024
hip+0.023
 Office+0.022
Neg logit contributions
ls-0.043
Run-0.043
mun-0.042
 bitten-0.040
 distracted-0.040
Token got
Token ID1392
Feature activation-3.84e-02

Pos logit contributions
ine+0.002
 birthday+0.002
hip+0.002
 X+0.002
 Office+0.002
Neg logit contributions
ls-0.001
Run-0.001
mun-0.001
P-0.001
 bitten-0.001
Token very
Token ID845
Feature activation-5.37e-02

Pos logit contributions
Run+0.077
ls+0.072
mun+0.067
Hey+0.065
awed+0.064
Neg logit contributions
ine-0.056
 birthday-0.051
 for-0.047
 end-0.045
 eye-0.044
Token tired
Token ID10032
Feature activation-1.13e-02

Pos logit contributions
Run+0.095
mun+0.087
Hey+0.087
aro+0.080
awed+0.078
Neg logit contributions
 birthday-0.090
ine-0.085
 eye-0.078
 sleep-0.074
 for-0.072
Token and
Token ID290
Feature activation-3.58e-02

Pos logit contributions
ls+0.018
Run+0.018
mun+0.017
P+0.015
 bitten+0.015
Neg logit contributions
ine-0.019
 birthday-0.018
 Office-0.018
 Jupiter-0.017
 X-0.017
Token fell
Token ID3214
Feature activation-1.87e-02

Pos logit contributions
Run+0.056
ls+0.055
mun+0.054
P+0.052
Hey+0.049
Neg logit contributions
ine-0.062
 birthday-0.061
 sleep-0.055
 desk-0.054
 Jupiter-0.053
Token asleep
Token ID16039
Feature activation-7.10e-03

Pos logit contributions
Run+0.036
mun+0.035
ls+0.034
P+0.032
Hey+0.031
Neg logit contributions
ine-0.028
 birthday-0.026
 Jupiter-0.024
hip-0.023
 moon-0.023
Token.
Token ID13
Feature activation+3.65e-03

Pos logit contributions
Run+0.012
ls+0.012
mun+0.011
P+0.010
Hey+0.010
Neg logit contributions
ine-0.012
 birthday-0.011
 Office-0.010
hip-0.010
 X-0.010
Token so
Token ID523
Feature activation-2.74e-02

Pos logit contributions
ls+0.016
Run+0.016
mun+0.015
P+0.014
Hey+0.013
Neg logit contributions
ine-0.017
 birthday-0.016
 Office-0.015
hip-0.015
 X-0.015
Token sad
Token ID6507
Feature activation-6.55e-03

Pos logit contributions
Run+0.043
ls+0.042
mun+0.042
P+0.037
Hey+0.036
Neg logit contributions
ine-0.049
 birthday-0.048
 eye-0.042
hip-0.042
 end-0.041
Token that
Token ID326
Feature activation+1.23e-03

Pos logit contributions
ls+0.012
Run+0.012
mun+0.011
P+0.010
Hey+0.010
Neg logit contributions
ine-0.010
 birthday-0.009
 Office-0.009
hip-0.009
 X-0.009
Token she
Token ID673
Feature activation-1.63e-02

Pos logit contributions
ine+0.003
hip+0.003
 X+0.003
 Office+0.003
rap+0.003
Neg logit contributions
ls-0.001
Run-0.001
mun-0.001
 distracted-0.001
Hey-0.001
Token never
Token ID1239
Feature activation-3.26e-03

Pos logit contributions
Run+0.027
ls+0.027
mun+0.027
P+0.024
Hey+0.023
Neg logit contributions
ine-0.028
 birthday-0.023
 Office-0.023
 desk-0.022
 eye-0.022
Token got
Token ID1392
Feature activation-5.36e-02

Pos logit contributions
ls+0.005
Run+0.005
mun+0.004
P+0.004
Hey+0.004
Neg logit contributions
ine-0.006
 birthday-0.006
hip-0.006
 Office-0.006
 X-0.005
Token her
Token ID607
Feature activation-1.41e-03

Pos logit contributions
Run+0.124
mun+0.115
ls+0.105
Sorry+0.103
Hey+0.103
Neg logit contributions
ine-0.064
 birthday-0.057
 for-0.054
 sleep-0.049
 aunt-0.049
Token skirt
Token ID23967
Feature activation-2.20e-02

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
P+0.003
Hey+0.003
Neg logit contributions
ine-0.001
 birthday-0.001
hip-0.001
 Office-0.001
 X-0.001
Token back
Token ID736
Feature activation-1.72e-02

Pos logit contributions
Run+0.042
ls+0.042
mun+0.041
P+0.038
Hey+0.037
Neg logit contributions
ine-0.031
 birthday-0.029
hip-0.027
 Office-0.027
 X-0.026
Token.
Token ID13
Feature activation-2.21e-03

Pos logit contributions
Run+0.035
mun+0.033
ls+0.033
P+0.031
Hey+0.030
Neg logit contributions
ine-0.025
 birthday-0.023
hip-0.021
 eye-0.020
 desk-0.020
Token Lily
Token ID20037
Feature activation-9.38e-03

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
Hey+0.003
 bitten+0.003
Neg logit contributions
ine-0.004
 X-0.003
hip-0.003
 birthday-0.003
rap-0.003
Token the
Token ID262
Feature activation-6.77e-03

Pos logit contributions
Run+0.022
ls+0.021
mun+0.021
P+0.019
awed+0.018
Neg logit contributions
ine-0.017
 birthday-0.016
 X-0.016
hip-0.015
 Office-0.015
Token toy
Token ID13373
Feature activation-7.93e-03

Pos logit contributions
Run+0.016
mun+0.016
ls+0.016
P+0.015
awed+0.014
Neg logit contributions
 birthday-0.007
ine-0.006
 desk-0.006
 eye-0.006
 X-0.006
Token away
Token ID1497
Feature activation-2.74e-03

Pos logit contributions
Run+0.016
mun+0.016
ls+0.015
P+0.014
Hey+0.014
Neg logit contributions
ine-0.010
 birthday-0.010
 X-0.009
hip-0.009
 Office-0.009
Token and
Token ID290
Feature activation-2.57e-02

Pos logit contributions
Run+0.005
ls+0.005
mun+0.005
P+0.005
Hey+0.005
Neg logit contributions
ine-0.004
 birthday-0.003
hip-0.003
 Office-0.003
 X-0.003
Token made
Token ID925
Feature activation-4.41e-04

Pos logit contributions
ls+0.045
Run+0.044
mun+0.043
P+0.041
Hey+0.038
Neg logit contributions
ine-0.040
 birthday-0.037
 desk-0.036
 X-0.036
 Office-0.035
Token him
Token ID683
Feature activation-5.36e-02

Pos logit contributions
Run+0.001
ls+0.001
mun+0.001
P+0.001
Hey+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
hip-0.001
 Office-0.001
 X-0.001
Token go
Token ID467
Feature activation-1.29e-02

Pos logit contributions
Run+0.126
ls+0.121
mun+0.120
Hey+0.117
P+0.117
Neg logit contributions
 birthday-0.062
ine-0.056
 sleep-0.053
 eye-0.053
 desk-0.050
Token back
Token ID736
Feature activation-8.89e-03

Pos logit contributions
Run+0.031
mun+0.029
ls+0.029
Hey+0.028
P+0.027
Neg logit contributions
ine-0.014
 birthday-0.013
 desk-0.011
 eye-0.011
hip-0.011
Token to
Token ID284
Feature activation-1.77e-02

Pos logit contributions
Run+0.020
ls+0.020
mun+0.019
P+0.018
Hey+0.018
Neg logit contributions
ine-0.010
 birthday-0.009
 Office-0.008
hip-0.008
 Jupiter-0.008
Token the
Token ID262
Feature activation-9.07e-03

Pos logit contributions
Run+0.027
ls+0.027
mun+0.026
 bitten+0.024
P+0.023
Neg logit contributions
ine-0.032
 birthday-0.029
hip-0.028
 Office-0.028
 X-0.027
Token store
Token ID3650
Feature activation-8.18e-03

Pos logit contributions
ls+0.017
Run+0.017
mun+0.016
P+0.015
Hey+0.014
Neg logit contributions
ine-0.014
 birthday-0.013
hip-0.012
 Office-0.012
 X-0.012
Token day
Token ID1110
Feature activation-9.36e-03

Pos logit contributions
ine+0.017
hip+0.015
 birthday+0.015
 Jupiter+0.014
 Office+0.014
Neg logit contributions
Run-0.029
ls-0.028
mun-0.027
P-0.026
Hey-0.025
Token,
Token ID11
Feature activation-7.00e-03

Pos logit contributions
Run+0.019
ls+0.019
mun+0.018
P+0.017
Hey+0.017
Neg logit contributions
ine-0.012
 birthday-0.011
 Office-0.010
hip-0.010
 X-0.010
Token the
Token ID262
Feature activation-4.21e-03

Pos logit contributions
ls+0.014
Run+0.013
mun+0.013
P+0.012
 bitten+0.012
Neg logit contributions
 birthday-0.010
ine-0.009
 desk-0.008
 moon-0.008
 Jupiter-0.008
Token puppy
Token ID26188
Feature activation-1.95e-02

Pos logit contributions
ls+0.008
Run+0.008
mun+0.008
P+0.007
 bitten+0.007
Neg logit contributions
 birthday-0.006
ine-0.006
hip-0.006
 X-0.005
 Office-0.005
Token got
Token ID1392
Feature activation-4.63e-02

Pos logit contributions
Run+0.032
mun+0.031
ls+0.031
aro+0.029
P+0.028
Neg logit contributions
ine-0.033
 eye-0.027
 Office-0.027
 desk-0.027
 end-0.027
Token very
Token ID845
Feature activation-5.36e-02

Pos logit contributions
Run+0.087
mun+0.086
ls+0.075
Hey+0.075
Sorry+0.072
Neg logit contributions
ine-0.068
 for-0.062
 end-0.058
 eye-0.056
 even-0.055
Token sick
Token ID6639
Feature activation-2.07e-02

Pos logit contributions
mun+0.104
Run+0.099
Hey+0.095
+0.083
rob+0.081
Neg logit contributions
 birthday-0.075
ine-0.073
 eye-0.070
 sleep-0.066
 for-0.064
Token and
Token ID290
Feature activation-4.09e-02

Pos logit contributions
Run+0.038
ls+0.038
mun+0.037
P+0.034
Hey+0.033
Neg logit contributions
ine-0.031
 birthday-0.029
hip-0.027
 Office-0.027
 X-0.027
Token had
Token ID550
Feature activation-3.61e-02

Pos logit contributions
mun+0.071
Run+0.070
P+0.066
ls+0.064
Hey+0.060
Neg logit contributions
ine-0.064
 birthday-0.058
 sleep-0.055
 desk-0.055
 pattern-0.054
Token to
Token ID284
Feature activation-2.91e-02

Pos logit contributions
Run+0.079
mun+0.078
ls+0.077
led+0.069
P+0.068
Neg logit contributions
ine-0.044
 birthday-0.044
 moon-0.035
 pattern-0.035
 desk-0.034
Token go
Token ID467
Feature activation-1.83e-02

Pos logit contributions
Run+0.062
ls+0.061
mun+0.059
P+0.056
 bitten+0.055
Neg logit contributions
ine-0.034
 birthday-0.030
 Office-0.029
hip-0.028
 Jupiter-0.028
TokenUnfortunately
Token ID13898
Feature activation+1.64e-02

Pos logit contributions
ls+0.005
Run+0.005
mun+0.004
P+0.004
 bitten+0.004
Neg logit contributions
ine-0.005
 birthday-0.005
hip-0.005
 Office-0.005
 X-0.004
Token,
Token ID11
Feature activation-5.17e-03

Pos logit contributions
ine+0.025
 birthday+0.024
hip+0.022
 Office+0.022
 X+0.021
Neg logit contributions
ls-0.030
Run-0.030
mun-0.028
led-0.026
P-0.025
Token the
Token ID262
Feature activation-2.46e-04

Pos logit contributions
ls+0.009
Run+0.009
mun+0.009
P+0.008
Hey+0.008
Neg logit contributions
ine-0.008
 birthday-0.008
 Office-0.007
 Jupiter-0.007
 X-0.007
Token infant
Token ID11212
Feature activation-3.13e-03

Pos logit contributions
ls+0.000
Run+0.000
mun+0.000
P+0.000
awed+0.000
Neg logit contributions
 birthday-0.000
 moon-0.000
ine-0.000
 end-0.000
 pattern-0.000
Token never
Token ID1239
Feature activation-7.96e-03

Pos logit contributions
ls+0.005
Run+0.005
mun+0.005
P+0.004
Hey+0.004
Neg logit contributions
ine-0.005
 birthday-0.005
 Office-0.005
hip-0.005
 X-0.005
Token got
Token ID1392
Feature activation-5.35e-02

Pos logit contributions
ls+0.012
Run+0.012
mun+0.012
P+0.011
Hey+0.010
Neg logit contributions
ine-0.014
 birthday-0.013
 eye-0.012
hip-0.012
 Office-0.012
Token to
Token ID284
Feature activation-3.30e-02

Pos logit contributions
Run+0.125
mun+0.122
Hey+0.110
P+0.108
Sorry+0.106
Neg logit contributions
 for-0.056
ine-0.056
 birthday-0.053
 sleep-0.048
 end-0.044
Token see
Token ID766
Feature activation-9.99e-03

Pos logit contributions
ls+0.068
Run+0.067
mun+0.065
P+0.061
led+0.060
Neg logit contributions
ine-0.044
 birthday-0.042
 X-0.037
hip-0.037
 Jupiter-0.037
Token his
Token ID465
Feature activation-4.07e-03

Pos logit contributions
Run+0.018
ls+0.018
mun+0.017
P+0.016
Hey+0.015
Neg logit contributions
ine-0.016
 birthday-0.015
 Office-0.014
hip-0.014
 Jupiter-0.014
Token friend
Token ID1545
Feature activation-1.60e-02

Pos logit contributions
Run+0.009
ls+0.009
mun+0.008
P+0.008
Hey+0.008
Neg logit contributions
ine-0.005
 birthday-0.005
 Office-0.004
hip-0.004
 X-0.004
Token again
Token ID757
Feature activation-1.10e-02

Pos logit contributions
ls+0.026
Run+0.026
mun+0.025
 bitten+0.022
Hey+0.022
Neg logit contributions
ine-0.027
 birthday-0.025
 Office-0.024
hip-0.024
 X-0.024
Token and
Token ID290
Feature activation-2.50e-02

Pos logit contributions
ine+0.004
 X+0.004
 Office+0.004
hip+0.004
 Jupiter+0.004
Neg logit contributions
ls-0.003
Run-0.002
mun-0.002
led-0.002
 bitten-0.002
Token gave
Token ID2921
Feature activation-1.23e-02

Pos logit contributions
ls+0.048
mun+0.047
Run+0.047
P+0.044
awed+0.043
Neg logit contributions
 birthday-0.034
 for-0.033
 eye-0.031
 be-0.030
 end-0.030
Token back
Token ID736
Feature activation-2.24e-02

Pos logit contributions
Run+0.026
ls+0.025
mun+0.025
P+0.022
awed+0.022
Neg logit contributions
ine-0.015
 birthday-0.015
 end-0.012
hip-0.012
 Office-0.012
Token the
Token ID262
Feature activation-4.21e-04

Pos logit contributions
Run+0.047
mun+0.044
ls+0.044
awed+0.040
led+0.040
Neg logit contributions
ine-0.029
 birthday-0.028
 Office-0.025
 Jupiter-0.025
 X-0.024
Token shiny
Token ID22441
Feature activation+1.17e-02

Pos logit contributions
ls+0.001
Run+0.001
mun+0.001
P+0.001
Hey+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
hip-0.001
 Office-0.001
 X-0.001
Token things
Token ID1243
Feature activation-5.35e-02

Pos logit contributions
ine+0.019
hip+0.017
 birthday+0.017
 Office+0.016
 Jupiter+0.016
Neg logit contributions
ls-0.021
Run-0.019
mun-0.019
led-0.018
P-0.017
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
mun+0.126
Run+0.120
ls+0.110
awed+0.109
P+0.109
Neg logit contributions
 birthday-0.060
 for-0.055
 end-0.052
hip-0.047
 eye-0.046
Token
Token ID198
Feature activation+8.11e-03

Pos logit contributions
ls+0.024
Run+0.023
mun+0.023
P+0.021
led+0.021
Neg logit contributions
ine-0.017
 birthday-0.016
 Office-0.014
 Jupiter-0.014
hip-0.014
Token
Token ID198
Feature activation-5.32e-03

Pos logit contributions
ine+0.017
 birthday+0.016
 Office+0.015
hip+0.015
 X+0.015
Neg logit contributions
ls-0.010
Run-0.010
mun-0.009
P-0.009
 bitten-0.008
TokenThey
Token ID2990
Feature activation-1.06e-02

Pos logit contributions
ls+0.009
Run+0.009
mun+0.008
P+0.007
 bitten+0.007
Neg logit contributions
ine-0.009
 birthday-0.009
 Office-0.008
hip-0.008
 X-0.008
Token hugged
Token ID41130
Feature activation-6.96e-03

Pos logit contributions
ls+0.019
Run+0.019
mun+0.018
aro+0.016
 bitten+0.016
Neg logit contributions
ine-0.017
 eye-0.015
 birthday-0.015
 end-0.015
 X-0.015
Token me
Token ID502
Feature activation-7.49e-03

Pos logit contributions
ine+0.003
 birthday+0.002
hip+0.002
 Office+0.002
 X+0.002
Neg logit contributions
Run-0.002
ls-0.002
mun-0.002
P-0.002
Hey-0.002
Token!
Token ID0
Feature activation+1.23e-03

Pos logit contributions
ls+0.011
Run+0.010
mun+0.010
P+0.009
 bitten+0.009
Neg logit contributions
ine-0.013
 birthday-0.013
 X-0.013
 Jupiter-0.012
 Office-0.012
Token Aren
Token ID9843
Feature activation+6.95e-03

Pos logit contributions
ine+0.002
 birthday+0.002
hip+0.002
 Office+0.002
 X+0.002
Neg logit contributions
Run-0.002
ls-0.002
mun-0.002
P-0.002
Hey-0.002
Tokenâ
Token ID22940
Feature activation-2.16e-02

Pos logit contributions
ine+0.013
 birthday+0.013
hip+0.012
 Office+0.012
rap+0.012
Neg logit contributions
ls-0.010
Run-0.010
mun-0.009
led-0.008
 bitten-0.008
Token
Token ID26391
Feature activation-2.23e-02

Pos logit contributions
ls+0.039
Run+0.037
mun+0.036
 bitten+0.033
awed+0.032
Neg logit contributions
ine-0.035
 birthday-0.032
hip-0.031
 Office-0.030
 X-0.029
Token
Token ID8151
Feature activation-5.32e-02

Pos logit contributions
ls+0.036
Run+0.036
mun+0.034
P+0.031
Hey+0.030
Neg logit contributions
ine-0.039
 birthday-0.036
hip-0.034
 Office-0.034
 X-0.034
Tokent
Token ID83
Feature activation-1.53e-02

Pos logit contributions
Run+0.082
ls+0.072
mun+0.072
P+0.067
 bitten+0.065
Neg logit contributions
ine-0.096
 birthday-0.089
hip-0.084
 Today-0.081
ary-0.077
Token we
Token ID356
Feature activation-2.76e-02

Pos logit contributions
Run+0.027
ls+0.027
mun+0.026
P+0.024
awed+0.023
Neg logit contributions
ine-0.024
 birthday-0.023
hip-0.021
 Office-0.021
 Jupiter-0.020
Token both
Token ID1111
Feature activation-2.59e-02

Pos logit contributions
Run+0.057
ls+0.056
mun+0.054
awed+0.050
P+0.049
Neg logit contributions
ine-0.037
 birthday-0.036
hip-0.032
 Office-0.030
 X-0.030
Token so
Token ID523
Feature activation-3.93e-02

Pos logit contributions
ls+0.052
Run+0.052
mun+0.050
P+0.046
Hey+0.045
Neg logit contributions
ine-0.035
 birthday-0.032
 Office-0.029
hip-0.029
 X-0.029
Token excited
Token ID6568
Feature activation-9.96e-03

Pos logit contributions
Run+0.075
ls+0.073
mun+0.073
P+0.066
 bitten+0.066
Neg logit contributions
 birthday-0.057
ine-0.055
hip-0.047
 Office-0.046
 X-0.046
Token.
Token ID13
Feature activation-3.32e-03

Pos logit contributions
Run+0.014
mun+0.014
ls+0.013
P+0.013
Hey+0.013
Neg logit contributions
ine-0.011
 birthday-0.010
 eye-0.010
 X-0.009
 pattern-0.009
Token It
Token ID632
Feature activation+3.65e-03

Pos logit contributions
Run+0.006
ls+0.006
mun+0.006
P+0.005
Hey+0.005
Neg logit contributions
ine-0.005
 birthday-0.005
 Office-0.004
hip-0.004
 X-0.004
Token hurt
Token ID5938
Feature activation-3.58e-03

Pos logit contributions
ine+0.006
 pattern+0.005
 birthday+0.005
 Jupiter+0.005
 moon+0.005
Neg logit contributions
Run-0.007
ls-0.006
mun-0.006
P-0.006
Hey-0.006
Token and
Token ID290
Feature activation-2.89e-02

Pos logit contributions
ls+0.006
Run+0.006
mun+0.006
P+0.006
Hey+0.005
Neg logit contributions
ine-0.006
 birthday-0.005
 Office-0.005
hip-0.005
 X-0.005
Token made
Token ID925
Feature activation-3.68e-04

Pos logit contributions
mun+0.046
Run+0.045
ls+0.044
P+0.042
Hey+0.040
Neg logit contributions
ine-0.050
 birthday-0.045
 X-0.043
 pattern-0.043
 Jupiter-0.042
Token him
Token ID683
Feature activation-5.31e-02

Pos logit contributions
ls+0.001
Run+0.001
mun+0.001
P+0.001
 bitten+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
 Office-0.001
hip-0.001
 X-0.001
Token sad
Token ID6507
Feature activation-1.74e-02

Pos logit contributions
Run+0.106
mun+0.104
ls+0.102
Hey+0.100
P+0.099
Neg logit contributions
 birthday-0.074
ine-0.073
 eye-0.069
 sleep-0.066
 pattern-0.063
Token.
Token ID13
Feature activation-1.11e-02

Pos logit contributions
ls+0.027
Run+0.026
mun+0.024
 bitten+0.022
led+0.021
Neg logit contributions
ine-0.032
 birthday-0.030
 Office-0.029
hip-0.029
 X-0.029
Token D
Token ID360
Feature activation-1.67e-02

Pos logit contributions
Run+0.021
ls+0.021
mun+0.021
Hey+0.018
led+0.018
Neg logit contributions
ine-0.016
 birthday-0.015
 Today-0.014
 Jupiter-0.013
 Office-0.013
Tokenucky
Token ID5309
Feature activation-1.37e-02

Pos logit contributions
Run+0.032
ls+0.030
mun+0.030
aro+0.027
 bitten+0.027
Neg logit contributions
ine-0.024
 birthday-0.022
 eye-0.021
hip-0.021
 pattern-0.020
Token started
Token ID2067
Feature activation-2.92e-02

Pos logit contributions
Run+0.025
ls+0.024
mun+0.023
P+0.022
Hey+0.022
Neg logit contributions
ine-0.022
 end-0.018
 eye-0.018
 birthday-0.017
 pattern-0.017
Token but
Token ID475
Feature activation+3.47e-03

Pos logit contributions
ls+0.049
Run+0.049
mun+0.047
P+0.043
Hey+0.042
Neg logit contributions
ine-0.042
 birthday-0.039
hip-0.036
 Office-0.036
 X-0.036
Token Buddy
Token ID36896
Feature activation-2.62e-02

Pos logit contributions
ine+0.006
 birthday+0.006
 Jupiter+0.006
 X+0.006
hip+0.006
Neg logit contributions
ls-0.005
Run-0.005
mun-0.005
P-0.005
awed-0.004
Token just
Token ID655
Feature activation-3.02e-03

Pos logit contributions
Run+0.042
ls+0.042
mun+0.040
P+0.037
Hey+0.036
Neg logit contributions
ine-0.047
 birthday-0.041
 Office-0.040
hip-0.039
 X-0.038
Token loved
Token ID6151
Feature activation-1.06e-02

Pos logit contributions
Run+0.004
ls+0.004
mun+0.004
led+0.003
 bitten+0.003
Neg logit contributions
ine-0.006
 birthday-0.006
hip-0.005
 Office-0.005
 Jupiter-0.005
Token the
Token ID262
Feature activation-2.38e-02

Pos logit contributions
ls+0.017
Run+0.017
mun+0.016
P+0.015
 bitten+0.014
Neg logit contributions
ine-0.018
 birthday-0.017
hip-0.016
 Office-0.016
 X-0.016
Token mud
Token ID17492
Feature activation-5.29e-02

Pos logit contributions
ls+0.050
Run+0.048
mun+0.046
P+0.044
led+0.043
Neg logit contributions
 birthday-0.032
ine-0.031
 pattern-0.029
 eye-0.028
 moon-0.027
Token too
Token ID1165
Feature activation-1.40e-02

Pos logit contributions
Run+0.101
mun+0.098
ls+0.096
P+0.090
Hey+0.088
Neg logit contributions
ine-0.076
 birthday-0.074
 pattern-0.067
hip-0.065
 X-0.065
Token much
Token ID881
Feature activation-2.86e-03

Pos logit contributions
mun+0.028
Run+0.028
ls+0.027
P+0.025
Hey+0.025
Neg logit contributions
ine-0.019
 birthday-0.018
hip-0.016
 Office-0.015
 X-0.015
Token.
Token ID13
Feature activation+1.15e-03

Pos logit contributions
ls+0.004
Run+0.004
mun+0.004
 bitten+0.003
led+0.003
Neg logit contributions
ine-0.006
 birthday-0.005
 Jupiter-0.005
 Office-0.005
hip-0.005
Token
Token ID198
Feature activation+1.83e-02

Pos logit contributions
ine+0.002
 birthday+0.002
 Office+0.002
hip+0.002
 Jupiter+0.002
Neg logit contributions
Run-0.002
ls-0.002
mun-0.002
P-0.002
 bitten-0.002
Token
Token ID198
Feature activation+8.24e-03

Pos logit contributions
ine+0.036
 birthday+0.034
 Office+0.033
 X+0.032
hip+0.032
Neg logit contributions
ls-0.026
Run-0.025
mun-0.024
P-0.022
Hey-0.021
Token she
Token ID673
Feature activation-4.46e-03

Pos logit contributions
Run+0.013
ls+0.013
mun+0.013
P+0.011
Hey+0.011
Neg logit contributions
ine-0.026
 birthday-0.024
hip-0.023
 Office-0.023
 X-0.023
Token decided
Token ID3066
Feature activation-5.15e-03

Pos logit contributions
Run+0.007
mun+0.007
ls+0.007
Hey+0.006
P+0.006
Neg logit contributions
ine-0.008
 Office-0.007
 birthday-0.007
 moon-0.007
 eye-0.006
Token to
Token ID284
Feature activation-1.75e-02

Pos logit contributions
Run+0.012
mun+0.011
ls+0.011
P+0.010
Hey+0.010
Neg logit contributions
 birthday-0.006
ine-0.006
hip-0.005
 desk-0.005
 for-0.005
Token give
Token ID1577
Feature activation-1.40e-02

Pos logit contributions
ls+0.039
Run+0.039
mun+0.037
P+0.035
led+0.034
Neg logit contributions
ine-0.020
 birthday-0.019
hip-0.017
 X-0.017
 Office-0.017
Token the
Token ID262
Feature activation-1.29e-02

Pos logit contributions
Run+0.031
ls+0.029
mun+0.028
P+0.026
awed+0.026
Neg logit contributions
ine-0.016
 birthday-0.015
hip-0.014
 end-0.014
rap-0.013
Token boy
Token ID2933
Feature activation-5.29e-02

Pos logit contributions
ls+0.028
aro+0.027
Run+0.027
P+0.026
mun+0.026
Neg logit contributions
 birthday-0.017
 end-0.016
 moon-0.015
 eye-0.015
 X-0.014
Token one
Token ID530
Feature activation-1.87e-02

Pos logit contributions
Run+0.105
ls+0.103
mun+0.100
P+0.090
Hey+0.090
Neg logit contributions
ine-0.075
 birthday-0.070
hip-0.063
 Office-0.063
 X-0.062
Token of
Token ID286
Feature activation-1.75e-02

Pos logit contributions
Run+0.037
mun+0.036
ls+0.035
P+0.032
Hey+0.032
Neg logit contributions
 birthday-0.026
ine-0.025
 eye-0.023
 end-0.022
 X-0.022
Token her
Token ID607
Feature activation-4.17e-03

Pos logit contributions
Run+0.035
ls+0.035
mun+0.033
led+0.031
P+0.030
Neg logit contributions
ine-0.025
 birthday-0.023
hip-0.021
 X-0.020
rap-0.020
Token own
Token ID898
Feature activation-1.59e-02

Pos logit contributions
Run+0.008
ls+0.008
mun+0.008
Hey+0.007
P+0.007
Neg logit contributions
ine-0.006
 birthday-0.006
 X-0.005
hip-0.005
 Office-0.005
Token.
Token ID13
Feature activation-2.38e-03

Pos logit contributions
Run+0.032
ls+0.030
mun+0.030
Hey+0.028
P+0.028
Neg logit contributions
ine-0.023
 birthday-0.022
 eye-0.019
hip-0.019
 X-0.019
Token!"
Token ID2474
Feature activation-8.60e-03

Pos logit contributions
ine+0.026
 birthday+0.024
hip+0.023
 Office+0.023
uries+0.023
Neg logit contributions
ls-0.021
Run-0.021
mun-0.018
P-0.016
Hey-0.016
Token
Token ID198
Feature activation+1.33e-03

Pos logit contributions
Run+0.017
mun+0.016
ls+0.016
P+0.015
led+0.015
Neg logit contributions
ine-0.012
 birthday-0.012
 X-0.011
 Jupiter-0.011
 Today-0.010
Token
Token ID198
Feature activation-8.37e-03

Pos logit contributions
ine+0.003
 birthday+0.002
 Office+0.002
hip+0.002
 X+0.002
Neg logit contributions
ls-0.002
Run-0.002
mun-0.002
P-0.002
Hey-0.002
Token"
Token ID1
Feature activation-1.12e-02

Pos logit contributions
Run+0.014
ls+0.014
mun+0.014
P+0.013
Hey+0.012
Neg logit contributions
ine-0.014
 birthday-0.013
hip-0.012
 Office-0.012
 X-0.012
TokenThis
Token ID1212
Feature activation+1.57e-02

Pos logit contributions
ls+0.013
Run+0.012
mun+0.012
 bitten+0.010
led+0.010
Neg logit contributions
ine-0.025
 birthday-0.024
hip-0.023
 Office-0.023
 X-0.023
Token is
Token ID318
Feature activation+3.01e-02

Pos logit contributions
ine+0.025
 birthday+0.023
hip+0.023
rap+0.022
sing+0.022
Neg logit contributions
Run-0.026
ls-0.026
mun-0.025
Hey-0.023
P-0.023
Token the
Token ID262
Feature activation-1.23e-03

Pos logit contributions
ine+0.056
hip+0.054
sing+0.051
 desk+0.051
 eye+0.051
Neg logit contributions
Run-0.037
ls-0.037
P-0.033
Hey-0.033
 distracted-0.031
Token best
Token ID1266
Feature activation+7.34e-03

Pos logit contributions
Run+0.002
mun+0.002
ls+0.002
P+0.002
Hey+0.002
Neg logit contributions
ine-0.002
rap-0.002
hip-0.002
 Office-0.002
artments-0.002
Token gift
Token ID6979
Feature activation-4.25e-03

Pos logit contributions
ine+0.010
hip+0.009
 Office+0.008
uries+0.008
ika+0.008
Neg logit contributions
ls-0.014
Run-0.013
mun-0.012
P-0.011
Hey-0.011
Token ever
Token ID1683
Feature activation-8.18e-03

Pos logit contributions
ls+0.007
Run+0.007
mun+0.006
 bitten+0.006
P+0.006
Neg logit contributions
ine-0.008
 birthday-0.007
hip-0.007
 Office-0.007
 Jupiter-0.006
Token!"
Token ID2474
Feature activation-6.27e-03

Pos logit contributions
ls+0.013
Run+0.012
mun+0.012
led+0.011
P+0.011
Neg logit contributions
ine-0.015
 birthday-0.014
 Office-0.013
hip-0.013
 Jupiter-0.013
Token From
Token ID3574
Feature activation+1.07e-02

Pos logit contributions
ine+0.015
 birthday+0.014
hip+0.013
 Office+0.013
 X+0.013
Neg logit contributions
Run-0.017
ls-0.017
mun-0.016
P-0.015
Hey-0.015
Token that
Token ID326
Feature activation+2.75e-02

Pos logit contributions
ine+0.015
 birthday+0.014
hip+0.014
 Office+0.014
 X+0.013
Neg logit contributions
ls-0.021
Run-0.021
mun-0.019
P-0.018
Hey-0.018
Token day
Token ID1110
Feature activation+3.29e-04

Pos logit contributions
ine+0.047
artments+0.047
urger+0.046
uries+0.046
rap+0.044
Neg logit contributions
mun-0.040
Run-0.040
ls-0.039
P-0.037
Hey-0.035
Token on
Token ID319
Feature activation+3.28e-03

Pos logit contributions
ine+0.001
 birthday+0.001
hip+0.001
 Office+0.001
 X+0.001
Neg logit contributions
Run-0.000
ls-0.000
mun-0.000
P-0.000
Hey-0.000
Token,
Token ID11
Feature activation+7.88e-03

Pos logit contributions
ine+0.004
 birthday+0.004
 Office+0.003
hip+0.003
 Jupiter+0.003
Neg logit contributions
Run-0.007
ls-0.007
mun-0.007
P-0.006
 bitten-0.006
Token Lily
Token ID20037
Feature activation+3.94e-02

Pos logit contributions
ine+0.012
 birthday+0.012
 Office+0.011
hip+0.011
 Jupiter+0.011
Neg logit contributions
Run-0.014
ls-0.014
mun-0.013
P-0.012
 bitten-0.012
Token loved
Token ID6151
Feature activation+1.13e-02

Pos logit contributions
 birthday+0.075
sing+0.070
 X+0.069
ine+0.068
 Today+0.067
Neg logit contributions
mun-0.053
ls-0.051
 distracted-0.048
led-0.048
P-0.047
Token playing
Token ID2712
Feature activation+4.04e-03

Pos logit contributions
ine+0.022
hip+0.020
 eye+0.019
uries+0.019
 Today+0.019
Neg logit contributions
ls-0.016
Run-0.015
mun-0.015
 bitten-0.014
 distracted-0.014
Token in
Token ID287
Feature activation+1.97e-02

Pos logit contributions
ine+0.007
 birthday+0.007
hip+0.007
 Office+0.006
 X+0.006
Neg logit contributions
ls-0.006
Run-0.006
mun-0.006
P-0.005
Hey-0.005
Token the
Token ID262
Feature activation+1.67e-03

Pos logit contributions
ine+0.046
urger+0.046
hip+0.044
uries+0.043
 Office+0.042
Neg logit contributions
Run-0.017
ls-0.016
mun-0.015
 bitten-0.015
P-0.014
Token empty
Token ID6565
Feature activation+2.71e-02

Pos logit contributions
ine+0.002
hip+0.002
 birthday+0.002
 Office+0.002
 Jupiter+0.002
Neg logit contributions
ls-0.003
Run-0.003
mun-0.003
P-0.003
Hey-0.003
Token,
Token ID11
Feature activation-3.10e-04

Pos logit contributions
ine+0.024
hip+0.022
 birthday+0.022
 X+0.022
 Jupiter+0.022
Neg logit contributions
ls-0.027
mun-0.026
Run-0.026
P-0.023
led-0.023
Token the
Token ID262
Feature activation+4.31e-03

Pos logit contributions
Run+0.001
ls+0.001
mun+0.001
P+0.001
 bitten+0.000
Neg logit contributions
ine-0.000
 birthday-0.000
 Jupiter-0.000
 Office-0.000
hip-0.000
Token teacher
Token ID4701
Feature activation+1.29e-02

Pos logit contributions
ine+0.006
 birthday+0.005
hip+0.005
 Office+0.005
 Jupiter+0.005
Neg logit contributions
ls-0.008
Run-0.008
mun-0.008
P-0.007
Hey-0.007
Token asked
Token ID1965
Feature activation+1.95e-02

Pos logit contributions
ine+0.025
 birthday+0.024
hip+0.023
 Office+0.023
 X+0.022
Neg logit contributions
ls-0.018
Run-0.018
mun-0.017
P-0.015
Hey-0.015
Token Lily
Token ID20037
Feature activation+2.44e-02

Pos logit contributions
ine+0.035
 X+0.032
 birthday+0.031
 Office+0.031
sing+0.031
Neg logit contributions
ls-0.029
Run-0.028
mun-0.028
Hey-0.025
led-0.024
Token's
Token ID338
Feature activation+2.83e-02

Pos logit contributions
ine+0.046
 X+0.045
sing+0.043
 oval+0.042
 Office+0.042
Neg logit contributions
ls-0.032
mun-0.031
Run-0.026
 distracted-0.026
 bitten-0.024
Token mom
Token ID1995
Feature activation+2.81e-02

Pos logit contributions
ine+0.044
hip+0.039
rap+0.035
uries+0.034
artments+0.034
Neg logit contributions
ls-0.047
mun-0.045
Run-0.045
 bitten-0.042
 distracted-0.041
Token to
Token ID284
Feature activation+1.05e-02

Pos logit contributions
ine+0.044
 X+0.043
sing+0.042
 Jupiter+0.042
 Office+0.041
Neg logit contributions
mun-0.044
ls-0.042
Run-0.035
 distracted-0.035
led-0.035
Token send
Token ID3758
Feature activation+1.16e-02

Pos logit contributions
ine+0.016
 Office+0.014
 Jupiter+0.014
 birthday+0.013
hip+0.013
Neg logit contributions
Run-0.018
mun-0.017
P-0.016
Hey-0.016
ls-0.016
Token a
Token ID257
Feature activation+7.76e-03

Pos logit contributions
ine+0.020
 birthday+0.019
 Office+0.018
hip+0.018
 X+0.018
Neg logit contributions
ls-0.018
Run-0.018
mun-0.017
P-0.016
Hey-0.016
Token note
Token ID3465
Feature activation+2.05e-02

Pos logit contributions
ine+0.010
hip+0.009
 Office+0.009
rap+0.009
 Jupiter+0.008
Neg logit contributions
ls-0.015
Run-0.015
mun-0.015
P-0.014
 bitten-0.013
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
ls+0.003
Run+0.003
mun+0.003
 bitten+0.002
P+0.002
Neg logit contributions
ine-0.003
 birthday-0.003
 Office-0.003
hip-0.003
 X-0.003
Token From
Token ID3574
Feature activation+8.48e-03

Pos logit contributions
ine+0.021
 birthday+0.019
 Office+0.018
hip+0.018
 X+0.018
Neg logit contributions
ls-0.019
Run-0.019
mun-0.018
P-0.017
Hey-0.016
Token then
Token ID788
Feature activation+1.64e-02

Pos logit contributions
ine+0.012
 birthday+0.011
hip+0.010
 Office+0.010
 X+0.010
Neg logit contributions
ls-0.017
Run-0.016
mun-0.016
P-0.015
Hey-0.014
Token on
Token ID319
Feature activation+1.01e-03

Pos logit contributions
ine+0.039
 birthday+0.037
 Office+0.036
 Jupiter+0.036
hip+0.036
Neg logit contributions
ls-0.016
Run-0.016
mun-0.014
P-0.014
Hey-0.012
Token,
Token ID11
Feature activation+5.08e-03

Pos logit contributions
ine+0.001
 birthday+0.001
hip+0.001
 end+0.001
 Office+0.001
Neg logit contributions
Run-0.002
ls-0.002
mun-0.002
 bitten-0.002
P-0.002
Token Lily
Token ID20037
Feature activation+3.63e-02

Pos logit contributions
ine+0.008
 birthday+0.008
hip+0.007
 Jupiter+0.007
 Office+0.007
Neg logit contributions
Run-0.009
ls-0.009
mun-0.009
P-0.008
 bitten-0.008
Token always
Token ID1464
Feature activation+1.78e-02

Pos logit contributions
 birthday+0.061
ine+0.061
hip+0.057
 X+0.057
 Jupiter+0.056
Neg logit contributions
ls-0.057
mun-0.056
Run-0.055
P-0.052
 distracted-0.049
Token brushed
Token ID31782
Feature activation+1.55e-02

Pos logit contributions
 birthday+0.037
ine+0.037
hip+0.036
 Jupiter+0.036
 X+0.035
Neg logit contributions
mun-0.020
Run-0.020
ls-0.018
 bitten-0.018
 distracted-0.018
Token Max
Token ID5436
Feature activation+7.09e-03

Pos logit contributions
 birthday+0.028
ine+0.027
 X+0.026
 Office+0.026
 Jupiter+0.026
Neg logit contributions
ls-0.024
mun-0.022
led-0.021
Run-0.021
Hey-0.019
Token softly
Token ID26625
Feature activation+2.69e-02

Pos logit contributions
 X+0.013
ine+0.013
 birthday+0.013
 Jupiter+0.013
 Office+0.012
Neg logit contributions
ls-0.010
led-0.009
mun-0.009
Run-0.009
 distracted-0.008
Token and
Token ID290
Feature activation+8.68e-03

Pos logit contributions
 Jupiter+0.051
rap+0.050
ine+0.049
 astronaut+0.049
 X+0.049
Neg logit contributions
ls-0.030
led-0.029
Run-0.028
mun-0.027
 bitten-0.027
Token in
Token ID287
Feature activation+9.49e-03

Pos logit contributions
Run+0.021
ls+0.020
mun+0.020
Hey+0.018
Sorry+0.017
Neg logit contributions
 birthday-0.012
 for-0.011
ine-0.011
rap-0.011
 Jupiter-0.011
Token the
Token ID262
Feature activation-1.74e-03

Pos logit contributions
ine+0.018
 birthday+0.016
hip+0.016
 Office+0.015
rap+0.015
Neg logit contributions
Run-0.014
ls-0.013
mun-0.013
 bitten-0.012
P-0.012
Token jungle
Token ID20712
Feature activation+9.91e-04

Pos logit contributions
P+0.003
mun+0.003
Run+0.003
Hey+0.002
ls+0.002
Neg logit contributions
ine-0.003
hip-0.003
rap-0.003
artments-0.003
uries-0.003
Token.
Token ID13
Feature activation+1.62e-02

Pos logit contributions
ine+0.001
 birthday+0.001
hip+0.001
 Office+0.001
 Jupiter+0.001
Neg logit contributions
Run-0.002
ls-0.002
mun-0.002
Hey-0.002
P-0.002
Token
Token ID198
Feature activation+1.65e-02

Pos logit contributions
ine+0.028
 birthday+0.026
hip+0.025
rap+0.024
 Office+0.024
Neg logit contributions
Run-0.026
ls-0.025
mun-0.025
Hey-0.023
P-0.022
TokenOne
Token ID3198
Feature activation+2.95e-02

Pos logit contributions
ine+0.030
 birthday+0.028
hip+0.026
 Office+0.026
 X+0.026
Neg logit contributions
ls-0.025
Run-0.025
mun-0.024
P-0.022
Hey-0.021
Token day
Token ID1110
Feature activation-3.84e-03

Pos logit contributions
ine+0.043
artments+0.040
uries+0.040
hip+0.040
rap+0.039
Neg logit contributions
Run-0.055
ls-0.055
P-0.051
mun-0.049
Hey-0.047
Token,
Token ID11
Feature activation+3.00e-03

Pos logit contributions
mun+0.009
ls+0.009
Run+0.009
awed+0.009
P+0.009
Neg logit contributions
 birthday-0.003
ine-0.003
 X-0.003
 moon-0.002
 Jupiter-0.002
Token he
Token ID339
Feature activation+1.03e-02

Pos logit contributions
 birthday+0.004
ine+0.004
 X+0.004
 moon+0.004
 Jupiter+0.004
Neg logit contributions
ls-0.006
mun-0.006
Run-0.005
P-0.005
awed-0.005
Token heard
Token ID2982
Feature activation+1.65e-02

Pos logit contributions
ine+0.016
 birthday+0.014
 Office+0.014
 X+0.013
 Jupiter+0.013
Neg logit contributions
ls-0.019
Run-0.018
mun-0.018
aro-0.016
Sorry-0.015
Token a
Token ID257
Feature activation-7.33e-03

Pos logit contributions
ine+0.027
 birthday+0.025
hip+0.024
 Office+0.024
rap+0.023
Neg logit contributions
Run-0.028
ls-0.028
mun-0.026
Hey-0.024
 bitten-0.024
Token
Token ID198
Feature activation+1.50e-02

Pos logit contributions
ls+0.007
Run+0.007
mun+0.007
P+0.006
Hey+0.006
Neg logit contributions
ine-0.009
 birthday-0.009
hip-0.008
 X-0.008
 Office-0.008
Token
Token ID198
Feature activation+9.44e-03

Pos logit contributions
 birthday+0.032
ine+0.031
 X+0.030
 Office+0.029
 eye+0.029
Neg logit contributions
ls-0.018
Run-0.018
mun-0.017
P-0.017
 distracted-0.014
TokenWhen
Token ID2215
Feature activation-9.84e-04

Pos logit contributions
 birthday+0.018
ine+0.018
 X+0.017
 Office+0.017
hip+0.016
Neg logit contributions
Run-0.014
ls-0.013
P-0.013
mun-0.012
Hey-0.011
Token it
Token ID340
Feature activation+1.03e-02

Pos logit contributions
ls+0.001
Run+0.001
mun+0.001
P+0.001
Hey+0.001
Neg logit contributions
ine-0.002
 birthday-0.002
hip-0.002
 Office-0.002
uries-0.002
Token started
Token ID2067
Feature activation-1.35e-02

Pos logit contributions
ine+0.017
 birthday+0.016
 Office+0.015
hip+0.015
 X+0.015
Neg logit contributions
Run-0.017
ls-0.017
mun-0.017
P-0.015
 bitten-0.015
Token getting
Token ID1972
Feature activation+2.49e-03

Pos logit contributions
Run+0.026
ls+0.025
mun+0.025
P+0.024
Hey+0.023
Neg logit contributions
ine-0.020
 birthday-0.019
 Jupiter-0.017
 pattern-0.017
hip-0.016
Token dark
Token ID3223
Feature activation+7.86e-03

Pos logit contributions
ine+0.003
 birthday+0.003
 Jupiter+0.003
 moon+0.003
 end+0.003
Neg logit contributions
Run-0.005
mun-0.005
ls-0.005
awed-0.004
P-0.004
Token,
Token ID11
Feature activation-6.76e-03

Pos logit contributions
ine+0.011
 birthday+0.010
 Office+0.009
hip+0.009
 X+0.009
Neg logit contributions
ls-0.015
Run-0.015
mun-0.015
P-0.013
Hey-0.013
Token Mild
Token ID29087
Feature activation+1.65e-02

Pos logit contributions
ls+0.012
Run+0.012
mun+0.011
P+0.010
Hey+0.010
Neg logit contributions
ine-0.011
 birthday-0.010
hip-0.010
 Office-0.010
 X-0.009
Token and
Token ID290
Feature activation+1.72e-03

Pos logit contributions
 birthday+0.028
ine+0.027
 X+0.025
hip+0.025
 Jupiter+0.025
Neg logit contributions
ls-0.026
Run-0.026
mun-0.025
led-0.024
Hey-0.023
Token her
Token ID607
Feature activation+5.12e-03

Pos logit contributions
ine+0.003
 birthday+0.003
hip+0.003
 Office+0.003
uries+0.002
Neg logit contributions
ls-0.003
Run-0.003
mun-0.003
 bitten-0.003
 distracted-0.002
Token with
Token ID351
Feature activation-7.99e-03

Pos logit contributions
ls+0.009
Run+0.009
mun+0.009
P+0.008
Hey+0.008
Neg logit contributions
ine-0.010
 birthday-0.009
hip-0.009
 Office-0.009
 X-0.009
Token hair
Token ID4190
Feature activation+6.43e-03

Pos logit contributions
Run+0.014
ls+0.014
mun+0.013
P+0.012
Hey+0.012
Neg logit contributions
ine-0.013
 birthday-0.012
 Office-0.011
hip-0.011
 X-0.011
Token.
Token ID13
Feature activation-4.12e-03

Pos logit contributions
ine+0.010
 birthday+0.009
 Office+0.009
hip+0.009
 X+0.009
Neg logit contributions
ls-0.012
Run-0.011
mun-0.011
P-0.010
led-0.010
Token
Token ID220
Feature activation-1.97e-03

Pos logit contributions
ls+0.008
Run+0.007
mun+0.007
P+0.007
led+0.007
Neg logit contributions
ine-0.006
 birthday-0.006
 Office-0.006
hip-0.005
 Jupiter-0.005
Token
Token ID198
Feature activation+5.88e-03

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
P+0.003
Hey+0.003
Neg logit contributions
ine-0.003
 birthday-0.003
hip-0.003
 Office-0.003
 X-0.003
Token
Token ID198
Feature activation-2.82e-03

Pos logit contributions
ine+0.010
 birthday+0.009
hip+0.009
 Office+0.009
 Jupiter+0.009
Neg logit contributions
ls-0.010
Run-0.009
mun-0.009
P-0.008
 bitten-0.008
TokenThe
Token ID464
Feature activation+9.39e-03

Pos logit contributions
ls+0.005
Run+0.005
mun+0.004
led+0.004
 bitten+0.004
Neg logit contributions
ine-0.005
 birthday-0.004
hip-0.004
 Office-0.004
 Jupiter-0.004
Token bunny
Token ID44915
Feature activation+1.59e-02

Pos logit contributions
ine+0.016
 birthday+0.014
hip+0.014
 Office+0.014
 Jupiter+0.014
Neg logit contributions
Run-0.016
ls-0.016
mun-0.015
P-0.014
Hey-0.013
Token and
Token ID290
Feature activation+1.65e-03

Pos logit contributions
ine+0.029
 birthday+0.028
hip+0.026
 Office+0.026
 X+0.026
Neg logit contributions
ls-0.024
Run-0.024
mun-0.023
P-0.021
Hey-0.020
Token the
Token ID262
Feature activation-1.98e-04

Pos logit contributions
ine+0.003
 birthday+0.003
hip+0.002
 Office+0.002
 X+0.002
Neg logit contributions
ls-0.003
Run-0.003
mun-0.003
P-0.002
Hey-0.002
Token k
Token ID479
Feature activation-7.66e-03

Pos logit contributions
Run+0.000
ls+0.000
mun+0.000
P+0.000
Hey+0.000
Neg logit contributions
ine-0.000
 birthday-0.000
hip-0.000
 Office-0.000
 Jupiter-0.000
Token but
Token ID475
Feature activation+8.51e-03

Pos logit contributions
Run+0.020
ls+0.019
mun+0.018
P+0.016
 distracted+0.016
Neg logit contributions
ine-0.020
 birthday-0.018
 Office-0.018
hip-0.018
rap-0.018
Token warned
Token ID7728
Feature activation+2.35e-02

Pos logit contributions
ine+0.016
 birthday+0.015
 Office+0.014
hip+0.014
 Jupiter+0.014
Neg logit contributions
ls-0.013
Run-0.013
mun-0.012
P-0.011
 bitten-0.011
Token her
Token ID607
Feature activation+1.78e-02

Pos logit contributions
 X+0.045
 Office+0.044
ine+0.043
 Jupiter+0.042
 birthday+0.041
Neg logit contributions
ls-0.032
Run-0.030
mun-0.030
Hey-0.028
led-0.027
Token that
Token ID326
Feature activation+1.33e-02

Pos logit contributions
ine+0.031
 X+0.029
 birthday+0.028
 Office+0.028
hip+0.028
Neg logit contributions
ls-0.028
Run-0.026
mun-0.025
 bitten-0.023
 distracted-0.023
Token the
Token ID262
Feature activation+1.03e-02

Pos logit contributions
ine+0.027
 birthday+0.025
hip+0.024
 Office+0.024
 X+0.024
Neg logit contributions
ls-0.017
Run-0.017
mun-0.017
P-0.014
Hey-0.014
Token toy
Token ID13373
Feature activation+1.09e-02

Pos logit contributions
ine+0.014
 birthday+0.013
hip+0.012
 Office+0.012
 Jupiter+0.012
Neg logit contributions
Run-0.020
ls-0.020
mun-0.019
P-0.018
Hey-0.018
Token was
Token ID373
Feature activation+7.99e-03

Pos logit contributions
ine+0.019
 birthday+0.018
hip+0.017
 X+0.017
 Office+0.017
Neg logit contributions
ls-0.017
Run-0.016
mun-0.016
Hey-0.014
 bitten-0.014
Token very
Token ID845
Feature activation-2.15e-02

Pos logit contributions
ine+0.012
 birthday+0.011
 Office+0.010
hip+0.010
 Jupiter+0.010
Neg logit contributions
Run-0.015
ls-0.015
mun-0.015
P-0.014
Hey-0.013
Token special
Token ID2041
Feature activation-3.35e-03

Pos logit contributions
Run+0.045
mun+0.044
ls+0.043
Hey+0.040
P+0.039
Neg logit contributions
ine-0.029
 birthday-0.026
 Office-0.023
hip-0.023
 X-0.023
Token and
Token ID290
Feature activation-5.93e-03

Pos logit contributions
ls+0.005
Run+0.005
mun+0.005
 bitten+0.004
P+0.004
Neg logit contributions
ine-0.006
hip-0.005
 Jupiter-0.005
rap-0.005
 Office-0.005
Token she
Token ID673
Feature activation-4.96e-03

Pos logit contributions
ls+0.009
Run+0.009
mun+0.008
 bitten+0.007
P+0.007
Neg logit contributions
ine-0.011
 birthday-0.010
hip-0.010
 Office-0.010
 X-0.010
Token back
Token ID736
Feature activation-1.17e-02

Pos logit contributions
mun+0.010
Run+0.010
ls+0.010
P+0.009
led+0.009
Neg logit contributions
ine-0.007
 birthday-0.007
hip-0.006
 Jupiter-0.006
 X-0.006
Token to
Token ID284
Feature activation-9.35e-03

Pos logit contributions
Run+0.025
mun+0.024
ls+0.023
P+0.023
Hey+0.023
Neg logit contributions
ine-0.014
 birthday-0.013
 Jupiter-0.012
 sleep-0.012
 for-0.012
Token the
Token ID262
Feature activation-7.18e-03

Pos logit contributions
Run+0.015
ls+0.015
mun+0.014
 bitten+0.013
P+0.013
Neg logit contributions
ine-0.016
 birthday-0.015
 Office-0.014
hip-0.014
 X-0.014
Token house
Token ID2156
Feature activation+1.29e-02

Pos logit contributions
Run+0.013
ls+0.013
mun+0.013
P+0.012
Hey+0.012
Neg logit contributions
ine-0.011
 birthday-0.010
hip-0.009
 Office-0.009
 X-0.009
Token,
Token ID11
Feature activation-6.51e-03

Pos logit contributions
ine+0.020
 birthday+0.018
 Office+0.017
hip+0.017
 X+0.017
Neg logit contributions
ls-0.023
Run-0.022
mun-0.022
 bitten-0.019
P-0.019
Token but
Token ID475
Feature activation+4.39e-03

Pos logit contributions
ls+0.012
Run+0.012
mun+0.011
P+0.011
Hey+0.010
Neg logit contributions
ine-0.010
 birthday-0.009
 Office-0.009
hip-0.008
 X-0.008
Token she
Token ID673
Feature activation-4.24e-03

Pos logit contributions
ine+0.008
 birthday+0.008
 Office+0.007
hip+0.007
 X+0.007
Neg logit contributions
ls-0.006
Run-0.006
mun-0.006
P-0.005
Hey-0.005
Token slipped
Token ID18859
Feature activation-1.20e-02

Pos logit contributions
Run+0.007
ls+0.007
mun+0.007
P+0.006
Hey+0.006
Neg logit contributions
ine-0.007
 birthday-0.007
 Office-0.006
hip-0.006
 X-0.006
Token and
Token ID290
Feature activation-3.55e-02

Pos logit contributions
ls+0.024
mun+0.024
Run+0.024
P+0.022
led+0.021
Neg logit contributions
ine-0.016
 birthday-0.014
hip-0.013
 X-0.013
 Office-0.013
Token fell
Token ID3214
Feature activation-1.26e-02

Pos logit contributions
mun+0.063
ls+0.056
Run+0.055
ins+0.054
awed+0.053
Neg logit contributions
ine-0.048
 sleep-0.047
 end-0.044
 for-0.042
 desk-0.042
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
mun+0.025
Run+0.024
ls+0.023
P+0.022
Hey+0.022
Neg logit contributions
ine-0.017
 birthday-0.015
 for-0.015
 moon-0.015
 pattern-0.015
Token magic
Token ID5536
Feature activation+2.83e-02

Pos logit contributions
ine+0.007
 birthday+0.006
 Office+0.006
hip+0.006
 X+0.006
Neg logit contributions
ls-0.010
Run-0.010
mun-0.009
P-0.008
Hey-0.008
Token wand
Token ID11569
Feature activation-8.12e-03

Pos logit contributions
ine+0.042
 birthday+0.037
hip+0.036
 Office+0.035
 X+0.034
Neg logit contributions
ls-0.055
Run-0.053
mun-0.050
Hey-0.047
P-0.047
Token again
Token ID757
Feature activation-2.55e-03

Pos logit contributions
Run+0.014
ls+0.014
mun+0.013
P+0.012
Hey+0.012
Neg logit contributions
ine-0.013
 birthday-0.013
hip-0.012
 Office-0.012
 X-0.012
Token!"
Token ID2474
Feature activation-1.39e-02

Pos logit contributions
ls+0.004
Run+0.004
mun+0.004
P+0.004
Hey+0.003
Neg logit contributions
ine-0.004
 birthday-0.004
hip-0.004
 Office-0.004
 X-0.004
Token They
Token ID1119
Feature activation-5.20e-03

Pos logit contributions
Run+0.029
ls+0.029
mun+0.029
P+0.027
 bitten+0.027
Neg logit contributions
ine-0.017
 birthday-0.017
 Today-0.015
 desk-0.014
 Grand-0.014
Token went
Token ID1816
Feature activation-8.14e-04

Pos logit contributions
ls+0.008
Run+0.008
mun+0.008
Hey+0.007
P+0.007
Neg logit contributions
ine-0.010
 birthday-0.009
 X-0.009
 eye-0.008
hip-0.008
Token to
Token ID284
Feature activation-4.91e-03

Pos logit contributions
Run+0.002
ls+0.002
mun+0.002
P+0.001
Hey+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
 X-0.001
 Office-0.001
 Jupiter-0.001
Token dry
Token ID5894
Feature activation-9.04e-03

Pos logit contributions
Run+0.008
ls+0.007
mun+0.007
 bitten+0.006
 distracted+0.006
Neg logit contributions
ine-0.009
hip-0.008
 birthday-0.008
 Office-0.008
urger-0.008
Token themselves
Token ID2405
Feature activation-2.95e-04

Pos logit contributions
Run+0.015
ls+0.015
mun+0.014
P+0.013
Hey+0.013
Neg logit contributions
ine-0.015
 birthday-0.014
 Office-0.014
hip-0.014
 X-0.013
Token and
Token ID290
Feature activation-1.57e-02

Pos logit contributions
ls+0.000
Run+0.000
mun+0.000
 bitten+0.000
P+0.000
Neg logit contributions
ine-0.001
 birthday-0.000
 Office-0.000
hip-0.000
 X-0.000
Token find
Token ID1064
Feature activation+4.97e-03

Pos logit contributions
ls+0.026
Run+0.026
mun+0.025
P+0.023
Hey+0.022
Neg logit contributions
ine-0.027
 birthday-0.025
 Office-0.023
 X-0.023
hip-0.023
Token
Token ID198
Feature activation+1.20e-02

Pos logit contributions
ine+0.002
 birthday+0.002
 Office+0.002
 Jupiter+0.002
hip+0.002
Neg logit contributions
ls-0.004
Run-0.004
mun-0.004
P-0.003
led-0.003
TokenBut
Token ID1537
Feature activation+1.27e-02

Pos logit contributions
ine+0.023
 birthday+0.022
hip+0.021
 Office+0.021
 X+0.021
Neg logit contributions
ls-0.017
Run-0.017
mun-0.016
P-0.015
Hey-0.014
Token then
Token ID788
Feature activation-1.71e-02

Pos logit contributions
ine+0.018
 birthday+0.017
hip+0.016
 Office+0.016
 X+0.015
Neg logit contributions
Run-0.025
ls-0.024
mun-0.023
 distracted-0.021
Hey-0.021
Token something
Token ID1223
Feature activation-1.23e-02

Pos logit contributions
ls+0.034
Run+0.034
mun+0.032
P+0.030
Hey+0.029
Neg logit contributions
ine-0.024
 birthday-0.022
hip-0.020
 Office-0.020
 X-0.020
Token happened
Token ID3022
Feature activation-1.61e-02

Pos logit contributions
mun+0.023
Run+0.021
ls+0.021
P+0.019
awed+0.019
Neg logit contributions
ine-0.017
 end-0.016
 birthday-0.016
 pattern-0.016
 believe-0.015
Token that
Token ID326
Feature activation-6.35e-03

Pos logit contributions
ls+0.030
Run+0.030
mun+0.030
P+0.027
Hey+0.026
Neg logit contributions
ine-0.024
 birthday-0.021
hip-0.020
 Office-0.020
 X-0.020
Token changed
Token ID3421
Feature activation-9.92e-03

Pos logit contributions
mun+0.011
Run+0.011
ls+0.011
P+0.011
awed+0.010
Neg logit contributions
ine-0.009
 moon-0.009
 Jupiter-0.009
 birthday-0.009
 X-0.009
Token the
Token ID262
Feature activation-8.78e-03

Pos logit contributions
Run+0.017
ls+0.017
mun+0.017
P+0.015
Hey+0.015
Neg logit contributions
ine-0.016
 birthday-0.015
hip-0.014
 Office-0.014
 Jupiter-0.014
Token race
Token ID3234
Feature activation-1.51e-02

Pos logit contributions
ls+0.017
Run+0.017
mun+0.017
led+0.015
P+0.015
Neg logit contributions
 birthday-0.012
ine-0.012
 pattern-0.011
 X-0.011
 moon-0.011
Token.
Token ID13
Feature activation-6.20e-03

Pos logit contributions
ls+0.027
Run+0.027
mun+0.027
P+0.025
 bitten+0.023
Neg logit contributions
ine-0.023
 birthday-0.022
hip-0.021
 Office-0.020
 X-0.020
Token The
Token ID383
Feature activation-2.11e-03

Pos logit contributions
ls+0.013
mun+0.013
Run+0.013
P+0.012
led+0.012
Neg logit contributions
ine-0.008
 birthday-0.007
 Office-0.007
 Jupiter-0.007
hip-0.007
Token not
Token ID407
Feature activation-1.85e-03

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
P+0.003
Hey+0.003
Neg logit contributions
ine-0.003
 birthday-0.002
 Office-0.002
hip-0.002
 X-0.002
Token want
Token ID765
Feature activation+5.02e-03

Pos logit contributions
Run+0.004
ls+0.003
mun+0.003
 bitten+0.003
 bothered+0.003
Neg logit contributions
ine-0.003
 birthday-0.002
 Office-0.002
 Jupiter-0.002
rap-0.002
Token to
Token ID284
Feature activation-2.40e-02

Pos logit contributions
ine+0.009
 birthday+0.008
hip+0.008
 Office+0.008
 X+0.008
Neg logit contributions
ls-0.008
Run-0.008
mun-0.008
P-0.007
Hey-0.007
Token play
Token ID711
Feature activation-1.28e-02

Pos logit contributions
Run+0.050
ls+0.049
mun+0.048
P+0.044
 bitten+0.044
Neg logit contributions
ine-0.030
 birthday-0.027
 Office-0.026
hip-0.025
 Jupiter-0.025
Token with
Token ID351
Feature activation-2.14e-02

Pos logit contributions
Run+0.021
ls+0.021
mun+0.020
P+0.018
Hey+0.017
Neg logit contributions
ine-0.022
 birthday-0.021
 Office-0.019
hip-0.019
 X-0.019
Token the
Token ID262
Feature activation-5.30e-03

Pos logit contributions
Run+0.037
ls+0.037
mun+0.036
P+0.034
Hey+0.032
Neg logit contributions
ine-0.034
 birthday-0.032
 Office-0.029
hip-0.029
 X-0.029
Token baby
Token ID5156
Feature activation-1.23e-02

Pos logit contributions
ls+0.010
Run+0.010
mun+0.009
P+0.008
Hey+0.008
Neg logit contributions
ine-0.008
 birthday-0.008
hip-0.007
 Office-0.007
 X-0.007
Token sheep
Token ID15900
Feature activation-2.59e-02

Pos logit contributions
Run+0.025
ls+0.023
P+0.023
mun+0.023
Hey+0.022
Neg logit contributions
ine-0.016
 birthday-0.014
 moon-0.013
 Office-0.013
 Jupiter-0.013
Token.
Token ID13
Feature activation-9.68e-03

Pos logit contributions
Run+0.048
ls+0.047
mun+0.046
P+0.043
Hey+0.042
Neg logit contributions
ine-0.039
 birthday-0.037
hip-0.034
 Office-0.034
 Jupiter-0.033
Token They
Token ID1119
Feature activation-2.06e-02

Pos logit contributions
ls+0.019
Run+0.019
mun+0.018
P+0.017
 bitten+0.017
Neg logit contributions
ine-0.014
 birthday-0.013
 Office-0.012
hip-0.012
 Jupiter-0.011
Token wanted
Token ID2227
Feature activation+1.03e-03

Pos logit contributions
ls+0.032
Run+0.031
mun+0.030
P+0.027
Hey+0.026
Neg logit contributions
ine-0.038
 birthday-0.034
 X-0.034
 Office-0.033
 eye-0.033
Token to
Token ID284
Feature activation+1.33e-02

Pos logit contributions
ls+0.013
Run+0.013
mun+0.013
P+0.011
Hey+0.011
Neg logit contributions
ine-0.011
 birthday-0.010
 Office-0.009
hip-0.009
 X-0.009
Token explore
Token ID7301
Feature activation+1.17e-02

Pos logit contributions
ine+0.023
 Birthday+0.021
 Office+0.021
urger+0.021
 birthday+0.019
Neg logit contributions
Run-0.021
ls-0.020
mun-0.019
Hey-0.019
 hurt-0.018
Token.
Token ID13
Feature activation-2.49e-03

Pos logit contributions
ine+0.020
 birthday+0.019
 Office+0.018
hip+0.018
rap+0.017
Neg logit contributions
ls-0.019
Run-0.019
mun-0.018
Hey-0.016
led-0.016
Token Tom
Token ID4186
Feature activation+5.08e-03

Pos logit contributions
ls+0.004
Run+0.004
mun+0.004
P+0.004
led+0.004
Neg logit contributions
ine-0.004
 birthday-0.004
 Office-0.003
hip-0.003
 X-0.003
Token started
Token ID2067
Feature activation-8.50e-03

Pos logit contributions
 birthday+0.010
ine+0.010
hip+0.009
 Office+0.009
 X+0.009
Neg logit contributions
ls-0.007
Run-0.007
mun-0.007
led-0.006
 distracted-0.006
Token to
Token ID284
Feature activation-4.68e-03

Pos logit contributions
ls+0.013
Run+0.013
mun+0.013
P+0.012
 bitten+0.011
Neg logit contributions
ine-0.015
 birthday-0.014
hip-0.013
 Office-0.013
 X-0.013
Token explore
Token ID7301
Feature activation+1.62e-02

Pos logit contributions
Run+0.009
mun+0.008
ls+0.008
Hey+0.007
 bitten+0.007
Neg logit contributions
ine-0.007
 birthday-0.006
 Office-0.006
rap-0.006
hip-0.006
Token the
Token ID262
Feature activation+1.70e-03

Pos logit contributions
ine+0.030
 birthday+0.029
hip+0.027
 Office+0.027
 Birthday+0.027
Neg logit contributions
ls-0.025
Run-0.024
mun-0.023
led-0.020
Hey-0.020
Token camp
Token ID1413
Feature activation+2.34e-02

Pos logit contributions
ine+0.003
hip+0.002
 birthday+0.002
rap+0.002
 X+0.002
Neg logit contributions
ls-0.003
mun-0.003
Run-0.003
Hey-0.003
P-0.003
Token,
Token ID11
Feature activation+1.02e-03

Pos logit contributions
ine+0.039
 birthday+0.037
rap+0.034
hip+0.034
 X+0.034
Neg logit contributions
ls-0.039
Run-0.037
mun-0.037
led-0.033
P-0.031
Token discovering
Token ID21611
Feature activation+1.13e-02

Pos logit contributions
ine+0.002
 birthday+0.002
hip+0.001
 Office+0.001
rap+0.001
Neg logit contributions
Run-0.002
mun-0.002
ls-0.002
 bitten-0.002
 distracted-0.001
Token sour
Token ID11348
Feature activation-2.19e-02

Pos logit contributions
Run+0.023
ls+0.022
mun+0.022
Hey+0.020
P+0.020
Neg logit contributions
 birthday-0.020
ine-0.019
 X-0.018
 eye-0.017
 pattern-0.017
Token candy
Token ID18550
Feature activation-1.20e-02

Pos logit contributions
Run+0.035
ls+0.035
mun+0.034
P+0.030
Hey+0.029
Neg logit contributions
ine-0.038
 birthday-0.036
 X-0.033
hip-0.033
 Office-0.033
Token to
Token ID284
Feature activation+9.15e-03

Pos logit contributions
Run+0.023
mun+0.022
ls+0.021
P+0.019
Hey+0.019
Neg logit contributions
ine-0.018
 birthday-0.017
hip-0.016
 X-0.015
 Office-0.015
Token share
Token ID2648
Feature activation-5.31e-03

Pos logit contributions
ine+0.014
 Jupiter+0.014
 Office+0.013
 birthday+0.013
 astronaut+0.013
Neg logit contributions
mun-0.015
ls-0.014
Run-0.014
 bitten-0.014
P-0.013
Token.
Token ID13
Feature activation+5.54e-03

Pos logit contributions
ls+0.008
Run+0.008
mun+0.008
P+0.007
Hey+0.007
Neg logit contributions
ine-0.009
 birthday-0.009
 Office-0.008
hip-0.008
 Jupiter-0.008
Token Tim
Token ID5045
Feature activation-5.71e-03

Pos logit contributions
ine+0.010
hip+0.009
 birthday+0.009
rap+0.009
 Office+0.009
Neg logit contributions
Run-0.008
ls-0.008
mun-0.007
Hey-0.007
 distracted-0.006
Token and
Token ID290
Feature activation+9.96e-03

Pos logit contributions
Run+0.009
ls+0.009
mun+0.009
P+0.008
Hey+0.008
Neg logit contributions
ine-0.010
 birthday-0.009
 Office-0.009
hip-0.009
 X-0.009
Token Jake
Token ID14757
Feature activation-3.94e-03

Pos logit contributions
ine+0.012
 birthday+0.011
uries+0.010
hip+0.010
 Office+0.010
Neg logit contributions
Run-0.021
ls-0.020
mun-0.020
Hey-0.019
led-0.018
Token both
Token ID1111
Feature activation-1.33e-02

Pos logit contributions
Run+0.006
ls+0.006
mun+0.005
P+0.005
Hey+0.005
Neg logit contributions
ine-0.007
 birthday-0.007
hip-0.007
 Office-0.007
 Jupiter-0.007
Token loved
Token ID6151
Feature activation-1.17e-02

Pos logit contributions
ls+0.019
Run+0.019
mun+0.018
P+0.016
Hey+0.016
Neg logit contributions
ine-0.025
 birthday-0.024
 Office-0.022
hip-0.022
 X-0.022
Token the
Token ID262
Feature activation-1.82e-02

Pos logit contributions
ls+0.019
Run+0.018
mun+0.018
 bitten+0.016
P+0.016
Neg logit contributions
ine-0.021
 birthday-0.019
hip-0.018
 Office-0.018
 Jupiter-0.018
Token.
Token ID13
Feature activation-4.43e-03

Pos logit contributions
Run+0.017
mun+0.016
ls+0.016
P+0.016
led+0.015
Neg logit contributions
 birthday-0.015
ine-0.014
 Jupiter-0.013
 moon-0.013
 X-0.013
Token They
Token ID1119
Feature activation+1.11e-03

Pos logit contributions
ls+0.008
Run+0.008
mun+0.008
P+0.007
 bitten+0.007
Neg logit contributions
ine-0.007
 birthday-0.006
 Office-0.006
hip-0.006
 X-0.006
Token are
Token ID389
Feature activation+7.91e-03

Pos logit contributions
ine+0.001
 birthday+0.001
 X+0.001
hip+0.001
 Office+0.001
Neg logit contributions
ls-0.002
Run-0.002
mun-0.002
P-0.002
 bitten-0.002
Token glad
Token ID9675
Feature activation-1.28e-02

Pos logit contributions
ine+0.012
 birthday+0.011
 Jupiter+0.010
 Office+0.010
 X+0.010
Neg logit contributions
Run-0.015
ls-0.015
mun-0.015
Hey-0.013
P-0.013
Token they
Token ID484
Feature activation-1.32e-02

Pos logit contributions
Run+0.020
mun+0.020
ls+0.019
P+0.019
led+0.018
Neg logit contributions
 birthday-0.023
ine-0.022
 Jupiter-0.020
 moon-0.020
 X-0.020
Token have
Token ID423
Feature activation-1.73e-02

Pos logit contributions
ls+0.025
Run+0.025
mun+0.024
P+0.022
awed+0.021
Neg logit contributions
ine-0.018
 birthday-0.017
hip-0.017
 Office-0.017
 X-0.016
Token float
Token ID12178
Feature activation-3.54e-03

Pos logit contributions
Run+0.036
ls+0.036
mun+0.035
Hey+0.031
P+0.031
Neg logit contributions
ine-0.023
 birthday-0.022
 Office-0.019
hip-0.019
 X-0.019
Tokenies
Token ID444
Feature activation-2.39e-03

Pos logit contributions
Run+0.006
mun+0.006
ls+0.006
P+0.005
Hey+0.005
Neg logit contributions
ine-0.006
 birthday-0.005
hip-0.005
 X-0.005
 Office-0.005
Token.
Token ID13
Feature activation-4.85e-04

Pos logit contributions
Run+0.004
ls+0.004
mun+0.004
P+0.004
Hey+0.004
Neg logit contributions
ine-0.004
 birthday-0.004
hip-0.003
 Office-0.003
 X-0.003
Token They
Token ID1119
Feature activation+3.86e-03

Pos logit contributions
Run+0.001
ls+0.001
mun+0.001
P+0.001
Hey+0.001
Neg logit contributions
ine-0.001
 birthday-0.001
hip-0.001
 Office-0.001
 X-0.001
Token are
Token ID389
Feature activation+1.28e-02

Pos logit contributions
ine+0.005
 birthday+0.004
 X+0.004
hip+0.004
 Office+0.004
Neg logit contributions
ls-0.008
Run-0.008
mun-0.008
P-0.007
led-0.007
Token hairy
Token ID43873
Feature activation-1.79e-02

Pos logit contributions
Run+0.042
ls+0.041
mun+0.041
P+0.038
Hey+0.036
Neg logit contributions
ine-0.038
 birthday-0.036
hip-0.033
 X-0.032
 pattern-0.032
Token and
Token ID290
Feature activation-2.09e-03

Pos logit contributions
ls+0.031
Run+0.030
mun+0.029
 bitten+0.026
P+0.026
Neg logit contributions
ine-0.029
 birthday-0.027
 Office-0.026
hip-0.026
 Jupiter-0.026
Token bent
Token ID17157
Feature activation+1.22e-03

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
P+0.003
Hey+0.003
Neg logit contributions
ine-0.004
 birthday-0.003
hip-0.003
 Office-0.003
 X-0.003
Token.
Token ID13
Feature activation+1.30e-04

Pos logit contributions
ine+0.002
 birthday+0.002
 Office+0.002
hip+0.002
 X+0.002
Neg logit contributions
ls-0.002
Run-0.002
mun-0.002
P-0.001
Hey-0.001
Token Lily
Token ID20037
Feature activation-1.01e-02

Pos logit contributions
ine+0.000
 birthday+0.000
 Office+0.000
 Jupiter+0.000
hip+0.000
Neg logit contributions
ls-0.000
Run-0.000
mun-0.000
P-0.000
led-0.000
Token got
Token ID1392
Feature activation-3.75e-02

Pos logit contributions
Run+0.017
ls+0.017
mun+0.016
P+0.015
Hey+0.015
Neg logit contributions
ine-0.017
 pattern-0.014
 desk-0.014
 eye-0.013
 Office-0.013
Token upset
Token ID9247
Feature activation-1.44e-02

Pos logit contributions
mun+0.063
Run+0.060
ls+0.058
P+0.054
awed+0.054
Neg logit contributions
 for-0.053
ine-0.053
 eye-0.052
 end-0.052
 birthday-0.050
Token and
Token ID290
Feature activation-3.53e-02

Pos logit contributions
ls+0.028
Run+0.028
mun+0.027
P+0.025
Hey+0.024
Neg logit contributions
ine-0.020
 birthday-0.019
 Office-0.018
hip-0.017
 X-0.017
Token told
Token ID1297
Feature activation+2.46e-02

Pos logit contributions
P+0.059
mun+0.057
Run+0.057
ls+0.056
+0.054
Neg logit contributions
ine-0.052
 for-0.049
 end-0.044
 desk-0.044
 be-0.044
Token Max
Token ID5436
Feature activation-8.87e-03

Pos logit contributions
ine+0.047
 birthday+0.044
 Office+0.044
uries+0.042
 eye+0.042
Neg logit contributions
ls-0.034
Run-0.033
mun-0.031
Hey-0.029
 bitten-0.028
Token he
Token ID339
Feature activation-1.65e-02

Pos logit contributions
Run+0.017
ls+0.016
mun+0.016
P+0.015
Hey+0.015
Neg logit contributions
ine-0.013
 birthday-0.012
hip-0.011
 Office-0.011
 Jupiter-0.011
Token for
Token ID329
Feature activation+8.03e-03

Pos logit contributions
Run+0.021
ls+0.021
mun+0.020
P+0.019
Hey+0.018
Neg logit contributions
ine-0.014
 birthday-0.013
hip-0.012
 Office-0.012
 X-0.012
Token a
Token ID257
Feature activation-1.25e-02

Pos logit contributions
ine+0.016
hip+0.016
 Office+0.015
 Birthday+0.015
rap+0.015
Neg logit contributions
Run-0.010
ls-0.009
 bitten-0.009
 killed-0.009
 distracted-0.008
Token while
Token ID981
Feature activation-1.42e-03

Pos logit contributions
Run+0.022
ls+0.022
mun+0.021
P+0.019
 bitten+0.019
Neg logit contributions
ine-0.020
 birthday-0.018
 Office-0.018
hip-0.018
 X-0.017
Token,
Token ID11
Feature activation-2.35e-02

Pos logit contributions
ls+0.003
Run+0.003
mun+0.003
P+0.003
Hey+0.003
Neg logit contributions
ine-0.002
 birthday-0.002
 X-0.002
hip-0.002
 Jupiter-0.002
Token he
Token ID339
Feature activation-1.17e-02

Pos logit contributions
ls+0.035
mun+0.034
Run+0.033
P+0.031
awed+0.030
Neg logit contributions
ine-0.042
 birthday-0.041
 X-0.039
 Jupiter-0.039
 moon-0.038
Token started
Token ID2067
Feature activation-3.84e-02

Pos logit contributions
ls+0.018
Run+0.018
mun+0.018
P+0.016
Hey+0.015
Neg logit contributions
ine-0.022
 birthday-0.019
 Office-0.018
 X-0.018
 eye-0.018
Token to
Token ID284
Feature activation-3.49e-02

Pos logit contributions
Run+0.081
ls+0.078
mun+0.076
Hey+0.075
P+0.074
Neg logit contributions
ine-0.052
 birthday-0.045
 desk-0.039
 pattern-0.039
 Jupiter-0.039
Token get
Token ID651
Feature activation-3.90e-02

Pos logit contributions
ls+0.071
Run+0.070
mun+0.068
P+0.064
Hey+0.062
Neg logit contributions
ine-0.046
 birthday-0.042
hip-0.039
 Office-0.039
 X-0.038
Token tired
Token ID10032
Feature activation-1.18e-02

Pos logit contributions
Run+0.067
mun+0.066
ls+0.062
awed+0.058
Hey+0.057
Neg logit contributions
ine-0.061
 birthday-0.055
 eye-0.055
 end-0.055
 sleep-0.054
Token.
Token ID13
Feature activation-3.83e-03

Pos logit contributions
ls+0.019
Run+0.019
mun+0.018
P+0.016
Hey+0.016
Neg logit contributions
ine-0.021
 birthday-0.019
 Office-0.019
hip-0.018
 Jupiter-0.018
Token He
Token ID679
Feature activation-1.89e-02

Pos logit contributions
Run+0.005
ls+0.005
mun+0.005
Hey+0.005
 bitten+0.005
Neg logit contributions
ine-0.007
 birthday-0.007
hip-0.006
rap-0.006
 Office-0.006
Token a
Token ID257
Feature activation-1.53e-02

Pos logit contributions
Run+0.008
ls+0.008
mun+0.007
P+0.007
Hey+0.007
Neg logit contributions
ine-0.008
 birthday-0.007
hip-0.007
 Office-0.007
 X-0.007
Token big
Token ID1263
Feature activation-8.03e-03

Pos logit contributions
ls+0.021
Run+0.020
P+0.019
mun+0.019
 bitten+0.017
Neg logit contributions
ine-0.029
hip-0.028
 Office-0.028
 birthday-0.027
uries-0.027
Token rock
Token ID3881
Feature activation-1.70e-02

Pos logit contributions
ls+0.009
Run+0.009
P+0.007
mun+0.007
 slapped+0.007
Neg logit contributions
uries-0.017
ine-0.017
hip-0.017
 Office-0.016
 birthday-0.016
Token.
Token ID13
Feature activation-2.79e-04

Pos logit contributions
mun+0.037
ls+0.036
Run+0.035
P+0.032
Sorry+0.032
Neg logit contributions
ine-0.021
 eye-0.018
 X-0.018
hip-0.018
 birthday-0.017
Token Jane
Token ID12091
Feature activation-9.45e-03

Pos logit contributions
Run+0.000
ls+0.000
mun+0.000
Hey+0.000
 bitten+0.000
Neg logit contributions
ine-0.000
 birthday-0.000
hip-0.000
 Office-0.000
 Jupiter-0.000
Token got
Token ID1392
Feature activation-3.62e-02

Pos logit contributions
Run+0.015
ls+0.015
mun+0.015
P+0.014
Hey+0.013
Neg logit contributions
ine-0.017
 birthday-0.014
 Office-0.014
 X-0.014
 desk-0.014
Token scared
Token ID12008
Feature activation-1.34e-02

Pos logit contributions
mun+0.063
Run+0.060
ls+0.060
awed+0.057
P+0.054
Neg logit contributions
ine-0.050
 for-0.049
 birthday-0.048
 eye-0.048
 name-0.047
Token and
Token ID290
Feature activation-3.69e-02

Pos logit contributions
ls+0.025
Run+0.025
mun+0.024
P+0.022
Hey+0.022
Neg logit contributions
ine-0.020
 birthday-0.018
 Office-0.018
hip-0.017
 X-0.017
Token started
Token ID2067
Feature activation-3.62e-02

Pos logit contributions
mun+0.058
P+0.054
ls+0.054
+0.050
Run+0.049
Neg logit contributions
ine-0.062
 sleep-0.056
 for-0.055
 imagine-0.054
 desk-0.053
Token to
Token ID284
Feature activation-4.19e-02

Pos logit contributions
Run+0.081
ls+0.079
Hey+0.077
mun+0.077
P+0.075
Neg logit contributions
ine-0.045
 birthday-0.034
 sleep-0.033
 desk-0.033
 pattern-0.033
Token cry
Token ID3960
Feature activation-3.71e-03

Pos logit contributions
ls+0.079
Run+0.073
mun+0.073
P+0.071
led+0.069
Neg logit contributions
ine-0.065
 birthday-0.058
 sleep-0.056
 X-0.055
 eye-0.055
Token were
Token ID547
Feature activation+2.72e-02

Pos logit contributions
ine+0.004
 birthday+0.004
hip+0.004
 Office+0.004
 X+0.004
Neg logit contributions
Run-0.004
ls-0.003
mun-0.003
P-0.003
Hey-0.003
Token not
Token ID407
Feature activation-3.78e-03

Pos logit contributions
ine+0.042
 birthday+0.039
hip+0.037
 Office+0.037
 X+0.037
Neg logit contributions
ls-0.048
Run-0.048
mun-0.047
P-0.043
 bitten-0.042
Token ashamed
Token ID22461
Feature activation-1.42e-02

Pos logit contributions
Run+0.006
ls+0.005
mun+0.005
P+0.005
Hey+0.005
Neg logit contributions
ine-0.007
 birthday-0.007
 Jupiter-0.006
 Office-0.006
hip-0.006
Token anymore
Token ID7471
Feature activation-1.35e-02

Pos logit contributions
Run+0.030
mun+0.030
P+0.029
ls+0.029
awed+0.027
Neg logit contributions
 birthday-0.018
ine-0.016
 end-0.015
 for-0.015
 about-0.015
Token.
Token ID13
Feature activation+6.23e-03

Pos logit contributions
ls+0.021
Run+0.021
mun+0.020
P+0.018
 bitten+0.018
Neg logit contributions
ine-0.024
 birthday-0.022
hip-0.021
 Office-0.021
 X-0.021
Token<|endoftext|>
Token ID50256
Feature activation-3.70e-02

Pos logit contributions
ine+0.012
 birthday+0.011
hip+0.010
 X+0.010
 Office+0.010
Neg logit contributions
Run-0.009
ls-0.009
mun-0.009
P-0.008
Hey-0.008
TokenL
Token ID43
Feature activation-1.80e-02

Pos logit contributions
Run+0.055
ls+0.054
mun+0.050
 slapped+0.047
led+0.047
Neg logit contributions
 birthday-0.064
 Today-0.062
ine-0.062
 Office-0.059
 moon-0.059
Tokenily
Token ID813
Feature activation+6.76e-03

Pos logit contributions
ls+0.031
Run+0.030
mun+0.029
P+0.027
Hey+0.026
Neg logit contributions
ine-0.029
 birthday-0.028
hip-0.027
 Office-0.027
 Jupiter-0.026
Token and
Token ID290
Feature activation-1.52e-02

Pos logit contributions
ine+0.011
 birthday+0.010
hip+0.010
 Office+0.010
 Jupiter+0.010
Neg logit contributions
ls-0.011
Run-0.011
mun-0.011
P-0.010
Hey-0.010
Token Tom
Token ID4186
Feature activation-8.07e-05

Pos logit contributions
ls+0.032
mun+0.031
Run+0.030
led+0.028
awed+0.028
Neg logit contributions
 birthday-0.020
ine-0.019
 X-0.017
hip-0.017
 eye-0.016
Token were
Token ID547
Feature activation-2.39e-02

Pos logit contributions
ls+0.000
Run+0.000
mun+0.000
P+0.000
Hey+0.000
Neg logit contributions
ine-0.000
 birthday-0.000
hip-0.000
 Office-0.000
 X-0.000
Token girl
Token ID2576
Feature activation-3.89e-03

Pos logit contributions
ine+0.009
 birthday+0.008
hip+0.008
 Office+0.008
 X+0.008
Neg logit contributions
Run-0.010
ls-0.010
mun-0.009
P-0.009
Hey-0.008
Token was
Token ID373
Feature activation-3.08e-03

Pos logit contributions
ls+0.005
Run+0.005
mun+0.005
P+0.005
Hey+0.004
Neg logit contributions
ine-0.008
 birthday-0.007
 Office-0.007
hip-0.007
 X-0.007
Token scared
Token ID12008
Feature activation-1.50e-02

Pos logit contributions
Run+0.003
ls+0.003
mun+0.003
 bitten+0.003
P+0.003
Neg logit contributions
ine-0.007
 birthday-0.007
hip-0.006
 Office-0.006
 X-0.006
Token and
Token ID290
Feature activation-3.25e-02

Pos logit contributions
ls+0.028
Run+0.027
mun+0.026
Hey+0.023
 bitten+0.023
Neg logit contributions
ine-0.023
 birthday-0.021
 Office-0.021
hip-0.020
 X-0.020
Token began
Token ID2540
Feature activation-2.78e-02

Pos logit contributions
ls+0.049
mun+0.048
P+0.048
Run+0.046
Hey+0.042
Neg logit contributions
ine-0.058
 Jupiter-0.050
 birthday-0.049
 for-0.049
 end-0.048
Token to
Token ID284
Feature activation-3.52e-02

Pos logit contributions
Run+0.054
ls+0.053
mun+0.051
Hey+0.049
P+0.048
Neg logit contributions
ine-0.042
 birthday-0.037
 Jupiter-0.033
 Office-0.033
hip-0.033
Token back
Token ID736
Feature activation-1.10e-02

Pos logit contributions
ls+0.061
Run+0.060
mun+0.058
P+0.054
led+0.051
Neg logit contributions
ine-0.058
 birthday-0.054
hip-0.050
 Office-0.050
 X-0.050
Token away
Token ID1497
Feature activation-9.22e-03

Pos logit contributions
ls+0.020
Run+0.020
mun+0.019
P+0.017
 bitten+0.017
Neg logit contributions
ine-0.017
 birthday-0.016
hip-0.015
 Office-0.015
 X-0.014
Token slowly
Token ID6364
Feature activation+7.37e-03

Pos logit contributions
Run+0.016
ls+0.016
mun+0.015
P+0.014
Hey+0.013
Neg logit contributions
ine-0.015
 birthday-0.014
hip-0.013
 Office-0.013
 X-0.013
Token.
Token ID13
Feature activation-2.84e-04

Pos logit contributions
ine+0.013
 birthday+0.012
 Office+0.011
hip+0.011
 Jupiter+0.011
Neg logit contributions
ls-0.012
Run-0.012
mun-0.011
 bitten-0.010
P-0.010
Token She
Token ID1375
Feature activation-6.75e-03

Pos logit contributions
Run+0.000
ls+0.000
mun+0.000
Hey+0.000
 distracted+0.000
Neg logit contributions
ine-0.000
 birthday-0.000
 X-0.000
hip-0.000
urger-0.000