TOP ACTIVATIONS
MAX = 6.030

who had died a year ago . It was very precious
someone a very long time ago . The little girl thought
who lived a long time ago ," mom said .
important person a long time ago . She was excited and
were from a long time ago and they help us remember
her birthday a few days ago . She liked to practice
same dress a long time ago . <|endoftext|> Once upon a
this house a long time ago ." L ily
had bought a long time ago . They did not know
would take out her depend able oven m itt s and
it away a long time ago . It is not good
they lived a long time ago . Lily listened carefully and
a ship that sank long ago . She said they could
is from a long time ago , when a king lived
She had a brush with pink brist les that she loved
were burying something from long ago that made them feel sad
people did a long time ago . They lived in castles
and plants that lived long ago ," she says .
. She said wine was for grown - ups . She
to live a long time ago ." The girl

MIN ACTIVATIONS
MIN = -5.928

so grateful , he gave Fin the shiny strip as a
higher . He showed T oni how to lift up her
ran over and gave the rabbit a big hug . But
ran over and gave the rabbit a big hug . But
The man showed Sam how to hold the gun
Tim my said , showing Billy the toy arrow . "
He wanted to teach the squirrel , so he placed the
and D ucky taught P andy how to qu ack like
agged his tail , showing Tim that he should try the
a new friend and showed Benny around . As
. Tim my told Nem o that he ate a lot
and hills , which made Caroline a bit less uncomfortable .
silly ! Then Bob showed Jim how to make the toy
And so , Andy taught Billy how to crawl safely and
The little boy told Bill about the big cooler and
The bear told Tony that he needed to find
. Nut ty showed Bub bles a shiny shield that he
K iki would teach T oto how to push his wings
He said whist ling made him happy and calm . L
safe ! He gave the knight a big hug and said

INTERVAL 3.040 - 6.030
CONTAINS 0.243%

milk . " Grand ma , you are very
day , Mia 's mom was baking a cake for Mia
saw Mom my carry some dirty plates to the dish washer
trash . " No pasta for you today . You will
but it was too heavy for her . Her daddy helped

INTERVAL 0.051 - 3.040
CONTAINS 48.202%

â € The little girl was very excited
day Anna would wake up early and go to the rock
. She hugged them and asked them what happened .
the ground and noticed it was a leak .
we should always let bugs stay outside where they belong ."

INTERVAL -2.938 - 0.051
CONTAINS 51.019%

of catch when Pete suddenly dropped his glove in the dirt
a brave and tough girl and she liked going on adventures
saw a butterfly . She started to chase the butterfly and
that it was now organized . He smiled and closed the
you make things vanish ?" The magician smiled and said ,

INTERVAL -5.928 - -2.938
CONTAINS 0.535%

loved to play his drum and make lots of noise !
" I 'll make the tail , you make the fin
. He could hear the birds ch ir ping from the
lion . He liked to wander around his jungle looking for
happy ! She thanked the bears for their help and they
Token who
Token ID508
Feature activation+3.10e-01

Pos logit contributions
 cups+0.030
 frost+0.027
ugs+0.027
 twins+0.026
ays+0.026
Neg logit contributions
 shortcut-0.031
lot-0.030
 duel-0.029
 gall-0.028
 noble-0.028
Token had
Token ID550
Feature activation+2.61e+00

Pos logit contributions
 cups+0.008
 frost+0.007
ugs+0.007
ays+0.007
 sisters+0.007
Neg logit contributions
 shortcut-0.008
lot-0.008
 duel-0.008
 gall-0.007
 noble-0.007
Token died
Token ID3724
Feature activation+2.19e+00

Pos logit contributions
 cups+0.065
 frost+0.059
ugs+0.057
 twins+0.057
 sisters+0.057
Neg logit contributions
 shortcut-0.073
lot-0.070
 duel-0.068
 unnecessary-0.065
 farewell-0.065
Token a
Token ID257
Feature activation+7.27e-01

Pos logit contributions
 cups+0.048
 frost+0.044
 twins+0.042
ugs+0.041
 sisters+0.041
Neg logit contributions
 shortcut-0.068
lot-0.066
 duel-0.065
 noble-0.062
 unnecessary-0.062
Token year
Token ID614
Feature activation+7.30e-01

Pos logit contributions
 cups+0.019
 frost+0.017
 twins+0.017
 sisters+0.016
 mug+0.016
Neg logit contributions
lot-0.020
 shortcut-0.020
 duel-0.019
 farewell-0.018
 unnecessary-0.018
Token ago
Token ID2084
Feature activation+6.03e+00

Pos logit contributions
 cups+0.017
 frost+0.015
 twins+0.015
ugs+0.014
 sisters+0.014
Neg logit contributions
 shortcut-0.022
lot-0.021
 duel-0.020
 farewell-0.020
 unnecessary-0.020
Token.
Token ID13
Feature activation+2.28e+00

Pos logit contributions
 cups+0.163
izz+0.162
ays+0.161
olk+0.157
 frost+0.152
Neg logit contributions
lot-0.135
 shortcut-0.131
 gall-0.123
 duel-0.118
 farewell-0.116
Token It
Token ID632
Feature activation+4.74e-01

Pos logit contributions
 cups+0.064
els+0.061
el+0.060
 frost+0.059
ugs+0.058
Neg logit contributions
lot-0.053
 shortcut-0.052
 Sir-0.050
 Skip-0.048
 noble-0.047
Token was
Token ID373
Feature activation+3.29e+00

Pos logit contributions
 cups+0.014
ugs+0.013
ums+0.013
 frost+0.012
ays+0.012
Neg logit contributions
 shortcut-0.011
lot-0.011
 duel-0.010
 gall-0.010
 noble-0.010
Token very
Token ID845
Feature activation+2.15e+00

Pos logit contributions
 cups+0.091
ugs+0.087
ays+0.083
 frost+0.082
ums+0.081
Neg logit contributions
 shortcut-0.080
lot-0.078
 duel-0.074
 noble-0.072
 farewell-0.071
Token precious
Token ID14186
Feature activation+2.09e+00

Pos logit contributions
 cups+0.064
ugs+0.060
 frost+0.060
ays+0.059
ions+0.059
Neg logit contributions
 shortcut-0.047
lot-0.045
 noble-0.043
 duel-0.043
 farewell-0.042
Token someone
Token ID2130
Feature activation+1.87e-01

Pos logit contributions
 cups+0.037
 frost+0.034
 twins+0.033
ugs+0.032
 sisters+0.032
Neg logit contributions
 shortcut-0.044
lot-0.042
 duel-0.041
 unnecessary-0.039
 farewell-0.039
Token a
Token ID257
Feature activation-1.98e-01

Pos logit contributions
 cups+0.004
 frost+0.004
 twins+0.004
 sisters+0.004
el+0.004
Neg logit contributions
 shortcut-0.006
lot-0.005
 duel-0.005
 farewell-0.005
 noble-0.005
Token very
Token ID845
Feature activation+2.19e-01

Pos logit contributions
lot+0.006
 shortcut+0.005
 unnecessary+0.005
 farewell+0.005
 duel+0.005
Neg logit contributions
 cups-0.005
 frost-0.005
 mug-0.004
 twins-0.004
 kettle-0.004
Token long
Token ID890
Feature activation+1.97e+00

Pos logit contributions
 cups+0.006
 twins+0.005
 frost+0.005
 sisters+0.005
 folders+0.005
Neg logit contributions
lot-0.005
 shortcut-0.005
 gall-0.005
 duel-0.005
 Skip-0.005
Token time
Token ID640
Feature activation+2.39e+00

Pos logit contributions
 cups+0.056
 frost+0.052
 twins+0.050
 sisters+0.050
ugs+0.049
Neg logit contributions
 shortcut-0.047
lot-0.046
 duel-0.044
 unnecessary-0.043
 noble-0.042
Token ago
Token ID2084
Feature activation+5.85e+00

Pos logit contributions
 cups+0.036
 frost+0.031
 twins+0.029
 sisters+0.028
ugs+0.028
Neg logit contributions
 shortcut-0.090
lot-0.088
 duel-0.086
 unnecessary-0.083
 farewell-0.083
Token.
Token ID13
Feature activation+7.10e-01

Pos logit contributions
 cups+0.150
ays+0.139
 frost+0.139
izz+0.137
ugs+0.136
Neg logit contributions
lot-0.145
 shortcut-0.145
 gall-0.134
 duel-0.132
 further-0.128
Token The
Token ID383
Feature activation-1.99e-01

Pos logit contributions
 cups+0.019
 frost+0.017
ugs+0.017
els+0.017
 sisters+0.017
Neg logit contributions
 shortcut-0.018
lot-0.018
 duel-0.017
 farewell-0.016
 further-0.016
Token little
Token ID1310
Feature activation+6.62e-01

Pos logit contributions
 shortcut+0.005
lot+0.005
 duel+0.005
 unnecessary+0.005
 gall+0.005
Neg logit contributions
 cups-0.006
 twins-0.005
 frost-0.005
 sisters-0.005
 folders-0.005
Token girl
Token ID2576
Feature activation+4.12e-01

Pos logit contributions
ugs+0.015
 cups+0.015
 frost+0.014
ays+0.014
izz+0.014
Neg logit contributions
 shortcut-0.019
 further-0.019
 higher-0.018
 farther-0.018
 closer-0.017
Token thought
Token ID1807
Feature activation-8.04e-01

Pos logit contributions
 cups+0.011
 frost+0.010
 sisters+0.009
 twins+0.009
el+0.009
Neg logit contributions
 shortcut-0.011
lot-0.011
 duel-0.011
 unnecessary-0.010
 further-0.010
Token who
Token ID508
Feature activation-1.01e-01

Pos logit contributions
 cups+0.037
olk+0.034
 frost+0.033
ays+0.033
ugs+0.033
Neg logit contributions
lot-0.032
 shortcut-0.030
 gall-0.029
 higher-0.029
 further-0.028
Token lived
Token ID5615
Feature activation+1.41e+00

Pos logit contributions
 shortcut+0.003
lot+0.003
 duel+0.003
 unnecessary+0.003
 farewell+0.003
Neg logit contributions
 cups-0.003
 frost-0.002
 twins-0.002
 sisters-0.002
ugs-0.002
Token a
Token ID257
Feature activation+4.64e-01

Pos logit contributions
 cups+0.030
 twins+0.027
 frost+0.026
 sisters+0.025
el+0.025
Neg logit contributions
 shortcut-0.046
lot-0.044
 duel-0.043
 unnecessary-0.042
 farewell-0.041
Token long
Token ID890
Feature activation+2.37e+00

Pos logit contributions
 cups+0.011
 twins+0.011
 frost+0.010
 sisters+0.010
 mug+0.010
Neg logit contributions
 shortcut-0.013
lot-0.013
 unnecessary-0.012
 duel-0.012
 farewell-0.012
Token time
Token ID640
Feature activation+3.04e+00

Pos logit contributions
 cups+0.060
 frost+0.054
ugs+0.053
ays+0.052
 twins+0.051
Neg logit contributions
 shortcut-0.065
lot-0.063
 duel-0.061
 gall-0.058
 farewell-0.058
Token ago
Token ID2084
Feature activation+5.79e+00

Pos logit contributions
 cups+0.072
 frost+0.065
ugs+0.065
olk+0.064
ays+0.063
Neg logit contributions
 shortcut-0.083
lot-0.082
 further-0.078
 duel-0.078
 higher-0.077
Token,"
Token ID553
Feature activation+1.25e+00

Pos logit contributions
izz+0.154
 cups+0.154
ays+0.149
olk+0.147
 frost+0.145
Neg logit contributions
lot-0.130
 shortcut-0.127
 gall-0.120
 duel-0.116
 further-0.115
Token mom
Token ID1995
Feature activation+2.62e+00

Pos logit contributions
ugs+0.028
izz+0.025
 cups+0.025
ays+0.024
cloth+0.023
Neg logit contributions
lot-0.036
 duel-0.036
 farewell-0.035
 gall-0.035
 closer-0.035
Token said
Token ID531
Feature activation+7.90e-01

Pos logit contributions
 cups+0.071
olk+0.067
ugs+0.067
ays+0.067
 folders+0.063
Neg logit contributions
lot-0.064
 gall-0.060
 closer-0.057
 duel-0.056
 shortcut-0.056
Token.
Token ID13
Feature activation+1.07e+00

Pos logit contributions
 cups+0.020
 frost+0.018
ugs+0.017
 twins+0.017
ays+0.017
Neg logit contributions
 shortcut-0.022
lot-0.021
 duel-0.020
 farewell-0.020
 further-0.020
Token
Token ID198
Feature activation+3.73e-03

Pos logit contributions
 cups+0.033
el+0.032
 frost+0.032
ugs+0.032
els+0.032
Neg logit contributions
lot-0.021
 shortcut-0.020
 Skip-0.018
 duel-0.018
 closer-0.018
Token important
Token ID1593
Feature activation-4.05e-01

Pos logit contributions
 shortcut+0.006
lot+0.006
 duel+0.006
 unnecessary+0.006
 gall+0.006
Neg logit contributions
 cups-0.005
 frost-0.005
 twins-0.005
 sisters-0.005
el-0.005
Token person
Token ID1048
Feature activation+7.16e-01

Pos logit contributions
 unnecessary+0.010
lot+0.010
 shortcut+0.010
 duel+0.010
 higher+0.010
Neg logit contributions
 cups-0.010
 twins-0.010
 mug-0.009
 folders-0.009
 frost-0.009
Token a
Token ID257
Feature activation-8.65e-01

Pos logit contributions
 cups+0.017
 frost+0.015
 twins+0.015
 sisters+0.015
ugs+0.015
Neg logit contributions
 shortcut-0.021
lot-0.020
 duel-0.019
 unnecessary-0.019
 farewell-0.019
Token long
Token ID890
Feature activation+1.67e+00

Pos logit contributions
lot+0.022
 nearer+0.022
 unnecessary+0.022
 farther+0.022
 higher+0.021
Neg logit contributions
 cups-0.021
 twins-0.020
 sisters-0.019
 mug-0.019
 cup-0.019
Token time
Token ID640
Feature activation+2.83e+00

Pos logit contributions
 cups+0.043
 twins+0.040
 frost+0.039
 sisters+0.039
cloth+0.038
Neg logit contributions
 shortcut-0.043
lot-0.042
 unnecessary-0.041
 duel-0.041
 higher-0.040
Token ago
Token ID2084
Feature activation+5.61e+00

Pos logit contributions
 cups+0.045
 frost+0.039
ugs+0.037
ays+0.035
 twins+0.035
Neg logit contributions
 shortcut-0.103
lot-0.101
 duel-0.098
 gall-0.095
 further-0.095
Token.
Token ID13
Feature activation+6.47e-01

Pos logit contributions
 cups+0.141
 frost+0.129
ays+0.128
izz+0.126
ugs+0.124
Neg logit contributions
 shortcut-0.144
lot-0.144
 gall-0.135
 duel-0.134
 farewell-0.129
Token She
Token ID1375
Feature activation+1.67e+00

Pos logit contributions
els+0.017
 frost+0.017
 cups+0.017
ugs+0.016
el+0.016
Neg logit contributions
lot-0.015
 shortcut-0.015
 farewell-0.014
 Skip-0.014
 duel-0.013
Token was
Token ID373
Feature activation+1.76e+00

Pos logit contributions
 cups+0.040
 frost+0.036
ugs+0.036
 twins+0.034
ays+0.034
Neg logit contributions
 shortcut-0.048
lot-0.047
 duel-0.045
 gall-0.044
 noble-0.043
Token excited
Token ID6568
Feature activation+6.55e-01

Pos logit contributions
 cups+0.053
 frost+0.049
 twins+0.047
ugs+0.047
 sisters+0.047
Neg logit contributions
 shortcut-0.041
lot-0.039
 duel-0.037
 unnecessary-0.035
 farewell-0.035
Token and
Token ID290
Feature activation-1.26e-01

Pos logit contributions
 cups+0.018
 frost+0.017
 twins+0.016
 sisters+0.016
ugs+0.016
Neg logit contributions
 shortcut-0.017
lot-0.016
 duel-0.016
 unnecessary-0.015
 noble-0.015
Token were
Token ID547
Feature activation+2.36e+00

Pos logit contributions
 cups+0.030
 frost+0.028
ugs+0.027
 twins+0.027
ays+0.027
Neg logit contributions
 shortcut-0.027
lot-0.026
 duel-0.025
 noble-0.024
 gall-0.024
Token from
Token ID422
Feature activation+2.75e+00

Pos logit contributions
 cups+0.069
 frost+0.064
ugs+0.062
 twins+0.062
 sisters+0.061
Neg logit contributions
 shortcut-0.056
lot-0.054
 duel-0.052
 unnecessary-0.049
 noble-0.049
Token a
Token ID257
Feature activation+9.37e-01

Pos logit contributions
 cups+0.067
 frost+0.061
ugs+0.059
 twins+0.058
 sisters+0.058
Neg logit contributions
 shortcut-0.078
lot-0.075
 duel-0.073
 unnecessary-0.070
 farewell-0.070
Token long
Token ID890
Feature activation+2.63e+00

Pos logit contributions
 cups+0.025
 twins+0.022
 frost+0.022
 sisters+0.021
 mug+0.021
Neg logit contributions
lot-0.026
 shortcut-0.025
 unnecessary-0.023
 Skip-0.023
 gall-0.023
Token time
Token ID640
Feature activation+3.02e+00

Pos logit contributions
 cups+0.072
 frost+0.067
ugs+0.066
ays+0.065
 twins+0.065
Neg logit contributions
 shortcut-0.065
lot-0.062
 duel-0.061
 farewell-0.058
 further-0.057
Token ago
Token ID2084
Feature activation+5.58e+00

Pos logit contributions
 cups+0.071
 frost+0.065
ugs+0.062
 twins+0.062
 sisters+0.061
Neg logit contributions
 shortcut-0.087
lot-0.084
 duel-0.082
 unnecessary-0.078
 further-0.078
Token and
Token ID290
Feature activation+1.28e+00

Pos logit contributions
 cups+0.139
 frost+0.127
ays+0.125
ugs+0.124
izz+0.123
Neg logit contributions
 shortcut-0.145
lot-0.144
 duel-0.134
 gall-0.133
 farewell-0.130
Token they
Token ID484
Feature activation+1.14e+00

Pos logit contributions
 cups+0.032
 frost+0.029
 twins+0.028
ugs+0.028
 sisters+0.028
Neg logit contributions
 shortcut-0.036
lot-0.034
 duel-0.033
 unnecessary-0.032
 farewell-0.032
Token help
Token ID1037
Feature activation+1.72e+00

Pos logit contributions
 cups+0.029
 frost+0.027
 twins+0.026
ugs+0.026
 sisters+0.026
Neg logit contributions
 shortcut-0.031
lot-0.030
 duel-0.029
 unnecessary-0.027
 farewell-0.027
Token us
Token ID514
Feature activation-5.52e-01

Pos logit contributions
 cups+0.040
 frost+0.036
ugs+0.035
 twins+0.034
ays+0.034
Neg logit contributions
 shortcut-0.051
lot-0.049
 duel-0.048
 unnecessary-0.046
 further-0.046
Token remember
Token ID3505
Feature activation+1.51e+00

Pos logit contributions
 shortcut+0.015
lot+0.014
 duel+0.014
 farewell+0.013
 noble+0.013
Neg logit contributions
 cups-0.015
 frost-0.013
 twins-0.013
 sisters-0.013
ugs-0.013
Token her
Token ID607
Feature activation+8.91e-01

Pos logit contributions
 cups+0.048
 frost+0.042
 twins+0.039
ugs+0.039
 sisters+0.038
Neg logit contributions
 shortcut-0.108
lot-0.105
 duel-0.102
 unnecessary-0.099
 farewell-0.099
Token birthday
Token ID10955
Feature activation+2.77e+00

Pos logit contributions
 cups+0.019
 twins+0.017
 sisters+0.017
 frost+0.017
 mug+0.016
Neg logit contributions
 shortcut-0.028
lot-0.028
 duel-0.027
 unnecessary-0.026
 noble-0.025
Token a
Token ID257
Feature activation-7.61e-01

Pos logit contributions
 cups+0.066
 frost+0.060
ays+0.059
ugs+0.058
 twins+0.058
Neg logit contributions
 shortcut-0.078
lot-0.076
 duel-0.073
 gall-0.070
 farewell-0.069
Token few
Token ID1178
Feature activation+2.47e+00

Pos logit contributions
lot+0.020
 unnecessary+0.019
 shortcut+0.019
 nearer+0.019
 ashore+0.018
Neg logit contributions
 cups-0.020
 mug-0.019
 frost-0.018
 twins-0.018
 sisters-0.017
Token days
Token ID1528
Feature activation+3.21e+00

Pos logit contributions
 cups+0.056
 frost+0.051
ugs+0.049
 twins+0.048
ays+0.048
Neg logit contributions
 shortcut-0.074
lot-0.071
 duel-0.070
 farewell-0.067
 noble-0.067
Token ago
Token ID2084
Feature activation+5.48e+00

Pos logit contributions
 cups+0.081
 frost+0.075
olk+0.072
 folders+0.072
ays+0.071
Neg logit contributions
 shortcut-0.081
lot-0.080
 gall-0.078
 further-0.076
 duel-0.075
Token.
Token ID13
Feature activation+6.54e-01

Pos logit contributions
 cups+0.141
ays+0.135
izz+0.131
 frost+0.129
olk+0.128
Neg logit contributions
lot-0.134
 shortcut-0.130
 gall-0.126
 farewell-0.120
 duel-0.120
Token She
Token ID1375
Feature activation+1.36e+00

Pos logit contributions
 cups+0.018
 frost+0.017
ugs+0.016
ays+0.015
 twins+0.015
Neg logit contributions
 shortcut-0.016
lot-0.016
 duel-0.015
 farewell-0.015
 further-0.015
Token liked
Token ID8288
Feature activation+1.98e-01

Pos logit contributions
 cups+0.035
 frost+0.032
 twins+0.031
 sisters+0.031
ugs+0.031
Neg logit contributions
 shortcut-0.037
lot-0.036
 duel-0.035
 unnecessary-0.033
 farewell-0.033
Token to
Token ID284
Feature activation-7.58e-02

Pos logit contributions
 cups+0.005
 frost+0.004
 twins+0.004
 sisters+0.004
 mug+0.004
Neg logit contributions
 shortcut-0.006
lot-0.006
 duel-0.005
 unnecessary-0.005
 noble-0.005
Token practice
Token ID3357
Feature activation+3.96e-01

Pos logit contributions
 shortcut+0.002
lot+0.002
 duel+0.002
 farewell+0.002
 unnecessary+0.002
Neg logit contributions
 cups-0.002
 frost-0.002
 twins-0.002
ugs-0.002
 sisters-0.002
Token same
Token ID976
Feature activation+2.29e+00

Pos logit contributions
 cups+0.014
 frost+0.013
ugs+0.012
 twins+0.012
 sisters+0.012
Neg logit contributions
 shortcut-0.022
lot-0.021
 duel-0.020
 farewell-0.019
 gall-0.019
Token dress
Token ID6576
Feature activation+1.00e+00

Pos logit contributions
 cups+0.036
ugs+0.031
 frost+0.031
ays+0.030
 twins+0.029
Neg logit contributions
 shortcut-0.084
lot-0.081
 duel-0.079
 farewell-0.078
 gall-0.077
Token a
Token ID257
Feature activation+6.66e-01

Pos logit contributions
 cups+0.023
 frost+0.020
 twins+0.019
ugs+0.019
 sisters+0.019
Neg logit contributions
 shortcut-0.031
lot-0.030
 duel-0.029
 unnecessary-0.029
 noble-0.028
Token long
Token ID890
Feature activation+2.83e+00

Pos logit contributions
 cups+0.015
 frost+0.014
 twins+0.013
 sisters+0.013
 mug+0.013
Neg logit contributions
 shortcut-0.020
lot-0.020
 duel-0.019
 unnecessary-0.019
 farewell-0.018
Token time
Token ID640
Feature activation+2.12e+00

Pos logit contributions
 cups+0.058
 frost+0.051
ugs+0.050
 twins+0.050
ays+0.049
Neg logit contributions
 shortcut-0.091
lot-0.089
 duel-0.086
 farewell-0.083
 gall-0.083
Token ago
Token ID2084
Feature activation+5.23e+00

Pos logit contributions
 cups+0.048
 frost+0.044
ugs+0.042
 twins+0.042
ays+0.042
Neg logit contributions
 shortcut-0.062
lot-0.061
 duel-0.058
 further-0.057
 gall-0.057
Token.
Token ID13
Feature activation+3.16e-01

Pos logit contributions
 cups+0.132
ays+0.122
izz+0.122
 frost+0.121
ugs+0.118
Neg logit contributions
lot-0.133
 shortcut-0.131
 gall-0.124
 duel-0.123
 farewell-0.119
Token<|endoftext|>
Token ID50256
Feature activation-4.67e-01

Pos logit contributions
 cups+0.009
 frost+0.008
els+0.008
ugs+0.008
cloth+0.008
Neg logit contributions
lot-0.007
 shortcut-0.007
 farewell-0.007
 duel-0.007
 noble-0.007
TokenOnce
Token ID7454
Feature activation+2.04e-01

Pos logit contributions
 shortcut+0.013
lot+0.013
 duel+0.012
 unnecessary+0.012
 noble+0.012
Neg logit contributions
 cups-0.011
 frost-0.010
 twins-0.010
ugs-0.010
 sisters-0.010
Token upon
Token ID2402
Feature activation-9.27e-01

Pos logit contributions
 cups+0.007
 frost+0.006
 twins+0.006
ugs+0.006
 sisters+0.006
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.003
 farewell-0.003
 further-0.003
Token a
Token ID257
Feature activation-2.31e-01

Pos logit contributions
 shortcut+0.029
lot+0.028
 duel+0.027
 unnecessary+0.026
 noble+0.026
Neg logit contributions
 cups-0.020
 frost-0.019
 twins-0.018
 sisters-0.017
ugs-0.017
Token this
Token ID428
Feature activation+5.55e-01

Pos logit contributions
 cups+0.048
ugs+0.046
 frost+0.045
ays+0.045
ums+0.044
Neg logit contributions
 shortcut-0.047
 duel-0.042
 further-0.042
lot-0.042
 unnecessary-0.041
Token house
Token ID2156
Feature activation+1.52e+00

Pos logit contributions
 cups+0.013
 frost+0.012
 twins+0.011
el+0.011
 mug+0.011
Neg logit contributions
lot-0.017
 shortcut-0.017
 duel-0.016
 unnecessary-0.015
 Skip-0.015
Token a
Token ID257
Feature activation+3.25e-01

Pos logit contributions
 cups+0.037
 frost+0.033
 twins+0.032
ugs+0.032
 sisters+0.032
Neg logit contributions
 shortcut-0.044
lot-0.042
 duel-0.041
 unnecessary-0.039
 farewell-0.039
Token long
Token ID890
Feature activation+2.54e+00

Pos logit contributions
 cups+0.009
 frost+0.008
 twins+0.008
 mug+0.007
 sisters+0.007
Neg logit contributions
lot-0.009
 shortcut-0.009
 unnecessary-0.008
 farewell-0.008
 duel-0.008
Token time
Token ID640
Feature activation+2.15e+00

Pos logit contributions
 cups+0.072
 frost+0.067
ays+0.066
ugs+0.066
 twins+0.065
Neg logit contributions
 shortcut-0.062
lot-0.058
 duel-0.056
 gall-0.054
 farewell-0.054
Token ago
Token ID2084
Feature activation+5.22e+00

Pos logit contributions
 cups+0.035
 frost+0.030
ugs+0.029
ays+0.028
el+0.028
Neg logit contributions
 shortcut-0.077
lot-0.075
 duel-0.073
 further-0.071
 gall-0.071
Token."
Token ID526
Feature activation+6.40e-01

Pos logit contributions
 cups+0.134
izz+0.127
ays+0.125
 frost+0.124
olk+0.122
Neg logit contributions
lot-0.128
 shortcut-0.126
 gall-0.117
 duel-0.116
 farewell-0.112
Token
Token ID198
Feature activation+2.46e-01

Pos logit contributions
 cups+0.017
 frost+0.017
els+0.016
el+0.016
ugs+0.016
Neg logit contributions
 shortcut-0.015
lot-0.015
 farewell-0.014
 duel-0.013
 higher-0.013
Token
Token ID198
Feature activation+1.44e+00

Pos logit contributions
 cups+0.007
 frost+0.006
 sisters+0.006
ugs+0.006
 twins+0.006
Neg logit contributions
 shortcut-0.006
lot-0.006
 duel-0.006
 farewell-0.006
 gall-0.005
TokenL
Token ID43
Feature activation+1.17e+00

Pos logit contributions
 cups+0.032
 frost+0.030
ays+0.029
ugs+0.028
el+0.028
Neg logit contributions
 shortcut-0.043
lot-0.042
Hop-0.042
 duel-0.041
 further-0.040
Tokenily
Token ID813
Feature activation+8.50e-01

Pos logit contributions
 cups+0.021
 frost+0.018
 kettle+0.017
 mug+0.016
 Lem+0.016
Neg logit contributions
lot-0.039
 shortcut-0.039
 further-0.037
 farewell-0.037
Hop-0.036
Token had
Token ID550
Feature activation+2.67e+00

Pos logit contributions
 cups+0.033
 frost+0.031
 twins+0.029
 sisters+0.029
ugs+0.028
Neg logit contributions
 shortcut-0.041
lot-0.039
 duel-0.038
 unnecessary-0.037
 farewell-0.036
Token bought
Token ID5839
Feature activation+1.55e+00

Pos logit contributions
 cups+0.067
 frost+0.062
 twins+0.058
 sisters+0.057
cloth+0.056
Neg logit contributions
 shortcut-0.075
lot-0.073
 duel-0.070
 unnecessary-0.068
 noble-0.067
Token a
Token ID257
Feature activation-7.03e-02

Pos logit contributions
 cups+0.037
 frost+0.034
 twins+0.032
 sisters+0.031
el+0.031
Neg logit contributions
 shortcut-0.046
lot-0.045
 duel-0.044
 unnecessary-0.042
 noble-0.042
Token long
Token ID890
Feature activation+2.00e+00

Pos logit contributions
lot+0.002
 shortcut+0.002
 duel+0.002
 unnecessary+0.002
 farewell+0.002
Neg logit contributions
 cups-0.002
 frost-0.002
 mug-0.002
 twins-0.002
 kettle-0.001
Token time
Token ID640
Feature activation+2.04e+00

Pos logit contributions
 cups+0.049
 frost+0.044
 twins+0.043
ugs+0.043
 sisters+0.042
Neg logit contributions
 shortcut-0.057
lot-0.055
 duel-0.053
 farewell-0.051
 gall-0.051
Token ago
Token ID2084
Feature activation+5.20e+00

Pos logit contributions
 cups+0.046
 frost+0.042
ugs+0.040
 sisters+0.040
 twins+0.039
Neg logit contributions
 shortcut-0.060
lot-0.058
 duel-0.056
 further-0.054
 gall-0.054
Token.
Token ID13
Feature activation+7.40e-01

Pos logit contributions
 cups+0.130
ays+0.121
 frost+0.121
izz+0.120
ugs+0.118
Neg logit contributions
 shortcut-0.131
lot-0.131
 gall-0.122
 duel-0.121
 further-0.118
Token They
Token ID1119
Feature activation+1.42e+00

Pos logit contributions
 cups+0.020
 frost+0.019
ugs+0.018
 sisters+0.018
els+0.018
Neg logit contributions
 shortcut-0.019
lot-0.018
 duel-0.017
 Skip-0.017
 farewell-0.017
Token did
Token ID750
Feature activation-3.60e-01

Pos logit contributions
 cups+0.037
ugs+0.034
 frost+0.034
ays+0.033
 twins+0.033
Neg logit contributions
 shortcut-0.037
lot-0.036
 duel-0.035
 gall-0.034
 farewell-0.033
Token not
Token ID407
Feature activation+7.24e-01

Pos logit contributions
 shortcut+0.011
lot+0.010
Hop+0.010
 duel+0.010
 unnecessary+0.010
Neg logit contributions
 cups-0.009
 frost-0.008
 twins-0.008
 sisters-0.007
cloth-0.007
Token know
Token ID760
Feature activation-2.39e-01

Pos logit contributions
 cups+0.023
 frost+0.021
ugs+0.021
ays+0.021
 twins+0.021
Neg logit contributions
 shortcut-0.015
lot-0.015
 duel-0.014
 gall-0.013
 unnecessary-0.013
Token would
Token ID561
Feature activation+1.95e+00

Pos logit contributions
 cups+0.090
ums+0.088
ugs+0.088
ays+0.088
ions+0.085
Neg logit contributions
lot-0.033
 shortcut-0.032
'd-0.032
 gall-0.031
 duel-0.031
Token take
Token ID1011
Feature activation+1.69e+00

Pos logit contributions
 cups+0.046
ays+0.045
ugs+0.044
 twins+0.042
ums+0.041
Neg logit contributions
 shortcut-0.053
 duel-0.051
 gall-0.050
lot-0.050
 farewell-0.049
Token out
Token ID503
Feature activation+1.18e+00

Pos logit contributions
 cups+0.044
ays+0.040
 sisters+0.040
 twins+0.040
ugs+0.039
Neg logit contributions
 shortcut-0.043
lot-0.042
 duel-0.040
 farewell-0.040
 unnecessary-0.040
Token her
Token ID607
Feature activation+4.25e+00

Pos logit contributions
ays+0.033
ums+0.032
ugs+0.031
 tub+0.030
izz+0.030
Neg logit contributions
 farewell-0.023
 further-0.023
 closer-0.023
 higher-0.022
 shortcut-0.022
Token depend
Token ID4745
Feature activation+1.96e+00

Pos logit contributions
ays+0.079
izz+0.076
ums+0.076
cloth+0.072
ions+0.071
Neg logit contributions
 shortcut-0.125
 closer-0.119
 hike-0.117
 higher-0.115
 noble-0.115
Tokenable
Token ID540
Feature activation+5.16e+00

Pos logit contributions
 cups+0.060
 sisters+0.058
 twins+0.055
 frost+0.054
ums+0.053
Neg logit contributions
 shortcut-0.038
 further-0.036
away-0.036
lot-0.035
 unnecessary-0.034
Token oven
Token ID14361
Feature activation+3.16e+00

Pos logit contributions
izz+0.104
ays+0.104
ions+0.092
ugs+0.089
ums+0.089
Neg logit contributions
 shortcut-0.155
 farewell-0.142
 hike-0.142
 duel-0.142
 spear-0.139
Token m
Token ID285
Feature activation+2.98e+00

Pos logit contributions
izz+0.113
 cups+0.110
ays+0.108
 sisters+0.107
 frost+0.106
Neg logit contributions
lot-0.044
 shortcut-0.043
 higher-0.037
 hike-0.037
 gall-0.036
Tokenitt
Token ID715
Feature activation+6.63e-01

Pos logit contributions
 cups+0.060
 folders+0.054
 sisters+0.053
 tub+0.051
 mug+0.051
Neg logit contributions
lot-0.083
to-0.083
 shortcut-0.081
stru-0.080
 farewell-0.080
Tokens
Token ID82
Feature activation+2.56e+00

Pos logit contributions
 cups+0.016
 frost+0.014
 twins+0.014
 sisters+0.014
ugs+0.013
Neg logit contributions
 shortcut-0.019
lot-0.019
 duel-0.018
 gall-0.017
 unnecessary-0.017
Token and
Token ID290
Feature activation+1.16e+00

Pos logit contributions
 cups+0.069
 frost+0.064
ays+0.064
 sisters+0.064
 twins+0.063
Neg logit contributions
 shortcut-0.061
lot-0.058
 duel-0.054
 further-0.054
 higher-0.054
Token it
Token ID340
Feature activation-5.32e-01

Pos logit contributions
 cups+0.004
 frost+0.004
 twins+0.004
ugs+0.004
 sisters+0.004
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.003
 unnecessary-0.003
 farewell-0.003
Token away
Token ID1497
Feature activation+6.26e-01

Pos logit contributions
lot+0.014
 shortcut+0.014
 duel+0.014
 noble+0.013
 gall+0.013
Neg logit contributions
 cups-0.014
 frost-0.013
cloth-0.012
 twins-0.012
ugs-0.012
Token a
Token ID257
Feature activation+3.23e-01

Pos logit contributions
 cups+0.015
 frost+0.013
 twins+0.013
ugs+0.013
 sisters+0.013
Neg logit contributions
 shortcut-0.018
lot-0.018
 duel-0.017
 unnecessary-0.016
 gall-0.016
Token long
Token ID890
Feature activation+2.67e+00

Pos logit contributions
 cups+0.008
 frost+0.007
 mug+0.007
 twins+0.007
 sisters+0.007
Neg logit contributions
lot-0.009
 shortcut-0.009
 duel-0.008
 farewell-0.008
 gall-0.008
Token time
Token ID640
Feature activation+2.28e+00

Pos logit contributions
 cups+0.085
 frost+0.080
ays+0.080
ugs+0.079
 sisters+0.079
Neg logit contributions
 shortcut-0.056
lot-0.051
 duel-0.050
 farewell-0.047
 further-0.046
Token ago
Token ID2084
Feature activation+5.16e+00

Pos logit contributions
 cups+0.051
 frost+0.047
ugs+0.046
el+0.045
ays+0.045
Neg logit contributions
 shortcut-0.066
lot-0.064
 duel-0.062
 further-0.061
 higher-0.060
Token.
Token ID13
Feature activation+1.21e+00

Pos logit contributions
 cups+0.134
izz+0.127
 frost+0.126
ays+0.124
olk+0.123
Neg logit contributions
 shortcut-0.124
lot-0.124
 duel-0.113
 gall-0.112
 farewell-0.109
Token It
Token ID632
Feature activation-9.77e-01

Pos logit contributions
 cups+0.034
 frost+0.030
ugs+0.030
els+0.029
 sisters+0.028
Neg logit contributions
lot-0.029
 shortcut-0.029
 Skip-0.027
 noble-0.026
 duel-0.026
Token is
Token ID318
Feature activation+2.44e+00

Pos logit contributions
 shortcut+0.027
lot+0.026
 duel+0.025
 unnecessary+0.024
 farewell+0.024
Neg logit contributions
 cups-0.025
 frost-0.023
 twins-0.022
 sisters-0.022
ugs-0.021
Token not
Token ID407
Feature activation+7.46e-01

Pos logit contributions
 cups+0.073
 frost+0.068
ugs+0.068
ays+0.066
 twins+0.066
Neg logit contributions
 shortcut-0.056
lot-0.053
 duel-0.051
 unnecessary-0.048
 noble-0.048
Token good
Token ID922
Feature activation+2.14e+00

Pos logit contributions
 cups+0.021
 frost+0.020
 twins+0.019
ugs+0.019
 sisters+0.019
Neg logit contributions
 shortcut-0.018
lot-0.017
 duel-0.017
 unnecessary-0.016
 farewell-0.016
Token they
Token ID484
Feature activation-1.32e+00

Pos logit contributions
 shortcut+0.039
lot+0.037
 duel+0.036
 gall+0.034
 farewell+0.034
Neg logit contributions
 cups-0.031
 twins-0.027
 sisters-0.026
 frost-0.026
ugs-0.026
Token lived
Token ID5615
Feature activation+7.05e-02

Pos logit contributions
 shortcut+0.044
lot+0.043
 unnecessary+0.042
 farewell+0.041
 duel+0.039
Neg logit contributions
 cups-0.027
 frost-0.025
 sisters-0.023
 twins-0.022
 paints-0.022
Token a
Token ID257
Feature activation-1.74e+00

Pos logit contributions
 cups+0.002
 twins+0.001
 frost+0.001
 sisters+0.001
ions+0.001
Neg logit contributions
 shortcut-0.002
 unnecessary-0.002
lot-0.002
 duel-0.002
 farewell-0.002
Token long
Token ID890
Feature activation+1.45e-01

Pos logit contributions
 unnecessary+0.046
lot+0.046
 shortcut+0.045
 nearer+0.044
rated+0.043
Neg logit contributions
 cups-0.047
 twins-0.043
 sisters-0.041
 frost-0.039
 mug-0.039
Token time
Token ID640
Feature activation+1.41e+00

Pos logit contributions
 cups+0.004
 frost+0.003
 twins+0.003
 sisters+0.003
 mug+0.003
Neg logit contributions
 shortcut-0.004
 unnecessary-0.004
lot-0.004
 duel-0.004
 nearer-0.004
Token ago
Token ID2084
Feature activation+5.14e+00

Pos logit contributions
 cups+0.031
 frost+0.028
 twins+0.027
 sisters+0.026
ugs+0.026
Neg logit contributions
 shortcut-0.044
lot-0.043
 duel-0.041
 unnecessary-0.040
 farewell-0.040
Token.
Token ID13
Feature activation+9.68e-01

Pos logit contributions
 cups+0.124
 frost+0.114
ays+0.111
ugs+0.111
izz+0.109
Neg logit contributions
 shortcut-0.138
lot-0.137
 duel-0.129
 gall-0.128
 further-0.124
Token Lily
Token ID20037
Feature activation-7.47e-01

Pos logit contributions
 cups+0.027
 frost+0.026
els+0.026
ugs+0.025
el+0.025
Neg logit contributions
lot-0.021
 shortcut-0.021
 farewell-0.020
Hop-0.019
 Skip-0.019
Token listened
Token ID16399
Feature activation-1.07e+00

Pos logit contributions
 shortcut+0.021
lot+0.020
 duel+0.020
 unnecessary+0.020
 further+0.019
Neg logit contributions
 cups-0.019
 frost-0.017
 sisters-0.017
 twins-0.017
el-0.016
Token carefully
Token ID7773
Feature activation-4.81e-01

Pos logit contributions
 shortcut+0.033
 duel+0.030
 unnecessary+0.030
lot+0.030
 gall+0.028
Neg logit contributions
 cups-0.025
 frost-0.022
 twins-0.021
 sisters-0.021
 folders-0.020
Token and
Token ID290
Feature activation+1.33e-01

Pos logit contributions
 shortcut+0.016
lot+0.016
 duel+0.016
 unnecessary+0.015
Hop+0.015
Neg logit contributions
 cups-0.009
 frost-0.008
 twins-0.008
 sisters-0.008
cloth-0.007
Token a
Token ID257
Feature activation+2.17e-02

Pos logit contributions
 cups+0.036
 frost+0.033
 twins+0.032
 sisters+0.031
 folders+0.030
Neg logit contributions
 shortcut-0.040
lot-0.039
 duel-0.037
 noble-0.035
 farewell-0.035
Token ship
Token ID4074
Feature activation+7.47e-01

Pos logit contributions
 cups+0.001
 frost+0.000
 mug+0.000
 twins+0.000
 kettle+0.000
Neg logit contributions
lot-0.001
 shortcut-0.001
 farewell-0.001
 unnecessary-0.001
 duel-0.001
Token that
Token ID326
Feature activation+1.42e-01

Pos logit contributions
 cups+0.015
 frost+0.013
 twins+0.012
cloth+0.012
 sisters+0.012
Neg logit contributions
 shortcut-0.025
lot-0.025
 unnecessary-0.024
 duel-0.024
 noble-0.023
Token sank
Token ID30895
Feature activation+1.43e+00

Pos logit contributions
 cups+0.004
 frost+0.003
 twins+0.003
 sisters+0.003
ugs+0.003
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.004
 unnecessary-0.004
 farewell-0.003
Token long
Token ID890
Feature activation+2.69e+00

Pos logit contributions
 cups+0.032
 twins+0.028
cloth+0.027
 frost+0.027
 sisters+0.026
Neg logit contributions
 shortcut-0.044
lot-0.044
 duel-0.043
 noble-0.042
 Skip-0.042
Token ago
Token ID2084
Feature activation+5.12e+00

Pos logit contributions
 cups+0.059
 frost+0.053
ugs+0.051
 twins+0.051
 sisters+0.051
Neg logit contributions
 shortcut-0.082
lot-0.079
 duel-0.077
 further-0.074
 gall-0.074
Token.
Token ID13
Feature activation+1.67e+00

Pos logit contributions
 cups+0.127
ays+0.119
 frost+0.118
izz+0.118
ugs+0.115
Neg logit contributions
 shortcut-0.132
lot-0.132
 duel-0.122
 gall-0.120
 farewell-0.119
Token She
Token ID1375
Feature activation+8.10e-01

Pos logit contributions
 cups+0.042
els+0.042
 sisters+0.041
el+0.040
 frost+0.040
Neg logit contributions
 shortcut-0.038
lot-0.038
 farewell-0.036
 Skip-0.036
 higher-0.035
Token said
Token ID531
Feature activation-6.12e-01

Pos logit contributions
 cups+0.020
 frost+0.018
ugs+0.017
 twins+0.017
 sisters+0.017
Neg logit contributions
 shortcut-0.023
lot-0.023
 duel-0.022
 gall-0.021
 noble-0.021
Token they
Token ID484
Feature activation-3.55e-01

Pos logit contributions
 shortcut+0.017
lot+0.017
 duel+0.016
 farewell+0.016
 unnecessary+0.016
Neg logit contributions
 cups-0.015
 frost-0.014
ugs-0.013
 twins-0.013
 sisters-0.013
Token could
Token ID714
Feature activation+3.19e-02

Pos logit contributions
 shortcut+0.010
lot+0.009
 duel+0.009
 unnecessary+0.009
 farewell+0.009
Neg logit contributions
 cups-0.009
 frost-0.008
 twins-0.008
ugs-0.008
 sisters-0.008
Token is
Token ID318
Feature activation+1.44e+00

Pos logit contributions
 shortcut+0.040
lot+0.039
 duel+0.038
 unnecessary+0.036
 farewell+0.036
Neg logit contributions
 cups-0.035
 frost-0.032
 twins-0.030
 sisters-0.030
ugs-0.029
Token from
Token ID422
Feature activation+2.11e+00

Pos logit contributions
 cups+0.041
ugs+0.039
 frost+0.038
ays+0.037
 sisters+0.037
Neg logit contributions
 shortcut-0.034
lot-0.033
 duel-0.031
 noble-0.030
 gall-0.029
Token a
Token ID257
Feature activation-2.72e-01

Pos logit contributions
 cups+0.050
 frost+0.046
 twins+0.044
 sisters+0.043
ugs+0.043
Neg logit contributions
 shortcut-0.062
lot-0.060
 duel-0.058
 farewell-0.055
 unnecessary-0.055
Token long
Token ID890
Feature activation+1.87e+00

Pos logit contributions
lot+0.008
 shortcut+0.007
 unnecessary+0.007
 farewell+0.007
 nearer+0.007
Neg logit contributions
 cups-0.007
 frost-0.006
 mug-0.006
 twins-0.006
 tub-0.006
Token time
Token ID640
Feature activation+2.02e+00

Pos logit contributions
 cups+0.052
 frost+0.047
ugs+0.046
 twins+0.046
 sisters+0.046
Neg logit contributions
 shortcut-0.047
lot-0.045
 duel-0.044
 gall-0.041
 farewell-0.041
Token ago
Token ID2084
Feature activation+5.11e+00

Pos logit contributions
 cups+0.042
 frost+0.038
ugs+0.037
 sisters+0.036
ays+0.036
Neg logit contributions
 shortcut-0.062
lot-0.060
 duel-0.058
 further-0.057
 unnecessary-0.056
Token,
Token ID11
Feature activation+3.54e-01

Pos logit contributions
 cups+0.130
izz+0.125
ays+0.123
 frost+0.123
olk+0.119
Neg logit contributions
lot-0.122
 shortcut-0.122
 gall-0.111
 duel-0.111
 farewell-0.110
Token when
Token ID618
Feature activation+1.24e+00

Pos logit contributions
 cups+0.010
ugs+0.010
 frost+0.009
 sisters+0.009
 twins+0.009
Neg logit contributions
 shortcut-0.008
lot-0.008
 duel-0.008
 further-0.007
 higher-0.007
Token a
Token ID257
Feature activation-2.44e-01

Pos logit contributions
 cups+0.031
 frost+0.028
 twins+0.027
 sisters+0.027
ugs+0.026
Neg logit contributions
 shortcut-0.035
lot-0.034
 duel-0.032
 farewell-0.031
 gall-0.031
Token king
Token ID5822
Feature activation+6.07e-01

Pos logit contributions
lot+0.007
 shortcut+0.007
 unnecessary+0.006
 duel+0.006
 gall+0.006
Neg logit contributions
 cups-0.006
 frost-0.006
 twins-0.006
 mug-0.005
 sisters-0.005
Token lived
Token ID5615
Feature activation+1.61e+00

Pos logit contributions
 cups+0.016
 frost+0.014
 twins+0.014
ugs+0.014
 sisters+0.014
Neg logit contributions
 shortcut-0.016
lot-0.016
 duel-0.015
 unnecessary-0.015
 farewell-0.014
Token She
Token ID1375
Feature activation-1.98e-01

Pos logit contributions
els+0.021
 cups+0.020
 frost+0.020
ays+0.019
ries+0.019
Neg logit contributions
 shortcut-0.032
 higher-0.031
lot-0.031
 Nice-0.031
 steep-0.031
Token had
Token ID550
Feature activation+2.09e+00

Pos logit contributions
lot+0.004
 shortcut+0.004
 gall+0.004
 duel+0.004
 noble+0.003
Neg logit contributions
 cups-0.007
ays-0.006
ugs-0.006
 frost-0.006
els-0.006
Token a
Token ID257
Feature activation+1.31e+00

Pos logit contributions
ays+0.067
ugs+0.066
els+0.066
 cups+0.065
cloth+0.062
Neg logit contributions
lot-0.040
 shortcut-0.038
 gall-0.038
 unnecessary-0.037
 duel-0.035
Token brush
Token ID14093
Feature activation+2.02e+00

Pos logit contributions
 cups+0.036
 frost+0.034
ugs+0.033
ays+0.033
 twins+0.032
Neg logit contributions
 shortcut-0.033
lot-0.031
 duel-0.030
 noble-0.029
 gall-0.029
Token with
Token ID351
Feature activation+1.80e+00

Pos logit contributions
 cups+0.059
izz+0.056
ugs+0.055
ays+0.055
 frost+0.055
Neg logit contributions
 shortcut-0.042
 higher-0.040
lot-0.040
 gall-0.039
 further-0.038
Token pink
Token ID11398
Feature activation+5.10e+00

Pos logit contributions
 cups+0.038
ugs+0.038
ays+0.037
 twins+0.036
 kettle+0.035
Neg logit contributions
 shortcut-0.049
 gall-0.047
 unnecessary-0.047
 farewell-0.047
lot-0.046
Token brist
Token ID48197
Feature activation+2.75e+00

Pos logit contributions
izz+0.117
auc+0.111
ays+0.110
ugs+0.103
 twins+0.101
Neg logit contributions
 higher-0.129
 shortcut-0.128
 may-0.124
 gall-0.122
 closer-0.121
Tokenles
Token ID829
Feature activation+1.52e+00

Pos logit contributions
 cups+0.102
 twins+0.098
 sisters+0.097
 kettle+0.096
cloth+0.091
Neg logit contributions
 higher-0.031
 unnecessary-0.028
lot-0.028
 further-0.028
 shortcut-0.027
Token that
Token ID326
Feature activation+1.25e+00

Pos logit contributions
 cups+0.041
izz+0.039
ugs+0.039
 frost+0.037
ays+0.037
Neg logit contributions
 shortcut-0.034
 higher-0.033
 gall-0.032
 further-0.032
lot-0.032
Token she
Token ID673
Feature activation+8.78e-01

Pos logit contributions
 cups+0.030
ays+0.027
 twins+0.027
ugs+0.027
 frost+0.026
Neg logit contributions
 shortcut-0.034
lot-0.032
 higher-0.032
 duel-0.031
 gall-0.031
Token loved
Token ID6151
Feature activation+1.04e+00

Pos logit contributions
 cups+0.027
ays+0.025
 frost+0.025
ugs+0.025
 twins+0.025
Neg logit contributions
 shortcut-0.018
lot-0.017
 duel-0.016
 noble-0.016
 gall-0.016
Token were
Token ID547
Feature activation+1.39e+00

Pos logit contributions
 shortcut+0.014
lot+0.014
 duel+0.014
 farewell+0.013
 gall+0.013
Neg logit contributions
 cups-0.011
 twins-0.010
 frost-0.010
 sisters-0.010
el-0.010
Token burying
Token ID48360
Feature activation+5.16e-01

Pos logit contributions
 cups+0.038
 twins+0.035
 frost+0.035
 sisters+0.034
ugs+0.033
Neg logit contributions
 shortcut-0.036
lot-0.035
 duel-0.033
 farewell-0.032
 gall-0.031
Token something
Token ID1223
Feature activation-7.09e-01

Pos logit contributions
 cups+0.012
 frost+0.011
 twins+0.010
 sisters+0.010
el+0.010
Neg logit contributions
 shortcut-0.015
lot-0.015
 duel-0.015
 gall-0.014
 farewell-0.014
Token from
Token ID422
Feature activation+1.38e+00

Pos logit contributions
lot+0.019
 shortcut+0.019
 duel+0.018
 farewell+0.018
 gall+0.017
Neg logit contributions
 cups-0.019
 frost-0.017
 twins-0.016
 sisters-0.016
cloth-0.016
Token long
Token ID890
Feature activation+2.69e+00

Pos logit contributions
 cups+0.032
 frost+0.028
 twins+0.027
 sisters+0.026
el+0.026
Neg logit contributions
lot-0.043
 shortcut-0.042
 duel-0.040
 farewell-0.039
 noble-0.038
Token ago
Token ID2084
Feature activation+5.07e+00

Pos logit contributions
 cups+0.060
 frost+0.056
ugs+0.053
 twins+0.053
 sisters+0.052
Neg logit contributions
 shortcut-0.081
lot-0.077
 duel-0.075
 further-0.073
 unnecessary-0.072
Token that
Token ID326
Feature activation+4.85e-01

Pos logit contributions
 cups+0.122
 frost+0.114
ays+0.111
ugs+0.109
izz+0.108
Neg logit contributions
 shortcut-0.138
lot-0.135
 duel-0.127
 gall-0.125
 further-0.123
Token made
Token ID925
Feature activation+1.19e+00

Pos logit contributions
 cups+0.012
 frost+0.010
 twins+0.010
cloth+0.010
 sisters+0.010
Neg logit contributions
lot-0.014
 shortcut-0.014
 duel-0.013
 farewell-0.013
 unnecessary-0.013
Token them
Token ID606
Feature activation-2.32e+00

Pos logit contributions
 cups+0.029
 frost+0.026
 twins+0.026
 sisters+0.025
el+0.025
Neg logit contributions
 shortcut-0.034
lot-0.034
 duel-0.033
 noble-0.031
 gall-0.031
Token feel
Token ID1254
Feature activation-8.15e-01

Pos logit contributions
 shortcut+0.058
lot+0.057
 duel+0.054
 noble+0.051
 gall+0.051
Neg logit contributions
 cups-0.066
 frost-0.060
 twins-0.059
 sisters-0.058
cloth-0.056
Token sad
Token ID6507
Feature activation+4.14e-01

Pos logit contributions
lot+0.020
 shortcut+0.020
 duel+0.019
 gall+0.017
 hike+0.017
Neg logit contributions
 cups-0.024
 twins-0.022
 frost-0.022
 sisters-0.022
cloth-0.021
Token people
Token ID661
Feature activation+5.70e-01

Pos logit contributions
 cups+0.007
 frost+0.007
 twins+0.006
 sisters+0.006
ugs+0.006
Neg logit contributions
 shortcut-0.009
lot-0.008
 duel-0.008
 unnecessary-0.008
 farewell-0.008
Token did
Token ID750
Feature activation+1.97e+00

Pos logit contributions
 cups+0.013
 frost+0.012
 twins+0.012
 sisters+0.011
el+0.011
Neg logit contributions
 shortcut-0.017
lot-0.016
 duel-0.016
 farewell-0.016
 unnecessary-0.015
Token a
Token ID257
Feature activation-2.49e-01

Pos logit contributions
 cups+0.051
 frost+0.046
ugs+0.045
ays+0.044
 sisters+0.044
Neg logit contributions
 shortcut-0.053
lot-0.051
 duel-0.049
 noble-0.047
 further-0.047
Token long
Token ID890
Feature activation+1.84e+00

Pos logit contributions
 shortcut+0.007
lot+0.007
 farther+0.006
 further+0.006
 gall+0.006
Neg logit contributions
 frost-0.006
 cups-0.006
 twins-0.006
 mug-0.005
 folders-0.005
Token time
Token ID640
Feature activation+2.30e+00

Pos logit contributions
 cups+0.051
 frost+0.047
 twins+0.045
 sisters+0.045
ugs+0.045
Neg logit contributions
 shortcut-0.047
lot-0.045
 duel-0.043
 unnecessary-0.041
 noble-0.041
Token ago
Token ID2084
Feature activation+5.05e+00

Pos logit contributions
 cups+0.052
 frost+0.047
ugs+0.046
ays+0.045
izz+0.044
Neg logit contributions
 shortcut-0.067
lot-0.066
 duel-0.063
 further-0.061
 gall-0.061
Token.
Token ID13
Feature activation+1.02e+00

Pos logit contributions
 cups+0.133
izz+0.127
 frost+0.123
ays+0.122
olk+0.122
Neg logit contributions
lot-0.122
 shortcut-0.120
 duel-0.111
 gall-0.109
 higher-0.106
Token They
Token ID1119
Feature activation-2.21e-01

Pos logit contributions
 cups+0.025
 frost+0.023
ugs+0.023
el+0.022
els+0.022
Neg logit contributions
 shortcut-0.028
lot-0.027
 duel-0.026
 noble-0.025
 Skip-0.025
Token lived
Token ID5615
Feature activation+1.18e+00

Pos logit contributions
 shortcut+0.006
lot+0.006
 duel+0.006
 unnecessary+0.006
 farewell+0.006
Neg logit contributions
 cups-0.005
 frost-0.005
 twins-0.005
 sisters-0.005
ugs-0.004
Token in
Token ID287
Feature activation+2.00e+00

Pos logit contributions
 cups+0.029
 twins+0.026
 frost+0.026
 sisters+0.025
 folders+0.024
Neg logit contributions
 shortcut-0.035
lot-0.033
 duel-0.032
 unnecessary-0.032
 farewell-0.030
Token castles
Token ID49203
Feature activation+6.99e-01

Pos logit contributions
 cups+0.056
ugs+0.056
ays+0.054
izz+0.053
 frost+0.052
Neg logit contributions
 shortcut-0.046
lot-0.043
 duel-0.043
 noble-0.041
 further-0.041
Token and
Token ID290
Feature activation-6.63e-01

Pos logit contributions
 cups+0.021
 frost+0.019
ugs+0.019
 twins+0.018
 sisters+0.018
Neg logit contributions
 shortcut-0.025
lot-0.024
 duel-0.023
 unnecessary-0.022
 gall-0.022
Token plants
Token ID6134
Feature activation+1.00e+00

Pos logit contributions
 shortcut+0.019
lot+0.019
 duel+0.018
 unnecessary+0.017
 farewell+0.017
Neg logit contributions
 cups-0.016
 frost-0.014
ugs-0.014
 twins-0.014
 sisters-0.013
Token that
Token ID326
Feature activation-1.61e-01

Pos logit contributions
 cups+0.025
ugs+0.023
 frost+0.023
 sisters+0.022
izz+0.022
Neg logit contributions
 shortcut-0.027
lot-0.026
 duel-0.024
 further-0.024
 higher-0.024
Token lived
Token ID5615
Feature activation+5.71e-01

Pos logit contributions
 shortcut+0.005
lot+0.005
 duel+0.005
 unnecessary+0.004
 farewell+0.004
Neg logit contributions
 cups-0.004
 frost-0.003
 twins-0.003
 sisters-0.003
ugs-0.003
Token long
Token ID890
Feature activation+2.70e+00

Pos logit contributions
 cups+0.013
 twins+0.011
 frost+0.011
els+0.011
 folders+0.010
Neg logit contributions
 shortcut-0.018
lot-0.017
 duel-0.017
 unnecessary-0.016
 farewell-0.016
Token ago
Token ID2084
Feature activation+5.02e+00

Pos logit contributions
 cups+0.057
 frost+0.051
ugs+0.049
 sisters+0.048
 twins+0.048
Neg logit contributions
 shortcut-0.085
lot-0.083
 duel-0.079
 gall-0.077
 unnecessary-0.076
Token,"
Token ID553
Feature activation+1.52e+00

Pos logit contributions
 cups+0.124
izz+0.121
ays+0.118
 frost+0.116
ugs+0.116
Neg logit contributions
lot-0.125
 shortcut-0.124
 duel-0.113
 gall-0.113
 further-0.111
Token she
Token ID673
Feature activation+1.81e+00

Pos logit contributions
ugs+0.040
ays+0.039
ase+0.037
ump+0.037
izz+0.036
Neg logit contributions
 closer-0.041
 shortcut-0.037
 spear-0.037
 further-0.037
 nearer-0.036
Token says
Token ID1139
Feature activation+6.46e-01

Pos logit contributions
 cups+0.041
ugs+0.039
 frost+0.037
ions+0.036
olk+0.035
Neg logit contributions
 shortcut-0.054
lot-0.053
 duel-0.050
 gall-0.049
 noble-0.049
Token.
Token ID13
Feature activation+6.36e-01

Pos logit contributions
 cups+0.015
 frost+0.014
ugs+0.013
ays+0.013
 twins+0.013
Neg logit contributions
 shortcut-0.019
lot-0.018
 duel-0.018
 farewell-0.017
 gall-0.017
Token
Token ID198
Feature activation+1.28e-01

Pos logit contributions
 cups+0.020
 frost+0.019
ugs+0.019
els+0.019
cloth+0.019
Neg logit contributions
 shortcut-0.013
lot-0.013
 Skip-0.012
 duel-0.011
Hop-0.011
Token.
Token ID13
Feature activation+8.91e-01

Pos logit contributions
izz+0.072
ays+0.070
 cups+0.070
ugs+0.069
ums+0.068
Neg logit contributions
 shortcut-0.057
 further-0.052
lot-0.051
 higher-0.049
 spear-0.049
Token She
Token ID1375
Feature activation+1.45e+00

Pos logit contributions
 cups+0.022
 frost+0.020
els+0.020
ugs+0.020
 sisters+0.020
Neg logit contributions
lot-0.023
 shortcut-0.023
 Skip-0.021
 farewell-0.021
 unnecessary-0.021
Token said
Token ID531
Feature activation+9.83e-02

Pos logit contributions
 cups+0.041
ugs+0.040
ays+0.038
ums+0.038
ions+0.037
Neg logit contributions
 shortcut-0.033
lot-0.032
 gall-0.031
 duel-0.030
'd-0.030
Token wine
Token ID8237
Feature activation+1.02e+00

Pos logit contributions
 cups+0.002
ugs+0.002
ays+0.002
 frost+0.002
el+0.002
Neg logit contributions
 shortcut-0.003
lot-0.002
 farewell-0.002
 duel-0.002
 higher-0.002
Token was
Token ID373
Feature activation+3.31e+00

Pos logit contributions
 cups+0.028
 frost+0.026
ugs+0.026
ays+0.025
 sisters+0.025
Neg logit contributions
 shortcut-0.025
lot-0.023
 duel-0.022
 noble-0.021
 farewell-0.021
Token for
Token ID329
Feature activation+5.00e+00

Pos logit contributions
 cups+0.103
ugs+0.101
ays+0.099
 frost+0.095
ums+0.095
Neg logit contributions
 shortcut-0.065
lot-0.062
 unnecessary-0.059
 noble-0.058
 duel-0.058
Token grown
Token ID7334
Feature activation+1.62e+00

Pos logit contributions
ays+0.106
 cups+0.105
ugs+0.102
 frost+0.098
cloth+0.096
Neg logit contributions
 shortcut-0.137
 unnecessary-0.131
 duel-0.130
 further-0.128
lot-0.127
Token-
Token ID12
Feature activation+1.67e+00

Pos logit contributions
 cups+0.043
 frost+0.041
 sisters+0.039
 twins+0.039
ays+0.039
Neg logit contributions
 shortcut-0.040
lot-0.040
 unnecessary-0.037
 further-0.037
 duel-0.037
Tokenups
Token ID4739
Feature activation+9.17e-01

Pos logit contributions
 cups+0.040
 frost+0.037
 sisters+0.035
 twins+0.035
ugs+0.034
Neg logit contributions
 shortcut-0.047
lot-0.047
 duel-0.045
 unnecessary-0.044
 further-0.043
Token.
Token ID13
Feature activation+1.26e+00

Pos logit contributions
 cups+0.022
 frost+0.020
ugs+0.019
 twins+0.019
 sisters+0.019
Neg logit contributions
 shortcut-0.026
lot-0.025
 duel-0.024
 gall-0.023
 unnecessary-0.023
Token She
Token ID1375
Feature activation+1.79e+00

Pos logit contributions
 cups+0.029
 sisters+0.026
els+0.026
 frost+0.026
ugs+0.026
Neg logit contributions
lot-0.034
 shortcut-0.033
 Skip-0.032
 farewell-0.031
 Sir-0.031
Token to
Token ID284
Feature activation+7.71e-01

Pos logit contributions
 cups+0.013
 frost+0.012
 twins+0.011
 sisters+0.011
 folders+0.011
Neg logit contributions
 shortcut-0.013
lot-0.013
 duel-0.012
 gall-0.012
 noble-0.012
Token live
Token ID2107
Feature activation+2.87e-01

Pos logit contributions
 cups+0.020
 frost+0.019
 twins+0.018
 sisters+0.018
el+0.018
Neg logit contributions
 shortcut-0.020
lot-0.020
 duel-0.018
 unnecessary-0.018
Hop-0.018
Token a
Token ID257
Feature activation-5.85e-01

Pos logit contributions
 cups+0.006
 twins+0.006
 frost+0.006
 folders+0.006
 sisters+0.005
Neg logit contributions
 shortcut-0.009
lot-0.009
 duel-0.008
 gall-0.008
 farewell-0.008
Token long
Token ID890
Feature activation+1.63e+00

Pos logit contributions
lot+0.016
 shortcut+0.016
 unnecessary+0.016
 farewell+0.015
Hop+0.015
Neg logit contributions
 cups-0.013
 twins-0.013
 frost-0.012
 sisters-0.012
 printer-0.012
Token time
Token ID640
Feature activation+1.84e+00

Pos logit contributions
 cups+0.043
 frost+0.040
 twins+0.038
 sisters+0.037
 folders+0.037
Neg logit contributions
 shortcut-0.043
lot-0.041
 unnecessary-0.041
 duel-0.039
 further-0.039
Token ago
Token ID2084
Feature activation+4.99e+00

Pos logit contributions
 cups+0.036
 frost+0.032
 twins+0.030
 sisters+0.030
ugs+0.029
Neg logit contributions
 shortcut-0.062
lot-0.060
 duel-0.059
 unnecessary-0.057
 farewell-0.056
Token."
Token ID526
Feature activation+2.81e-01

Pos logit contributions
 cups+0.125
 frost+0.115
izz+0.113
ays+0.112
ugs+0.110
Neg logit contributions
 shortcut-0.130
lot-0.130
 duel-0.121
 gall-0.119
 farewell-0.116
Token
Token ID198
Feature activation-2.16e-01

Pos logit contributions
 cups+0.009
ays+0.009
els+0.009
ugs+0.008
 frost+0.008
Neg logit contributions
lot-0.005
 shortcut-0.005
 duel-0.005
 farewell-0.005
 Skip-0.005
Token
Token ID198
Feature activation+6.69e-01

Pos logit contributions
 shortcut+0.006
lot+0.005
 duel+0.005
 gall+0.005
 farewell+0.005
Neg logit contributions
 cups-0.006
 frost-0.005
 sisters-0.005
ugs-0.005
 twins-0.005
TokenThe
Token ID464
Feature activation-1.49e-01

Pos logit contributions
 cups+0.017
ays+0.015
 frost+0.015
ugs+0.015
 sisters+0.014
Neg logit contributions
 shortcut-0.019
lot-0.019
 duel-0.018
 noble-0.017
Hop-0.017
Token girl
Token ID2576
Feature activation+7.25e-01

Pos logit contributions
 shortcut+0.005
lot+0.005
 unnecessary+0.005
 duel+0.005
 gall+0.004
Neg logit contributions
 cups-0.003
 twins-0.003
 sisters-0.003
 frost-0.003
 folders-0.003
Token so
Token ID523
Feature activation-9.56e-01

Pos logit contributions
 cups+0.003
 twins+0.003
 frost+0.003
 sisters+0.002
cloth+0.002
Neg logit contributions
 shortcut-0.003
lot-0.003
 duel-0.003
 unnecessary-0.003
 hike-0.003
Token grateful
Token ID14066
Feature activation-9.94e-01

Pos logit contributions
 shortcut+0.027
lot+0.026
 unnecessary+0.026
 duel+0.025
 hike+0.025
Neg logit contributions
 cups-0.023
 twins-0.023
 frost-0.021
 sisters-0.020
izz-0.020
Token,
Token ID11
Feature activation-2.21e+00

Pos logit contributions
 shortcut+0.033
 duel+0.032
lot+0.031
 unnecessary+0.031
 noble+0.030
Neg logit contributions
 cups-0.020
 twins-0.018
 frost-0.017
el-0.016
 sisters-0.016
Token he
Token ID339
Feature activation-1.62e+00

Pos logit contributions
 shortcut+0.072
lot+0.072
 duel+0.071
 unnecessary+0.070
 quest+0.066
Neg logit contributions
 cups-0.041
 twins-0.040
 sisters-0.034
 frost-0.034
el-0.033
Token gave
Token ID2921
Feature activation-2.43e+00

Pos logit contributions
 unnecessary+0.052
 shortcut+0.051
 duel+0.048
lot+0.048
 further+0.046
Neg logit contributions
 cups-0.033
 twins-0.031
 sisters-0.031
 frost-0.030
el-0.027
Token Fin
Token ID4463
Feature activation-5.93e+00

Pos logit contributions
 shortcut+0.063
lot+0.062
 quest+0.060
 duel+0.059
 noble+0.057
Neg logit contributions
el-0.065
 cups-0.059
 twins-0.058
els-0.054
 frost-0.053
Token the
Token ID262
Feature activation-1.93e+00

Pos logit contributions
 shortcut+0.216
 duel+0.207
 unnecessary+0.199
 steep+0.196
rated+0.194
Neg logit contributions
 cups-0.099
 twins-0.088
 frost-0.078
 sisters-0.077
cloth-0.072
Token shiny
Token ID22441
Feature activation-1.21e+00

Pos logit contributions
lot+0.065
 unnecessary+0.061
 shortcut+0.061
 higher+0.059
 farewell+0.058
Neg logit contributions
 cups-0.040
 twins-0.037
 sisters-0.034
 mug-0.032
 frost-0.032
Token strip
Token ID10283
Feature activation-2.12e+00

Pos logit contributions
lot+0.035
 shortcut+0.034
 unnecessary+0.034
 duel+0.032
 Skip+0.031
Neg logit contributions
 cups-0.030
 frost-0.027
el-0.026
 mug-0.025
 twins-0.025
Token as
Token ID355
Feature activation-2.58e+00

Pos logit contributions
 shortcut+0.069
lot+0.068
 duel+0.066
 unnecessary+0.065
 noble+0.063
Neg logit contributions
 cups-0.043
 frost-0.038
 twins-0.036
ugs-0.036
cloth-0.035
Token a
Token ID257
Feature activation-2.00e+00

Pos logit contributions
 shortcut+0.105
rated+0.103
 duel+0.102
itionally+0.099
lot+0.098
Neg logit contributions
 cups-0.029
 twins-0.028
 ink-0.023
ions-0.022
 frost-0.021
Token higher
Token ID2440
Feature activation-1.82e+00

Pos logit contributions
lot+0.050
 shortcut+0.047
 farewell+0.045
 quest+0.041
 noble+0.040
Neg logit contributions
 cups-0.122
 twins-0.109
 sisters-0.109
 pitcher-0.107
 girls-0.107
Token.
Token ID13
Feature activation+9.12e-02

Pos logit contributions
 shortcut+0.063
lot+0.061
 duel+0.061
 unnecessary+0.058
 noble+0.058
Neg logit contributions
 cups-0.035
 twins-0.030
 frost-0.030
el-0.029
ays-0.028
Token He
Token ID679
Feature activation-1.83e+00

Pos logit contributions
 cups+0.002
 frost+0.002
 twins+0.002
ugs+0.002
 sisters+0.002
Neg logit contributions
 shortcut-0.003
lot-0.003
 duel-0.003
 unnecessary-0.003
 further-0.003
Token showed
Token ID3751
Feature activation-1.11e+00

Pos logit contributions
 unnecessary+0.062
 shortcut+0.062
lot+0.059
 farewell+0.057
 noble+0.057
Neg logit contributions
 cups-0.032
 twins-0.028
 sisters-0.027
 frost-0.027
 painted-0.026
Token T
Token ID309
Feature activation-7.26e-01

Pos logit contributions
 shortcut+0.034
lot+0.032
 duel+0.031
 noble+0.030
 unnecessary+0.030
Neg logit contributions
 cups-0.026
 frost-0.024
el-0.023
 sisters-0.023
 twins-0.023
Tokenoni
Token ID14651
Feature activation-5.86e+00

Pos logit contributions
 shortcut+0.024
 duel+0.023
 further+0.023
 unnecessary+0.022
 farther+0.022
Neg logit contributions
 cups-0.014
 frost-0.014
ugs-0.014
ums-0.013
 twins-0.013
Token how
Token ID703
Feature activation-2.69e+00

Pos logit contributions
 shortcut+0.189
rated+0.189
lot+0.187
 unnecessary+0.183
 duel+0.182
Neg logit contributions
 cups-0.113
 twins-0.104
 sisters-0.099
ions-0.089
 her-0.088
Token to
Token ID284
Feature activation-2.26e+00

Pos logit contributions
 shortcut+0.078
lot+0.078
 duel+0.075
 hike+0.075
 quest+0.075
Neg logit contributions
 cups-0.054
 twins-0.050
 brushes-0.048
 paints-0.046
apy-0.046
Token lift
Token ID10303
Feature activation-2.38e+00

Pos logit contributions
lot+0.061
 unnecessary+0.060
 shortcut+0.057
 noble+0.054
stru+0.054
Neg logit contributions
 cups-0.053
 shampoo-0.048
 frost-0.048
el-0.048
 pitcher-0.048
Token up
Token ID510
Feature activation-2.79e+00

Pos logit contributions
 shortcut+0.067
lot+0.065
 duel+0.063
 noble+0.060
 farewell+0.059
Neg logit contributions
 cups-0.062
 frost-0.056
el-0.055
els-0.053
 twins-0.052
Token her
Token ID607
Feature activation-5.95e-01

Pos logit contributions
lot+0.092
 shortcut+0.089
 duel+0.087
 noble+0.084
 unnecessary+0.082
Neg logit contributions
 cups-0.060
 frost-0.050
el-0.050
els-0.048
 twins-0.048
Token ran
Token ID4966
Feature activation-2.56e+00

Pos logit contributions
 unnecessary+0.042
 shortcut+0.040
 duel+0.038
lot+0.037
 higher+0.036
Neg logit contributions
 cups-0.026
 sisters-0.023
 frost-0.023
 twins-0.022
el-0.021
Token over
Token ID625
Feature activation-1.82e+00

Pos logit contributions
 unnecessary+0.077
 duel+0.073
lot+0.072
 noble+0.071
 shortcut+0.070
Neg logit contributions
 cups-0.053
 twins-0.046
cloth-0.042
 frost-0.042
 her-0.041
Token and
Token ID290
Feature activation-1.59e+00

Pos logit contributions
lot+0.062
 unnecessary+0.060
rated+0.059
 shortcut+0.059
 duel+0.058
Neg logit contributions
 cups-0.038
ions-0.030
 twins-0.030
 frost-0.029
el-0.028
Token gave
Token ID2921
Feature activation-2.37e+00

Pos logit contributions
 unnecessary+0.051
 shortcut+0.051
lot+0.048
 duel+0.047
 farewell+0.046
Neg logit contributions
 cups-0.033
 sisters-0.027
 twins-0.026
 frost-0.025
 girls-0.025
Token the
Token ID262
Feature activation-1.34e+00

Pos logit contributions
 shortcut+0.084
lot+0.080
 steep+0.077
 duel+0.077
 unnecessary+0.075
Neg logit contributions
 cups-0.046
el-0.042
 frost-0.041
 twins-0.038
 sisters-0.036
Token rabbit
Token ID22746
Feature activation-5.86e+00

Pos logit contributions
lot+0.039
 shortcut+0.037
 unnecessary+0.037
 farewell+0.037
 duel+0.036
Neg logit contributions
 cups-0.033
 twins-0.032
 sisters-0.032
 mug-0.030
 ink-0.027
Token a
Token ID257
Feature activation-2.66e+00

Pos logit contributions
 shortcut+0.195
 duel+0.186
lot+0.185
 unnecessary+0.184
 steep+0.170
Neg logit contributions
 cups-0.122
 twins-0.103
 frost-0.103
 sisters-0.101
 mug-0.094
Token big
Token ID1263
Feature activation-1.77e+00

Pos logit contributions
lot+0.076
 unnecessary+0.075
 duel+0.071
 shortcut+0.070
 further+0.070
Neg logit contributions
 cups-0.063
 mug-0.059
 frost-0.055
 kettle-0.054
 cup-0.054
Token hug
Token ID16225
Feature activation-5.69e-01

Pos logit contributions
lot+0.048
 shortcut+0.046
 unnecessary+0.045
 duel+0.044
Hop+0.041
Neg logit contributions
 cups-0.047
 frost-0.043
 mug-0.042
 twins-0.040
 sisters-0.040
Token.
Token ID13
Feature activation-1.57e+00

Pos logit contributions
 shortcut+0.019
lot+0.018
 duel+0.018
 unnecessary+0.018
 noble+0.018
Neg logit contributions
 cups-0.011
 frost-0.010
 twins-0.009
 sisters-0.009
ugs-0.009
Token But
Token ID887
Feature activation-4.38e-01

Pos logit contributions
 shortcut+0.053
 unnecessary+0.051
 duel+0.050
lot+0.048
 further+0.048
Neg logit contributions
 cups-0.032
 twins-0.028
 frost-0.027
ugs-0.026
ays-0.026
Token ran
Token ID4966
Feature activation-2.74e+00

Pos logit contributions
 unnecessary+0.051
 shortcut+0.048
lot+0.044
 duel+0.044
 higher+0.044
Neg logit contributions
 cups-0.026
 sisters-0.023
 twins-0.022
 frost-0.022
 painted-0.022
Token over
Token ID625
Feature activation-1.90e+00

Pos logit contributions
 unnecessary+0.087
 duel+0.082
lot+0.082
 noble+0.080
 shortcut+0.079
Neg logit contributions
 cups-0.051
 twins-0.044
 her-0.039
cloth-0.038
 frost-0.038
Token and
Token ID290
Feature activation-1.65e+00

Pos logit contributions
lot+0.066
 unnecessary+0.064
rated+0.064
 duel+0.062
 shortcut+0.062
Neg logit contributions
 cups-0.039
ions-0.031
 twins-0.030
 paints-0.029
el-0.028
Token gave
Token ID2921
Feature activation-2.65e+00

Pos logit contributions
 unnecessary+0.056
 shortcut+0.055
lot+0.052
 farewell+0.052
 duel+0.049
Neg logit contributions
 cups-0.033
 sisters-0.026
 twins-0.025
 kettle-0.024
 frost-0.023
Token the
Token ID262
Feature activation-1.68e+00

Pos logit contributions
 shortcut+0.100
lot+0.095
 steep+0.094
rig+0.093
 duel+0.092
Neg logit contributions
 cups-0.045
el-0.043
 frost-0.040
 twins-0.037
ries-0.035
Token rabbit
Token ID22746
Feature activation-5.80e+00

Pos logit contributions
lot+0.048
 shortcut+0.046
 farewell+0.046
 unnecessary+0.045
 duel+0.044
Neg logit contributions
 cups-0.042
 twins-0.041
 sisters-0.041
 mug-0.038
 girls-0.035
Token a
Token ID257
Feature activation-2.53e+00

Pos logit contributions
 shortcut+0.194
lot+0.184
 duel+0.184
 unnecessary+0.182
 steep+0.169
Neg logit contributions
 cups-0.123
 frost-0.104
 twins-0.102
 sisters-0.101
 mug-0.094
Token big
Token ID1263
Feature activation-1.77e+00

Pos logit contributions
lot+0.073
 unnecessary+0.071
rated+0.067
 further+0.066
 shortcut+0.066
Neg logit contributions
 cups-0.061
 mug-0.057
 frost-0.053
 kettle-0.052
 cup-0.051
Token hug
Token ID16225
Feature activation-2.20e-01

Pos logit contributions
lot+0.047
 shortcut+0.045
 unnecessary+0.043
 duel+0.043
 noble+0.040
Neg logit contributions
 cups-0.048
 frost-0.044
 mug-0.042
 twins-0.042
 sisters-0.041
Token.
Token ID13
Feature activation-1.98e+00

Pos logit contributions
 shortcut+0.007
lot+0.007
 duel+0.007
 unnecessary+0.007
 noble+0.007
Neg logit contributions
 cups-0.004
 frost-0.004
 twins-0.004
ugs-0.004
 sisters-0.004
Token But
Token ID887
Feature activation-5.69e-01

Pos logit contributions
 shortcut+0.063
 unnecessary+0.062
 duel+0.059
lot+0.057
 further+0.056
Neg logit contributions
 cups-0.045
 twins-0.039
 frost-0.039
ugs-0.036
 sisters-0.036
Token
Token ID198
Feature activation-2.14e+00

Pos logit contributions
 shortcut+0.046
lot+0.043
 duel+0.043
 unnecessary+0.042
 noble+0.041
Neg logit contributions
 cups-0.036
 twins-0.031
 frost-0.031
ugs-0.031
ays-0.030
Token
Token ID198
Feature activation-1.30e+00

Pos logit contributions
 shortcut+0.074
 unnecessary+0.073
 duel+0.072
lot+0.070
 noble+0.069
Neg logit contributions
 cups-0.036
izz-0.032
ugs-0.032
 twins-0.032
Mom-0.030
TokenThe
Token ID464
Feature activation-1.03e+00

Pos logit contributions
 shortcut+0.037
lot+0.036
 duel+0.035
 unnecessary+0.034
 farewell+0.034
Neg logit contributions
 cups-0.031
 frost-0.027
 twins-0.027
ugs-0.027
 sisters-0.026
Token man
Token ID582
Feature activation-1.01e+00

Pos logit contributions
 unnecessary+0.033
 shortcut+0.032
lot+0.032
 duel+0.031
Hop+0.031
Neg logit contributions
 twins-0.021
 cups-0.020
 frost-0.018
 sisters-0.018
 mug-0.016
Token showed
Token ID3751
Feature activation-2.23e+00

Pos logit contributions
 unnecessary+0.034
 shortcut+0.033
lot+0.031
 duel+0.030
 noble+0.029
Neg logit contributions
 cups-0.021
 frost-0.019
 twins-0.018
 sisters-0.017
el-0.017
Token Sam
Token ID3409
Feature activation-5.69e+00

Pos logit contributions
 shortcut+0.081
lot+0.078
 noble+0.076
 duel+0.076
 unnecessary+0.074
Neg logit contributions
 cups-0.038
el-0.036
els-0.035
ays-0.031
 twins-0.031
Token how
Token ID703
Feature activation-1.88e+00

Pos logit contributions
 shortcut+0.191
 unnecessary+0.183
 duel+0.179
rated+0.174
lot+0.171
Neg logit contributions
 cups-0.118
 twins-0.103
 sisters-0.097
 frost-0.097
el-0.096
Token to
Token ID284
Feature activation-1.73e+00

Pos logit contributions
 shortcut+0.058
lot+0.056
rated+0.055
 duel+0.054
Left+0.053
Neg logit contributions
 cups-0.039
 twins-0.037
 sisters-0.034
 folders-0.033
 paints-0.032
Token hold
Token ID1745
Feature activation-1.09e+00

Pos logit contributions
 unnecessary+0.039
lot+0.038
 shortcut+0.035
 farewell+0.035
 noble+0.033
Neg logit contributions
 cups-0.048
el-0.045
 glue-0.045
 frost-0.045
 paint-0.045
Token the
Token ID262
Feature activation-4.99e-01

Pos logit contributions
 shortcut+0.032
lot+0.032
 duel+0.031
 noble+0.030
 gall+0.029
Neg logit contributions
 cups-0.026
 frost-0.022
el-0.022
 twins-0.022
els-0.022
Token gun
Token ID2485
Feature activation-2.27e+00

Pos logit contributions
lot+0.011
 gall+0.010
 shortcut+0.010
 farewell+0.010
 duel+0.010
Neg logit contributions
 cups-0.015
 twins-0.015
 frost-0.014
 sisters-0.013
 mug-0.013
Token Tim
Token ID5045
Feature activation-1.95e-01

Pos logit contributions
 cups+0.006
ugs+0.006
 frost+0.006
ays+0.006
ions+0.005
Neg logit contributions
 shortcut-0.003
lot-0.003
 gall-0.003
 duel-0.003
 spear-0.003
Tokenmy
Token ID1820
Feature activation+6.48e-01

Pos logit contributions
 shortcut+0.006
lot+0.005
 duel+0.005
 unnecessary+0.005
 noble+0.005
Neg logit contributions
 cups-0.005
 frost-0.004
 twins-0.004
 sisters-0.004
ugs-0.004
Token said
Token ID531
Feature activation-1.18e+00

Pos logit contributions
 cups+0.014
 frost+0.012
ugs+0.012
 twins+0.011
ays+0.011
Neg logit contributions
 shortcut-0.021
lot-0.020
 duel-0.019
 gall-0.019
 farewell-0.019
Token,
Token ID11
Feature activation-1.25e+00

Pos logit contributions
 shortcut+0.037
lot+0.036
 duel+0.035
 unnecessary+0.034
 noble+0.033
Neg logit contributions
 cups-0.026
 frost-0.023
 twins-0.022
 sisters-0.022
ugs-0.022
Token showing
Token ID4478
Feature activation-7.92e-01

Pos logit contributions
 shortcut+0.039
 unnecessary+0.038
lot+0.036
 duel+0.035
 further+0.034
Neg logit contributions
 cups-0.029
 frost-0.026
 twins-0.026
 sisters-0.025
 mug-0.024
Token Billy
Token ID15890
Feature activation-5.66e+00

Pos logit contributions
 shortcut+0.020
lot+0.019
 duel+0.018
 unnecessary+0.017
 noble+0.017
Neg logit contributions
 cups-0.022
 frost-0.020
 twins-0.020
el-0.020
 sisters-0.020
Token the
Token ID262
Feature activation-1.06e+00

Pos logit contributions
 shortcut+0.160
lot+0.146
 duel+0.146
 unnecessary+0.143
 noble+0.139
Neg logit contributions
 cups-0.147
 twins-0.134
 frost-0.132
 sisters-0.130
cloth-0.125
Token toy
Token ID13373
Feature activation-7.14e-01

Pos logit contributions
 shortcut+0.029
lot+0.028
Hop+0.027
 unnecessary+0.027
 duel+0.027
Neg logit contributions
 cups-0.025
 twins-0.023
 frost-0.023
 sisters-0.022
el-0.022
Token arrow
Token ID15452
Feature activation-2.31e+00

Pos logit contributions
 shortcut+0.018
lot+0.018
 duel+0.017
 unnecessary+0.017
 noble+0.016
Neg logit contributions
 cups-0.019
 frost-0.018
 twins-0.018
 sisters-0.017
el-0.017
Token.
Token ID13
Feature activation-5.41e-01

Pos logit contributions
 shortcut+0.066
lot+0.064
 duel+0.061
 unnecessary+0.061
 noble+0.059
Neg logit contributions
 cups-0.057
 frost-0.052
 twins-0.050
 sisters-0.049
el-0.048
Token "
Token ID366
Feature activation-1.43e+00

Pos logit contributions
 shortcut+0.017
 unnecessary+0.016
lot+0.016
 duel+0.016
 further+0.015
Neg logit contributions
 cups-0.012
 twins-0.011
 frost-0.011
 sisters-0.010
 mug-0.010
Token He
Token ID679
Feature activation-6.03e-01

Pos logit contributions
 shortcut+0.016
lot+0.015
 duel+0.015
 unnecessary+0.014
 noble+0.014
Neg logit contributions
 cups-0.014
 frost-0.013
 twins-0.012
ugs-0.012
 sisters-0.012
Token wanted
Token ID2227
Feature activation-1.59e+00

Pos logit contributions
 shortcut+0.020
 unnecessary+0.019
lot+0.018
 duel+0.018
 farewell+0.017
Neg logit contributions
 cups-0.013
 frost-0.011
 sisters-0.011
 twins-0.011
el-0.010
Token to
Token ID284
Feature activation-1.64e+00

Pos logit contributions
 shortcut+0.050
rated+0.050
lot+0.050
 duel+0.047
rig+0.046
Neg logit contributions
 cups-0.036
 frost-0.033
 twins-0.030
 sisters-0.028
 ink-0.028
Token teach
Token ID4545
Feature activation-1.24e+00

Pos logit contributions
 shortcut+0.038
lot+0.038
 unnecessary+0.035
 duel+0.035
 farewell+0.034
Neg logit contributions
 cups-0.047
 frost-0.044
 sisters-0.042
 twins-0.042
el-0.041
Token the
Token ID262
Feature activation-2.38e+00

Pos logit contributions
lot+0.039
 shortcut+0.039
 duel+0.036
 noble+0.035
rated+0.035
Neg logit contributions
 cups-0.028
 frost-0.025
 twins-0.025
el-0.025
 sisters-0.024
Token squirrel
Token ID33039
Feature activation-5.66e+00

Pos logit contributions
lot+0.063
 duel+0.060
 unnecessary+0.060
 shortcut+0.057
 farewell+0.057
Neg logit contributions
 twins-0.063
 cups-0.060
 sisters-0.057
 girls-0.054
 mug-0.053
Token,
Token ID11
Feature activation-2.61e+00

Pos logit contributions
 shortcut+0.175
 duel+0.165
lot+0.164
Left+0.158
 unnecessary+0.157
Neg logit contributions
 cups-0.128
 twins-0.114
 sisters-0.110
 folders-0.105
 frost-0.104
Token so
Token ID523
Feature activation-1.04e+00

Pos logit contributions
lot+0.074
 shortcut+0.074
 duel+0.070
 unnecessary+0.067
 farewell+0.067
Neg logit contributions
 cups-0.066
 twins-0.058
 sisters-0.057
 frost-0.057
 folders-0.054
Token he
Token ID339
Feature activation-1.04e+00

Pos logit contributions
 shortcut+0.021
lot+0.019
 duel+0.019
 unnecessary+0.017
 farewell+0.017
Neg logit contributions
 cups-0.035
 frost-0.032
 twins-0.032
 sisters-0.031
ugs-0.031
Token placed
Token ID4624
Feature activation-8.25e-01

Pos logit contributions
 unnecessary+0.034
 shortcut+0.034
lot+0.032
 duel+0.031
 farewell+0.030
Neg logit contributions
 cups-0.021
 frost-0.018
 sisters-0.018
 twins-0.017
el-0.016
Token the
Token ID262
Feature activation-9.33e-01

Pos logit contributions
 shortcut+0.027
lot+0.027
 duel+0.026
 noble+0.025
 unnecessary+0.025
Neg logit contributions
 cups-0.017
 frost-0.016
el-0.015
ions-0.014
 twins-0.014
Token and
Token ID290
Feature activation-2.04e+00

Pos logit contributions
 shortcut+0.065
 unnecessary+0.063
lot+0.060
 duel+0.059
 nearer+0.057
Neg logit contributions
 cups-0.036
 twins-0.031
 sisters-0.031
 frost-0.029
 mug-0.027
Token D
Token ID360
Feature activation-1.41e+00

Pos logit contributions
 shortcut+0.062
lot+0.059
 duel+0.057
 unnecessary+0.055
 noble+0.054
Neg logit contributions
 cups-0.048
 frost-0.042
 twins-0.042
 sisters-0.042
el-0.040
Tokenucky
Token ID5309
Feature activation-2.68e+00

Pos logit contributions
 shortcut+0.055
lot+0.054
 duel+0.053
 gall+0.051
 noble+0.050
Neg logit contributions
 cups-0.021
ugs-0.017
 sisters-0.017
el-0.017
 twins-0.017
Token taught
Token ID7817
Feature activation-2.36e-01

Pos logit contributions
 shortcut+0.080
lot+0.076
 unnecessary+0.075
 duel+0.073
 nearer+0.071
Neg logit contributions
 cups-0.066
 sisters-0.060
 twins-0.059
 frost-0.057
ugs-0.052
Token P
Token ID350
Feature activation-1.83e+00

Pos logit contributions
 shortcut+0.007
lot+0.006
 duel+0.006
 unnecessary+0.006
 farewell+0.006
Neg logit contributions
 cups-0.006
 frost-0.005
ugs-0.005
 twins-0.005
 sisters-0.005
Tokenandy
Token ID10757
Feature activation-5.64e+00

Pos logit contributions
 shortcut+0.064
lot+0.062
 duel+0.061
 gall+0.059
 farewell+0.058
Neg logit contributions
 cups-0.035
ugs-0.031
ums-0.029
 sisters-0.029
el-0.029
Token how
Token ID703
Feature activation-1.59e+00

Pos logit contributions
 shortcut+0.173
 duel+0.161
lot+0.159
 unnecessary+0.154
Left+0.147
Neg logit contributions
 cups-0.137
 twins-0.125
 sisters-0.125
 frost-0.116
ugs-0.109
Token to
Token ID284
Feature activation-2.70e+00

Pos logit contributions
 shortcut+0.046
lot+0.043
 duel+0.043
 gall+0.040
 farewell+0.039
Neg logit contributions
 cups-0.038
 twins-0.036
 sisters-0.035
ugs-0.033
 frost-0.032
Token qu
Token ID627
Feature activation-2.63e+00

Pos logit contributions
lot+0.063
 unnecessary+0.060
 shortcut+0.058
 farewell+0.053
 duel+0.052
Neg logit contributions
 cups-0.079
el-0.073
 frost-0.071
 tub-0.071
cloth-0.070
Tokenack
Token ID441
Feature activation-2.60e+00

Pos logit contributions
 shortcut+0.074
lot+0.071
 duel+0.068
 farewell+0.065
 gall+0.065
Neg logit contributions
 cups-0.068
ays-0.061
ugs-0.060
el-0.060
 frost-0.060
Token like
Token ID588
Feature activation-2.38e-01

Pos logit contributions
 shortcut+0.083
lot+0.082
 duel+0.077
 noble+0.073
 unnecessary+0.073
Neg logit contributions
 cups-0.059
 sisters-0.052
 twins-0.051
 frost-0.050
ions-0.048
Tokenagged
Token ID14655
Feature activation-2.83e+00

Pos logit contributions
 shortcut+0.028
lot+0.027
 duel+0.026
 noble+0.025
 unnecessary+0.025
Neg logit contributions
 cups-0.016
 frost-0.014
ugs-0.014
 twins-0.013
 sisters-0.013
Token his
Token ID465
Feature activation-1.96e-01

Pos logit contributions
 shortcut+0.077
 unnecessary+0.075
lot+0.073
 duel+0.071
Hop+0.071
Neg logit contributions
 cups-0.071
 twins-0.065
 frost-0.063
 sisters-0.063
cloth-0.060
Token tail
Token ID7894
Feature activation-1.29e+00

Pos logit contributions
 shortcut+0.005
lot+0.005
 duel+0.005
 unnecessary+0.005
 further+0.005
Neg logit contributions
 cups-0.005
 frost-0.005
 twins-0.005
 sisters-0.005
ugs-0.005
Token,
Token ID11
Feature activation-2.04e+00

Pos logit contributions
 shortcut+0.049
 unnecessary+0.048
lot+0.046
 duel+0.045
 noble+0.043
Neg logit contributions
 cups-0.021
 twins-0.019
 frost-0.018
 sisters-0.018
cloth-0.017
Token showing
Token ID4478
Feature activation-2.82e+00

Pos logit contributions
 shortcut+0.068
 unnecessary+0.065
lot+0.064
 duel+0.061
 farewell+0.059
Neg logit contributions
 cups-0.046
 frost-0.039
 twins-0.039
 sisters-0.038
cloth-0.036
Token Tim
Token ID5045
Feature activation-5.63e+00

Pos logit contributions
lot+0.090
 steep+0.089
 shortcut+0.088
rig+0.088
 hike+0.086
Neg logit contributions
 her-0.049
el-0.049
 twins-0.047
 cups-0.047
 Mom-0.047
Token that
Token ID326
Feature activation-2.06e+00

Pos logit contributions
 shortcut+0.179
lot+0.165
 duel+0.164
 unnecessary+0.162
 noble+0.160
Neg logit contributions
 cups-0.129
 twins-0.118
 sisters-0.109
 frost-0.107
cloth-0.106
Token he
Token ID339
Feature activation-1.52e+00

Pos logit contributions
lot+0.054
 shortcut+0.053
 unnecessary+0.053
 farewell+0.053
rig+0.050
Neg logit contributions
 cups-0.055
 twins-0.049
 frost-0.048
 sisters-0.048
 mug-0.046
Token should
Token ID815
Feature activation-3.81e-01

Pos logit contributions
 shortcut+0.051
 unnecessary+0.049
 farewell+0.048
 further+0.047
 higher+0.047
Neg logit contributions
 cups-0.028
 sisters-0.027
 twins-0.027
 frost-0.026
el-0.025
Token try
Token ID1949
Feature activation-3.48e+00

Pos logit contributions
 shortcut+0.009
lot+0.009
 duel+0.008
 unnecessary+0.008
 further+0.008
Neg logit contributions
 cups-0.011
 frost-0.010
 twins-0.010
 sisters-0.010
ugs-0.009
Token the
Token ID262
Feature activation-1.99e+00

Pos logit contributions
rated+0.114
 shortcut+0.111
 farewell+0.110
Hop+0.109
lot+0.106
Neg logit contributions
 cups-0.067
 twins-0.061
 ink-0.059
els-0.058
 sisters-0.057
Token a
Token ID257
Feature activation-1.16e+00

Pos logit contributions
 shortcut+0.001
 duel+0.001
lot+0.001
 higher+0.001
 farewell+0.001
Neg logit contributions
 cups-0.001
 frost-0.001
 twins-0.000
 sisters-0.000
 ink-0.000
Token new
Token ID649
Feature activation-2.67e-02

Pos logit contributions
lot+0.034
Left+0.034
rated+0.034
Hop+0.032
 unnecessary+0.031
Neg logit contributions
 cups-0.024
 mug-0.023
 twins-0.022
 cup-0.022
 sisters-0.021
Token friend
Token ID1545
Feature activation-3.86e-01

Pos logit contributions
lot+0.001
 shortcut+0.001
 duel+0.001
 unnecessary+0.001
 farewell+0.001
Neg logit contributions
 cups-0.001
 twins-0.001
 frost-0.001
 sisters-0.001
 mug-0.001
Token and
Token ID290
Feature activation-1.55e+00

Pos logit contributions
 shortcut+0.012
lot+0.012
 duel+0.012
 unnecessary+0.011
 farewell+0.011
Neg logit contributions
 cups-0.008
 twins-0.007
 frost-0.007
 sisters-0.007
 mug-0.007
Token showed
Token ID3751
Feature activation-2.28e+00

Pos logit contributions
 shortcut+0.044
lot+0.043
 duel+0.041
 unnecessary+0.041
 higher+0.039
Neg logit contributions
 cups-0.038
 sisters-0.034
 twins-0.033
 frost-0.032
 folders-0.031
Token Benny
Token ID44275
Feature activation-5.63e+00

Pos logit contributions
 shortcut+0.063
lot+0.062
 duel+0.058
 steep+0.057
 gall+0.055
Neg logit contributions
 cups-0.058
el-0.057
 sisters-0.053
 twins-0.053
 ink-0.051
Token around
Token ID1088
Feature activation-1.23e+00

Pos logit contributions
 shortcut+0.165
lot+0.154
 unnecessary+0.152
 duel+0.152
 steep+0.144
Neg logit contributions
 cups-0.138
 twins-0.123
 sisters-0.122
 frost-0.117
cloth-0.117
Token.
Token ID13
Feature activation-1.54e+00

Pos logit contributions
 shortcut+0.042
lot+0.042
 duel+0.039
 unnecessary+0.038
 farewell+0.038
Neg logit contributions
 cups-0.025
 twins-0.022
 sisters-0.021
el-0.021
 mug-0.020
Token
Token ID198
Feature activation-1.79e+00

Pos logit contributions
 shortcut+0.043
lot+0.041
 duel+0.040
 unnecessary+0.039
 gall+0.038
Neg logit contributions
 cups-0.040
 twins-0.035
 sisters-0.035
 frost-0.034
ays-0.033
Token
Token ID198
Feature activation-8.63e-01

Pos logit contributions
 shortcut+0.057
lot+0.054
 duel+0.054
 unnecessary+0.053
 farewell+0.051
Neg logit contributions
 cups-0.037
 twins-0.033
 frost-0.032
ugs-0.032
izz-0.031
TokenAs
Token ID1722
Feature activation-4.30e-01

Pos logit contributions
 shortcut+0.023
lot+0.022
 duel+0.021
 unnecessary+0.020
 farewell+0.020
Neg logit contributions
 cups-0.023
 frost-0.021
 twins-0.021
 sisters-0.020
ugs-0.020
Token.
Token ID13
Feature activation-6.07e-01

Pos logit contributions
 shortcut+0.003
lot+0.003
 duel+0.003
 gall+0.003
 unnecessary+0.003
Neg logit contributions
 cups-0.002
 frost-0.002
 twins-0.002
ugs-0.002
 sisters-0.002
Token Tim
Token ID5045
Feature activation-1.43e+00

Pos logit contributions
 shortcut+0.018
lot+0.017
 unnecessary+0.016
 duel+0.016
 noble+0.015
Neg logit contributions
 cups-0.016
 frost-0.014
 twins-0.013
 sisters-0.013
 mug-0.013
Tokenmy
Token ID1820
Feature activation-4.64e-01

Pos logit contributions
 shortcut+0.050
 duel+0.047
 unnecessary+0.047
lot+0.046
 farewell+0.046
Neg logit contributions
 cups-0.026
el-0.024
 twins-0.024
 frost-0.024
 sisters-0.024
Token told
Token ID1297
Feature activation-1.08e+00

Pos logit contributions
 shortcut+0.014
lot+0.013
 duel+0.013
 unnecessary+0.013
 farewell+0.012
Neg logit contributions
 cups-0.011
 frost-0.010
 twins-0.010
 sisters-0.010
ugs-0.009
Token Nem
Token ID22547
Feature activation-4.46e+00

Pos logit contributions
 shortcut+0.029
lot+0.028
 duel+0.027
 unnecessary+0.026
 noble+0.026
Neg logit contributions
 cups-0.029
 frost-0.027
 twins-0.026
 sisters-0.026
el-0.025
Tokeno
Token ID78
Feature activation-5.62e+00

Pos logit contributions
 shortcut+0.161
 further+0.156
 unnecessary+0.155
 farewell+0.152
 duel+0.151
Neg logit contributions
ugs-0.073
 cups-0.071
 twins-0.066
ums-0.063
 mug-0.058
Token that
Token ID326
Feature activation-2.33e+00

Pos logit contributions
 shortcut+0.197
lot+0.182
 unnecessary+0.181
 duel+0.180
Left+0.176
Neg logit contributions
 cups-0.107
 twins-0.099
 sisters-0.093
 frost-0.084
 folders-0.082
Token he
Token ID339
Feature activation-2.02e+00

Pos logit contributions
 shortcut+0.049
lot+0.047
 duel+0.044
 unnecessary+0.043
 farewell+0.042
Neg logit contributions
 cups-0.076
 frost-0.070
 twins-0.069
 sisters-0.068
ugs-0.066
Token ate
Token ID15063
Feature activation-1.84e+00

Pos logit contributions
 shortcut+0.059
 unnecessary+0.056
 farewell+0.056
 further+0.055
 duel+0.055
Neg logit contributions
 cups-0.047
 twins-0.044
 frost-0.044
 sisters-0.042
el-0.041
Token a
Token ID257
Feature activation-2.79e+00

Pos logit contributions
lot+0.062
 shortcut+0.061
 duel+0.059
 unnecessary+0.056
 noble+0.055
Neg logit contributions
 cups-0.041
 frost-0.038
 twins-0.033
 icing-0.032
el-0.031
Token lot
Token ID1256
Feature activation-1.35e+00

Pos logit contributions
lot+0.088
 unnecessary+0.084
 shortcut+0.080
 Skip+0.079
 duel+0.076
Neg logit contributions
 cups-0.069
 frost-0.064
 pan-0.057
 mug-0.057
 cup-0.057
Token and
Token ID290
Feature activation-3.25e+00

Pos logit contributions
 shortcut+0.062
lot+0.061
 duel+0.060
 unnecessary+0.059
 noble+0.057
Neg logit contributions
 cups-0.038
 frost-0.032
 twins-0.032
 sisters-0.031
ugs-0.031
Token hills
Token ID18639
Feature activation-2.36e+00

Pos logit contributions
lot+0.094
 shortcut+0.092
 duel+0.086
 farewell+0.082
 unnecessary+0.082
Neg logit contributions
 cups-0.083
 sisters-0.075
 twins-0.074
 frost-0.068
 girls-0.065
Token,
Token ID11
Feature activation-2.66e+00

Pos logit contributions
 shortcut+0.079
 unnecessary+0.079
lot+0.077
 duel+0.076
 noble+0.074
Neg logit contributions
 cups-0.045
ions-0.038
 twins-0.038
 frost-0.037
el-0.036
Token which
Token ID543
Feature activation-1.51e+00

Pos logit contributions
 shortcut+0.078
lot+0.077
 duel+0.073
 unnecessary+0.073
 farewell+0.070
Neg logit contributions
 cups-0.065
 twins-0.056
 sisters-0.054
 paints-0.052
 frost-0.052
Token made
Token ID925
Feature activation-9.21e-01

Pos logit contributions
lot+0.040
 shortcut+0.040
 duel+0.038
 unnecessary+0.037
 farewell+0.037
Neg logit contributions
 cups-0.039
 sisters-0.035
 twins-0.035
 frost-0.035
el-0.034
Token Caroline
Token ID29037
Feature activation-5.62e+00

Pos logit contributions
 shortcut+0.030
lot+0.030
 duel+0.028
 noble+0.027
 unnecessary+0.026
Neg logit contributions
 cups-0.020
 twins-0.017
 sisters-0.017
el-0.017
 frost-0.017
Token a
Token ID257
Feature activation-3.13e+00

Pos logit contributions
 shortcut+0.149
lot+0.140
 duel+0.138
 unnecessary+0.133
Left+0.126
Neg logit contributions
 cups-0.151
 sisters-0.139
 twins-0.136
 frost-0.134
el-0.125
Token bit
Token ID1643
Feature activation-1.42e+00

Pos logit contributions
lot+0.088
 shortcut+0.082
 duel+0.078
 Skip+0.076
 unnecessary+0.075
Neg logit contributions
 cups-0.080
 twins-0.074
 sisters-0.072
 frost-0.069
 mug-0.068
Token less
Token ID1342
Feature activation-1.30e+00

Pos logit contributions
lot+0.040
 shortcut+0.040
 duel+0.037
 farewell+0.034
 gall+0.034
Neg logit contributions
 cups-0.038
 twins-0.034
 frost-0.033
 sisters-0.033
cloth-0.031
Token uncomfortable
Token ID12916
Feature activation+1.95e-01

Pos logit contributions
 shortcut+0.034
lot+0.033
 duel+0.032
 farewell+0.030
Hop+0.029
Neg logit contributions
 cups-0.036
 twins-0.033
 sisters-0.032
 frost-0.032
cloth-0.030
Token.
Token ID13
Feature activation-1.75e-01

Pos logit contributions
 cups+0.004
 frost+0.003
 twins+0.003
 sisters+0.003
ugs+0.003
Neg logit contributions
 shortcut-0.007
lot-0.007
 duel-0.006
 noble-0.006
 farewell-0.006
Token silly
Token ID14397
Feature activation-1.37e+00

Pos logit contributions
 shortcut+0.020
lot+0.019
 duel+0.018
 unnecessary+0.017
 hike+0.017
Neg logit contributions
 cups-0.019
 frost-0.018
 twins-0.017
izz-0.016
 sisters-0.016
Token!
Token ID0
Feature activation-1.26e+00

Pos logit contributions
 shortcut+0.042
lot+0.041
 duel+0.041
 unnecessary+0.039
 farewell+0.039
Neg logit contributions
 cups-0.031
 twins-0.027
 frost-0.027
 sisters-0.026
ugs-0.026
Token Then
Token ID3244
Feature activation-5.19e-01

Pos logit contributions
 shortcut+0.036
 unnecessary+0.034
 duel+0.034
lot+0.033
 further+0.032
Neg logit contributions
 cups-0.031
 twins-0.029
 frost-0.028
ugs-0.027
ays-0.027
Token Bob
Token ID5811
Feature activation-8.79e-02

Pos logit contributions
 shortcut+0.014
lot+0.013
 duel+0.013
 unnecessary+0.012
 noble+0.012
Neg logit contributions
 cups-0.014
 twins-0.013
 frost-0.013
 sisters-0.013
el-0.012
Token showed
Token ID3751
Feature activation-2.34e+00

Pos logit contributions
 shortcut+0.003
lot+0.003
 unnecessary+0.003
 duel+0.002
 noble+0.002
Neg logit contributions
 cups-0.002
 frost-0.002
 sisters-0.002
 twins-0.002
el-0.002
Token Jim
Token ID5395
Feature activation-5.62e+00

Pos logit contributions
 shortcut+0.067
 unnecessary+0.064
 duel+0.064
lot+0.063
 noble+0.061
Neg logit contributions
 cups-0.053
el-0.053
 frost-0.051
 twins-0.051
els-0.048
Token how
Token ID703
Feature activation-1.75e+00

Pos logit contributions
 shortcut+0.176
 unnecessary+0.163
 duel+0.162
lot+0.160
 steep+0.150
Neg logit contributions
 cups-0.133
 twins-0.119
 sisters-0.118
 frost-0.117
el-0.108
Token to
Token ID284
Feature activation+2.54e-02

Pos logit contributions
 shortcut+0.050
lot+0.049
 duel+0.047
Hop+0.044
 hike+0.044
Neg logit contributions
 cups-0.043
 twins-0.039
 sisters-0.037
 frost-0.037
 paints-0.036
Token make
Token ID787
Feature activation-1.71e+00

Pos logit contributions
 cups+0.001
 frost+0.001
 twins+0.001
 sisters+0.001
ugs+0.001
Neg logit contributions
 shortcut-0.001
lot-0.001
 duel-0.001
 unnecessary-0.001
 noble-0.001
Token the
Token ID262
Feature activation-2.92e-01

Pos logit contributions
 shortcut+0.052
lot+0.051
 duel+0.048
 farewell+0.047
Hop+0.046
Neg logit contributions
 cups-0.041
 frost-0.037
 twins-0.036
 sisters-0.035
el-0.034
Token toy
Token ID13373
Feature activation-8.76e-01

Pos logit contributions
 shortcut+0.009
 higher+0.008
 duel+0.008
 further+0.008
lot+0.008
Neg logit contributions
 cups-0.007
ugs-0.006
ays-0.006
izz-0.006
 frost-0.006
TokenAnd
Token ID1870
Feature activation-1.30e+00

Pos logit contributions
 shortcut+0.023
lot+0.022
 duel+0.022
 unnecessary+0.021
 farewell+0.021
Neg logit contributions
 cups-0.021
 frost-0.019
 twins-0.018
ugs-0.018
 sisters-0.018
Token so
Token ID523
Feature activation-1.35e+00

Pos logit contributions
 shortcut+0.041
lot+0.040
 duel+0.040
 unnecessary+0.039
 noble+0.037
Neg logit contributions
 cups-0.027
 twins-0.024
 sisters-0.024
 frost-0.023
el-0.022
Token,
Token ID11
Feature activation-1.18e+00

Pos logit contributions
 shortcut+0.043
lot+0.042
 duel+0.041
 unnecessary+0.039
 farewell+0.039
Neg logit contributions
 cups-0.028
 twins-0.025
 sisters-0.025
el-0.024
 frost-0.024
Token Andy
Token ID12382
Feature activation-8.04e-01

Pos logit contributions
lot+0.033
 shortcut+0.033
 duel+0.032
 unnecessary+0.031
 noble+0.030
Neg logit contributions
 cups-0.028
 twins-0.026
 sisters-0.026
 frost-0.026
el-0.025
Token taught
Token ID7817
Feature activation-7.37e-01

Pos logit contributions
 shortcut+0.025
 duel+0.024
 unnecessary+0.024
lot+0.024
 noble+0.022
Neg logit contributions
 cups-0.018
 sisters-0.016
 frost-0.016
 twins-0.015
el-0.015
Token Billy
Token ID15890
Feature activation-5.58e+00

Pos logit contributions
 shortcut+0.019
lot+0.019
 duel+0.018
 noble+0.017
 unnecessary+0.017
Neg logit contributions
 cups-0.020
el-0.019
 frost-0.019
 twins-0.018
 sisters-0.018
Token how
Token ID703
Feature activation-1.39e+00

Pos logit contributions
 shortcut+0.173
 duel+0.160
lot+0.157
 unnecessary+0.155
 steep+0.150
Neg logit contributions
 cups-0.128
 twins-0.117
 sisters-0.116
 frost-0.110
 folders-0.106
Token to
Token ID284
Feature activation-7.66e-01

Pos logit contributions
 shortcut+0.043
 duel+0.041
lot+0.040
 farther+0.040
 farewell+0.039
Neg logit contributions
 cups-0.028
 twins-0.028
 sisters-0.027
ugs-0.025
 ink-0.024
Token crawl
Token ID27318
Feature activation-2.57e+00

Pos logit contributions
lot+0.019
 shortcut+0.018
 duel+0.017
 unnecessary+0.017
 farewell+0.017
Neg logit contributions
 cups-0.021
 sisters-0.019
el-0.019
 frost-0.019
cloth-0.019
Token safely
Token ID11512
Feature activation-1.18e+00

Pos logit contributions
lot+0.082
 duel+0.075
 shortcut+0.073
 noble+0.070
 unnecessary+0.069
Neg logit contributions
 cups-0.060
ions-0.055
 ink-0.049
cloth-0.049
 folders-0.049
Token and
Token ID290
Feature activation-1.62e+00

Pos logit contributions
lot+0.040
 shortcut+0.039
 duel+0.039
 unnecessary+0.038
 noble+0.037
Neg logit contributions
 cups-0.022
 twins-0.018
 sisters-0.018
cloth-0.018
el-0.018
Token
Token ID198
Feature activation-7.79e-01

Pos logit contributions
 shortcut+0.045
 unnecessary+0.043
 duel+0.043
lot+0.042
 farewell+0.041
Neg logit contributions
 cups-0.025
 twins-0.023
 frost-0.023
izz-0.023
el-0.022
TokenThe
Token ID464
Feature activation-1.56e+00

Pos logit contributions
 shortcut+0.023
 unnecessary+0.022
lot+0.022
 duel+0.022
 farewell+0.021
Neg logit contributions
 cups-0.016
 frost-0.015
 twins-0.015
ugs-0.014
el-0.014
Token little
Token ID1310
Feature activation-4.70e-01

Pos logit contributions
 unnecessary+0.053
 shortcut+0.052
 duel+0.051
lot+0.051
rated+0.049
Neg logit contributions
 twins-0.031
 cups-0.029
 frost-0.027
 sisters-0.025
 pan-0.022
Token boy
Token ID2933
Feature activation+1.04e-01

Pos logit contributions
 shortcut+0.014
lot+0.013
 duel+0.013
 unnecessary+0.013
 farewell+0.013
Neg logit contributions
 cups-0.011
 frost-0.010
 twins-0.010
 sisters-0.010
el-0.009
Token told
Token ID1297
Feature activation-1.11e+00

Pos logit contributions
 cups+0.002
 frost+0.002
el+0.002
 twins+0.002
 sisters+0.002
Neg logit contributions
 shortcut-0.004
 unnecessary-0.004
 duel-0.003
lot-0.003
 further-0.003
Token Bill
Token ID3941
Feature activation-5.56e+00

Pos logit contributions
 shortcut+0.030
lot+0.029
 duel+0.029
 unnecessary+0.029
 noble+0.028
Neg logit contributions
 cups-0.026
el-0.026
 frost-0.025
els-0.025
 twins-0.024
Token about
Token ID546
Feature activation-1.69e+00

Pos logit contributions
 shortcut+0.190
 duel+0.181
 unnecessary+0.175
Left+0.174
 farther+0.165
Neg logit contributions
 twins-0.106
 cups-0.102
 sisters-0.100
 frost-0.090
els-0.089
Token the
Token ID262
Feature activation-6.78e-01

Pos logit contributions
lot+0.055
 shortcut+0.055
 duel+0.052
 unnecessary+0.051
rated+0.051
Neg logit contributions
 cups-0.035
 frost-0.032
 twins-0.031
 sisters-0.030
 mug-0.029
Token big
Token ID1263
Feature activation-4.00e-02

Pos logit contributions
lot+0.019
 shortcut+0.018
 duel+0.018
 unnecessary+0.017
 farewell+0.017
Neg logit contributions
 cups-0.018
 twins-0.016
 frost-0.016
 mug-0.015
 tub-0.015
Token cooler
Token ID19346
Feature activation+1.48e-01

Pos logit contributions
 shortcut+0.001
lot+0.001
 duel+0.001
 further+0.001
 gall+0.001
Neg logit contributions
 cups-0.001
 frost-0.001
ugs-0.001
 sisters-0.001
ays-0.001
Token and
Token ID290
Feature activation-1.52e+00

Pos logit contributions
 cups+0.003
 frost+0.003
 twins+0.003
cloth+0.003
 sisters+0.003
Neg logit contributions
 shortcut-0.005
 unnecessary-0.005
lot-0.005
 duel-0.005
 farewell-0.004
Token
Token ID198
Feature activation-2.05e+00

Pos logit contributions
 shortcut+0.055
 unnecessary+0.051
 duel+0.051
lot+0.050
 gall+0.048
Neg logit contributions
 cups-0.041
 twins-0.037
ays-0.036
ums-0.035
 sisters-0.035
Token
Token ID198
Feature activation-1.03e+00

Pos logit contributions
 shortcut+0.068
 unnecessary+0.065
 duel+0.065
lot+0.063
 higher+0.061
Neg logit contributions
 cups-0.038
 twins-0.035
izz-0.034
ums-0.033
ugs-0.032
TokenThe
Token ID464
Feature activation-1.96e+00

Pos logit contributions
 shortcut+0.029
lot+0.027
 duel+0.027
 unnecessary+0.026
 gall+0.025
Neg logit contributions
 cups-0.025
 twins-0.022
 frost-0.022
 sisters-0.022
ugs-0.022
Token bear
Token ID6842
Feature activation-1.08e+00

Pos logit contributions
lot+0.051
 shortcut+0.050
 unnecessary+0.048
 duel+0.047
to+0.046
Neg logit contributions
 twins-0.055
 cups-0.053
 sisters-0.053
 frost-0.049
 girls-0.046
Token told
Token ID1297
Feature activation-1.86e+00

Pos logit contributions
 shortcut+0.033
 unnecessary+0.032
lot+0.031
 duel+0.031
 further+0.029
Neg logit contributions
 cups-0.024
 twins-0.022
 frost-0.022
 sisters-0.022
el-0.021
Token Tony
Token ID8832
Feature activation-5.55e+00

Pos logit contributions
 shortcut+0.056
lot+0.052
 duel+0.051
 unnecessary+0.050
 steep+0.049
Neg logit contributions
 cups-0.043
el-0.041
 sisters-0.040
 twins-0.039
ums-0.038
Token that
Token ID326
Feature activation-2.24e+00

Pos logit contributions
 shortcut+0.191
 duel+0.186
 unnecessary+0.174
Left+0.173
 rematch+0.166
Neg logit contributions
 cups-0.095
 twins-0.093
 sisters-0.092
cloth-0.083
ions-0.075
Token he
Token ID339
Feature activation-1.09e+00

Pos logit contributions
 shortcut+0.063
lot+0.061
 farewell+0.058
 unnecessary+0.058
 duel+0.057
Neg logit contributions
 cups-0.058
 twins-0.054
 sisters-0.054
 frost-0.050
ions-0.050
Token needed
Token ID2622
Feature activation+2.38e-01

Pos logit contributions
 shortcut+0.034
 duel+0.032
 unnecessary+0.031
 higher+0.031
lot+0.031
Neg logit contributions
 cups-0.022
 twins-0.022
 sisters-0.022
 frost-0.021
el-0.021
Token to
Token ID284
Feature activation+3.66e-01

Pos logit contributions
 cups+0.006
 frost+0.006
 twins+0.005
 sisters+0.005
ions+0.005
Neg logit contributions
lot-0.007
 shortcut-0.006
 duel-0.006
rated-0.006
 noble-0.006
Token find
Token ID1064
Feature activation-1.74e+00

Pos logit contributions
 cups+0.010
 frost+0.009
 twins+0.009
 sisters+0.009
cloth+0.009
Neg logit contributions
lot-0.009
 shortcut-0.009
 duel-0.008
 unnecessary-0.008
 noble-0.008
Token.
Token ID13
Feature activation-1.69e+00

Pos logit contributions
 shortcut+0.089
lot+0.085
 unnecessary+0.085
 duel+0.084
 noble+0.082
Neg logit contributions
 cups-0.031
 twins-0.029
 sisters-0.026
el-0.023
 frost-0.022
Token Nut
Token ID11959
Feature activation-3.70e+00

Pos logit contributions
 shortcut+0.051
 duel+0.049
lot+0.049
 unnecessary+0.048
 farewell+0.047
Neg logit contributions
 cups-0.038
 twins-0.033
 frost-0.032
ugs-0.031
 sisters-0.031
Tokenty
Token ID774
Feature activation-1.16e+00

Pos logit contributions
 unnecessary+0.101
 shortcut+0.099
 farewell+0.094
 duel+0.094
 rematch+0.088
Neg logit contributions
el-0.086
 twins-0.085
izz-0.082
 cups-0.082
els-0.082
Token showed
Token ID3751
Feature activation-2.63e+00

Pos logit contributions
 shortcut+0.044
 unnecessary+0.044
 duel+0.040
lot+0.040
 higher+0.039
Neg logit contributions
 cups-0.019
 sisters-0.017
 twins-0.017
 frost-0.015
el-0.015
Token Bub
Token ID19645
Feature activation-3.19e+00

Pos logit contributions
 shortcut+0.068
lot+0.066
 duel+0.064
 steep+0.062
 noble+0.061
Neg logit contributions
el-0.070
 cups-0.067
els-0.062
 twins-0.061
 frost-0.061
Tokenbles
Token ID7689
Feature activation-5.53e+00

Pos logit contributions
 farewell+0.117
 shortcut+0.111
 duel+0.108
 steep+0.107
 unnecessary+0.107
Neg logit contributions
 cups-0.046
 twins-0.045
els-0.045
ugs-0.045
izz-0.044
Token a
Token ID257
Feature activation-2.24e+00

Pos logit contributions
 shortcut+0.173
lot+0.170
 unnecessary+0.166
 duel+0.165
rated+0.155
Neg logit contributions
 cups-0.124
 twins-0.111
 sisters-0.109
el-0.105
 frost-0.104
Token shiny
Token ID22441
Feature activation-4.80e-01

Pos logit contributions
 unnecessary+0.064
lot+0.062
Left+0.061
 further+0.060
 farewell+0.060
Neg logit contributions
 mug-0.050
 cups-0.048
 tub-0.043
 cup-0.043
 twins-0.043
Token shield
Token ID7614
Feature activation-6.58e-01

Pos logit contributions
lot+0.012
 shortcut+0.012
 duel+0.011
 unnecessary+0.011
 noble+0.011
Neg logit contributions
 cups-0.013
 frost-0.012
el-0.012
 mug-0.012
 twins-0.011
Token that
Token ID326
Feature activation-1.93e+00

Pos logit contributions
 shortcut+0.021
 unnecessary+0.020
lot+0.020
 duel+0.020
 noble+0.019
Neg logit contributions
 cups-0.013
 frost-0.012
 twins-0.012
 sisters-0.011
el-0.011
Token he
Token ID339
Feature activation-1.64e+00

Pos logit contributions
 unnecessary+0.055
 shortcut+0.054
 farewell+0.052
lot+0.051
 duel+0.050
Neg logit contributions
 cups-0.047
 frost-0.042
 twins-0.040
 sisters-0.040
el-0.039
Token K
Token ID509
Feature activation-9.07e-01

Pos logit contributions
 shortcut+0.025
lot+0.024
 duel+0.023
 unnecessary+0.022
 noble+0.022
Neg logit contributions
 cups-0.022
 twins-0.020
 sisters-0.019
 frost-0.019
ions-0.018
Tokeniki
Token ID5580
Feature activation-1.30e+00

Pos logit contributions
 shortcut+0.026
 duel+0.025
 unnecessary+0.024
 farther+0.024
 further+0.024
Neg logit contributions
el-0.022
 cups-0.021
ugs-0.021
 frost-0.020
els-0.020
Token would
Token ID561
Feature activation-1.27e+00

Pos logit contributions
 shortcut+0.039
 unnecessary+0.036
 duel+0.036
lot+0.035
 noble+0.034
Neg logit contributions
 cups-0.030
 sisters-0.028
 frost-0.027
 twins-0.027
el-0.027
Token teach
Token ID4545
Feature activation-2.34e-01

Pos logit contributions
 unnecessary+0.034
lot+0.034
 shortcut+0.032
Left+0.030
 duel+0.030
Neg logit contributions
 sisters-0.027
 cups-0.027
 frost-0.027
 paint-0.027
el-0.027
Token T
Token ID309
Feature activation-1.66e+00

Pos logit contributions
 shortcut+0.008
lot+0.008
 duel+0.008
 noble+0.008
 unnecessary+0.007
Neg logit contributions
 cups-0.004
 frost-0.004
el-0.004
 twins-0.004
 sisters-0.004
Tokenoto
Token ID2069
Feature activation-5.53e+00

Pos logit contributions
 duel+0.047
 shortcut+0.047
 unnecessary+0.044
 farther+0.044
lot+0.043
Neg logit contributions
ugs-0.039
el-0.038
ums-0.038
izz-0.037
els-0.036
Token how
Token ID703
Feature activation-2.24e+00

Pos logit contributions
 shortcut+0.172
lot+0.161
 duel+0.160
Left+0.151
 unnecessary+0.151
Neg logit contributions
 cups-0.122
 sisters-0.115
 twins-0.113
 folders-0.103
 frost-0.102
Token to
Token ID284
Feature activation-1.28e+00

Pos logit contributions
 shortcut+0.068
 duel+0.066
lot+0.065
 hike+0.062
Left+0.060
Neg logit contributions
 cups-0.045
 twins-0.044
 sisters-0.042
 paints-0.038
ugs-0.038
Token push
Token ID4574
Feature activation-1.53e+00

Pos logit contributions
lot+0.031
 shortcut+0.031
 unnecessary+0.030
 farewell+0.028
 noble+0.028
Neg logit contributions
 cups-0.033
 frost-0.032
 sisters-0.031
el-0.031
 twins-0.030
Token his
Token ID465
Feature activation-7.46e-01

Pos logit contributions
 shortcut+0.042
lot+0.041
 duel+0.039
 noble+0.038
 farewell+0.037
Neg logit contributions
 cups-0.040
 frost-0.036
el-0.036
els-0.035
 twins-0.035
Token wings
Token ID12098
Feature activation-2.23e+00

Pos logit contributions
lot+0.023
 shortcut+0.022
 duel+0.021
Hop+0.020
 noble+0.020
Neg logit contributions
 cups-0.017
 mug-0.015
 sisters-0.015
 frost-0.014
 twins-0.014
Token He
Token ID679
Feature activation-1.31e+00

Pos logit contributions
 cups+0.020
 sisters+0.019
 frost+0.019
ugs+0.019
els+0.019
Neg logit contributions
lot-0.021
 shortcut-0.020
 farewell-0.019
 duel-0.019
 higher-0.019
Token said
Token ID531
Feature activation-2.07e+00

Pos logit contributions
 shortcut+0.037
 unnecessary+0.035
lot+0.033
 duel+0.033
 further+0.032
Neg logit contributions
 cups-0.033
 frost-0.031
 twins-0.031
 sisters-0.030
el-0.028
Token whist
Token ID32262
Feature activation-1.29e+00

Pos logit contributions
 shortcut+0.059
lot+0.056
 duel+0.053
 unnecessary+0.052
rated+0.050
Neg logit contributions
 cups-0.053
 sisters-0.048
 frost-0.048
 twins-0.048
 folders-0.044
Tokenling
Token ID1359
Feature activation-1.78e+00

Pos logit contributions
 shortcut+0.042
 duel+0.039
lot+0.038
 gall+0.038
 hike+0.038
Neg logit contributions
 cups-0.027
 twins-0.026
ays-0.025
 frost-0.025
 sisters-0.025
Token made
Token ID925
Feature activation-4.70e-01

Pos logit contributions
 shortcut+0.061
 duel+0.057
lot+0.057
 unnecessary+0.056
 hike+0.054
Neg logit contributions
 cups-0.034
 twins-0.031
 frost-0.030
 sisters-0.030
cloth-0.029
Token him
Token ID683
Feature activation-5.53e+00

Pos logit contributions
 shortcut+0.015
lot+0.014
 duel+0.014
 noble+0.013
 steep+0.013
Neg logit contributions
 cups-0.010
 frost-0.010
 twins-0.010
 sisters-0.009
el-0.009
Token happy
Token ID3772
Feature activation-1.13e+00

Pos logit contributions
 shortcut+0.155
lot+0.146
Hop+0.143
 Skip+0.140
 duel+0.139
Neg logit contributions
 cups-0.128
 twins-0.125
 sisters-0.120
 frost-0.114
 hair-0.110
Token and
Token ID290
Feature activation-2.15e+00

Pos logit contributions
 shortcut+0.042
lot+0.040
 unnecessary+0.039
 duel+0.039
 noble+0.038
Neg logit contributions
 cups-0.019
 twins-0.018
 sisters-0.017
 frost-0.017
 mug-0.014
Token calm
Token ID9480
Feature activation-8.70e-01

Pos logit contributions
 shortcut+0.058
lot+0.053
 duel+0.052
 hike+0.051
 unnecessary+0.050
Neg logit contributions
 cups-0.056
 twins-0.054
 sisters-0.054
 frost-0.050
cloth-0.048
Token.
Token ID13
Feature activation-3.41e-01

Pos logit contributions
 shortcut+0.031
 duel+0.030
lot+0.030
 noble+0.028
 unnecessary+0.027
Neg logit contributions
 cups-0.016
 twins-0.014
 frost-0.014
 sisters-0.013
el-0.013
Token L
Token ID406
Feature activation+2.13e-01

Pos logit contributions
lot+0.007
 shortcut+0.007
 farewell+0.006
 duel+0.006
 gall+0.006
Neg logit contributions
 cups-0.010
els-0.010
 frost-0.010
ries-0.010
ugs-0.009
Token safe
Token ID3338
Feature activation-8.53e-02

Pos logit contributions
 shortcut+0.108
 unnecessary+0.106
lot+0.106
 duel+0.105
Left+0.101
Neg logit contributions
 cups-0.070
 twins-0.068
 sisters-0.065
 frost-0.060
el-0.058
Token!
Token ID0
Feature activation-1.22e+00

Pos logit contributions
 shortcut+0.003
 duel+0.003
lot+0.003
 unnecessary+0.003
 noble+0.003
Neg logit contributions
 cups-0.002
 twins-0.002
 frost-0.001
 sisters-0.001
cloth-0.001
Token He
Token ID679
Feature activation-1.30e+00

Pos logit contributions
 shortcut+0.035
 duel+0.033
 unnecessary+0.033
lot+0.032
 further+0.032
Neg logit contributions
 cups-0.029
 twins-0.027
ugs-0.026
ays-0.025
 frost-0.025
Token gave
Token ID2921
Feature activation-2.09e+00

Pos logit contributions
 unnecessary+0.043
 shortcut+0.043
 duel+0.041
lot+0.040
 nearer+0.038
Neg logit contributions
 cups-0.024
 frost-0.023
 sisters-0.021
 twins-0.021
el-0.021
Token the
Token ID262
Feature activation-1.45e+00

Pos logit contributions
 shortcut+0.074
lot+0.070
 steep+0.069
 duel+0.069
rig+0.068
Neg logit contributions
el-0.040
 cups-0.037
 frost-0.034
 twins-0.034
 sisters-0.031
Token knight
Token ID22062
Feature activation-5.50e+00

Pos logit contributions
 unnecessary+0.040
to+0.039
rated+0.039
lot+0.039
 higher+0.039
Neg logit contributions
 twins-0.035
 cups-0.033
 mug-0.032
 sisters-0.032
 folders-0.029
Token a
Token ID257
Feature activation-2.56e+00

Pos logit contributions
 shortcut+0.179
 duel+0.172
 unnecessary+0.170
lot+0.161
 steep+0.161
Neg logit contributions
 cups-0.109
 frost-0.099
 twins-0.096
 sisters-0.095
 folders-0.087
Token big
Token ID1263
Feature activation-2.17e+00

Pos logit contributions
lot+0.072
 unnecessary+0.071
 duel+0.069
rated+0.069
 shortcut+0.067
Neg logit contributions
 cups-0.059
 mug-0.056
 frost-0.054
 cup-0.051
 twins-0.050
Token hug
Token ID16225
Feature activation-7.02e-01

Pos logit contributions
lot+0.061
 unnecessary+0.057
 shortcut+0.057
 duel+0.056
Hop+0.052
Neg logit contributions
 cups-0.055
 frost-0.051
 mug-0.050
 twins-0.047
 tub-0.047
Token and
Token ID290
Feature activation-1.40e+00

Pos logit contributions
 shortcut+0.024
 duel+0.023
 unnecessary+0.023
lot+0.023
 noble+0.022
Neg logit contributions
 cups-0.013
 frost-0.012
 twins-0.012
cloth-0.011
 mug-0.011
Token said
Token ID531
Feature activation-2.94e+00

Pos logit contributions
 shortcut+0.045
lot+0.043
 duel+0.042
 unnecessary+0.041
 noble+0.039
Neg logit contributions
 cups-0.030
 frost-0.027
 sisters-0.026
 twins-0.026
 mug-0.024
Token milk
Token ID7545
Feature activation+3.14e+00

Pos logit contributions
ays+0.045
ugs+0.045
 cups+0.044
izz+0.043
ums+0.043
Neg logit contributions
 shortcut-0.071
 further-0.067
 higher-0.067
 farewell-0.067
 unnecessary-0.066
Token.
Token ID13
Feature activation+1.51e+00

Pos logit contributions
ums+0.090
izz+0.089
els+0.089
 folders+0.089
 sisters+0.089
Neg logit contributions
 shortcut-0.054
 closer-0.053
 higher-0.052
lot-0.051
 gall-0.050
Token
Token ID198
Feature activation+4.73e-01

Pos logit contributions
 cups+0.040
 sisters+0.038
 frost+0.038
el+0.038
els+0.037
Neg logit contributions
 shortcut-0.034
lot-0.034
 farewell-0.032
 Skip-0.031
Left-0.030
Token
Token ID198
Feature activation+2.47e+00

Pos logit contributions
 cups+0.012
 frost+0.011
 twins+0.011
 sisters+0.011
ugs+0.011
Neg logit contributions
 shortcut-0.013
lot-0.012
 duel-0.012
 gall-0.011
 farewell-0.011
Token"
Token ID1
Feature activation+2.88e+00

Pos logit contributions
 cups+0.076
ays+0.074
 sisters+0.073
 frost+0.073
el+0.072
Neg logit contributions
 shortcut-0.051
Hop-0.048
Left-0.045
lot-0.045
 duel-0.044
TokenGrand
Token ID23581
Feature activation+3.05e+00

Pos logit contributions
 cups+0.077
ays+0.074
ugs+0.074
ums+0.073
el+0.072
Neg logit contributions
Left-0.070
 shortcut-0.070
lot-0.068
Hop-0.068
 duel-0.066
Tokenma
Token ID2611
Feature activation+1.15e-01

Pos logit contributions
 cups+0.065
 pitcher+0.057
olk+0.056
ays+0.056
 kettle+0.054
Neg logit contributions
lot-0.081
 farewell-0.081
 shortcut-0.080
 duel-0.076
ph-0.076
Token,
Token ID11
Feature activation+5.98e-01

Pos logit contributions
 cups+0.003
 frost+0.003
 twins+0.002
ugs+0.002
 sisters+0.002
Neg logit contributions
 shortcut-0.003
lot-0.003
 duel-0.003
 unnecessary-0.003
 farewell-0.003
Token you
Token ID345
Feature activation+2.50e-01

Pos logit contributions
 cups+0.017
ays+0.017
ugs+0.017
ums+0.016
els+0.016
Neg logit contributions
 shortcut-0.014
lot-0.012
 duel-0.012
 farewell-0.012
 noble-0.012
Token are
Token ID389
Feature activation+1.99e+00

Pos logit contributions
 cups+0.006
 frost+0.005
 twins+0.005
 sisters+0.005
cloth+0.005
Neg logit contributions
 shortcut-0.008
lot-0.007
 duel-0.007
 unnecessary-0.007
 further-0.007
Token very
Token ID845
Feature activation+1.74e+00

Pos logit contributions
 cups+0.063
ugs+0.061
ays+0.060
el+0.058
els+0.057
Neg logit contributions
 shortcut-0.039
lot-0.037
 duel-0.036
 noble-0.035
 gall-0.035
Token day
Token ID1110
Feature activation+8.13e-01

Pos logit contributions
 cups+0.034
ays+0.033
els+0.033
 frost+0.032
 folders+0.031
Neg logit contributions
 shortcut-0.020
lot-0.018
 noble-0.018
 hike-0.017
 duel-0.017
Token,
Token ID11
Feature activation+8.58e-01

Pos logit contributions
 cups+0.023
 frost+0.022
 folders+0.022
 twins+0.021
 sisters+0.021
Neg logit contributions
 shortcut-0.018
 farewell-0.017
lot-0.016
 higher-0.016
 further-0.016
Token Mia
Token ID32189
Feature activation+1.84e+00

Pos logit contributions
ugs+0.018
els+0.018
 frost+0.018
ays+0.018
 folders+0.018
Neg logit contributions
 closer-0.023
 higher-0.023
 shortcut-0.022
 further-0.022
 farewell-0.022
Token's
Token ID338
Feature activation+2.95e+00

Pos logit contributions
 cups+0.056
ugs+0.055
ays+0.055
ums+0.055
ions+0.055
Neg logit contributions
lot-0.036
 shortcut-0.034
 hike-0.034
 gall-0.032
'd-0.032
Token mom
Token ID1995
Feature activation+1.85e+00

Pos logit contributions
ays+0.064
ions+0.062
els+0.061
izz+0.060
 cups+0.060
Neg logit contributions
 shortcut-0.087
 duel-0.078
lot-0.078
 farewell-0.077
 hike-0.077
Token was
Token ID373
Feature activation+3.34e+00

Pos logit contributions
 cups+0.050
ays+0.048
ugs+0.047
ums+0.046
 frost+0.046
Neg logit contributions
lot-0.044
 shortcut-0.043
'd-0.041
 gall-0.041
 duel-0.041
Token baking
Token ID16871
Feature activation+3.08e+00

Pos logit contributions
ays+0.090
 cups+0.088
ugs+0.086
olk+0.085
els+0.085
Neg logit contributions
 shortcut-0.076
 higher-0.075
 further-0.074
 duel-0.073
 daring-0.072
Token a
Token ID257
Feature activation+1.36e+00

Pos logit contributions
ays+0.085
ugs+0.082
 cups+0.079
 sisters+0.079
el+0.079
Neg logit contributions
 higher-0.069
 further-0.062
 farther-0.060
 shortcut-0.059
 closer-0.058
Token cake
Token ID12187
Feature activation+3.19e+00

Pos logit contributions
 cups+0.027
ays+0.026
ugs+0.026
els+0.025
el+0.025
Neg logit contributions
 shortcut-0.042
 duel-0.040
lot-0.039
 noble-0.038
 higher-0.038
Token for
Token ID329
Feature activation+3.84e+00

Pos logit contributions
ays+0.098
 cups+0.094
els+0.094
izz+0.094
el+0.094
Neg logit contributions
 higher-0.056
 shortcut-0.053
 hike-0.053
 gall-0.050
 further-0.050
Token Mia
Token ID32189
Feature activation+2.06e+00

Pos logit contributions
ays+0.072
ugs+0.069
 cups+0.067
ums+0.065
 Lem+0.064
Neg logit contributions
 further-0.108
 shortcut-0.107
 duel-0.106
 farewell-0.105
 gall-0.105
Token saw
Token ID2497
Feature activation+2.31e+00

Pos logit contributions
ugs+0.082
ums+0.081
ions+0.081
 cups+0.081
ays+0.080
Neg logit contributions
lot-0.048
 shortcut-0.047
'd-0.045
 duel-0.043
 gall-0.043
Token Mom
Token ID11254
Feature activation+6.93e-01

Pos logit contributions
ugs+0.077
ums+0.073
ays+0.072
aters+0.070
cloth+0.069
Neg logit contributions
 further-0.037
 closer-0.035
 higher-0.034
 shortcut-0.032
 nearer-0.032
Tokenmy
Token ID1820
Feature activation+3.74e-01

Pos logit contributions
 cups+0.020
ugs+0.019
 frost+0.019
olk+0.018
ays+0.018
Neg logit contributions
 shortcut-0.014
lot-0.014
 gall-0.014
 higher-0.014
 further-0.013
Token carry
Token ID3283
Feature activation+2.62e+00

Pos logit contributions
 cups+0.009
 frost+0.008
ugs+0.008
 twins+0.008
 sisters+0.008
Neg logit contributions
 shortcut-0.011
lot-0.010
 duel-0.010
 gall-0.010
 further-0.010
Token some
Token ID617
Feature activation+2.20e+00

Pos logit contributions
ays+0.061
ums+0.058
 sisters+0.058
 cups+0.057
ugs+0.057
Neg logit contributions
 further-0.068
 higher-0.065
 farther-0.064
 shortcut-0.064
 closer-0.063
Token dirty
Token ID11841
Feature activation+3.58e+00

Pos logit contributions
ays+0.046
ums+0.044
ugs+0.043
izz+0.043
 cups+0.043
Neg logit contributions
 shortcut-0.066
 further-0.061
 farther-0.059
 unnecessary-0.059
 spear-0.058
Token plates
Token ID13375
Feature activation+2.82e+00

Pos logit contributions
izz+0.070
ums+0.070
ays+0.069
ena+0.067
illian+0.066
Neg logit contributions
 further-0.100
 farther-0.093
 shortcut-0.092
 higher-0.088
 closer-0.088
Token to
Token ID284
Feature activation+2.88e+00

Pos logit contributions
olk+0.080
 frost+0.080
ums+0.080
els+0.079
 sisters+0.078
Neg logit contributions
 further-0.058
 shortcut-0.057
 farther-0.054
 closer-0.053
 higher-0.053
Token the
Token ID262
Feature activation+2.47e+00

Pos logit contributions
ays+0.082
ugs+0.081
ums+0.074
els+0.072
 twins+0.071
Neg logit contributions
 shortcut-0.064
 further-0.063
 duel-0.060
 closer-0.060
 hike-0.060
Token dish
Token ID9433
Feature activation+2.39e+00

Pos logit contributions
ugs+0.039
ays+0.038
els+0.038
el+0.037
izz+0.037
Neg logit contributions
 shortcut-0.087
 further-0.081
 duel-0.081
lot-0.080
 farther-0.079
Tokenwasher
Token ID45146
Feature activation+2.33e+00

Pos logit contributions
 frost+0.063
els+0.062
olk+0.061
 icing+0.060
 cups+0.060
Neg logit contributions
 shortcut-0.051
lot-0.048
 further-0.047
 hike-0.045
 ashore-0.045
Token trash
Token ID13913
Feature activation+1.57e+00

Pos logit contributions
 cups+0.019
ugs+0.017
 frost+0.017
 sisters+0.017
 twins+0.016
Neg logit contributions
 shortcut-0.031
lot-0.029
 duel-0.029
 farewell-0.028
 further-0.028
Token.
Token ID13
Feature activation+1.26e+00

Pos logit contributions
ugs+0.039
 cups+0.039
ums+0.039
 sisters+0.038
 folders+0.037
Neg logit contributions
 shortcut-0.040
 further-0.037
 higher-0.036
 farther-0.036
lot-0.035
Token "
Token ID366
Feature activation+4.46e-01

Pos logit contributions
 cups+0.023
 frost+0.022
 sisters+0.021
ugs+0.020
els+0.020
Neg logit contributions
 shortcut-0.041
lot-0.040
 duel-0.038
 Skip-0.038
 higher-0.038
TokenNo
Token ID2949
Feature activation+3.73e-01

Pos logit contributions
 cups+0.012
ugs+0.012
 sisters+0.012
ums+0.011
els+0.011
Neg logit contributions
 shortcut-0.011
Hop-0.011
lot-0.010
Left-0.010
 duel-0.010
Token pasta
Token ID26296
Feature activation-6.34e-01

Pos logit contributions
 cups+0.009
 frost+0.008
 twins+0.008
ugs+0.008
 sisters+0.008
Neg logit contributions
 shortcut-0.011
lot-0.010
 duel-0.010
 unnecessary-0.010
 farewell-0.010
Token for
Token ID329
Feature activation+3.16e+00

Pos logit contributions
 shortcut+0.018
lot+0.018
 duel+0.017
 farewell+0.017
 unnecessary+0.017
Neg logit contributions
 cups-0.015
 frost-0.014
 twins-0.013
ugs-0.013
 sisters-0.013
Token you
Token ID345
Feature activation+1.22e-02

Pos logit contributions
ays+0.086
 cups+0.084
ugs+0.082
ums+0.079
 frost+0.079
Neg logit contributions
 shortcut-0.072
 duel-0.068
 unnecessary-0.064
 gall-0.064
 defeat-0.063
Token today
Token ID1909
Feature activation+1.92e+00

Pos logit contributions
 cups+0.000
 frost+0.000
 twins+0.000
ugs+0.000
 sisters+0.000
Neg logit contributions
 shortcut-0.000
lot-0.000
 duel-0.000
 noble-0.000
 farewell-0.000
Token.
Token ID13
Feature activation+5.39e-01

Pos logit contributions
 cups+0.049
 frost+0.046
ugs+0.044
 sisters+0.044
 twins+0.044
Neg logit contributions
 shortcut-0.051
lot-0.049
 duel-0.047
 further-0.045
 gall-0.045
Token You
Token ID921
Feature activation+4.86e-02

Pos logit contributions
 cups+0.014
ugs+0.014
els+0.013
ries+0.013
 frost+0.013
Neg logit contributions
lot-0.012
 shortcut-0.011
 duel-0.011
 Skip-0.011
Left-0.011
Token will
Token ID481
Feature activation+4.04e-01

Pos logit contributions
 cups+0.001
 frost+0.001
 twins+0.001
ugs+0.001
 sisters+0.001
Neg logit contributions
 shortcut-0.001
lot-0.001
 duel-0.001
 unnecessary-0.001
 farewell-0.001
Token but
Token ID475
Feature activation+9.74e-01

Pos logit contributions
 shortcut+0.021
lot+0.021
 duel+0.019
 farewell+0.019
 noble+0.018
Neg logit contributions
 cups-0.026
 frost-0.024
 twins-0.023
 sisters-0.023
cloth-0.022
Token it
Token ID340
Feature activation-6.47e-01

Pos logit contributions
 cups+0.025
 frost+0.024
ugs+0.023
 twins+0.023
 sisters+0.022
Neg logit contributions
 shortcut-0.026
lot-0.025
 duel-0.024
 further-0.023
 farewell-0.023
Token was
Token ID373
Feature activation+1.14e+00

Pos logit contributions
 shortcut+0.018
lot+0.017
 duel+0.017
 unnecessary+0.016
 farewell+0.016
Neg logit contributions
 cups-0.017
 frost-0.015
 twins-0.015
el-0.014
 sisters-0.014
Token too
Token ID1165
Feature activation+4.57e-01

Pos logit contributions
 cups+0.029
 frost+0.027
 twins+0.026
 sisters+0.025
ugs+0.025
Neg logit contributions
 shortcut-0.031
lot-0.031
 duel-0.029
 farewell-0.028
 unnecessary-0.028
Token heavy
Token ID4334
Feature activation+7.43e-01

Pos logit contributions
 cups+0.016
 frost+0.015
ugs+0.014
 twins+0.014
 mug+0.014
Neg logit contributions
 shortcut-0.008
lot-0.008
 duel-0.008
 unnecessary-0.007
 farewell-0.007
Token for
Token ID329
Feature activation+3.82e+00

Pos logit contributions
 cups+0.017
 frost+0.015
 twins+0.015
ugs+0.014
 sisters+0.014
Neg logit contributions
 shortcut-0.023
lot-0.022
 duel-0.021
 noble-0.020
 farewell-0.020
Token her
Token ID607
Feature activation+9.78e-01

Pos logit contributions
 cups+0.061
ays+0.056
ums+0.053
ugs+0.053
 frost+0.052
Neg logit contributions
 shortcut-0.133
 unnecessary-0.127
 further-0.126
 farewell-0.125
 gall-0.125
Token.
Token ID13
Feature activation+1.29e+00

Pos logit contributions
 cups+0.022
 frost+0.020
 twins+0.019
ugs+0.019
 sisters+0.019
Neg logit contributions
 shortcut-0.030
lot-0.029
 duel-0.028
 unnecessary-0.027
 farewell-0.027
Token Her
Token ID2332
Feature activation+1.02e+00

Pos logit contributions
 cups+0.021
 frost+0.020
ugs+0.018
ions+0.018
els+0.018
Neg logit contributions
 shortcut-0.044
lot-0.044
 farewell-0.042
 duel-0.042
 higher-0.041
Token daddy
Token ID35822
Feature activation+6.57e-01

Pos logit contributions
 cups+0.020
 frost+0.019
ions+0.018
izz+0.018
ays+0.018
Neg logit contributions
 shortcut-0.034
lot-0.032
 duel-0.031
 farewell-0.030
 spear-0.030
Token helped
Token ID4193
Feature activation+1.22e+00

Pos logit contributions
 cups+0.014
 frost+0.012
 twins+0.012
 sisters+0.012
ugs+0.011
Neg logit contributions
 shortcut-0.021
lot-0.020
 duel-0.020
 unnecessary-0.020
 farewell-0.019
Tokenâ
Token ID22940
Feature activation-9.85e-01

Pos logit contributions
 cups+0.014
 frost+0.013
ugs+0.013
els+0.012
 sisters+0.012
Neg logit contributions
 shortcut-0.018
lot-0.017
 duel-0.016
 higher-0.016
 further-0.016
Token€
Token ID26391
Feature activation+8.92e-02

Pos logit contributions
 shortcut+0.022
lot+0.021
 duel+0.020
 unnecessary+0.019
 farewell+0.018
Neg logit contributions
 cups-0.031
 frost-0.029
 twins-0.028
ugs-0.028
 sisters-0.027
Token
Token ID220
Feature activation-6.27e-01

Pos logit contributions
 cups+0.002
 frost+0.002
 twins+0.002
 sisters+0.002
ugs+0.002
Neg logit contributions
 shortcut-0.002
lot-0.002
 duel-0.002
 unnecessary-0.002
 gall-0.002
Token
Token ID198
Feature activation+1.32e-01

Pos logit contributions
 shortcut+0.019
 duel+0.018
lot+0.018
 farewell+0.017
 noble+0.017
Neg logit contributions
 cups-0.014
 twins-0.013
ugs-0.013
 frost-0.013
ays-0.012
Token
Token ID198
Feature activation+1.28e+00

Pos logit contributions
 cups+0.003
 frost+0.003
 twins+0.003
ugs+0.003
 sisters+0.002
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.004
 unnecessary-0.004
 noble-0.004
TokenThe
Token ID464
Feature activation+9.04e-01

Pos logit contributions
 cups+0.037
 frost+0.036
ays+0.035
els+0.035
 sisters+0.034
Neg logit contributions
 shortcut-0.030
Hop-0.028
lot-0.027
 duel-0.027
 further-0.027
Token little
Token ID1310
Feature activation+1.85e+00

Pos logit contributions
 cups+0.027
 frost+0.025
 twins+0.024
 sisters+0.024
ugs+0.024
Neg logit contributions
 shortcut-0.021
lot-0.020
 duel-0.019
 unnecessary-0.018
 gall-0.018
Token girl
Token ID2576
Feature activation+1.37e+00

Pos logit contributions
ugs+0.050
ays+0.048
izz+0.048
els+0.048
ions+0.048
Neg logit contributions
 further-0.045
 higher-0.044
 shortcut-0.043
 farther-0.043
 closer-0.043
Token was
Token ID373
Feature activation+1.53e+00

Pos logit contributions
 cups+0.033
 frost+0.030
ugs+0.029
 twins+0.029
 sisters+0.029
Neg logit contributions
 shortcut-0.038
lot-0.037
 duel-0.036
 gall-0.035
 noble-0.034
Token very
Token ID845
Feature activation+1.04e+00

Pos logit contributions
 cups+0.044
 frost+0.041
 twins+0.040
 sisters+0.039
ugs+0.039
Neg logit contributions
 shortcut-0.037
lot-0.035
 duel-0.034
 unnecessary-0.032
 farewell-0.032
Token excited
Token ID6568
Feature activation+1.06e+00

Pos logit contributions
 cups+0.032
 frost+0.029
 twins+0.029
 sisters+0.028
ugs+0.028
Neg logit contributions
 shortcut-0.024
lot-0.023
 duel-0.022
 unnecessary-0.021
 farewell-0.021
Token day
Token ID1110
Feature activation+1.38e+00

Pos logit contributions
 cups+0.022
 frost+0.020
 twins+0.020
 mug+0.020
 sisters+0.019
Neg logit contributions
lot-0.024
 shortcut-0.024
 duel-0.023
 unnecessary-0.023
 noble-0.022
Token Anna
Token ID11735
Feature activation-4.45e-02

Pos logit contributions
 cups+0.034
 frost+0.031
ugs+0.030
 twins+0.029
 sisters+0.029
Neg logit contributions
 shortcut-0.039
lot-0.038
 duel-0.036
 farewell-0.035
 further-0.035
Token would
Token ID561
Feature activation-2.58e-01

Pos logit contributions
 shortcut+0.001
lot+0.001
 duel+0.001
 unnecessary+0.001
 further+0.001
Neg logit contributions
 cups-0.001
 frost-0.001
 sisters-0.001
 twins-0.001
ugs-0.001
Token wake
Token ID7765
Feature activation-1.14e+00

Pos logit contributions
 shortcut+0.006
lot+0.006
 unnecessary+0.006
 duel+0.006
 further+0.006
Neg logit contributions
 cups-0.007
 sisters-0.007
 frost-0.006
 twins-0.006
cloth-0.006
Token up
Token ID510
Feature activation-3.42e-01

Pos logit contributions
 shortcut+0.028
lot+0.026
 duel+0.026
'd+0.025
 steep+0.024
Neg logit contributions
 cups-0.033
 twins-0.031
 sisters-0.031
el-0.030
els-0.030
Token early
Token ID1903
Feature activation+8.31e-01

Pos logit contributions
 shortcut+0.011
lot+0.011
 duel+0.011
 unnecessary+0.010
Hop+0.010
Neg logit contributions
 cups-0.007
 sisters-0.006
 twins-0.006
 frost-0.006
el-0.006
Token and
Token ID290
Feature activation-1.77e+00

Pos logit contributions
 cups+0.018
 frost+0.016
 twins+0.015
 sisters+0.015
 mug+0.015
Neg logit contributions
 shortcut-0.026
 unnecessary-0.025
lot-0.025
 duel-0.024
 farewell-0.023
Token go
Token ID467
Feature activation-1.96e+00

Pos logit contributions
 shortcut+0.047
lot+0.045
 unnecessary+0.044
 duel+0.042
Hop+0.041
Neg logit contributions
 cups-0.045
 sisters-0.042
 frost-0.041
 twins-0.040
 paints-0.040
Token to
Token ID284
Feature activation-9.83e-01

Pos logit contributions
lot+0.057
 shortcut+0.056
 duel+0.055
 unnecessary+0.055
rated+0.053
Neg logit contributions
 cups-0.045
 twins-0.039
 sisters-0.038
 frost-0.038
ugs-0.037
Token the
Token ID262
Feature activation-1.24e+00

Pos logit contributions
lot+0.032
 shortcut+0.032
Hop+0.030
 duel+0.030
 farewell+0.030
Neg logit contributions
 cups-0.020
 sisters-0.018
 twins-0.017
 frost-0.017
 mug-0.016
Token rock
Token ID3881
Feature activation-1.39e+00

Pos logit contributions
lot+0.038
rated+0.035
 shortcut+0.035
 farewell+0.035
 unnecessary+0.034
Neg logit contributions
 cups-0.028
 twins-0.026
 mug-0.025
 tub-0.025
 sisters-0.025
Token.
Token ID13
Feature activation+1.72e+00

Pos logit contributions
 cups+0.032
 frost+0.030
ugs+0.029
 sisters+0.028
 twins+0.028
Neg logit contributions
 shortcut-0.036
lot-0.035
 duel-0.033
 gall-0.032
 further-0.032
Token She
Token ID1375
Feature activation+1.42e+00

Pos logit contributions
 cups+0.035
ugs+0.034
 frost+0.033
 sisters+0.033
el+0.032
Neg logit contributions
 shortcut-0.053
lot-0.052
 farewell-0.050
 duel-0.048
 Skip-0.047
Token hugged
Token ID41130
Feature activation+6.93e-01

Pos logit contributions
 cups+0.036
 frost+0.033
 twins+0.032
 sisters+0.031
cloth+0.030
Neg logit contributions
 shortcut-0.040
lot-0.038
 duel-0.037
 unnecessary-0.036
 farewell-0.035
Token them
Token ID606
Feature activation-1.45e+00

Pos logit contributions
 cups+0.017
 frost+0.016
ugs+0.015
 twins+0.015
 sisters+0.015
Neg logit contributions
 shortcut-0.020
lot-0.019
 duel-0.018
 farewell-0.018
 further-0.017
Token and
Token ID290
Feature activation+7.25e-01

Pos logit contributions
 unnecessary+0.053
 shortcut+0.053
 duel+0.052
lot+0.052
 noble+0.049
Neg logit contributions
 cups-0.026
 twins-0.023
 sisters-0.022
 frost-0.022
 mug-0.020
Token asked
Token ID1965
Feature activation+6.28e-01

Pos logit contributions
 cups+0.017
ugs+0.016
 frost+0.015
ays+0.015
ums+0.015
Neg logit contributions
 shortcut-0.021
lot-0.021
 duel-0.020
 gall-0.020
 further-0.019
Token them
Token ID606
Feature activation-2.35e+00

Pos logit contributions
 cups+0.014
 twins+0.012
 frost+0.012
 sisters+0.011
cloth+0.011
Neg logit contributions
 shortcut-0.021
lot-0.021
 duel-0.020
 noble-0.019
 gall-0.019
Token what
Token ID644
Feature activation-1.56e-01

Pos logit contributions
lot+0.071
 shortcut+0.070
 duel+0.069
 unnecessary+0.067
Hop+0.064
Neg logit contributions
 cups-0.060
 twins-0.054
 sisters-0.053
cloth-0.048
 frost-0.045
Token happened
Token ID3022
Feature activation+1.98e-01

Pos logit contributions
 shortcut+0.004
lot+0.004
 duel+0.004
 gall+0.004
 noble+0.004
Neg logit contributions
 cups-0.004
 frost-0.004
ugs-0.003
 twins-0.003
 sisters-0.003
Token.
Token ID13
Feature activation+1.65e+00

Pos logit contributions
 cups+0.003
 twins+0.003
 frost+0.003
 sisters+0.003
cloth+0.002
Neg logit contributions
 shortcut-0.007
 duel-0.007
lot-0.007
 unnecessary-0.007
 farewell-0.007
Token
Token ID198
Feature activation+8.03e-01

Pos logit contributions
 cups+0.042
 frost+0.039
ugs+0.038
 sisters+0.037
el+0.037
Neg logit contributions
 shortcut-0.044
lot-0.043
 duel-0.041
 farewell-0.039
 Skip-0.039
Token the
Token ID262
Feature activation+4.30e-01

Pos logit contributions
 cups+0.005
 frost+0.005
 twins+0.004
ugs+0.004
 sisters+0.004
Neg logit contributions
 shortcut-0.006
lot-0.005
 duel-0.005
 farewell-0.005
 noble-0.005
Token ground
Token ID2323
Feature activation+3.22e-01

Pos logit contributions
 cups+0.009
 frost+0.008
ugs+0.008
 sisters+0.008
 twins+0.008
Neg logit contributions
 shortcut-0.014
lot-0.013
 duel-0.013
 unnecessary-0.012
 further-0.012
Token and
Token ID290
Feature activation-8.21e-01

Pos logit contributions
 cups+0.007
 frost+0.006
 twins+0.005
cloth+0.005
el+0.005
Neg logit contributions
 shortcut-0.011
lot-0.010
 duel-0.010
 unnecessary-0.010
 farewell-0.010
Token noticed
Token ID6810
Feature activation-7.27e-01

Pos logit contributions
 shortcut+0.024
lot+0.023
 unnecessary+0.023
 duel+0.022
 farewell+0.022
Neg logit contributions
 cups-0.020
 frost-0.019
 twins-0.018
 sisters-0.018
cloth-0.016
Token it
Token ID340
Feature activation-7.45e-01

Pos logit contributions
 shortcut+0.023
lot+0.023
 duel+0.022
 unnecessary+0.021
 noble+0.021
Neg logit contributions
 cups-0.017
 frost-0.016
 twins-0.015
els-0.014
ions-0.014
Token was
Token ID373
Feature activation+2.01e+00

Pos logit contributions
 shortcut+0.024
lot+0.023
 duel+0.023
 unnecessary+0.023
 farewell+0.022
Neg logit contributions
 cups-0.016
 frost-0.015
el-0.014
 twins-0.014
els-0.013
Token a
Token ID257
Feature activation-2.35e-01

Pos logit contributions
 cups+0.053
 frost+0.048
ugs+0.048
 twins+0.047
 sisters+0.047
Neg logit contributions
 shortcut-0.053
lot-0.051
 duel-0.049
 gall-0.047
 noble-0.047
Token leak
Token ID13044
Feature activation+1.40e+00

Pos logit contributions
lot+0.007
 shortcut+0.007
 duel+0.006
 farewell+0.006
 unnecessary+0.006
Neg logit contributions
 cups-0.006
 frost-0.005
 twins-0.005
 mug-0.005
 sisters-0.005
Token.
Token ID13
Feature activation-1.06e+00

Pos logit contributions
 cups+0.032
 frost+0.029
 twins+0.028
ugs+0.028
 sisters+0.028
Neg logit contributions
 shortcut-0.042
lot-0.040
 duel-0.039
 gall-0.038
 further-0.037
Token
Token ID198
Feature activation-1.32e+00

Pos logit contributions
 shortcut+0.032
 unnecessary+0.031
 duel+0.030
lot+0.029
 gall+0.028
Neg logit contributions
 cups-0.025
 frost-0.022
 twins-0.021
ays-0.021
ugs-0.020
Token
Token ID198
Feature activation-5.21e-01

Pos logit contributions
 shortcut+0.045
 unnecessary+0.045
 duel+0.043
lot+0.042
 gall+0.041
Neg logit contributions
 cups-0.023
 twins-0.020
 frost-0.020
Mom-0.020
izz-0.020
Token we
Token ID356
Feature activation+1.63e-01

Pos logit contributions
lot+0.007
 shortcut+0.007
 duel+0.006
 farewell+0.006
 gall+0.006
Neg logit contributions
 cups-0.006
 frost-0.005
 twins-0.005
 sisters-0.005
 mug-0.005
Token should
Token ID815
Feature activation+1.23e+00

Pos logit contributions
 cups+0.005
 frost+0.004
 twins+0.004
ugs+0.004
 sisters+0.004
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.004
 unnecessary-0.003
 farewell-0.003
Token always
Token ID1464
Feature activation-4.06e-01

Pos logit contributions
 cups+0.037
 frost+0.034
ugs+0.033
 twins+0.033
ays+0.033
Neg logit contributions
 shortcut-0.028
lot-0.027
 duel-0.026
 farewell-0.025
 gall-0.025
Token let
Token ID1309
Feature activation-2.30e-01

Pos logit contributions
 shortcut+0.010
lot+0.010
 duel+0.010
 unnecessary+0.010
 farewell+0.009
Neg logit contributions
 cups-0.010
 frost-0.010
 twins-0.010
 sisters-0.010
el-0.009
Token bugs
Token ID11316
Feature activation-2.09e+00

Pos logit contributions
lot+0.007
 shortcut+0.007
 duel+0.007
 noble+0.007
 steep+0.007
Neg logit contributions
 cups-0.005
 twins-0.004
 sisters-0.004
 frost-0.004
el-0.004
Token stay
Token ID2652
Feature activation+6.45e-01

Pos logit contributions
lot+0.056
 shortcut+0.055
 duel+0.052
 noble+0.051
 farewell+0.050
Neg logit contributions
 cups-0.055
 frost-0.050
 sisters-0.048
 twins-0.047
cloth-0.047
Token outside
Token ID2354
Feature activation+7.74e-01

Pos logit contributions
 cups+0.021
 frost+0.019
 twins+0.019
 sisters+0.019
el+0.018
Neg logit contributions
 shortcut-0.014
lot-0.013
 duel-0.013
 gall-0.012
 farewell-0.012
Token where
Token ID810
Feature activation-3.07e-01

Pos logit contributions
 cups+0.020
 frost+0.018
 twins+0.017
ugs+0.017
 sisters+0.017
Neg logit contributions
 shortcut-0.021
lot-0.021
 duel-0.020
 unnecessary-0.019
 farewell-0.019
Token they
Token ID484
Feature activation-3.77e-01

Pos logit contributions
 shortcut+0.009
lot+0.009
 duel+0.008
 gall+0.008
 farewell+0.008
Neg logit contributions
 cups-0.007
 frost-0.007
 twins-0.007
 sisters-0.006
ugs-0.006
Token belong
Token ID5594
Feature activation+9.45e-01

Pos logit contributions
lot+0.012
 shortcut+0.012
 farewell+0.011
 duel+0.011
 Skip+0.011
Neg logit contributions
 cups-0.008
 frost-0.008
 sisters-0.007
 twins-0.007
 paints-0.007
Token."
Token ID526
Feature activation+9.32e-01

Pos logit contributions
 cups+0.020
 twins+0.018
 frost+0.017
 sisters+0.017
ays+0.016
Neg logit contributions
 shortcut-0.031
 duel-0.030
lot-0.030
 gall-0.028
 farewell-0.028
Token of
Token ID286
Feature activation-1.11e+00

Pos logit contributions
 cups+0.032
 frost+0.030
 twins+0.028
ugs+0.028
 sisters+0.028
Neg logit contributions
 shortcut-0.045
lot-0.043
 duel-0.042
 further-0.041
 gall-0.040
Token catch
Token ID4929
Feature activation-4.90e-01

Pos logit contributions
 farewell+0.036
lot+0.034
 shortcut+0.033
rated+0.033
stru+0.031
Neg logit contributions
 cups-0.025
 twins-0.023
 dough-0.022
 sisters-0.022
 frost-0.021
Token when
Token ID618
Feature activation+4.29e-01

Pos logit contributions
 shortcut+0.016
lot+0.016
 duel+0.015
 farewell+0.015
 noble+0.015
Neg logit contributions
 cups-0.010
 twins-0.009
 frost-0.008
 sisters-0.008
 mug-0.008
Token Pete
Token ID17542
Feature activation-1.93e-02

Pos logit contributions
 cups+0.008
 twins+0.007
 sisters+0.007
 frost+0.007
el+0.006
Neg logit contributions
 shortcut-0.015
lot-0.014
 duel-0.014
 noble-0.013
 unnecessary-0.013
Token suddenly
Token ID6451
Feature activation+7.31e-01

Pos logit contributions
 shortcut+0.001
 unnecessary+0.001
 farewell+0.001
 duel+0.001
lot+0.001
Neg logit contributions
 cups-0.000
 frost-0.000
 twins-0.000
 sisters-0.000
 mug-0.000
Token dropped
Token ID5710
Feature activation-2.91e-01

Pos logit contributions
 cups+0.018
 frost+0.016
 twins+0.015
 sisters+0.015
el+0.015
Neg logit contributions
 shortcut-0.021
 duel-0.020
lot-0.020
 noble-0.019
 unnecessary-0.019
Token his
Token ID465
Feature activation+2.05e-01

Pos logit contributions
 shortcut+0.008
lot+0.008
 duel+0.008
 noble+0.007
 unnecessary+0.007
Neg logit contributions
 cups-0.007
 frost-0.007
 twins-0.007
 sisters-0.006
el-0.006
Token glove
Token ID29144
Feature activation+7.56e-01

Pos logit contributions
 cups+0.005
 twins+0.005
 mug+0.005
 frost+0.005
 sisters+0.005
Neg logit contributions
 shortcut-0.006
lot-0.006
 duel-0.005
 further-0.005
 unnecessary-0.005
Token in
Token ID287
Feature activation+1.55e+00

Pos logit contributions
 cups+0.017
 frost+0.015
 twins+0.015
ugs+0.014
 sisters+0.014
Neg logit contributions
 shortcut-0.023
lot-0.022
 duel-0.022
 unnecessary-0.021
 farewell-0.021
Token the
Token ID262
Feature activation+6.50e-01

Pos logit contributions
ugs+0.045
 cups+0.044
 frost+0.042
ums+0.042
ays+0.042
Neg logit contributions
 shortcut-0.037
 duel-0.033
 further-0.033
lot-0.033
 farther-0.032
Token dirt
Token ID13647
Feature activation+1.29e+00

Pos logit contributions
 cups+0.016
 frost+0.015
 twins+0.014
ugs+0.014
 sisters+0.014
Neg logit contributions
 shortcut-0.019
lot-0.018
 duel-0.017
 unnecessary-0.017
 farewell-0.017
Token a
Token ID257
Feature activation+4.43e-02

Pos logit contributions
 cups+0.057
ays+0.056
ugs+0.055
els+0.054
 frost+0.054
Neg logit contributions
 shortcut-0.028
 gall-0.026
lot-0.026
 higher-0.025
 duel-0.025
Token brave
Token ID14802
Feature activation+3.37e-01

Pos logit contributions
 cups+0.001
 frost+0.001
 twins+0.001
 sisters+0.001
ugs+0.001
Neg logit contributions
 shortcut-0.001
lot-0.001
 duel-0.001
 unnecessary-0.001
 farewell-0.001
Token and
Token ID290
Feature activation-1.97e+00

Pos logit contributions
 cups+0.007
 frost+0.006
 twins+0.006
 sisters+0.005
ugs+0.005
Neg logit contributions
 shortcut-0.011
lot-0.011
 duel-0.011
 unnecessary-0.010
 farewell-0.010
Token tough
Token ID5802
Feature activation+5.95e-01

Pos logit contributions
 shortcut+0.047
lot+0.047
 duel+0.045
 unnecessary+0.043
 farewell+0.042
Neg logit contributions
 cups-0.055
 twins-0.051
 sisters-0.050
 frost-0.050
 mug-0.048
Token girl
Token ID2576
Feature activation+1.31e+00

Pos logit contributions
 cups+0.012
 twins+0.011
 sisters+0.010
 frost+0.010
 mug+0.010
Neg logit contributions
lot-0.020
 shortcut-0.019
 duel-0.019
 unnecessary-0.018
 nearer-0.018
Token and
Token ID290
Feature activation-7.64e-01

Pos logit contributions
 cups+0.035
ays+0.033
 frost+0.033
ugs+0.032
 icing+0.032
Neg logit contributions
 shortcut-0.031
lot-0.030
 gall-0.029
 noble-0.029
 duel-0.028
Token she
Token ID673
Feature activation-8.30e-01

Pos logit contributions
 shortcut+0.018
lot+0.017
 duel+0.017
 noble+0.016
 gall+0.016
Neg logit contributions
 cups-0.022
 frost-0.021
ays-0.021
ugs-0.020
els-0.020
Token liked
Token ID8288
Feature activation-1.92e-01

Pos logit contributions
 shortcut+0.019
lot+0.019
 duel+0.018
 gall+0.017
 noble+0.017
Neg logit contributions
 cups-0.025
 frost-0.023
ugs-0.022
 twins-0.022
 sisters-0.022
Token going
Token ID1016
Feature activation-2.39e+00

Pos logit contributions
 shortcut+0.005
lot+0.005
 duel+0.005
 farewell+0.004
 unnecessary+0.004
Neg logit contributions
 cups-0.005
 frost-0.005
 twins-0.005
 sisters-0.005
 mug-0.005
Token on
Token ID319
Feature activation-7.55e-02

Pos logit contributions
lot+0.080
 duel+0.079
 farewell+0.079
 shortcut+0.076
Hop+0.075
Neg logit contributions
 cups-0.043
 twins-0.039
 sisters-0.035
ugs-0.035
 paints-0.034
Token adventures
Token ID17545
Feature activation+4.60e-02

Pos logit contributions
 shortcut+0.002
lot+0.002
 duel+0.001
 farewell+0.001
 noble+0.001
Neg logit contributions
 cups-0.002
 frost-0.002
 twins-0.002
 sisters-0.002
 folders-0.002
Token saw
Token ID2497
Feature activation+5.43e-01

Pos logit contributions
 cups+0.033
 frost+0.030
 twins+0.028
 sisters+0.028
cloth+0.027
Neg logit contributions
 shortcut-0.035
lot-0.033
 duel-0.032
 unnecessary-0.031
 farewell-0.031
Token a
Token ID257
Feature activation-1.01e+00

Pos logit contributions
 cups+0.015
ugs+0.014
 frost+0.013
ays+0.013
cloth+0.013
Neg logit contributions
 shortcut-0.013
lot-0.013
 duel-0.012
 further-0.012
 unnecessary-0.012
Token butterfly
Token ID35113
Feature activation-1.12e+00

Pos logit contributions
lot+0.028
 shortcut+0.027
 farewell+0.026
 duel+0.026
 unnecessary+0.026
Neg logit contributions
 cups-0.025
 frost-0.023
 mug-0.023
 twins-0.023
 sisters-0.022
Token.
Token ID13
Feature activation+1.62e-02

Pos logit contributions
 shortcut+0.031
lot+0.029
 duel+0.029
 gall+0.027
 further+0.027
Neg logit contributions
 cups-0.029
 frost-0.026
ugs-0.025
 twins-0.025
 sisters-0.025
Token She
Token ID1375
Feature activation-4.05e-01

Pos logit contributions
 cups+0.000
 frost+0.000
els+0.000
 sisters+0.000
 twins+0.000
Neg logit contributions
 shortcut-0.000
lot-0.000
 duel-0.000
 gall-0.000
 higher-0.000
Token started
Token ID2067
Feature activation-7.31e-01

Pos logit contributions
 shortcut+0.013
 unnecessary+0.012
lot+0.012
 duel+0.012
 further+0.011
Neg logit contributions
 cups-0.009
 frost-0.008
 sisters-0.008
 twins-0.008
el-0.008
Token to
Token ID284
Feature activation-2.42e-01

Pos logit contributions
 shortcut+0.019
lot+0.019
 noble+0.017
 duel+0.017
 unnecessary+0.017
Neg logit contributions
 cups-0.020
 twins-0.018
 sisters-0.018
 frost-0.017
 mug-0.017
Token chase
Token ID15505
Feature activation-9.19e-01

Pos logit contributions
lot+0.006
 shortcut+0.006
 unnecessary+0.005
 noble+0.005
 farewell+0.005
Neg logit contributions
 cups-0.007
 mug-0.006
 sisters-0.006
 frost-0.006
 twins-0.006
Token the
Token ID262
Feature activation-1.59e-01

Pos logit contributions
 shortcut+0.026
lot+0.026
 duel+0.025
 noble+0.024
 gall+0.023
Neg logit contributions
 cups-0.023
 frost-0.020
 twins-0.020
el-0.020
 sisters-0.019
Token butterfly
Token ID35113
Feature activation-3.14e+00

Pos logit contributions
 shortcut+0.003
lot+0.003
 duel+0.003
 further+0.003
 unnecessary+0.003
Neg logit contributions
 cups-0.005
 frost-0.005
ugs-0.005
 twins-0.005
ays-0.005
Token and
Token ID290
Feature activation-1.05e+00

Pos logit contributions
lot+0.111
 shortcut+0.104
 duel+0.103
 noble+0.102
 unnecessary+0.101
Neg logit contributions
 cups-0.057
 mug-0.046
 sisters-0.045
el-0.045
 twins-0.044
Token that
Token ID326
Feature activation-2.12e+00

Pos logit contributions
 shortcut+0.033
lot+0.032
 duel+0.030
 unnecessary+0.029
 farewell+0.029
Neg logit contributions
 cups-0.027
 frost-0.025
 twins-0.024
 sisters-0.024
ays-0.023
Token it
Token ID340
Feature activation-4.11e-01

Pos logit contributions
 shortcut+0.055
lot+0.054
 duel+0.051
 unnecessary+0.050
 farewell+0.049
Neg logit contributions
 cups-0.058
 frost-0.053
 twins-0.052
 sisters-0.051
 folders-0.049
Token was
Token ID373
Feature activation+2.71e+00

Pos logit contributions
 shortcut+0.012
lot+0.012
 duel+0.011
 unnecessary+0.011
 farewell+0.011
Neg logit contributions
 cups-0.010
 frost-0.009
 twins-0.009
 sisters-0.008
el-0.008
Token now
Token ID783
Feature activation+8.01e-01

Pos logit contributions
 cups+0.078
ugs+0.074
 frost+0.072
ays+0.071
 twins+0.069
Neg logit contributions
 shortcut-0.064
lot-0.061
 duel-0.059
 noble-0.056
 further-0.056
Token organized
Token ID8389
Feature activation+1.31e-01

Pos logit contributions
 cups+0.021
 frost+0.020
 twins+0.019
ugs+0.019
 sisters+0.019
Neg logit contributions
 shortcut-0.021
lot-0.020
 duel-0.020
 unnecessary-0.019
 farewell-0.019
Token.
Token ID13
Feature activation-1.07e+00

Pos logit contributions
 cups+0.003
 frost+0.002
 twins+0.002
cloth+0.002
 sisters+0.002
Neg logit contributions
 shortcut-0.004
lot-0.004
 duel-0.004
 unnecessary-0.004
 farewell-0.004
Token He
Token ID679
Feature activation-3.46e-02

Pos logit contributions
 shortcut+0.029
lot+0.027
 duel+0.027
 unnecessary+0.026
 further+0.025
Neg logit contributions
 cups-0.029
 frost-0.026
 twins-0.026
ugs-0.025
 sisters-0.025
Token smiled
Token ID13541
Feature activation-7.03e-01

Pos logit contributions
 shortcut+0.001
lot+0.001
 duel+0.001
 gall+0.001
 noble+0.001
Neg logit contributions
 cups-0.001
 frost-0.001
ugs-0.001
 twins-0.001
ays-0.001
Token and
Token ID290
Feature activation-4.55e-01

Pos logit contributions
 shortcut+0.024
 duel+0.023
lot+0.022
 unnecessary+0.022
 gall+0.021
Neg logit contributions
 cups-0.014
 twins-0.013
 frost-0.012
 sisters-0.012
el-0.011
Token closed
Token ID4838
Feature activation-6.21e-01

Pos logit contributions
 shortcut+0.012
lot+0.012
 duel+0.011
 gall+0.011
 farewell+0.011
Neg logit contributions
 cups-0.012
 frost-0.011
ugs-0.011
ays-0.010
 twins-0.010
Token the
Token ID262
Feature activation+6.10e-01

Pos logit contributions
 shortcut+0.017
lot+0.017
 duel+0.016
 unnecessary+0.016
 noble+0.015
Neg logit contributions
 cups-0.016
 frost-0.015
 twins-0.014
el-0.014
cloth-0.014
Token you
Token ID345
Feature activation-5.14e-01

Pos logit contributions
 shortcut+0.006
lot+0.005
 duel+0.005
 farewell+0.005
 unnecessary+0.005
Neg logit contributions
 cups-0.004
 twins-0.003
 frost-0.003
 sisters-0.003
ions-0.003
Token make
Token ID787
Feature activation+3.91e-01

Pos logit contributions
 shortcut+0.014
lot+0.014
 unnecessary+0.013
 duel+0.013
 farewell+0.012
Neg logit contributions
 cups-0.013
 frost-0.012
 sisters-0.012
 twins-0.012
el-0.011
Token things
Token ID1243
Feature activation-1.44e+00

Pos logit contributions
 cups+0.009
 frost+0.009
 twins+0.008
 sisters+0.008
el+0.008
Neg logit contributions
 shortcut-0.012
lot-0.011
 duel-0.011
 farewell-0.010
 gall-0.010
Token vanish
Token ID39572
Feature activation-1.16e+00

Pos logit contributions
 shortcut+0.036
lot+0.035
 duel+0.033
 unnecessary+0.032
 noble+0.032
Neg logit contributions
 cups-0.040
 frost-0.037
 twins-0.036
ugs-0.036
 sisters-0.035
Token?"
Token ID1701
Feature activation-6.32e-01

Pos logit contributions
 shortcut+0.039
lot+0.038
 duel+0.037
 farewell+0.037
 gall+0.036
Neg logit contributions
 cups-0.022
 frost-0.019
 twins-0.019
 sisters-0.018
el-0.018
Token The
Token ID383
Feature activation-7.22e-01

Pos logit contributions
 shortcut+0.020
lot+0.018
 duel+0.018
 unnecessary+0.018
 noble+0.017
Neg logit contributions
 cups-0.014
 frost-0.013
 twins-0.012
 sisters-0.012
ugs-0.012
Token magician
Token ID42055
Feature activation+4.98e-01

Pos logit contributions
 shortcut+0.020
lot+0.020
 unnecessary+0.019
 duel+0.018
 steep+0.018
Neg logit contributions
 cups-0.018
 twins-0.017
 frost-0.017
 sisters-0.017
 folders-0.016
Token smiled
Token ID13541
Feature activation-6.75e-01

Pos logit contributions
 cups+0.011
 frost+0.010
 twins+0.010
 sisters+0.010
el+0.009
Neg logit contributions
 shortcut-0.016
lot-0.015
 unnecessary-0.015
 duel-0.014
 further-0.014
Token and
Token ID290
Feature activation-2.13e-01

Pos logit contributions
 shortcut+0.025
 duel+0.023
 unnecessary+0.023
lot+0.023
 hike+0.022
Neg logit contributions
 cups-0.012
 twins-0.011
 frost-0.011
 sisters-0.010
el-0.010
Token said
Token ID531
Feature activation-1.82e+00

Pos logit contributions
 shortcut+0.008
lot+0.007
 duel+0.007
 unnecessary+0.007
 farewell+0.007
Neg logit contributions
 cups-0.004
 frost-0.003
 twins-0.003
 sisters-0.003
el-0.003
Token,
Token ID11
Feature activation-1.34e+00

Pos logit contributions
 shortcut+0.069
lot+0.065
 unnecessary+0.065
 duel+0.062
 hike+0.062
Neg logit contributions
 cups-0.030
 frost-0.026
 sisters-0.026
 twins-0.026
ums-0.022
Token loved
Token ID6151
Feature activation-1.34e+00

Pos logit contributions
 shortcut+0.029
lot+0.026
 duel+0.026
 unnecessary+0.025
 farewell+0.025
Neg logit contributions
 cups-0.049
 frost-0.046
 twins-0.046
 sisters-0.045
el-0.045
Token to
Token ID284
Feature activation-1.78e+00

Pos logit contributions
 shortcut+0.036
lot+0.035
 farewell+0.033
 duel+0.033
 unnecessary+0.032
Neg logit contributions
 cups-0.035
 twins-0.032
 sisters-0.032
 frost-0.031
 mug-0.030
Token play
Token ID711
Feature activation-1.93e+00

Pos logit contributions
 shortcut+0.043
lot+0.043
 unnecessary+0.040
 duel+0.039
 further+0.038
Neg logit contributions
 cups-0.050
 frost-0.046
 sisters-0.046
 twins-0.045
 mug-0.044
Token his
Token ID465
Feature activation-1.02e+00

Pos logit contributions
 shortcut+0.061
lot+0.061
 duel+0.058
 unnecessary+0.057
 farewell+0.057
Neg logit contributions
 cups-0.041
 frost-0.036
 twins-0.036
 sisters-0.035
 mug-0.034
Token drum
Token ID13026
Feature activation-1.32e+00

Pos logit contributions
lot+0.033
 shortcut+0.032
 farewell+0.031
 unnecessary+0.031
 duel+0.031
Neg logit contributions
 cups-0.020
 twins-0.019
 sisters-0.018
 mug-0.018
 frost-0.017
Token and
Token ID290
Feature activation-3.15e+00

Pos logit contributions
 shortcut+0.042
lot+0.041
 unnecessary+0.040
 farewell+0.039
 duel+0.038
Neg logit contributions
 cups-0.028
 sisters-0.025
 frost-0.024
 twins-0.024
 folders-0.023
Token make
Token ID787
Feature activation-2.48e+00

Pos logit contributions
lot+0.085
 shortcut+0.084
 unnecessary+0.079
 farewell+0.078
 duel+0.077
Neg logit contributions
 cups-0.080
 sisters-0.074
 twins-0.072
 frost-0.071
 paints-0.069
Token lots
Token ID6041
Feature activation-3.41e+00

Pos logit contributions
 shortcut+0.070
lot+0.067
 duel+0.063
 farewell+0.062
 noble+0.061
Neg logit contributions
 cups-0.062
 twins-0.060
 sisters-0.058
 frost-0.057
el-0.053
Token of
Token ID286
Feature activation-2.56e+00

Pos logit contributions
 shortcut+0.139
lot+0.139
 duel+0.134
 farewell+0.129
 unnecessary+0.129
Neg logit contributions
 cups-0.043
 frost-0.035
 twins-0.035
 sisters-0.033
ugs-0.031
Token noise
Token ID7838
Feature activation-1.09e+00

Pos logit contributions
lot+0.075
 shortcut+0.073
 duel+0.070
 farewell+0.067
 noble+0.065
Neg logit contributions
 cups-0.063
 twins-0.060
 sisters-0.058
 frost-0.057
 folders-0.053
Token!
Token ID0
Feature activation-1.94e+00

Pos logit contributions
 shortcut+0.037
lot+0.036
 duel+0.035
 farewell+0.034
 noble+0.034
Neg logit contributions
 cups-0.021
 frost-0.019
 twins-0.019
 sisters-0.018
ugs-0.017
Token "
Token ID366
Feature activation-2.37e+00

Pos logit contributions
 shortcut+0.012
lot+0.012
 duel+0.011
 farewell+0.011
 Skip+0.010
Neg logit contributions
 cups-0.015
 frost-0.015
el-0.014
els-0.014
ugs-0.014
TokenI
Token ID40
Feature activation-1.10e-01

Pos logit contributions
 shortcut+0.073
lot+0.071
 duel+0.068
 unnecessary+0.066
 farewell+0.066
Neg logit contributions
 cups-0.052
 frost-0.048
 twins-0.045
ugs-0.044
 sisters-0.044
Token'll
Token ID1183
Feature activation-2.50e-01

Pos logit contributions
 shortcut+0.003
lot+0.002
 duel+0.002
 unnecessary+0.002
 farewell+0.002
Neg logit contributions
 cups-0.003
 frost-0.003
 twins-0.003
 sisters-0.003
cloth-0.003
Token make
Token ID787
Feature activation-8.84e-01

Pos logit contributions
 shortcut+0.006
 duel+0.006
lot+0.005
 gall+0.005
 defeat+0.005
Neg logit contributions
 cups-0.007
ugs-0.007
ays-0.007
 frost-0.007
 folders-0.007
Token the
Token ID262
Feature activation-1.62e+00

Pos logit contributions
 shortcut+0.024
lot+0.023
 duel+0.022
 unnecessary+0.021
 farewell+0.021
Neg logit contributions
 cups-0.023
 frost-0.021
 twins-0.020
ugs-0.020
 sisters-0.020
Token tail
Token ID7894
Feature activation-3.37e+00

Pos logit contributions
 shortcut+0.046
lot+0.045
 unnecessary+0.043
 duel+0.042
Hop+0.042
Neg logit contributions
 cups-0.042
 twins-0.039
 frost-0.039
 sisters-0.036
 mug-0.035
Token,
Token ID11
Feature activation-9.00e-01

Pos logit contributions
 shortcut+0.084
lot+0.083
 duel+0.077
 unnecessary+0.077
 farewell+0.075
Neg logit contributions
 cups-0.095
 frost-0.086
 twins-0.086
 sisters-0.084
cloth-0.081
Token you
Token ID345
Feature activation-1.60e+00

Pos logit contributions
 shortcut+0.024
lot+0.023
 duel+0.022
 unnecessary+0.021
 farewell+0.021
Neg logit contributions
 cups-0.024
 frost-0.022
 twins-0.022
 sisters-0.021
cloth-0.021
Token make
Token ID787
Feature activation-1.17e+00

Pos logit contributions
 shortcut+0.045
lot+0.045
 unnecessary+0.042
 farewell+0.041
 duel+0.041
Neg logit contributions
 cups-0.039
 frost-0.037
 twins-0.034
 sisters-0.033
 icing-0.032
Token the
Token ID262
Feature activation-1.89e+00

Pos logit contributions
 shortcut+0.035
lot+0.034
 duel+0.032
 noble+0.031
 farewell+0.031
Neg logit contributions
 cups-0.028
 frost-0.026
 twins-0.025
 sisters-0.024
el-0.024
Token fin
Token ID957
Feature activation-3.87e+00

Pos logit contributions
 shortcut+0.054
lot+0.052
 unnecessary+0.049
 duel+0.048
Hop+0.048
Neg logit contributions
 cups-0.049
 twins-0.046
 frost-0.046
 sisters-0.043
 mug-0.041
Token.
Token ID13
Feature activation-1.71e+00

Pos logit contributions
 cups+0.010
 frost+0.008
 twins+0.008
ugs+0.008
cloth+0.008
Neg logit contributions
 shortcut-0.017
lot-0.017
 unnecessary-0.017
 duel-0.017
 noble-0.016
Token He
Token ID679
Feature activation-7.88e-01

Pos logit contributions
 shortcut+0.053
 unnecessary+0.051
 duel+0.050
lot+0.049
 gall+0.048
Neg logit contributions
 cups-0.041
 twins-0.035
 frost-0.034
 tub-0.033
 sisters-0.033
Token could
Token ID714
Feature activation-1.18e+00

Pos logit contributions
 shortcut+0.024
 unnecessary+0.023
 duel+0.023
lot+0.022
 ashore+0.020
Neg logit contributions
 cups-0.019
 frost-0.017
 sisters-0.016
cloth-0.016
 twins-0.015
Token hear
Token ID3285
Feature activation-2.86e+00

Pos logit contributions
 shortcut+0.029
lot+0.028
 duel+0.027
 unnecessary+0.027
 noble+0.025
Neg logit contributions
 cups-0.033
 frost-0.031
 sisters-0.030
 twins-0.030
cloth-0.029
Token the
Token ID262
Feature activation-1.45e+00

Pos logit contributions
lot+0.108
 shortcut+0.108
 duel+0.105
 rematch+0.095
 noble+0.094
Neg logit contributions
 cups-0.052
 frost-0.051
 twins-0.044
 candles-0.043
 icing-0.039
Token birds
Token ID10087
Feature activation-3.05e+00

Pos logit contributions
lot+0.042
 duel+0.039
 shortcut+0.039
 unnecessary+0.036
 noble+0.035
Neg logit contributions
 cups-0.039
 twins-0.035
 frost-0.035
 sisters-0.033
 mug-0.033
Token ch
Token ID442
Feature activation-3.14e-01

Pos logit contributions
 shortcut+0.087
lot+0.086
 duel+0.081
 unnecessary+0.080
 noble+0.079
Neg logit contributions
 cups-0.077
 frost-0.070
el-0.069
 sisters-0.069
 twins-0.068
Tokenir
Token ID343
Feature activation-9.90e-01

Pos logit contributions
 shortcut+0.010
lot+0.010
 unnecessary+0.009
 farewell+0.009
 further+0.009
Neg logit contributions
 cups-0.006
 frost-0.006
 twins-0.006
 sisters-0.006
 folders-0.005
Tokenping
Token ID13886
Feature activation-1.65e+00

Pos logit contributions
 shortcut+0.028
 duel+0.026
lot+0.025
 noble+0.025
 unnecessary+0.024
Neg logit contributions
 cups-0.025
el-0.024
 frost-0.024
ugs-0.023
 twins-0.023
Token from
Token ID422
Feature activation-7.13e-01

Pos logit contributions
 shortcut+0.061
lot+0.061
 duel+0.060
 unnecessary+0.058
 noble+0.058
Neg logit contributions
 cups-0.028
 frost-0.025
 twins-0.023
el-0.023
 sisters-0.022
Token the
Token ID262
Feature activation+1.41e-01

Pos logit contributions
lot+0.023
 shortcut+0.022
 duel+0.022
 noble+0.020
 unnecessary+0.020
Neg logit contributions
 cups-0.016
 frost-0.015
 twins-0.014
el-0.013
 mug-0.013
Token lion
Token ID18744
Feature activation-9.98e-01

Pos logit contributions
 shortcut+0.020
lot+0.019
 duel+0.019
 unnecessary+0.019
 noble+0.018
Neg logit contributions
 cups-0.015
 frost-0.014
 twins-0.014
 sisters-0.013
 mug-0.013
Token.
Token ID13
Feature activation-1.47e+00

Pos logit contributions
 shortcut+0.033
lot+0.032
 duel+0.031
 unnecessary+0.031
 farewell+0.030
Neg logit contributions
 cups-0.020
 frost-0.018
 twins-0.017
 sisters-0.016
ugs-0.016
Token He
Token ID679
Feature activation-2.35e+00

Pos logit contributions
 shortcut+0.042
lot+0.039
 unnecessary+0.039
 duel+0.038
 further+0.038
Neg logit contributions
 cups-0.036
 frost-0.034
 twins-0.033
 sisters-0.031
ugs-0.031
Token liked
Token ID8288
Feature activation-2.37e+00

Pos logit contributions
 shortcut+0.084
 unnecessary+0.083
 further+0.080
 nearer+0.080
 duel+0.077
Neg logit contributions
 cups-0.039
 twins-0.038
 frost-0.038
 sisters-0.033
el-0.032
Token to
Token ID284
Feature activation-2.36e+00

Pos logit contributions
rated+0.076
 shortcut+0.075
lot+0.072
 farewell+0.071
 nearer+0.070
Neg logit contributions
 cups-0.049
 twins-0.047
 frost-0.044
 sisters-0.044
 eggs-0.040
Token wander
Token ID27776
Feature activation-3.01e+00

Pos logit contributions
 shortcut+0.048
lot+0.047
 unnecessary+0.044
 duel+0.042
 further+0.042
Neg logit contributions
 cups-0.076
 frost-0.072
 sisters-0.069
 twins-0.069
el-0.067
Token around
Token ID1088
Feature activation-1.66e+00

Pos logit contributions
lot+0.088
 duel+0.088
 shortcut+0.088
 unnecessary+0.085
'd+0.084
Neg logit contributions
 cups-0.066
 twins-0.057
 frost-0.054
 folders-0.053
 ink-0.052
Token his
Token ID465
Feature activation-8.25e-01

Pos logit contributions
 shortcut+0.059
lot+0.059
 unnecessary+0.058
 farewell+0.056
 duel+0.055
Neg logit contributions
 cups-0.028
 twins-0.025
el-0.025
 frost-0.023
 sisters-0.022
Token jungle
Token ID20712
Feature activation-3.46e-01

Pos logit contributions
 shortcut+0.024
 nearer+0.023
 unnecessary+0.022
 farewell+0.022
 duel+0.022
Neg logit contributions
 cups-0.020
 twins-0.019
el-0.018
 frost-0.018
 sisters-0.017
Token looking
Token ID2045
Feature activation-3.18e+00

Pos logit contributions
 shortcut+0.012
 unnecessary+0.012
lot+0.011
 farewell+0.011
 duel+0.011
Neg logit contributions
 cups-0.007
 twins-0.006
 sisters-0.005
 frost-0.005
 folders-0.005
Token for
Token ID329
Feature activation-1.67e+00

Pos logit contributions
lot+0.128
 duel+0.126
 shortcut+0.126
chery+0.118
Hop+0.117
Neg logit contributions
 cups-0.049
 twins-0.043
 frost-0.035
 sisters-0.033
overs-0.031
Token happy
Token ID3772
Feature activation+5.94e-01

Pos logit contributions
 cups+0.008
 twins+0.007
 frost+0.007
 sisters+0.007
ugs+0.007
Neg logit contributions
 shortcut-0.008
lot-0.007
 duel-0.007
 unnecessary-0.007
 hike-0.007
Token!
Token ID0
Feature activation+1.08e+00

Pos logit contributions
 cups+0.015
 frost+0.013
 twins+0.013
 sisters+0.013
ugs+0.013
Neg logit contributions
 shortcut-0.017
lot-0.016
 duel-0.016
 unnecessary-0.015
 noble-0.015
Token She
Token ID1375
Feature activation-1.26e-02

Pos logit contributions
 cups+0.029
 frost+0.026
 sisters+0.025
els+0.024
ugs+0.024
Neg logit contributions
lot-0.028
 shortcut-0.028
 farewell-0.026
 duel-0.026
 gall-0.025
Token thanked
Token ID26280
Feature activation+5.44e-01

Pos logit contributions
 shortcut+0.000
 unnecessary+0.000
lot+0.000
 duel+0.000
 further+0.000
Neg logit contributions
 cups-0.000
 frost-0.000
 sisters-0.000
 twins-0.000
el-0.000
Token the
Token ID262
Feature activation-8.77e-01

Pos logit contributions
 cups+0.012
 frost+0.011
 twins+0.010
 sisters+0.010
ugs+0.010
Neg logit contributions
 shortcut-0.017
lot-0.017
 duel-0.016
 unnecessary-0.015
 noble-0.015
Token bears
Token ID13062
Feature activation-3.23e+00

Pos logit contributions
 shortcut+0.027
lot+0.026
 duel+0.025
 unnecessary+0.024
 farewell+0.024
Neg logit contributions
 cups-0.020
 frost-0.018
 twins-0.017
 sisters-0.017
ugs-0.017
Token for
Token ID329
Feature activation+6.85e-01

Pos logit contributions
 shortcut+0.112
 unnecessary+0.109
lot+0.106
 duel+0.106
Left+0.098
Neg logit contributions
 cups-0.063
 sisters-0.057
 twins-0.057
 frost-0.053
 mug-0.049
Token their
Token ID511
Feature activation-1.44e-01

Pos logit contributions
 cups+0.016
 frost+0.014
 twins+0.014
 sisters+0.014
ugs+0.014
Neg logit contributions
 shortcut-0.020
lot-0.020
 duel-0.019
 unnecessary-0.018
 farewell-0.018
Token help
Token ID1037
Feature activation+6.37e-01

Pos logit contributions
lot+0.004
 shortcut+0.004
 higher+0.003
Hop+0.003
 nearer+0.003
Neg logit contributions
 sisters-0.004
 twins-0.004
 cups-0.004
 folders-0.003
 frost-0.003
Token and
Token ID290
Feature activation-2.47e-01

Pos logit contributions
 cups+0.013
 frost+0.012
 twins+0.012
 sisters+0.012
ugs+0.011
Neg logit contributions
 shortcut-0.020
lot-0.020
 duel-0.019
 unnecessary-0.018
 noble-0.018
Token they
Token ID484
Feature activation+2.98e-01

Pos logit contributions
 shortcut+0.007
lot+0.007
 duel+0.007
 unnecessary+0.006
 farewell+0.006
Neg logit contributions
 cups-0.006
 frost-0.006
 twins-0.005
ugs-0.005
 sisters-0.005