TOP ACTIVATIONS
MAX = 0.082

have some ?" S ue said , " Yes
Tim my quickly yelled " ST OP !" and ran to
them back ." S ue looked at the rake
my car ." S ue is angry and hurt
smiles . <|endoftext|> Tim and Kim were twins who liked to
did it . We made a new tower . A better
The voice said , " cong ratulations , you can have
deep breath and shouted " ST OP !"
need help !" S ue is scared . She
wander . J ill started to feel jealous
. The sign said " ST OP ". "
said sadly . S ally tried to cheer him
she heard a loud , gru ff voice calling her name
Tom and said in a gru ff voice , " Get
a dessert ?" J ill was delighted . "
bad dog ." S ue hugged Tom and cried
. They thought it was a good day at the park
were happy . The conflict was over , and they played
learned their lesson . They never played with the ladder and
her room . S ue hears Tom 's me

MIN ACTIVATIONS
MIN = -0.107

, please , please ?" they asked . Mom
is our rad ish ?" they asked . " Who took
we go , dad ?" they asked . "
? Can we go ?" they asked . Mom
the park , mom ?" they asked . " Please ,
ppy and one called Cotton . They lived in a green
go to the zoo ?" they said . Mom
a balloon , mom ?" they asked . "
lot ion . Anna liked the bottle so her mom bought
some cocoa , please ?" they asked their mom .
some gum , please ?" they asked . The
ice cream , please ?" they said . The
can we have spaghetti ?" they asked . "
a big pig . The big pig had a penny .
we play with you ?" they asked . "
!" He gave her all the sa us ages and she
mom my started to cut the l imes . Lily was
help ! She picked up the bag , and it was
a cookie , mom ?" they asked . "
She bought him a pair of new shoes . They were

INTERVAL 0.035 - 0.082
CONTAINS 1.482%

end . <|endoftext|> Once upon a time there was a bad
how to fight . They fought for a little while ,
end . <|endoftext|> Once upon a time there were two little
was like a dream come true . <|endoftext|> Once , there
outside and run around in the park . One day ,

INTERVAL -0.012 - 0.035
CONTAINS 77.026%

and grateful that she hugged her mom my tight . From
smiled and said , " I 'm good ! What brings
. She thought he might fall and hurt himself . She
that he was a sailor and sailed around in his boat
smiled and nodded . Together , the woman

INTERVAL -0.060 - -0.012
CONTAINS 21.392%

this is my plate . I like flowers more than animals
the bunny felt sad . Its friends came to comfort it
to it . Lily goes first and slides down fast .
pup , and he was very unique . He liked to
. At home , M addy 's dad had

INTERVAL -0.107 - -0.060
CONTAINS 0.099%

. But the b unn ies didn 't wake up
. He flew for a long time until he found a
Lily tried to put on her sweater . It was a
the house . Every day , Fl uffy would jump and
a basket . He placed the basket of l imes at
Token have
Token ID423
Feature activation+2.40e-02

Pos logit contributions
Acc+0.044
 Acc+0.044
rou+0.037
 collapse+0.037
tr+0.036
Neg logit contributions
 pouch-0.032
 couple-0.032
 goggles-0.031
 banks-0.031
 plunge-0.030
Token some
Token ID617
Feature activation-4.03e-03

Pos logit contributions
 Acc+0.054
Acc+0.052
rou+0.048
tr+0.047
 collapse+0.046
Neg logit contributions
 couple-0.026
 goggles-0.026
 plunge-0.024
 kernels-0.024
 pouch-0.024
Token?"
Token ID1701
Feature activation+1.08e-02

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.008
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
Token
Token ID198
Feature activation+4.19e-02

Pos logit contributions
 Acc+0.022
Acc+0.022
rou+0.020
 collapse+0.019
tr+0.019
Neg logit contributions
 couple-0.013
 pouch-0.013
gers-0.012
 goggles-0.012
 banks-0.012
Token
Token ID198
Feature activation+5.20e-02

Pos logit contributions
 Acc+0.109
Acc+0.105
rou+0.099
tr+0.097
 collapse+0.094
Neg logit contributions
 couple-0.024
 pouch-0.022
ittens-0.019
 Holly-0.018
gers-0.018
TokenS
Token ID50
Feature activation+8.23e-02

Pos logit contributions
 Acc+0.126
rou+0.111
Acc+0.110
ccoli+0.101
arter+0.099
Neg logit contributions
 couple-0.055
gers-0.044
 pouch-0.042
 middle-0.039
 villagers-0.035
Tokenue
Token ID518
Feature activation+2.40e-02

Pos logit contributions
 Acc+0.154
Acc+0.140
rou+0.113
 collapse+0.107
aling+0.107
Neg logit contributions
owers-0.117
 bag-0.116
re-0.109
ittens-0.106
ve-0.104
Token said
Token ID531
Feature activation+1.03e-02

Pos logit contributions
 Acc+0.055
Acc+0.051
 collapse+0.048
rou+0.046
 unavoidable+0.045
Neg logit contributions
 couple-0.025
 pouch-0.024
 therapist-0.024
 bag-0.024
 pair-0.023
Token,
Token ID11
Feature activation+4.08e-03

Pos logit contributions
 Acc+0.022
Acc+0.021
 collapse+0.019
tr+0.019
rou+0.019
Neg logit contributions
 couple-0.012
 pouch-0.011
 plunge-0.011
gers-0.011
 goggles-0.011
Token "
Token ID366
Feature activation-1.19e-02

Pos logit contributions
 Acc+0.009
Acc+0.009
 collapse+0.008
rou+0.008
 solved+0.007
Neg logit contributions
 couple-0.005
 pouch-0.004
 goggles-0.004
gers-0.004
 therapist-0.004
TokenYes
Token ID5297
Feature activation+1.05e-02

Pos logit contributions
 pouch+0.021
 couple+0.021
 goggles+0.020
gers+0.020
 banks+0.020
Neg logit contributions
 Acc-0.018
Acc-0.018
 collapse-0.014
rou-0.014
tr-0.013
Token Tim
Token ID5045
Feature activation-8.01e-03

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.004
Acc-0.004
 collapse-0.003
rou-0.003
tr-0.003
Tokenmy
Token ID1820
Feature activation-2.83e-02

Pos logit contributions
 couple+0.020
 pouch+0.020
gers+0.019
 goggles+0.019
 plunge+0.019
Neg logit contributions
 Acc-0.007
Acc-0.006
 collapse-0.004
rou-0.004
tr-0.004
Token quickly
Token ID2952
Feature activation+2.20e-02

Pos logit contributions
gers+0.048
 goggles+0.047
ippers+0.047
 couple+0.047
 plunge+0.046
Neg logit contributions
 Acc-0.044
Acc-0.043
 solved-0.037
rou-0.036
 over-0.036
Token yelled
Token ID22760
Feature activation+2.54e-02

Pos logit contributions
 Acc+0.047
Acc+0.045
 collapse+0.038
tr+0.038
rou+0.038
Neg logit contributions
 sleeves-0.027
 couple-0.026
 pouch-0.026
 banks-0.025
 legs-0.025
Token "
Token ID366
Feature activation+6.29e-03

Pos logit contributions
 Acc+0.052
Acc+0.051
 collapse+0.046
tr+0.044
ris+0.043
Neg logit contributions
 legs-0.029
 goggles-0.029
 Holly-0.029
 shoes-0.028
 couple-0.028
TokenST
Token ID2257
Feature activation+7.87e-02

Pos logit contributions
 Acc+0.011
Acc+0.011
 collapse+0.009
rou+0.009
 solved+0.009
Neg logit contributions
 couple-0.010
 pouch-0.009
gers-0.009
 plunge-0.009
 goggles-0.009
TokenOP
Token ID3185
Feature activation-1.63e-03

Pos logit contributions
Acc+0.115
 Acc+0.106
 clearly+0.081
 cal+0.081
 over+0.081
Neg logit contributions
ippers-0.136
ittens-0.133
 pouch-0.131
 sleeves-0.128
aces-0.126
Token!"
Token ID2474
Feature activation-1.07e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.003
Acc-0.002
 collapse-0.002
rou-0.002
tr-0.002
Token and
Token ID290
Feature activation-5.28e-03

Pos logit contributions
 pouch+0.016
 couple+0.016
gers+0.016
 goggles+0.015
 plunge+0.015
Neg logit contributions
 Acc-0.020
Acc-0.018
 collapse-0.016
tr-0.015
rou-0.015
Token ran
Token ID4966
Feature activation+1.84e-02

Pos logit contributions
 pouch+0.008
 couple+0.008
gers+0.008
 goggles+0.008
 plunge+0.007
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.008
tr-0.008
Token to
Token ID284
Feature activation+1.90e-02

Pos logit contributions
 Acc+0.037
Acc+0.035
rou+0.030
tr+0.028
 collapse+0.028
Neg logit contributions
gers-0.026
 couple-0.026
 pouch-0.025
 goggles-0.025
 banks-0.025
Token them
Token ID606
Feature activation+3.38e-03

Pos logit contributions
 Acc+0.059
Acc+0.058
 solved+0.049
tr+0.049
 collapse+0.048
Neg logit contributions
 goggles-0.039
gers-0.038
 banks-0.038
 kernels-0.038
 pills-0.038
Token back
Token ID736
Feature activation+2.53e-02

Pos logit contributions
 Acc+0.006
Acc+0.006
rou+0.005
 collapse+0.005
tr+0.005
Neg logit contributions
 couple-0.005
 pouch-0.005
gers-0.005
 goggles-0.005
 banks-0.005
Token."
Token ID526
Feature activation+1.38e-03

Pos logit contributions
 Acc+0.053
Acc+0.052
 collapse+0.042
rou+0.042
 solved+0.042
Neg logit contributions
 banks-0.032
 pouch-0.031
 couple-0.031
gers-0.031
 kernels-0.030
Token
Token ID198
Feature activation+2.55e-02

Pos logit contributions
 Acc+0.003
Acc+0.003
rou+0.002
 collapse+0.002
tr+0.002
Neg logit contributions
 couple-0.002
 pouch-0.002
gers-0.002
 goggles-0.002
 plunge-0.002
Token
Token ID198
Feature activation+2.97e-02

Pos logit contributions
 Acc+0.059
Acc+0.059
rou+0.052
tr+0.052
 solved+0.052
Neg logit contributions
 pouch-0.022
 couple-0.022
gers-0.020
ippers-0.020
 goggles-0.020
TokenS
Token ID50
Feature activation+7.70e-02

Pos logit contributions
 Acc+0.064
Acc+0.061
rou+0.055
tr+0.053
 collapse+0.051
Neg logit contributions
 couple-0.038
gers-0.036
 pouch-0.034
 legs-0.032
 goggles-0.032
Tokenue
Token ID518
Feature activation+3.16e-02

Pos logit contributions
 Acc+0.139
Acc+0.126
rou+0.098
 collapse+0.097
 over+0.096
Neg logit contributions
owers-0.115
 bag-0.112
re-0.111
gers-0.104
ittens-0.103
Token looked
Token ID3114
Feature activation+4.33e-03

Pos logit contributions
 Acc+0.071
Acc+0.066
 collapse+0.061
 unavoidable+0.060
rou+0.059
Neg logit contributions
 couple-0.032
 legs-0.032
 sleeves-0.031
 bag-0.031
 pair-0.031
Token at
Token ID379
Feature activation+3.31e-02

Pos logit contributions
 Acc+0.008
Acc+0.007
 collapse+0.006
tr+0.006
rou+0.006
Neg logit contributions
 pouch-0.007
 couple-0.007
gers-0.007
 goggles-0.007
 banks-0.006
Token the
Token ID262
Feature activation-1.03e-02

Pos logit contributions
 Acc+0.058
Acc+0.053
tr+0.049
rou+0.046
 solved+0.046
Neg logit contributions
 shoes-0.053
 goggles-0.052
 couple-0.051
 plunge-0.050
 bag-0.050
Token rake
Token ID47885
Feature activation+8.72e-03

Pos logit contributions
 couple+0.013
 pouch+0.013
gers+0.013
 goggles+0.012
 plunge+0.012
Neg logit contributions
 Acc-0.021
Acc-0.020
 collapse-0.018
rou-0.017
tr-0.017
Token my
Token ID616
Feature activation-1.77e-02

Pos logit contributions
 pouch+0.001
 couple+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token car
Token ID1097
Feature activation+5.34e-03

Pos logit contributions
 pouch+0.023
 couple+0.023
 plunge+0.023
gers+0.023
 goggles+0.023
Neg logit contributions
 Acc-0.033
Acc-0.031
 collapse-0.027
rou-0.027
tr-0.027
Token."
Token ID526
Feature activation+1.02e-02

Pos logit contributions
 Acc+0.010
Acc+0.009
 collapse+0.008
rou+0.007
 solved+0.007
Neg logit contributions
 pouch-0.008
 couple-0.008
gers-0.008
 goggles-0.008
 banks-0.008
Token
Token ID198
Feature activation+2.36e-02

Pos logit contributions
 Acc+0.021
Acc+0.020
rou+0.018
 collapse+0.017
tr+0.017
Neg logit contributions
 couple-0.013
 pouch-0.012
gers-0.012
 goggles-0.012
 banks-0.011
Token
Token ID198
Feature activation+3.34e-02

Pos logit contributions
 Acc+0.053
Acc+0.051
tr+0.046
 collapse+0.046
rou+0.045
Neg logit contributions
 couple-0.022
 pouch-0.022
gers-0.021
 Holly-0.020
 banks-0.020
TokenS
Token ID50
Feature activation+7.63e-02

Pos logit contributions
 Acc+0.069
Acc+0.060
rou+0.055
tr+0.053
ccoli+0.052
Neg logit contributions
gers-0.043
 couple-0.043
 pouch-0.040
 banks-0.039
 pair-0.038
Tokenue
Token ID518
Feature activation+2.27e-02

Pos logit contributions
 Acc+0.137
Acc+0.128
 collapse+0.101
 false+0.100
 over+0.099
Neg logit contributions
owers-0.108
ittens-0.103
re-0.102
ve-0.101
 bag-0.098
Token is
Token ID318
Feature activation-7.34e-04

Pos logit contributions
 Acc+0.045
Acc+0.043
 collapse+0.037
 unavoidable+0.036
sm+0.036
Neg logit contributions
 pouch-0.030
 goggles-0.030
gers-0.029
 banks-0.029
 bag-0.029
Token angry
Token ID7954
Feature activation+1.14e-02

Pos logit contributions
 pouch+0.001
 couple+0.001
 goggles+0.001
gers+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.002
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token and
Token ID290
Feature activation-1.17e-02

Pos logit contributions
 Acc+0.022
Acc+0.022
 collapse+0.019
tr+0.018
 solved+0.018
Neg logit contributions
gers-0.015
 couple-0.015
 pouch-0.015
 legs-0.014
 goggles-0.014
Token hurt
Token ID5938
Feature activation+1.37e-02

Pos logit contributions
 couple+0.016
 pouch+0.016
gers+0.015
 goggles+0.015
 plunge+0.015
Neg logit contributions
 Acc-0.023
Acc-0.022
 collapse-0.019
rou-0.019
 solved-0.018
Token smiles
Token ID21845
Feature activation+2.06e-02

Pos logit contributions
 Acc+0.069
Acc+0.069
 collapse+0.060
 unavoidable+0.059
tr+0.057
Neg logit contributions
 pouch-0.042
 bag-0.040
 goggles-0.040
 Holly-0.039
 donkey-0.038
Token.
Token ID13
Feature activation+1.14e-02

Pos logit contributions
 Acc+0.044
Acc+0.042
 collapse+0.036
rou+0.035
tr+0.035
Neg logit contributions
 couple-0.025
 pouch-0.025
gers-0.024
 legs-0.024
 goggles-0.023
Token<|endoftext|>
Token ID50256
Feature activation-2.80e-02

Pos logit contributions
 Acc+0.024
Acc+0.024
 collapse+0.022
rou+0.021
 solved+0.021
Neg logit contributions
 pouch-0.012
 couple-0.012
 goggles-0.012
gers-0.012
 Holly-0.011
TokenTim
Token ID14967
Feature activation+5.30e-02

Pos logit contributions
 couple+0.042
 pouch+0.041
 banks+0.040
 goggles+0.040
 sleeves+0.040
Neg logit contributions
 Acc-0.049
Acc-0.046
sm-0.040
 over-0.039
 collapse-0.039
Token and
Token ID290
Feature activation+1.69e-02

Pos logit contributions
 Acc+0.113
Acc+0.111
 collapse+0.108
 unavoidable+0.100
tr+0.098
Neg logit contributions
 bag-0.058
 pouch-0.057
gers-0.055
 goggles-0.054
 socks-0.054
Token Kim
Token ID6502
Feature activation+7.59e-02

Pos logit contributions
 Acc+0.030
Acc+0.030
rou+0.027
 collapse+0.026
tr+0.025
Neg logit contributions
 goggles-0.025
 couple-0.025
 pouch-0.024
gers-0.024
 pair-0.023
Token were
Token ID547
Feature activation-8.99e-03

Pos logit contributions
 Acc+0.174
Acc+0.171
rou+0.162
tr+0.160
 collapse+0.159
Neg logit contributions
 shoes-0.067
 banks-0.067
 bag-0.065
 bank-0.064
 goggles-0.061
Token twins
Token ID20345
Feature activation-3.39e-04

Pos logit contributions
 pouch+0.013
 couple+0.013
gers+0.013
 goggles+0.013
 banks+0.012
Neg logit contributions
 Acc-0.017
Acc-0.016
rou-0.014
 collapse-0.014
 solved-0.013
Token who
Token ID508
Feature activation+6.86e-03

Pos logit contributions
 pouch+0.001
 couple+0.001
gers+0.001
 goggles+0.001
 banks+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.000
rou-0.000
tr-0.000
Token liked
Token ID8288
Feature activation+1.78e-02

Pos logit contributions
 Acc+0.013
Acc+0.013
 collapse+0.011
rou+0.011
tr+0.011
Neg logit contributions
 couple-0.009
gers-0.009
 pouch-0.009
 goggles-0.009
 therapist-0.009
Token to
Token ID284
Feature activation+1.81e-02

Pos logit contributions
 Acc+0.031
Acc+0.029
 collapse+0.026
rou+0.025
tr+0.025
Neg logit contributions
 couple-0.028
 goggles-0.028
 kernels-0.026
 pair-0.026
 potatoes-0.026
Token did
Token ID750
Feature activation+2.55e-02

Pos logit contributions
 Acc+0.022
Acc+0.021
 collapse+0.019
rou+0.018
tr+0.018
Neg logit contributions
 pouch-0.013
 goggles-0.013
gers-0.013
 couple-0.013
 plunge-0.012
Token it
Token ID340
Feature activation+3.63e-02

Pos logit contributions
 Acc+0.054
Acc+0.052
tr+0.045
rou+0.044
sm+0.043
Neg logit contributions
 goggles-0.029
gers-0.029
 legs-0.029
 shoes-0.028
 couple-0.028
Token.
Token ID13
Feature activation+2.84e-02

Pos logit contributions
 Acc+0.065
Acc+0.061
rou+0.047
sm+0.045
 unavoidable+0.044
Neg logit contributions
gers-0.056
 couple-0.055
 pair-0.053
ner-0.053
 plunge-0.053
Token We
Token ID775
Feature activation+1.54e-02

Pos logit contributions
Acc+0.052
 Acc+0.047
rou+0.047
 collapse+0.046
sm+0.043
Neg logit contributions
gers-0.038
 pouch-0.038
 goggles-0.038
 plunge-0.036
 couple-0.036
Token made
Token ID925
Feature activation+3.73e-02

Pos logit contributions
 Acc+0.030
Acc+0.028
 collapse+0.024
rou+0.024
 unavoidable+0.023
Neg logit contributions
 goggles-0.021
 pouch-0.021
gers-0.021
 couple-0.020
 banks-0.020
Token a
Token ID257
Feature activation+7.55e-02

Pos logit contributions
 Acc+0.082
Acc+0.081
tr+0.072
sm+0.068
 unavoidable+0.067
Neg logit contributions
 legs-0.038
 plunge-0.037
 kernels-0.037
 couple-0.037
 shoes-0.037
Token new
Token ID649
Feature activation+4.27e-02

Pos logit contributions
 Acc+0.191
Acc+0.185
ressive+0.169
sm+0.167
 resolved+0.161
Neg logit contributions
 bag-0.063
 pair-0.061
 couple-0.060
 line-0.054
 idea-0.054
Token tower
Token ID10580
Feature activation+3.64e-02

Pos logit contributions
Acc+0.079
 Acc+0.079
rou+0.065
 unavoidable+0.063
sm+0.062
Neg logit contributions
 couple-0.067
 pair-0.064
 legs-0.061
 pouch-0.061
 bag-0.060
Token.
Token ID13
Feature activation+4.02e-02

Pos logit contributions
 Acc+0.078
Acc+0.077
 unavoidable+0.061
rou+0.060
 collapse+0.059
Neg logit contributions
 couple-0.044
gers-0.044
 pouch-0.041
 pair-0.041
ner-0.040
Token A
Token ID317
Feature activation+6.08e-02

Pos logit contributions
Acc+0.078
rou+0.070
 Acc+0.070
 collapse+0.068
sm+0.067
Neg logit contributions
 goggles-0.050
 pouch-0.049
gers-0.049
ner-0.046
 couple-0.046
Token better
Token ID1365
Feature activation+2.40e-02

Pos logit contributions
 Acc+0.137
Acc+0.135
 unavoidable+0.113
sm+0.113
rou+0.113
Neg logit contributions
 couple-0.071
 pair-0.069
 bag-0.064
 legs-0.061
 biggest-0.060
TokenThe
Token ID464
Feature activation+8.63e-05

Pos logit contributions
 Acc+0.019
Acc+0.018
 collapse+0.016
rou+0.015
tr+0.015
Neg logit contributions
 couple-0.017
 pouch-0.017
gers-0.016
 therapist-0.016
 goggles-0.016
Token voice
Token ID3809
Feature activation+2.00e-02

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 couple-0.000
 pouch-0.000
gers-0.000
 goggles-0.000
 plunge-0.000
Token said
Token ID531
Feature activation+9.68e-03

Pos logit contributions
 Acc+0.041
Acc+0.041
rou+0.034
 collapse+0.033
tr+0.032
Neg logit contributions
 couple-0.026
 pouch-0.024
gers-0.024
 therapist-0.024
 plunge-0.023
Token,
Token ID11
Feature activation+3.85e-03

Pos logit contributions
 Acc+0.017
Acc+0.017
 collapse+0.015
rou+0.014
tr+0.014
Neg logit contributions
 couple-0.014
 pouch-0.014
gers-0.014
 therapist-0.013
 goggles-0.013
Token "
Token ID366
Feature activation+8.92e-03

Pos logit contributions
 Acc+0.007
Acc+0.007
 collapse+0.006
rou+0.006
tr+0.005
Neg logit contributions
 couple-0.006
 pouch-0.006
gers-0.005
 goggles-0.005
 therapist-0.005
Tokencong
Token ID36801
Feature activation+7.47e-02

Pos logit contributions
 Acc+0.016
Acc+0.015
 collapse+0.013
rou+0.012
tr+0.012
Neg logit contributions
 couple-0.014
 pouch-0.013
gers-0.013
 therapist-0.013
 goggles-0.013
Tokenratulations
Token ID30167
Feature activation-2.35e-02

Pos logit contributions
Acc+0.119
 collapse+0.111
 Acc+0.106
 gone+0.105
sm+0.103
Neg logit contributions
 pouch-0.110
 shoes-0.110
 penn-0.107
ippers-0.106
 banks-0.105
Token,
Token ID11
Feature activation+5.43e-03

Pos logit contributions
 pouch+0.047
 couple+0.046
 goggles+0.046
 banks+0.045
gers+0.045
Neg logit contributions
 Acc-0.029
Acc-0.026
rou-0.022
 collapse-0.021
 solved-0.021
Token you
Token ID345
Feature activation-6.38e-03

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.008
rou+0.008
tr+0.008
Neg logit contributions
 couple-0.008
 pouch-0.008
gers-0.008
 plunge-0.008
 goggles-0.008
Token can
Token ID460
Feature activation+2.69e-03

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.012
Acc-0.011
 collapse-0.010
rou-0.010
 solved-0.009
Token have
Token ID423
Feature activation+1.51e-02

Pos logit contributions
 Acc+0.005
Acc+0.005
rou+0.004
 collapse+0.004
tr+0.004
Neg logit contributions
 pouch-0.004
 couple-0.004
 goggles-0.004
 plunge-0.004
gers-0.004
Token deep
Token ID2769
Feature activation-1.60e-02

Pos logit contributions
 Acc+0.007
Acc+0.006
 collapse+0.006
rou+0.005
 solved+0.005
Neg logit contributions
 pouch-0.004
 couple-0.004
 goggles-0.004
 plunge-0.004
gers-0.004
Token breath
Token ID8033
Feature activation-1.62e-02

Pos logit contributions
 couple+0.017
 pouch+0.016
gers+0.016
 therapist+0.015
 goggles+0.015
Neg logit contributions
 Acc-0.035
Acc-0.033
 collapse-0.031
 solved-0.030
rou-0.029
Token and
Token ID290
Feature activation+8.26e-03

Pos logit contributions
 pouch+0.027
 couple+0.027
gers+0.027
 goggles+0.027
 banks+0.026
Neg logit contributions
 Acc-0.025
Acc-0.024
 collapse-0.020
rou-0.020
 solved-0.019
Token shouted
Token ID17293
Feature activation+1.36e-02

Pos logit contributions
 Acc+0.016
Acc+0.015
 collapse+0.013
rou+0.013
tr+0.013
Neg logit contributions
 couple-0.011
 goggles-0.011
 pouch-0.011
gers-0.011
 plunge-0.011
Token "
Token ID366
Feature activation+1.87e-02

Pos logit contributions
 Acc+0.026
Acc+0.026
 collapse+0.022
rou+0.022
tr+0.021
Neg logit contributions
 couple-0.018
 pouch-0.018
 goggles-0.018
gers-0.018
 Holly-0.017
TokenST
Token ID2257
Feature activation+7.46e-02

Pos logit contributions
 Acc+0.036
Acc+0.032
 collapse+0.028
rou+0.027
 solved+0.027
Neg logit contributions
 couple-0.028
 pouch-0.027
gers-0.027
 therapist-0.026
 plunge-0.026
TokenOP
Token ID3185
Feature activation-1.78e-02

Pos logit contributions
Acc+0.120
 Acc+0.108
rou+0.077
 blurry+0.076
 collapsing+0.076
Neg logit contributions
ippers-0.129
 pouch-0.128
aces-0.125
gers-0.122
 plunge-0.120
Token!"
Token ID2474
Feature activation-1.31e-02

Pos logit contributions
 couple+0.033
 pouch+0.033
 goggles+0.032
gers+0.032
 plunge+0.031
Neg logit contributions
 Acc-0.024
Acc-0.023
rou-0.019
 collapse-0.019
tr-0.018
Token
Token ID220
Feature activation-1.12e-02

Pos logit contributions
 pouch+0.019
 couple+0.019
gers+0.019
 goggles+0.019
 plunge+0.019
Neg logit contributions
 Acc-0.023
Acc-0.022
 collapse-0.019
tr-0.018
rou-0.018
Token
Token ID198
Feature activation+4.39e-03

Pos logit contributions
 couple+0.017
 pouch+0.017
gers+0.017
 goggles+0.017
 plunge+0.016
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
rou-0.015
tr-0.015
Token
Token ID198
Feature activation+8.62e-03

Pos logit contributions
 Acc+0.008
Acc+0.008
 collapse+0.007
rou+0.007
tr+0.006
Neg logit contributions
 couple-0.006
 pouch-0.006
gers-0.006
 goggles-0.006
 therapist-0.006
Token need
Token ID761
Feature activation+3.05e-02

Pos logit contributions
 Acc+0.019
Acc+0.018
 collapse+0.016
rou+0.015
tr+0.015
Neg logit contributions
 pouch-0.013
gers-0.013
 goggles-0.012
 couple-0.012
 pair-0.012
Token help
Token ID1037
Feature activation+1.68e-02

Pos logit contributions
 Acc+0.061
Acc+0.061
rou+0.052
 collapse+0.051
tr+0.050
Neg logit contributions
 goggles-0.040
 pouch-0.039
 pills-0.039
 sleeves-0.039
 legs-0.038
Token!"
Token ID2474
Feature activation+1.19e-02

Pos logit contributions
 Acc+0.035
Acc+0.034
rou+0.029
 collapse+0.028
tr+0.027
Neg logit contributions
 pouch-0.021
 couple-0.021
 legs-0.021
 goggles-0.020
 sleeves-0.020
Token
Token ID198
Feature activation+3.00e-02

Pos logit contributions
 Acc+0.026
Acc+0.025
rou+0.023
tr+0.022
 collapse+0.021
Neg logit contributions
 couple-0.014
 pouch-0.013
 goggles-0.013
gers-0.013
 legs-0.012
Token
Token ID198
Feature activation+3.39e-02

Pos logit contributions
 Acc+0.070
Acc+0.069
 solved+0.061
rou+0.061
tr+0.061
Neg logit contributions
 pouch-0.026
 couple-0.023
gers-0.023
ittens-0.023
ippers-0.023
TokenS
Token ID50
Feature activation+7.42e-02

Pos logit contributions
 Acc+0.075
Acc+0.068
rou+0.062
tr+0.059
 unavoidable+0.058
Neg logit contributions
gers-0.040
 couple-0.038
 pouch-0.038
 legs-0.036
ittens-0.034
Tokenue
Token ID518
Feature activation+3.18e-02

Pos logit contributions
 Acc+0.131
Acc+0.125
rou+0.093
 unavoidable+0.091
bul+0.090
Neg logit contributions
 bag-0.112
owers-0.109
ittens-0.105
re-0.104
 lid-0.104
Token is
Token ID318
Feature activation+6.71e-03

Pos logit contributions
 Acc+0.062
Acc+0.059
 unavoidable+0.052
 collapse+0.050
 resolved+0.049
Neg logit contributions
 pouch-0.041
 legs-0.041
 banks-0.041
 bag-0.041
 sleeves-0.040
Token scared
Token ID12008
Feature activation+1.65e-02

Pos logit contributions
 Acc+0.015
Acc+0.014
 collapse+0.012
rou+0.012
tr+0.012
Neg logit contributions
 pouch-0.008
 goggles-0.007
 couple-0.007
gers-0.007
 pair-0.007
Token.
Token ID13
Feature activation+8.64e-03

Pos logit contributions
 Acc+0.035
Acc+0.033
tr+0.029
 solved+0.028
rou+0.028
Neg logit contributions
 legs-0.020
 couple-0.019
 sleeves-0.019
 pouch-0.019
gers-0.019
Token She
Token ID1375
Feature activation+1.18e-02

Pos logit contributions
 Acc+0.019
Acc+0.019
rou+0.017
 collapse+0.016
 solved+0.016
Neg logit contributions
 pouch-0.009
 couple-0.009
gers-0.009
 Holly-0.008
 goggles-0.008
Token wander
Token ID27776
Feature activation-1.69e-02

Pos logit contributions
 Acc+0.007
Acc+0.007
rou+0.006
 collapse+0.006
tr+0.006
Neg logit contributions
 pouch-0.006
 couple-0.006
 goggles-0.006
 banks-0.006
 plunge-0.006
Token.
Token ID13
Feature activation+2.69e-02

Pos logit contributions
 couple+0.029
 pouch+0.029
 plunge+0.028
 therapist+0.028
 goggles+0.028
Neg logit contributions
 Acc-0.025
Acc-0.023
 collapse-0.020
 over-0.020
 solved-0.020
Token
Token ID220
Feature activation+2.10e-02

Pos logit contributions
 Acc+0.056
Acc+0.056
rou+0.047
 collapse+0.047
sm+0.045
Neg logit contributions
 couple-0.032
 pouch-0.031
gers-0.030
 goggles-0.029
 Holly-0.029
Token
Token ID198
Feature activation+1.05e-02

Pos logit contributions
 Acc+0.043
Acc+0.042
 collapse+0.037
rou+0.036
tr+0.035
Neg logit contributions
 pouch-0.025
ippers-0.024
 couple-0.024
gers-0.023
 goggles-0.023
Token
Token ID198
Feature activation+2.25e-02

Pos logit contributions
 Acc+0.021
Acc+0.020
 collapse+0.018
rou+0.017
tr+0.017
Neg logit contributions
 couple-0.013
 pouch-0.013
gers-0.013
 goggles-0.013
 banks-0.012
TokenJ
Token ID41
Feature activation+7.33e-02

Pos logit contributions
 Acc+0.046
Acc+0.041
 collapse+0.035
rou+0.034
tr+0.034
Neg logit contributions
 couple-0.032
gers-0.031
 pouch-0.029
 goggles-0.028
 therapist-0.028
Tokenill
Token ID359
Feature activation+1.79e-03

Pos logit contributions
 Acc+0.133
Acc+0.127
 collapse+0.108
 cal+0.102
 unavoidable+0.096
Neg logit contributions
re-0.119
aces-0.111
ippers-0.110
bles-0.109
ittens-0.102
Token started
Token ID2067
Feature activation-1.34e-02

Pos logit contributions
 Acc+0.003
Acc+0.003
 collapse+0.003
rou+0.002
tr+0.002
Neg logit contributions
 pouch-0.003
 couple-0.003
gers-0.003
 goggles-0.003
 banks-0.003
Token to
Token ID284
Feature activation-7.81e-03

Pos logit contributions
gers+0.025
 pouch+0.025
 couple+0.025
 goggles+0.024
 plunge+0.024
Neg logit contributions
 Acc-0.019
Acc-0.018
 collapse-0.015
rou-0.015
 solved-0.014
Token feel
Token ID1254
Feature activation-1.42e-02

Pos logit contributions
 couple+0.011
 pouch+0.011
gers+0.011
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.015
Acc-0.014
 collapse-0.012
rou-0.012
 solved-0.012
Token jealous
Token ID19354
Feature activation-1.32e-02

Pos logit contributions
 couple+0.022
 pouch+0.022
gers+0.022
 sleeves+0.021
 goggles+0.021
Neg logit contributions
 Acc-0.024
Acc-0.023
 collapse-0.020
rou-0.019
 solved-0.019
Token.
Token ID13
Feature activation-5.92e-03

Pos logit contributions
 couple+0.028
 pouch+0.028
 goggles+0.027
gers+0.027
 plunge+0.027
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.018
tr-0.017
rou-0.017
Token The
Token ID383
Feature activation-1.94e-02

Pos logit contributions
 couple+0.010
 pouch+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.007
tr-0.007
Token sign
Token ID1051
Feature activation-6.16e-04

Pos logit contributions
 Holly+0.030
 pouch+0.030
 therapist+0.029
gers+0.029
 pills+0.029
Neg logit contributions
 Acc-0.031
Acc-0.029
 collapse-0.026
tr-0.026
 solved-0.025
Token said
Token ID531
Feature activation+3.16e-03

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token "
Token ID366
Feature activation-7.36e-03

Pos logit contributions
 Acc+0.006
Acc+0.006
rou+0.005
 collapse+0.005
tr+0.005
Neg logit contributions
 couple-0.004
 pouch-0.004
gers-0.004
 goggles-0.004
 plunge-0.004
TokenST
Token ID2257
Feature activation+7.31e-02

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.011
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.013
Acc-0.012
 collapse-0.010
rou-0.010
tr-0.010
TokenOP
Token ID3185
Feature activation-2.14e-02

Pos logit contributions
Acc+0.104
 Acc+0.095
rou+0.079
 collapse+0.072
 collapsing+0.065
Neg logit contributions
 pouch-0.136
gers-0.134
ippers-0.134
aces-0.129
 banks-0.125
Token".
Token ID1911
Feature activation-1.74e-02

Pos logit contributions
 couple+0.035
 pouch+0.035
 goggles+0.035
gers+0.034
 plunge+0.034
Neg logit contributions
 Acc-0.034
Acc-0.032
rou-0.028
 collapse-0.028
tr-0.027
Token
Token ID198
Feature activation-1.02e-02

Pos logit contributions
 couple+0.035
 pouch+0.035
 therapist+0.035
gers+0.034
 goggles+0.034
Neg logit contributions
 Acc-0.022
Acc-0.020
 collapse-0.017
 solved-0.016
tr-0.015
Token
Token ID198
Feature activation-4.83e-03

Pos logit contributions
 couple+0.016
 pouch+0.016
gers+0.016
 goggles+0.015
 plunge+0.015
Neg logit contributions
 Acc-0.017
Acc-0.017
 collapse-0.014
rou-0.014
 solved-0.014
Token"
Token ID1
Feature activation-3.74e-03

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.009
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.007
Token said
Token ID531
Feature activation+2.31e-04

Pos logit contributions
gers+0.044
 plunge+0.042
 goggles+0.042
ippers+0.042
 pouch+0.041
Neg logit contributions
 Acc-0.052
Acc-0.049
 solved-0.044
 fixed-0.043
 over-0.042
Token sadly
Token ID21098
Feature activation+3.99e-03

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 couple-0.000
 pouch-0.000
gers-0.000
 goggles-0.000
 therapist-0.000
Token.
Token ID13
Feature activation+2.11e-02

Pos logit contributions
 Acc+0.007
Acc+0.007
 collapse+0.006
rou+0.006
tr+0.005
Neg logit contributions
 pouch-0.006
 couple-0.006
gers-0.006
 goggles-0.006
 banks-0.006
Token
Token ID198
Feature activation+3.40e-02

Pos logit contributions
Acc+0.046
 Acc+0.045
 collapse+0.042
rou+0.041
 solved+0.040
Neg logit contributions
 pouch-0.021
 couple-0.020
 Holly-0.019
 legs-0.019
ner-0.018
Token
Token ID198
Feature activation+3.61e-02

Pos logit contributions
 Acc+0.075
Acc+0.073
 collapse+0.070
tr+0.066
rou+0.065
Neg logit contributions
ittens-0.031
 pouch-0.031
 couple-0.028
 Holly-0.027
 therapist-0.027
TokenS
Token ID50
Feature activation+7.30e-02

Pos logit contributions
 Acc+0.079
Acc+0.072
 collapse+0.072
tr+0.069
rou+0.068
Neg logit contributions
 couple-0.033
gers-0.032
ittens-0.031
 pouch-0.031
 therapist-0.029
Tokenally
Token ID453
Feature activation+1.97e-02

Pos logit contributions
 Acc+0.133
Acc+0.120
 over+0.120
 collapse+0.119
 explode+0.114
Neg logit contributions
ittens-0.092
re-0.086
ve-0.080
owers-0.079
ancer-0.076
Token tried
Token ID3088
Feature activation+1.91e-02

Pos logit contributions
 Acc+0.037
Acc+0.035
 collapse+0.031
rou+0.030
 unavoidable+0.029
Neg logit contributions
 pouch-0.028
 couple-0.027
 therapist-0.027
 pair-0.026
 sleeves-0.026
Token to
Token ID284
Feature activation+8.60e-03

Pos logit contributions
 Acc+0.034
Acc+0.032
tr+0.028
rou+0.027
 collapse+0.027
Neg logit contributions
 couple-0.028
 goggles-0.028
 pouch-0.028
 Holly-0.028
 pair-0.027
Token cheer
Token ID14042
Feature activation+2.02e-02

Pos logit contributions
 Acc+0.015
Acc+0.014
rou+0.012
tr+0.012
 collapse+0.012
Neg logit contributions
 plunge-0.013
 couple-0.013
 goggles-0.013
 pouch-0.013
 pair-0.013
Token him
Token ID683
Feature activation+1.36e-02

Pos logit contributions
 Acc+0.038
Acc+0.037
rou+0.033
 solved+0.033
 collapse+0.032
Neg logit contributions
gers-0.027
 shoes-0.027
 Holly-0.027
ippers-0.026
 goggles-0.026
Token she
Token ID673
Feature activation+6.45e-03

Pos logit contributions
 pouch+0.023
gers+0.022
 couple+0.022
 goggles+0.022
ippers+0.022
Neg logit contributions
 Acc-0.025
Acc-0.024
 collapse-0.021
 solved-0.020
tr-0.020
Token heard
Token ID2982
Feature activation+2.72e-02

Pos logit contributions
 Acc+0.013
Acc+0.012
 collapse+0.010
rou+0.010
tr+0.010
Neg logit contributions
 couple-0.009
 pouch-0.009
 goggles-0.009
gers-0.009
 sleeves-0.008
Token a
Token ID257
Feature activation-8.63e-04

Pos logit contributions
 Acc+0.058
Acc+0.057
rou+0.048
 collapse+0.046
tr+0.044
Neg logit contributions
 couple-0.034
gers-0.031
 plunge-0.031
 shoes-0.030
 pouch-0.030
Token loud
Token ID7812
Feature activation-7.75e-04

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.002
Acc-0.002
 collapse-0.001
rou-0.001
tr-0.001
Token,
Token ID11
Feature activation+2.06e-02

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.002
Acc-0.002
 collapse-0.001
rou-0.001
tr-0.001
Token gru
Token ID22848
Feature activation+7.28e-02

Pos logit contributions
 Acc+0.041
Acc+0.038
rou+0.033
 collapse+0.032
 resolved+0.032
Neg logit contributions
 couple-0.028
 pouch-0.027
 goggles-0.027
gers-0.026
 legs-0.026
Tokenff
Token ID487
Feature activation+7.50e-03

Pos logit contributions
Acc+0.155
 Acc+0.142
 resolved+0.133
 unavoidable+0.132
 solved+0.125
Neg logit contributions
 pouch-0.089
 legs-0.082
ippers-0.078
 neck-0.076
re-0.074
Token voice
Token ID3809
Feature activation+1.86e-02

Pos logit contributions
 Acc+0.020
Acc+0.019
rou+0.017
 solved+0.017
 collapse+0.017
Neg logit contributions
 couple-0.005
 pouch-0.005
gers-0.005
 plunge-0.004
 goggles-0.004
Token calling
Token ID4585
Feature activation-1.67e-04

Pos logit contributions
 Acc+0.033
Acc+0.032
rou+0.026
 solved+0.025
 collapse+0.024
Neg logit contributions
 couple-0.029
 pouch-0.028
gers-0.028
 plunge-0.028
 goggles-0.028
Token her
Token ID607
Feature activation+1.74e-02

Pos logit contributions
 couple+0.000
 pouch+0.000
gers+0.000
 goggles+0.000
 plunge+0.000
Neg logit contributions
 Acc-0.000
Acc-0.000
rou-0.000
 collapse-0.000
tr-0.000
Token name
Token ID1438
Feature activation+1.38e-02

Pos logit contributions
Acc+0.041
 Acc+0.040
 collapse+0.033
rou+0.032
 solved+0.032
Neg logit contributions
 pouch-0.019
 couple-0.017
gers-0.017
 goggles-0.017
 pair-0.017
Token Tom
Token ID4186
Feature activation+6.09e-03

Pos logit contributions
 Acc+0.009
Acc+0.008
rou+0.007
tr+0.007
 collapse+0.007
Neg logit contributions
 couple-0.009
 pouch-0.008
gers-0.008
 goggles-0.008
 plunge-0.008
Token and
Token ID290
Feature activation-3.86e-03

Pos logit contributions
 Acc+0.011
Acc+0.011
 collapse+0.009
rou+0.009
tr+0.009
Neg logit contributions
 pouch-0.009
 couple-0.009
 goggles-0.009
gers-0.009
 banks-0.009
Token said
Token ID531
Feature activation+1.04e-02

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.008
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.007
Token in
Token ID287
Feature activation-3.86e-03

Pos logit contributions
 Acc+0.021
Acc+0.021
 collapse+0.018
rou+0.018
tr+0.018
Neg logit contributions
 couple-0.012
 pouch-0.012
gers-0.012
 goggles-0.012
 plunge-0.012
Token a
Token ID257
Feature activation-1.71e-02

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.007
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token gru
Token ID22848
Feature activation+7.23e-02

Pos logit contributions
gers+0.027
 couple+0.027
 pouch+0.026
 goggles+0.026
ner+0.026
Neg logit contributions
 Acc-0.028
Acc-0.027
 collapse-0.024
 solved-0.022
rou-0.022
Tokenff
Token ID487
Feature activation-1.24e-02

Pos logit contributions
Acc+0.149
 Acc+0.136
 resolved+0.126
 unavoidable+0.124
 solved+0.124
Neg logit contributions
 pouch-0.091
 legs-0.082
ippers-0.082
gers-0.080
 couple-0.080
Token voice
Token ID3809
Feature activation+1.20e-02

Pos logit contributions
 pouch+0.009
gers+0.009
 couple+0.009
 goggles+0.009
 plunge+0.008
Neg logit contributions
 Acc-0.031
Acc-0.030
 collapse-0.028
rou-0.027
tr-0.027
Token,
Token ID11
Feature activation-1.25e-02

Pos logit contributions
 Acc+0.024
Acc+0.024
 collapse+0.020
rou+0.020
 solved+0.020
Neg logit contributions
 couple-0.015
 pouch-0.015
gers-0.015
 plunge-0.014
 therapist-0.014
Token "
Token ID366
Feature activation+1.03e-02

Pos logit contributions
 pouch+0.016
gers+0.016
 couple+0.016
 goggles+0.016
 plunge+0.016
Neg logit contributions
 Acc-0.025
Acc-0.024
 collapse-0.020
rou-0.020
tr-0.020
TokenGet
Token ID3855
Feature activation+4.48e-03

Pos logit contributions
 Acc+0.018
Acc+0.017
 collapse+0.014
rou+0.014
 solved+0.014
Neg logit contributions
 couple-0.016
gers-0.016
 pouch-0.016
 therapist-0.016
 goggles-0.016
Token a
Token ID257
Feature activation+4.54e-02

Pos logit contributions
 Acc+0.014
Acc+0.014
 collapse+0.012
tr+0.012
rou+0.012
Neg logit contributions
 couple-0.010
 pouch-0.010
 goggles-0.009
gers-0.009
 plunge-0.009
Token dessert
Token ID23084
Feature activation-4.65e-03

Pos logit contributions
Acc+0.094
 Acc+0.093
 resolved+0.076
 unavoidable+0.076
sm+0.075
Neg logit contributions
 couple-0.063
 pouch-0.058
 bag-0.058
 pair-0.058
 bank-0.053
Token?"
Token ID1701
Feature activation+1.19e-02

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.008
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
Token
Token ID198
Feature activation+4.11e-02

Pos logit contributions
 Acc+0.027
Acc+0.026
 collapse+0.023
rou+0.023
tr+0.023
Neg logit contributions
 couple-0.012
 pouch-0.012
 goggles-0.011
gers-0.011
 pair-0.011
Token
Token ID198
Feature activation+4.98e-02

Pos logit contributions
 Acc+0.098
Acc+0.096
 collapse+0.090
tr+0.089
rou+0.086
Neg logit contributions
 couple-0.031
 Holly-0.027
 pair-0.026
 pouch-0.025
ner-0.024
TokenJ
Token ID41
Feature activation+7.23e-02

Pos logit contributions
 Acc+0.092
Acc+0.082
arter+0.078
 collapse+0.076
 cleans+0.076
Neg logit contributions
 couple-0.072
 pair-0.061
 therapist-0.061
gers-0.060
ner-0.057
Tokenill
Token ID359
Feature activation+1.57e-02

Pos logit contributions
 Acc+0.118
Acc+0.118
 collapse+0.109
 cal+0.102
 unavoidable+0.097
Neg logit contributions
re-0.112
aces-0.107
ippers-0.104
ve-0.103
bles-0.101
Token was
Token ID373
Feature activation+2.99e-04

Pos logit contributions
 Acc+0.030
Acc+0.029
 collapse+0.025
rou+0.024
tr+0.024
Neg logit contributions
 couple-0.021
 pouch-0.021
 therapist-0.020
 pair-0.020
 banks-0.020
Token delighted
Token ID20707
Feature activation-1.63e-03

Pos logit contributions
 Acc+0.001
Acc+0.001
 collapse+0.001
tr+0.001
rou+0.001
Neg logit contributions
 pouch-0.000
 couple-0.000
 goggles-0.000
gers-0.000
 plunge-0.000
Token.
Token ID13
Feature activation+2.35e-02

Pos logit contributions
 couple+0.003
 pouch+0.003
gers+0.002
 goggles+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.002
rou-0.002
tr-0.002
Token "
Token ID366
Feature activation+7.71e-03

Pos logit contributions
Acc+0.051
 Acc+0.051
 collapse+0.048
rou+0.045
sm+0.045
Neg logit contributions
 couple-0.022
 pouch-0.022
ner-0.021
 Holly-0.021
gers-0.020
Token bad
Token ID2089
Feature activation-1.26e-02

Pos logit contributions
 Acc+0.053
Acc+0.052
 collapse+0.045
rou+0.044
 solved+0.043
Neg logit contributions
 couple-0.032
 pouch-0.031
 snail-0.031
 pair-0.030
 donkey-0.030
Token dog
Token ID3290
Feature activation+4.02e-03

Pos logit contributions
 pouch+0.014
 couple+0.014
 goggles+0.014
gers+0.014
 plunge+0.014
Neg logit contributions
 Acc-0.027
Acc-0.026
 collapse-0.023
tr-0.022
rou-0.022
Token."
Token ID526
Feature activation+1.26e-02

Pos logit contributions
 Acc+0.007
Acc+0.007
 collapse+0.006
rou+0.006
tr+0.006
Neg logit contributions
 pouch-0.006
 couple-0.006
gers-0.006
 goggles-0.006
 therapist-0.006
Token
Token ID198
Feature activation+2.48e-02

Pos logit contributions
 Acc+0.027
Acc+0.026
rou+0.024
 collapse+0.023
tr+0.023
Neg logit contributions
 couple-0.014
 pouch-0.014
gers-0.013
 goggles-0.013
 Holly-0.013
Token
Token ID198
Feature activation+3.12e-02

Pos logit contributions
 Acc+0.057
Acc+0.056
tr+0.049
rou+0.049
 collapse+0.049
Neg logit contributions
 pouch-0.023
 couple-0.021
gers-0.021
 goggles-0.020
 legs-0.020
TokenS
Token ID50
Feature activation+7.22e-02

Pos logit contributions
 Acc+0.063
Acc+0.056
rou+0.049
tr+0.049
 collapse+0.049
Neg logit contributions
gers-0.042
 couple-0.042
 pouch-0.041
 legs-0.039
 snail-0.037
Tokenue
Token ID518
Feature activation+2.82e-02

Pos logit contributions
 Acc+0.131
Acc+0.117
rou+0.090
 over+0.087
aling+0.083
Neg logit contributions
 bag-0.114
owers-0.109
re-0.106
ittens-0.106
 lid-0.104
Token hugged
Token ID41130
Feature activation+3.40e-02

Pos logit contributions
 Acc+0.066
Acc+0.062
 unavoidable+0.056
 collapse+0.056
rou+0.055
Neg logit contributions
 legs-0.027
 pouch-0.027
 bag-0.026
 sleeves-0.026
 couple-0.025
Token Tom
Token ID4186
Feature activation+2.18e-02

Pos logit contributions
 Acc+0.062
Acc+0.055
rou+0.051
tr+0.047
 solved+0.047
Neg logit contributions
 shoes-0.055
gers-0.054
 couple-0.054
 ones-0.054
 legs-0.053
Token and
Token ID290
Feature activation+3.61e-02

Pos logit contributions
 Acc+0.045
Acc+0.042
 collapse+0.038
tr+0.036
rou+0.036
Neg logit contributions
 pouch-0.029
 legs-0.027
 bag-0.027
 couple-0.027
gers-0.026
Token cried
Token ID16896
Feature activation+1.97e-02

Pos logit contributions
 Acc+0.088
Acc+0.084
rou+0.078
 collapse+0.075
tr+0.073
Neg logit contributions
 couple-0.033
 shoes-0.031
 bag-0.030
 goggles-0.030
 legs-0.029
Token.
Token ID13
Feature activation+1.82e-02

Pos logit contributions
 Acc+0.027
Acc+0.026
tr+0.023
 collapse+0.023
rou+0.022
Neg logit contributions
 couple-0.018
 Holly-0.017
gers-0.017
 pills-0.016
 pouch-0.016
Token They
Token ID1119
Feature activation+1.18e-02

Pos logit contributions
Acc+0.039
 Acc+0.039
 collapse+0.036
rou+0.035
tr+0.034
Neg logit contributions
 pouch-0.019
 Holly-0.018
 couple-0.018
gers-0.018
 pills-0.017
Token thought
Token ID1807
Feature activation+6.66e-03

Pos logit contributions
 Acc+0.025
Acc+0.024
 collapse+0.021
tr+0.020
rou+0.020
Neg logit contributions
 couple-0.015
 pouch-0.014
 goggles-0.014
 pair-0.013
gers-0.013
Token it
Token ID340
Feature activation+1.65e-02

Pos logit contributions
 Acc+0.013
Acc+0.012
tr+0.011
 collapse+0.010
rou+0.010
Neg logit contributions
 couple-0.009
 goggles-0.009
gers-0.009
 plunge-0.009
 Holly-0.009
Token was
Token ID373
Feature activation+2.85e-02

Pos logit contributions
 Acc+0.034
Acc+0.032
rou+0.029
 collapse+0.028
 unavoidable+0.026
Neg logit contributions
gers-0.020
 couple-0.020
 goggles-0.020
 pouch-0.020
 pair-0.019
Token a
Token ID257
Feature activation+7.22e-02

Pos logit contributions
 Acc+0.060
Acc+0.058
tr+0.052
 collapse+0.052
rou+0.051
Neg logit contributions
 goggles-0.033
 Holly-0.032
 pills-0.031
 potatoes-0.031
 pair-0.031
Token good
Token ID922
Feature activation+1.17e-02

Pos logit contributions
 Acc+0.165
Acc+0.161
 collapse+0.161
ayed+0.160
atters+0.155
Neg logit contributions
 couple-0.067
 pair-0.067
 snail-0.064
 bag-0.061
 pig-0.057
Token day
Token ID1110
Feature activation+1.44e-02

Pos logit contributions
 Acc+0.029
Acc+0.029
 collapse+0.026
rou+0.025
 unavoidable+0.024
Neg logit contributions
 couple-0.010
 pair-0.010
gers-0.009
 pouch-0.009
 goggles-0.009
Token at
Token ID379
Feature activation+2.78e-02

Pos logit contributions
 Acc+0.028
Acc+0.027
 collapse+0.023
rou+0.022
tr+0.021
Neg logit contributions
 couple-0.022
 pouch-0.020
gers-0.020
 goggles-0.019
 bag-0.019
Token the
Token ID262
Feature activation+3.22e-02

Pos logit contributions
 Acc+0.059
Acc+0.057
tr+0.052
rou+0.051
sm+0.049
Neg logit contributions
 couple-0.032
 pair-0.031
 plunge-0.030
 goggles-0.030
ner-0.029
Token park
Token ID3952
Feature activation+1.17e-02

Pos logit contributions
 Acc+0.076
Acc+0.074
 resolved+0.063
 collapse+0.062
sm+0.062
Neg logit contributions
 bag-0.034
 bank-0.033
 banks-0.032
 couple-0.032
 ones-0.031
Token were
Token ID547
Feature activation+1.31e-02

Pos logit contributions
 Acc+0.045
Acc+0.044
rou+0.037
tr+0.037
 collapse+0.037
Neg logit contributions
 couple-0.029
 legs-0.029
 goggles-0.028
gers-0.028
 shoes-0.026
Token happy
Token ID3772
Feature activation+2.14e-02

Pos logit contributions
 Acc+0.028
Acc+0.027
tr+0.024
 collapse+0.023
ray+0.023
Neg logit contributions
 couple-0.015
 pouch-0.014
 goggles-0.014
 pair-0.014
 legs-0.014
Token.
Token ID13
Feature activation+3.87e-02

Pos logit contributions
 Acc+0.041
Acc+0.040
tr+0.035
 collapse+0.034
 unavoidable+0.034
Neg logit contributions
 couple-0.030
 legs-0.028
gers-0.027
 goggles-0.027
 shoes-0.027
Token The
Token ID383
Feature activation-9.41e-03

Pos logit contributions
Acc+0.074
 Acc+0.069
rou+0.069
 collapse+0.066
 cal+0.061
Neg logit contributions
 banks-0.049
 pouch-0.046
gers-0.044
 couple-0.044
ittens-0.043
Token conflict
Token ID5358
Feature activation+3.60e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.014
 goggles+0.014
 banks+0.014
Neg logit contributions
 Acc-0.016
Acc-0.016
 collapse-0.013
rou-0.013
 solved-0.013
Token was
Token ID373
Feature activation+7.21e-02

Pos logit contributions
Acc+0.080
 Acc+0.080
rou+0.065
 collapse+0.062
Un+0.061
Neg logit contributions
gers-0.042
 pair-0.039
 bag-0.039
 couple-0.038
 pouch-0.038
Token over
Token ID625
Feature activation+2.50e-02

Pos logit contributions
 Acc+0.162
Acc+0.146
rou+0.131
ray+0.121
anne+0.118
Neg logit contributions
 none-0.084
 most-0.082
 friends-0.076
 best-0.075
 all-0.075
Token,
Token ID11
Feature activation+2.63e-02

Pos logit contributions
 Acc+0.058
Acc+0.058
tr+0.056
 collapse+0.052
rou+0.050
Neg logit contributions
 couple-0.022
ippers-0.021
kins-0.021
 bag-0.020
re-0.020
Token and
Token ID290
Feature activation+3.15e-02

Pos logit contributions
 Acc+0.056
Acc+0.053
sm+0.046
ray+0.046
rou+0.046
Neg logit contributions
 legs-0.029
 couple-0.029
 goggles-0.028
 pouch-0.028
gers-0.027
Token they
Token ID484
Feature activation+1.61e-02

Pos logit contributions
 Acc+0.060
Acc+0.058
rou+0.050
sm+0.047
tr+0.046
Neg logit contributions
 couple-0.043
 legs-0.042
 banks-0.041
 plunge-0.041
 pouch-0.040
Token played
Token ID2826
Feature activation-1.16e-03

Pos logit contributions
 Acc+0.035
Acc+0.033
rou+0.028
 collapse+0.028
sm+0.027
Neg logit contributions
 couple-0.019
 pouch-0.019
 goggles-0.018
 banks-0.018
gers-0.018
Token learned
Token ID4499
Feature activation+3.55e-02

Pos logit contributions
 couple+0.015
 pouch+0.015
gers+0.015
 goggles+0.015
 plunge+0.015
Neg logit contributions
 Acc-0.015
Acc-0.014
 collapse-0.012
rou-0.012
 solved-0.011
Token their
Token ID511
Feature activation+1.77e-02

Pos logit contributions
 Acc+0.067
Acc+0.066
 collapse+0.059
rou+0.058
 solved+0.057
Neg logit contributions
 pills-0.046
 goggles-0.045
 shoes-0.045
 penn-0.042
 couple-0.042
Token lesson
Token ID11483
Feature activation+2.24e-02

Pos logit contributions
 Acc+0.036
Acc+0.035
 solved+0.031
 collapse+0.031
rou+0.030
Neg logit contributions
 goggles-0.021
 couple-0.020
 banks-0.020
 pouch-0.020
 plunge-0.020
Token.
Token ID13
Feature activation+1.16e-02

Pos logit contributions
 Acc+0.047
Acc+0.044
 collapse+0.039
tr+0.038
rou+0.038
Neg logit contributions
gers-0.029
 couple-0.027
 therapist-0.026
 pouch-0.025
 banks-0.025
Token They
Token ID1119
Feature activation+8.39e-03

Pos logit contributions
Acc+0.024
 Acc+0.024
rou+0.022
 collapse+0.022
 solved+0.021
Neg logit contributions
 pouch-0.013
 couple-0.013
gers-0.012
 goggles-0.012
 Holly-0.012
Token never
Token ID1239
Feature activation+7.21e-02

Pos logit contributions
 Acc+0.016
Acc+0.015
 collapse+0.013
rou+0.013
tr+0.013
Neg logit contributions
 pouch-0.012
 couple-0.012
gers-0.011
 goggles-0.011
 banks-0.011
Token played
Token ID2826
Feature activation+8.88e-03

Pos logit contributions
 Acc+0.174
Acc+0.170
rou+0.142
 unavoidable+0.140
rap+0.139
Neg logit contributions
 begged-0.053
 legs-0.053
 goggles-0.053
 pair-0.053
re-0.051
Token with
Token ID351
Feature activation+2.12e-02

Pos logit contributions
 Acc+0.017
Acc+0.016
rou+0.014
 collapse+0.014
tr+0.014
Neg logit contributions
gers-0.012
 goggles-0.012
 couple-0.012
 pouch-0.012
 banks-0.012
Token the
Token ID262
Feature activation-9.59e-03

Pos logit contributions
 Acc+0.043
Acc+0.042
rou+0.036
tr+0.036
 collapse+0.036
Neg logit contributions
 goggles-0.027
 pills-0.026
 banks-0.026
 penn-0.025
 shoes-0.025
Token ladder
Token ID18002
Feature activation+5.10e-02

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.010
 goggles+0.010
 therapist+0.010
Neg logit contributions
 Acc-0.021
Acc-0.020
 collapse-0.018
rou-0.017
tr-0.017
Token and
Token ID290
Feature activation+3.42e-02

Pos logit contributions
 Acc+0.081
Acc+0.073
rou+0.052
isdom+0.052
sm+0.051
Neg logit contributions
 bag-0.091
gers-0.090
 bank-0.090
 couple-0.089
 pair-0.087
Token her
Token ID607
Feature activation-3.62e-03

Pos logit contributions
 Acc+0.017
Acc+0.016
tr+0.014
rou+0.014
 collapse+0.013
Neg logit contributions
 couple-0.012
 pouch-0.012
 goggles-0.012
gers-0.011
 sleeves-0.011
Token room
Token ID2119
Feature activation+6.52e-03

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.005
 goggles+0.005
 plunge+0.004
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
Token.
Token ID13
Feature activation+2.19e-02

Pos logit contributions
 Acc+0.011
Acc+0.011
 collapse+0.009
rou+0.009
tr+0.009
Neg logit contributions
 couple-0.010
 pouch-0.010
gers-0.010
 goggles-0.010
 banks-0.010
Token
Token ID198
Feature activation+1.52e-02

Pos logit contributions
Acc+0.051
 Acc+0.051
rou+0.048
 collapse+0.044
 solved+0.043
Neg logit contributions
 pouch-0.019
 couple-0.018
gers-0.017
 goggles-0.017
 banks-0.017
Token
Token ID198
Feature activation+2.87e-02

Pos logit contributions
 Acc+0.032
Acc+0.031
rou+0.027
tr+0.027
 collapse+0.027
Neg logit contributions
 pouch-0.017
 couple-0.017
gers-0.016
 goggles-0.016
 banks-0.016
TokenS
Token ID50
Feature activation+7.16e-02

Pos logit contributions
 Acc+0.064
Acc+0.058
rou+0.052
 collapse+0.049
tr+0.048
Neg logit contributions
 couple-0.034
gers-0.034
 pouch-0.033
 therapist-0.030
 goggles-0.030
Tokenue
Token ID518
Feature activation+1.59e-02

Pos logit contributions
 Acc+0.131
Acc+0.116
rou+0.086
 over+0.085
 collapse+0.085
Neg logit contributions
 bag-0.114
owers-0.107
ittens-0.105
re-0.105
gers-0.103
Token hears
Token ID23180
Feature activation+3.07e-02

Pos logit contributions
 Acc+0.027
Acc+0.026
rou+0.021
 collapse+0.020
tr+0.020
Neg logit contributions
 pouch-0.026
gers-0.025
 banks-0.025
 goggles-0.025
 therapist-0.025
Token Tom
Token ID4186
Feature activation+5.76e-03

Pos logit contributions
 Acc+0.053
Acc+0.053
rou+0.044
tr+0.043
 solved+0.039
Neg logit contributions
 couple-0.051
 pills-0.048
 shoes-0.048
 Holly-0.047
 goggles-0.047
Token's
Token ID338
Feature activation-1.03e-02

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.008
rou+0.008
tr+0.008
Neg logit contributions
 pouch-0.009
 couple-0.008
 goggles-0.008
 plunge-0.008
gers-0.008
Token me
Token ID502
Feature activation+1.70e-02

Pos logit contributions
 pouch+0.012
 couple+0.012
gers+0.012
 goggles+0.012
 plunge+0.011
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.018
tr-0.018
rou-0.018
Token,
Token ID11
Feature activation-7.18e-03

Pos logit contributions
Acc+0.019
 Acc+0.019
rou+0.016
 collapse+0.016
sm+0.015
Neg logit contributions
 pouch-0.011
 goggles-0.011
 couple-0.011
 plunge-0.011
gers-0.011
Token please
Token ID3387
Feature activation+4.39e-03

Pos logit contributions
 pouch+0.014
 couple+0.014
gers+0.014
 goggles+0.013
 banks+0.013
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.007
rou-0.007
tr-0.007
Token,
Token ID11
Feature activation-6.58e-03

Pos logit contributions
Acc+0.008
 Acc+0.008
rou+0.007
 collapse+0.007
tr+0.006
Neg logit contributions
 pouch-0.006
 couple-0.006
 goggles-0.006
gers-0.006
 plunge-0.006
Token please
Token ID3387
Feature activation+3.33e-03

Pos logit contributions
 pouch+0.012
 couple+0.012
gers+0.012
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.007
tr-0.007
Token?"
Token ID1701
Feature activation-4.19e-02

Pos logit contributions
 Acc+0.006
Acc+0.006
rou+0.005
 collapse+0.005
tr+0.005
Neg logit contributions
 pouch-0.005
 couple-0.005
 goggles-0.005
gers-0.005
 plunge-0.005
Token they
Token ID484
Feature activation-1.07e-01

Pos logit contributions
 pouch+0.096
 plunge+0.091
gers+0.091
 couple+0.090
ner+0.090
Neg logit contributions
 Acc-0.044
Acc-0.034
 Tom-0.029
 Tim-0.028
 Sam-0.028
Token asked
Token ID1965
Feature activation-2.89e-03

Pos logit contributions
gers+0.212
ippers+0.212
angers+0.204
rob+0.202
robe+0.201
Neg logit contributions
 Tom-0.108
 fixed-0.105
 over-0.104
 pressed-0.100
 Tim-0.100
Token.
Token ID13
Feature activation-2.08e-02

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.005
Acc-0.004
 collapse-0.004
rou-0.004
tr-0.004
Token
Token ID198
Feature activation+4.97e-03

Pos logit contributions
 pouch+0.042
 couple+0.041
gers+0.041
 plunge+0.040
ner+0.040
Neg logit contributions
 Acc-0.027
Acc-0.024
 collapse-0.019
tr-0.019
 over-0.018
Token
Token ID198
Feature activation-3.49e-03

Pos logit contributions
 Acc+0.010
Acc+0.009
 collapse+0.008
rou+0.008
tr+0.008
Neg logit contributions
 couple-0.007
 pouch-0.007
 goggles-0.006
gers-0.006
 plunge-0.006
TokenMom
Token ID29252
Feature activation+1.63e-02

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token is
Token ID318
Feature activation+9.96e-03

Pos logit contributions
Acc+0.063
 Acc+0.062
rou+0.053
 collapse+0.050
tr+0.050
Neg logit contributions
 plunge-0.030
 bag-0.030
 goggles-0.030
 pair-0.029
gers-0.029
Token our
Token ID674
Feature activation+1.51e-02

Pos logit contributions
 Acc+0.017
Acc+0.016
rou+0.014
tr+0.014
 collapse+0.013
Neg logit contributions
 pouch-0.016
 couple-0.016
 goggles-0.016
 plunge-0.016
 pills-0.015
Token rad
Token ID2511
Feature activation-2.44e-02

Pos logit contributions
 Acc+0.031
Acc+0.031
 collapse+0.027
rou+0.027
tr+0.027
Neg logit contributions
 pouch-0.019
 couple-0.018
 banks-0.017
 penn-0.017
 bag-0.017
Tokenish
Token ID680
Feature activation-3.73e-04

Pos logit contributions
 couple+0.039
gers+0.038
 goggles+0.037
 therapist+0.037
ner+0.036
Neg logit contributions
 Acc-0.042
Acc-0.040
rou-0.035
 collapse-0.032
tr-0.032
Token?"
Token ID1701
Feature activation-3.10e-02

Pos logit contributions
 pouch+0.001
 couple+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.000
rou-0.000
tr-0.000
Token they
Token ID484
Feature activation-1.05e-01

Pos logit contributions
 pouch+0.068
 plunge+0.067
 sleeves+0.065
 banks+0.065
 kernels+0.064
Neg logit contributions
 Acc-0.034
Acc-0.027
 collapse-0.026
 solved-0.021
 over-0.021
Token asked
Token ID1965
Feature activation+1.25e-03

Pos logit contributions
gers+0.227
angers+0.210
robe+0.210
ancer+0.207
 therapist+0.205
Neg logit contributions
 Tom-0.095
 over-0.095
 mad-0.090
 Tim-0.089
 collapse-0.088
Token.
Token ID13
Feature activation-9.55e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
rou+0.002
 collapse+0.002
tr+0.002
Neg logit contributions
 couple-0.002
 pouch-0.002
gers-0.002
 goggles-0.002
 plunge-0.002
Token "
Token ID366
Feature activation+1.78e-03

Pos logit contributions
 pouch+0.015
 couple+0.014
gers+0.014
 goggles+0.014
 plunge+0.014
Neg logit contributions
 Acc-0.017
Acc-0.016
 collapse-0.014
tr-0.013
 solved-0.013
TokenWho
Token ID8241
Feature activation+3.56e-02

Pos logit contributions
 Acc+0.003
Acc+0.003
rou+0.002
 collapse+0.002
 solved+0.002
Neg logit contributions
 couple-0.003
 pouch-0.003
gers-0.003
 goggles-0.003
 banks-0.003
Token took
Token ID1718
Feature activation+1.78e-02

Pos logit contributions
Acc+0.076
 Acc+0.075
rou+0.062
 collapse+0.059
 unavoidable+0.058
Neg logit contributions
 goggles-0.045
 pouch-0.043
gers-0.043
 bag-0.043
 pair-0.041
Token we
Token ID356
Feature activation-5.79e-03

Pos logit contributions
 couple+0.024
 pouch+0.024
gers+0.023
 goggles+0.023
 banks+0.022
Neg logit contributions
 Acc-0.023
Acc-0.021
 collapse-0.018
tr-0.017
 solved-0.017
Token go
Token ID467
Feature activation-2.78e-02

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.009
Acc-0.008
 collapse-0.007
rou-0.006
tr-0.006
Token,
Token ID11
Feature activation-3.26e-02

Pos logit contributions
 therapist+0.051
 couple+0.051
 pouch+0.050
 sleeves+0.050
 kernels+0.049
Neg logit contributions
 Acc-0.036
Acc-0.034
 over-0.032
 solved-0.029
rou-0.028
Token dad
Token ID9955
Feature activation-1.68e-02

Pos logit contributions
 pouch+0.067
gers+0.066
ippers+0.065
 banks+0.064
 sleeves+0.064
Neg logit contributions
 Acc-0.041
Acc-0.035
 over-0.030
 collapse-0.030
 solved-0.028
Token?"
Token ID1701
Feature activation-5.59e-02

Pos logit contributions
 pouch+0.030
 goggles+0.029
 couple+0.029
gers+0.029
 sleeves+0.029
Neg logit contributions
 Acc-0.025
Acc-0.023
 solved-0.019
rou-0.019
 collapse-0.019
Token they
Token ID484
Feature activation-1.04e-01

Pos logit contributions
 pouch+0.146
 plunge+0.143
ippers+0.142
 banks+0.141
 kernels+0.141
Neg logit contributions
 Acc-0.037
 Tim-0.028
Acc-0.025
 Sarah-0.021
 Tom-0.021
Token asked
Token ID1965
Feature activation-9.99e-03

Pos logit contributions
gers+0.223
ippers+0.220
robe+0.212
 pouch+0.207
angers+0.206
Neg logit contributions
 Tom-0.094
 Tim-0.094
 fixed-0.094
 over-0.094
 Sarah-0.093
Token.
Token ID13
Feature activation-3.96e-02

Pos logit contributions
 pouch+0.018
 couple+0.017
gers+0.017
 goggles+0.017
 plunge+0.017
Neg logit contributions
 Acc-0.015
Acc-0.015
 collapse-0.012
rou-0.012
 solved-0.011
Token
Token ID198
Feature activation-5.89e-03

Pos logit contributions
 plunge+0.077
 pouch+0.077
gers+0.077
 banks+0.076
 kernels+0.076
Neg logit contributions
 Acc-0.052
Acc-0.044
 over-0.039
tr-0.033
 solved-0.033
Token
Token ID198
Feature activation-1.62e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.008
 plunge+0.008
Neg logit contributions
 Acc-0.011
Acc-0.010
 collapse-0.009
rou-0.008
tr-0.008
Token"
Token ID1
Feature activation-3.27e-02

Pos logit contributions
 pouch+0.026
 couple+0.025
 plunge+0.025
gers+0.025
 goggles+0.025
Neg logit contributions
 Acc-0.027
Acc-0.025
 solved-0.021
 collapse-0.021
tr-0.020
Token?
Token ID30
Feature activation-1.20e-02

Pos logit contributions
 couple+0.002
 pouch+0.002
gers+0.002
 goggles+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.002
Acc-0.002
 collapse-0.002
rou-0.002
tr-0.002
Token Can
Token ID1680
Feature activation-8.94e-03

Pos logit contributions
 couple+0.020
 pouch+0.020
gers+0.019
 goggles+0.019
 plunge+0.019
Neg logit contributions
 Acc-0.020
Acc-0.018
 collapse-0.015
tr-0.015
rou-0.015
Token we
Token ID356
Feature activation-4.44e-03

Pos logit contributions
 couple+0.014
 pouch+0.014
gers+0.014
 goggles+0.014
 banks+0.013
Neg logit contributions
 Acc-0.015
Acc-0.014
 collapse-0.012
rou-0.012
tr-0.012
Token go
Token ID467
Feature activation-1.68e-02

Pos logit contributions
 pouch+0.008
 couple+0.008
gers+0.008
 goggles+0.008
 plunge+0.008
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token?"
Token ID1701
Feature activation-5.02e-02

Pos logit contributions
 couple+0.029
 pouch+0.029
gers+0.028
 therapist+0.028
 goggles+0.028
Neg logit contributions
 Acc-0.025
Acc-0.024
 over-0.020
 collapse-0.020
rou-0.020
Token they
Token ID484
Feature activation-1.03e-01

Pos logit contributions
 pouch+0.120
 plunge+0.118
 banks+0.118
gers+0.117
ve+0.116
Neg logit contributions
 Acc-0.043
 Tim-0.034
 Tom-0.032
 Sam-0.032
Acc-0.030
Token asked
Token ID1965
Feature activation-5.27e-03

Pos logit contributions
gers+0.227
ippers+0.216
robe+0.209
 plunge+0.207
 pouch+0.207
Neg logit contributions
 Tom-0.095
 over-0.092
 Lisa-0.088
 Tim-0.087
 pressed-0.086
Token.
Token ID13
Feature activation-2.70e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.008
Acc-0.008
 collapse-0.007
rou-0.006
tr-0.006
Token
Token ID198
Feature activation+5.20e-03

Pos logit contributions
 plunge+0.049
gers+0.049
 pouch+0.049
 couple+0.049
 banks+0.049
Neg logit contributions
 Acc-0.039
Acc-0.036
 over-0.029
tr-0.029
 collapse-0.028
Token
Token ID198
Feature activation-5.71e-03

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.009
rou+0.009
tr+0.008
Neg logit contributions
 couple-0.007
 pouch-0.007
gers-0.007
 goggles-0.006
 plunge-0.006
TokenMom
Token ID29252
Feature activation+1.51e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.008
 goggles+0.008
 plunge+0.008
Neg logit contributions
 Acc-0.010
Acc-0.010
 collapse-0.008
rou-0.008
tr-0.008
Token the
Token ID262
Feature activation+1.76e-02

Pos logit contributions
 Acc+0.008
Acc+0.008
rou+0.007
 collapse+0.006
tr+0.006
Neg logit contributions
 couple-0.007
 pouch-0.007
 goggles-0.007
 plunge-0.007
gers-0.007
Token park
Token ID3952
Feature activation+1.50e-02

Pos logit contributions
 Acc+0.039
Acc+0.038
rou+0.032
 collapse+0.031
sm+0.031
Neg logit contributions
 pouch-0.022
 couple-0.021
 bag-0.020
gers-0.020
 goggles-0.020
Token,
Token ID11
Feature activation-1.00e-02

Pos logit contributions
 Acc+0.027
Acc+0.026
 collapse+0.022
rou+0.021
tr+0.020
Neg logit contributions
 couple-0.023
 pouch-0.023
gers-0.023
 goggles-0.022
 pair-0.022
Token mom
Token ID1995
Feature activation+8.16e-03

Pos logit contributions
 pouch+0.018
 couple+0.017
gers+0.017
 goggles+0.017
 banks+0.017
Neg logit contributions
 Acc-0.015
Acc-0.014
 collapse-0.012
rou-0.012
tr-0.012
Token?"
Token ID1701
Feature activation-3.27e-02

Pos logit contributions
 Acc+0.015
Acc+0.014
 collapse+0.012
rou+0.012
tr+0.012
Neg logit contributions
 couple-0.012
 pouch-0.012
gers-0.012
 goggles-0.011
 pair-0.011
Token they
Token ID484
Feature activation-1.02e-01

Pos logit contributions
 pouch+0.059
gers+0.059
 plunge+0.058
 banks+0.058
 kernels+0.057
Neg logit contributions
 Acc-0.050
Acc-0.044
 collapse-0.037
 over-0.036
 solved-0.035
Token asked
Token ID1965
Feature activation-8.52e-03

Pos logit contributions
gers+0.234
ippers+0.215
 plunge+0.208
robe+0.204
 pouch+0.203
Neg logit contributions
 over-0.093
 fixed-0.091
 Tom-0.089
 Sarah-0.088
 Tim-0.085
Token.
Token ID13
Feature activation-2.47e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.014
 goggles+0.014
 plunge+0.014
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.010
rou-0.010
 solved-0.010
Token "
Token ID366
Feature activation-2.30e-02

Pos logit contributions
 plunge+0.044
gers+0.044
 pouch+0.043
 banks+0.043
 goggles+0.042
Neg logit contributions
 Acc-0.038
Acc-0.035
tr-0.029
 over-0.028
 collapse-0.028
TokenPlease
Token ID5492
Feature activation+7.64e-03

Pos logit contributions
 pouch+0.043
 banks+0.042
 plunge+0.042
gers+0.041
 pills+0.041
Neg logit contributions
Acc-0.032
 Acc-0.032
tr-0.023
 collapse-0.023
rou-0.023
Token,
Token ID11
Feature activation-7.75e-04

Pos logit contributions
Acc+0.016
 Acc+0.016
rou+0.013
 collapse+0.013
 solved+0.013
Neg logit contributions
 pouch-0.009
gers-0.009
 goggles-0.009
 couple-0.009
 plunge-0.009
Tokenppy
Token ID14097
Feature activation-3.65e-03

Pos logit contributions
 Acc+0.014
Acc+0.013
 collapse+0.012
tr+0.011
rou+0.011
Neg logit contributions
 pouch-0.008
 couple-0.008
gers-0.008
 goggles-0.008
ippers-0.008
Token and
Token ID290
Feature activation-2.28e-02

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token one
Token ID530
Feature activation+1.04e-02

Pos logit contributions
 pouch+0.040
gers+0.039
ippers+0.039
 couple+0.039
 sleeves+0.039
Neg logit contributions
 Acc-0.036
Acc-0.034
 solved-0.028
 collapse-0.027
sm-0.026
Token called
Token ID1444
Feature activation-3.40e-02

Pos logit contributions
 Acc+0.022
Acc+0.022
 collapse+0.019
rou+0.018
 solved+0.018
Neg logit contributions
 pouch-0.013
 couple-0.013
 pair-0.012
 goggles-0.012
gers-0.012
Token Cotton
Token ID23915
Feature activation-3.19e-02

Pos logit contributions
 couple+0.048
 sleeves+0.048
gers+0.046
 pouch+0.046
 goggles+0.045
Neg logit contributions
 Acc-0.062
Acc-0.056
 over-0.050
tr-0.050
rou-0.049
Token.
Token ID13
Feature activation-1.02e-01

Pos logit contributions
 plunge+0.061
gers+0.061
 sleeves+0.060
 couple+0.060
 goggles+0.060
Neg logit contributions
 Acc-0.041
Acc-0.038
rou-0.032
 over-0.031
 solved-0.031
Token They
Token ID1119
Feature activation-2.92e-02

Pos logit contributions
iest+0.210
 plunge+0.194
kins+0.190
 pair+0.183
fs+0.183
Neg logit contributions
 over-0.109
 Acc-0.091
 Tim-0.086
 Sam-0.081
 Good-0.081
Token lived
Token ID5615
Feature activation-1.27e-02

Pos logit contributions
ippers+0.049
 pouch+0.048
 couple+0.048
gers+0.048
 therapist+0.048
Neg logit contributions
 Acc-0.046
Acc-0.044
 solved-0.039
 over-0.037
 resolved-0.036
Token in
Token ID287
Feature activation-3.66e-02

Pos logit contributions
 couple+0.020
 pouch+0.020
gers+0.020
 plunge+0.020
 goggles+0.019
Neg logit contributions
 Acc-0.021
Acc-0.020
rou-0.017
 solved-0.017
 collapse-0.017
Token a
Token ID257
Feature activation-3.25e-03

Pos logit contributions
ippers+0.073
 plunge+0.073
gers+0.073
 sleeves+0.073
 couple+0.071
Neg logit contributions
 Acc-0.044
Acc-0.040
 solved-0.035
 over-0.033
 collapse-0.033
Token green
Token ID4077
Feature activation-2.38e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
Token go
Token ID467
Feature activation-2.39e-02

Pos logit contributions
 pouch+0.000
 couple+0.000
 goggles+0.000
gers+0.000
 plunge+0.000
Neg logit contributions
 Acc-0.000
Acc-0.000
 collapse-0.000
rou-0.000
tr-0.000
Token to
Token ID284
Feature activation-5.94e-03

Pos logit contributions
 therapist+0.045
 couple+0.045
 pouch+0.044
 sleeves+0.044
gers+0.044
Neg logit contributions
 Acc-0.030
Acc-0.028
 over-0.026
 solved-0.024
 collapse-0.023
Token the
Token ID262
Feature activation-3.08e-03

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.009
Acc-0.009
 collapse-0.008
rou-0.007
tr-0.007
Token zoo
Token ID26626
Feature activation-1.20e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token?"
Token ID1701
Feature activation-3.80e-02

Pos logit contributions
 pouch+0.021
 couple+0.020
 goggles+0.020
gers+0.020
 sleeves+0.020
Neg logit contributions
 Acc-0.018
Acc-0.017
 collapse-0.015
rou-0.014
tr-0.014
Token they
Token ID484
Feature activation-1.01e-01

Pos logit contributions
 pouch+0.080
 banks+0.079
 plunge+0.079
gers+0.078
 sleeves+0.077
Neg logit contributions
 Acc-0.044
Acc-0.035
 over-0.032
 collapse-0.032
 Sarah-0.030
Token said
Token ID531
Feature activation-1.48e-02

Pos logit contributions
gers+0.214
ippers+0.203
 pouch+0.197
 plunge+0.192
robe+0.189
Neg logit contributions
 fixed-0.111
 over-0.106
 Sarah-0.099
 Tim-0.098
 solved-0.097
Token.
Token ID13
Feature activation-3.44e-02

Pos logit contributions
 pouch+0.027
 couple+0.027
gers+0.026
 goggles+0.026
 banks+0.026
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.017
rou-0.016
 solved-0.016
Token
Token ID198
Feature activation-4.10e-03

Pos logit contributions
gers+0.066
 couple+0.066
 pouch+0.066
 plunge+0.065
 banks+0.065
Neg logit contributions
 Acc-0.046
Acc-0.039
 over-0.035
 collapse-0.033
tr-0.032
Token
Token ID198
Feature activation-2.92e-03

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
TokenMom
Token ID29252
Feature activation-4.47e-03

Pos logit contributions
 couple+0.005
 pouch+0.005
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token a
Token ID257
Feature activation-1.04e-02

Pos logit contributions
 couple+0.004
 pouch+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token balloon
Token ID21190
Feature activation-1.88e-03

Pos logit contributions
gers+0.014
 pouch+0.014
 couple+0.014
 goggles+0.014
 banks+0.014
Neg logit contributions
 Acc-0.019
Acc-0.019
 collapse-0.016
tr-0.016
rou-0.016
Token,
Token ID11
Feature activation-2.88e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.003
 solved-0.003
Token mom
Token ID1995
Feature activation+4.76e-03

Pos logit contributions
 pouch+0.056
gers+0.056
 banks+0.055
 sleeves+0.054
 plunge+0.054
Neg logit contributions
 Acc-0.037
Acc-0.033
 collapse-0.028
 over-0.028
tr-0.027
Token?"
Token ID1701
Feature activation-4.98e-02

Pos logit contributions
 Acc+0.008
Acc+0.008
 collapse+0.007
rou+0.006
tr+0.006
Neg logit contributions
 couple-0.007
 pouch-0.007
gers-0.007
 goggles-0.007
 pair-0.007
Token they
Token ID484
Feature activation-1.01e-01

Pos logit contributions
 pouch+0.116
 plunge+0.113
 banks+0.112
ippers+0.112
gers+0.111
Neg logit contributions
 Acc-0.046
Acc-0.035
 Tom-0.034
 Tim-0.031
 Lisa-0.030
Token asked
Token ID1965
Feature activation-1.36e-02

Pos logit contributions
gers+0.219
ippers+0.207
robe+0.202
 pouch+0.200
 plunge+0.199
Neg logit contributions
 Tom-0.090
 over-0.089
 fixed-0.085
 pressed-0.085
 Lisa-0.082
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
 pouch+0.025
 couple+0.025
gers+0.024
 goggles+0.024
 plunge+0.024
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.015
rou-0.015
 solved-0.015
Token
Token ID198
Feature activation+1.64e-02

Pos logit contributions
 pouch+0.037
 plunge+0.037
gers+0.036
 couple+0.036
 banks+0.036
Neg logit contributions
 Acc-0.033
Acc-0.031
 collapse-0.025
 over-0.025
tr-0.025
Token
Token ID198
Feature activation+2.82e-03

Pos logit contributions
 Acc+0.035
Acc+0.035
rou+0.031
 collapse+0.030
tr+0.030
Neg logit contributions
 couple-0.018
 pouch-0.018
gers-0.017
 goggles-0.017
ippers-0.017
Token"
Token ID1
Feature activation-2.24e-02

Pos logit contributions
 Acc+0.006
Acc+0.005
rou+0.005
 collapse+0.005
tr+0.005
Neg logit contributions
 couple-0.004
 pouch-0.004
gers-0.004
 goggles-0.004
 therapist-0.003
Token lot
Token ID1256
Feature activation-3.16e-02

Pos logit contributions
 couple+0.018
 pouch+0.017
gers+0.017
 goggles+0.017
 plunge+0.017
Neg logit contributions
 Acc-0.017
Acc-0.017
 collapse-0.014
rou-0.014
 solved-0.013
Tokenion
Token ID295
Feature activation-2.33e-02

Pos logit contributions
 couple+0.065
 goggles+0.063
 pouch+0.062
gers+0.062
 banks+0.061
Neg logit contributions
 Acc-0.037
Acc-0.035
rou-0.031
sm-0.029
 over-0.028
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
 couple+0.043
 pouch+0.042
 goggles+0.042
 plunge+0.041
gers+0.041
Neg logit contributions
 Acc-0.032
Acc-0.030
tr-0.025
 collapse-0.025
rou-0.025
Token Anna
Token ID11735
Feature activation-2.09e-02

Pos logit contributions
 pouch+0.043
 plunge+0.042
gers+0.042
 couple+0.042
 goggles+0.042
Neg logit contributions
 Acc-0.034
Acc-0.031
 over-0.025
tr-0.025
 collapse-0.024
Token liked
Token ID8288
Feature activation-4.05e-02

Pos logit contributions
gers+0.031
 pouch+0.030
 plunge+0.030
 goggles+0.030
 couple+0.030
Neg logit contributions
 Acc-0.038
Acc-0.036
 solved-0.031
rou-0.030
 collapse-0.030
Token the
Token ID262
Feature activation-9.99e-02

Pos logit contributions
ippers+0.080
gers+0.080
 goggles+0.076
 plunge+0.076
 couple+0.075
Neg logit contributions
 Acc-0.053
Acc-0.051
 over-0.042
sm-0.040
rou-0.040
Token bottle
Token ID9294
Feature activation-1.29e-02

Pos logit contributions
sters+0.198
 goggles+0.186
ippers+0.181
 crew+0.181
 couple+0.181
Neg logit contributions
 over-0.119
 Acc-0.104
tr-0.101
 better-0.097
 fixed-0.095
Token so
Token ID523
Feature activation-2.87e-03

Pos logit contributions
 couple+0.020
 pouch+0.020
 goggles+0.019
gers+0.019
 plunge+0.019
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.018
rou-0.018
 solved-0.018
Token her
Token ID607
Feature activation-5.67e-02

Pos logit contributions
 couple+0.006
 pouch+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.004
Acc-0.003
 collapse-0.003
rou-0.002
tr-0.002
Token mom
Token ID1995
Feature activation-2.55e-03

Pos logit contributions
ner+0.126
 donkey+0.125
 goggles+0.124
 plunge+0.122
rob+0.121
Neg logit contributions
 Acc-0.060
Acc-0.046
tr-0.042
rou-0.037
 over-0.037
Token bought
Token ID5839
Feature activation-5.05e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token some
Token ID617
Feature activation-1.94e-02

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.001
rou+0.001
tr+0.001
Neg logit contributions
 couple-0.001
 pouch-0.001
 goggles-0.001
gers-0.001
 plunge-0.001
Token cocoa
Token ID35845
Feature activation-1.51e-02

Pos logit contributions
 pouch+0.033
 couple+0.033
gers+0.032
 banks+0.031
 goggles+0.031
Neg logit contributions
 Acc-0.031
Acc-0.028
 collapse-0.024
rou-0.024
 over-0.023
Token,
Token ID11
Feature activation-1.60e-02

Pos logit contributions
 pouch+0.025
 couple+0.025
gers+0.024
 goggles+0.024
 plunge+0.024
Neg logit contributions
 Acc-0.024
Acc-0.023
 collapse-0.019
tr-0.019
rou-0.019
Token please
Token ID3387
Feature activation+1.38e-02

Pos logit contributions
 pouch+0.034
gers+0.034
 couple+0.034
 therapist+0.033
 goggles+0.033
Neg logit contributions
 Acc-0.018
Acc-0.016
 collapse-0.013
tr-0.012
 solved-0.012
Token?"
Token ID1701
Feature activation-4.63e-02

Pos logit contributions
Acc+0.027
 Acc+0.026
rou+0.023
 collapse+0.022
tr+0.021
Neg logit contributions
 pouch-0.019
 couple-0.019
 goggles-0.018
 pair-0.018
 plunge-0.018
Token they
Token ID484
Feature activation-9.93e-02

Pos logit contributions
 pouch+0.115
 plunge+0.114
 banks+0.113
ippers+0.113
 kernels+0.112
Neg logit contributions
 Acc-0.035
 Tim-0.027
 Sarah-0.026
Acc-0.023
 Sam-0.021
Token asked
Token ID1965
Feature activation-8.59e-03

Pos logit contributions
gers+0.219
ippers+0.204
rob+0.201
 pouch+0.196
robe+0.196
Neg logit contributions
 over-0.093
 fixed-0.092
 Tim-0.092
 Acc-0.089
 Tom-0.089
Token their
Token ID511
Feature activation-1.21e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.015
 goggles+0.015
 plunge+0.015
Neg logit contributions
 Acc-0.013
Acc-0.012
 collapse-0.010
rou-0.010
 solved-0.009
Token mom
Token ID1995
Feature activation-7.71e-03

Pos logit contributions
 pouch+0.019
 couple+0.019
gers+0.019
ner+0.019
 goggles+0.019
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
rou-0.016
tr-0.015
Token.
Token ID13
Feature activation-2.68e-02

Pos logit contributions
 pouch+0.013
 couple+0.013
gers+0.013
 goggles+0.013
 plunge+0.013
Neg logit contributions
 Acc-0.012
Acc-0.012
 collapse-0.010
rou-0.009
tr-0.009
Token
Token ID198
Feature activation+7.91e-03

Pos logit contributions
 pouch+0.050
gers+0.050
 couple+0.050
 plunge+0.049
 kernels+0.049
Neg logit contributions
 Acc-0.039
Acc-0.034
 over-0.029
 collapse-0.027
 solved-0.026
Token some
Token ID617
Feature activation-2.53e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token gum
Token ID27142
Feature activation-2.04e-02

Pos logit contributions
 couple+0.037
 pouch+0.037
gers+0.037
 banks+0.037
 plunge+0.036
Neg logit contributions
 Acc-0.043
Acc-0.039
rou-0.035
 solved-0.035
 fixed-0.035
Token,
Token ID11
Feature activation-1.18e-02

Pos logit contributions
 couple+0.035
 pouch+0.034
 goggles+0.033
 plunge+0.033
gers+0.033
Neg logit contributions
 Acc-0.031
Acc-0.029
tr-0.025
 collapse-0.025
rou-0.024
Token please
Token ID3387
Feature activation+1.41e-02

Pos logit contributions
 pouch+0.020
 couple+0.019
gers+0.019
 goggles+0.019
 banks+0.019
Neg logit contributions
 Acc-0.019
Acc-0.018
 collapse-0.015
rou-0.015
 solved-0.014
Token?"
Token ID1701
Feature activation-2.91e-02

Pos logit contributions
Acc+0.028
 Acc+0.028
rou+0.023
 collapse+0.023
tr+0.022
Neg logit contributions
 couple-0.019
 pouch-0.019
 goggles-0.018
 pair-0.018
gers-0.018
Token they
Token ID484
Feature activation-9.90e-02

Pos logit contributions
 pouch+0.078
 plunge+0.076
ippers+0.075
 goggles+0.074
 banks+0.074
Neg logit contributions
 Acc-0.018
Acc-0.014
 collapse-0.009
 over-0.009
 solved-0.009
Token asked
Token ID1965
Feature activation+8.14e-03

Pos logit contributions
ippers+0.210
gers+0.206
robe+0.199
angers+0.195
iest+0.190
Neg logit contributions
 Tom-0.100
 Sarah-0.094
 fixed-0.093
 over-0.092
 Tim-0.092
Token.
Token ID13
Feature activation-6.65e-03

Pos logit contributions
 Acc+0.014
Acc+0.014
rou+0.012
 collapse+0.011
tr+0.011
Neg logit contributions
 couple-0.013
gers-0.013
 pouch-0.012
 goggles-0.012
 plunge-0.012
Token
Token ID198
Feature activation+1.41e-02

Pos logit contributions
 pouch+0.012
 couple+0.012
gers+0.012
 goggles+0.012
 plunge+0.012
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.007
tr-0.007
Token
Token ID198
Feature activation+8.84e-03

Pos logit contributions
 Acc+0.030
Acc+0.029
rou+0.026
 collapse+0.025
tr+0.025
Neg logit contributions
 couple-0.017
gers-0.016
 pouch-0.016
 goggles-0.016
ner-0.015
TokenThe
Token ID464
Feature activation-2.72e-02

Pos logit contributions
 Acc+0.017
Acc+0.016
rou+0.014
 collapse+0.014
tr+0.013
Neg logit contributions
 couple-0.013
gers-0.013
 pouch-0.013
 goggles-0.012
 banks-0.012
Token ice
Token ID4771
Feature activation-5.50e-03

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.011
Acc-0.011
 collapse-0.009
rou-0.009
tr-0.009
Token cream
Token ID8566
Feature activation+1.35e-03

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.009
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.007
Token,
Token ID11
Feature activation-6.21e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.002
rou+0.002
 solved+0.002
Neg logit contributions
 couple-0.002
 pouch-0.002
gers-0.002
 goggles-0.002
 plunge-0.002
Token please
Token ID3387
Feature activation+1.67e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.011
Acc-0.010
 collapse-0.009
rou-0.009
tr-0.009
Token?"
Token ID1701
Feature activation-1.82e-02

Pos logit contributions
Acc+0.034
 Acc+0.033
rou+0.029
 collapse+0.028
tr+0.027
Neg logit contributions
 couple-0.022
 pouch-0.021
 goggles-0.020
gers-0.020
 pair-0.020
Token they
Token ID484
Feature activation-9.89e-02

Pos logit contributions
 pouch+0.042
gers+0.041
 plunge+0.041
 couple+0.040
 sleeves+0.040
Neg logit contributions
 Acc-0.018
Acc-0.016
 collapse-0.013
 solved-0.011
 over-0.011
Token said
Token ID531
Feature activation+5.92e-03

Pos logit contributions
gers+0.212
ippers+0.207
robe+0.195
angers+0.195
 plunge+0.194
Neg logit contributions
 Tom-0.095
 over-0.095
 fixed-0.094
 apologized-0.088
 gone-0.086
Token.
Token ID13
Feature activation-8.26e-03

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.009
rou+0.008
tr+0.008
Neg logit contributions
 couple-0.009
 pouch-0.009
 goggles-0.009
gers-0.009
 plunge-0.009
Token
Token ID198
Feature activation+1.64e-02

Pos logit contributions
 pouch+0.013
 couple+0.013
gers+0.012
 goggles+0.012
 plunge+0.012
Neg logit contributions
 Acc-0.014
Acc-0.014
 collapse-0.012
tr-0.011
rou-0.011
Token
Token ID198
Feature activation+1.05e-02

Pos logit contributions
 Acc+0.035
Acc+0.034
rou+0.030
tr+0.030
 collapse+0.030
Neg logit contributions
 couple-0.018
 pouch-0.018
gers-0.017
 goggles-0.017
ner-0.017
TokenThe
Token ID464
Feature activation-1.49e-02

Pos logit contributions
 Acc+0.019
Acc+0.018
rou+0.016
tr+0.015
 collapse+0.015
Neg logit contributions
 couple-0.016
 pouch-0.016
gers-0.016
 goggles-0.016
 banks-0.015
Token can
Token ID460
Feature activation-1.89e-02

Pos logit contributions
Acc+0.029
 Acc+0.029
rou+0.025
 collapse+0.025
sm+0.023
Neg logit contributions
 goggles-0.020
 couple-0.020
 pouch-0.020
 pair-0.020
 plunge-0.020
Token we
Token ID356
Feature activation+8.96e-04

Pos logit contributions
 pouch+0.033
 couple+0.032
gers+0.032
 goggles+0.031
 banks+0.031
Neg logit contributions
 Acc-0.031
Acc-0.028
 collapse-0.024
 solved-0.022
tr-0.022
Token have
Token ID423
Feature activation+1.42e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.001
rou+0.001
tr+0.001
Neg logit contributions
 pouch-0.001
 couple-0.001
 goggles-0.001
gers-0.001
 plunge-0.001
Token spaghetti
Token ID48241
Feature activation-1.02e-02

Pos logit contributions
 Acc+0.003
Acc+0.003
 collapse+0.002
rou+0.002
tr+0.002
Neg logit contributions
 couple-0.002
 pouch-0.002
 goggles-0.002
gers-0.002
 plunge-0.002
Token?"
Token ID1701
Feature activation-4.28e-02

Pos logit contributions
 pouch+0.017
 couple+0.017
gers+0.017
 goggles+0.017
 plunge+0.017
Neg logit contributions
 Acc-0.016
Acc-0.015
 collapse-0.013
rou-0.012
tr-0.012
Token they
Token ID484
Feature activation-9.81e-02

Pos logit contributions
 pouch+0.101
gers+0.094
 plunge+0.093
ancer+0.092
 sleeves+0.092
Neg logit contributions
 Acc-0.044
Acc-0.035
 collapse-0.029
 over-0.028
 Lisa-0.027
Token asked
Token ID1965
Feature activation-5.42e-03

Pos logit contributions
gers+0.204
ippers+0.192
 pouch+0.189
rob+0.186
robe+0.182
Neg logit contributions
 over-0.108
 Acc-0.105
Acc-0.100
 fixed-0.099
 Tim-0.093
Token.
Token ID13
Feature activation-1.54e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.009
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.007
Token
Token ID198
Feature activation+1.54e-02

Pos logit contributions
 pouch+0.027
 couple+0.027
gers+0.026
 plunge+0.026
 goggles+0.026
Neg logit contributions
 Acc-0.024
Acc-0.022
 collapse-0.019
tr-0.018
 solved-0.018
Token
Token ID198
Feature activation+7.62e-03

Pos logit contributions
 Acc+0.033
Acc+0.032
rou+0.029
 collapse+0.028
tr+0.028
Neg logit contributions
 couple-0.018
 pouch-0.017
 goggles-0.017
gers-0.017
ippers-0.016
Token"
Token ID1
Feature activation-2.72e-02

Pos logit contributions
 Acc+0.013
Acc+0.013
rou+0.011
 collapse+0.011
tr+0.010
Neg logit contributions
 couple-0.012
 pouch-0.012
 goggles-0.012
gers-0.012
 therapist-0.011
Token a
Token ID257
Feature activation+1.38e-02

Pos logit contributions
 Acc+0.029
Acc+0.027
 collapse+0.025
rou+0.024
tr+0.023
Neg logit contributions
 couple-0.018
 goggles-0.018
 pouch-0.017
gers-0.017
 pair-0.017
Token big
Token ID1263
Feature activation+1.53e-02

Pos logit contributions
 Acc+0.032
Acc+0.031
 collapse+0.029
rou+0.027
tr+0.027
Neg logit contributions
 couple-0.014
 pouch-0.013
 pair-0.013
gers-0.012
 snail-0.012
Token pig
Token ID12967
Feature activation-1.22e-02

Pos logit contributions
 Acc+0.034
Acc+0.034
 collapse+0.030
rou+0.029
tr+0.028
Neg logit contributions
 couple-0.017
 pouch-0.017
 bag-0.016
 pair-0.016
gers-0.016
Token.
Token ID13
Feature activation-1.50e-02

Pos logit contributions
 couple+0.022
 pouch+0.022
gers+0.022
 goggles+0.022
 plunge+0.022
Neg logit contributions
 Acc-0.018
Acc-0.017
tr-0.014
rou-0.014
 collapse-0.014
Token The
Token ID383
Feature activation-8.36e-02

Pos logit contributions
 couple+0.027
 pouch+0.027
gers+0.027
 plunge+0.027
 goggles+0.027
Neg logit contributions
 Acc-0.022
Acc-0.020
 collapse-0.017
tr-0.017
 solved-0.016
Token big
Token ID1263
Feature activation-9.80e-02

Pos logit contributions
ner+0.136
angers+0.121
asers+0.113
ippers+0.112
 trainers+0.111
Neg logit contributions
 fixed-0.131
 gone-0.128
Acc-0.121
 Acc-0.121
 ruined-0.120
Token pig
Token ID12967
Feature activation-1.16e-02

Pos logit contributions
angers+0.157
ner+0.151
ippers+0.149
fs+0.147
ancer+0.143
Neg logit contributions
 fixed-0.140
 gone-0.133
 ash-0.131
 snow-0.131
 our-0.130
Token had
Token ID550
Feature activation-2.87e-02

Pos logit contributions
 couple+0.019
 pouch+0.019
 plunge+0.018
 goggles+0.018
gers+0.018
Neg logit contributions
 Acc-0.019
Acc-0.018
tr-0.015
 collapse-0.015
 solved-0.015
Token a
Token ID257
Feature activation-2.44e-02

Pos logit contributions
ner+0.057
 therapist+0.056
ancer+0.056
ippers+0.056
gers+0.055
Neg logit contributions
Acc-0.034
 Acc-0.034
tr-0.031
 solved-0.030
 fixed-0.028
Token penny
Token ID26369
Feature activation-3.02e-02

Pos logit contributions
ner+0.036
gers+0.035
 plunge+0.034
 therapist+0.034
 banks+0.034
Neg logit contributions
 Acc-0.043
Acc-0.041
tr-0.037
 solved-0.035
 collapse-0.035
Token.
Token ID13
Feature activation-1.48e-02

Pos logit contributions
 couple+0.059
 plunge+0.058
 Holly+0.056
 goggles+0.056
 banks+0.056
Neg logit contributions
 Acc-0.037
Acc-0.035
tr-0.033
rou-0.031
 over-0.030
Token we
Token ID356
Feature activation+1.16e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.014
 goggles+0.014
 banks+0.014
Neg logit contributions
 Acc-0.016
Acc-0.016
 collapse-0.013
rou-0.013
tr-0.013
Token play
Token ID711
Feature activation+2.18e-03

Pos logit contributions
 Acc+0.021
Acc+0.021
rou+0.017
 collapse+0.017
sm+0.016
Neg logit contributions
 pouch-0.018
 couple-0.017
 goggles-0.017
 banks-0.017
 plunge-0.017
Token with
Token ID351
Feature activation+9.61e-03

Pos logit contributions
 Acc+0.004
Acc+0.004
 collapse+0.003
rou+0.003
tr+0.003
Neg logit contributions
 couple-0.003
 pouch-0.003
 goggles-0.003
gers-0.003
 plunge-0.003
Token you
Token ID345
Feature activation+1.37e-02

Pos logit contributions
 Acc+0.018
Acc+0.018
tr+0.015
 collapse+0.015
rou+0.015
Neg logit contributions
 goggles-0.014
 couple-0.013
 plunge-0.013
 pouch-0.013
gers-0.013
Token?"
Token ID1701
Feature activation-1.80e-02

Pos logit contributions
 Acc+0.025
Acc+0.024
 collapse+0.021
rou+0.020
tr+0.019
Neg logit contributions
 couple-0.020
 goggles-0.020
 pouch-0.020
gers-0.020
 pair-0.019
Token they
Token ID484
Feature activation-9.76e-02

Pos logit contributions
 pouch+0.038
gers+0.037
 sleeves+0.037
 couple+0.037
 plunge+0.037
Neg logit contributions
 Acc-0.021
Acc-0.019
 collapse-0.016
 solved-0.014
 over-0.014
Token asked
Token ID1965
Feature activation-5.81e-03

Pos logit contributions
gers+0.217
ippers+0.210
 plunge+0.198
robe+0.196
angers+0.194
Neg logit contributions
 over-0.101
 Tom-0.086
 fixed-0.085
 Sarah-0.084
 Tim-0.084
Token.
Token ID13
Feature activation-1.57e-02

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.009
Acc-0.009
 collapse-0.007
rou-0.007
tr-0.007
Token
Token ID198
Feature activation+3.73e-03

Pos logit contributions
 pouch+0.025
 couple+0.025
gers+0.025
 plunge+0.025
 sleeves+0.025
Neg logit contributions
 Acc-0.026
Acc-0.024
 collapse-0.020
tr-0.020
 solved-0.020
Token
Token ID198
Feature activation+2.89e-03

Pos logit contributions
 Acc+0.007
Acc+0.007
rou+0.006
 collapse+0.006
tr+0.006
Neg logit contributions
 couple-0.005
 pouch-0.005
 goggles-0.005
gers-0.005
 plunge-0.005
Token"
Token ID1
Feature activation-1.67e-02

Pos logit contributions
 Acc+0.005
Acc+0.005
rou+0.005
 collapse+0.004
tr+0.004
Neg logit contributions
 couple-0.004
 pouch-0.004
 goggles-0.004
gers-0.004
 banks-0.004
Token!"
Token ID2474
Feature activation-8.04e-03

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.008
tr-0.007
Token He
Token ID679
Feature activation+2.54e-04

Pos logit contributions
 pouch+0.013
 couple+0.013
gers+0.013
 goggles+0.013
 plunge+0.013
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.011
rou-0.010
tr-0.010
Token gave
Token ID2921
Feature activation-1.06e-02

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 couple-0.000
 pouch-0.000
gers-0.000
 goggles-0.000
 therapist-0.000
Token her
Token ID607
Feature activation-4.70e-02

Pos logit contributions
 pouch+0.018
 couple+0.018
gers+0.018
 goggles+0.018
 banks+0.017
Neg logit contributions
 Acc-0.016
Acc-0.015
 collapse-0.013
rou-0.012
 solved-0.012
Token all
Token ID477
Feature activation+7.58e-03

Pos logit contributions
ner+0.082
gers+0.080
ippers+0.080
angers+0.079
 sleeves+0.079
Neg logit contributions
 Acc-0.063
Acc-0.061
 over-0.059
sm-0.054
 fixed-0.054
Token the
Token ID262
Feature activation-9.76e-02

Pos logit contributions
 Acc+0.014
Acc+0.014
 collapse+0.012
tr+0.011
rou+0.011
Neg logit contributions
 couple-0.011
 pouch-0.011
 goggles-0.011
gers-0.011
 plunge-0.010
Token sa
Token ID473
Feature activation-5.81e-02

Pos logit contributions
angers+0.188
robe+0.164
ner+0.163
anger+0.161
izards+0.160
Neg logit contributions
 fixed-0.115
 better-0.110
 thanks-0.110
 cal-0.108
 over-0.102
Tokenus
Token ID385
Feature activation-5.77e-02

Pos logit contributions
 plunge+0.099
 goggles+0.096
 Holly+0.096
 couple+0.094
 snail+0.093
Neg logit contributions
 Acc-0.084
rou-0.076
Acc-0.073
ray-0.071
sm-0.070
Tokenages
Token ID1095
Feature activation-8.71e-04

Pos logit contributions
 plunge+0.075
ner+0.070
 Holly+0.070
gers+0.067
 pair+0.067
Neg logit contributions
 Acc-0.108
Acc-0.100
sm-0.096
ray-0.092
rou-0.090
Token and
Token ID290
Feature activation-2.15e-02

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.001
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token she
Token ID673
Feature activation-3.13e-03

Pos logit contributions
 pouch+0.033
gers+0.032
 plunge+0.032
ippers+0.032
 goggles+0.032
Neg logit contributions
 Acc-0.037
Acc-0.035
 solved-0.030
 collapse-0.030
rou-0.029
Token mom
Token ID1995
Feature activation+4.31e-03

Pos logit contributions
ner+0.054
gers+0.054
 goggles+0.054
 plunge+0.053
 banks+0.052
Neg logit contributions
 Acc-0.040
Acc-0.036
tr-0.032
rou-0.031
 solved-0.029
Tokenmy
Token ID1820
Feature activation-8.21e-03

Pos logit contributions
 Acc+0.005
Acc+0.004
 collapse+0.003
rou+0.003
tr+0.003
Neg logit contributions
 pouch-0.010
 couple-0.010
 pair-0.009
 sleeves-0.009
gers-0.009
Token started
Token ID2067
Feature activation-2.14e-02

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.011
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.016
Acc-0.015
 collapse-0.013
rou-0.013
 solved-0.013
Token to
Token ID284
Feature activation-2.17e-02

Pos logit contributions
gers+0.043
 pouch+0.042
 plunge+0.041
 goggles+0.041
 couple+0.041
Neg logit contributions
 Acc-0.028
Acc-0.027
 collapse-0.021
 over-0.021
rou-0.021
Token cut
Token ID2005
Feature activation-4.77e-02

Pos logit contributions
gers+0.037
 pouch+0.036
 couple+0.035
ippers+0.035
 goggles+0.035
Neg logit contributions
 Acc-0.036
Acc-0.033
 collapse-0.029
 solved-0.028
 fixed-0.027
Token the
Token ID262
Feature activation-9.76e-02

Pos logit contributions
 couple+0.089
 plunge+0.088
 goggles+0.086
 pouch+0.086
gers+0.084
Neg logit contributions
 over-0.058
 Acc-0.057
Acc-0.056
tr-0.050
 collapse-0.050
Token l
Token ID300
Feature activation-5.90e-02

Pos logit contributions
 crew+0.179
 goggles+0.176
ancer+0.174
 suitcase+0.173
ippers+0.173
Neg logit contributions
 over-0.105
tr-0.092
 pays-0.092
 solved-0.090
 finishes-0.088
Tokenimes
Token ID999
Feature activation-5.55e-03

Pos logit contributions
 plunge+0.086
 suitcase+0.086
 banks+0.085
 couple+0.084
 goggles+0.084
Neg logit contributions
Acc-0.091
sm-0.090
rou-0.089
 solved-0.085
 Acc-0.084
Token.
Token ID13
Feature activation-1.63e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.009
Acc-0.009
 collapse-0.007
rou-0.007
tr-0.007
Token Lily
Token ID20037
Feature activation-1.02e-02

Pos logit contributions
 pouch+0.028
 couple+0.028
gers+0.028
 goggles+0.028
 plunge+0.027
Neg logit contributions
 Acc-0.026
Acc-0.024
tr-0.020
 collapse-0.020
 solved-0.019
Token was
Token ID373
Feature activation-2.56e-02

Pos logit contributions
 pouch+0.016
 couple+0.016
gers+0.016
 goggles+0.016
 plunge+0.015
Neg logit contributions
 Acc-0.018
Acc-0.017
 collapse-0.014
rou-0.014
 solved-0.014
Token help
Token ID1037
Feature activation-2.59e-02

Pos logit contributions
gers+0.040
 pouch+0.039
 couple+0.038
 sleeves+0.038
 therapist+0.038
Neg logit contributions
 Acc-0.037
Acc-0.033
 collapse-0.030
 solved-0.030
 over-0.029
Token!
Token ID0
Feature activation-1.67e-02

Pos logit contributions
gers+0.049
 pouch+0.048
 goggles+0.048
 plunge+0.048
 couple+0.046
Neg logit contributions
 Acc-0.035
Acc-0.033
 collapse-0.029
 over-0.028
 solved-0.027
Token She
Token ID1375
Feature activation-1.97e-02

Pos logit contributions
gers+0.026
 couple+0.026
 goggles+0.026
 pouch+0.026
 plunge+0.026
Neg logit contributions
 Acc-0.029
Acc-0.027
tr-0.022
 collapse-0.022
 over-0.021
Token picked
Token ID6497
Feature activation-4.97e-02

Pos logit contributions
 pouch+0.031
gers+0.031
 plunge+0.030
 couple+0.030
ippers+0.030
Neg logit contributions
 Acc-0.033
Acc-0.032
 solved-0.028
rou-0.026
tr-0.026
Token up
Token ID510
Feature activation-4.65e-02

Pos logit contributions
 pouch+0.096
 sleeves+0.094
ippers+0.092
gers+0.092
 banks+0.091
Neg logit contributions
 over-0.059
Acc-0.055
 Acc-0.051
sm-0.047
 solved-0.044
Token the
Token ID262
Feature activation-9.74e-02

Pos logit contributions
 pouch+0.090
 plunge+0.090
ippers+0.089
ner+0.088
ancer+0.087
Neg logit contributions
Acc-0.053
 Acc-0.050
 over-0.048
sm-0.046
 fixed-0.045
Token bag
Token ID6131
Feature activation-1.32e-02

Pos logit contributions
ner+0.157
angers+0.134
 Holly+0.132
folk+0.131
unn+0.130
Neg logit contributions
 over-0.135
 pays-0.131
 fixed-0.119
sm-0.116
 what-0.113
Token,
Token ID11
Feature activation-2.77e-02

Pos logit contributions
 pouch+0.022
 couple+0.022
 goggles+0.022
gers+0.022
 plunge+0.021
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
rou-0.016
tr-0.016
Token and
Token ID290
Feature activation-1.47e-02

Pos logit contributions
gers+0.042
 pouch+0.040
ippers+0.039
 plunge+0.038
 banks+0.037
Neg logit contributions
Acc-0.053
 Acc-0.051
 solved-0.043
 over-0.043
 fixed-0.042
Token it
Token ID340
Feature activation-3.27e-02

Pos logit contributions
 pouch+0.023
gers+0.022
 plunge+0.022
 couple+0.022
 goggles+0.022
Neg logit contributions
 Acc-0.025
Acc-0.025
 solved-0.021
 collapse-0.021
rou-0.020
Token was
Token ID373
Feature activation-9.15e-03

Pos logit contributions
ippers+0.061
 plunge+0.060
 pouch+0.060
 couple+0.060
ancer+0.060
Neg logit contributions
 Acc-0.041
Acc-0.041
 solved-0.039
 fixed-0.037
tr-0.036
Token a
Token ID257
Feature activation+2.00e-02

Pos logit contributions
 Acc+0.019
Acc+0.018
rou+0.016
 collapse+0.016
tr+0.015
Neg logit contributions
 couple-0.009
 goggles-0.009
 pouch-0.009
gers-0.009
 plunge-0.009
Token cookie
Token ID19751
Feature activation+3.53e-03

Pos logit contributions
 Acc+0.041
Acc+0.039
 collapse+0.034
rou+0.033
 solved+0.032
Neg logit contributions
 couple-0.028
 pouch-0.027
 bag-0.027
 pair-0.026
 goggles-0.025
Token,
Token ID11
Feature activation-8.37e-04

Pos logit contributions
 Acc+0.006
Acc+0.006
 collapse+0.005
rou+0.005
 solved+0.005
Neg logit contributions
 couple-0.005
 pouch-0.005
gers-0.005
 goggles-0.005
 therapist-0.005
Token mom
Token ID1995
Feature activation+1.08e-02

Pos logit contributions
 couple+0.001
 pouch+0.001
 goggles+0.001
gers+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.002
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token?"
Token ID1701
Feature activation-2.79e-02

Pos logit contributions
 Acc+0.020
Acc+0.020
 collapse+0.017
rou+0.016
tr+0.016
Neg logit contributions
 couple-0.016
 pouch-0.015
gers-0.015
 pair-0.015
 goggles-0.015
Token they
Token ID484
Feature activation-9.69e-02

Pos logit contributions
 pouch+0.062
 plunge+0.060
 kernels+0.060
gers+0.059
 banks+0.059
Neg logit contributions
 Acc-0.032
Acc-0.027
 collapse-0.021
 over-0.020
 solved-0.020
Token asked
Token ID1965
Feature activation-4.40e-03

Pos logit contributions
gers+0.216
ippers+0.199
 plunge+0.197
robe+0.194
rob+0.190
Neg logit contributions
 Tom-0.090
 over-0.089
 fixed-0.084
 Lisa-0.083
 gone-0.079
Token.
Token ID13
Feature activation-9.29e-03

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.005
Token
Token ID198
Feature activation+2.34e-02

Pos logit contributions
 pouch+0.014
 couple+0.014
gers+0.014
 goggles+0.014
 plunge+0.014
Neg logit contributions
 Acc-0.016
Acc-0.015
 collapse-0.013
tr-0.013
rou-0.013
Token
Token ID198
Feature activation+1.18e-02

Pos logit contributions
 Acc+0.052
Acc+0.052
rou+0.047
tr+0.046
 collapse+0.046
Neg logit contributions
 pouch-0.023
 couple-0.022
 goggles-0.021
gers-0.021
 Holly-0.021
Token"
Token ID1
Feature activation-2.15e-02

Pos logit contributions
 Acc+0.025
Acc+0.025
rou+0.022
 collapse+0.021
tr+0.021
Neg logit contributions
 couple-0.014
 pouch-0.014
gers-0.014
 goggles-0.013
 sleeves-0.012
Token She
Token ID1375
Feature activation+7.47e-03

Pos logit contributions
 Acc+0.013
Acc+0.013
rou+0.012
 collapse+0.012
 solved+0.011
Neg logit contributions
 pouch-0.006
 couple-0.006
 goggles-0.006
gers-0.006
 banks-0.006
Token bought
Token ID5839
Feature activation-1.60e-02

Pos logit contributions
 Acc+0.014
Acc+0.013
 collapse+0.012
rou+0.011
tr+0.011
Neg logit contributions
 couple-0.011
 pouch-0.011
gers-0.010
 goggles-0.010
 therapist-0.010
Token him
Token ID683
Feature activation-2.44e-03

Pos logit contributions
 pouch+0.026
gers+0.026
 couple+0.025
 goggles+0.025
 banks+0.025
Neg logit contributions
 Acc-0.026
Acc-0.025
 collapse-0.020
rou-0.020
 solved-0.020
Token a
Token ID257
Feature activation+9.50e-04

Pos logit contributions
 couple+0.004
 pouch+0.004
gers+0.004
 goggles+0.004
 plunge+0.003
Neg logit contributions
 Acc-0.004
Acc-0.004
 collapse-0.004
rou-0.004
tr-0.003
Token pair
Token ID5166
Feature activation-3.15e-02

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.002
rou+0.002
 solved+0.001
Neg logit contributions
 pouch-0.001
 couple-0.001
 goggles-0.001
gers-0.001
 plunge-0.001
Token of
Token ID286
Feature activation-9.61e-02

Pos logit contributions
 banks+0.055
 pouch+0.053
 goggles+0.053
 sleeves+0.053
 plunge+0.053
Neg logit contributions
 Acc-0.045
Acc-0.042
 over-0.039
 solved-0.037
sm-0.037
Token new
Token ID649
Feature activation-8.57e-02

Pos logit contributions
 banks+0.141
ner+0.134
owers+0.131
ider+0.129
ancer+0.129
Neg logit contributions
 fixed-0.139
 Acc-0.131
 over-0.127
Acc-0.126
 how-0.124
Token shoes
Token ID10012
Feature activation-4.25e-02

Pos logit contributions
ner+0.111
ancer+0.107
 plunge+0.096
fters+0.095
owers+0.095
Neg logit contributions
 over-0.140
 fixed-0.137
 Acc-0.134
tr-0.127
rou-0.124
Token.
Token ID13
Feature activation-1.54e-03

Pos logit contributions
 plunge+0.083
 goggles+0.082
 couple+0.081
 pouch+0.081
 banks+0.081
Neg logit contributions
 Acc-0.050
Acc-0.045
 over-0.044
 fixed-0.040
tr-0.040
Token They
Token ID1119
Feature activation-2.95e-02

Pos logit contributions
 pouch+0.002
 couple+0.002
 goggles+0.002
gers+0.002
 banks+0.002
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.003
 solved-0.002
Token were
Token ID547
Feature activation-3.99e-02

Pos logit contributions
 pouch+0.051
gers+0.050
 plunge+0.050
ippers+0.050
 couple+0.050
Neg logit contributions
 Acc-0.043
Acc-0.042
 solved-0.037
 fixed-0.036
 over-0.036
Token end
Token ID886
Feature activation+3.59e-02

Pos logit contributions
 pouch+0.053
ippers+0.053
 plunge+0.053
 goggles+0.053
gers+0.052
Neg logit contributions
 Acc-0.042
Acc-0.037
rou-0.031
 collapse-0.031
 solved-0.030
Token.
Token ID13
Feature activation+2.56e-02

Pos logit contributions
Acc+0.085
 Acc+0.080
 collapse+0.068
rou+0.067
 unavoidable+0.066
Neg logit contributions
gers-0.034
 pills-0.034
 pouch-0.034
 therapist-0.033
 couple-0.033
Token<|endoftext|>
Token ID50256
Feature activation-2.12e-02

Pos logit contributions
 Acc+0.054
Acc+0.054
rou+0.048
 collapse+0.047
 solved+0.044
Neg logit contributions
 pouch-0.028
 couple-0.028
 therapist-0.026
ner-0.025
 Holly-0.025
TokenOnce
Token ID7454
Feature activation-5.44e-03

Pos logit contributions
 couple+0.031
 pouch+0.030
 goggles+0.030
gers+0.030
 plunge+0.029
Neg logit contributions
 Acc-0.038
Acc-0.036
 collapse-0.031
 solved-0.031
tr-0.030
Token upon
Token ID2402
Feature activation-5.05e-03

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.005
Neg logit contributions
 Acc-0.012
Acc-0.012
 collapse-0.010
rou-0.010
tr-0.010
Token a
Token ID257
Feature activation+3.80e-02

Pos logit contributions
 pouch+0.008
 couple+0.008
gers+0.008
 goggles+0.008
 plunge+0.007
Neg logit contributions
 Acc-0.009
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.007
Token time
Token ID640
Feature activation+6.66e-03

Pos logit contributions
 Acc+0.089
Acc+0.081
 collapse+0.077
rou+0.075
 resolved+0.072
Neg logit contributions
 couple-0.038
gers-0.035
 goggles-0.034
 pair-0.034
 pouch-0.033
Token there
Token ID612
Feature activation+6.99e-03

Pos logit contributions
 Acc+0.013
Acc+0.013
 collapse+0.011
rou+0.011
tr+0.011
Neg logit contributions
 pouch-0.008
 couple-0.008
 goggles-0.008
gers-0.008
 plunge-0.008
Token was
Token ID373
Feature activation+9.70e-03

Pos logit contributions
 Acc+0.013
Acc+0.013
rou+0.011
 collapse+0.011
tr+0.011
Neg logit contributions
 pouch-0.010
 couple-0.009
gers-0.009
 goggles-0.009
 plunge-0.009
Token a
Token ID257
Feature activation+2.03e-02

Pos logit contributions
 Acc+0.019
Acc+0.018
 collapse+0.016
rou+0.016
tr+0.015
Neg logit contributions
 couple-0.013
 goggles-0.013
 pouch-0.013
gers-0.013
 plunge-0.012
Token bad
Token ID2089
Feature activation+1.68e-03

Pos logit contributions
 Acc+0.051
Acc+0.049
 collapse+0.045
rou+0.043
tr+0.043
Neg logit contributions
 couple-0.018
 pouch-0.017
 pair-0.016
 snail-0.016
gers-0.015
Token how
Token ID703
Feature activation+3.04e-02

Pos logit contributions
 Acc+0.018
Acc+0.017
tr+0.014
 collapse+0.014
rou+0.013
Neg logit contributions
 goggles-0.029
gers-0.029
 couple-0.028
 banks-0.028
 pouch-0.028
Token to
Token ID284
Feature activation+1.17e-02

Pos logit contributions
 Acc+0.056
Acc+0.054
tr+0.050
 collapse+0.048
rou+0.045
Neg logit contributions
 goggles-0.039
gers-0.039
 legs-0.039
 couple-0.038
 plunge-0.037
Token fight
Token ID1907
Feature activation+4.13e-02

Pos logit contributions
 Acc+0.020
Acc+0.020
rou+0.016
 collapse+0.016
sm+0.016
Neg logit contributions
 pouch-0.018
 goggles-0.018
 banks-0.018
gers-0.017
 plunge-0.017
Token.
Token ID13
Feature activation+1.67e-02

Pos logit contributions
Acc+0.088
 Acc+0.088
rou+0.072
tr+0.069
sm+0.068
Neg logit contributions
 legs-0.049
gers-0.049
 ones-0.048
 people-0.047
 pieces-0.046
Token They
Token ID1119
Feature activation+2.11e-02

Pos logit contributions
Acc+0.036
 Acc+0.036
 collapse+0.032
rou+0.031
sm+0.030
Neg logit contributions
 pouch-0.018
 couple-0.017
 banks-0.017
gers-0.016
 goggles-0.016
Token fought
Token ID8350
Feature activation+5.77e-02

Pos logit contributions
 Acc+0.044
Acc+0.041
 collapse+0.035
rou+0.035
tr+0.034
Neg logit contributions
 couple-0.028
 pouch-0.027
 pair-0.026
 goggles-0.026
 banks-0.026
Token for
Token ID329
Feature activation+2.11e-02

Pos logit contributions
 Acc+0.117
Acc+0.115
sm+0.087
rou+0.087
tr+0.084
Neg logit contributions
 legs-0.081
 pieces-0.075
 people-0.075
gers-0.073
 banks-0.072
Token a
Token ID257
Feature activation+9.39e-03

Pos logit contributions
 Acc+0.042
Acc+0.040
rou+0.035
tr+0.034
 collapse+0.033
Neg logit contributions
 couple-0.028
 banks-0.027
 goggles-0.027
 pouch-0.027
 plunge-0.027
Token little
Token ID1310
Feature activation+9.55e-03

Pos logit contributions
 Acc+0.023
Acc+0.022
rou+0.020
 collapse+0.020
tr+0.019
Neg logit contributions
 couple-0.008
 pouch-0.008
 goggles-0.007
gers-0.007
 pair-0.007
Token while
Token ID981
Feature activation+7.52e-03

Pos logit contributions
 Acc+0.022
Acc+0.021
rou+0.018
 collapse+0.018
tr+0.018
Neg logit contributions
 couple-0.010
 pouch-0.010
gers-0.009
 plunge-0.009
 goggles-0.009
Token,
Token ID11
Feature activation+1.22e-02

Pos logit contributions
 Acc+0.015
Acc+0.014
 collapse+0.012
tr+0.012
rou+0.012
Neg logit contributions
 couple-0.010
 pouch-0.010
gers-0.010
 goggles-0.009
 sleeves-0.009
Token end
Token ID886
Feature activation+3.23e-02

Pos logit contributions
 Acc+0.015
Acc+0.015
 collapse+0.013
rou+0.012
 solved+0.012
Neg logit contributions
 couple-0.009
 pouch-0.009
gers-0.008
 goggles-0.008
 pair-0.008
Token.
Token ID13
Feature activation+1.23e-02

Pos logit contributions
Acc+0.076
 Acc+0.074
rou+0.061
 collapse+0.061
 unavoidable+0.060
Neg logit contributions
gers-0.034
 pair-0.031
 pills-0.031
 legs-0.031
 goggles-0.031
Token<|endoftext|>
Token ID50256
Feature activation-8.72e-03

Pos logit contributions
 Acc+0.025
Acc+0.024
 collapse+0.021
rou+0.021
 solved+0.020
Neg logit contributions
 pouch-0.016
 couple-0.016
 goggles-0.014
 Holly-0.014
gers-0.014
TokenOnce
Token ID7454
Feature activation-8.19e-03

Pos logit contributions
 couple+0.011
 pouch+0.011
gers+0.011
 goggles+0.011
 plunge+0.010
Neg logit contributions
 Acc-0.017
Acc-0.017
 collapse-0.014
rou-0.014
 solved-0.014
Token upon
Token ID2402
Feature activation-3.49e-03

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.008
Neg logit contributions
 Acc-0.018
Acc-0.017
 collapse-0.015
rou-0.015
 solved-0.015
Token a
Token ID257
Feature activation+3.90e-02

Pos logit contributions
 couple+0.005
 pouch+0.005
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token time
Token ID640
Feature activation+1.42e-02

Pos logit contributions
 Acc+0.094
Acc+0.085
 collapse+0.080
rou+0.078
 resolved+0.076
Neg logit contributions
 couple-0.037
gers-0.035
 pair-0.033
 goggles-0.033
 pouch-0.033
Token there
Token ID612
Feature activation+1.54e-02

Pos logit contributions
 Acc+0.030
Acc+0.030
rou+0.026
 collapse+0.026
tr+0.025
Neg logit contributions
 pouch-0.015
 plunge-0.015
 goggles-0.015
 couple-0.015
gers-0.015
Token were
Token ID547
Feature activation-4.27e-03

Pos logit contributions
 Acc+0.031
Acc+0.031
rou+0.027
 collapse+0.026
tr+0.025
Neg logit contributions
 pouch-0.019
gers-0.018
 couple-0.018
 plunge-0.018
 goggles-0.018
Token two
Token ID734
Feature activation-1.36e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.008
tr-0.008
Token little
Token ID1310
Feature activation-2.51e-02

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.009
Neg logit contributions
 Acc-0.034
Acc-0.033
rou-0.030
 collapse-0.030
 solved-0.029
Token was
Token ID373
Feature activation+2.19e-02

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token like
Token ID588
Feature activation+8.85e-03

Pos logit contributions
 Acc+0.048
Acc+0.046
tr+0.039
ray+0.038
 collapse+0.038
Neg logit contributions
 pouch-0.025
 couple-0.025
 pair-0.024
 goggles-0.023
gers-0.023
Token a
Token ID257
Feature activation+4.13e-02

Pos logit contributions
 Acc+0.017
Acc+0.016
 collapse+0.014
tr+0.014
rou+0.013
Neg logit contributions
 couple-0.012
 pouch-0.012
gers-0.012
 goggles-0.012
 plunge-0.012
Token dream
Token ID4320
Feature activation+3.87e-03

Pos logit contributions
 Acc+0.087
Acc+0.084
 collapse+0.070
 unavoidable+0.067
rou+0.066
Neg logit contributions
 couple-0.061
 pouch-0.055
 pair-0.054
 bag-0.051
 donkey-0.049
Token come
Token ID1282
Feature activation+3.26e-02

Pos logit contributions
 Acc+0.007
Acc+0.006
 collapse+0.005
rou+0.005
tr+0.005
Neg logit contributions
 couple-0.006
 pouch-0.006
gers-0.006
 plunge-0.006
 goggles-0.006
Token true
Token ID2081
Feature activation+3.81e-02

Pos logit contributions
 Acc+0.056
Acc+0.054
 collapse+0.040
 solved+0.039
rou+0.039
Neg logit contributions
 pouch-0.054
 couple-0.053
gers-0.052
 plunge-0.051
 pair-0.051
Token.
Token ID13
Feature activation+1.13e-02

Pos logit contributions
Acc+0.075
 Acc+0.075
 collapse+0.059
 unavoidable+0.054
tr+0.052
Neg logit contributions
 couple-0.059
 pouch-0.056
gers-0.056
 bag-0.055
 pair-0.055
Token<|endoftext|>
Token ID50256
Feature activation-1.99e-02

Pos logit contributions
 Acc+0.022
Acc+0.022
 collapse+0.020
rou+0.019
tr+0.019
Neg logit contributions
 pouch-0.015
 couple-0.014
gers-0.013
 Holly-0.013
 plunge-0.013
TokenOnce
Token ID7454
Feature activation-5.99e-03

Pos logit contributions
 couple+0.028
 pouch+0.028
 goggles+0.027
gers+0.027
 plunge+0.027
Neg logit contributions
 Acc-0.036
Acc-0.035
 collapse-0.030
 solved-0.030
rou-0.029
Token,
Token ID11
Feature activation+1.68e-02

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.011
rou-0.011
tr-0.011
Token there
Token ID612
Feature activation+2.70e-03

Pos logit contributions
 Acc+0.043
Acc+0.041
 collapse+0.038
rou+0.037
tr+0.037
Neg logit contributions
 pouch-0.012
 pair-0.011
 goggles-0.011
 couple-0.011
gers-0.011
Token outside
Token ID2354
Feature activation-1.34e-02

Pos logit contributions
 pouch+0.040
 couple+0.039
 plunge+0.039
gers+0.038
 banks+0.038
Neg logit contributions
 Acc-0.028
Acc-0.027
 solved-0.021
tr-0.021
 over-0.021
Token and
Token ID290
Feature activation-1.15e-02

Pos logit contributions
 pouch+0.022
 couple+0.022
gers+0.022
 plunge+0.021
 goggles+0.021
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.017
 over-0.017
tr-0.017
Token run
Token ID1057
Feature activation-2.64e-02

Pos logit contributions
 pouch+0.018
 couple+0.018
gers+0.018
 goggles+0.017
 therapist+0.017
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
rou-0.015
 solved-0.015
Token around
Token ID1088
Feature activation-1.48e-02

Pos logit contributions
 pouch+0.046
 therapist+0.045
 couple+0.045
ancer+0.044
 plunge+0.044
Neg logit contributions
 Acc-0.036
Acc-0.035
 over-0.035
 solved-0.032
sm-0.030
Token in
Token ID287
Feature activation+7.08e-03

Pos logit contributions
 pouch+0.027
 couple+0.026
gers+0.026
 goggles+0.026
 plunge+0.026
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.017
 over-0.016
 solved-0.016
Token the
Token ID262
Feature activation+4.27e-02

Pos logit contributions
 Acc+0.013
Acc+0.013
tr+0.011
rou+0.011
 collapse+0.011
Neg logit contributions
 couple-0.010
 pouch-0.010
 goggles-0.010
 pair-0.010
gers-0.010
Token park
Token ID3952
Feature activation-2.17e-02

Pos logit contributions
 Acc+0.106
Acc+0.099
 collapse+0.092
 resolved+0.091
rou+0.089
Neg logit contributions
 pouch-0.041
 bag-0.039
gers-0.039
 stream-0.038
 goggles-0.038
Token.
Token ID13
Feature activation-5.70e-04

Pos logit contributions
 pouch+0.042
 plunge+0.042
 couple+0.042
 goggles+0.041
 therapist+0.041
Neg logit contributions
 Acc-0.027
Acc-0.026
 over-0.022
tr-0.021
 collapse-0.021
Token One
Token ID1881
Feature activation+2.70e-02

Pos logit contributions
 pouch+0.001
 couple+0.001
gers+0.001
 goggles+0.001
 banks+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
 solved-0.001
Token day
Token ID1110
Feature activation-2.41e-03

Pos logit contributions
 Acc+0.070
Acc+0.067
 collapse+0.061
rou+0.060
 solved+0.058
Neg logit contributions
 couple-0.023
 pair-0.021
gers-0.021
 pouch-0.020
 goggles-0.020
Token,
Token ID11
Feature activation-1.60e-02

Pos logit contributions
 couple+0.003
 pouch+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.005
Acc-0.004
 collapse-0.004
rou-0.004
tr-0.004
Token and
Token ID290
Feature activation-9.23e-03

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.008
tr+0.008
rou+0.008
Neg logit contributions
 couple-0.009
 pouch-0.008
gers-0.008
 therapist-0.008
 goggles-0.008
Token grateful
Token ID14066
Feature activation-5.85e-04

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.015
 goggles+0.015
 plunge+0.014
Neg logit contributions
 Acc-0.015
Acc-0.015
 collapse-0.012
rou-0.012
 solved-0.012
Token that
Token ID326
Feature activation-6.48e-03

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.001
 goggles+0.001
 therapist+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token she
Token ID673
Feature activation+8.46e-05

Pos logit contributions
 pouch+0.008
 couple+0.008
gers+0.008
 goggles+0.008
 plunge+0.007
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.011
rou-0.011
 solved-0.011
Token hugged
Token ID41130
Feature activation+6.53e-03

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 pouch-0.000
 couple-0.000
gers-0.000
 goggles-0.000
 therapist-0.000
Token her
Token ID607
Feature activation+6.14e-03

Pos logit contributions
 Acc+0.012
Acc+0.011
 collapse+0.010
rou+0.010
tr+0.010
Neg logit contributions
 couple-0.009
 pouch-0.009
gers-0.009
 Holly-0.009
 therapist-0.009
Token mom
Token ID1995
Feature activation+1.43e-02

Pos logit contributions
 Acc+0.012
Acc+0.012
 collapse+0.010
rou+0.010
 unavoidable+0.010
Neg logit contributions
 pouch-0.008
 couple-0.008
 pair-0.008
 goggles-0.008
 bag-0.008
Tokenmy
Token ID1820
Feature activation-1.91e-03

Pos logit contributions
 Acc+0.017
Acc+0.016
 collapse+0.013
rou+0.011
 solved+0.011
Neg logit contributions
 pouch-0.031
 couple-0.030
 pair-0.029
gers-0.029
 bag-0.029
Token tight
Token ID5381
Feature activation-1.75e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.002
tr-0.002
Token.
Token ID13
Feature activation+2.73e-02

Pos logit contributions
 couple+0.031
 pouch+0.030
 goggles+0.030
gers+0.030
 plunge+0.030
Neg logit contributions
 Acc-0.026
Acc-0.024
 collapse-0.020
rou-0.020
tr-0.020
Token From
Token ID3574
Feature activation+1.54e-02

Pos logit contributions
 Acc+0.048
Acc+0.048
 collapse+0.047
rou+0.043
 solved+0.042
Neg logit contributions
 pouch-0.037
 Holly-0.035
 couple-0.034
 therapist-0.032
 sleeves-0.032
Token smiled
Token ID13541
Feature activation+1.84e-02

Pos logit contributions
 Acc+0.024
Acc+0.024
 collapse+0.020
rou+0.020
tr+0.019
Neg logit contributions
 couple-0.017
 pouch-0.017
 goggles-0.016
gers-0.016
 banks-0.016
Token and
Token ID290
Feature activation+4.12e-03

Pos logit contributions
 Acc+0.037
Acc+0.036
 collapse+0.032
tr+0.030
 unavoidable+0.029
Neg logit contributions
 couple-0.025
 pouch-0.024
gers-0.023
 goggles-0.023
 pair-0.022
Token said
Token ID531
Feature activation+5.30e-03

Pos logit contributions
 Acc+0.009
Acc+0.009
 collapse+0.008
rou+0.008
tr+0.008
Neg logit contributions
 couple-0.004
 pouch-0.004
 goggles-0.004
gers-0.004
 therapist-0.004
Token,
Token ID11
Feature activation-8.08e-03

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.009
rou+0.009
tr+0.009
Neg logit contributions
 couple-0.007
 pouch-0.007
gers-0.007
 goggles-0.007
 plunge-0.006
Token "
Token ID366
Feature activation+2.67e-03

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.016
Acc-0.016
 collapse-0.013
rou-0.013
tr-0.013
TokenI
Token ID40
Feature activation-3.36e-03

Pos logit contributions
 Acc+0.005
Acc+0.004
 collapse+0.004
rou+0.004
 solved+0.004
Neg logit contributions
 couple-0.004
 pouch-0.004
gers-0.004
 goggles-0.004
 therapist-0.004
Token'm
Token ID1101
Feature activation-1.24e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.005
Token good
Token ID922
Feature activation-1.80e-02

Pos logit contributions
gers+0.019
 pouch+0.019
 couple+0.019
 goggles+0.018
 banks+0.018
Neg logit contributions
 Acc-0.022
Acc-0.021
 collapse-0.018
rou-0.018
 solved-0.017
Token!
Token ID0
Feature activation+1.44e-02

Pos logit contributions
 pouch+0.030
 banks+0.029
gers+0.029
 goggles+0.029
 sleeves+0.028
Neg logit contributions
 Acc-0.030
Acc-0.028
rou-0.024
 collapse-0.023
 solved-0.023
Token What
Token ID1867
Feature activation+1.21e-02

Pos logit contributions
Acc+0.027
 Acc+0.027
 collapse+0.024
rou+0.023
 solved+0.023
Neg logit contributions
 pouch-0.019
 couple-0.018
 goggles-0.018
 therapist-0.018
gers-0.018
Token brings
Token ID6774
Feature activation+9.09e-03

Pos logit contributions
 Acc+0.023
Acc+0.023
 collapse+0.020
rou+0.019
tr+0.019
Neg logit contributions
 goggles-0.016
 couple-0.016
 plunge-0.016
 pair-0.015
gers-0.015
Token.
Token ID13
Feature activation-5.60e-03

Pos logit contributions
 Acc+0.012
Acc+0.011
rou+0.009
 collapse+0.009
tr+0.009
Neg logit contributions
 pouch-0.010
 couple-0.009
 goggles-0.009
gers-0.009
 banks-0.009
Token She
Token ID1375
Feature activation+1.77e-04

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.011
Acc-0.011
 collapse-0.009
rou-0.009
tr-0.009
Token thought
Token ID1807
Feature activation+2.24e-03

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 pouch-0.000
 couple-0.000
 goggles-0.000
gers-0.000
 sleeves-0.000
Token he
Token ID339
Feature activation-4.52e-03

Pos logit contributions
 Acc+0.005
Acc+0.005
rou+0.004
 collapse+0.004
tr+0.004
Neg logit contributions
 couple-0.003
 goggles-0.003
 pouch-0.003
gers-0.003
 plunge-0.003
Token might
Token ID1244
Feature activation+3.11e-03

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.006
Neg logit contributions
 Acc-0.008
Acc-0.008
 collapse-0.007
rou-0.006
tr-0.006
Token fall
Token ID2121
Feature activation+2.61e-02

Pos logit contributions
 Acc+0.005
Acc+0.005
rou+0.004
tr+0.004
 collapse+0.004
Neg logit contributions
 pouch-0.005
 couple-0.005
 goggles-0.005
gers-0.005
 plunge-0.005
Token and
Token ID290
Feature activation+2.55e-02

Pos logit contributions
 Acc+0.053
Acc+0.051
rou+0.042
 solved+0.041
sm+0.040
Neg logit contributions
gers-0.035
 pouch-0.034
 goggles-0.033
 legs-0.033
 sleeves-0.033
Token hurt
Token ID5938
Feature activation+7.72e-03

Pos logit contributions
 Acc+0.050
Acc+0.048
rou+0.042
ris+0.040
tr+0.040
Neg logit contributions
 plunge-0.033
 goggles-0.032
 legs-0.031
 pouch-0.031
ner-0.030
Token himself
Token ID2241
Feature activation+5.19e-03

Pos logit contributions
 Acc+0.014
Acc+0.013
rou+0.011
tr+0.011
 solved+0.011
Neg logit contributions
 goggles-0.011
gers-0.011
 pouch-0.011
 couple-0.011
 plunge-0.011
Token.
Token ID13
Feature activation-6.33e-03

Pos logit contributions
 Acc+0.009
Acc+0.009
rou+0.007
 collapse+0.007
 solved+0.007
Neg logit contributions
 pouch-0.008
 couple-0.008
gers-0.008
 goggles-0.008
 banks-0.008
Token She
Token ID1375
Feature activation-3.58e-03

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.009
 goggles+0.009
 plunge+0.009
Neg logit contributions
 Acc-0.012
Acc-0.011
 collapse-0.010
rou-0.009
tr-0.009
Token that
Token ID326
Feature activation-1.39e-02

Pos logit contributions
 Acc+0.023
Acc+0.022
rou+0.018
 collapse+0.018
tr+0.018
Neg logit contributions
 couple-0.017
 goggles-0.016
gers-0.016
 pouch-0.016
 plunge-0.016
Token he
Token ID339
Feature activation-1.64e-02

Pos logit contributions
 pouch+0.021
 couple+0.021
gers+0.020
ippers+0.020
 goggles+0.020
Neg logit contributions
 Acc-0.024
Acc-0.023
 collapse-0.020
 solved-0.020
tr-0.019
Token was
Token ID373
Feature activation-3.19e-04

Pos logit contributions
 pouch+0.027
 couple+0.027
 plunge+0.026
 goggles+0.026
gers+0.026
Neg logit contributions
 Acc-0.026
Acc-0.025
 solved-0.021
 collapse-0.021
rou-0.021
Token a
Token ID257
Feature activation+2.69e-02

Pos logit contributions
 couple+0.000
 pouch+0.000
 goggles+0.000
gers+0.000
 plunge+0.000
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.000
rou-0.000
tr-0.000
Token sailor
Token ID43272
Feature activation-1.26e-02

Pos logit contributions
 Acc+0.064
Acc+0.062
 collapse+0.056
rou+0.053
tr+0.052
Neg logit contributions
 couple-0.029
 pouch-0.027
 pair-0.026
 snail-0.026
 goggles-0.025
Token and
Token ID290
Feature activation-7.86e-03

Pos logit contributions
 pouch+0.021
 couple+0.021
gers+0.021
 goggles+0.021
 plunge+0.021
Neg logit contributions
 Acc-0.019
Acc-0.018
 collapse-0.016
rou-0.015
 solved-0.015
Token sailed
Token ID33835
Feature activation-1.27e-02

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.015
Acc-0.015
 collapse-0.013
rou-0.013
 solved-0.012
Token around
Token ID1088
Feature activation+3.81e-03

Pos logit contributions
 pouch+0.022
 couple+0.021
 therapist+0.021
 goggles+0.021
gers+0.021
Neg logit contributions
 Acc-0.019
Acc-0.018
 collapse-0.016
 solved-0.015
rou-0.015
Token in
Token ID287
Feature activation-2.37e-03

Pos logit contributions
 Acc+0.007
Acc+0.006
 collapse+0.005
tr+0.005
rou+0.005
Neg logit contributions
 couple-0.006
 pouch-0.006
gers-0.006
 goggles-0.006
 plunge-0.006
Token his
Token ID465
Feature activation-6.50e-03

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.004
Acc-0.004
 collapse-0.003
rou-0.003
tr-0.003
Token boat
Token ID8848
Feature activation-3.36e-02

Pos logit contributions
 couple+0.008
 pouch+0.008
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.014
Acc-0.013
 collapse-0.012
rou-0.011
tr-0.011
Token smiled
Token ID13541
Feature activation+9.34e-03

Pos logit contributions
 Acc+0.009
Acc+0.009
 collapse+0.007
rou+0.007
tr+0.007
Neg logit contributions
 couple-0.008
 pouch-0.007
 goggles-0.007
gers-0.007
 therapist-0.007
Token and
Token ID290
Feature activation-5.01e-03

Pos logit contributions
 Acc+0.017
Acc+0.017
 collapse+0.015
tr+0.014
rou+0.014
Neg logit contributions
 couple-0.014
 pouch-0.013
gers-0.013
 goggles-0.013
 pair-0.013
Token nodded
Token ID14464
Feature activation+2.71e-03

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.006
Neg logit contributions
 Acc-0.010
Acc-0.010
 collapse-0.009
rou-0.009
tr-0.009
Token.
Token ID13
Feature activation+8.57e-03

Pos logit contributions
 Acc+0.005
Acc+0.004
 collapse+0.004
rou+0.004
tr+0.004
Neg logit contributions
 couple-0.004
 pouch-0.004
gers-0.004
 goggles-0.004
 pair-0.004
Token
Token ID220
Feature activation+1.71e-02

Pos logit contributions
 Acc+0.018
Acc+0.018
 collapse+0.016
rou+0.016
tr+0.015
Neg logit contributions
 pouch-0.010
 couple-0.010
 therapist-0.009
gers-0.009
 goggles-0.009
Token
Token ID198
Feature activation-4.91e-03

Pos logit contributions
 Acc+0.034
Acc+0.034
 collapse+0.029
rou+0.028
tr+0.028
Neg logit contributions
 pouch-0.021
ippers-0.020
 couple-0.020
gers-0.019
 goggles-0.019
Token
Token ID198
Feature activation+4.63e-03

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.009
Acc-0.009
 collapse-0.007
rou-0.007
tr-0.007
TokenTogether
Token ID41631
Feature activation-7.31e-03

Pos logit contributions
 Acc+0.008
Acc+0.007
 collapse+0.006
rou+0.006
tr+0.006
Neg logit contributions
 couple-0.008
 pouch-0.007
gers-0.007
 therapist-0.007
 goggles-0.007
Token,
Token ID11
Feature activation+4.55e-03

Pos logit contributions
 pouch+0.010
 couple+0.010
gers+0.010
 goggles+0.010
 plunge+0.010
Neg logit contributions
 Acc-0.014
Acc-0.013
 collapse-0.011
rou-0.011
 solved-0.011
Token the
Token ID262
Feature activation-5.22e-04

Pos logit contributions
 Acc+0.009
Acc+0.009
 collapse+0.007
rou+0.007
tr+0.007
Neg logit contributions
 couple-0.006
 pouch-0.006
 goggles-0.006
gers-0.006
 plunge-0.006
Token woman
Token ID2415
Feature activation+1.93e-02

Pos logit contributions
 couple+0.001
 pouch+0.001
gers+0.000
 goggles+0.000
 therapist+0.000
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token this
Token ID428
Feature activation-3.65e-02

Pos logit contributions
 Acc+0.018
Acc+0.017
rou+0.014
 collapse+0.014
tr+0.014
Neg logit contributions
 couple-0.013
 goggles-0.013
 pouch-0.013
 plunge-0.013
gers-0.013
Token is
Token ID318
Feature activation+2.83e-02

Pos logit contributions
 couple+0.074
 pouch+0.074
 banks+0.074
 plunge+0.074
ner+0.073
Neg logit contributions
 Acc-0.039
Acc-0.036
tr-0.030
 solved-0.029
 over-0.028
Token my
Token ID616
Feature activation-1.82e-02

Pos logit contributions
 Acc+0.054
Acc+0.053
rou+0.042
 collapse+0.040
sm+0.040
Neg logit contributions
 goggles-0.043
 pills-0.043
 Holly-0.040
 pouch-0.040
 pair-0.040
Token plate
Token ID7480
Feature activation+1.77e-02

Pos logit contributions
 pouch+0.022
 couple+0.022
gers+0.021
 banks+0.021
ner+0.021
Neg logit contributions
 Acc-0.037
Acc-0.035
rou-0.031
 collapse-0.030
tr-0.030
Token.
Token ID13
Feature activation-8.22e-03

Pos logit contributions
 Acc+0.033
Acc+0.032
rou+0.025
 collapse+0.025
tr+0.024
Neg logit contributions
gers-0.026
 pouch-0.026
 couple-0.026
 goggles-0.025
 pair-0.025
Token I
Token ID314
Feature activation-1.84e-02

Pos logit contributions
 couple+0.012
 pouch+0.012
gers+0.012
 goggles+0.012
 plunge+0.012
Neg logit contributions
 Acc-0.015
Acc-0.014
 collapse-0.012
rou-0.012
tr-0.012
Token like
Token ID588
Feature activation-1.58e-02

Pos logit contributions
 couple+0.030
 pouch+0.030
 plunge+0.029
gers+0.029
 sleeves+0.028
Neg logit contributions
 Acc-0.030
Acc-0.029
 solved-0.025
 collapse-0.025
rou-0.025
Token flowers
Token ID12734
Feature activation-1.32e-02

Pos logit contributions
 pouch+0.028
 couple+0.028
gers+0.028
 plunge+0.027
 banks+0.027
Neg logit contributions
 Acc-0.023
Acc-0.022
 collapse-0.017
rou-0.017
 solved-0.017
Token more
Token ID517
Feature activation+1.30e-03

Pos logit contributions
 couple+0.022
 pouch+0.022
gers+0.022
 plunge+0.022
 goggles+0.022
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
rou-0.016
tr-0.016
Token than
Token ID621
Feature activation+4.65e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.002
rou+0.002
tr+0.002
Neg logit contributions
 couple-0.002
 pouch-0.002
gers-0.002
 goggles-0.002
 banks-0.002
Token animals
Token ID4695
Feature activation-9.25e-03

Pos logit contributions
 Acc+0.008
Acc+0.008
 collapse+0.007
rou+0.007
tr+0.007
Neg logit contributions
 couple-0.007
 pouch-0.007
 goggles-0.007
gers-0.007
 plunge-0.007
Token the
Token ID262
Feature activation-5.02e-02

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.011
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.011
Acc-0.011
 collapse-0.009
rou-0.009
 solved-0.009
Token bunny
Token ID44915
Feature activation-3.34e-02

Pos logit contributions
ippers+0.067
angers+0.066
gers+0.064
 goggles+0.060
ancer+0.059
Neg logit contributions
 Acc-0.089
Acc-0.079
 fixed-0.076
 over-0.074
 lightning-0.074
Token felt
Token ID2936
Feature activation-5.41e-03

Pos logit contributions
gers+0.059
ippers+0.059
 couple+0.058
 goggles+0.058
 plunge+0.058
Neg logit contributions
 Acc-0.046
Acc-0.045
 solved-0.042
 collapse-0.040
 over-0.039
Token sad
Token ID6507
Feature activation+2.11e-03

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.010
Acc-0.010
 collapse-0.009
rou-0.008
tr-0.008
Token.
Token ID13
Feature activation-1.99e-02

Pos logit contributions
 Acc+0.004
Acc+0.004
 collapse+0.003
tr+0.003
rou+0.003
Neg logit contributions
 pouch-0.003
 couple-0.003
gers-0.003
 goggles-0.003
 banks-0.003
Token Its
Token ID6363
Feature activation-3.36e-02

Pos logit contributions
 couple+0.037
 pouch+0.036
 goggles+0.036
gers+0.036
 pair+0.036
Neg logit contributions
 Acc-0.030
Acc-0.028
 collapse-0.022
 solved-0.022
tr-0.021
Token friends
Token ID2460
Feature activation-5.14e-03

Pos logit contributions
ner+0.041
ancer+0.037
 therapist+0.037
 aisle+0.036
 couple+0.036
Neg logit contributions
 Acc-0.067
Acc-0.063
 collapse-0.061
 over-0.060
tr-0.059
Token came
Token ID1625
Feature activation+1.26e-02

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.008
tr-0.008
Token to
Token ID284
Feature activation+2.72e-02

Pos logit contributions
 Acc+0.023
Acc+0.021
 collapse+0.018
tr+0.018
rou+0.017
Neg logit contributions
 pouch-0.020
 couple-0.019
 plunge-0.018
 goggles-0.018
 legs-0.018
Token comfort
Token ID4467
Feature activation+1.12e-02

Pos logit contributions
 Acc+0.059
Acc+0.057
tr+0.056
rou+0.056
 collapse+0.051
Neg logit contributions
 couple-0.029
 plunge-0.028
 goggles-0.026
 pair-0.025
 banks-0.025
Token it
Token ID340
Feature activation+1.31e-02

Pos logit contributions
 Acc+0.020
Acc+0.019
rou+0.017
 collapse+0.017
tr+0.017
Neg logit contributions
gers-0.017
 pouch-0.017
 couple-0.016
 goggles-0.016
 plunge-0.016
Token to
Token ID284
Feature activation-1.16e-02

Pos logit contributions
 couple+0.002
 pouch+0.002
gers+0.002
 goggles+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token it
Token ID340
Feature activation-1.79e-03

Pos logit contributions
 pouch+0.020
 couple+0.019
gers+0.019
 therapist+0.019
 goggles+0.019
Neg logit contributions
 Acc-0.018
Acc-0.017
 collapse-0.015
rou-0.014
 solved-0.014
Token.
Token ID13
Feature activation-3.61e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.002
rou-0.002
 solved-0.002
Token Lily
Token ID20037
Feature activation-5.26e-03

Pos logit contributions
 couple+0.069
 plunge+0.067
 therapist+0.067
gers+0.067
 pouch+0.067
Neg logit contributions
 Acc-0.050
Acc-0.044
 over-0.038
 Tim-0.032
 collapse-0.032
Token goes
Token ID2925
Feature activation-1.70e-03

Pos logit contributions
 pouch+0.008
 couple+0.008
gers+0.008
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.010
Acc-0.009
 collapse-0.008
rou-0.008
tr-0.007
Token first
Token ID717
Feature activation-1.92e-02

Pos logit contributions
 couple+0.003
 pouch+0.002
gers+0.002
 goggles+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.002
tr-0.002
Token and
Token ID290
Feature activation+1.07e-02

Pos logit contributions
 pouch+0.034
 couple+0.033
 goggles+0.033
gers+0.033
 plunge+0.033
Neg logit contributions
 Acc-0.027
Acc-0.026
 collapse-0.022
rou-0.022
 over-0.022
Token slides
Token ID19392
Feature activation-9.23e-03

Pos logit contributions
 Acc+0.019
Acc+0.018
 collapse+0.016
tr+0.015
 solved+0.015
Neg logit contributions
 goggles-0.016
 pouch-0.016
 couple-0.016
gers-0.016
 plunge-0.016
Token down
Token ID866
Feature activation+1.40e-02

Pos logit contributions
 couple+0.016
 pouch+0.016
gers+0.016
 goggles+0.016
 plunge+0.015
Neg logit contributions
 Acc-0.014
Acc-0.013
 collapse-0.011
rou-0.011
tr-0.011
Token fast
Token ID3049
Feature activation-8.70e-03

Pos logit contributions
 Acc+0.028
Acc+0.026
tr+0.023
rou+0.022
 collapse+0.021
Neg logit contributions
gers-0.019
 couple-0.019
 pouch-0.019
ner-0.019
 goggles-0.018
Token.
Token ID13
Feature activation-2.90e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
 goggles+0.015
gers+0.015
 plunge+0.014
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.011
rou-0.010
 solved-0.010
Token pup
Token ID15552
Feature activation-6.79e-03

Pos logit contributions
 Acc+0.071
Acc+0.068
 collapse+0.064
 resolved+0.061
 unavoidable+0.061
Neg logit contributions
 couple-0.024
 pair-0.021
 pouch-0.021
 snail-0.020
 donkey-0.019
Token,
Token ID11
Feature activation+1.14e-02

Pos logit contributions
 pouch+0.012
 couple+0.012
gers+0.012
 goggles+0.012
 plunge+0.011
Neg logit contributions
 Acc-0.010
Acc-0.010
 collapse-0.008
rou-0.008
tr-0.008
Token and
Token ID290
Feature activation-3.54e-02

Pos logit contributions
 Acc+0.019
Acc+0.017
 collapse+0.015
rou+0.015
tr+0.014
Neg logit contributions
 pouch-0.019
 goggles-0.018
 pair-0.018
 couple-0.018
gers-0.017
Token he
Token ID339
Feature activation-1.92e-03

Pos logit contributions
 banks+0.069
ippers+0.067
 sleeves+0.065
 pouch+0.065
 plunge+0.065
Neg logit contributions
 Acc-0.050
Acc-0.049
 solved-0.038
sm-0.036
 over-0.036
Token was
Token ID373
Feature activation-3.01e-02

Pos logit contributions
 pouch+0.003
 couple+0.003
gers+0.003
 goggles+0.003
 plunge+0.003
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.002
tr-0.002
Token very
Token ID845
Feature activation-3.57e-02

Pos logit contributions
gers+0.055
 couple+0.054
 sleeves+0.053
 kernels+0.053
 plunge+0.052
Neg logit contributions
 Acc-0.047
Acc-0.044
rou-0.036
 over-0.036
tr-0.035
Token unique
Token ID3748
Feature activation-1.86e-02

Pos logit contributions
 banks+0.057
 sleeves+0.056
 kernels+0.056
 couple+0.055
gers+0.055
Neg logit contributions
 Acc-0.063
Acc-0.057
rou-0.051
 collapse-0.046
 solved-0.045
Token.
Token ID13
Feature activation-1.03e-02

Pos logit contributions
 pouch+0.034
 couple+0.034
 plunge+0.034
 goggles+0.034
gers+0.033
Neg logit contributions
 Acc-0.026
Acc-0.024
rou-0.021
tr-0.020
 collapse-0.019
Token He
Token ID679
Feature activation-2.93e-03

Pos logit contributions
 couple+0.014
 pouch+0.014
gers+0.014
 plunge+0.014
 goggles+0.014
Neg logit contributions
 Acc-0.020
Acc-0.019
 collapse-0.016
tr-0.016
rou-0.016
Token liked
Token ID8288
Feature activation+2.99e-03

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.004
 goggles+0.004
 plunge+0.004
Neg logit contributions
 Acc-0.005
Acc-0.005
 collapse-0.004
rou-0.004
tr-0.004
Token to
Token ID284
Feature activation-1.40e-02

Pos logit contributions
 Acc+0.005
Acc+0.004
 collapse+0.004
rou+0.004
tr+0.004
Neg logit contributions
 couple-0.005
 pouch-0.005
 goggles-0.005
gers-0.005
 plunge-0.005
Token.
Token ID13
Feature activation-7.44e-03

Pos logit contributions
 Acc+0.026
Acc+0.025
 collapse+0.022
 solved+0.020
tr+0.020
Neg logit contributions
 couple-0.021
 pouch-0.020
gers-0.020
 legs-0.019
ippers-0.019
Token
Token ID198
Feature activation-8.77e-03

Pos logit contributions
 couple+0.011
 pouch+0.011
gers+0.011
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.013
Acc-0.013
 collapse-0.011
rou-0.011
tr-0.010
Token
Token ID198
Feature activation-9.71e-03

Pos logit contributions
 pouch+0.013
 couple+0.013
gers+0.013
 goggles+0.013
 plunge+0.013
Neg logit contributions
 Acc-0.015
Acc-0.015
 collapse-0.012
rou-0.012
 solved-0.012
TokenAt
Token ID2953
Feature activation-1.39e-03

Pos logit contributions
 pouch+0.016
 couple+0.016
gers+0.016
 goggles+0.016
 plunge+0.016
Neg logit contributions
 Acc-0.016
Acc-0.015
 collapse-0.012
rou-0.012
 solved-0.012
Token home
Token ID1363
Feature activation-1.45e-02

Pos logit contributions
 couple+0.002
 pouch+0.002
 goggles+0.002
gers+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.002
Acc-0.002
 collapse-0.002
rou-0.002
tr-0.002
Token,
Token ID11
Feature activation-3.37e-02

Pos logit contributions
 pouch+0.022
 goggles+0.022
 couple+0.022
gers+0.022
 banks+0.021
Neg logit contributions
 Acc-0.026
Acc-0.024
 collapse-0.021
rou-0.020
 solved-0.020
Token M
Token ID337
Feature activation-7.21e-03

Pos logit contributions
 pouch+0.069
 banks+0.067
gers+0.067
 bank+0.066
 donkey+0.066
Neg logit contributions
 Acc-0.042
Acc-0.039
 over-0.033
rou-0.032
 collapse-0.030
Tokenaddy
Token ID13218
Feature activation+9.09e-03

Pos logit contributions
 pouch+0.011
 couple+0.011
gers+0.011
 goggles+0.011
 plunge+0.011
Neg logit contributions
 Acc-0.012
Acc-0.012
 collapse-0.010
rou-0.010
tr-0.009
Token's
Token ID338
Feature activation-1.33e-03

Pos logit contributions
 Acc+0.017
Acc+0.016
 collapse+0.014
rou+0.013
tr+0.013
Neg logit contributions
 couple-0.013
 pouch-0.013
gers-0.013
 goggles-0.012
 therapist-0.012
Token dad
Token ID9955
Feature activation+1.30e-02

Pos logit contributions
 couple+0.002
 pouch+0.002
gers+0.002
 goggles+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.002
Acc-0.002
 collapse-0.002
rou-0.002
tr-0.002
Token had
Token ID550
Feature activation-1.62e-02

Pos logit contributions
 Acc+0.026
Acc+0.025
 collapse+0.022
tr+0.021
rou+0.021
Neg logit contributions
 couple-0.017
 pouch-0.016
 pair-0.016
gers-0.016
 therapist-0.016
Token.
Token ID13
Feature activation-3.86e-03

Pos logit contributions
 Acc+0.007
Acc+0.007
 collapse+0.005
rou+0.005
tr+0.005
Neg logit contributions
 couple-0.007
 pouch-0.007
gers-0.006
 goggles-0.006
 banks-0.006
Token
Token ID220
Feature activation+6.67e-03

Pos logit contributions
 pouch+0.006
 couple+0.006
gers+0.006
 goggles+0.006
 plunge+0.005
Neg logit contributions
 Acc-0.007
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.005
Token But
Token ID887
Feature activation-9.35e-03

Pos logit contributions
 Acc+0.013
Acc+0.012
 collapse+0.011
rou+0.010
tr+0.010
Neg logit contributions
 pouch-0.009
 couple-0.009
gers-0.009
 goggles-0.009
 banks-0.009
Token the
Token ID262
Feature activation-5.35e-02

Pos logit contributions
 pouch+0.015
 couple+0.015
gers+0.015
 goggles+0.015
 sleeves+0.015
Neg logit contributions
 Acc-0.015
Acc-0.015
 collapse-0.012
rou-0.012
 solved-0.012
Token b
Token ID275
Feature activation-2.50e-02

Pos logit contributions
ippers+0.079
angers+0.073
 goggles+0.071
gers+0.071
ner+0.069
Neg logit contributions
 Acc-0.094
Acc-0.086
 collapse-0.082
 solved-0.080
 cal-0.080
Tokenunn
Token ID20935
Feature activation-6.78e-02

Pos logit contributions
 goggles+0.020
 couple+0.018
 plunge+0.018
gers+0.018
 pair+0.017
Neg logit contributions
 Acc-0.063
Acc-0.061
rou-0.057
sm-0.055
 collapse-0.054
Tokenies
Token ID444
Feature activation-2.60e-02

Pos logit contributions
 goggles+0.113
 plunge+0.109
 trainers+0.108
gers+0.107
ner+0.106
Neg logit contributions
 Acc-0.108
Acc-0.097
rou-0.092
sm-0.090
 over-0.081
Token didn
Token ID1422
Feature activation+1.24e-02

Pos logit contributions
gers+0.038
 couple+0.037
 pouch+0.037
 goggles+0.037
ippers+0.036
Neg logit contributions
 Acc-0.047
Acc-0.044
 collapse-0.039
 solved-0.038
rou-0.038
Token't
Token ID470
Feature activation-1.68e-02

Pos logit contributions
 Acc+0.032
Acc+0.031
rou+0.028
tr+0.027
 solved+0.027
Neg logit contributions
 pouch-0.008
 plunge-0.008
ner-0.008
 couple-0.008
gers-0.008
Token wake
Token ID7765
Feature activation+7.62e-03

Pos logit contributions
 pouch+0.029
gers+0.028
 goggles+0.028
 couple+0.028
 therapist+0.027
Neg logit contributions
 Acc-0.026
Acc-0.025
 collapse-0.022
 solved-0.021
rou-0.020
Token up
Token ID510
Feature activation+6.90e-04

Pos logit contributions
 Acc+0.012
Acc+0.012
tr+0.010
rou+0.010
 collapse+0.010
Neg logit contributions
 couple-0.013
 pouch-0.012
gers-0.012
 plunge-0.012
 banks-0.012
Token.
Token ID13
Feature activation-2.90e-03

Pos logit contributions
 couple+0.063
 pouch+0.062
 sleeves+0.061
gers+0.061
 plunge+0.060
Neg logit contributions
 Acc-0.039
Acc-0.037
rou-0.033
 over-0.033
tr-0.031
Token He
Token ID679
Feature activation-3.44e-02

Pos logit contributions
 pouch+0.004
 couple+0.004
gers+0.004
 goggles+0.004
 banks+0.003
Neg logit contributions
 Acc-0.006
Acc-0.006
 collapse-0.005
rou-0.005
tr-0.005
Token flew
Token ID13112
Feature activation-1.77e-02

Pos logit contributions
ippers+0.059
ner+0.058
 couple+0.058
 plunge+0.057
gers+0.056
Neg logit contributions
 Acc-0.047
Acc-0.047
 solved-0.045
 fixed-0.042
 over-0.042
Token for
Token ID329
Feature activation-2.30e-02

Pos logit contributions
 pouch+0.032
 couple+0.032
 therapist+0.031
 sleeves+0.031
ippers+0.031
Neg logit contributions
 Acc-0.024
Acc-0.023
 over-0.020
 collapse-0.019
 solved-0.019
Token a
Token ID257
Feature activation-4.04e-02

Pos logit contributions
 pouch+0.038
 therapist+0.037
gers+0.037
ippers+0.036
 couple+0.036
Neg logit contributions
 Acc-0.036
Acc-0.035
 collapse-0.031
 solved-0.031
sm-0.029
Token long
Token ID890
Feature activation-6.09e-02

Pos logit contributions
ippers+0.048
gers+0.047
 therapist+0.045
angers+0.043
 pills+0.043
Neg logit contributions
Acc-0.079
 Acc-0.076
 solved-0.071
tr-0.068
 fixed-0.067
Token time
Token ID640
Feature activation-2.78e-02

Pos logit contributions
sters+0.127
ancer+0.126
 therapist+0.123
ippers+0.123
angers+0.121
Neg logit contributions
Acc-0.055
 solved-0.053
 over-0.051
 Acc-0.051
 fixed-0.051
Token until
Token ID1566
Feature activation+2.13e-03

Pos logit contributions
gers+0.049
 pouch+0.049
 therapist+0.048
 goggles+0.048
 couple+0.047
Neg logit contributions
 Acc-0.040
Acc-0.039
 solved-0.033
rou-0.032
 over-0.032
Token he
Token ID339
Feature activation-5.00e-03

Pos logit contributions
 Acc+0.005
Acc+0.005
rou+0.004
tr+0.004
 collapse+0.004
Neg logit contributions
 couple-0.002
 pouch-0.002
gers-0.002
 goggles-0.002
 plunge-0.002
Token found
Token ID1043
Feature activation-1.26e-02

Pos logit contributions
 pouch+0.007
 couple+0.007
gers+0.007
 goggles+0.007
 plunge+0.007
Neg logit contributions
 Acc-0.009
Acc-0.009
 collapse-0.007
rou-0.007
tr-0.007
Token a
Token ID257
Feature activation-1.60e-02

Pos logit contributions
 couple+0.020
 pouch+0.020
gers+0.020
 sleeves+0.020
 goggles+0.020
Neg logit contributions
 Acc-0.020
Acc-0.020
 collapse-0.017
 solved-0.016
rou-0.016
Token Lily
Token ID20037
Feature activation+2.45e-03

Pos logit contributions
 pouch+0.049
 banks+0.048
 couple+0.047
gers+0.047
 plunge+0.047
Neg logit contributions
 Acc-0.024
Acc-0.023
 solved-0.017
tr-0.017
 collapse-0.017
Token tried
Token ID3088
Feature activation+2.01e-02

Pos logit contributions
 Acc+0.005
Acc+0.004
 collapse+0.004
rou+0.004
tr+0.004
Neg logit contributions
 pouch-0.004
 couple-0.004
gers-0.003
 goggles-0.003
 sleeves-0.003
Token to
Token ID284
Feature activation-2.75e-02

Pos logit contributions
 Acc+0.038
Acc+0.035
 collapse+0.032
 unavoidable+0.030
rou+0.030
Neg logit contributions
 legs-0.028
 pair-0.028
 pouch-0.028
 goggles-0.028
 shoes-0.028
Token put
Token ID1234
Feature activation-3.60e-02

Pos logit contributions
gers+0.046
 couple+0.045
 pouch+0.044
 kernels+0.044
 therapist+0.043
Neg logit contributions
 Acc-0.043
Acc-0.041
 collapse-0.036
 solved-0.035
 over-0.034
Token on
Token ID319
Feature activation-2.53e-02

Pos logit contributions
 couple+0.067
 banks+0.065
 plunge+0.063
gers+0.061
sters+0.061
Neg logit contributions
 Acc-0.047
Acc-0.046
 over-0.046
 fixed-0.041
 solved-0.039
Token her
Token ID607
Feature activation-6.60e-02

Pos logit contributions
 couple+0.049
gers+0.046
 pouch+0.046
 plunge+0.046
ner+0.045
Neg logit contributions
 Acc-0.034
Acc-0.032
 collapse-0.026
 over-0.026
rou-0.026
Token sweater
Token ID35685
Feature activation-1.82e-02

Pos logit contributions
ner+0.112
 kernels+0.112
 couple+0.111
 banks+0.110
 plunge+0.109
Neg logit contributions
 Acc-0.084
Acc-0.084
tr-0.082
 over-0.080
 fixed-0.076
Token.
Token ID13
Feature activation-1.26e-02

Pos logit contributions
 couple+0.031
 pouch+0.031
 goggles+0.030
gers+0.030
 plunge+0.030
Neg logit contributions
 Acc-0.028
Acc-0.026
tr-0.023
 collapse-0.022
 over-0.022
Token It
Token ID632
Feature activation-1.44e-02

Pos logit contributions
 couple+0.018
 pouch+0.018
gers+0.018
 goggles+0.017
 plunge+0.017
Neg logit contributions
 Acc-0.024
Acc-0.022
 collapse-0.019
tr-0.019
rou-0.018
Token was
Token ID373
Feature activation-7.65e-04

Pos logit contributions
 couple+0.025
 pouch+0.024
gers+0.024
 goggles+0.024
ippers+0.024
Neg logit contributions
 Acc-0.023
Acc-0.022
 collapse-0.019
tr-0.019
 solved-0.018
Token a
Token ID257
Feature activation+6.66e-03

Pos logit contributions
 pouch+0.001
 couple+0.001
 goggles+0.001
gers+0.001
 plunge+0.001
Neg logit contributions
 Acc-0.001
Acc-0.001
 collapse-0.001
rou-0.001
tr-0.001
Token the
Token ID262
Feature activation+1.20e-02

Pos logit contributions
 pouch+0.009
 couple+0.009
gers+0.008
 goggles+0.008
 plunge+0.008
Neg logit contributions
 Acc-0.008
Acc-0.008
 collapse-0.007
rou-0.007
tr-0.006
Token house
Token ID2156
Feature activation-2.12e-02

Pos logit contributions
 Acc+0.027
Acc+0.025
 collapse+0.022
rou+0.022
 resolved+0.021
Neg logit contributions
 pouch-0.014
gers-0.013
 goggles-0.013
 pair-0.013
 couple-0.013
Token.
Token ID13
Feature activation-3.77e-02

Pos logit contributions
 couple+0.041
 pouch+0.041
 therapist+0.040
 goggles+0.040
gers+0.040
Neg logit contributions
 Acc-0.028
Acc-0.026
 collapse-0.022
 over-0.022
 solved-0.021
Token Every
Token ID3887
Feature activation-1.61e-03

Pos logit contributions
 couple+0.066
kins+0.065
gers+0.064
 plunge+0.063
 sleeves+0.062
Neg logit contributions
 Acc-0.059
Acc-0.053
tr-0.050
 over-0.046
 collapse-0.043
Token day
Token ID1110
Feature activation-3.87e-02

Pos logit contributions
 couple+0.002
 pouch+0.002
gers+0.002
 goggles+0.002
 plunge+0.002
Neg logit contributions
 Acc-0.003
Acc-0.003
 collapse-0.003
rou-0.003
 solved-0.003
Token,
Token ID11
Feature activation-6.59e-02

Pos logit contributions
 couple+0.071
 banks+0.070
 pouch+0.069
 sleeves+0.068
ippers+0.068
Neg logit contributions
 Acc-0.049
Acc-0.043
 over-0.040
 collapse-0.040
tr-0.040
Token Fl
Token ID1610
Feature activation+3.67e-03

Pos logit contributions
ippers+0.154
 couple+0.148
 sleeves+0.147
 pouch+0.145
kins+0.144
Neg logit contributions
 Acc-0.050
Acc-0.046
 Tim-0.045
 over-0.040
 collapse-0.039
Tokenuffy
Token ID15352
Feature activation+2.78e-04

Pos logit contributions
 Acc+0.007
Acc+0.007
 collapse+0.006
 solved+0.006
rou+0.006
Neg logit contributions
 pouch-0.005
gers-0.004
 couple-0.004
ippers-0.004
 goggles-0.004
Token would
Token ID561
Feature activation+7.98e-03

Pos logit contributions
 Acc+0.000
Acc+0.000
 collapse+0.000
rou+0.000
tr+0.000
Neg logit contributions
 pouch-0.000
 couple-0.000
gers-0.000
 goggles-0.000
 banks-0.000
Token jump
Token ID4391
Feature activation+1.18e-03

Pos logit contributions
 Acc+0.014
Acc+0.013
rou+0.011
 collapse+0.011
tr+0.010
Neg logit contributions
 pouch-0.013
 couple-0.012
 goggles-0.012
 plunge-0.012
 pair-0.012
Token and
Token ID290
Feature activation+2.45e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.002
rou+0.002
tr+0.002
Neg logit contributions
 pouch-0.002
gers-0.002
 couple-0.002
 goggles-0.002
 plunge-0.002
Token a
Token ID257
Feature activation-1.75e-02

Pos logit contributions
 couple+0.015
 pouch+0.014
gers+0.014
 goggles+0.014
 plunge+0.014
Neg logit contributions
 Acc-0.014
Acc-0.013
 collapse-0.011
rou-0.011
 solved-0.011
Token basket
Token ID7988
Feature activation-1.55e-02

Pos logit contributions
gers+0.017
 couple+0.017
 goggles+0.016
 pouch+0.016
 plunge+0.016
Neg logit contributions
 Acc-0.039
Acc-0.038
 collapse-0.034
rou-0.033
 solved-0.033
Token.
Token ID13
Feature activation-3.99e-03

Pos logit contributions
 couple+0.027
 pouch+0.026
gers+0.026
 goggles+0.026
 plunge+0.026
Neg logit contributions
 Acc-0.023
Acc-0.022
 collapse-0.019
rou-0.019
 solved-0.018
Token He
Token ID679
Feature activation+8.07e-03

Pos logit contributions
 pouch+0.005
 couple+0.005
gers+0.005
 goggles+0.005
 plunge+0.005
Neg logit contributions
 Acc-0.008
Acc-0.007
 collapse-0.006
rou-0.006
tr-0.006
Token placed
Token ID4624
Feature activation-4.01e-02

Pos logit contributions
 Acc+0.015
Acc+0.014
 collapse+0.012
rou+0.012
tr+0.012
Neg logit contributions
 pouch-0.012
 couple-0.012
gers-0.011
 goggles-0.011
 therapist-0.011
Token the
Token ID262
Feature activation-8.60e-02

Pos logit contributions
 pouch+0.070
 couple+0.070
gers+0.069
 therapist+0.068
ippers+0.068
Neg logit contributions
 Acc-0.054
Acc-0.052
 over-0.048
rou-0.044
 collapse-0.043
Token basket
Token ID7988
Feature activation+5.44e-03

Pos logit contributions
ippers+0.140
ner+0.140
ancer+0.139
 suitcase+0.134
 goggles+0.134
Neg logit contributions
rou-0.100
sm-0.093
 over-0.093
 pays-0.088
Acc-0.088
Token of
Token ID286
Feature activation-7.30e-02

Pos logit contributions
 Acc+0.010
Acc+0.010
 collapse+0.008
rou+0.008
tr+0.008
Neg logit contributions
 pouch-0.008
 couple-0.008
gers-0.007
 goggles-0.007
 plunge-0.007
Token l
Token ID300
Feature activation-3.10e-02

Pos logit contributions
ancer+0.123
ner+0.117
ippers+0.114
robe+0.112
angers+0.111
Neg logit contributions
 Acc-0.097
 over-0.092
sm-0.091
Acc-0.086
 thanks-0.086
Tokenimes
Token ID999
Feature activation+1.03e-03

Pos logit contributions
 couple+0.036
 banks+0.035
 plunge+0.034
 goggles+0.034
 pouch+0.033
Neg logit contributions
 Acc-0.061
Acc-0.060
rou-0.056
sm-0.055
 solved-0.053
Token at
Token ID379
Feature activation+3.84e-03

Pos logit contributions
 Acc+0.002
Acc+0.002
 collapse+0.002
rou+0.002
tr+0.002
Neg logit contributions
 pouch-0.001
 couple-0.001
gers-0.001
 goggles-0.001
 plunge-0.001