TOP ACTIVATIONS
MAX = 0.069

as fast as he could to the other side . He
late with his friend , M andy .
But then , M andy found a four -
were in the kitchen . M andy wanted to help make
was a happy march , full of shouts , laughter and
dad about her new friend Vin ny , the magic vine
. Suddenly , M andy heard a voice .
looks like a sport I can do there .
like the march was going to be delayed . While she
in the rain ," said M andy .
her mom for permitting her to march . The
look for a sport he could do . Then he saw
shows them a big tub full of water and foam .
to discuss the berries , M illy thought they were delicious
like the sunscreen because it felt tight on his skin .
to permit the little girl to march . The
Let 's go home !" Mack enzie and her mom my
like to join me on an adventure ?"
. Suddenly , M andy noticed something shiny on
j ogging is a way to exercise and stay healthy .

MIN ACTIVATIONS
MIN = -0.069

On the way , A str id saw a pond filled
they arrived home , A str id put her balloon in
! Mom my and A str id said good bye and
ie . " Hello , little girl . I
a big bird named O str ich wanted to swim in
Miss O str ich smiled and said ,
. They saw lots of old things like swords and shields
balloon . A str id held onto her balloon
around and saw lots of interesting things in the room .
a box of shiny or nam ents that spark led in
the lake . But O str ich was scared of water
� € Hi Miss O str ich ! That sounds funny
. " Bad wind , bad wind !" L
Miss O str ich told a very funny
N ina was panic - str icken . She screamed for
new words every day . Conf used is just a word
cherry tree with lots of cher ries on it . She
went she searched for or nam ents , and the more
€ Miss O str ich said , � �
explore . She noticed the old chest and kn elt down

INTERVAL 0.034 - 0.069
CONTAINS 0.468%

continued to climb until they couldn 't climb any higher .
play in the park . One day , they see a
and the frog worked together to clean the pond . They
some special plants growing on his body . He took out
her qualities and skill . M illy was so happy to

INTERVAL -0.000 - 0.034
CONTAINS 50.057%

€ � � sw imming , playing and making lots of
dad give them a net and a bucket . Ben and
, you can borrow one . Here , take this one
Their mom and dad looked at them with sad eyes .
have . He was very loyal and always helped his friends

INTERVAL -0.035 - -0.000
CONTAINS 49.191%

ever day and saw something new and amazing each time .
to John . When they heard him they cheered ! But
One day , Daisy was playing dolls with her friend .
is okay . We can set our blocks again later ."
we can play with the rain later . When it stops

INTERVAL -0.069 - -0.035
CONTAINS 0.285%

that night , Tim my started to cry . His knee
" Hey , that 's mine !" Sara shouts
is electric . She says that Sara can use her regular
Then , they all went back to the party and had
every day to supply the av oc ados . And the
Token as
Token ID355
Feature activation-1.29e-04

Pos logit contributions
 agreement+0.039
date+0.038
 Everyday+0.038
 particular+0.037
 invite+0.037
Neg logit contributions
w-0.017
ls-0.013
 many-0.012
ged-0.012
s-0.012
Token fast
Token ID3049
Feature activation+1.10e-02

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
aged+0.000
llers+0.000
Neg logit contributions
 agreement-0.000
 particular-0.000
date-0.000
 Everyday-0.000
 invite-0.000
Token as
Token ID355
Feature activation+3.86e-03

Pos logit contributions
 agreement+0.018
 particular+0.017
date+0.016
 Everyday+0.016
 invite+0.015
Neg logit contributions
w-0.016
aged-0.014
ller-0.014
ich-0.014
ged-0.014
Token he
Token ID339
Feature activation+2.39e-02

Pos logit contributions
 agreement+0.009
date+0.008
 Everyday+0.008
 particular+0.007
 invite+0.007
Neg logit contributions
w-0.004
ich-0.003
ller-0.003
nam-0.003
aged-0.003
Token could
Token ID714
Feature activation+5.21e-02

Pos logit contributions
 agreement+0.046
date+0.040
 particular+0.038
 Everyday+0.037
 basis+0.036
Neg logit contributions
w-0.032
aged-0.028
 Blood-0.028
ich-0.027
ged-0.027
Token to
Token ID284
Feature activation+6.87e-02

Pos logit contributions
 agreement+0.098
date+0.094
 Everyday+0.094
 checkout+0.093
 eater+0.086
Neg logit contributions
uts-0.052
w-0.048
 Blood-0.047
 touching-0.047
EEP-0.046
Token the
Token ID262
Feature activation+4.41e-02

Pos logit contributions
umm+0.124
date+0.122
 Everyday+0.118
 eater+0.111
 Whenever+0.110
Neg logit contributions
 touching-0.063
 screaming-0.059
 crossing-0.058
 hitting-0.057
 pressing-0.057
Token other
Token ID584
Feature activation+2.54e-02

Pos logit contributions
 Everyday+0.079
rily+0.076
 particular+0.075
rene+0.075
 Whenever+0.074
Neg logit contributions
ich-0.044
w-0.039
 screaming-0.038
RR-0.037
 shouting-0.037
Token side
Token ID1735
Feature activation+2.70e-02

Pos logit contributions
rily+0.039
 surely+0.037
rene+0.035
 Everyday+0.035
date+0.034
Neg logit contributions
 pressing-0.029
 fighting-0.028
 reading-0.028
ich-0.028
EEP-0.028
Token.
Token ID13
Feature activation+2.37e-03

Pos logit contributions
 agreement+0.049
 invite+0.040
 offer+0.040
 Everyday+0.040
date+0.040
Neg logit contributions
w-0.033
uts-0.032
 pressing-0.031
ich-0.030
 crying-0.030
Token He
Token ID679
Feature activation+1.99e-02

Pos logit contributions
 agreement+0.004
 particular+0.004
date+0.004
 invite+0.004
 offer+0.004
Neg logit contributions
w-0.003
ller-0.003
aged-0.003
llers-0.003
ich-0.003
Token late
Token ID2739
Feature activation+1.31e-02

Pos logit contributions
latable+0.051
 agreement+0.050
date+0.049
 particular+0.047
 Everyday+0.045
Neg logit contributions
 reading-0.029
 crying-0.028
 breathing-0.028
 screaming-0.027
w-0.027
Token with
Token ID351
Feature activation+1.79e-02

Pos logit contributions
 agreement+0.021
 particular+0.019
date+0.019
 Everyday+0.019
 invite+0.018
Neg logit contributions
w-0.019
ller-0.017
ich-0.017
aged-0.017
llers-0.017
Token his
Token ID465
Feature activation+1.14e-02

Pos logit contributions
 agreement+0.037
date+0.036
 particular+0.033
 invite+0.033
 Everyday+0.032
Neg logit contributions
w-0.016
 bleeding-0.015
 crying-0.015
 flashing-0.015
nam-0.015
Token friend
Token ID1545
Feature activation+1.51e-02

Pos logit contributions
 agreement+0.019
 Everyday+0.017
date+0.017
 particular+0.017
 Whenever+0.016
Neg logit contributions
w-0.016
ich-0.016
 elbows-0.014
ller-0.014
aged-0.014
Token,
Token ID11
Feature activation+1.54e-02

Pos logit contributions
 agreement+0.032
 particular+0.029
 Everyday+0.027
 surely+0.026
 offer+0.026
Neg logit contributions
w-0.013
 reading-0.013
ich-0.013
 crying-0.012
 Pur-0.012
Token M
Token ID337
Feature activation+6.59e-02

Pos logit contributions
 agreement+0.029
 particular+0.027
date+0.026
 Everyday+0.025
 offer+0.025
Neg logit contributions
ich-0.017
nam-0.015
w-0.015
 Pur-0.015
 crying-0.015
Tokenandy
Token ID10757
Feature activation+2.37e-02

Pos logit contributions
 agreement+0.118
 surely+0.105
date+0.101
 fill+0.100
 tow+0.099
Neg logit contributions
ich-0.089
w-0.082
ller-0.076
uts-0.075
llers-0.075
Token.
Token ID13
Feature activation+1.10e-02

Pos logit contributions
 agreement+0.043
 particular+0.040
date+0.038
 Everyday+0.038
 offer+0.036
Neg logit contributions
w-0.029
ich-0.029
ller-0.028
aged-0.027
llers-0.026
Token
Token ID220
Feature activation-7.89e-03

Pos logit contributions
 agreement+0.022
date+0.020
 particular+0.020
 invite+0.019
 offer+0.019
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
llers-0.010
aged-0.010
Token
Token ID198
Feature activation+1.49e-02

Pos logit contributions
w+0.011
ich+0.009
ller+0.009
aged+0.009
llers+0.009
Neg logit contributions
 agreement-0.014
 particular-0.013
 Everyday-0.013
date-0.013
 invite-0.012
Token
Token ID198
Feature activation+2.11e-02

Pos logit contributions
 agreement+0.030
 particular+0.028
date+0.027
 Everyday+0.027
 invite+0.027
Neg logit contributions
w-0.017
ich-0.014
ller-0.013
aged-0.013
irens-0.013
Token
Token ID198
Feature activation+1.75e-02

Pos logit contributions
w+0.006
ich+0.005
ller+0.005
aged+0.005
llers+0.005
Neg logit contributions
 agreement-0.008
 particular-0.008
 Everyday-0.008
date-0.007
 invite-0.007
Token
Token ID198
Feature activation+2.35e-02

Pos logit contributions
 agreement+0.032
 particular+0.030
 Everyday+0.029
date+0.028
 invite+0.028
Neg logit contributions
w-0.023
ich-0.020
ller-0.020
aged-0.019
llers-0.019
TokenBut
Token ID1537
Feature activation+3.30e-02

Pos logit contributions
 agreement+0.043
 particular+0.039
 Everyday+0.038
date+0.038
 invite+0.038
Neg logit contributions
w-0.031
ich-0.027
aged-0.026
irens-0.026
llers-0.025
Token then
Token ID788
Feature activation+9.72e-03

Pos logit contributions
 agreement+0.056
date+0.049
 Everyday+0.048
 particular+0.048
 offer+0.048
Neg logit contributions
w-0.046
aged-0.042
ich-0.041
ller-0.038
llers-0.038
Token,
Token ID11
Feature activation+2.25e-02

Pos logit contributions
 agreement+0.018
 particular+0.016
 Everyday+0.016
date+0.016
 invite+0.015
Neg logit contributions
w-0.013
ich-0.011
aged-0.011
ller-0.011
llers-0.011
Token M
Token ID337
Feature activation+6.41e-02

Pos logit contributions
 agreement+0.047
 particular+0.042
 Everyday+0.042
date+0.041
 invite+0.040
Neg logit contributions
w-0.024
ich-0.020
aged-0.020
ller-0.019
irens-0.018
Tokenandy
Token ID10757
Feature activation+2.19e-02

Pos logit contributions
 agreement+0.107
date+0.094
 particular+0.088
 surely+0.087
 invite+0.086
Neg logit contributions
w-0.098
ich-0.096
ouncing-0.086
inches-0.086
ller-0.085
Token found
Token ID1043
Feature activation+4.69e-03

Pos logit contributions
 agreement+0.043
 particular+0.039
date+0.037
 Everyday+0.037
 offer+0.035
Neg logit contributions
w-0.027
ich-0.024
aged-0.023
ller-0.023
llers-0.022
Token a
Token ID257
Feature activation+4.45e-04

Pos logit contributions
 agreement+0.010
 particular+0.010
 invite+0.009
date+0.009
 Everyday+0.009
Neg logit contributions
w-0.004
aged-0.004
ich-0.004
nam-0.003
irens-0.003
Token four
Token ID1440
Feature activation+7.98e-03

Pos logit contributions
 agreement+0.001
 Everyday+0.001
date+0.001
 particular+0.001
 invite+0.001
Neg logit contributions
w-0.001
ich-0.001
ller-0.001
nam-0.001
 bleeding-0.001
Token-
Token ID12
Feature activation-2.21e-04

Pos logit contributions
 agreement+0.015
date+0.014
 Everyday+0.013
 surely+0.013
 particular+0.013
Neg logit contributions
w-0.009
ich-0.008
nam-0.008
aged-0.008
ller-0.008
Token were
Token ID547
Feature activation-1.16e-02

Pos logit contributions
w+0.017
ich+0.014
aged+0.013
ller+0.013
llers+0.013
Neg logit contributions
 agreement-0.019
 particular-0.017
date-0.016
 Everyday-0.015
 offer-0.015
Token in
Token ID287
Feature activation+6.00e-03

Pos logit contributions
w+0.011
ich+0.010
aged+0.009
ller+0.009
llers+0.009
Neg logit contributions
 agreement-0.025
 particular-0.024
 Everyday-0.023
 offer-0.022
date-0.022
Token the
Token ID262
Feature activation-7.96e-03

Pos logit contributions
 agreement+0.011
date+0.011
 Everyday+0.010
 particular+0.010
 invite+0.010
Neg logit contributions
nam-0.006
 reading-0.005
w-0.005
 crying-0.005
 praying-0.005
Token kitchen
Token ID9592
Feature activation+4.24e-03

Pos logit contributions
w+0.012
ich+0.011
ller+0.010
irens+0.010
llers+0.010
Neg logit contributions
 agreement-0.013
 particular-0.011
 Everyday-0.011
 offer-0.011
date-0.011
Token.
Token ID13
Feature activation+4.41e-02

Pos logit contributions
 agreement+0.007
 particular+0.006
 Everyday+0.006
date+0.006
 invite+0.006
Neg logit contributions
w-0.006
ich-0.006
aged-0.006
ller-0.006
llers-0.006
Token M
Token ID337
Feature activation+6.37e-02

Pos logit contributions
 agreement+0.089
date+0.088
 choice+0.080
 offer+0.079
 promise+0.079
Neg logit contributions
 Blood-0.042
elt-0.033
AAAA-0.029
 Ding-0.029
 Boom-0.029
Tokenandy
Token ID10757
Feature activation+3.23e-03

Pos logit contributions
 agreement+0.112
date+0.099
 surely+0.098
 fill+0.096
 invite+0.092
Neg logit contributions
ich-0.090
w-0.089
azes-0.079
inches-0.078
ouncing-0.078
Token wanted
Token ID2227
Feature activation+1.59e-03

Pos logit contributions
 agreement+0.006
 particular+0.005
date+0.005
 Everyday+0.005
 offer+0.004
Neg logit contributions
w-0.005
ller-0.004
ich-0.004
aged-0.004
llers-0.004
Token to
Token ID284
Feature activation+1.85e-02

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
date+0.003
 invite+0.003
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
llers-0.002
Token help
Token ID1037
Feature activation-5.12e-05

Pos logit contributions
 agreement+0.030
rily+0.028
lated+0.027
mer+0.026
rant+0.026
Neg logit contributions
 Blood-0.020
 touching-0.019
 reading-0.018
w-0.018
 crying-0.018
Token make
Token ID787
Feature activation-5.67e-03

Pos logit contributions
w+0.000
aged+0.000
ller+0.000
ich+0.000
irens+0.000
Neg logit contributions
 agreement-0.000
 Everyday-0.000
 particular-0.000
date-0.000
 surely-0.000
Token was
Token ID373
Feature activation+2.49e-02

Pos logit contributions
 agreement+0.045
date+0.040
 particular+0.038
 Everyday+0.037
 offer+0.036
Neg logit contributions
w-0.039
ich-0.033
aged-0.033
AAAA-0.032
ller-0.032
Token a
Token ID257
Feature activation+3.11e-02

Pos logit contributions
 agreement+0.052
date+0.047
 offer+0.045
 particular+0.045
 Everyday+0.045
Neg logit contributions
w-0.025
ich-0.022
 flashing-0.022
 Blood-0.021
aged-0.021
Token happy
Token ID3772
Feature activation+2.11e-02

Pos logit contributions
 agreement+0.054
 Everyday+0.050
 particular+0.047
date+0.046
 Whenever+0.045
Neg logit contributions
w-0.042
ich-0.042
ller-0.037
AAAA-0.035
RR-0.035
Token march
Token ID9960
Feature activation+3.40e-02

Pos logit contributions
 agreement+0.030
 Everyday+0.029
date+0.027
 particular+0.027
 Whenever+0.025
Neg logit contributions
ich-0.032
w-0.031
ller-0.029
AAAA-0.028
EEP-0.028
Token,
Token ID11
Feature activation+7.99e-03

Pos logit contributions
 agreement+0.066
 particular+0.057
 Everyday+0.057
rily+0.053
 offer+0.053
Neg logit contributions
w-0.040
EEP-0.035
ich-0.035
ller-0.033
 Pur-0.033
Token full
Token ID1336
Feature activation+6.37e-02

Pos logit contributions
 agreement+0.015
date+0.013
 particular+0.013
 Everyday+0.013
 offer+0.012
Neg logit contributions
w-0.010
ich-0.010
ller-0.009
llers-0.009
aged-0.009
Token of
Token ID286
Feature activation+3.37e-02

Pos logit contributions
 agreement+0.083
rene+0.083
date+0.082
 surely+0.079
 particular+0.078
Neg logit contributions
irens-0.089
 flashing-0.089
EEP-0.089
uts-0.085
 Pur-0.084
Token shouts
Token ID34757
Feature activation+2.30e-02

Pos logit contributions
 agreement+0.062
date+0.062
 particular+0.061
 Everyday+0.058
 surely+0.058
Neg logit contributions
w-0.034
irens-0.032
aged-0.031
 flashing-0.030
nam-0.030
Token,
Token ID11
Feature activation+1.45e-03

Pos logit contributions
 agreement+0.038
 particular+0.035
 Everyday+0.034
date+0.033
 invite+0.032
Neg logit contributions
w-0.034
ich-0.031
ller-0.030
aged-0.029
llers-0.029
Token laughter
Token ID20263
Feature activation-2.84e-03

Pos logit contributions
 agreement+0.003
 particular+0.003
date+0.002
 Everyday+0.002
 invite+0.002
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
llers-0.002
aged-0.002
Token and
Token ID290
Feature activation+5.36e-03

Pos logit contributions
w+0.004
ich+0.004
ller+0.004
aged+0.003
llers+0.003
Neg logit contributions
 agreement-0.005
 particular-0.005
 Everyday-0.004
date-0.004
 offer-0.004
Token dad
Token ID9955
Feature activation+2.11e-03

Pos logit contributions
w+0.019
RR+0.017
ller+0.016
irens+0.016
ich+0.016
Neg logit contributions
 agreement-0.016
 invite-0.015
 particular-0.014
 Everyday-0.014
 offer-0.014
Token about
Token ID546
Feature activation+8.41e-03

Pos logit contributions
 agreement+0.003
 particular+0.003
date+0.003
 Everyday+0.003
 invite+0.003
Neg logit contributions
w-0.003
ich-0.003
ller-0.003
aged-0.003
llers-0.003
Token her
Token ID607
Feature activation+1.53e-02

Pos logit contributions
 agreement+0.015
 particular+0.014
date+0.014
 Everyday+0.013
 invite+0.013
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
llers-0.010
aged-0.009
Token new
Token ID649
Feature activation-6.67e-03

Pos logit contributions
 agreement+0.023
 Everyday+0.021
date+0.021
 particular+0.021
cloth+0.019
Neg logit contributions
w-0.023
ich-0.022
 reading-0.021
ller-0.020
aged-0.020
Token friend
Token ID1545
Feature activation+6.26e-03

Pos logit contributions
w+0.010
ich+0.009
ller+0.009
aged+0.009
llers+0.009
Neg logit contributions
 agreement-0.011
 particular-0.010
 Everyday-0.010
date-0.010
 invite-0.009
Token Vin
Token ID11820
Feature activation+6.32e-02

Pos logit contributions
 agreement+0.013
 particular+0.012
 Everyday+0.011
ainer+0.011
date+0.011
Neg logit contributions
 reading-0.006
ich-0.006
 crying-0.006
 Pur-0.006
w-0.005
Tokenny
Token ID3281
Feature activation-3.14e-03

Pos logit contributions
 agreement+0.088
date+0.079
 Everyday+0.075
 particular+0.073
 fill+0.073
Neg logit contributions
w-0.102
EEP-0.089
RR-0.086
ouncing-0.085
 Pur-0.083
Token,
Token ID11
Feature activation+2.25e-03

Pos logit contributions
ich+0.004
 reading+0.003
w+0.003
 crying+0.003
aged+0.003
Neg logit contributions
 agreement-0.006
 particular-0.006
 Everyday-0.006
date-0.005
 surely-0.005
Token the
Token ID262
Feature activation+2.34e-03

Pos logit contributions
 agreement+0.004
date+0.004
 particular+0.004
 Everyday+0.004
 offer+0.003
Neg logit contributions
ich-0.003
 crying-0.002
w-0.002
 bleeding-0.002
ller-0.002
Token magic
Token ID5536
Feature activation-2.54e-03

Pos logit contributions
 agreement+0.004
date+0.003
 particular+0.003
 Everyday+0.003
 invite+0.003
Neg logit contributions
w-0.003
ich-0.003
AAAA-0.003
irens-0.003
ller-0.003
Token vine
Token ID17793
Feature activation+2.03e-02

Pos logit contributions
w+0.003
ich+0.003
ller+0.002
uts+0.002
aged+0.002
Neg logit contributions
 agreement-0.005
date-0.004
 Everyday-0.004
 particular-0.004
 invite-0.004
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
 agreement+0.024
date+0.023
 invite+0.022
 Everyday+0.022
 particular+0.022
Neg logit contributions
w-0.014
 bleeding-0.011
 touching-0.011
 reading-0.011
ich-0.011
Token
Token ID198
Feature activation+1.84e-02

Pos logit contributions
 agreement+0.025
 particular+0.022
date+0.022
 invite+0.021
 offer+0.021
Neg logit contributions
w-0.016
 Blood-0.015
aged-0.015
ich-0.014
llers-0.014
Token
Token ID198
Feature activation+2.50e-02

Pos logit contributions
 agreement+0.037
 particular+0.034
date+0.033
 invite+0.033
 Everyday+0.033
Neg logit contributions
w-0.021
ich-0.018
aged-0.016
ller-0.016
irens-0.016
TokenSuddenly
Token ID38582
Feature activation+2.85e-03

Pos logit contributions
 agreement+0.051
 particular+0.045
 offer+0.045
 invite+0.045
date+0.044
Neg logit contributions
w-0.029
ich-0.024
aged-0.023
 Blood-0.023
irens-0.022
Token,
Token ID11
Feature activation+2.24e-02

Pos logit contributions
 agreement+0.004
 particular+0.004
 Everyday+0.004
date+0.004
 invite+0.004
Neg logit contributions
w-0.004
ich-0.004
ller-0.004
llers-0.004
aged-0.004
Token M
Token ID337
Feature activation+6.30e-02

Pos logit contributions
 agreement+0.042
date+0.038
 particular+0.037
 Everyday+0.037
 invite+0.036
Neg logit contributions
w-0.029
ich-0.024
aged-0.024
 Blood-0.023
ller-0.022
Tokenandy
Token ID10757
Feature activation+1.63e-02

Pos logit contributions
 agreement+0.110
date+0.098
 surely+0.092
 particular+0.092
 invite+0.092
Neg logit contributions
ich-0.092
w-0.091
OO-0.078
inches-0.076
EEP-0.076
Token heard
Token ID2982
Feature activation+1.58e-02

Pos logit contributions
 agreement+0.029
 particular+0.027
 Everyday+0.026
date+0.026
 invite+0.025
Neg logit contributions
w-0.022
ich-0.020
ller-0.019
aged-0.019
llers-0.019
Token a
Token ID257
Feature activation+9.78e-03

Pos logit contributions
 agreement+0.030
 particular+0.028
date+0.027
 Everyday+0.027
 invite+0.026
Neg logit contributions
w-0.018
ich-0.016
ller-0.016
aged-0.016
irens-0.015
Token voice
Token ID3809
Feature activation+9.37e-03

Pos logit contributions
 agreement+0.016
 particular+0.015
date+0.015
 Everyday+0.015
 invite+0.014
Neg logit contributions
w-0.013
ich-0.012
ller-0.012
RR-0.011
irens-0.011
Token.
Token ID13
Feature activation+2.67e-02

Pos logit contributions
 agreement+0.016
 particular+0.014
 Everyday+0.014
date+0.014
 invite+0.013
Neg logit contributions
w-0.013
ich-0.011
ller-0.011
irens-0.011
aged-0.011
Token looks
Token ID3073
Feature activation+3.47e-02

Pos logit contributions
 agreement+0.026
date+0.022
 particular+0.022
 offer+0.020
 Everyday+0.020
Neg logit contributions
w-0.021
ller-0.019
llers-0.018
ich-0.018
RR-0.018
Token like
Token ID588
Feature activation+2.68e-02

Pos logit contributions
date+0.061
 agreement+0.057
 particular+0.056
 basis+0.055
 Everyday+0.051
Neg logit contributions
nam-0.039
w-0.038
 fighting-0.036
 crying-0.035
 bleeding-0.035
Token a
Token ID257
Feature activation+2.77e-02

Pos logit contributions
 agreement+0.059
 particular+0.054
date+0.053
 Everyday+0.051
 invite+0.049
Neg logit contributions
w-0.024
nam-0.021
ich-0.019
 happened-0.019
 bleeding-0.019
Token sport
Token ID6332
Feature activation+1.92e-02

Pos logit contributions
 agreement+0.045
 Everyday+0.043
 particular+0.039
date+0.039
 Whenever+0.039
Neg logit contributions
w-0.037
ich-0.035
ller-0.034
RR-0.033
AAAA-0.031
Token I
Token ID314
Feature activation+2.72e-02

Pos logit contributions
 agreement+0.040
 Everyday+0.034
rene+0.034
 particular+0.033
ainer+0.033
Neg logit contributions
w-0.022
ich-0.019
uts-0.018
inches-0.017
 reading-0.017
Token can
Token ID460
Feature activation+6.29e-02

Pos logit contributions
 agreement+0.056
rily+0.049
ainer+0.048
umm+0.047
date+0.046
Neg logit contributions
w-0.034
 happened-0.026
inches-0.023
 tri-0.023
 screamed-0.023
Token do
Token ID466
Feature activation+3.83e-02

Pos logit contributions
 agreement+0.097
rily+0.085
cloth+0.080
umm+0.080
 particular+0.079
Neg logit contributions
uts-0.089
w-0.087
inches-0.078
 touching-0.074
 crying-0.074
Token there
Token ID612
Feature activation+2.26e-02

Pos logit contributions
 agreement+0.078
 particular+0.070
 Everyday+0.069
date+0.066
 invite+0.063
Neg logit contributions
w-0.046
uts-0.039
aged-0.037
ller-0.036
 reading-0.036
Token.
Token ID13
Feature activation+5.15e-04

Pos logit contributions
 agreement+0.045
 particular+0.038
latable+0.037
date+0.037
 Everyday+0.037
Neg logit contributions
w-0.027
uts-0.023
ich-0.022
 reading-0.022
ller-0.022
Token
Token ID198
Feature activation-1.08e-02

Pos logit contributions
 agreement+0.001
 particular+0.001
date+0.001
 invite+0.001
 Everyday+0.001
Neg logit contributions
w-0.001
ller-0.001
ich-0.001
llers-0.001
aged-0.001
Token
Token ID198
Feature activation-9.42e-03

Pos logit contributions
w+0.016
ich+0.014
aged+0.014
ller+0.014
llers+0.013
Neg logit contributions
 agreement-0.018
 particular-0.017
 Everyday-0.017
date-0.016
 invite-0.015
Token like
Token ID588
Feature activation+3.28e-02

Pos logit contributions
 agreement+0.053
date+0.052
 Everyday+0.046
 particular+0.046
 basis+0.045
Neg logit contributions
nam-0.028
 bleeding-0.027
 crying-0.027
 flashing-0.026
w-0.026
Token the
Token ID262
Feature activation+3.04e-02

Pos logit contributions
 agreement+0.073
date+0.067
 particular+0.065
 Everyday+0.065
 offer+0.062
Neg logit contributions
w-0.024
nam-0.024
 flashing-0.023
 reading-0.022
ich-0.022
Token march
Token ID9960
Feature activation+3.90e-02

Pos logit contributions
 agreement+0.052
date+0.051
rily+0.049
ainer+0.047
 Everyday+0.047
Neg logit contributions
w-0.033
ich-0.033
 flashing-0.029
nam-0.029
 Pur-0.029
Token was
Token ID373
Feature activation+3.28e-02

Pos logit contributions
 agreement+0.075
 Everyday+0.061
ainer+0.060
 particular+0.060
latable+0.059
Neg logit contributions
w-0.045
uts-0.043
 happened-0.041
 flashing-0.040
EEP-0.040
Token going
Token ID1016
Feature activation+4.97e-02

Pos logit contributions
 agreement+0.067
date+0.060
 Everyday+0.060
 offer+0.057
 eater+0.057
Neg logit contributions
 flashing-0.031
w-0.031
 pressing-0.030
 bleeding-0.029
 touching-0.029
Token to
Token ID284
Feature activation+6.25e-02

Pos logit contributions
 agreement+0.103
 particular+0.087
date+0.087
 Everyday+0.087
 offer+0.086
Neg logit contributions
w-0.055
irens-0.049
ich-0.047
ller-0.047
llers-0.044
Token be
Token ID307
Feature activation+1.42e-02

Pos logit contributions
 agreement+0.101
date+0.100
 eater+0.095
rily+0.093
 Everyday+0.091
Neg logit contributions
w-0.070
 pressing-0.067
 crossing-0.066
AAAA-0.066
 Pur-0.066
Token delayed
Token ID11038
Feature activation+1.38e-02

Pos logit contributions
 agreement+0.027
 Everyday+0.025
 particular+0.024
 offer+0.023
date+0.023
Neg logit contributions
w-0.017
ich-0.016
aged-0.015
irens-0.015
llers-0.014
Token.
Token ID13
Feature activation-3.54e-03

Pos logit contributions
 agreement+0.027
 Everyday+0.025
date+0.024
 particular+0.024
 invite+0.024
Neg logit contributions
w-0.014
nam-0.013
 reading-0.013
ich-0.013
aged-0.013
Token While
Token ID2893
Feature activation+1.08e-02

Pos logit contributions
w+0.005
ich+0.004
ller+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.006
 particular-0.006
date-0.006
 invite-0.005
 Everyday-0.005
Token she
Token ID673
Feature activation+2.61e-02

Pos logit contributions
 agreement+0.023
 particular+0.021
 Everyday+0.021
date+0.021
 offer+0.020
Neg logit contributions
w-0.011
ich-0.009
aged-0.008
ller-0.008
irens-0.008
Token in
Token ID287
Feature activation+1.74e-02

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
aged+0.000
llers+0.000
Neg logit contributions
 agreement-0.000
 particular-0.000
 Everyday-0.000
date-0.000
 invite-0.000
Token the
Token ID262
Feature activation+5.69e-03

Pos logit contributions
 agreement+0.035
 particular+0.033
date+0.033
 Everyday+0.032
 offer+0.031
Neg logit contributions
w-0.014
ich-0.014
irens-0.014
llers-0.014
ller-0.014
Token rain
Token ID6290
Feature activation-2.10e-02

Pos logit contributions
 agreement+0.010
 particular+0.009
 surely+0.009
 offer+0.009
date+0.009
Neg logit contributions
w-0.007
ich-0.007
irens-0.006
 bleeding-0.006
AAAA-0.006
Token,"
Token ID553
Feature activation-9.77e-06

Pos logit contributions
w+0.024
ich+0.020
ller+0.020
llers+0.019
aged+0.019
Neg logit contributions
 agreement-0.043
 particular-0.039
 Everyday-0.037
date-0.037
 invite-0.036
Token said
Token ID531
Feature activation+1.38e-02

Feature not active on prev token;
no predictions were affected.
Token M
Token ID337
Feature activation+6.25e-02

Pos logit contributions
 agreement+0.021
 Everyday+0.018
 particular+0.018
date+0.018
 invite+0.017
Neg logit contributions
w-0.021
aged-0.020
 Blood-0.018
ich-0.018
ller-0.018
Tokenandy
Token ID10757
Feature activation+9.78e-03

Pos logit contributions
 agreement+0.104
date+0.095
 surely+0.092
 fill+0.089
 invite+0.088
Neg logit contributions
w-0.091
ich-0.087
inches-0.080
OO-0.078
uts-0.078
Token.
Token ID13
Feature activation+8.88e-03

Pos logit contributions
 agreement+0.015
 particular+0.014
date+0.013
 Everyday+0.013
 invite+0.012
Neg logit contributions
w-0.015
aged-0.015
ller-0.014
ich-0.014
llers-0.013
Token
Token ID220
Feature activation-2.24e-02

Pos logit contributions
 agreement+0.016
 particular+0.014
 Everyday+0.014
date+0.014
 invite+0.013
Neg logit contributions
w-0.012
ich-0.011
ller-0.011
aged-0.010
llers-0.010
Token
Token ID198
Feature activation+5.83e-03

Pos logit contributions
w+0.035
ich+0.032
ller+0.030
llers+0.030
RR+0.029
Neg logit contributions
 agreement-0.034
 particular-0.034
 Everyday-0.033
date-0.030
 invite-0.030
Token
Token ID198
Feature activation+1.32e-02

Pos logit contributions
 agreement+0.010
 particular+0.010
 Everyday+0.009
date+0.009
 invite+0.009
Neg logit contributions
w-0.008
ich-0.007
ller-0.007
llers-0.007
aged-0.007
Token her
Token ID607
Feature activation+3.65e-02

Pos logit contributions
 agreement+0.031
date+0.028
 particular+0.028
 Everyday+0.028
latable+0.027
Neg logit contributions
w-0.023
ich-0.021
aged-0.021
ller-0.021
llers-0.021
Token mom
Token ID1995
Feature activation+2.48e-02

Pos logit contributions
date+0.069
 agreement+0.068
rant+0.067
 Everyday+0.065
igator+0.064
Neg logit contributions
ich-0.036
 reading-0.032
 crying-0.031
ller-0.031
nam-0.030
Token for
Token ID329
Feature activation+2.13e-02

Pos logit contributions
 agreement+0.051
latable+0.049
 particular+0.049
date+0.048
ainer+0.048
Neg logit contributions
 reading-0.021
 crying-0.019
 praying-0.018
 sitting-0.017
ich-0.017
Token permitting
Token ID29256
Feature activation+1.10e-02

Pos logit contributions
 Everyday+0.043
 particular+0.043
 agreement+0.043
date+0.040
 invite+0.039
Neg logit contributions
w-0.020
ich-0.018
aged-0.017
nam-0.017
 reading-0.017
Token her
Token ID607
Feature activation+4.46e-02

Pos logit contributions
 agreement+0.022
 particular+0.021
 Everyday+0.021
date+0.020
latable+0.020
Neg logit contributions
w-0.010
irens-0.009
aged-0.008
 reading-0.008
 Pur-0.008
Token to
Token ID284
Feature activation+6.23e-02

Pos logit contributions
 agreement+0.094
 particular+0.087
date+0.084
 Everyday+0.082
rily+0.079
Neg logit contributions
w-0.043
uts-0.038
aged-0.037
EEP-0.037
ich-0.037
Token march
Token ID9960
Feature activation+3.78e-02

Pos logit contributions
rily+0.089
 eater+0.086
 agreement+0.083
igator+0.081
 particular+0.079
Neg logit contributions
w-0.084
 riding-0.080
 Pur-0.079
 crossing-0.077
EEP-0.077
Token.
Token ID13
Feature activation+4.90e-04

Pos logit contributions
 agreement+0.070
 Everyday+0.066
 particular+0.065
 eater+0.061
latable+0.061
Neg logit contributions
w-0.037
uts-0.033
ller-0.033
irens-0.033
aged-0.032
Token
Token ID198
Feature activation-2.01e-03

Pos logit contributions
 agreement+0.001
 particular+0.001
date+0.001
 invite+0.001
 offer+0.001
Neg logit contributions
w-0.001
ich-0.000
ller-0.000
llers-0.000
aged-0.000
Token
Token ID198
Feature activation+4.20e-03

Pos logit contributions
w+0.003
ich+0.002
ller+0.002
aged+0.002
llers+0.002
Neg logit contributions
 agreement-0.004
 particular-0.003
date-0.003
 Everyday-0.003
 invite-0.003
TokenThe
Token ID464
Feature activation+2.44e-02

Pos logit contributions
 agreement+0.008
 particular+0.007
 fill+0.007
 offer+0.007
 Everyday+0.007
Neg logit contributions
w-0.005
ich-0.005
aged-0.004
irens-0.004
AAAA-0.004
Token look
Token ID804
Feature activation+8.78e-03

Pos logit contributions
rily+0.078
rant+0.075
umm+0.074
year+0.070
cloth+0.069
Neg logit contributions
 riding-0.047
 fighting-0.047
 hitting-0.047
 pushing-0.046
 touching-0.045
Token for
Token ID329
Feature activation+3.02e-02

Pos logit contributions
 agreement+0.016
date+0.016
 particular+0.016
 Everyday+0.015
 invite+0.015
Neg logit contributions
w-0.010
nam-0.009
ller-0.008
aged-0.008
inches-0.008
Token a
Token ID257
Feature activation+2.36e-02

Pos logit contributions
 particular+0.056
 agreement+0.056
date+0.055
 surely+0.053
 Everyday+0.053
Neg logit contributions
nam-0.030
w-0.027
llers-0.027
irens-0.026
ller-0.025
Token sport
Token ID6332
Feature activation+2.32e-02

Pos logit contributions
 agreement+0.039
 Everyday+0.036
 particular+0.034
date+0.033
cloth+0.032
Neg logit contributions
w-0.029
ich-0.028
ller-0.026
EEP-0.026
AAAA-0.026
Token he
Token ID339
Feature activation+2.38e-02

Pos logit contributions
 agreement+0.050
 Everyday+0.045
 particular+0.044
rene+0.044
ainer+0.044
Neg logit contributions
w-0.020
uts-0.019
ich-0.019
inches-0.018
 reading-0.017
Token could
Token ID714
Feature activation+6.23e-02

Pos logit contributions
 agreement+0.051
ainer+0.046
date+0.042
rily+0.041
 particular+0.041
Neg logit contributions
w-0.028
 tri-0.023
 happened-0.022
 screamed-0.022
ounced-0.021
Token do
Token ID466
Feature activation+3.83e-02

Pos logit contributions
umm+0.098
 agreement+0.098
isher+0.097
rily+0.097
rant+0.092
Neg logit contributions
uts-0.071
 touching-0.064
w-0.062
 riding-0.061
 swinging-0.060
Token.
Token ID13
Feature activation-7.26e-03

Pos logit contributions
 agreement+0.074
 Everyday+0.067
 particular+0.066
date+0.061
 invite+0.061
Neg logit contributions
w-0.048
aged-0.042
uts-0.041
 reading-0.038
ich-0.037
Token Then
Token ID3244
Feature activation-1.43e-02

Pos logit contributions
w+0.013
ich+0.012
ller+0.011
aged+0.011
llers+0.011
Neg logit contributions
 agreement-0.010
 particular-0.009
 Everyday-0.009
date-0.008
 invite-0.008
Token he
Token ID339
Feature activation+2.07e-02

Pos logit contributions
w+0.021
ich+0.020
ller+0.018
llers+0.018
aged+0.018
Neg logit contributions
 agreement-0.024
 particular-0.023
 Everyday-0.021
date-0.021
 invite-0.021
Token saw
Token ID2497
Feature activation+1.89e-03

Pos logit contributions
 agreement+0.043
 particular+0.039
 Everyday+0.038
date+0.038
 offer+0.037
Neg logit contributions
w-0.023
ich-0.019
ounced-0.018
aged-0.018
inches-0.018
Token shows
Token ID2523
Feature activation+5.91e-03

Pos logit contributions
 agreement+0.017
date+0.017
 particular+0.016
 basis+0.015
 Everyday+0.015
Neg logit contributions
w-0.005
inches-0.005
okes-0.004
ller-0.004
ls-0.003
Token them
Token ID606
Feature activation+1.58e-02

Pos logit contributions
 agreement+0.009
 particular+0.008
date+0.008
 Everyday+0.008
 invite+0.007
Neg logit contributions
w-0.009
aged-0.008
irens-0.008
ller-0.008
llers-0.008
Token a
Token ID257
Feature activation+1.31e-02

Pos logit contributions
 agreement+0.032
date+0.029
 particular+0.029
 Everyday+0.027
 invite+0.026
Neg logit contributions
w-0.017
aged-0.015
llers-0.014
inches-0.014
uts-0.014
Token big
Token ID1263
Feature activation+2.57e-02

Pos logit contributions
 agreement+0.021
 Everyday+0.019
date+0.019
 particular+0.018
 invite+0.017
Neg logit contributions
w-0.019
ich-0.017
ller-0.017
AAAA-0.016
 bleeding-0.016
Token tub
Token ID12202
Feature activation+1.94e-02

Pos logit contributions
 agreement+0.037
 particular+0.035
date+0.034
 Everyday+0.034
 invite+0.032
Neg logit contributions
w-0.040
ller-0.035
AAAA-0.034
 bleeding-0.033
ich-0.033
Token full
Token ID1336
Feature activation+6.21e-02

Pos logit contributions
 agreement+0.042
rene+0.036
 Everyday+0.036
 particular+0.035
 basis+0.034
Neg logit contributions
w-0.020
uts-0.019
o-0.017
aged-0.016
ages-0.016
Token of
Token ID286
Feature activation+2.53e-02

Pos logit contributions
rene+0.103
 agreement+0.096
date+0.095
 surely+0.092
zzy+0.090
Neg logit contributions
w-0.071
uts-0.067
 bleeding-0.067
s-0.065
EEP-0.065
Token water
Token ID1660
Feature activation+1.62e-02

Pos logit contributions
date+0.045
 agreement+0.044
 surely+0.042
 particular+0.042
 Everyday+0.039
Neg logit contributions
w-0.034
 elbows-0.026
nam-0.026
 bleeding-0.026
 blood-0.026
Token and
Token ID290
Feature activation+7.20e-03

Pos logit contributions
 agreement+0.036
 particular+0.033
date+0.031
 Everyday+0.030
rene+0.030
Neg logit contributions
w-0.018
uts-0.016
inches-0.015
ages-0.014
aged-0.014
Token foam
Token ID19828
Feature activation+9.23e-03

Pos logit contributions
 agreement+0.013
date+0.013
 particular+0.012
 Everyday+0.011
 basis+0.011
Neg logit contributions
w-0.009
aged-0.008
nam-0.008
 elbows-0.008
llers-0.008
Token.
Token ID13
Feature activation-4.16e-04

Pos logit contributions
 agreement+0.018
 particular+0.017
 Everyday+0.016
date+0.016
 surely+0.015
Neg logit contributions
w-0.012
uts-0.011
aged-0.011
inches-0.010
ages-0.010
Token to
Token ID284
Feature activation+1.97e-02

Pos logit contributions
w+0.005
ich+0.004
aged+0.004
ller+0.004
llers+0.004
Neg logit contributions
 agreement-0.007
 particular-0.007
date-0.006
 Everyday-0.006
 invite-0.006
Token discuss
Token ID2112
Feature activation+7.72e-03

Pos logit contributions
date+0.033
 agreement+0.030
tted+0.029
rily+0.029
kin+0.029
Neg logit contributions
nam-0.023
 touching-0.021
 reading-0.020
 examining-0.019
 fighting-0.019
Token the
Token ID262
Feature activation+6.14e-04

Pos logit contributions
 agreement+0.015
date+0.015
 particular+0.014
 invite+0.014
 surely+0.014
Neg logit contributions
w-0.008
nam-0.008
ller-0.007
llers-0.007
ich-0.007
Token berries
Token ID36322
Feature activation+1.94e-02

Pos logit contributions
 agreement+0.001
date+0.001
 surely+0.001
 particular+0.001
 invite+0.001
Neg logit contributions
w-0.001
ich-0.001
aged-0.001
llers-0.001
 Blood-0.001
Token,
Token ID11
Feature activation-6.28e-03

Pos logit contributions
 agreement+0.040
date+0.037
 particular+0.036
 Everyday+0.034
 invite+0.033
Neg logit contributions
uts-0.022
w-0.019
ls-0.017
es-0.017
 crying-0.017
Token M
Token ID337
Feature activation+6.17e-02

Pos logit contributions
nam+0.007
 reading+0.007
llers+0.006
 touching+0.006
 crying+0.006
Neg logit contributions
date-0.012
 agreement-0.011
 particular-0.011
 surely-0.010
 date-0.010
Tokenilly
Token ID6548
Feature activation+3.32e-03

Pos logit contributions
 agreement+0.106
 particular+0.092
 surely+0.092
date+0.092
 tow+0.091
Neg logit contributions
w-0.087
ich-0.086
llers-0.077
inches-0.075
uts-0.075
Token thought
Token ID1807
Feature activation+2.98e-02

Pos logit contributions
 agreement+0.006
 particular+0.006
date+0.006
 surely+0.005
 Everyday+0.005
Neg logit contributions
aged-0.004
w-0.004
ich-0.003
ller-0.003
uts-0.003
Token they
Token ID484
Feature activation+9.73e-03

Pos logit contributions
 agreement+0.072
date+0.070
ainer+0.061
 surely+0.060
 host+0.060
Neg logit contributions
w-0.021
 happened-0.021
ls-0.019
ons-0.018
uts-0.017
Token were
Token ID547
Feature activation+4.81e-03

Pos logit contributions
 agreement+0.021
date+0.018
 particular+0.018
ainer+0.017
 basis+0.017
Neg logit contributions
w-0.011
nam-0.009
aged-0.009
uts-0.008
llers-0.008
Token delicious
Token ID12625
Feature activation+8.38e-03

Pos logit contributions
 agreement+0.009
date+0.009
 particular+0.009
 Everyday+0.008
 surely+0.008
Neg logit contributions
nam-0.006
 reading-0.005
ich-0.005
w-0.005
 praying-0.005
Token like
Token ID588
Feature activation+3.69e-04

Pos logit contributions
w+0.006
aged+0.005
EEP+0.005
uts+0.005
ich+0.005
Neg logit contributions
 agreement-0.005
 particular-0.005
 Everyday-0.005
date-0.005
latable-0.004
Token the
Token ID262
Feature activation+2.33e-02

Pos logit contributions
 agreement+0.001
date+0.001
 particular+0.001
 Everyday+0.001
 offer+0.001
Neg logit contributions
w-0.000
llers-0.000
irens-0.000
ller-0.000
ich-0.000
Token sunscreen
Token ID44221
Feature activation+1.76e-02

Pos logit contributions
date+0.040
 agreement+0.037
 surely+0.037
 Everyday+0.037
iona+0.035
Neg logit contributions
w-0.025
 flashing-0.024
 blood-0.024
 elbows-0.023
 crying-0.023
Token because
Token ID780
Feature activation+1.85e-02

Pos logit contributions
 agreement+0.037
 particular+0.031
date+0.030
 Everyday+0.030
iona+0.029
Neg logit contributions
w-0.020
uts-0.019
 bleeding-0.018
ich-0.018
aged-0.018
Token it
Token ID340
Feature activation+1.69e-02

Pos logit contributions
 agreement+0.045
date+0.040
 invite+0.039
 offer+0.038
 basis+0.037
Neg logit contributions
w-0.016
irens-0.012
ller-0.011
uts-0.011
nam-0.011
Token felt
Token ID2936
Feature activation+6.17e-02

Pos logit contributions
 agreement+0.034
date+0.027
 particular+0.026
 Everyday+0.026
 offer+0.025
Neg logit contributions
w-0.024
llers-0.020
aged-0.020
ller-0.020
ich-0.019
Token tight
Token ID5381
Feature activation+1.73e-02

Pos logit contributions
date+0.136
 agreement+0.129
 particular+0.116
 basis+0.114
iona+0.112
Neg logit contributions
 bleeding-0.057
 hurting-0.048
 crying-0.047
 breathing-0.045
w-0.045
Token on
Token ID319
Feature activation+1.25e-02

Pos logit contributions
 agreement+0.033
 particular+0.032
date+0.029
 Everyday+0.029
 invite+0.028
Neg logit contributions
w-0.022
aged-0.021
ich-0.019
uts-0.019
 bleeding-0.019
Token his
Token ID465
Feature activation+1.48e-02

Pos logit contributions
 agreement+0.024
date+0.023
 particular+0.022
 Everyday+0.022
 invite+0.022
Neg logit contributions
w-0.015
aged-0.013
ller-0.012
ich-0.012
llers-0.012
Token skin
Token ID4168
Feature activation+1.69e-02

Pos logit contributions
 agreement+0.026
 Everyday+0.025
 particular+0.025
date+0.024
 invite+0.023
Neg logit contributions
w-0.019
ich-0.018
aged-0.018
 elbows-0.018
ller-0.017
Token.
Token ID13
Feature activation-6.32e-03

Pos logit contributions
 agreement+0.034
 particular+0.030
date+0.030
 invite+0.028
 Everyday+0.028
Neg logit contributions
w-0.021
uts-0.020
aged-0.019
ller-0.019
 bleeding-0.019
Token to
Token ID284
Feature activation+4.71e-02

Pos logit contributions
 agreement+0.043
 particular+0.040
date+0.039
 Everyday+0.039
 invite+0.037
Neg logit contributions
w-0.025
ller-0.023
ich-0.022
aged-0.022
llers-0.021
Token permit
Token ID8749
Feature activation+1.99e-03

Pos logit contributions
rily+0.078
umm+0.078
 eater+0.076
ited+0.072
rant+0.070
Neg logit contributions
 Hel-0.048
w-0.048
 riding-0.046
 pressing-0.046
EEP-0.046
Token the
Token ID262
Feature activation+1.61e-02

Pos logit contributions
 agreement+0.004
 particular+0.004
date+0.004
 Everyday+0.004
 invite+0.003
Neg logit contributions
w-0.002
ller-0.002
llers-0.002
irens-0.002
nam-0.002
Token little
Token ID1310
Feature activation+3.46e-03

Pos logit contributions
 agreement+0.031
rant+0.029
rily+0.029
cloth+0.028
date+0.028
Neg logit contributions
ich-0.016
w-0.014
EEP-0.013
 happened-0.013
 crying-0.013
Token girl
Token ID2576
Feature activation+3.07e-02

Pos logit contributions
 agreement+0.007
 Everyday+0.006
latable+0.006
 particular+0.006
date+0.006
Neg logit contributions
w-0.003
ich-0.003
 crying-0.003
nam-0.003
 happened-0.002
Token to
Token ID284
Feature activation+6.15e-02

Pos logit contributions
 agreement+0.078
 particular+0.070
ainer+0.068
date+0.068
 Everyday+0.067
Neg logit contributions
aged-0.016
 reading-0.015
uts-0.015
w-0.015
ages-0.014
Token march
Token ID9960
Feature activation+2.16e-02

Pos logit contributions
rily+0.103
umm+0.099
 eater+0.099
 agreement+0.093
rant+0.089
Neg logit contributions
w-0.070
EEP-0.068
 Hel-0.064
 Pur-0.063
 riding-0.063
Token.
Token ID13
Feature activation-7.87e-03

Pos logit contributions
 agreement+0.042
 Everyday+0.039
 particular+0.038
latable+0.037
ainer+0.036
Neg logit contributions
w-0.023
nam-0.020
aged-0.019
irens-0.018
 crying-0.018
Token
Token ID198
Feature activation-4.27e-03

Pos logit contributions
w+0.012
ich+0.011
ller+0.010
aged+0.010
llers+0.010
Neg logit contributions
 agreement-0.013
 particular-0.012
 Everyday-0.012
date-0.011
 invite-0.011
Token
Token ID198
Feature activation+2.09e-03

Pos logit contributions
w+0.005
ich+0.005
ller+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.008
 particular-0.008
 Everyday-0.007
date-0.007
 invite-0.007
TokenThe
Token ID464
Feature activation+1.69e-02

Pos logit contributions
 agreement+0.004
 particular+0.003
date+0.003
 Everyday+0.003
 offer+0.003
Neg logit contributions
w-0.003
ich-0.003
ller-0.003
llers-0.003
RR-0.003
Token Let
Token ID3914
Feature activation+3.17e-02

Pos logit contributions
 agreement+0.037
 particular+0.036
date+0.034
 choice+0.033
 offer+0.032
Neg logit contributions
ich-0.017
w-0.016
EEP-0.016
llers-0.016
 Pur-0.016
Token's
Token ID338
Feature activation+1.77e-02

Pos logit contributions
 agreement+0.068
 particular+0.064
 Everyday+0.060
date+0.060
 offer+0.059
Neg logit contributions
w-0.035
ich-0.027
aged-0.024
ller-0.024
EEP-0.024
Token go
Token ID467
Feature activation+1.10e-02

Pos logit contributions
 agreement+0.029
rily+0.028
year+0.028
mer+0.027
 particular+0.027
Neg logit contributions
w-0.019
 Brown-0.017
 reading-0.016
 riding-0.016
 not-0.015
Token home
Token ID1363
Feature activation+2.33e-03

Pos logit contributions
 agreement+0.022
 particular+0.021
 Everyday+0.020
date+0.019
 unison+0.019
Neg logit contributions
w-0.012
ich-0.010
shadow-0.010
enn-0.010
 reading-0.009
Token!"
Token ID2474
Feature activation+2.56e-02

Pos logit contributions
 agreement+0.004
 particular+0.003
 Everyday+0.003
date+0.003
 invite+0.003
Neg logit contributions
w-0.004
ich-0.003
ller-0.003
aged-0.003
llers-0.003
Token Mack
Token ID16640
Feature activation+6.15e-02

Pos logit contributions
 agreement+0.046
 particular+0.043
date+0.043
 invite+0.040
 Everyday+0.040
Neg logit contributions
w-0.033
ich-0.028
aged-0.028
inches-0.027
ller-0.027
Tokenenzie
Token ID26389
Feature activation+2.04e-02

Pos logit contributions
 agreement+0.107
date+0.106
 tow+0.104
 checkout+0.097
 fill+0.096
Neg logit contributions
w-0.089
RR-0.063
awed-0.063
es-0.061
ouncing-0.061
Token and
Token ID290
Feature activation+5.62e-03

Pos logit contributions
 agreement+0.037
 particular+0.035
date+0.034
 Everyday+0.031
 offer+0.030
Neg logit contributions
aged-0.025
w-0.024
ller-0.023
ages-0.022
ich-0.022
Token her
Token ID607
Feature activation+2.67e-02

Pos logit contributions
date+0.010
 agreement+0.010
latable+0.009
 particular+0.009
 Everyday+0.009
Neg logit contributions
w-0.006
aged-0.006
 Pur-0.005
nam-0.005
 crying-0.005
Token mom
Token ID1995
Feature activation+9.54e-04

Pos logit contributions
 agreement+0.055
date+0.054
 Everyday+0.052
 particular+0.051
rily+0.051
Neg logit contributions
ich-0.022
w-0.021
 crying-0.021
 bleeding-0.020
aged-0.019
Tokenmy
Token ID1820
Feature activation+1.62e-03

Pos logit contributions
 agreement+0.002
 particular+0.002
date+0.002
 Everyday+0.002
 offer+0.002
Neg logit contributions
w-0.001
aged-0.001
ich-0.001
ller-0.001
llers-0.001
Token like
Token ID588
Feature activation+2.12e-02

Pos logit contributions
 agreement+0.037
 checkout+0.037
umm+0.036
rily+0.036
 particular+0.035
Neg logit contributions
w-0.032
 reading-0.024
 whist-0.023
 listening-0.022
shadow-0.022
Token to
Token ID284
Feature activation+4.66e-02

Pos logit contributions
 agreement+0.046
date+0.042
 particular+0.042
 surely+0.040
rily+0.039
Neg logit contributions
w-0.020
 reading-0.016
ich-0.014
 praying-0.014
irens-0.014
Token join
Token ID4654
Feature activation+7.14e-04

Pos logit contributions
rily+0.075
 agreement+0.069
cloth+0.069
umm+0.069
 eater+0.068
Neg logit contributions
w-0.056
ich-0.050
 flying-0.050
 Pur-0.049
 whist-0.049
Token me
Token ID502
Feature activation+3.05e-02

Pos logit contributions
 agreement+0.001
 particular+0.001
 Everyday+0.001
 surely+0.001
date+0.001
Neg logit contributions
w-0.001
ich-0.001
RR-0.001
 reading-0.001
icken-0.001
Token on
Token ID319
Feature activation+2.41e-02

Pos logit contributions
 agreement+0.063
 particular+0.058
 Everyday+0.055
 Whenever+0.053
cloth+0.052
Neg logit contributions
w-0.032
ich-0.028
 reading-0.025
ller-0.025
inches-0.024
Token an
Token ID281
Feature activation+6.12e-02

Pos logit contributions
 agreement+0.047
 particular+0.042
date+0.040
 offer+0.040
 Everyday+0.040
Neg logit contributions
w-0.029
ich-0.026
ller-0.025
llers-0.024
irens-0.023
Token adventure
Token ID8855
Feature activation+3.02e-02

Pos logit contributions
cloth+0.089
date+0.085
 agreement+0.081
 surely+0.080
tted+0.076
Neg logit contributions
ich-0.075
ller-0.075
w-0.073
 praying-0.068
llers-0.067
Token?"
Token ID1701
Feature activation+1.02e-02

Pos logit contributions
 agreement+0.071
 particular+0.062
 Everyday+0.061
 Whenever+0.061
 tow+0.060
Neg logit contributions
w-0.023
ich-0.019
ller-0.017
 reading-0.016
enn-0.014
Token
Token ID220
Feature activation-5.32e-03

Pos logit contributions
 agreement+0.018
 particular+0.017
date+0.016
 Everyday+0.015
 invite+0.015
Neg logit contributions
w-0.014
ller-0.012
ich-0.012
llers-0.012
aged-0.012
Token
Token ID198
Feature activation+6.90e-03

Pos logit contributions
w+0.007
ich+0.006
ller+0.006
aged+0.006
llers+0.006
Neg logit contributions
 agreement-0.010
 particular-0.009
 Everyday-0.009
date-0.009
 invite-0.008
Token
Token ID198
Feature activation+9.03e-03

Pos logit contributions
 agreement+0.013
 particular+0.012
 Everyday+0.012
date+0.011
 invite+0.011
Neg logit contributions
w-0.009
ich-0.008
ller-0.008
aged-0.007
llers-0.007
Token.
Token ID13
Feature activation+2.10e-02

Pos logit contributions
w+0.001
ich+0.001
uts+0.001
 reading+0.001
ller+0.001
Neg logit contributions
 agreement-0.002
 particular-0.001
 Everyday-0.001
date-0.001
 offer-0.001
Token
Token ID198
Feature activation+1.50e-02

Pos logit contributions
 agreement+0.037
date+0.033
 particular+0.033
 offer+0.032
 invite+0.031
Neg logit contributions
w-0.026
 Blood-0.025
ich-0.024
aged-0.024
llers-0.024
Token
Token ID198
Feature activation+2.17e-02

Pos logit contributions
 agreement+0.030
 particular+0.028
date+0.027
 Everyday+0.027
 invite+0.026
Neg logit contributions
w-0.017
ich-0.015
ller-0.014
aged-0.014
llers-0.014
TokenSuddenly
Token ID38582
Feature activation+2.24e-03

Pos logit contributions
 agreement+0.043
 offer+0.039
 Everyday+0.038
 particular+0.038
 invite+0.037
Neg logit contributions
w-0.026
ich-0.021
 Blood-0.020
aged-0.019
RR-0.019
Token,
Token ID11
Feature activation+1.92e-02

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
date+0.003
 invite+0.003
Neg logit contributions
w-0.004
ich-0.003
ller-0.003
llers-0.003
aged-0.003
Token M
Token ID337
Feature activation+6.10e-02

Pos logit contributions
 agreement+0.035
 particular+0.031
date+0.031
 Everyday+0.030
 offer+0.030
Neg logit contributions
w-0.026
ich-0.022
aged-0.021
ller-0.021
 Blood-0.021
Tokenandy
Token ID10757
Feature activation+3.47e-03

Pos logit contributions
 agreement+0.098
date+0.086
 particular+0.081
 surely+0.079
 invite+0.079
Neg logit contributions
w-0.098
ich-0.096
inches-0.086
ouncing-0.084
icken-0.084
Token noticed
Token ID6810
Feature activation+1.69e-02

Pos logit contributions
 agreement+0.005
 particular+0.005
 Everyday+0.005
 invite+0.005
date+0.005
Neg logit contributions
w-0.005
ich-0.005
irens-0.005
ller-0.005
nam-0.005
Token something
Token ID1223
Feature activation+9.47e-04

Pos logit contributions
 agreement+0.035
 particular+0.032
date+0.031
 invite+0.031
 Everyday+0.031
Neg logit contributions
w-0.019
ich-0.016
ller-0.016
aged-0.015
llers-0.015
Token shiny
Token ID22441
Feature activation-2.35e-03

Pos logit contributions
 agreement+0.002
date+0.002
 Everyday+0.002
 particular+0.001
 invite+0.001
Neg logit contributions
w-0.001
ich-0.001
 bleeding-0.001
ller-0.001
aged-0.001
Token on
Token ID319
Feature activation+3.05e-02

Pos logit contributions
w+0.003
 bleeding+0.002
ller+0.002
 happened+0.002
awed+0.002
Neg logit contributions
 agreement-0.005
date-0.005
 invite-0.005
 Everyday-0.005
 particular-0.005
Token j
Token ID474
Feature activation+6.95e-03

Pos logit contributions
w+0.011
ich+0.010
ller+0.010
aged+0.010
llers+0.009
Neg logit contributions
 agreement-0.014
 particular-0.013
 Everyday-0.012
date-0.012
 invite-0.012
Tokenogging
Token ID30853
Feature activation+2.87e-02

Pos logit contributions
 agreement+0.013
 particular+0.012
 invite+0.011
 Everyday+0.011
date+0.011
Neg logit contributions
w-0.010
ich-0.008
inches-0.008
irens-0.008
aged-0.008
Token is
Token ID318
Feature activation+2.26e-02

Pos logit contributions
 agreement+0.051
 particular+0.044
date+0.041
 Everyday+0.041
 invite+0.041
Neg logit contributions
w-0.043
aged-0.036
ich-0.036
ller-0.035
uts-0.034
Token a
Token ID257
Feature activation+1.87e-02

Pos logit contributions
 agreement+0.048
date+0.046
rily+0.042
 particular+0.041
 Everyday+0.040
Neg logit contributions
w-0.021
 bleeding-0.019
aged-0.019
ich-0.017
llers-0.017
Token way
Token ID835
Feature activation+1.81e-02

Pos logit contributions
 agreement+0.032
 Everyday+0.031
rily+0.029
date+0.029
 Whenever+0.027
Neg logit contributions
ich-0.023
w-0.023
RR-0.021
AAAA-0.020
EEP-0.020
Token to
Token ID284
Feature activation+6.07e-02

Pos logit contributions
 agreement+0.038
date+0.034
 particular+0.033
 Everyday+0.032
 invite+0.031
Neg logit contributions
w-0.021
ich-0.018
uts-0.017
ller-0.017
aged-0.016
Token exercise
Token ID5517
Feature activation+3.66e-02

Pos logit contributions
date+0.086
rily+0.085
 agreement+0.083
 eater+0.080
umm+0.080
Neg logit contributions
w-0.081
 crossing-0.077
 touching-0.076
 Pur-0.072
 crying-0.072
Token and
Token ID290
Feature activation+3.78e-03

Pos logit contributions
 agreement+0.067
 particular+0.059
date+0.058
 Everyday+0.058
 invite+0.057
Neg logit contributions
w-0.048
uts-0.038
ich-0.038
inches-0.038
irens-0.038
Token stay
Token ID2652
Feature activation+5.49e-03

Pos logit contributions
 agreement+0.006
 particular+0.005
date+0.005
 Everyday+0.005
rily+0.004
Neg logit contributions
w-0.006
llers-0.005
aged-0.005
ich-0.005
ller-0.005
Token healthy
Token ID5448
Feature activation+1.65e-02

Pos logit contributions
 Everyday+0.010
 agreement+0.009
 particular+0.009
date+0.009
 Whenever+0.009
Neg logit contributions
w-0.007
aged-0.006
ich-0.006
shadow-0.006
 bleeding-0.006
Token.
Token ID13
Feature activation-2.29e-03

Pos logit contributions
 agreement+0.034
 particular+0.034
 Everyday+0.031
date+0.031
 tow+0.030
Neg logit contributions
w-0.016
uts-0.014
ages-0.013
ich-0.013
 reading-0.013
TokenOn
Token ID2202
Feature activation+1.72e-02

Pos logit contributions
 agreement+0.009
 particular+0.008
 Everyday+0.008
 offer+0.008
 invite+0.007
Neg logit contributions
w-0.006
ich-0.005
aged-0.005
irens-0.005
uts-0.005
Token the
Token ID262
Feature activation+1.34e-02

Pos logit contributions
 agreement+0.036
date+0.034
 offer+0.033
 invite+0.033
tted+0.031
Neg logit contributions
w-0.015
irens-0.011
ich-0.011
 flashing-0.010
 praying-0.010
Token way
Token ID835
Feature activation-7.21e-03

Pos logit contributions
 agreement+0.019
date+0.016
 particular+0.015
 Everyday+0.015
cloth+0.015
Neg logit contributions
w-0.022
ich-0.020
ller-0.020
llers-0.019
EEP-0.018
Token,
Token ID11
Feature activation+5.90e-03

Pos logit contributions
w+0.010
ich+0.009
aged+0.009
ller+0.008
EEP+0.008
Neg logit contributions
 agreement-0.013
 particular-0.011
 Everyday-0.011
date-0.011
 invite-0.011
Token A
Token ID317
Feature activation+7.20e-03

Pos logit contributions
 agreement+0.010
 particular+0.009
 Everyday+0.009
date+0.009
 invite+0.009
Neg logit contributions
w-0.009
ich-0.008
ller-0.007
llers-0.007
aged-0.007
Tokenstr
Token ID2536
Feature activation-6.93e-02

Pos logit contributions
 agreement+0.012
 Everyday+0.010
 particular+0.010
date+0.010
 invite+0.010
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
llers-0.009
RR-0.009
Tokenid
Token ID312
Feature activation+3.01e-03

Pos logit contributions
nam+0.091
w+0.084
 elbows+0.083
AAAA+0.080
 viol+0.078
Neg logit contributions
 Everyday-0.089
ini-0.087
 particular-0.086
 definitely-0.085
 agreement-0.084
Token saw
Token ID2497
Feature activation+1.67e-02

Pos logit contributions
 agreement+0.005
 particular+0.005
date+0.005
 Everyday+0.005
 invite+0.004
Neg logit contributions
w-0.005
ich-0.004
aged-0.004
ller-0.004
llers-0.004
Token a
Token ID257
Feature activation-1.08e-02

Pos logit contributions
 agreement+0.033
 particular+0.030
date+0.029
 invite+0.029
 Everyday+0.029
Neg logit contributions
w-0.020
ich-0.017
aged-0.017
irens-0.016
ller-0.016
Token pond
Token ID16723
Feature activation+1.27e-03

Pos logit contributions
w+0.017
ich+0.016
ller+0.015
llers+0.015
aged+0.015
Neg logit contributions
 agreement-0.017
 particular-0.015
 Everyday-0.015
date-0.014
 invite-0.014
Token filled
Token ID5901
Feature activation+1.33e-02

Pos logit contributions
 agreement+0.002
 Everyday+0.002
 particular+0.002
date+0.002
 Whenever+0.002
Neg logit contributions
w-0.001
ich-0.001
 crying-0.001
 reading-0.001
 screaming-0.001
Token they
Token ID484
Feature activation+1.64e-02

Pos logit contributions
 agreement+0.016
 particular+0.015
 Everyday+0.014
date+0.014
 invite+0.014
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
aged-0.009
llers-0.009
Token arrived
Token ID5284
Feature activation-1.30e-02

Pos logit contributions
 agreement+0.030
 particular+0.027
date+0.027
 Everyday+0.026
 offer+0.025
Neg logit contributions
w-0.022
AAAA-0.016
ich-0.015
awed-0.015
RR-0.015
Token home
Token ID1363
Feature activation-1.75e-02

Pos logit contributions
w+0.019
ich+0.017
ller+0.017
aged+0.016
llers+0.016
Neg logit contributions
 agreement-0.022
 particular-0.020
 Everyday-0.019
date-0.019
 invite-0.019
Token,
Token ID11
Feature activation+1.27e-03

Pos logit contributions
w+0.021
ich+0.020
aged+0.019
ller+0.019
llers+0.018
Neg logit contributions
 agreement-0.033
 particular-0.031
 Everyday-0.030
date-0.029
 invite-0.028
Token A
Token ID317
Feature activation+1.31e-02

Pos logit contributions
 agreement+0.002
 particular+0.002
 Everyday+0.002
date+0.002
 invite+0.002
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
llers-0.002
Tokenstr
Token ID2536
Feature activation-6.44e-02

Pos logit contributions
 agreement+0.021
 Everyday+0.019
 particular+0.019
date+0.019
 invite+0.018
Neg logit contributions
w-0.019
ich-0.018
ller-0.017
llers-0.016
AAAA-0.016
Tokenid
Token ID312
Feature activation+6.47e-03

Pos logit contributions
nam+0.088
w+0.084
 satellites+0.077
 elbows+0.077
AAAA+0.076
Neg logit contributions
 agreement-0.081
 Everyday-0.080
 particular-0.078
 expert-0.075
ini-0.074
Token put
Token ID1234
Feature activation+1.50e-03

Pos logit contributions
 agreement+0.011
 particular+0.010
 Everyday+0.010
date+0.010
 invite+0.010
Neg logit contributions
w-0.010
ich-0.009
aged-0.008
ller-0.008
irens-0.008
Token her
Token ID607
Feature activation+1.39e-02

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
 invite+0.003
 basis+0.003
Neg logit contributions
w-0.002
aged-0.001
ich-0.001
awed-0.001
 bleeding-0.001
Token balloon
Token ID21190
Feature activation-1.13e-02

Pos logit contributions
 agreement+0.023
date+0.021
 Everyday+0.021
 particular+0.021
 invite+0.020
Neg logit contributions
w-0.021
ich-0.020
ller-0.018
aged-0.017
nam-0.017
Token in
Token ID287
Feature activation+2.05e-02

Pos logit contributions
ich+0.013
w+0.013
aged+0.012
 bleeding+0.011
ller+0.011
Neg logit contributions
 agreement-0.022
 Everyday-0.021
 particular-0.021
 Whenever-0.020
 invite-0.019
Token!
Token ID0
Feature activation+1.10e-02

Pos logit contributions
w+0.006
ich+0.006
ller+0.006
uts+0.005
llers+0.005
Neg logit contributions
 agreement-0.010
 particular-0.009
date-0.009
 Everyday-0.009
 invite-0.009
Token Mom
Token ID11254
Feature activation+8.87e-03

Pos logit contributions
 agreement+0.018
 particular+0.016
 Everyday+0.015
date+0.015
 invite+0.015
Neg logit contributions
w-0.017
ich-0.015
ller-0.015
aged-0.015
llers-0.015
Tokenmy
Token ID1820
Feature activation+3.81e-03

Pos logit contributions
 agreement+0.019
 particular+0.018
date+0.018
 offer+0.017
 Everyday+0.017
Neg logit contributions
w-0.009
ller-0.008
ich-0.008
llers-0.007
aged-0.007
Token and
Token ID290
Feature activation-1.23e-02

Pos logit contributions
 agreement+0.007
 particular+0.007
date+0.006
 Everyday+0.006
 offer+0.006
Neg logit contributions
w-0.005
ich-0.005
ller-0.004
aged-0.004
llers-0.004
Token A
Token ID317
Feature activation+1.24e-02

Pos logit contributions
w+0.019
ich+0.017
ller+0.017
llers+0.016
aged+0.016
Neg logit contributions
 agreement-0.020
 particular-0.019
 Everyday-0.018
date-0.017
 invite-0.017
Tokenstr
Token ID2536
Feature activation-6.34e-02

Pos logit contributions
 agreement+0.020
 particular+0.018
 Everyday+0.018
date+0.017
 invite+0.017
Neg logit contributions
w-0.018
ich-0.018
ller-0.016
llers-0.016
aged-0.016
Tokenid
Token ID312
Feature activation+1.75e-02

Pos logit contributions
nam+0.084
w+0.080
AAAA+0.077
 satellites+0.075
 elbows+0.073
Neg logit contributions
 agreement-0.079
date-0.079
 Everyday-0.078
ini-0.077
 particular-0.076
Token said
Token ID531
Feature activation+2.86e-02

Pos logit contributions
 agreement+0.029
 particular+0.027
date+0.027
 Everyday+0.026
 invite+0.025
Neg logit contributions
w-0.028
aged-0.025
ich-0.024
ller-0.022
ged-0.021
Token good
Token ID922
Feature activation+1.57e-03

Pos logit contributions
 agreement+0.059
 particular+0.056
date+0.053
 Everyday+0.053
 invite+0.051
Neg logit contributions
w-0.030
aged-0.027
ich-0.025
ller-0.024
uts-0.024
Token bye
Token ID33847
Feature activation+1.92e-03

Pos logit contributions
 agreement+0.002
 particular+0.002
 Everyday+0.002
date+0.002
 invite+0.002
Neg logit contributions
w-0.003
ich-0.002
ller-0.002
aged-0.002
llers-0.002
Token and
Token ID290
Feature activation+5.93e-03

Pos logit contributions
 agreement+0.004
 particular+0.004
 Everyday+0.003
date+0.003
 invite+0.003
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
EEP-0.002
Tokenie
Token ID494
Feature activation-1.49e-02

Pos logit contributions
w+0.025
ich+0.022
ller+0.021
aged+0.020
irens+0.020
Neg logit contributions
 agreement-0.016
 Everyday-0.014
 particular-0.014
 offer-0.012
date-0.012
Token.
Token ID13
Feature activation-3.31e-02

Pos logit contributions
w+0.020
ich+0.019
ller+0.019
aged+0.018
 bleeding+0.017
Neg logit contributions
 agreement-0.027
 particular-0.026
 Everyday-0.024
date-0.023
 invite-0.022
Token
Token ID198
Feature activation-1.17e-02

Pos logit contributions
w+0.066
ich+0.060
 elbows+0.057
irens+0.054
inches+0.053
Neg logit contributions
 agreement-0.043
 Everyday-0.043
 particular-0.039
 offer-0.037
 invite-0.037
Token
Token ID198
Feature activation-6.85e-03

Pos logit contributions
w+0.018
ich+0.016
aged+0.015
ller+0.015
llers+0.015
Neg logit contributions
 agreement-0.019
 particular-0.018
 Everyday-0.018
date-0.017
 invite-0.016
Token"
Token ID1
Feature activation-3.22e-02

Pos logit contributions
w+0.010
ich+0.009
ller+0.008
aged+0.008
llers+0.008
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.011
date-0.010
 invite-0.010
TokenHello
Token ID15496
Feature activation-6.29e-02

Pos logit contributions
inches+0.059
aged+0.056
ouncing+0.055
w+0.055
uts+0.055
Neg logit contributions
 agreement-0.042
 particular-0.041
 Everyday-0.037
date-0.034
 offer-0.034
Token,
Token ID11
Feature activation-4.54e-02

Pos logit contributions
w+0.118
 elbows+0.111
irens+0.108
aged+0.107
llers+0.103
Neg logit contributions
 agreement-0.078
 particular-0.075
 fill-0.069
 Everyday-0.068
 invite-0.068
Token little
Token ID1310
Feature activation-4.86e-02

Pos logit contributions
w+0.083
aged+0.075
 elbows+0.074
irens+0.073
ich+0.073
Neg logit contributions
 agreement-0.062
 particular-0.057
 Everyday-0.053
 offer-0.053
 invite-0.052
Token girl
Token ID2576
Feature activation-2.13e-02

Pos logit contributions
w+0.096
llers+0.091
irens+0.091
uts+0.090
aged+0.090
Neg logit contributions
 agreement-0.056
 particular-0.052
 offer-0.047
 invite-0.045
 fill-0.045
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
w+0.028
ich+0.025
ller+0.024
aged+0.023
llers+0.022
Neg logit contributions
 agreement-0.040
 particular-0.038
 Everyday-0.036
date-0.036
 invite-0.034
Token I
Token ID314
Feature activation-1.16e-02

Pos logit contributions
w+0.032
ich+0.028
ller+0.027
aged+0.027
llers+0.027
Neg logit contributions
 agreement-0.035
 particular-0.033
 Everyday-0.031
date-0.031
 invite-0.030
Token a
Token ID257
Feature activation+1.18e-02

Pos logit contributions
w+0.008
ich+0.007
ller+0.007
llers+0.007
irens+0.007
Neg logit contributions
 agreement-0.010
 particular-0.009
 Everyday-0.009
date-0.009
 invite-0.009
Token big
Token ID1263
Feature activation+1.53e-02

Pos logit contributions
 agreement+0.013
 particular+0.013
date+0.011
 Everyday+0.011
 invite+0.011
Neg logit contributions
w-0.024
aged-0.021
llers-0.021
ller-0.021
ich-0.021
Token bird
Token ID6512
Feature activation-6.52e-03

Pos logit contributions
 agreement+0.027
 particular+0.024
 Everyday+0.024
date+0.024
 invite+0.023
Neg logit contributions
w-0.021
ich-0.019
ller-0.018
llers-0.017
RR-0.017
Token named
Token ID3706
Feature activation-1.32e-02

Pos logit contributions
w+0.013
ich+0.012
ller+0.012
llers+0.012
aged+0.012
Neg logit contributions
 agreement-0.007
 particular-0.006
 Everyday-0.006
date-0.006
 invite-0.006
Token O
Token ID440
Feature activation+4.89e-03

Pos logit contributions
w+0.024
aged+0.022
 elbows+0.022
irens+0.022
ich+0.022
Neg logit contributions
 agreement-0.017
 Everyday-0.016
 particular-0.016
 fill-0.016
 invite-0.015
Tokenstr
Token ID2536
Feature activation-6.28e-02

Pos logit contributions
 agreement+0.009
 particular+0.009
date+0.009
 Everyday+0.009
 invite+0.009
Neg logit contributions
w-0.005
nam-0.005
ller-0.005
aged-0.005
llers-0.004
Tokenich
Token ID488
Feature activation+3.84e-03

Pos logit contributions
nam+0.062
w+0.053
AAAA+0.050
 elbows+0.048
destruct+0.045
Neg logit contributions
berry-0.107
date-0.107
 particular-0.102
 Everyday-0.098
ini-0.098
Token wanted
Token ID2227
Feature activation+1.71e-03

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
date+0.002
 invite+0.002
Neg logit contributions
w-0.008
 elbows-0.008
irens-0.008
nam-0.008
ich-0.008
Token to
Token ID284
Feature activation+1.52e-02

Pos logit contributions
 agreement+0.002
 particular+0.002
 Everyday+0.002
 invite+0.002
 fill+0.002
Neg logit contributions
w-0.003
ich-0.003
ller-0.003
llers-0.003
aged-0.003
Token swim
Token ID9422
Feature activation+1.25e-02

Pos logit contributions
 agreement+0.022
 particular+0.018
 Everyday+0.017
date+0.017
rily+0.017
Neg logit contributions
w-0.024
ich-0.022
ller-0.022
llers-0.021
 Blood-0.021
Token in
Token ID287
Feature activation+2.14e-02

Pos logit contributions
 agreement+0.027
 Everyday+0.024
 particular+0.024
 Whenever+0.023
 tow+0.023
Neg logit contributions
w-0.013
nam-0.009
ls-0.008
ich-0.008
RR-0.008
Token
Token ID220
Feature activation-6.01e-03

Pos logit contributions
w+0.045
nam+0.043
 elbows+0.039
 bleeding+0.039
 satellites+0.039
Neg logit contributions
 Everyday-0.035
date-0.033
 particular-0.033
 agreement-0.032
 invite-0.032
Token
Token ID198
Feature activation+8.14e-03

Pos logit contributions
w+0.008
ich+0.007
ller+0.007
aged+0.007
llers+0.007
Neg logit contributions
 agreement-0.011
 particular-0.010
 Everyday-0.010
date-0.009
 invite-0.009
Token
Token ID198
Feature activation+8.07e-03

Pos logit contributions
 agreement+0.017
 particular+0.016
date+0.015
 invite+0.015
 Everyday+0.015
Neg logit contributions
w-0.008
ich-0.007
ller-0.007
irens-0.007
RR-0.007
TokenMiss
Token ID17140
Feature activation-6.54e-03

Pos logit contributions
 agreement+0.015
 particular+0.014
 offer+0.014
 fill+0.014
 invite+0.013
Neg logit contributions
w-0.010
ich-0.008
RR-0.008
irens-0.008
ller-0.008
Token O
Token ID440
Feature activation+6.03e-03

Pos logit contributions
w+0.010
ich+0.009
ller+0.009
aged+0.009
llers+0.009
Neg logit contributions
 agreement-0.010
 particular-0.010
 Everyday-0.009
date-0.009
 invite-0.009
Tokenstr
Token ID2536
Feature activation-6.26e-02

Pos logit contributions
 agreement+0.013
 particular+0.012
 Everyday+0.012
date+0.012
 invite+0.012
Neg logit contributions
w-0.005
ich-0.005
nam-0.004
aged-0.004
ller-0.004
Tokenich
Token ID488
Feature activation+1.64e-02

Pos logit contributions
nam+0.059
w+0.049
AAAA+0.047
 satellites+0.045
 elbows+0.044
Neg logit contributions
 Everyday-0.104
 agreement-0.103
date-0.102
 particular-0.100
berry-0.099
Token smiled
Token ID13541
Feature activation+1.37e-02

Pos logit contributions
 agreement+0.032
 particular+0.029
date+0.028
 Everyday+0.027
 invite+0.027
Neg logit contributions
w-0.021
ller-0.018
ich-0.018
aged-0.017
llers-0.017
Token and
Token ID290
Feature activation-3.78e-03

Pos logit contributions
 particular+0.025
date+0.024
 agreement+0.024
 invite+0.023
 Everyday+0.022
Neg logit contributions
aged-0.014
ller-0.014
ich-0.013
w-0.013
 crying-0.013
Token said
Token ID531
Feature activation+2.23e-02

Pos logit contributions
w+0.004
ich+0.004
ller+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.008
 particular-0.007
date-0.007
 Everyday-0.007
 invite-0.007
Token,
Token ID11
Feature activation+5.87e-03

Pos logit contributions
 agreement+0.039
 particular+0.037
 Everyday+0.035
date+0.034
 invite+0.034
Neg logit contributions
w-0.031
ich-0.027
ller-0.027
aged-0.026
llers-0.026
Token.
Token ID13
Feature activation+2.05e-02

Pos logit contributions
w+0.011
ich+0.011
ller+0.011
uts+0.011
 reading+0.010
Neg logit contributions
 agreement-0.019
 particular-0.018
 Everyday-0.016
date-0.016
latable-0.016
Token They
Token ID1119
Feature activation-1.22e-02

Pos logit contributions
 agreement+0.039
 particular+0.038
date+0.035
 offer+0.033
 end+0.032
Neg logit contributions
ller-0.021
 Blood-0.021
EEP-0.021
aged-0.020
llers-0.020
Token saw
Token ID2497
Feature activation-5.61e-03

Pos logit contributions
w+0.019
ich+0.017
ller+0.017
aged+0.017
llers+0.016
Neg logit contributions
 agreement-0.019
 particular-0.018
 Everyday-0.017
date-0.017
 invite-0.016
Token lots
Token ID6041
Feature activation+2.15e-03

Pos logit contributions
w+0.009
ich+0.008
ller+0.008
aged+0.007
llers+0.007
Neg logit contributions
 agreement-0.009
 particular-0.008
 Everyday-0.008
date-0.008
 invite-0.008
Token of
Token ID286
Feature activation-2.22e-02

Pos logit contributions
 agreement+0.002
 particular+0.002
date+0.002
 Everyday+0.001
 invite+0.001
Neg logit contributions
w-0.005
ich-0.005
ller-0.005
irens-0.004
aged-0.004
Token old
Token ID1468
Feature activation-6.25e-02

Pos logit contributions
w+0.036
ich+0.033
ller+0.032
llers+0.031
aged+0.031
Neg logit contributions
 agreement-0.034
 particular-0.031
 Everyday-0.030
 invite-0.029
 offer-0.028
Token things
Token ID1243
Feature activation-2.35e-03

Pos logit contributions
w+0.122
shadow+0.119
RR+0.113
 Blood+0.113
EEP+0.113
Neg logit contributions
 of-0.061
 agreement-0.061
 invite-0.059
 offer-0.059
est-0.059
Token like
Token ID588
Feature activation-1.21e-03

Pos logit contributions
w+0.003
ich+0.003
ller+0.003
uts+0.003
aged+0.003
Neg logit contributions
 agreement-0.004
 particular-0.004
date-0.004
 Everyday-0.004
 invite-0.003
Token swords
Token ID19735
Feature activation-1.23e-02

Pos logit contributions
w+0.001
ich+0.001
ller+0.001
llers+0.001
aged+0.001
Neg logit contributions
 agreement-0.002
 particular-0.002
date-0.002
 Everyday-0.002
 invite-0.002
Token and
Token ID290
Feature activation-1.07e-02

Pos logit contributions
w+0.020
ich+0.018
ller+0.018
aged+0.017
llers+0.017
Neg logit contributions
 agreement-0.019
 particular-0.017
 Everyday-0.016
date-0.016
 invite-0.016
Token shields
Token ID21132
Feature activation-6.07e-03

Pos logit contributions
w+0.014
ich+0.012
ller+0.012
nam+0.012
llers+0.012
Neg logit contributions
 agreement-0.020
 particular-0.019
date-0.018
 Everyday-0.017
 invite-0.016
Token balloon
Token ID21190
Feature activation+1.49e-03

Pos logit contributions
 agreement+0.017
 particular+0.015
date+0.014
 Everyday+0.014
 invite+0.014
Neg logit contributions
w-0.013
ich-0.011
ller-0.011
inches-0.010
llers-0.010
Token.
Token ID13
Feature activation+1.33e-02

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.002
date+0.002
 invite+0.002
Neg logit contributions
w-0.002
ller-0.002
ich-0.002
uts-0.002
inches-0.002
Token
Token ID198
Feature activation-8.37e-03

Pos logit contributions
 agreement+0.026
 particular+0.024
 offer+0.023
date+0.023
 fill+0.023
Neg logit contributions
w-0.013
ller-0.013
 Blood-0.012
uts-0.012
llers-0.012
Token
Token ID198
Feature activation-8.22e-03

Pos logit contributions
w+0.009
ich+0.007
ller+0.007
irens+0.007
 Blood+0.007
Neg logit contributions
 agreement-0.017
 particular-0.016
date-0.016
 invite-0.016
 fill-0.016
TokenA
Token ID32
Feature activation+1.28e-02

Pos logit contributions
w+0.011
ich+0.010
ller+0.009
irens+0.009
RR+0.009
Neg logit contributions
 agreement-0.015
 particular-0.014
 Everyday-0.013
date-0.013
 invite-0.013
Tokenstr
Token ID2536
Feature activation-6.21e-02

Pos logit contributions
 agreement+0.021
 Everyday+0.019
 particular+0.019
date+0.018
 invite+0.018
Neg logit contributions
w-0.018
ich-0.017
ller-0.017
RR-0.015
AAAA-0.015
Tokenid
Token ID312
Feature activation+2.47e-03

Pos logit contributions
w+0.044
nam+0.044
 elbows+0.044
AAAA+0.044
 Whe+0.041
Neg logit contributions
 agreement-0.113
date-0.112
rily-0.106
 particular-0.105
ini-0.105
Token held
Token ID2714
Feature activation-1.68e-03

Pos logit contributions
 agreement+0.004
 particular+0.004
date+0.004
 Everyday+0.004
 invite+0.004
Neg logit contributions
w-0.004
ich-0.003
aged-0.003
ller-0.003
llers-0.003
Token onto
Token ID4291
Feature activation+7.41e-03

Pos logit contributions
w+0.002
aged+0.002
ich+0.002
ller+0.002
nam+0.002
Neg logit contributions
 agreement-0.003
date-0.003
 particular-0.003
 invite-0.003
 Everyday-0.003
Token her
Token ID607
Feature activation+1.94e-02

Pos logit contributions
 agreement+0.013
 particular+0.012
date+0.012
 Everyday+0.012
 invite+0.012
Neg logit contributions
w-0.010
ich-0.009
ller-0.009
aged-0.009
llers-0.008
Token balloon
Token ID21190
Feature activation-7.98e-03

Pos logit contributions
 agreement+0.034
date+0.033
 Everyday+0.033
 particular+0.032
 invite+0.031
Neg logit contributions
ich-0.024
ller-0.023
w-0.022
nam-0.021
 bleeding-0.021
Token around
Token ID1088
Feature activation-3.60e-03

Pos logit contributions
w+0.019
ich+0.017
ller+0.016
aged+0.016
llers+0.016
Neg logit contributions
 agreement-0.021
 particular-0.020
date-0.019
 Everyday-0.019
 invite-0.018
Token and
Token ID290
Feature activation-1.42e-02

Pos logit contributions
w+0.004
ller+0.004
ich+0.004
aged+0.004
 Blood+0.004
Neg logit contributions
 agreement-0.006
date-0.006
 Everyday-0.006
 particular-0.006
 invite-0.005
Token saw
Token ID2497
Feature activation+3.55e-03

Pos logit contributions
w+0.025
ich+0.024
irens+0.022
ller+0.022
uts+0.022
Neg logit contributions
 agreement-0.020
 particular-0.019
 invite-0.018
 Everyday-0.018
 fill-0.018
Token lots
Token ID6041
Feature activation+9.65e-04

Pos logit contributions
 agreement+0.007
 particular+0.006
date+0.006
 Everyday+0.006
 invite+0.006
Neg logit contributions
w-0.004
ich-0.004
ller-0.004
aged-0.004
nam-0.004
Token of
Token ID286
Feature activation-2.65e-02

Pos logit contributions
 agreement+0.001
date+0.001
 particular+0.001
 surely+0.001
 invite+0.001
Neg logit contributions
irens-0.002
ller-0.002
uts-0.002
 flashing-0.002
ls-0.002
Token interesting
Token ID3499
Feature activation-6.21e-02

Pos logit contributions
w+0.037
ich+0.034
ller+0.032
aged+0.031
llers+0.031
Neg logit contributions
 agreement-0.046
 particular-0.043
 Everyday-0.041
date-0.040
 invite-0.040
Token things
Token ID1243
Feature activation-1.10e-03

Pos logit contributions
w+0.114
ich+0.109
EEP+0.098
shadow+0.097
AAAA+0.095
Neg logit contributions
 particular-0.084
 agreement-0.082
 Everyday-0.078
 offer-0.077
 invite-0.076
Token in
Token ID287
Feature activation+1.15e-02

Pos logit contributions
w+0.001
uts+0.001
ller+0.001
ich+0.001
aged+0.001
Neg logit contributions
 agreement-0.002
 particular-0.002
date-0.002
 Everyday-0.002
 invite-0.002
Token the
Token ID262
Feature activation-3.73e-02

Pos logit contributions
 agreement+0.022
date+0.021
 Everyday+0.019
 invite+0.019
 particular+0.019
Neg logit contributions
w-0.012
irens-0.011
nam-0.011
ller-0.010
ich-0.010
Token room
Token ID2119
Feature activation+1.31e-02

Pos logit contributions
w+0.062
ich+0.058
ller+0.056
aged+0.055
llers+0.055
Neg logit contributions
 agreement-0.056
 particular-0.052
 Everyday-0.049
date-0.048
 invite-0.047
Token.
Token ID13
Feature activation+1.64e-02

Pos logit contributions
 agreement+0.026
 particular+0.023
 Everyday+0.023
date+0.022
iona+0.021
Neg logit contributions
w-0.014
aged-0.014
uts-0.014
ich-0.014
 reading-0.013
Token a
Token ID257
Feature activation-1.62e-02

Pos logit contributions
w+0.033
aged+0.029
ich+0.028
llers+0.028
ller+0.028
Neg logit contributions
 agreement-0.049
 particular-0.046
date-0.044
 Everyday-0.043
 offer-0.041
Token box
Token ID3091
Feature activation-2.22e-03

Pos logit contributions
w+0.022
ich+0.021
nam+0.019
 praying+0.019
ller+0.019
Neg logit contributions
 agreement-0.026
 Everyday-0.024
 particular-0.023
date-0.022
 Whenever-0.022
Token of
Token ID286
Feature activation-1.11e-02

Pos logit contributions
w+0.003
aged+0.003
 reading+0.003
uts+0.003
ich+0.003
Neg logit contributions
 agreement-0.004
 particular-0.003
 Everyday-0.003
date-0.003
 Whenever-0.003
Token shiny
Token ID22441
Feature activation-2.26e-02

Pos logit contributions
w+0.014
nam+0.012
aged+0.011
ller+0.011
 flashing+0.011
Neg logit contributions
 agreement-0.021
date-0.019
 particular-0.019
 Everyday-0.018
 surely-0.018
Token or
Token ID393
Feature activation-1.36e-02

Pos logit contributions
w+0.027
ller+0.023
ich+0.023
aged+0.023
llers+0.022
Neg logit contributions
 agreement-0.044
 particular-0.041
date-0.040
 Everyday-0.039
 invite-0.038
Tokennam
Token ID7402
Feature activation-6.20e-02

Pos logit contributions
w+0.011
ich+0.009
ller+0.009
aged+0.008
llers+0.008
Neg logit contributions
 agreement-0.032
 particular-0.030
 Everyday-0.029
date-0.029
 invite-0.028
Tokenents
Token ID658
Feature activation+5.19e-03

Pos logit contributions
AAAA+0.104
w+0.100
rob+0.094
 elbows+0.092
ses+0.091
Neg logit contributions
 agreement-0.079
 Everyday-0.079
 of-0.077
 particular-0.075
 fill-0.073
Token that
Token ID326
Feature activation-2.07e-02

Pos logit contributions
 agreement+0.011
 particular+0.009
 Everyday+0.009
date+0.009
 invite+0.009
Neg logit contributions
w-0.006
uts-0.006
ich-0.005
aged-0.005
ller-0.005
Token spark
Token ID9009
Feature activation-5.50e-03

Pos logit contributions
w+0.024
ich+0.019
aged+0.019
ller+0.019
nam+0.018
Neg logit contributions
 agreement-0.043
 particular-0.037
date-0.037
 Everyday-0.036
 invite-0.035
Tokenled
Token ID992
Feature activation+4.46e-03

Pos logit contributions
w+0.007
ller+0.006
aged+0.006
uts+0.005
llers+0.005
Neg logit contributions
 agreement-0.012
 particular-0.010
 Everyday-0.010
date-0.010
 invite-0.010
Token in
Token ID287
Feature activation+1.38e-02

Pos logit contributions
 agreement+0.010
 invite+0.009
 particular+0.009
 Everyday+0.009
date+0.009
Neg logit contributions
w-0.004
nam-0.004
 reading-0.003
aged-0.003
 flashing-0.003
Token the
Token ID262
Feature activation-3.85e-03

Pos logit contributions
date+0.050
tted+0.046
 host+0.046
 surely+0.046
 agreement+0.046
Neg logit contributions
 reading-0.013
 crying-0.010
 pressing-0.010
 crossing-0.010
 touching-0.009
Token lake
Token ID13546
Feature activation+6.85e-03

Pos logit contributions
w+0.007
ich+0.006
ller+0.006
llers+0.006
irens+0.006
Neg logit contributions
 agreement-0.005
 particular-0.005
date-0.004
 Everyday-0.004
 invite-0.004
Token.
Token ID13
Feature activation+1.28e-03

Pos logit contributions
 agreement+0.014
 particular+0.013
date+0.012
 Everyday+0.012
 surely+0.012
Neg logit contributions
w-0.006
ich-0.006
nam-0.006
 reading-0.005
 crying-0.005
Token But
Token ID887
Feature activation+4.24e-03

Pos logit contributions
 agreement+0.002
 particular+0.002
date+0.002
 invite+0.002
 Everyday+0.002
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
llers-0.002
Token O
Token ID440
Feature activation+6.19e-04

Pos logit contributions
 agreement+0.007
 particular+0.006
 Everyday+0.006
date+0.006
 invite+0.006
Neg logit contributions
w-0.006
ich-0.006
ller-0.006
aged-0.005
llers-0.005
Tokenstr
Token ID2536
Feature activation-6.19e-02

Pos logit contributions
 particular+0.001
 agreement+0.001
 Everyday+0.001
date+0.001
 offer+0.001
Neg logit contributions
nam-0.001
 elbows-0.001
irens-0.000
ller-0.000
w-0.000
Tokenich
Token ID488
Feature activation+2.03e-02

Pos logit contributions
nam+0.064
 elbows+0.050
w+0.048
AAAA+0.047
RR+0.044
Neg logit contributions
date-0.104
 Everyday-0.102
ini-0.101
 particular-0.101
 finally-0.098
Token was
Token ID373
Feature activation-7.75e-05

Pos logit contributions
 agreement+0.032
 particular+0.029
 Everyday+0.027
date+0.027
 invite+0.027
Neg logit contributions
w-0.033
ich-0.029
aged-0.029
ller-0.028
llers-0.027
Token scared
Token ID12008
Feature activation-4.49e-04

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
llers+0.000
EEP+0.000
Neg logit contributions
 agreement-0.000
 particular-0.000
 Everyday-0.000
 invite-0.000
 fill-0.000
Token of
Token ID286
Feature activation-1.68e-03

Pos logit contributions
w+0.001
ich+0.001
ller+0.001
aged+0.001
llers+0.001
Neg logit contributions
 agreement-0.001
 particular-0.001
date-0.001
 invite-0.001
 Everyday-0.001
Token water
Token ID1660
Feature activation+2.02e-02

Pos logit contributions
w+0.003
ich+0.002
ller+0.002
aged+0.002
llers+0.002
Neg logit contributions
 agreement-0.003
 particular-0.002
 Everyday-0.002
 invite-0.002
date-0.002
Token�
Token ID95
Feature activation-7.59e-03

Pos logit contributions
w+0.005
ich+0.004
aged+0.004
ller+0.004
llers+0.004
Neg logit contributions
 agreement-0.005
 particular-0.005
 Everyday-0.004
date-0.004
 invite-0.004
Token€
Token ID26391
Feature activation-2.67e-02

Pos logit contributions
w+0.012
ich+0.010
ller+0.010
aged+0.010
llers+0.010
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.011
date-0.011
 invite-0.011
TokenHi
Token ID17250
Feature activation-9.96e-03

Pos logit contributions
nam+0.038
w+0.036
 elbows+0.032
 satellites+0.032
destruct+0.031
Neg logit contributions
date-0.038
 Everyday-0.038
 particular-0.037
 invite-0.035
 agreement-0.035
Token Miss
Token ID4544
Feature activation-1.10e-02

Pos logit contributions
w+0.012
ich+0.011
ller+0.011
RR+0.010
llers+0.010
Neg logit contributions
 agreement-0.019
 particular-0.017
 Everyday-0.017
date-0.017
 invite-0.017
Token O
Token ID440
Feature activation+4.00e-03

Pos logit contributions
w+0.017
ich+0.015
ller+0.015
aged+0.014
llers+0.014
Neg logit contributions
 agreement-0.018
 particular-0.016
 Everyday-0.016
date-0.015
 invite-0.015
Tokenstr
Token ID2536
Feature activation-6.17e-02

Pos logit contributions
 agreement+0.009
 particular+0.009
date+0.008
 Everyday+0.008
 invite+0.008
Neg logit contributions
w-0.003
aged-0.002
nam-0.002
 elbows-0.002
ich-0.002
Tokenich
Token ID488
Feature activation+2.30e-02

Pos logit contributions
nam+0.050
AAAA+0.048
w+0.047
 satellites+0.042
Crash+0.041
Neg logit contributions
 agreement-0.108
date-0.106
 Everyday-0.105
berry-0.105
 particular-0.104
Token!
Token ID0
Feature activation+1.50e-02

Pos logit contributions
 agreement+0.044
 particular+0.042
date+0.039
 Everyday+0.039
latable+0.039
Neg logit contributions
w-0.027
ller-0.025
ich-0.025
 crying-0.021
 bleeding-0.021
Token That
Token ID1320
Feature activation-3.88e-04

Pos logit contributions
 agreement+0.028
 particular+0.026
date+0.025
 invite+0.024
 Everyday+0.024
Neg logit contributions
w-0.018
ich-0.016
ller-0.016
llers-0.015
RR-0.015
Token sounds
Token ID5238
Feature activation+2.09e-02

Pos logit contributions
w+0.001
ich+0.000
ller+0.000
llers+0.000
aged+0.000
Neg logit contributions
 agreement-0.001
 particular-0.001
date-0.001
 Everyday-0.001
 invite-0.001
Token funny
Token ID8258
Feature activation+9.74e-04

Pos logit contributions
 agreement+0.032
date+0.030
 particular+0.028
ainer+0.027
 Everyday+0.026
Neg logit contributions
w-0.031
 happened-0.028
nam-0.026
 bleeding-0.026
ller-0.026
Token.
Token ID13
Feature activation+1.43e-02

Pos logit contributions
w+0.035
ich+0.032
ller+0.031
aged+0.030
llers+0.030
Neg logit contributions
 agreement-0.027
 particular-0.025
 Everyday-0.023
date-0.023
 invite-0.022
Token
Token ID198
Feature activation+2.78e-03

Pos logit contributions
 agreement+0.022
 particular+0.020
 Everyday+0.019
 invite+0.018
date+0.018
Neg logit contributions
w-0.024
ich-0.022
ller-0.021
llers-0.020
aged-0.020
Token
Token ID198
Feature activation+1.45e-02

Pos logit contributions
 agreement+0.004
 particular+0.004
 Everyday+0.004
 invite+0.003
 offer+0.003
Neg logit contributions
w-0.005
ich-0.004
nam-0.004
ller-0.004
llers-0.004
Token"
Token ID1
Feature activation-1.35e-02

Pos logit contributions
 agreement+0.026
 particular+0.024
 Everyday+0.023
date+0.023
 invite+0.022
Neg logit contributions
w-0.020
ich-0.018
ller-0.017
aged-0.017
llers-0.016
TokenBad
Token ID22069
Feature activation-4.56e-02

Pos logit contributions
w+0.020
ich+0.018
ller+0.017
aged+0.017
llers+0.016
Neg logit contributions
 agreement-0.023
 particular-0.021
 Everyday-0.021
date-0.020
 invite-0.020
Token wind
Token ID2344
Feature activation-6.17e-02

Pos logit contributions
w+0.083
aged+0.078
nam+0.078
shadow+0.076
irens+0.074
Neg logit contributions
 agreement-0.057
 end-0.052
 particular-0.050
date-0.048
 offer-0.047
Token,
Token ID11
Feature activation+6.28e-03

Pos logit contributions
w+0.091
ich+0.084
ller+0.082
llers+0.081
nam+0.080
Neg logit contributions
 agreement-0.097
 particular-0.089
date-0.086
 Everyday-0.085
 invite-0.084
Token bad
Token ID2089
Feature activation-5.35e-02

Pos logit contributions
 agreement+0.011
date+0.011
 particular+0.010
 Everyday+0.010
 invite+0.009
Neg logit contributions
w-0.009
ich-0.008
ller-0.007
 Blood-0.007
RR-0.007
Token wind
Token ID2344
Feature activation-5.65e-02

Pos logit contributions
w+0.089
ich+0.084
aged+0.082
llers+0.081
ller+0.081
Neg logit contributions
 agreement-0.075
 particular-0.069
 offer-0.066
 invite-0.065
 Everyday-0.065
Token!"
Token ID2474
Feature activation+6.40e-03

Pos logit contributions
w+0.082
ich+0.075
ller+0.072
llers+0.072
irens+0.072
Neg logit contributions
 agreement-0.091
 particular-0.084
 Everyday-0.080
date-0.079
 invite-0.079
Token L
Token ID406
Feature activation+3.19e-02

Pos logit contributions
 agreement+0.010
 particular+0.009
 Everyday+0.008
 invite+0.008
date+0.008
Neg logit contributions
w-0.011
ich-0.010
ller-0.009
llers-0.009
aged-0.009
Token
Token ID220
Feature activation-3.29e-03

Pos logit contributions
w+0.033
nam+0.031
 bleeding+0.028
aged+0.028
destruct+0.028
Neg logit contributions
 Everyday-0.032
 agreement-0.032
date-0.031
 particular-0.031
 invite-0.031
Token
Token ID198
Feature activation+5.02e-03

Pos logit contributions
w+0.004
ich+0.004
ller+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.006
 particular-0.005
 Everyday-0.005
date-0.005
 invite-0.005
Token
Token ID198
Feature activation+8.14e-03

Pos logit contributions
 agreement+0.011
 particular+0.010
date+0.010
 invite+0.009
 fill+0.009
Neg logit contributions
w-0.005
ich-0.004
ller-0.004
irens-0.004
aged-0.004
TokenMiss
Token ID17140
Feature activation-8.86e-03

Pos logit contributions
 agreement+0.015
 fill+0.014
 particular+0.014
 invite+0.014
 offer+0.014
Neg logit contributions
w-0.010
ich-0.008
irens-0.008
RR-0.008
uts-0.007
Token O
Token ID440
Feature activation+4.33e-03

Pos logit contributions
w+0.013
ich+0.012
ller+0.012
aged+0.011
llers+0.011
Neg logit contributions
 agreement-0.015
 particular-0.013
 Everyday-0.013
date-0.013
 invite-0.012
Tokenstr
Token ID2536
Feature activation-6.16e-02

Pos logit contributions
 agreement+0.009
 particular+0.009
 Everyday+0.009
date+0.008
 invite+0.008
Neg logit contributions
w-0.004
ich-0.003
nam-0.003
aged-0.003
ller-0.003
Tokenich
Token ID488
Feature activation+1.79e-02

Pos logit contributions
nam+0.066
w+0.051
 elbows+0.050
 satellites+0.049
AAAA+0.048
Neg logit contributions
 Everyday-0.099
 agreement-0.097
date-0.097
ini-0.094
 particular-0.094
Token told
Token ID1297
Feature activation+3.60e-03

Pos logit contributions
 agreement+0.034
 particular+0.031
date+0.029
 Everyday+0.028
latable+0.027
Neg logit contributions
w-0.024
ller-0.020
ich-0.020
aged-0.019
llers-0.019
Token a
Token ID257
Feature activation+7.21e-03

Pos logit contributions
 agreement+0.007
 particular+0.007
date+0.006
 Everyday+0.006
latable+0.006
Neg logit contributions
w-0.004
ller-0.003
irens-0.003
aged-0.003
llers-0.003
Token very
Token ID845
Feature activation+1.91e-02

Pos logit contributions
 agreement+0.011
date+0.010
 particular+0.009
 Everyday+0.009
 invite+0.009
Neg logit contributions
w-0.011
ller-0.010
ich-0.010
RR-0.010
EEP-0.010
Token funny
Token ID8258
Feature activation+4.05e-03

Pos logit contributions
 agreement+0.032
date+0.031
 Everyday+0.028
 fill+0.028
 particular+0.027
Neg logit contributions
w-0.025
ller-0.023
ich-0.022
aged-0.022
uts-0.021
TokenN
Token ID45
Feature activation-9.22e-03

Pos logit contributions
w+0.009
ich+0.008
ller+0.008
irens+0.008
aged+0.008
Neg logit contributions
 agreement-0.013
 particular-0.012
 Everyday-0.012
 invite-0.012
date-0.012
Tokenina
Token ID1437
Feature activation+1.14e-02

Pos logit contributions
w+0.014
ich+0.012
nam+0.012
irens+0.012
 elbows+0.012
Neg logit contributions
 agreement-0.015
 particular-0.014
 Everyday-0.013
date-0.013
 offer-0.012
Token was
Token ID373
Feature activation+2.42e-03

Pos logit contributions
 agreement+0.020
 particular+0.018
date+0.018
 Everyday+0.017
 invite+0.016
Neg logit contributions
w-0.016
ller-0.015
aged-0.014
ich-0.014
llers-0.014
Token panic
Token ID13619
Feature activation+8.40e-03

Pos logit contributions
 particular+0.004
 agreement+0.004
date+0.004
 offer+0.004
 Everyday+0.004
Neg logit contributions
aged-0.003
 crying-0.003
ich-0.003
 reading-0.003
 bleeding-0.003
Token-
Token ID12
Feature activation-7.67e-03

Pos logit contributions
 agreement+0.014
 particular+0.013
date+0.013
 Everyday+0.013
 invite+0.013
Neg logit contributions
w-0.011
ller-0.010
aged-0.010
 bleeding-0.010
ich-0.010
Tokenstr
Token ID2536
Feature activation-6.16e-02

Pos logit contributions
w+0.011
ich+0.010
ller+0.009
aged+0.009
llers+0.009
Neg logit contributions
 agreement-0.014
 particular-0.012
 Everyday-0.012
date-0.012
 invite-0.012
Tokenicken
Token ID5973
Feature activation+2.03e-02

Pos logit contributions
AAAA+0.051
nam+0.048
w+0.047
Crash+0.044
RR+0.044
Neg logit contributions
 agreement-0.107
date-0.105
rily-0.104
ini-0.099
 particular-0.098
Token.
Token ID13
Feature activation+1.41e-02

Pos logit contributions
 particular+0.036
 agreement+0.036
date+0.033
 invite+0.032
 surely+0.031
Neg logit contributions
w-0.025
aged-0.024
ages-0.022
ller-0.021
icken-0.021
Token She
Token ID1375
Feature activation+1.44e-02

Pos logit contributions
 agreement+0.027
 particular+0.024
date+0.024
 invite+0.023
 offer+0.023
Neg logit contributions
w-0.016
ich-0.015
ller-0.015
aged-0.015
 Blood-0.014
Token screamed
Token ID25421
Feature activation+9.36e-03

Pos logit contributions
 agreement+0.029
date+0.027
 particular+0.027
 Everyday+0.025
 basis+0.025
Neg logit contributions
w-0.017
aged-0.015
 screamed-0.014
ller-0.014
 bleeding-0.014
Token for
Token ID329
Feature activation+8.53e-03

Pos logit contributions
 agreement+0.016
 particular+0.015
 Everyday+0.014
date+0.014
 invite+0.014
Neg logit contributions
w-0.013
aged-0.012
ich-0.012
ller-0.012
 bleeding-0.011
Token new
Token ID649
Feature activation-2.57e-02

Pos logit contributions
w+0.026
ich+0.024
ller+0.023
aged+0.023
llers+0.022
Neg logit contributions
 agreement-0.029
 particular-0.027
 Everyday-0.025
date-0.025
 invite-0.025
Token words
Token ID2456
Feature activation+4.53e-03

Pos logit contributions
w+0.042
ich+0.040
 Blood+0.037
awed+0.037
ounced+0.036
Neg logit contributions
 agreement-0.039
 offer-0.036
 particular-0.034
 invite-0.033
 Everyday-0.033
Token every
Token ID790
Feature activation+1.68e-02

Pos logit contributions
 agreement+0.008
 particular+0.007
date+0.007
 Everyday+0.007
latable+0.007
Neg logit contributions
w-0.006
ller-0.006
uts-0.006
ich-0.005
llers-0.005
Token day
Token ID1110
Feature activation+1.14e-02

Pos logit contributions
 agreement+0.022
 Everyday+0.020
 particular+0.019
date+0.018
latable+0.018
Neg logit contributions
w-0.032
ich-0.027
ller-0.026
RR-0.025
 reading-0.025
Token.
Token ID13
Feature activation+8.74e-04

Pos logit contributions
 agreement+0.021
 particular+0.020
date+0.019
 Everyday+0.018
latable+0.017
Neg logit contributions
w-0.016
uts-0.014
ller-0.013
ich-0.013
 reading-0.013
Token Conf
Token ID7326
Feature activation-6.12e-02

Pos logit contributions
 agreement+0.002
 particular+0.001
 Everyday+0.001
date+0.001
 invite+0.001
Neg logit contributions
w-0.001
ich-0.001
ller-0.001
aged-0.001
llers-0.001
Tokenused
Token ID1484
Feature activation+2.87e-03

Pos logit contributions
w+0.080
ller+0.071
ich+0.070
aged+0.068
llers+0.067
Neg logit contributions
 agreement-0.107
 particular-0.098
date-0.095
 Everyday-0.095
 invite-0.093
Token is
Token ID318
Feature activation-3.52e-04

Pos logit contributions
 agreement+0.005
date+0.005
 particular+0.005
 Everyday+0.005
 invite+0.004
Neg logit contributions
w-0.004
ich-0.003
ller-0.003
aged-0.003
RR-0.003
Token just
Token ID655
Feature activation+1.92e-04

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
aged+0.000
llers+0.000
Neg logit contributions
 agreement-0.001
 particular-0.001
date-0.001
 Everyday-0.001
 invite-0.001
Token a
Token ID257
Feature activation-8.78e-03

Pos logit contributions
 agreement+0.000
 particular+0.000
date+0.000
 Everyday+0.000
 invite+0.000
Neg logit contributions
w-0.000
ich-0.000
aged-0.000
ller-0.000
nam-0.000
Token word
Token ID1573
Feature activation+1.08e-02

Pos logit contributions
w+0.014
ich+0.013
ller+0.012
llers+0.012
aged+0.012
Neg logit contributions
 agreement-0.013
 particular-0.012
 Everyday-0.012
date-0.012
 invite-0.011
Token cherry
Token ID23612
Feature activation-2.75e-02

Pos logit contributions
w+0.034
aged+0.031
llers+0.030
irens+0.029
ller+0.029
Neg logit contributions
 agreement-0.025
 particular-0.024
date-0.022
 Everyday-0.021
 invite-0.021
Token tree
Token ID5509
Feature activation-1.92e-02

Pos logit contributions
w+0.048
ich+0.044
ller+0.043
aged+0.043
llers+0.043
Neg logit contributions
 agreement-0.036
 particular-0.034
 Everyday-0.032
date-0.031
 invite-0.030
Token with
Token ID351
Feature activation-2.44e-02

Pos logit contributions
w+0.030
ich+0.027
ller+0.026
aged+0.025
llers+0.025
Neg logit contributions
 agreement-0.031
 particular-0.029
 Everyday-0.027
date-0.027
 invite-0.026
Token lots
Token ID6041
Feature activation-2.73e-03

Pos logit contributions
w+0.037
ich+0.035
ller+0.034
llers+0.033
irens+0.032
Neg logit contributions
 agreement-0.039
 particular-0.036
 Everyday-0.035
 invite-0.032
 fill-0.032
Token of
Token ID286
Feature activation-3.14e-02

Pos logit contributions
w+0.007
ich+0.006
ller+0.006
aged+0.006
irens+0.006
Neg logit contributions
 agreement-0.002
 particular-0.002
date-0.002
 invite-0.001
 Everyday-0.001
Token cher
Token ID21382
Feature activation-6.09e-02

Pos logit contributions
ich+0.041
w+0.040
ller+0.038
llers+0.036
irens+0.036
Neg logit contributions
 agreement-0.056
 Everyday-0.053
 particular-0.052
 offer-0.049
 fill-0.048
Tokenries
Token ID1678
Feature activation-1.02e-02

Pos logit contributions
RR+0.088
EEP+0.087
ses+0.085
AAAA+0.081
sed+0.081
Neg logit contributions
 particular-0.076
 Everyday-0.073
 fill-0.072
 agreement-0.070
 longer-0.069
Token on
Token ID319
Feature activation+1.91e-03

Pos logit contributions
w+0.017
ich+0.015
ller+0.015
aged+0.014
llers+0.014
Neg logit contributions
 agreement-0.015
 particular-0.014
 Everyday-0.013
date-0.013
 invite-0.013
Token it
Token ID340
Feature activation+8.58e-04

Pos logit contributions
 agreement+0.003
 particular+0.003
date+0.003
 invite+0.003
 Everyday+0.003
Neg logit contributions
w-0.003
ller-0.003
ich-0.003
llers-0.003
aged-0.002
Token.
Token ID13
Feature activation-8.77e-03

Pos logit contributions
 agreement+0.002
 particular+0.001
date+0.001
 invite+0.001
 Everyday+0.001
Neg logit contributions
w-0.001
uts-0.001
ller-0.001
aged-0.001
ich-0.001
Token She
Token ID1375
Feature activation-1.37e-02

Pos logit contributions
w+0.015
ich+0.013
ller+0.012
aged+0.012
llers+0.012
Neg logit contributions
 agreement-0.014
 particular-0.013
 Everyday-0.012
date-0.012
 invite-0.012
Token went
Token ID1816
Feature activation-1.55e-02

Pos logit contributions
w+0.005
aged+0.004
ich+0.004
ller+0.004
 Blood+0.004
Neg logit contributions
 agreement-0.005
 particular-0.004
 Everyday-0.004
date-0.004
 offer-0.004
Token she
Token ID673
Feature activation+6.57e-03

Pos logit contributions
w+0.026
ich+0.025
ller+0.025
llers+0.024
aged+0.023
Neg logit contributions
 particular-0.021
 agreement-0.021
 Everyday-0.020
 invite-0.020
date-0.020
Token searched
Token ID16499
Feature activation-1.50e-02

Pos logit contributions
 agreement+0.014
 particular+0.011
 Everyday+0.011
date+0.011
ainer+0.011
Neg logit contributions
w-0.008
aged-0.007
ich-0.006
 Blood-0.006
ller-0.006
Token for
Token ID329
Feature activation-6.95e-03

Pos logit contributions
w+0.020
ich+0.017
aged+0.017
ller+0.017
llers+0.016
Neg logit contributions
 agreement-0.027
 particular-0.025
date-0.024
 Everyday-0.024
 invite-0.023
Token or
Token ID393
Feature activation-1.11e-02

Pos logit contributions
w+0.010
ich+0.009
ller+0.008
aged+0.008
llers+0.008
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.011
date-0.011
 invite-0.010
Tokennam
Token ID7402
Feature activation-6.08e-02

Pos logit contributions
w+0.004
ich+0.003
ller+0.002
aged+0.002
llers+0.002
Neg logit contributions
 agreement-0.031
 particular-0.029
 Everyday-0.028
date-0.028
 invite-0.028
Tokenents
Token ID658
Feature activation+9.65e-03

Pos logit contributions
AAAA+0.100
w+0.095
rob+0.087
RR+0.087
 elbows+0.087
Neg logit contributions
 Everyday-0.082
 agreement-0.081
 of-0.077
 particular-0.076
date-0.075
Token,
Token ID11
Feature activation-3.85e-03

Pos logit contributions
 agreement+0.019
 particular+0.017
 Everyday+0.016
date+0.016
 invite+0.016
Neg logit contributions
w-0.012
uts-0.011
aged-0.010
ich-0.010
ller-0.010
Token and
Token ID290
Feature activation-5.99e-03

Pos logit contributions
w+0.004
nam+0.004
ich+0.004
aged+0.004
 reading+0.004
Neg logit contributions
 agreement-0.008
date-0.007
 Everyday-0.007
 particular-0.007
 invite-0.006
Token the
Token ID262
Feature activation-2.03e-02

Pos logit contributions
w+0.009
ich+0.008
aged+0.007
nam+0.007
ller+0.007
Neg logit contributions
 agreement-0.010
 particular-0.009
date-0.009
 Everyday-0.009
 invite-0.009
Token more
Token ID517
Feature activation-1.54e-02

Pos logit contributions
w+0.031
ich+0.029
ller+0.028
aged+0.027
nam+0.027
Neg logit contributions
 agreement-0.032
 particular-0.029
date-0.028
 Everyday-0.028
 invite-0.027
Token€
Token ID26391
Feature activation-2.46e-02

Pos logit contributions
 agreement+0.027
 particular+0.025
 Everyday+0.023
date+0.023
 invite+0.023
Neg logit contributions
w-0.032
ich-0.029
ller-0.028
aged-0.027
llers-0.027
Token
Token ID198
Feature activation+4.87e-03

Pos logit contributions
w+0.042
nam+0.040
 bleeding+0.036
 elbows+0.036
destruct+0.035
Neg logit contributions
 Everyday-0.031
 agreement-0.028
 particular-0.028
 Whenever-0.028
 invite-0.028
Token
Token ID198
Feature activation+7.10e-03

Pos logit contributions
 agreement+0.010
 particular+0.009
date+0.009
 invite+0.009
 fill+0.008
Neg logit contributions
w-0.005
ich-0.005
ller-0.005
irens-0.004
aged-0.004
TokenMiss
Token ID17140
Feature activation-7.73e-03

Pos logit contributions
 agreement+0.013
 particular+0.012
 fill+0.012
 invite+0.012
 offer+0.012
Neg logit contributions
w-0.009
ich-0.008
RR-0.007
irens-0.007
ller-0.007
Token O
Token ID440
Feature activation+6.76e-03

Pos logit contributions
w+0.012
ich+0.011
ller+0.010
aged+0.010
llers+0.010
Neg logit contributions
 agreement-0.013
 particular-0.012
 Everyday-0.011
date-0.011
 invite-0.011
Tokenstr
Token ID2536
Feature activation-6.06e-02

Pos logit contributions
 agreement+0.014
 particular+0.014
date+0.014
 Everyday+0.013
 invite+0.013
Neg logit contributions
w-0.006
nam-0.005
ich-0.005
 elbows-0.005
aged-0.005
Tokenich
Token ID488
Feature activation+1.80e-02

Pos logit contributions
nam+0.066
 elbows+0.050
w+0.050
 satellites+0.049
AAAA+0.048
Neg logit contributions
 Everyday-0.097
 agreement-0.096
date-0.094
ini-0.093
 particular-0.093
Token said
Token ID531
Feature activation+2.68e-02

Pos logit contributions
 agreement+0.035
 particular+0.032
date+0.031
 Everyday+0.029
 offer+0.029
Neg logit contributions
w-0.023
ller-0.020
ich-0.019
aged-0.018
llers-0.018
Token,
Token ID11
Feature activation+6.28e-03

Pos logit contributions
 agreement+0.055
 particular+0.052
 Everyday+0.049
date+0.049
 invite+0.047
Neg logit contributions
w-0.029
aged-0.025
ller-0.024
ich-0.023
uts-0.023
Token �
Token ID6184
Feature activation+1.11e-02

Pos logit contributions
 agreement+0.013
 particular+0.012
date+0.012
 Everyday+0.011
 invite+0.011
Neg logit contributions
w-0.006
ller-0.006
ich-0.006
aged-0.006
llers-0.006
Token�
Token ID95
Feature activation+9.36e-03

Pos logit contributions
 agreement+0.018
 particular+0.016
 Everyday+0.015
date+0.015
 invite+0.015
Neg logit contributions
w-0.017
ich-0.016
ller-0.015
llers-0.015
aged-0.015
Token explore
Token ID7301
Feature activation+6.71e-04

Pos logit contributions
 agreement+0.015
 particular+0.014
 Everyday+0.013
date+0.013
 invite+0.012
Neg logit contributions
w-0.021
ich-0.020
ller-0.019
aged-0.019
llers-0.019
Token.
Token ID13
Feature activation-4.80e-03

Pos logit contributions
 agreement+0.001
 particular+0.001
 Everyday+0.001
date+0.001
 invite+0.001
Neg logit contributions
w-0.001
ich-0.001
ller-0.001
aged-0.001
llers-0.001
Token She
Token ID1375
Feature activation+4.36e-03

Pos logit contributions
w+0.009
ich+0.008
ller+0.007
aged+0.007
llers+0.007
Neg logit contributions
 agreement-0.007
 particular-0.006
 Everyday-0.006
date-0.006
 invite-0.006
Token noticed
Token ID6810
Feature activation-3.35e-03

Pos logit contributions
 agreement+0.008
date+0.007
 particular+0.007
 Everyday+0.007
 offer+0.007
Neg logit contributions
w-0.006
ich-0.005
aged-0.005
ller-0.005
llers-0.005
Token the
Token ID262
Feature activation-3.13e-02

Pos logit contributions
w+0.006
ich+0.005
ller+0.005
EEP+0.005
aged+0.005
Neg logit contributions
 agreement-0.005
 particular-0.004
 Everyday-0.004
date-0.004
 invite-0.004
Token old
Token ID1468
Feature activation-6.01e-02

Pos logit contributions
w+0.054
ich+0.049
ller+0.048
aged+0.048
llers+0.048
Neg logit contributions
 agreement-0.046
 particular-0.041
 Everyday-0.039
 offer-0.038
 invite-0.037
Token chest
Token ID7721
Feature activation-1.63e-02

Pos logit contributions
w+0.128
inches+0.120
ich+0.117
irens+0.114
llers+0.114
Neg logit contributions
 agreement-0.068
 particular-0.060
 offer-0.060
 Everyday-0.057
 fill-0.055
Token and
Token ID290
Feature activation-1.14e-02

Pos logit contributions
w+0.018
 reading+0.017
aged+0.017
ich+0.017
 crying+0.017
Neg logit contributions
 agreement-0.033
date-0.029
 Everyday-0.027
 particular-0.027
 basis-0.027
Token kn
Token ID638
Feature activation-2.28e-02

Pos logit contributions
w+0.018
ich+0.017
ller+0.016
llers+0.015
irens+0.015
Neg logit contributions
 agreement-0.018
 particular-0.017
 Everyday-0.016
 invite-0.015
 offer-0.015
Tokenelt
Token ID2120
Feature activation-2.64e-02

Pos logit contributions
w+0.027
AAAA+0.025
RR+0.024
rob+0.023
ller+0.022
Neg logit contributions
 agreement-0.039
 Everyday-0.039
 sill-0.039
 fill-0.037
date-0.037
Token down
Token ID866
Feature activation-5.81e-04

Pos logit contributions
w+0.034
ich+0.031
ller+0.030
llers+0.029
aged+0.028
Neg logit contributions
 agreement-0.049
 particular-0.045
 Everyday-0.043
date-0.042
 invite-0.042
Token continued
Token ID3767
Feature activation-6.48e-03

Pos logit contributions
 agreement+0.016
 particular+0.015
date+0.014
 Everyday+0.014
 invite+0.014
Neg logit contributions
w-0.013
ich-0.011
aged-0.011
ller-0.011
llers-0.010
Token to
Token ID284
Feature activation+1.77e-02

Pos logit contributions
w+0.009
ich+0.008
ller+0.007
aged+0.007
llers+0.007
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.010
date-0.010
 invite-0.010
Token climb
Token ID12080
Feature activation-1.16e-03

Pos logit contributions
 agreement+0.028
 particular+0.026
date+0.026
 Everyday+0.025
cloth+0.023
Neg logit contributions
w-0.025
ich-0.022
ller-0.022
llers-0.021
aged-0.020
Token until
Token ID1566
Feature activation+3.06e-02

Pos logit contributions
w+0.002
ller+0.001
ich+0.001
irens+0.001
llers+0.001
Neg logit contributions
 agreement-0.002
date-0.002
 particular-0.002
 invite-0.002
 Everyday-0.002
Token they
Token ID484
Feature activation+8.45e-03

Pos logit contributions
 agreement+0.061
 particular+0.055
date+0.055
 Everyday+0.053
 invite+0.053
Neg logit contributions
w-0.035
ich-0.030
ller-0.030
llers-0.029
irens-0.029
Token couldn
Token ID3521
Feature activation+4.16e-02

Pos logit contributions
 agreement+0.014
 particular+0.013
date+0.012
 Everyday+0.012
 invite+0.011
Neg logit contributions
w-0.013
ich-0.012
ller-0.011
aged-0.011
llers-0.011
Token't
Token ID470
Feature activation+1.38e-02

Pos logit contributions
 agreement+0.095
 particular+0.084
date+0.082
 invite+0.080
 offer+0.079
Neg logit contributions
w-0.039
ller-0.031
uts-0.031
ich-0.030
llers-0.029
Token climb
Token ID12080
Feature activation+6.68e-03

Pos logit contributions
 agreement+0.020
 particular+0.018
date+0.018
 Everyday+0.018
cloth+0.016
Neg logit contributions
w-0.021
ich-0.020
ller-0.019
uts-0.018
llers-0.018
Token any
Token ID597
Feature activation+7.69e-03

Pos logit contributions
 agreement+0.011
 particular+0.010
date+0.010
 invite+0.010
 Everyday+0.010
Neg logit contributions
w-0.010
ller-0.009
ich-0.009
llers-0.009
irens-0.008
Token higher
Token ID2440
Feature activation-1.01e-02

Pos logit contributions
 agreement+0.012
date+0.011
 Everyday+0.011
 particular+0.010
 invite+0.010
Neg logit contributions
w-0.011
ls-0.009
ich-0.009
irens-0.009
 elbows-0.009
Token.
Token ID13
Feature activation+9.67e-03

Pos logit contributions
w+0.013
ich+0.012
ller+0.012
llers+0.011
uts+0.010
Neg logit contributions
 agreement-0.019
date-0.017
 invite-0.016
 Everyday-0.016
 particular-0.016
Token play
Token ID711
Feature activation+1.57e-02

Pos logit contributions
rily+0.024
 agreement+0.023
 surely+0.020
 checkout+0.019
 refreshing+0.019
Neg logit contributions
w-0.020
 Blood-0.018
 elbows-0.018
 reading-0.017
 crying-0.016
Token in
Token ID287
Feature activation+6.58e-03

Pos logit contributions
 agreement+0.030
 particular+0.028
 Everyday+0.028
 offer+0.025
 invite+0.025
Neg logit contributions
w-0.019
ich-0.017
aged-0.016
RR-0.015
 elbows-0.015
Token the
Token ID262
Feature activation-1.44e-03

Pos logit contributions
 agreement+0.011
 particular+0.010
date+0.010
 Everyday+0.010
 invite+0.009
Neg logit contributions
w-0.009
ich-0.009
ller-0.008
llers-0.008
irens-0.008
Token park
Token ID3952
Feature activation+8.24e-03

Pos logit contributions
w+0.002
ich+0.002
irens+0.002
ller+0.002
llers+0.002
Neg logit contributions
 agreement-0.002
 particular-0.002
 Everyday-0.002
date-0.002
 invite-0.002
Token.
Token ID13
Feature activation+1.59e-02

Pos logit contributions
 agreement+0.016
 particular+0.014
 Everyday+0.014
date+0.014
latable+0.013
Neg logit contributions
w-0.009
ich-0.008
 reading-0.008
aged-0.007
uts-0.007
Token One
Token ID1881
Feature activation+3.91e-02

Pos logit contributions
 agreement+0.022
date+0.020
 particular+0.018
 offer+0.018
 invite+0.018
Neg logit contributions
w-0.025
llers-0.023
 Blood-0.023
ller-0.023
RR-0.023
Token day
Token ID1110
Feature activation-1.92e-02

Pos logit contributions
 agreement+0.059
date+0.050
rily+0.049
 Everyday+0.045
 invite+0.045
Neg logit contributions
w-0.068
ich-0.058
EEP-0.054
 Blood-0.054
aged-0.053
Token,
Token ID11
Feature activation+3.06e-03

Pos logit contributions
ich+0.038
ller+0.038
llers+0.037
w+0.036
 elbows+0.035
Neg logit contributions
 particular-0.022
 agreement-0.019
 Everyday-0.018
 of-0.018
 invite-0.017
Token they
Token ID484
Feature activation+1.07e-02

Pos logit contributions
 agreement+0.005
 particular+0.005
 Everyday+0.005
date+0.005
 invite+0.005
Neg logit contributions
w-0.004
ich-0.004
ller-0.004
llers-0.004
aged-0.004
Token see
Token ID766
Feature activation-1.73e-03

Pos logit contributions
 agreement+0.019
 particular+0.017
 Everyday+0.017
date+0.017
rily+0.016
Neg logit contributions
w-0.014
ich-0.011
AAAA-0.011
nam-0.011
 Blood-0.011
Token a
Token ID257
Feature activation-1.44e-02

Pos logit contributions
w+0.003
ller+0.002
ich+0.002
llers+0.002
aged+0.002
Neg logit contributions
 agreement-0.003
 particular-0.003
 Everyday-0.002
 invite-0.002
date-0.002
Token and
Token ID290
Feature activation-4.23e-03

Pos logit contributions
 agreement+0.051
 particular+0.048
date+0.046
 Everyday+0.043
 invite+0.041
Neg logit contributions
w-0.037
aged-0.034
ller-0.032
ich-0.030
llers-0.029
Token the
Token ID262
Feature activation+2.00e-02

Pos logit contributions
w+0.007
ich+0.006
ller+0.006
aged+0.006
llers+0.006
Neg logit contributions
 agreement-0.007
 particular-0.006
 Everyday-0.006
date-0.006
 invite-0.005
Token frog
Token ID21264
Feature activation+1.29e-02

Pos logit contributions
 agreement+0.031
date+0.028
 particular+0.028
 basis+0.027
 Everyday+0.027
Neg logit contributions
w-0.030
ich-0.027
 crying-0.025
 Blood-0.024
 bleeding-0.024
Token worked
Token ID3111
Feature activation-1.47e-03

Pos logit contributions
 agreement+0.023
 particular+0.022
date+0.021
 Everyday+0.020
 basis+0.019
Neg logit contributions
w-0.017
aged-0.016
ich-0.014
awed-0.014
ged-0.013
Token together
Token ID1978
Feature activation+7.51e-03

Pos logit contributions
w+0.001
aged+0.001
ller+0.001
irens+0.001
llers+0.001
Neg logit contributions
 agreement-0.003
 particular-0.003
date-0.003
 invite-0.003
 Everyday-0.003
Token to
Token ID284
Feature activation+3.78e-02

Pos logit contributions
 agreement+0.016
 particular+0.014
date+0.014
 Everyday+0.014
 invite+0.014
Neg logit contributions
aged-0.007
w-0.007
uts-0.006
 pressing-0.006
ged-0.006
Token clean
Token ID3424
Feature activation+9.12e-03

Pos logit contributions
 agreement+0.056
date+0.053
 particular+0.051
 Everyday+0.051
 eater+0.049
Neg logit contributions
w-0.045
aged-0.043
icken-0.043
ich-0.042
 whist-0.042
Token the
Token ID262
Feature activation+1.16e-02

Pos logit contributions
 agreement+0.018
date+0.018
 particular+0.017
 invite+0.017
 Everyday+0.017
Neg logit contributions
w-0.009
ller-0.008
aged-0.008
icken-0.008
llers-0.008
Token pond
Token ID16723
Feature activation+8.16e-03

Pos logit contributions
 particular+0.018
 agreement+0.017
date+0.017
 Everyday+0.016
 Whenever+0.015
Neg logit contributions
w-0.017
ich-0.017
 bleeding-0.015
 crying-0.015
ller-0.015
Token.
Token ID13
Feature activation+1.05e-02

Pos logit contributions
 agreement+0.016
 particular+0.015
date+0.015
 Everyday+0.014
 invite+0.014
Neg logit contributions
aged-0.009
w-0.008
ages-0.007
uts-0.007
icken-0.007
Token They
Token ID1119
Feature activation+2.26e-03

Pos logit contributions
 agreement+0.019
 particular+0.018
date+0.018
 offer+0.017
 invite+0.017
Neg logit contributions
 Blood-0.010
aged-0.010
llers-0.010
ller-0.010
uts-0.010
Token some
Token ID617
Feature activation+2.76e-02

Pos logit contributions
 agreement+0.006
 particular+0.005
date+0.005
 Everyday+0.005
 offer+0.005
Neg logit contributions
w-0.003
aged-0.003
ich-0.003
llers-0.003
 Blood-0.003
Token special
Token ID2041
Feature activation+1.24e-02

Pos logit contributions
 agreement+0.050
date+0.048
 surely+0.045
 particular+0.044
 basis+0.042
Neg logit contributions
w-0.038
 bleeding-0.034
 blood-0.032
 bruises-0.030
aged-0.030
Token plants
Token ID6134
Feature activation+1.09e-02

Pos logit contributions
 agreement+0.021
 surely+0.019
 particular+0.019
date+0.019
 Everyday+0.019
Neg logit contributions
w-0.019
aged-0.016
 elbows-0.015
inches-0.015
 bleeding-0.015
Token growing
Token ID3957
Feature activation+1.83e-02

Pos logit contributions
 agreement+0.024
 particular+0.021
date+0.020
 Everyday+0.019
ainer+0.019
Neg logit contributions
w-0.012
uts-0.011
 crying-0.010
aged-0.010
ich-0.010
Token on
Token ID319
Feature activation+3.80e-02

Pos logit contributions
 agreement+0.043
date+0.038
 invite+0.037
 particular+0.036
 Everyday+0.035
Neg logit contributions
w-0.019
ls-0.015
aged-0.013
uts-0.013
ing-0.013
Token his
Token ID465
Feature activation+3.60e-02

Pos logit contributions
 agreement+0.083
date+0.081
 invite+0.079
rene+0.075
 host+0.073
Neg logit contributions
w-0.027
RR-0.024
icken-0.023
 yourselves-0.021
 touching-0.021
Token body
Token ID1767
Feature activation+1.54e-02

Pos logit contributions
 agreement+0.062
 Everyday+0.061
date+0.060
 particular+0.057
rene+0.054
Neg logit contributions
ich-0.049
 elbows-0.043
w-0.041
 bleeding-0.040
 crying-0.040
Token.
Token ID13
Feature activation+1.86e-02

Pos logit contributions
 agreement+0.032
 particular+0.029
 invite+0.026
date+0.026
 Everyday+0.026
Neg logit contributions
uts-0.018
w-0.018
ich-0.016
ages-0.015
aged-0.015
Token He
Token ID679
Feature activation-1.45e-03

Pos logit contributions
 agreement+0.032
 particular+0.031
date+0.029
 invite+0.028
 offer+0.027
Neg logit contributions
w-0.024
ller-0.023
ich-0.023
llers-0.022
RR-0.022
Token took
Token ID1718
Feature activation+9.90e-03

Pos logit contributions
w+0.002
ller+0.002
ich+0.002
aged+0.002
llers+0.002
Neg logit contributions
 agreement-0.003
 particular-0.002
date-0.002
 Everyday-0.002
 offer-0.002
Token out
Token ID503
Feature activation+3.65e-04

Pos logit contributions
 agreement+0.021
 particular+0.019
date+0.018
 Everyday+0.018
 invite+0.018
Neg logit contributions
w-0.011
aged-0.008
irens-0.008
 Blood-0.007
 Pur-0.007
Token her
Token ID607
Feature activation+1.40e-02

Pos logit contributions
date+0.045
 agreement+0.041
 surely+0.040
 invite+0.040
 particular+0.040
Neg logit contributions
 flashing-0.016
nam-0.015
 pressing-0.015
 fighting-0.014
 crying-0.014
Token qualities
Token ID14482
Feature activation+5.43e-03

Pos logit contributions
date+0.023
 agreement+0.022
 Everyday+0.022
 particular+0.022
iona+0.020
Neg logit contributions
 reading-0.017
 crying-0.016
aged-0.016
ller-0.016
 praying-0.016
Token and
Token ID290
Feature activation-1.20e-02

Pos logit contributions
 agreement+0.011
date+0.010
 particular+0.010
 invite+0.009
ainer+0.009
Neg logit contributions
uts-0.005
w-0.005
 flashing-0.005
ller-0.005
 crying-0.005
Token skill
Token ID5032
Feature activation-1.19e-02

Pos logit contributions
w+0.014
aged+0.014
llers+0.013
ller+0.013
nam+0.013
Neg logit contributions
 agreement-0.022
 particular-0.020
date-0.020
 Everyday-0.019
latable-0.018
Token.
Token ID13
Feature activation+1.26e-02

Pos logit contributions
w+0.017
ich+0.015
aged+0.015
ller+0.015
llers+0.015
Neg logit contributions
 agreement-0.020
 particular-0.019
date-0.018
 Everyday-0.018
 invite-0.017
Token M
Token ID337
Feature activation+3.87e-02

Pos logit contributions
 agreement+0.024
 particular+0.021
date+0.021
 offer+0.021
 invite+0.020
Neg logit contributions
w-0.013
 Blood-0.013
llers-0.013
ich-0.013
aged-0.012
Tokenilly
Token ID6548
Feature activation+6.85e-03

Pos logit contributions
 agreement+0.059
date+0.052
 particular+0.052
 invite+0.049
 fill+0.048
Neg logit contributions
w-0.065
ich-0.061
ller-0.057
llers-0.056
aged-0.056
Token was
Token ID373
Feature activation+5.05e-03

Pos logit contributions
 agreement+0.013
 particular+0.012
date+0.011
 Everyday+0.011
 offer+0.010
Neg logit contributions
w-0.009
ller-0.008
aged-0.008
ich-0.008
llers-0.008
Token so
Token ID523
Feature activation-5.40e-03

Pos logit contributions
 agreement+0.009
 particular+0.009
date+0.008
 offer+0.008
 Everyday+0.008
Neg logit contributions
aged-0.006
 reading-0.006
ich-0.005
 pressing-0.005
 crying-0.005
Token happy
Token ID3772
Feature activation-2.16e-02

Pos logit contributions
w+0.006
aged+0.006
ich+0.006
ller+0.005
llers+0.005
Neg logit contributions
 agreement-0.010
 particular-0.010
date-0.010
 Everyday-0.009
 invite-0.009
Token to
Token ID284
Feature activation+1.88e-02

Pos logit contributions
w+0.029
ich+0.027
ller+0.026
llers+0.025
aged+0.025
Neg logit contributions
 agreement-0.039
 particular-0.036
date-0.034
 Everyday-0.034
 invite-0.033
Token€
Token ID26391
Feature activation-9.67e-03

Pos logit contributions
 agreement+0.035
 particular+0.032
 Everyday+0.030
date+0.030
 invite+0.029
Neg logit contributions
w-0.040
ich-0.035
ller-0.034
aged-0.034
llers-0.034
Token�
Token ID447
Feature activation+8.59e-03

Pos logit contributions
w+0.013
ich+0.011
ller+0.011
nam+0.011
aged+0.011
Neg logit contributions
 agreement-0.016
date-0.015
 particular-0.015
 Everyday-0.015
 invite-0.015
Token�
Token ID251
Feature activation+2.61e-02

Pos logit contributions
 agreement+0.008
 particular+0.007
 Everyday+0.006
date+0.006
 invite+0.005
Neg logit contributions
w-0.019
ich-0.018
ller-0.018
aged-0.018
llers-0.017
Tokensw
Token ID2032
Feature activation+8.86e-03

Pos logit contributions
 agreement+0.049
 particular+0.046
date+0.043
 invite+0.042
 fill+0.042
Neg logit contributions
w-0.031
irens-0.027
aged-0.026
 Pur-0.026
 Hel-0.025
Tokenimming
Token ID27428
Feature activation+8.23e-03

Pos logit contributions
 agreement+0.016
 particular+0.015
 invite+0.014
date+0.014
 Everyday+0.014
Neg logit contributions
w-0.012
ich-0.011
ller-0.011
llers-0.010
aged-0.010
Token,
Token ID11
Feature activation+9.93e-03

Pos logit contributions
 agreement+0.014
date+0.014
 particular+0.013
 invite+0.013
 Everyday+0.013
Neg logit contributions
w-0.011
aged-0.010
ged-0.008
irens-0.008
awed-0.008
Token playing
Token ID2712
Feature activation+1.00e-02

Pos logit contributions
 agreement+0.017
date+0.017
 particular+0.017
 Everyday+0.015
 offer+0.015
Neg logit contributions
w-0.012
ich-0.012
ller-0.012
aged-0.011
llers-0.011
Token and
Token ID290
Feature activation-3.11e-03

Pos logit contributions
 particular+0.019
 Everyday+0.018
 agreement+0.018
 invite+0.017
 surely+0.017
Neg logit contributions
w-0.011
aged-0.010
RR-0.009
AAAA-0.009
icken-0.009
Token making
Token ID1642
Feature activation+1.90e-02

Pos logit contributions
w+0.003
ich+0.003
aged+0.003
llers+0.003
ller+0.003
Neg logit contributions
 agreement-0.006
 particular-0.006
date-0.006
 Everyday-0.006
 offer-0.005
Token lots
Token ID6041
Feature activation+2.03e-02

Pos logit contributions
 agreement+0.035
 particular+0.034
date+0.032
 Everyday+0.032
 surely+0.030
Neg logit contributions
w-0.023
aged-0.021
ich-0.020
llers-0.020
ller-0.019
Token of
Token ID286
Feature activation+9.36e-03

Pos logit contributions
 agreement+0.026
date+0.023
 particular+0.022
 surely+0.022
 Everyday+0.020
Neg logit contributions
uts-0.033
irens-0.033
w-0.033
ller-0.032
ouncing-0.032
Token dad
Token ID9955
Feature activation-7.10e-03

Pos logit contributions
w+0.021
ller+0.019
llers+0.018
ich+0.018
irens+0.018
Neg logit contributions
 agreement-0.020
 particular-0.018
 invite-0.017
 fill-0.016
 offer-0.016
Token give
Token ID1577
Feature activation+1.34e-02

Pos logit contributions
w+0.011
 fighting+0.009
inches+0.009
aged+0.009
 reading+0.009
Neg logit contributions
 particular-0.010
 agreement-0.010
date-0.010
 Everyday-0.009
 basis-0.009
Token them
Token ID606
Feature activation+2.00e-02

Pos logit contributions
 agreement+0.022
 particular+0.021
date+0.020
 Everyday+0.020
 basis+0.019
Neg logit contributions
w-0.020
aged-0.017
inches-0.016
irens-0.016
ich-0.016
Token a
Token ID257
Feature activation+1.13e-02

Pos logit contributions
 agreement+0.041
 particular+0.039
date+0.039
 Everyday+0.037
 surely+0.035
Neg logit contributions
w-0.020
nam-0.016
aged-0.015
ich-0.015
irens-0.015
Token net
Token ID2010
Feature activation+2.00e-02

Pos logit contributions
 agreement+0.018
 Everyday+0.017
date+0.017
 particular+0.016
 Whenever+0.016
Neg logit contributions
w-0.016
ich-0.014
inches-0.013
AAAA-0.013
 Blood-0.013
Token and
Token ID290
Feature activation+3.56e-03

Pos logit contributions
 agreement+0.044
 particular+0.041
date+0.039
 Everyday+0.039
rene+0.037
Neg logit contributions
uts-0.018
w-0.017
ages-0.015
inches-0.014
aged-0.014
Token a
Token ID257
Feature activation+5.58e-03

Pos logit contributions
 agreement+0.007
date+0.007
 particular+0.007
 Everyday+0.006
 basis+0.006
Neg logit contributions
w-0.004
inches-0.003
nam-0.003
aged-0.003
destruct-0.003
Token bucket
Token ID19236
Feature activation+1.47e-02

Pos logit contributions
 agreement+0.009
 Everyday+0.009
 particular+0.009
date+0.008
 Whenever+0.008
Neg logit contributions
w-0.008
ich-0.007
AAAA-0.006
nam-0.006
inches-0.006
Token.
Token ID13
Feature activation+2.22e-02

Pos logit contributions
 agreement+0.030
date+0.029
 particular+0.028
 basis+0.026
 Everyday+0.025
Neg logit contributions
uts-0.016
w-0.015
inches-0.014
aged-0.014
ages-0.012
Token Ben
Token ID3932
Feature activation+2.96e-03

Pos logit contributions
 agreement+0.045
date+0.044
 particular+0.042
 surely+0.039
 basis+0.038
Neg logit contributions
irens-0.020
 Blood-0.020
uts-0.019
inches-0.019
AAAA-0.019
Token and
Token ID290
Feature activation-5.10e-03

Pos logit contributions
 agreement+0.005
date+0.005
 particular+0.005
 Everyday+0.004
 invite+0.004
Neg logit contributions
w-0.004
inches-0.004
ich-0.004
aged-0.004
ller-0.004
Token,
Token ID11
Feature activation+1.21e-02

Pos logit contributions
 agreement+0.044
 particular+0.041
 Everyday+0.040
date+0.039
 invite+0.037
Neg logit contributions
w-0.029
ich-0.024
aged-0.024
ller-0.023
EEP-0.023
Token you
Token ID345
Feature activation+8.53e-03

Pos logit contributions
 agreement+0.029
date+0.028
 particular+0.027
 Everyday+0.026
rene+0.025
Neg logit contributions
w-0.010
 accidents-0.007
ich-0.007
 fighting-0.006
EEP-0.006
Token can
Token ID460
Feature activation+7.74e-03

Pos logit contributions
 agreement+0.012
 particular+0.011
date+0.010
 Everyday+0.009
rily+0.009
Neg logit contributions
w-0.016
EEP-0.013
aged-0.013
RR-0.012
ller-0.012
Token borrow
Token ID8804
Feature activation+1.39e-02

Pos logit contributions
 agreement+0.012
 particular+0.011
 refreshing+0.010
igator+0.010
rily+0.010
Neg logit contributions
w-0.012
ller-0.010
llers-0.010
ich-0.010
uts-0.010
Token one
Token ID530
Feature activation+2.34e-02

Pos logit contributions
 agreement+0.027
 particular+0.025
date+0.024
 Everyday+0.024
 invite+0.023
Neg logit contributions
w-0.017
ller-0.015
ich-0.015
llers-0.014
irens-0.014
Token.
Token ID13
Feature activation+9.29e-03

Pos logit contributions
 agreement+0.043
 Everyday+0.038
date+0.036
rene+0.036
latable+0.035
Neg logit contributions
w-0.033
ich-0.030
ller-0.026
enn-0.025
 bleeding-0.024
Token Here
Token ID3423
Feature activation+1.56e-02

Pos logit contributions
 agreement+0.017
 particular+0.016
date+0.016
 fill+0.014
 offer+0.014
Neg logit contributions
w-0.010
ller-0.010
 Pur-0.009
llers-0.009
EEP-0.009
Token,
Token ID11
Feature activation-2.55e-03

Pos logit contributions
 agreement+0.032
 particular+0.030
date+0.029
 Everyday+0.028
 invite+0.026
Neg logit contributions
w-0.019
ich-0.015
ller-0.014
aged-0.014
inches-0.013
Token take
Token ID1011
Feature activation+4.50e-03

Pos logit contributions
w+0.003
nam+0.003
destruct+0.003
 Pur+0.002
ich+0.002
Neg logit contributions
 agreement-0.005
 particular-0.004
date-0.004
 Everyday-0.004
latable-0.004
Token this
Token ID428
Feature activation+6.85e-03

Pos logit contributions
 agreement+0.009
 particular+0.008
date+0.008
 Everyday+0.008
 invite+0.008
Neg logit contributions
w-0.006
aged-0.004
irens-0.004
ller-0.004
ich-0.004
Token one
Token ID530
Feature activation+2.47e-02

Pos logit contributions
 agreement+0.013
 Everyday+0.012
 particular+0.012
date+0.011
 Whenever+0.011
Neg logit contributions
w-0.008
ich-0.008
ller-0.007
aged-0.007
 bleeding-0.007
TokenTheir
Token ID14574
Feature activation+8.37e-03

Pos logit contributions
 agreement+0.014
 particular+0.013
 Everyday+0.012
date+0.012
 invite+0.012
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
llers-0.010
aged-0.009
Token mom
Token ID1995
Feature activation+5.31e-03

Pos logit contributions
 agreement+0.014
 particular+0.013
date+0.012
 Everyday+0.012
 invite+0.012
Neg logit contributions
w-0.012
ich-0.011
ller-0.011
aged-0.011
llers-0.011
Token and
Token ID290
Feature activation+1.82e-03

Pos logit contributions
 agreement+0.008
 particular+0.008
date+0.007
 Everyday+0.007
 invite+0.007
Neg logit contributions
w-0.008
ich-0.008
ller-0.008
aged-0.007
llers-0.007
Token dad
Token ID9955
Feature activation-1.06e-02

Pos logit contributions
 agreement+0.004
 particular+0.003
 Everyday+0.003
date+0.003
 invite+0.003
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
llers-0.002
Token looked
Token ID3114
Feature activation-1.13e-02

Pos logit contributions
w+0.016
ich+0.014
ller+0.014
aged+0.013
llers+0.013
Neg logit contributions
 agreement-0.018
 particular-0.016
 Everyday-0.015
date-0.015
 invite-0.015
Token at
Token ID379
Feature activation+3.22e-02

Pos logit contributions
w+0.013
ich+0.011
aged+0.011
ller+0.011
inches+0.011
Neg logit contributions
 agreement-0.022
 particular-0.021
date-0.020
 Everyday-0.020
 invite-0.019
Token them
Token ID606
Feature activation-3.60e-04

Pos logit contributions
date+0.058
 agreement+0.057
 particular+0.056
 invite+0.055
 Everyday+0.054
Neg logit contributions
w-0.036
irens-0.032
ich-0.031
aged-0.031
llers-0.031
Token with
Token ID351
Feature activation+9.73e-03

Pos logit contributions
 crying+0.000
aged+0.000
 bleeding+0.000
 fighting+0.000
 reading+0.000
Neg logit contributions
 particular-0.001
date-0.001
 agreement-0.001
 invite-0.001
 Everyday-0.001
Token sad
Token ID6507
Feature activation-1.13e-02

Pos logit contributions
date+0.019
 particular+0.018
 agreement+0.018
 Everyday+0.017
iona+0.017
Neg logit contributions
 elbows-0.009
 bleeding-0.009
 flashing-0.009
w-0.009
 blood-0.009
Token eyes
Token ID2951
Feature activation+2.52e-02

Pos logit contributions
w+0.011
aged+0.010
irens+0.009
ller+0.009
ich+0.009
Neg logit contributions
 agreement-0.023
 particular-0.023
date-0.022
 Everyday-0.022
 invite-0.021
Token.
Token ID13
Feature activation+2.33e-02

Pos logit contributions
 agreement+0.049
date+0.049
 particular+0.048
iona+0.045
 invite+0.045
Neg logit contributions
aged-0.026
uts-0.026
 bleeding-0.026
w-0.026
ller-0.024
Token have
Token ID423
Feature activation+1.30e-02

Pos logit contributions
w+0.009
ich+0.008
 crying+0.008
llers+0.008
ller+0.008
Neg logit contributions
 agreement-0.007
 particular-0.007
 Everyday-0.007
rily-0.006
date-0.006
Token.
Token ID13
Feature activation+3.45e-02

Pos logit contributions
 agreement+0.029
 particular+0.026
 surely+0.024
 Everyday+0.024
ainer+0.023
Neg logit contributions
w-0.012
uts-0.011
 reading-0.010
 happened-0.010
icken-0.009
Token He
Token ID679
Feature activation+1.61e-02

Pos logit contributions
 agreement+0.060
date+0.059
rily+0.057
 surely+0.054
 host+0.053
Neg logit contributions
EEP-0.037
llers-0.036
irens-0.035
ller-0.034
 Blood-0.034
Token was
Token ID373
Feature activation+7.50e-03

Pos logit contributions
 agreement+0.035
date+0.031
ainer+0.030
 particular+0.029
rily+0.029
Neg logit contributions
w-0.016
 tri-0.013
ller-0.013
 screamed-0.013
 happened-0.013
Token very
Token ID845
Feature activation+1.13e-02

Pos logit contributions
 agreement+0.015
date+0.013
 particular+0.013
 basis+0.013
 offer+0.013
Neg logit contributions
 reading-0.008
 crying-0.008
 praying-0.007
ich-0.007
nam-0.007
Token loyal
Token ID9112
Feature activation+4.91e-03

Pos logit contributions
 agreement+0.022
date+0.021
 Everyday+0.020
 particular+0.019
 basis+0.019
Neg logit contributions
w-0.013
aged-0.011
ich-0.011
nam-0.011
 Blood-0.010
Token and
Token ID290
Feature activation+1.30e-02

Pos logit contributions
 particular+0.011
 agreement+0.011
 surely+0.010
date+0.010
 Everyday+0.010
Neg logit contributions
uts-0.004
aged-0.003
ich-0.003
w-0.003
 crying-0.003
Token always
Token ID1464
Feature activation-1.27e-02

Pos logit contributions
 agreement+0.024
 particular+0.021
date+0.020
 Everyday+0.019
 date+0.018
Neg logit contributions
w-0.019
aged-0.017
nam-0.016
ages-0.016
ller-0.016
Token helped
Token ID4193
Feature activation+1.54e-02

Pos logit contributions
w+0.019
aged+0.017
ich+0.016
ller+0.016
llers+0.016
Neg logit contributions
 agreement-0.021
 particular-0.020
 Everyday-0.019
date-0.019
 invite-0.017
Token his
Token ID465
Feature activation+2.03e-02

Pos logit contributions
 agreement+0.029
 particular+0.026
 Everyday+0.026
date+0.024
 surely+0.024
Neg logit contributions
w-0.019
aged-0.017
irens-0.016
ich-0.016
llers-0.016
Token friends
Token ID2460
Feature activation+8.79e-03

Pos logit contributions
 agreement+0.042
rant+0.040
 particular+0.039
 Everyday+0.039
iona+0.038
Neg logit contributions
ich-0.018
aged-0.017
w-0.016
ller-0.015
ages-0.015
Token ever
Token ID1683
Feature activation-1.46e-02

Pos logit contributions
w+0.005
ller+0.004
aged+0.004
ich+0.004
llers+0.004
Neg logit contributions
 agreement-0.008
 particular-0.007
date-0.007
 Everyday-0.007
 invite-0.006
Token day
Token ID1110
Feature activation-6.73e-03

Pos logit contributions
w+0.025
ich+0.024
ller+0.022
aged+0.022
irens+0.022
Neg logit contributions
 agreement-0.020
 particular-0.019
 Everyday-0.018
 invite-0.017
 fill-0.017
Token and
Token ID290
Feature activation-1.34e-02

Pos logit contributions
w+0.009
ich+0.008
ller+0.008
uts+0.008
aged+0.008
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.011
date-0.011
 invite-0.010
Token saw
Token ID2497
Feature activation-6.81e-03

Pos logit contributions
w+0.022
ich+0.021
ller+0.020
llers+0.019
irens+0.019
Neg logit contributions
 agreement-0.020
 particular-0.018
 Everyday-0.017
 invite-0.017
date-0.017
Token something
Token ID1223
Feature activation-3.67e-04

Pos logit contributions
w+0.010
ich+0.009
ller+0.008
aged+0.008
llers+0.008
Neg logit contributions
 agreement-0.012
 particular-0.011
 Everyday-0.010
date-0.010
 invite-0.010
Token new
Token ID649
Feature activation-1.93e-02

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
 flashing+0.000
 crying+0.000
Neg logit contributions
 agreement-0.001
 Everyday-0.001
date-0.001
 particular-0.001
 basis-0.001
Token and
Token ID290
Feature activation-2.22e-03

Pos logit contributions
w+0.024
ich+0.023
ller+0.022
uts+0.021
aged+0.020
Neg logit contributions
 agreement-0.039
 particular-0.034
 Everyday-0.034
date-0.033
 invite-0.032
Token amazing
Token ID4998
Feature activation-7.20e-03

Pos logit contributions
w+0.003
nam+0.003
ller+0.003
ich+0.003
aged+0.003
Neg logit contributions
 agreement-0.004
date-0.003
 particular-0.003
 Everyday-0.003
 invite-0.003
Token each
Token ID1123
Feature activation+3.65e-03

Pos logit contributions
w+0.009
ller+0.008
ich+0.008
aged+0.008
llers+0.008
Neg logit contributions
 agreement-0.014
 Everyday-0.012
 particular-0.012
date-0.012
 invite-0.011
Token time
Token ID640
Feature activation-1.42e-02

Pos logit contributions
 agreement+0.006
 Everyday+0.005
latable+0.005
date+0.005
 rightful+0.005
Neg logit contributions
w-0.006
 pressing-0.005
 happened-0.004
 crying-0.004
 touching-0.004
Token.
Token ID13
Feature activation-1.22e-02

Pos logit contributions
w+0.020
ich+0.018
ller+0.018
aged+0.017
llers+0.017
Neg logit contributions
 agreement-0.025
 particular-0.023
 Everyday-0.022
date-0.022
 invite-0.021
Token to
Token ID284
Feature activation+2.07e-03

Pos logit contributions
 agreement+0.033
date+0.032
 particular+0.032
 Everyday+0.030
 invite+0.029
Neg logit contributions
ller-0.016
w-0.016
uts-0.015
aged-0.015
llers-0.015
Token John
Token ID1757
Feature activation+2.17e-04

Pos logit contributions
 agreement+0.004
 particular+0.003
date+0.003
 Everyday+0.003
 invite+0.003
Neg logit contributions
w-0.002
ller-0.002
llers-0.002
aged-0.002
ich-0.002
Token.
Token ID13
Feature activation-6.00e-03

Pos logit contributions
 agreement+0.000
 particular+0.000
date+0.000
 Everyday+0.000
latable+0.000
Neg logit contributions
aged-0.000
 crying-0.000
ller-0.000
w-0.000
 reading-0.000
Token When
Token ID1649
Feature activation-9.15e-03

Pos logit contributions
w+0.011
ich+0.010
aged+0.009
ller+0.009
inches+0.009
Neg logit contributions
 agreement-0.009
 Everyday-0.008
 particular-0.008
date-0.007
 invite-0.007
Token they
Token ID484
Feature activation+5.80e-03

Pos logit contributions
w+0.013
ich+0.012
ller+0.011
aged+0.011
llers+0.011
Neg logit contributions
 agreement-0.016
 particular-0.015
 Everyday-0.014
date-0.014
 invite-0.014
Token heard
Token ID2982
Feature activation-8.39e-03

Pos logit contributions
 agreement+0.010
 particular+0.009
date+0.009
 Everyday+0.008
 offer+0.008
Neg logit contributions
w-0.009
ich-0.007
aged-0.007
ller-0.007
RR-0.007
Token him
Token ID683
Feature activation+2.15e-03

Pos logit contributions
w+0.012
ich+0.010
ller+0.010
aged+0.010
llers+0.010
Neg logit contributions
 agreement-0.015
 particular-0.014
 Everyday-0.013
date-0.013
 invite-0.013
Token they
Token ID484
Feature activation-3.36e-03

Pos logit contributions
 agreement+0.004
 particular+0.003
date+0.003
 Everyday+0.003
 invite+0.003
Neg logit contributions
w-0.003
ich-0.003
ller-0.003
aged-0.003
llers-0.003
Token cheered
Token ID34767
Feature activation-3.29e-03

Pos logit contributions
w+0.005
ich+0.004
ller+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.006
 particular-0.005
date-0.005
 Everyday-0.005
 offer-0.005
Token!
Token ID0
Feature activation-5.40e-03

Pos logit contributions
w+0.004
ller+0.004
ich+0.004
aged+0.004
llers+0.004
Neg logit contributions
 agreement-0.006
 particular-0.005
 Everyday-0.005
date-0.005
 invite-0.005
Token But
Token ID887
Feature activation+1.22e-03

Pos logit contributions
w+0.010
ich+0.009
aged+0.008
 elbows+0.008
inches+0.008
Neg logit contributions
 agreement-0.007
 Everyday-0.007
 particular-0.007
date-0.006
 invite-0.006
Token One
Token ID1881
Feature activation+3.26e-02

Pos logit contributions
 agreement+0.016
date+0.014
 particular+0.013
 invite+0.013
 offer+0.013
Neg logit contributions
w-0.016
 Blood-0.015
llers-0.015
ich-0.015
ller-0.015
Token day
Token ID1110
Feature activation-1.47e-02

Pos logit contributions
 agreement+0.038
 particular+0.031
 Everyday+0.030
date+0.030
 invite+0.029
Neg logit contributions
w-0.066
ich-0.061
ller-0.058
aged-0.058
llers-0.057
Token,
Token ID11
Feature activation+1.16e-02

Pos logit contributions
ich+0.030
ller+0.030
llers+0.028
w+0.028
irens+0.027
Neg logit contributions
 particular-0.016
 agreement-0.016
date-0.015
 Everyday-0.014
 fill-0.013
Token Daisy
Token ID40355
Feature activation+2.99e-03

Pos logit contributions
 agreement+0.019
 particular+0.017
 Everyday+0.016
date+0.016
 invite+0.016
Neg logit contributions
w-0.018
ich-0.016
ller-0.016
aged-0.015
llers-0.015
Token was
Token ID373
Feature activation-2.88e-03

Pos logit contributions
 agreement+0.006
 particular+0.005
date+0.005
 Everyday+0.005
 offer+0.005
Neg logit contributions
w-0.004
ich-0.003
aged-0.003
ller-0.003
llers-0.003
Token playing
Token ID2712
Feature activation-2.76e-03

Pos logit contributions
w+0.003
ller+0.003
ich+0.003
llers+0.003
RR+0.003
Neg logit contributions
 agreement-0.006
 particular-0.005
date-0.005
 invite-0.005
 Everyday-0.005
Token dolls
Token ID36062
Feature activation+1.15e-03

Pos logit contributions
w+0.004
ich+0.004
ller+0.004
aged+0.004
llers+0.003
Neg logit contributions
 agreement-0.005
 particular-0.004
 Everyday-0.004
date-0.004
 invite-0.004
Token with
Token ID351
Feature activation-2.39e-03

Pos logit contributions
 agreement+0.002
 particular+0.002
 Everyday+0.002
date+0.002
 invite+0.002
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
aged-0.002
llers-0.001
Token her
Token ID607
Feature activation+1.46e-03

Pos logit contributions
w+0.003
nam+0.002
irens+0.002
ich+0.002
llers+0.002
Neg logit contributions
 agreement-0.005
 particular-0.004
 Everyday-0.004
 offer-0.004
date-0.004
Token friend
Token ID1545
Feature activation-3.53e-04

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
date+0.003
 offer+0.003
Neg logit contributions
w-0.002
ich-0.001
ller-0.001
aged-0.001
llers-0.001
Token.
Token ID13
Feature activation+7.13e-03

Pos logit contributions
w+0.000
ich+0.000
aged+0.000
ller+0.000
llers+0.000
Neg logit contributions
 agreement-0.001
 particular-0.001
 Everyday-0.001
 offer-0.001
date-0.001
Token is
Token ID318
Feature activation-2.91e-03

Pos logit contributions
w+0.018
ich+0.016
ller+0.016
llers+0.015
aged+0.015
Neg logit contributions
 agreement-0.018
 particular-0.016
date-0.016
 Everyday-0.015
 invite-0.015
Token okay
Token ID8788
Feature activation-1.98e-02

Pos logit contributions
w+0.004
ich+0.003
ller+0.003
aged+0.003
llers+0.003
Neg logit contributions
 agreement-0.005
 particular-0.005
date-0.005
 Everyday-0.005
 invite-0.004
Token.
Token ID13
Feature activation+1.20e-02

Pos logit contributions
w+0.033
ich+0.030
ller+0.029
aged+0.028
llers+0.028
Neg logit contributions
 agreement-0.030
 particular-0.027
 Everyday-0.026
date-0.026
 invite-0.025
Token We
Token ID775
Feature activation-2.88e-04

Pos logit contributions
 agreement+0.022
 particular+0.020
date+0.020
 Everyday+0.018
 invite+0.018
Neg logit contributions
w-0.015
ller-0.014
ich-0.014
llers-0.013
EEP-0.013
Token can
Token ID460
Feature activation-4.46e-03

Pos logit contributions
w+0.001
aged+0.000
ich+0.000
EEP+0.000
AAAA+0.000
Neg logit contributions
 agreement-0.000
 particular-0.000
date-0.000
latable-0.000
 Everyday-0.000
Token set
Token ID900
Feature activation-1.69e-03

Pos logit contributions
w+0.008
ich+0.007
ller+0.007
llers+0.007
aged+0.007
Neg logit contributions
 agreement-0.006
 particular-0.005
date-0.005
 Everyday-0.005
 offer-0.004
Token our
Token ID674
Feature activation-1.91e-02

Pos logit contributions
w+0.002
ich+0.002
ller+0.002
llers+0.002
aged+0.002
Neg logit contributions
 agreement-0.003
 particular-0.003
date-0.003
 Everyday-0.003
 invite-0.003
Token blocks
Token ID7021
Feature activation+1.68e-03

Pos logit contributions
w+0.028
ich+0.025
ller+0.024
aged+0.024
llers+0.024
Neg logit contributions
 agreement-0.032
 particular-0.030
 Everyday-0.028
date-0.028
 invite-0.027
Token again
Token ID757
Feature activation-8.27e-03

Pos logit contributions
 agreement+0.003
 particular+0.003
 Everyday+0.003
date+0.003
 invite+0.002
Neg logit contributions
w-0.002
ich-0.002
ller-0.002
uts-0.002
aged-0.002
Token later
Token ID1568
Feature activation+1.48e-02

Pos logit contributions
w+0.014
ich+0.013
ller+0.012
llers+0.012
aged+0.012
Neg logit contributions
 agreement-0.012
 particular-0.011
date-0.010
 Everyday-0.010
 invite-0.010
Token."
Token ID526
Feature activation-8.53e-04

Pos logit contributions
 agreement+0.026
 particular+0.024
 Everyday+0.022
date+0.022
 invite+0.021
Neg logit contributions
w-0.021
ich-0.019
ller-0.019
llers-0.018
uts-0.018
Token we
Token ID356
Feature activation+1.41e-02

Pos logit contributions
 agreement+0.034
date+0.034
 Everyday+0.031
rene+0.030
 particular+0.030
Neg logit contributions
w-0.010
 accidents-0.009
EEP-0.009
ich-0.008
 bleeding-0.008
Token can
Token ID460
Feature activation+1.50e-02

Pos logit contributions
rily+0.027
 particular+0.026
 agreement+0.026
date+0.023
iona+0.023
Neg logit contributions
w-0.018
EEP-0.014
Crash-0.014
 broke-0.014
 fighting-0.013
Token play
Token ID711
Feature activation+1.94e-02

Pos logit contributions
 agreement+0.025
 particular+0.022
date+0.021
rily+0.021
 Everyday+0.020
Neg logit contributions
w-0.022
llers-0.020
ller-0.019
EEP-0.019
ich-0.019
Token with
Token ID351
Feature activation+8.41e-03

Pos logit contributions
 agreement+0.038
 particular+0.038
 Everyday+0.037
 Whenever+0.034
 surely+0.033
Neg logit contributions
w-0.021
ich-0.018
AAAA-0.018
RR-0.017
 bleeding-0.016
Token the
Token ID262
Feature activation+8.42e-03

Pos logit contributions
 agreement+0.016
 particular+0.015
 Everyday+0.014
date+0.014
 invite+0.013
Neg logit contributions
w-0.011
ich-0.010
ller-0.009
llers-0.009
inches-0.009
Token rain
Token ID6290
Feature activation-1.84e-03

Pos logit contributions
 agreement+0.015
 particular+0.014
 Everyday+0.014
date+0.013
rily+0.013
Neg logit contributions
w-0.011
ich-0.010
 bleeding-0.010
 crying-0.010
 elbows-0.009
Token later
Token ID1568
Feature activation+1.78e-02

Pos logit contributions
w+0.002
ich+0.002
uts+0.002
ller+0.002
llers+0.002
Neg logit contributions
 agreement-0.004
 particular-0.003
 Everyday-0.003
date-0.003
 offer-0.003
Token.
Token ID13
Feature activation+1.82e-02

Pos logit contributions
 agreement+0.034
 particular+0.030
 Everyday+0.028
date+0.028
 invite+0.027
Neg logit contributions
w-0.024
ich-0.020
ller-0.020
uts-0.020
EEP-0.019
Token When
Token ID1649
Feature activation+1.73e-02

Pos logit contributions
 agreement+0.033
 particular+0.032
date+0.031
 offer+0.028
 Everyday+0.027
Neg logit contributions
EEP-0.021
w-0.020
ller-0.018
RR-0.018
ich-0.018
Token it
Token ID340
Feature activation+8.73e-03

Pos logit contributions
 agreement+0.039
 particular+0.036
 Everyday+0.036
date+0.036
 offer+0.033
Neg logit contributions
w-0.016
ich-0.013
EEP-0.011
llers-0.011
ller-0.011
Token stops
Token ID9911
Feature activation-2.06e-02

Pos logit contributions
 agreement+0.017
 particular+0.015
date+0.014
 Everyday+0.014
 offer+0.013
Neg logit contributions
w-0.012
ller-0.009
ich-0.009
aged-0.009
llers-0.009
Token that
Token ID326
Feature activation-2.93e-02

Pos logit contributions
 agreement+0.014
 particular+0.013
 Everyday+0.013
date+0.012
 invite+0.012
Neg logit contributions
w-0.011
ich-0.010
ller-0.010
aged-0.009
irens-0.009
Token night
Token ID1755
Feature activation-3.32e-02

Pos logit contributions
 elbows+0.073
w+0.073
ich+0.072
nam+0.071
inches+0.071
Neg logit contributions
 particular-0.020
 Everyday-0.017
 agreement-0.016
 invite-0.012
 end-0.012
Token,
Token ID11
Feature activation+2.79e-03

Pos logit contributions
w+0.065
nam+0.063
ich+0.062
 elbows+0.062
ller+0.058
Neg logit contributions
 particular-0.036
date-0.035
 fill-0.034
 of-0.034
 invite-0.034
Token Tim
Token ID5045
Feature activation-8.86e-03

Pos logit contributions
 agreement+0.005
 particular+0.005
 Everyday+0.005
date+0.004
 invite+0.004
Neg logit contributions
w-0.004
ich-0.003
ller-0.003
aged-0.003
llers-0.003
Tokenmy
Token ID1820
Feature activation-1.36e-02

Pos logit contributions
w+0.010
ich+0.008
aged+0.007
ller+0.007
 bleeding+0.007
Neg logit contributions
 agreement-0.019
 particular-0.018
date-0.017
 offer-0.017
 Everyday-0.017
Token started
Token ID2067
Feature activation-3.73e-02

Pos logit contributions
w+0.017
ich+0.014
aged+0.014
ller+0.014
llers+0.013
Neg logit contributions
 agreement-0.027
 particular-0.025
date-0.024
 Everyday-0.023
 offer-0.022
Token to
Token ID284
Feature activation-2.54e-02

Pos logit contributions
w+0.067
ich+0.059
RR+0.057
ller+0.055
AAAA+0.054
Neg logit contributions
 agreement-0.053
 particular-0.049
 Everyday-0.049
 invite-0.048
 fill-0.047
Token cry
Token ID3960
Feature activation-7.34e-03

Pos logit contributions
w+0.038
ich+0.034
ller+0.033
aged+0.033
llers+0.033
Neg logit contributions
 agreement-0.040
 particular-0.037
date-0.036
 Everyday-0.035
 invite-0.033
Token.
Token ID13
Feature activation+1.80e-03

Pos logit contributions
w+0.013
ich+0.012
ller+0.011
aged+0.011
llers+0.011
Neg logit contributions
 agreement-0.010
 particular-0.009
 Everyday-0.009
date-0.009
 invite-0.008
Token His
Token ID2399
Feature activation-4.24e-03

Pos logit contributions
 agreement+0.003
 particular+0.002
 Everyday+0.002
date+0.002
 invite+0.002
Neg logit contributions
w-0.003
ich-0.003
ller-0.003
aged-0.003
llers-0.003
Token knee
Token ID10329
Feature activation-9.08e-03

Pos logit contributions
w+0.006
ich+0.006
aged+0.005
ller+0.005
llers+0.005
Neg logit contributions
 agreement-0.007
 particular-0.007
date-0.007
 Everyday-0.006
 invite-0.006
Token
Token ID198
Feature activation-1.58e-02

Pos logit contributions
w+0.029
ich+0.026
ller+0.025
llers+0.025
nam+0.024
Neg logit contributions
 agreement-0.021
 Everyday-0.020
 particular-0.020
 Whenever-0.017
 offer-0.017
Token
Token ID198
Feature activation-6.13e-03

Pos logit contributions
w+0.024
ich+0.022
ller+0.020
llers+0.020
aged+0.020
Neg logit contributions
 agreement-0.026
 particular-0.025
 Everyday-0.024
 invite-0.022
 offer-0.022
Token"
Token ID1
Feature activation-1.65e-02

Pos logit contributions
w+0.008
ich+0.008
ller+0.008
llers+0.007
aged+0.007
Neg logit contributions
 agreement-0.011
 particular-0.010
 Everyday-0.010
date-0.009
 invite-0.009
TokenHey
Token ID10814
Feature activation-3.81e-02

Pos logit contributions
w+0.024
ller+0.024
ich+0.024
llers+0.023
aged+0.023
Neg logit contributions
 agreement-0.026
 particular-0.024
date-0.022
 Everyday-0.022
 invite-0.021
Token,
Token ID11
Feature activation-2.12e-02

Pos logit contributions
w+0.064
llers+0.061
ller+0.061
irens+0.061
aged+0.060
Neg logit contributions
 agreement-0.052
 particular-0.047
 invite-0.044
date-0.043
 offer-0.043
Token that
Token ID326
Feature activation-3.55e-02

Pos logit contributions
w+0.029
ich+0.026
ller+0.024
aged+0.024
llers+0.024
Neg logit contributions
 agreement-0.038
 particular-0.036
 Everyday-0.034
date-0.034
 invite-0.033
Token's
Token ID338
Feature activation-1.44e-02

Pos logit contributions
w+0.054
nam+0.052
ich+0.052
irens+0.052
aged+0.052
Neg logit contributions
 agreement-0.049
 particular-0.046
 invite-0.045
 Everyday-0.045
date-0.045
Token mine
Token ID6164
Feature activation-3.61e-02

Pos logit contributions
w+0.022
aged+0.021
ich+0.021
irens+0.020
ller+0.020
Neg logit contributions
 agreement-0.021
 particular-0.020
 invite-0.019
 fill-0.018
date-0.018
Token!"
Token ID2474
Feature activation-1.97e-02

Pos logit contributions
w+0.055
ich+0.050
aged+0.048
ller+0.048
llers+0.048
Neg logit contributions
 agreement-0.056
 particular-0.052
 Everyday-0.049
date-0.049
 invite-0.049
Token Sara
Token ID24799
Feature activation-1.05e-02

Pos logit contributions
w+0.036
ich+0.034
ller+0.032
llers+0.032
aged+0.031
Neg logit contributions
 agreement-0.026
 particular-0.025
 Everyday-0.024
 invite-0.022
 offer-0.022
Token shouts
Token ID34757
Feature activation-1.82e-02

Pos logit contributions
w+0.015
ich+0.014
nam+0.013
ller+0.012
aged+0.012
Neg logit contributions
 agreement-0.018
 Everyday-0.017
 invite-0.017
 offer-0.016
 fill-0.016
Token is
Token ID318
Feature activation-8.34e-03

Pos logit contributions
w+0.028
ich+0.025
ller+0.024
aged+0.024
llers+0.023
Neg logit contributions
 agreement-0.033
 particular-0.031
date-0.029
 Everyday-0.028
 invite-0.027
Token electric
Token ID5186
Feature activation-6.52e-04

Pos logit contributions
w+0.008
 bleeding+0.008
ich+0.007
aged+0.007
nam+0.007
Neg logit contributions
 agreement-0.018
date-0.018
 particular-0.017
 Everyday-0.016
 basis-0.016
Token.
Token ID13
Feature activation-7.70e-03

Pos logit contributions
w+0.001
aged+0.001
 bleeding+0.001
ller+0.001
irens+0.001
Neg logit contributions
 agreement-0.001
 particular-0.001
date-0.001
 invite-0.001
 Everyday-0.001
Token She
Token ID1375
Feature activation-1.99e-02

Pos logit contributions
w+0.014
ich+0.012
ller+0.012
llers+0.012
aged+0.012
Neg logit contributions
 agreement-0.011
 particular-0.010
 Everyday-0.010
date-0.009
 invite-0.009
Token says
Token ID1139
Feature activation-2.59e-04

Pos logit contributions
w+0.020
ller+0.017
ich+0.017
inches+0.016
aged+0.016
Neg logit contributions
 agreement-0.042
 particular-0.040
date-0.040
 Everyday-0.037
 invite-0.036
Token that
Token ID326
Feature activation-3.68e-02

Pos logit contributions
w+0.000
ich+0.000
ller+0.000
llers+0.000
RR+0.000
Neg logit contributions
 agreement-0.000
 particular-0.000
 Everyday-0.000
 invite-0.000
 offer-0.000
Token Sara
Token ID24799
Feature activation+1.46e-04

Pos logit contributions
w+0.058
ich+0.053
ller+0.052
llers+0.051
aged+0.051
Neg logit contributions
 agreement-0.054
 particular-0.052
 Everyday-0.050
 invite-0.048
 offer-0.047
Token can
Token ID460
Feature activation-5.40e-03

Pos logit contributions
 agreement+0.000
 particular+0.000
date+0.000
 Everyday+0.000
 invite+0.000
Neg logit contributions
w-0.000
ich-0.000
aged-0.000
ller-0.000
inches-0.000
Token use
Token ID779
Feature activation-1.18e-02

Pos logit contributions
w+0.009
ller+0.008
ich+0.008
llers+0.007
uts+0.007
Neg logit contributions
 agreement-0.009
 particular-0.007
date-0.007
 Everyday-0.007
rily-0.006
Token her
Token ID607
Feature activation-9.64e-03

Pos logit contributions
w+0.015
ller+0.012
ich+0.012
aged+0.012
llers+0.012
Neg logit contributions
 agreement-0.023
 particular-0.021
date-0.021
 Everyday-0.020
 invite-0.020
Token regular
Token ID3218
Feature activation-2.26e-02

Pos logit contributions
w+0.013
ich+0.011
aged+0.011
ller+0.011
inches+0.010
Neg logit contributions
 agreement-0.018
 particular-0.017
date-0.016
 Everyday-0.016
 invite-0.015
Token Then
Token ID3244
Feature activation-7.43e-03

Pos logit contributions
w+0.023
ich+0.021
 elbows+0.019
rob+0.018
RR+0.018
Neg logit contributions
 Everyday-0.012
 agreement-0.011
 particular-0.010
 Whenever-0.010
 Every-0.009
Token,
Token ID11
Feature activation+3.29e-04

Pos logit contributions
w+0.012
ich+0.012
ller+0.011
RR+0.011
llers+0.011
Neg logit contributions
 agreement-0.011
 particular-0.010
 Everyday-0.010
 invite-0.010
date-0.010
Token they
Token ID484
Feature activation-6.33e-03

Pos logit contributions
 agreement+0.001
 particular+0.001
 Everyday+0.001
date+0.001
 offer+0.001
Neg logit contributions
w-0.000
ich-0.000
aged-0.000
ller-0.000
nam-0.000
Token all
Token ID477
Feature activation-1.33e-02

Pos logit contributions
w+0.009
aged+0.008
ich+0.008
nam+0.007
 bleeding+0.007
Neg logit contributions
 agreement-0.011
 particular-0.011
 Everyday-0.009
date-0.009
 offer-0.009
Token went
Token ID1816
Feature activation-9.31e-03

Pos logit contributions
w+0.020
aged+0.018
ller+0.016
 bleeding+0.016
ich+0.016
Neg logit contributions
 agreement-0.021
 particular-0.020
date-0.019
 Everyday-0.018
 offer-0.018
Token back
Token ID736
Feature activation-3.49e-02

Pos logit contributions
w+0.012
ich+0.012
aged+0.011
irens+0.011
ller+0.011
Neg logit contributions
 agreement-0.016
 particular-0.016
 Everyday-0.015
date-0.014
 offer-0.014
Token to
Token ID284
Feature activation-9.87e-03

Pos logit contributions
w+0.074
RR+0.067
ller+0.065
ich+0.065
llers+0.064
Neg logit contributions
 invite-0.035
 fill-0.034
 agreement-0.033
 particular-0.032
date-0.032
Token the
Token ID262
Feature activation-8.04e-03

Pos logit contributions
w+0.015
ich+0.013
ller+0.012
aged+0.012
llers+0.012
Neg logit contributions
 agreement-0.017
 particular-0.016
 Everyday-0.015
date-0.015
 invite-0.015
Token party
Token ID2151
Feature activation-1.43e-02

Pos logit contributions
w+0.014
ich+0.014
 crying+0.013
 reading+0.013
 bleeding+0.013
Neg logit contributions
 agreement-0.010
 particular-0.010
 Everyday-0.009
iona-0.008
 Whenever-0.008
Token and
Token ID290
Feature activation-1.79e-02

Pos logit contributions
w+0.020
ich+0.019
aged+0.018
ller+0.018
llers+0.018
Neg logit contributions
 agreement-0.024
 particular-0.023
 Everyday-0.022
date-0.020
 offer-0.020
Token had
Token ID550
Feature activation-1.94e-02

Pos logit contributions
w+0.027
ich+0.024
aged+0.024
ller+0.024
llers+0.023
Neg logit contributions
 agreement-0.029
 particular-0.027
 Everyday-0.025
date-0.025
 offer-0.024
Token every
Token ID790
Feature activation+2.38e-02

Pos logit contributions
 agreement+0.007
 particular+0.006
latable+0.006
date+0.006
 invite+0.005
Neg logit contributions
 reading-0.004
ich-0.004
ller-0.004
w-0.004
aged-0.004
Token day
Token ID1110
Feature activation-1.09e-03

Pos logit contributions
latable+0.040
 agreement+0.037
ola+0.036
 invite+0.035
cloth+0.035
Neg logit contributions
w-0.027
 reading-0.026
 crying-0.025
 praying-0.025
ller-0.022
Token to
Token ID284
Feature activation+2.06e-02

Pos logit contributions
w+0.001
ich+0.001
aged+0.001
uts+0.001
ller+0.001
Neg logit contributions
 agreement-0.002
 particular-0.002
date-0.002
 Everyday-0.002
 invite-0.002
Token supply
Token ID5127
Feature activation+2.45e-03

Pos logit contributions
 agreement+0.027
date+0.026
 particular+0.024
latable+0.024
 Everyday+0.024
Neg logit contributions
w-0.029
 reading-0.029
 praying-0.028
ller-0.027
 touching-0.027
Token the
Token ID262
Feature activation-2.58e-02

Pos logit contributions
 agreement+0.004
 particular+0.004
date+0.004
 Everyday+0.004
 invite+0.004
Neg logit contributions
w-0.003
aged-0.003
ich-0.003
ller-0.003
irens-0.003
Token av
Token ID1196
Feature activation-3.63e-02

Pos logit contributions
w+0.039
ich+0.036
ller+0.035
aged+0.034
llers+0.034
Neg logit contributions
 agreement-0.041
 particular-0.038
date-0.036
 Everyday-0.036
 invite-0.035
Tokenoc
Token ID420
Feature activation-3.37e-02

Pos logit contributions
w+0.064
ich+0.059
 elbows+0.058
aged+0.057
inches+0.057
Neg logit contributions
 agreement-0.048
 particular-0.044
 Everyday-0.042
date-0.041
 offer-0.039
Tokenados
Token ID22484
Feature activation-4.17e-03

Pos logit contributions
w+0.032
ich+0.025
 elbows+0.025
AAAA+0.025
ller+0.024
Neg logit contributions
 agreement-0.071
date-0.065
 Everyday-0.065
 particular-0.064
 offer-0.064
Token.
Token ID13
Feature activation+4.11e-03

Pos logit contributions
w+0.005
uts+0.005
ich+0.005
nam+0.004
aged+0.004
Neg logit contributions
 agreement-0.008
 particular-0.007
date-0.007
 Everyday-0.007
 invite-0.006
Token And
Token ID843
Feature activation-4.63e-03

Pos logit contributions
 agreement+0.007
 particular+0.006
date+0.006
 invite+0.006
 fill+0.006
Neg logit contributions
ller-0.005
llers-0.005
w-0.005
 Blood-0.005
uts-0.005
Token the
Token ID262
Feature activation-6.64e-03

Pos logit contributions
w+0.006
aged+0.005
ller+0.005
ich+0.005
nam+0.005
Neg logit contributions
 agreement-0.008
date-0.008
 particular-0.007
latable-0.007
 invite-0.007