TOP ACTIVATIONS
MAX = 2.394

, and Jill knew the elderly neighbour was right . The
having fun together . The elderly rh in oc eros was
After the haircut , Lily looked in the
. After the haircut was finished , the three
." After the haircut , Lily looked in the
. When the elderly lady got home , she
on his face . The elderly woman thanked him and gave
closed . The elderly lady was so grateful and
her !" The elderly woman win ced in pain
The animals all thanked the elderly rh in oc eros for
help . Even though the elderly couple were feeling tired ,
. The elderly lady laughed and said ,
After the haircut was done , Tim my
goodbye . The elderly woman was so happy to
day ?" The elderly woman laughed and said ,
. The elderly person said , " You
and looked back . The elderly woman was not angry .
okay . The elderly man smiled weak ly and
feel excited . When the haircut was done , Little one
home . The elderly lady was very happy to

MIN ACTIVATIONS
MIN = -2.395

She takes out a magn ifying glass . She looks at
a flashlight and a magn ifying glass to examine their finds
He took out a magn ifying glass and a notebook and
the necklace and the magn ifying glass and started to examine
the necklace and the magn ifying glass and told him that
She took out a magn ifying glass and looked very carefully
flashlight , and a magn ifying glass . As
When Daisy put the magn ifying glass down , the cater
learned even more about how volcan oes work . She thought
job cleaning the garage ." Little Jimmy smiled . He had
. They wear masks and fins to see the fish and
said , holding the magn ifying glass over a black ant
examine it with the magn ifying glass ?" "
big book with pictures of volcan oes . Vol can oes
a furry , brown bear cub . One day
fox living in a big chim ney . Every day ,
big book with pictures of volcan oes . Vol can oes
bear who had a little cub . One day , little
a house with a big chim ney . The chim ney
. They had a magn ifying glass and a small box

INTERVAL 1.196 - 2.394
CONTAINS 0.151%

He found a big tree with an old branch reaching down
â € The old lady was astonished at Daisy
mom holding a big file with many papers .
. After a few hours , the witch stirred
there was a city filled with joy . Everyone was happy

INTERVAL -0.001 - 1.196
CONTAINS 49.146%

her paper and stamped them with the heart . She wrote
He loved to go for walks . One day , Tim
After playing for a while , Tim started to feel
. The end . <|endoftext|> Once upon a time there were
away . One day it stored a toy plane , a

INTERVAL -1.198 - -0.001
CONTAINS 50.580%

small house with his best friend , a little boy named
. They mixed them with a spoon . They said ,
Sarah , " If you can guess what 's inside the
time , there was a little boy named Tim my .
a great job , Lily . I 'm happy you 're

INTERVAL -2.395 - -1.198
CONTAINS 0.123%

it ! We have the best tower !" They see Lily
always be kind and think of others before we think of
appetite and lived in a swamp in the middle of the
his best to get rid of the big spot . Lucy
colorful cand ies and ch ocol ates . She said to
Token,
Token ID11
Feature activation-3.95e-02

Pos logit contributions
 couple+0.014
rob+0.012
 seek+0.011
 lady+0.010
thank+0.010
Neg logit contributions
aying-0.025
 fears-0.024
arse-0.024
 Ow-0.023
uning-0.023
Token and
Token ID290
Feature activation+1.83e-01

Pos logit contributions
aying+0.002
arse+0.002
 fears+0.002
uning+0.002
 Ow+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 lady-0.001
 gentleman-0.001
Token Jill
Token ID22772
Feature activation+4.55e-01

Pos logit contributions
rob+0.008
 couple+0.007
 seek+0.006
 gentleman+0.006
 trainer+0.006
Neg logit contributions
aying-0.009
 fears-0.009
arse-0.008
omed-0.008
uning-0.008
Token knew
Token ID2993
Feature activation+3.11e-01

Pos logit contributions
 couple+0.018
rob+0.017
 seek+0.015
 gentleman+0.015
 trainer+0.014
Neg logit contributions
aying-0.024
 fears-0.022
arse-0.022
uning-0.022
omed-0.022
Token the
Token ID262
Feature activation+2.74e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.009
 lady+0.009
 gentleman+0.008
Neg logit contributions
aying-0.018
arse-0.018
 fears-0.018
 Ow-0.017
uning-0.017
Token elderly
Token ID13830
Feature activation+2.39e+00

Pos logit contributions
rob+0.010
 couple+0.010
 seek+0.009
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.015
 fears-0.014
uning-0.014
arse-0.014
 Ow-0.013
Token neighbour
Token ID12250
Feature activation+1.01e+00

Pos logit contributions
rob+0.074
robat+0.053
 seek+0.052
Ar+0.050
seek+0.044
Neg logit contributions
aying-0.140
uning-0.136
 wipes-0.134
 fears-0.128
icking-0.127
Token was
Token ID373
Feature activation+1.01e-01

Pos logit contributions
 couple+0.037
rob+0.035
 seek+0.032
 gentleman+0.031
 lady+0.031
Neg logit contributions
aying-0.056
arse-0.054
 fears-0.053
 Ow-0.052
uning-0.052
Token right
Token ID826
Feature activation+2.37e-01

Pos logit contributions
rob+0.004
 couple+0.004
 seek+0.003
robat+0.003
 gentleman+0.003
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.005
uning-0.005
omed-0.005
Token.
Token ID13
Feature activation+5.01e-02

Pos logit contributions
rob+0.009
 couple+0.009
 seek+0.008
 gentleman+0.007
 trainer+0.007
Neg logit contributions
aying-0.013
 fears-0.012
arse-0.012
omed-0.012
uning-0.012
Token The
Token ID383
Feature activation+2.98e-01

Pos logit contributions
rob+0.002
 couple+0.002
 seek+0.002
 trainer+0.002
 gentleman+0.002
Neg logit contributions
aying-0.002
 fears-0.002
uning-0.002
arse-0.002
omed-0.002
Token having
Token ID1719
Feature activation-1.00e+00

Pos logit contributions
aying+0.040
arse+0.039
 fears+0.038
 Ow+0.038
omed+0.037
Neg logit contributions
 couple-0.025
rob-0.022
 seek-0.022
thank-0.020
 gentleman-0.020
Token fun
Token ID1257
Feature activation+4.68e-01

Pos logit contributions
aying+0.052
 fears+0.049
arse+0.049
uning+0.048
omed+0.047
Neg logit contributions
 couple-0.040
rob-0.039
 seek-0.035
 gentleman-0.033
 trainer-0.033
Token together
Token ID1978
Feature activation-1.32e-01

Pos logit contributions
 couple+0.018
rob+0.018
 gentleman+0.016
 trainer+0.015
 seek+0.015
Neg logit contributions
aying-0.024
 fears-0.022
omed-0.022
uning-0.022
adoes-0.022
Token.
Token ID13
Feature activation+1.27e-01

Pos logit contributions
aying+0.007
 fears+0.007
arse+0.007
uning+0.007
omed+0.007
Neg logit contributions
rob-0.005
 couple-0.005
 seek-0.004
 gentleman-0.004
 trainer-0.004
Token The
Token ID383
Feature activation+4.35e-01

Pos logit contributions
 couple+0.006
rob+0.005
 seek+0.005
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.006
arse-0.006
 fears-0.006
uning-0.006
omed-0.006
Token elderly
Token ID13830
Feature activation+2.38e+00

Pos logit contributions
 couple+0.013
rob+0.012
 seek+0.010
 gentleman+0.010
 trainer+0.010
Neg logit contributions
aying-0.028
 fears-0.026
arse-0.026
uning-0.026
omed-0.026
Token rh
Token ID9529
Feature activation+4.35e-01

Pos logit contributions
rob+0.074
 couple+0.063
 seek+0.063
robat+0.060
seek+0.058
Neg logit contributions
aying-0.122
adoes-0.117
 fears-0.111
uning-0.110
 wipes-0.109
Tokenin
Token ID259
Feature activation+5.23e-01

Pos logit contributions
 couple+0.019
rob+0.019
 seek+0.017
 gentleman+0.017
 trainer+0.016
Neg logit contributions
aying-0.020
 fears-0.019
arse-0.019
uning-0.018
omed-0.018
Tokenoc
Token ID420
Feature activation+2.60e-01

Pos logit contributions
rob+0.012
 couple+0.012
 gentleman+0.010
 seek+0.010
 businessman+0.009
Neg logit contributions
aying-0.034
 fears-0.033
ist-0.031
omed-0.031
 cleans-0.031
Tokeneros
Token ID27498
Feature activation+1.19e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.009
 gentleman+0.008
 lady+0.008
Neg logit contributions
aying-0.014
uning-0.013
 fears-0.013
adoes-0.013
arse-0.013
Token was
Token ID373
Feature activation-1.04e-01

Pos logit contributions
 couple+0.005
rob+0.004
 seek+0.004
 gentleman+0.004
 trainer+0.004
Neg logit contributions
aying-0.006
 fears-0.006
arse-0.006
uning-0.006
omed-0.006
Token
Token ID220
Feature activation+2.30e-02

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.001
 trainer+0.001
 gentleman+0.001
Neg logit contributions
aying-0.002
 fears-0.002
arse-0.001
omed-0.001
 wipes-0.001
Token
Token ID198
Feature activation-6.60e-01

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 gentleman+0.001
 lady+0.001
Neg logit contributions
aying-0.001
 fears-0.001
arse-0.001
uning-0.001
omed-0.001
Token
Token ID198
Feature activation-2.56e-01

Pos logit contributions
aying+0.032
 Ow+0.030
 wipes+0.030
omed+0.029
arse+0.029
Neg logit contributions
 couple-0.028
rob-0.028
 seek-0.024
 gentleman-0.024
 trainer-0.024
TokenAfter
Token ID3260
Feature activation+9.23e-01

Pos logit contributions
aying+0.014
 fears+0.013
arse+0.013
 wipes+0.013
omed+0.012
Neg logit contributions
rob-0.010
 couple-0.010
 gentleman-0.008
 trainer-0.008
robat-0.008
Token the
Token ID262
Feature activation+4.64e-01

Pos logit contributions
 couple+0.034
rob+0.033
 seek+0.029
 gentleman+0.028
 trainer+0.028
Neg logit contributions
aying-0.050
 fears-0.047
arse-0.047
uning-0.047
omed-0.046
Token haircut
Token ID45548
Feature activation+2.31e+00

Pos logit contributions
rob+0.021
 couple+0.019
 seek+0.018
 gentleman+0.018
 trainer+0.017
Neg logit contributions
aying-0.021
 cleans-0.020
 fears-0.020
arse-0.020
 wipes-0.020
Token,
Token ID11
Feature activation+7.81e-02

Pos logit contributions
 couple+0.088
rob+0.085
 seek+0.076
 gentleman+0.073
 trainer+0.072
Neg logit contributions
aying-0.123
 fears-0.116
arse-0.115
uning-0.113
omed-0.112
Token Lily
Token ID20037
Feature activation+5.03e-01

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.003
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.004
 fears-0.004
arse-0.004
uning-0.004
omed-0.004
Token looked
Token ID3114
Feature activation+2.28e-01

Pos logit contributions
 couple+0.020
rob+0.020
 seek+0.018
 gentleman+0.017
 trainer+0.017
Neg logit contributions
aying-0.026
 fears-0.024
arse-0.024
uning-0.023
omed-0.023
Token in
Token ID287
Feature activation-3.71e-01

Pos logit contributions
 couple+0.008
rob+0.007
 lady+0.007
 gentleman+0.007
 man+0.007
Neg logit contributions
aying-0.013
arse-0.012
 Ow-0.012
uning-0.012
adoes-0.011
Token the
Token ID262
Feature activation-8.54e-01

Pos logit contributions
aying+0.019
 fears+0.018
 wipes+0.018
 Ow+0.018
arse+0.018
Neg logit contributions
rob-0.015
 couple-0.014
 seek-0.013
 trainer-0.012
robat-0.012
Token.
Token ID13
Feature activation+2.49e-01

Pos logit contributions
aying+0.020
 fears+0.019
arse+0.018
uning+0.018
omed+0.018
Neg logit contributions
 couple-0.014
rob-0.014
 seek-0.012
 gentleman-0.012
 trainer-0.011
Token
Token ID198
Feature activation-2.14e-03

Pos logit contributions
 couple+0.010
rob+0.009
 gentleman+0.009
 lady+0.009
 man+0.008
Neg logit contributions
aying-0.013
arse-0.012
 fears-0.012
uning-0.012
omed-0.012
Token
Token ID198
Feature activation-2.31e-02

Pos logit contributions
aying+0.000
 fears+0.000
arse+0.000
uning+0.000
omed+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 seek-0.000
 gentleman-0.000
 lady-0.000
TokenAfter
Token ID3260
Feature activation+9.41e-01

Pos logit contributions
aying+0.001
 fears+0.001
arse+0.001
uning+0.001
omed+0.001
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 gentleman-0.001
 trainer-0.001
Token the
Token ID262
Feature activation+5.35e-01

Pos logit contributions
 couple+0.034
rob+0.032
 seek+0.029
 gentleman+0.027
 lady+0.027
Neg logit contributions
aying-0.054
 fears-0.051
arse-0.050
uning-0.050
omed-0.049
Token haircut
Token ID45548
Feature activation+2.30e+00

Pos logit contributions
 couple+0.023
rob+0.022
 seek+0.020
 gentleman+0.019
 trainer+0.019
Neg logit contributions
aying-0.026
 fears-0.025
arse-0.024
uning-0.024
omed-0.024
Token was
Token ID373
Feature activation+5.48e-01

Pos logit contributions
 couple+0.087
rob+0.083
 seek+0.074
 gentleman+0.071
 lady+0.070
Neg logit contributions
aying-0.124
 fears-0.117
arse-0.117
uning-0.115
 Ow-0.113
Token finished
Token ID5201
Feature activation+1.27e+00

Pos logit contributions
 couple+0.020
 seek+0.017
thank+0.017
rob+0.017
 lady+0.016
Neg logit contributions
aying-0.030
 fears-0.030
arse-0.029
 Ow-0.029
uning-0.029
Token,
Token ID11
Feature activation+1.95e-01

Pos logit contributions
 couple+0.038
rob+0.032
 seek+0.031
 man+0.029
 lady+0.029
Neg logit contributions
aying-0.081
 fears-0.078
 Ow-0.076
arse-0.076
uning-0.075
Token the
Token ID262
Feature activation-2.35e-01

Pos logit contributions
 couple+0.006
rob+0.005
 lady+0.005
 seek+0.005
 gentleman+0.005
Neg logit contributions
aying-0.012
 fears-0.011
arse-0.011
uning-0.011
omed-0.011
Token three
Token ID1115
Feature activation+7.18e-01

Pos logit contributions
aying+0.010
 fears+0.010
 fists+0.009
 wipes+0.009
arse+0.009
Neg logit contributions
rob-0.012
 seek-0.010
 couple-0.010
robat-0.010
 trainer-0.009
Token."
Token ID526
Feature activation+3.23e-01

Pos logit contributions
 couple+0.016
 seek+0.015
 lady+0.014
rob+0.014
thank+0.014
Neg logit contributions
aying-0.027
arse-0.026
 fears-0.026
omed-0.025
uning-0.025
Token
Token ID198
Feature activation+6.15e-02

Pos logit contributions
 couple+0.012
rob+0.011
 lady+0.010
 seek+0.010
 gentleman+0.010
Neg logit contributions
aying-0.018
 fears-0.017
arse-0.017
omed-0.016
uning-0.016
Token
Token ID198
Feature activation+5.86e-02

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 gentleman+0.002
 trainer+0.002
Neg logit contributions
aying-0.003
 Ow-0.003
arse-0.003
 fears-0.003
omed-0.003
TokenAfter
Token ID3260
Feature activation+1.07e+00

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 gentleman+0.002
 trainer+0.002
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
uning-0.003
omed-0.003
Token the
Token ID262
Feature activation+6.75e-01

Pos logit contributions
 couple+0.040
rob+0.039
 seek+0.035
 gentleman+0.033
 trainer+0.033
Neg logit contributions
aying-0.058
 fears-0.054
arse-0.054
uning-0.053
omed-0.053
Token haircut
Token ID45548
Feature activation+2.27e+00

Pos logit contributions
rob+0.030
 couple+0.028
 seek+0.026
 gentleman+0.025
 trainer+0.024
Neg logit contributions
aying-0.031
 fears-0.030
arse-0.029
uning-0.029
 wipes-0.029
Token,
Token ID11
Feature activation+1.75e-01

Pos logit contributions
 couple+0.090
rob+0.088
 seek+0.079
 gentleman+0.076
 trainer+0.075
Neg logit contributions
aying-0.117
 fears-0.109
uning-0.108
arse-0.107
omed-0.106
Token Lily
Token ID20037
Feature activation+4.87e-01

Pos logit contributions
 couple+0.007
rob+0.007
 seek+0.006
 gentleman+0.006
 trainer+0.006
Neg logit contributions
aying-0.009
 fears-0.008
arse-0.008
uning-0.008
omed-0.008
Token looked
Token ID3114
Feature activation+2.21e-01

Pos logit contributions
rob+0.021
 couple+0.020
 seek+0.018
 gentleman+0.017
 trainer+0.017
Neg logit contributions
aying-0.024
 fears-0.022
 cleans-0.022
arse-0.022
 wipes-0.022
Token in
Token ID287
Feature activation-3.16e-01

Pos logit contributions
 couple+0.008
rob+0.007
 lady+0.007
 gentleman+0.007
 man+0.007
Neg logit contributions
aying-0.012
arse-0.011
 Ow-0.011
uning-0.011
adoes-0.011
Token the
Token ID262
Feature activation-4.26e-01

Pos logit contributions
 fears+0.015
aying+0.015
 wipes+0.015
 fists+0.014
 Ow+0.014
Neg logit contributions
rob-0.014
 couple-0.013
 seek-0.012
robat-0.012
 trainer-0.011
Token.
Token ID13
Feature activation+1.22e-01

Pos logit contributions
aying+0.008
omed+0.007
 fears+0.007
uning+0.007
 wipes+0.007
Neg logit contributions
rob-0.011
 couple-0.010
Offic-0.009
 trainer-0.009
 businessman-0.009
Token
Token ID198
Feature activation+3.73e-02

Pos logit contributions
 couple+0.004
rob+0.004
 seek+0.004
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.007
 fears-0.007
arse-0.007
uning-0.007
omed-0.006
Token
Token ID198
Feature activation+1.05e-01

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 gentleman+0.001
 trainer+0.001
Neg logit contributions
aying-0.002
 fears-0.002
 Ow-0.002
arse-0.002
omed-0.002
TokenWhen
Token ID2215
Feature activation+2.49e-01

Pos logit contributions
rob+0.005
 couple+0.005
 gentleman+0.004
 seek+0.004
 trainer+0.004
Neg logit contributions
aying-0.005
arse-0.004
 fears-0.004
omed-0.004
 Ow-0.004
Token the
Token ID262
Feature activation+5.44e-01

Pos logit contributions
rob+0.010
 couple+0.009
 seek+0.009
 trainer+0.009
 gentleman+0.009
Neg logit contributions
aying-0.012
 Ow-0.011
omed-0.011
arse-0.011
 fears-0.011
Token elderly
Token ID13830
Feature activation+2.26e+00

Pos logit contributions
rob+0.020
robat+0.015
 seek+0.013
 couple+0.012
thank+0.011
Neg logit contributions
aying-0.032
 fears-0.031
uning-0.030
omed-0.030
 cleans-0.030
Token lady
Token ID10846
Feature activation+1.90e+00

Pos logit contributions
rob+0.086
robat+0.068
Ar+0.056
 seek+0.055
 yak+0.054
Neg logit contributions
aying-0.117
 cleans-0.114
 wipes-0.112
uning-0.111
omed-0.107
Token got
Token ID1392
Feature activation+4.72e-01

Pos logit contributions
rob+0.077
 couple+0.076
 seek+0.067
 gentleman+0.064
thank+0.064
Neg logit contributions
aying-0.096
omed-0.090
 fears-0.089
uning-0.088
arse-0.088
Token home
Token ID1363
Feature activation+1.05e+00

Pos logit contributions
rob+0.019
 couple+0.017
robat+0.015
 seek+0.015
 trainer+0.015
Neg logit contributions
aying-0.025
 fears-0.023
uning-0.023
omed-0.023
arse-0.023
Token,
Token ID11
Feature activation+7.93e-02

Pos logit contributions
rob+0.049
 couple+0.045
robat+0.041
 trainer+0.040
 seek+0.040
Neg logit contributions
aying-0.049
arse-0.045
 fears-0.045
uning-0.045
omed-0.045
Token she
Token ID673
Feature activation+1.62e-01

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 gentleman+0.002
 trainer+0.002
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.005
uning-0.005
omed-0.005
Token on
Token ID319
Feature activation-2.68e-01

Pos logit contributions
rob+0.021
 couple+0.021
 seek+0.019
 gentleman+0.018
 trainer+0.017
Neg logit contributions
aying-0.029
arse-0.027
 fears-0.027
uning-0.026
omed-0.026
Token his
Token ID465
Feature activation-3.08e-01

Pos logit contributions
aying+0.018
 fears+0.017
arse+0.017
uning+0.017
 Ow+0.017
Neg logit contributions
 couple-0.007
rob-0.006
 seek-0.005
 gentleman-0.005
 trainer-0.005
Token face
Token ID1986
Feature activation+2.07e-01

Pos logit contributions
aying+0.012
arse+0.011
uning+0.010
 fears+0.010
omed+0.010
Neg logit contributions
 couple-0.017
rob-0.016
 seek-0.015
 gentleman-0.015
 trainer-0.015
Token.
Token ID13
Feature activation+9.23e-02

Pos logit contributions
 couple+0.007
rob+0.007
 seek+0.006
 gentleman+0.006
 lady+0.005
Neg logit contributions
aying-0.012
 fears-0.011
arse-0.011
uning-0.011
 Ow-0.011
Token The
Token ID383
Feature activation+9.51e-02

Pos logit contributions
 couple+0.003
rob+0.003
 gentleman+0.003
 seek+0.003
 lady+0.002
Neg logit contributions
aying-0.005
arse-0.005
 fears-0.005
uning-0.005
adoes-0.005
Token elderly
Token ID13830
Feature activation+2.22e+00

Pos logit contributions
rob+0.003
 seek+0.002
 couple+0.002
robat+0.002
seek+0.002
Neg logit contributions
aying-0.006
 fears-0.006
 fists-0.006
 wipes-0.006
arse-0.006
Token woman
Token ID2415
Feature activation+6.63e-01

Pos logit contributions
rob+0.073
 seek+0.065
robat+0.059
 yak+0.054
Ar+0.053
Neg logit contributions
aying-0.122
 wipes-0.119
 cleans-0.117
adoes-0.112
 fears-0.112
Token thanked
Token ID26280
Feature activation-2.13e-01

Pos logit contributions
 couple+0.024
rob+0.022
 seek+0.020
 lady+0.020
 gentleman+0.019
Neg logit contributions
aying-0.038
arse-0.036
 fears-0.035
uning-0.035
 Ow-0.035
Token him
Token ID683
Feature activation-2.38e-01

Pos logit contributions
aying+0.011
 fears+0.011
arse+0.010
uning+0.010
omed+0.010
Neg logit contributions
 couple-0.009
rob-0.008
 seek-0.007
 gentleman-0.007
 lady-0.007
Token and
Token ID290
Feature activation-1.67e-01

Pos logit contributions
aying+0.015
arse+0.015
 Ow+0.015
 fears+0.014
uning+0.014
Neg logit contributions
 couple-0.007
rob-0.005
 lady-0.005
 gentleman-0.005
 seek-0.005
Token gave
Token ID2921
Feature activation-6.10e-01

Pos logit contributions
aying+0.009
 fears+0.008
 wipes+0.008
arse+0.008
 cleans+0.008
Neg logit contributions
rob-0.006
 couple-0.006
 seek-0.006
 gentleman-0.005
 trainer-0.005
Token closed
Token ID4838
Feature activation-3.60e-02

Pos logit contributions
 couple+0.004
rob+0.003
 seek+0.003
 lady+0.003
thank+0.003
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.005
uning-0.004
 Ow-0.004
Token.
Token ID13
Feature activation+1.74e-01

Pos logit contributions
aying+0.002
arse+0.002
 fears+0.002
uning+0.002
 Ow+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 lady-0.001
thank-0.001
Token
Token ID198
Feature activation+1.13e-01

Pos logit contributions
 couple+0.007
rob+0.006
 seek+0.006
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.009
 fears-0.009
arse-0.009
uning-0.009
omed-0.009
Token
Token ID198
Feature activation+1.82e-01

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.004
thank+0.004
 trainer+0.004
Neg logit contributions
aying-0.005
omed-0.005
 fears-0.005
 Ow-0.005
arse-0.005
TokenThe
Token ID464
Feature activation+2.56e-01

Pos logit contributions
rob+0.009
 couple+0.008
 king+0.008
 seek+0.008
 trainer+0.007
Neg logit contributions
aying-0.008
omed-0.007
arse-0.007
 fears-0.007
arding-0.007
Token elderly
Token ID13830
Feature activation+2.22e+00

Pos logit contributions
rob+0.011
 couple+0.008
amer+0.008
cest+0.008
 seek+0.008
Neg logit contributions
aying-0.012
 fears-0.012
omed-0.012
oken-0.012
 wipes-0.011
Token lady
Token ID10846
Feature activation+1.06e+00

Pos logit contributions
rob+0.060
Ar+0.049
robat+0.049
 seek+0.047
 yak+0.046
Neg logit contributions
 wipes-0.114
aying-0.113
 cleans-0.113
omed-0.108
adoes-0.106
Token was
Token ID373
Feature activation+6.38e-02

Pos logit contributions
 couple+0.039
rob+0.037
 seek+0.034
 gentleman+0.032
 lady+0.032
Neg logit contributions
aying-0.058
 fears-0.055
arse-0.055
uning-0.053
 Ow-0.053
Token so
Token ID523
Feature activation+3.28e-01

Pos logit contributions
rob+0.003
 couple+0.002
robat+0.002
 seek+0.002
 trainer+0.002
Neg logit contributions
aying-0.003
 fears-0.003
omed-0.003
 cleans-0.003
arse-0.003
Token grateful
Token ID14066
Feature activation-7.52e-02

Pos logit contributions
rob+0.011
 couple+0.010
 seek+0.009
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.020
 fears-0.019
arse-0.019
omed-0.018
uning-0.018
Token and
Token ID290
Feature activation-1.49e-04

Pos logit contributions
aying+0.004
 fears+0.003
omed+0.003
 wipes+0.003
uning+0.003
Neg logit contributions
rob-0.003
 couple-0.003
 seek-0.003
robat-0.003
 trainer-0.003
Token her
Token ID607
Feature activation-1.19e-01

Pos logit contributions
aying+0.002
omed+0.002
 fears+0.002
uning+0.002
arse+0.002
Neg logit contributions
 couple-0.001
 seek-0.001
rob-0.001
 lady-0.001
 man-0.001
Token!"
Token ID2474
Feature activation+5.18e-01

Pos logit contributions
omed+0.007
aying+0.007
arse+0.006
outed+0.006
ours+0.006
Neg logit contributions
 couple-0.004
 seek-0.004
 lady-0.004
 man-0.004
thank-0.003
Token
Token ID198
Feature activation+1.95e-01

Pos logit contributions
rob+0.020
 couple+0.019
 seek+0.017
 gentleman+0.016
 trainer+0.016
Neg logit contributions
aying-0.028
 fears-0.026
 Ow-0.026
uning-0.026
arse-0.025
Token
Token ID198
Feature activation+2.33e-01

Pos logit contributions
 couple+0.008
rob+0.007
 seek+0.006
 gentleman+0.006
thank+0.006
Neg logit contributions
aying-0.010
 Ow-0.010
arse-0.010
 fears-0.010
omed-0.009
TokenThe
Token ID464
Feature activation-6.63e-02

Pos logit contributions
rob+0.010
 couple+0.010
thank+0.009
 gentleman+0.008
 seek+0.008
Neg logit contributions
aying-0.011
 fears-0.010
arse-0.010
 Ow-0.010
omed-0.010
Token elderly
Token ID13830
Feature activation+2.21e+00

Pos logit contributions
aying+0.004
 fears+0.004
 fists+0.004
arse+0.004
 cleans+0.004
Neg logit contributions
rob-0.002
 seek-0.002
robat-0.002
 couple-0.002
thank-0.001
Token woman
Token ID2415
Feature activation+8.15e-01

Pos logit contributions
rob+0.060
 seek+0.054
robat+0.046
Ar+0.041
seek+0.040
Neg logit contributions
aying-0.133
 wipes-0.126
 cleans-0.125
uning-0.121
 fears-0.121
Token win
Token ID1592
Feature activation+6.60e-01

Pos logit contributions
 couple+0.030
 lady+0.026
 man+0.026
 gentleman+0.024
 seek+0.024
Neg logit contributions
aying-0.047
arse-0.046
 Ow-0.045
uning-0.044
 fears-0.043
Tokenced
Token ID771
Feature activation+6.08e-01

Pos logit contributions
 couple+0.025
rob+0.023
 seek+0.022
 lady+0.021
 gentleman+0.021
Neg logit contributions
aying-0.036
arse-0.034
 fears-0.034
 Ow-0.033
uning-0.033
Token in
Token ID287
Feature activation-2.33e-01

Pos logit contributions
 couple+0.018
rob+0.016
 lady+0.015
 seek+0.015
 man+0.015
Neg logit contributions
aying-0.037
ours-0.035
uning-0.035
 fears-0.035
arse-0.035
Token pain
Token ID2356
Feature activation+3.51e-01

Pos logit contributions
aying+0.013
 fears+0.012
arse+0.012
uning+0.012
omed+0.012
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 gentleman-0.007
 trainer-0.007
TokenThe
Token ID464
Feature activation-2.71e-02

Pos logit contributions
aying+0.009
arse+0.008
 fears+0.008
omed+0.008
 Ow+0.008
Neg logit contributions
 couple-0.009
rob-0.009
 gentleman-0.008
 seek-0.008
 lady-0.008
Token animals
Token ID4695
Feature activation-3.71e-01

Pos logit contributions
aying+0.001
 fears+0.001
arse+0.001
omed+0.001
uning+0.001
Neg logit contributions
rob-0.001
 couple-0.001
 seek-0.001
 gentleman-0.001
thank-0.001
Token all
Token ID477
Feature activation-1.32e-01

Pos logit contributions
aying+0.020
arse+0.020
 Ow+0.019
 fears+0.019
uning+0.019
Neg logit contributions
 couple-0.014
rob-0.013
 seek-0.012
 lady-0.012
 gentleman-0.011
Token thanked
Token ID26280
Feature activation+2.67e-01

Pos logit contributions
aying+0.007
arse+0.007
 fears+0.007
uning+0.007
 Ow+0.007
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.004
 gentleman-0.004
 trainer-0.004
Token the
Token ID262
Feature activation+8.72e-02

Pos logit contributions
 couple+0.009
rob+0.009
 trainer+0.007
 seek+0.007
 lady+0.007
Neg logit contributions
aying-0.016
arse-0.015
 fears-0.015
omed-0.015
ours-0.015
Token elderly
Token ID13830
Feature activation+2.17e+00

Pos logit contributions
rob+0.003
 couple+0.003
 seek+0.002
 gentleman+0.002
thank+0.002
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.005
uning-0.005
omed-0.005
Token rh
Token ID9529
Feature activation+3.80e-01

Pos logit contributions
rob+0.067
 seek+0.066
 couple+0.061
robat+0.057
seek+0.055
Neg logit contributions
aying-0.108
uning-0.101
adoes-0.101
 fears-0.099
 cleans-0.098
Tokenin
Token ID259
Feature activation+3.97e-01

Pos logit contributions
 couple+0.017
rob+0.017
 seek+0.015
 gentleman+0.015
 trainer+0.014
Neg logit contributions
aying-0.018
 fears-0.016
arse-0.016
uning-0.016
omed-0.016
Tokenoc
Token ID420
Feature activation+2.80e-01

Pos logit contributions
rob+0.010
 couple+0.009
 seek+0.009
thank+0.007
 businessman+0.007
Neg logit contributions
aying-0.026
 fears-0.024
uning-0.024
omed-0.023
arse-0.023
Tokeneros
Token ID27498
Feature activation+2.45e-01

Pos logit contributions
 couple+0.010
rob+0.010
 seek+0.009
 gentleman+0.008
 lady+0.008
Neg logit contributions
aying-0.015
uning-0.014
 fears-0.014
arse-0.014
adoes-0.014
Token for
Token ID329
Feature activation-2.85e-01

Pos logit contributions
 couple+0.007
rob+0.006
 seek+0.006
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.016
arse-0.015
 fears-0.015
 Ow-0.015
uning-0.014
Token help
Token ID1037
Feature activation+1.13e-01

Pos logit contributions
aying+0.046
omed+0.045
arse+0.044
uning+0.042
outed+0.042
Neg logit contributions
 couple-0.024
rob-0.021
 seek-0.021
 lady-0.020
 man-0.019
Token.
Token ID13
Feature activation+5.13e-02

Pos logit contributions
 couple+0.004
 seek+0.003
rob+0.003
 lady+0.003
 man+0.003
Neg logit contributions
aying-0.007
arse-0.006
omed-0.006
 fears-0.006
ours-0.006
Token Even
Token ID3412
Feature activation+5.51e-02

Pos logit contributions
 couple+0.002
 lady+0.002
 gentleman+0.002
 ranger+0.002
 man+0.002
Neg logit contributions
arse-0.003
aying-0.003
 fears-0.003
arding-0.003
iqu-0.003
Token though
Token ID996
Feature activation-5.33e-02

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 lady+0.002
 gentleman+0.002
Neg logit contributions
aying-0.003
arse-0.003
 fears-0.003
uning-0.003
 Ow-0.003
Token the
Token ID262
Feature activation+1.03e-01

Pos logit contributions
aying+0.003
arse+0.003
uning+0.003
omed+0.003
 fears+0.003
Neg logit contributions
 couple-0.002
 seek-0.001
 lady-0.001
rob-0.001
 man-0.001
Token elderly
Token ID13830
Feature activation+2.17e+00

Pos logit contributions
rob+0.004
 couple+0.003
 seek+0.003
 gentleman+0.003
robat+0.003
Neg logit contributions
aying-0.006
 fears-0.005
arse-0.005
uning-0.005
 fists-0.005
Token couple
Token ID3155
Feature activation+1.38e+00

Pos logit contributions
rob+0.047
 seek+0.044
robat+0.035
seek+0.032
Ar+0.025
Neg logit contributions
aying-0.139
 wipes-0.125
ours-0.124
 cleans-0.123
uning-0.122
Token were
Token ID547
Feature activation+3.86e-01

Pos logit contributions
 couple+0.055
rob+0.053
 seek+0.047
 gentleman+0.045
 trainer+0.045
Neg logit contributions
aying-0.072
 fears-0.068
arse-0.068
uning-0.067
omed-0.066
Token feeling
Token ID4203
Feature activation+8.35e-01

Pos logit contributions
 couple+0.013
 seek+0.011
 lady+0.010
 trainer+0.010
 man+0.010
Neg logit contributions
aying-0.023
 Ow-0.023
arse-0.023
omed-0.022
ours-0.022
Token tired
Token ID10032
Feature activation+3.43e-01

Pos logit contributions
 couple+0.027
 l+0.022
 knowing+0.022
 man+0.021
thank+0.021
Neg logit contributions
arse-0.050
uning-0.048
 Ow-0.047
ains-0.047
aying-0.047
Token,
Token ID11
Feature activation-6.93e-02

Pos logit contributions
 couple+0.011
rob+0.008
 man+0.008
 lady+0.008
 seek+0.008
Neg logit contributions
aying-0.021
arse-0.021
 fears-0.021
ours-0.020
 Ow-0.020
Token.
Token ID13
Feature activation+6.73e-02

Pos logit contributions
 couple+0.024
rob+0.021
 seek+0.021
 lady+0.020
 gentleman+0.019
Neg logit contributions
aying-0.040
 fears-0.038
arse-0.038
uning-0.037
 Ow-0.037
Token
Token ID220
Feature activation+4.47e-01

Pos logit contributions
 couple+0.003
rob+0.002
 gentleman+0.002
 lady+0.002
 seek+0.002
Neg logit contributions
aying-0.004
 fears-0.003
arse-0.003
uning-0.003
adoes-0.003
Token
Token ID198
Feature activation-2.76e-01

Pos logit contributions
 couple+0.006
rob+0.005
 lady+0.005
 gentleman+0.005
 man+0.005
Neg logit contributions
 fears-0.033
arse-0.032
aying-0.031
adoes-0.031
 fists-0.031
Token
Token ID198
Feature activation-8.34e-02

Pos logit contributions
aying+0.016
arse+0.015
 fears+0.015
omed+0.014
uning+0.014
Neg logit contributions
 couple-0.010
rob-0.009
 seek-0.008
 gentleman-0.008
 trainer-0.008
TokenThe
Token ID464
Feature activation+2.54e-02

Pos logit contributions
aying+0.004
arse+0.004
omed+0.004
 fears+0.004
 Ow+0.004
Neg logit contributions
rob-0.004
 couple-0.004
 seek-0.003
 trainer-0.003
 gentleman-0.003
Token elderly
Token ID13830
Feature activation+2.14e+00

Pos logit contributions
rob+0.001
robat+0.001
 seek+0.001
 couple+0.001
thank+0.001
Neg logit contributions
aying-0.002
 fears-0.001
 wipes-0.001
 fists-0.001
omed-0.001
Token lady
Token ID10846
Feature activation+1.31e+00

Pos logit contributions
rob+0.058
robat+0.044
 seek+0.043
 yak+0.038
Ar+0.036
Neg logit contributions
aying-0.130
 wipes-0.127
 cleans-0.121
omed-0.121
outed-0.117
Token laughed
Token ID13818
Feature activation+2.32e-01

Pos logit contributions
 couple+0.048
rob+0.043
 seek+0.040
 lady+0.039
 gentleman+0.039
Neg logit contributions
aying-0.074
arse-0.070
 fears-0.069
 Ow-0.068
uning-0.068
Token and
Token ID290
Feature activation+1.66e-01

Pos logit contributions
 couple+0.006
 lady+0.005
 man+0.005
 gentleman+0.005
 seek+0.005
Neg logit contributions
aying-0.015
 fears-0.015
arse-0.015
uning-0.015
ours-0.015
Token said
Token ID531
Feature activation+3.13e-01

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.004
 gentleman+0.004
 trainer+0.004
Neg logit contributions
aying-0.010
 fears-0.009
arse-0.009
uning-0.009
omed-0.009
Token,
Token ID11
Feature activation-1.45e-01

Pos logit contributions
 couple+0.009
 lady+0.008
 gentleman+0.007
rob+0.007
 man+0.007
Neg logit contributions
aying-0.020
 fears-0.020
arse-0.020
ains-0.019
oken-0.019
Token
Token ID220
Feature activation-1.52e-01

Pos logit contributions
aying+0.016
 fears+0.015
arse+0.015
uning+0.015
 fists+0.014
Neg logit contributions
 couple-0.013
rob-0.013
 seek-0.012
 gentleman-0.012
 lady-0.011
Token
Token ID198
Feature activation-8.33e-01

Pos logit contributions
 fears+0.010
aying+0.010
arse+0.010
uning+0.009
 fists+0.009
Neg logit contributions
rob-0.003
 couple-0.003
 seek-0.003
 lady-0.003
 gentleman-0.003
Token
Token ID198
Feature activation-3.88e-01

Pos logit contributions
aying+0.041
 Ow+0.039
omed+0.038
arse+0.038
 fears+0.037
Neg logit contributions
 couple-0.035
rob-0.033
 gentleman-0.030
 seek-0.030
 trainer-0.029
TokenAfter
Token ID3260
Feature activation+9.43e-01

Pos logit contributions
aying+0.017
arse+0.016
 fears+0.016
omed+0.016
 Ow+0.016
Neg logit contributions
 couple-0.018
rob-0.018
 gentleman-0.016
 seek-0.016
 trainer-0.015
Token the
Token ID262
Feature activation+5.72e-01

Pos logit contributions
 couple+0.036
rob+0.035
 seek+0.031
 gentleman+0.030
 trainer+0.029
Neg logit contributions
aying-0.050
 fears-0.047
arse-0.047
uning-0.046
omed-0.046
Token haircut
Token ID45548
Feature activation+2.14e+00

Pos logit contributions
rob+0.020
 couple+0.018
 seek+0.017
thank+0.016
keeping+0.016
Neg logit contributions
aying-0.031
 fears-0.030
 wipes-0.029
 fists-0.029
arse-0.029
Token was
Token ID373
Feature activation+4.92e-01

Pos logit contributions
 couple+0.088
rob+0.087
 seek+0.078
 gentleman+0.075
 trainer+0.074
Neg logit contributions
aying-0.104
 fears-0.097
uning-0.096
 cleans-0.095
 wipes-0.095
Token done
Token ID1760
Feature activation+2.48e-01

Pos logit contributions
 couple+0.019
rob+0.018
 seek+0.017
 lady+0.016
thank+0.016
Neg logit contributions
aying-0.026
 fears-0.025
arse-0.024
uning-0.024
 Ow-0.024
Token,
Token ID11
Feature activation+4.90e-03

Pos logit contributions
 couple+0.009
rob+0.008
 seek+0.007
 gentleman+0.007
 lady+0.007
Neg logit contributions
aying-0.015
 fears-0.014
arse-0.014
uning-0.014
 Ow-0.013
Token Tim
Token ID5045
Feature activation-1.97e-01

Pos logit contributions
 couple+0.000
rob+0.000
 lady+0.000
 man+0.000
 seek+0.000
Neg logit contributions
aying-0.000
 fears-0.000
arse-0.000
uning-0.000
oken-0.000
Tokenmy
Token ID1820
Feature activation-4.85e-01

Pos logit contributions
aying+0.008
 fears+0.008
arse+0.007
uning+0.007
omed+0.007
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.009
 trainer-0.009
Token goodbye
Token ID24829
Feature activation-5.02e-01

Pos logit contributions
arse+0.011
aying+0.011
 fears+0.010
 Ow+0.010
ains+0.010
Neg logit contributions
 couple-0.005
 lady-0.004
 man-0.004
 gentleman-0.004
rob-0.003
Token.
Token ID13
Feature activation+6.90e-02

Pos logit contributions
aying+0.027
 fears+0.026
arse+0.026
uning+0.025
omed+0.025
Neg logit contributions
 couple-0.018
rob-0.018
 seek-0.016
 gentleman-0.015
 trainer-0.015
Token
Token ID198
Feature activation-1.34e-01

Pos logit contributions
 couple+0.003
rob+0.003
 gentleman+0.003
 lady+0.003
 seek+0.003
Neg logit contributions
aying-0.003
arse-0.003
 fears-0.003
uning-0.003
iqu-0.003
Token
Token ID198
Feature activation-1.71e-01

Pos logit contributions
aying+0.007
arse+0.007
 fears+0.007
omed+0.007
 Ow+0.007
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.004
 gentleman-0.004
 trainer-0.004
TokenThe
Token ID464
Feature activation-2.63e-02

Pos logit contributions
aying+0.008
arse+0.007
 fears+0.007
omed+0.007
 Ow+0.007
Neg logit contributions
 couple-0.007
rob-0.007
 seek-0.006
 gentleman-0.006
 trainer-0.006
Token elderly
Token ID13830
Feature activation+2.14e+00

Pos logit contributions
aying+0.001
 fears+0.001
arse+0.001
uning+0.001
 fists+0.001
Neg logit contributions
rob-0.001
 couple-0.001
 seek-0.001
robat-0.001
 gentleman-0.001
Token woman
Token ID2415
Feature activation+5.69e-01

Pos logit contributions
rob+0.057
 seek+0.049
robat+0.043
seek+0.040
Ar+0.037
Neg logit contributions
aying-0.125
 wipes-0.120
 cleans-0.118
 fears-0.116
uning-0.115
Token was
Token ID373
Feature activation-1.52e-01

Pos logit contributions
 couple+0.021
rob+0.020
 seek+0.018
 gentleman+0.017
 trainer+0.017
Neg logit contributions
aying-0.031
arse-0.029
 fears-0.029
uning-0.029
 Ow-0.029
Token so
Token ID523
Feature activation+1.36e-01

Pos logit contributions
aying+0.008
 fears+0.007
arse+0.007
omed+0.007
 cleans+0.007
Neg logit contributions
rob-0.007
 couple-0.006
robat-0.005
 seek-0.005
 trainer-0.005
Token happy
Token ID3772
Feature activation-3.40e-01

Pos logit contributions
rob+0.005
 couple+0.004
 seek+0.004
robat+0.004
 gentleman+0.004
Neg logit contributions
aying-0.008
 fears-0.007
arse-0.007
omed-0.007
uning-0.007
Token to
Token ID284
Feature activation-1.37e-01

Pos logit contributions
aying+0.018
 fears+0.016
uning+0.016
arse+0.016
omed+0.016
Neg logit contributions
rob-0.014
 couple-0.013
 seek-0.011
 trainer-0.011
 gentleman-0.011
Token day
Token ID1110
Feature activation-1.12e+00

Pos logit contributions
aying+0.024
 fears+0.023
arse+0.023
uning+0.023
omed+0.022
Neg logit contributions
 couple-0.009
rob-0.008
 seek-0.007
 gentleman-0.006
 trainer-0.006
Token?"
Token ID1701
Feature activation-2.10e-01

Pos logit contributions
aying+0.057
arse+0.056
 fears+0.056
uning+0.054
 Ow+0.053
Neg logit contributions
 couple-0.046
rob-0.042
 seek-0.040
 lady-0.038
 gentleman-0.037
Token
Token ID198
Feature activation-2.01e-01

Pos logit contributions
aying+0.010
 fears+0.009
arse+0.009
uning+0.009
omed+0.009
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.008
 lady-0.008
Token
Token ID198
Feature activation-2.06e-01

Pos logit contributions
aying+0.010
 Ow+0.009
arse+0.009
 cleans+0.009
omed+0.009
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 trainer-0.007
 gentleman-0.007
TokenThe
Token ID464
Feature activation-1.60e-01

Pos logit contributions
aying+0.010
arse+0.009
 fears+0.009
omed+0.009
 Ow+0.009
Neg logit contributions
rob-0.009
 couple-0.008
 seek-0.007
 trainer-0.007
robat-0.007
Token elderly
Token ID13830
Feature activation+2.14e+00

Pos logit contributions
aying+0.007
 fears+0.007
 cleans+0.007
 ok+0.007
arse+0.006
Neg logit contributions
rob-0.008
robat-0.006
 seek-0.006
seek-0.006
 trainer-0.006
Token woman
Token ID2415
Feature activation+6.23e-01

Pos logit contributions
rob+0.075
 seek+0.073
robat+0.063
Ar+0.061
seek+0.059
Neg logit contributions
aying-0.107
 cleans-0.106
 wipes-0.103
adoes-0.095
uning-0.094
Token laughed
Token ID13818
Feature activation-2.65e-02

Pos logit contributions
 couple+0.018
 lady+0.015
 man+0.015
rob+0.014
 seek+0.014
Neg logit contributions
aying-0.040
arse-0.040
uning-0.038
 Ow-0.038
arding-0.038
Token and
Token ID290
Feature activation-2.05e-01

Pos logit contributions
aying+0.002
arse+0.002
 fears+0.002
uning+0.002
ours+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 lady-0.001
 man-0.001
 gentleman-0.001
Token said
Token ID531
Feature activation-7.87e-02

Pos logit contributions
aying+0.012
 fears+0.012
arse+0.012
 wipes+0.012
uning+0.012
Neg logit contributions
 couple-0.006
rob-0.006
 seek-0.005
 gentleman-0.005
 trainer-0.005
Token,
Token ID11
Feature activation-3.41e-01

Pos logit contributions
aying+0.005
arse+0.005
 fears+0.005
iqu+0.005
ains+0.005
Neg logit contributions
 couple-0.002
 lady-0.002
 man-0.001
rob-0.001
 gentleman-0.001
Token.
Token ID13
Feature activation+2.54e-01

Pos logit contributions
 couple+0.013
rob+0.013
 seek+0.011
 gentleman+0.011
 trainer+0.011
Neg logit contributions
aying-0.019
 fears-0.018
arse-0.018
uning-0.018
omed-0.017
Token
Token ID220
Feature activation+3.54e-01

Pos logit contributions
 couple+0.008
rob+0.007
 gentleman+0.007
 lady+0.007
 man+0.007
Neg logit contributions
 fears-0.015
aying-0.015
arse-0.014
uning-0.014
iqu-0.014
Token
Token ID198
Feature activation-2.61e-01

Pos logit contributions
 couple+0.005
rob+0.004
 lady+0.003
 man+0.003
 gentleman+0.003
Neg logit contributions
 fears-0.026
arse-0.025
aying-0.025
adoes-0.025
uning-0.024
Token
Token ID198
Feature activation-9.70e-02

Pos logit contributions
aying+0.016
 fears+0.015
arse+0.015
uning+0.015
omed+0.015
Neg logit contributions
 couple-0.008
rob-0.007
 seek-0.007
 gentleman-0.006
 lady-0.006
TokenThe
Token ID464
Feature activation+6.70e-02

Pos logit contributions
aying+0.005
 fears+0.005
arse+0.005
uning+0.005
omed+0.005
Neg logit contributions
 couple-0.004
rob-0.004
 seek-0.003
 gentleman-0.003
 trainer-0.003
Token elderly
Token ID13830
Feature activation+2.13e+00

Pos logit contributions
rob+0.002
 couple+0.002
 seek+0.002
robat+0.002
thank+0.001
Neg logit contributions
aying-0.004
 fears-0.004
arse-0.004
 fists-0.004
uning-0.004
Token person
Token ID1048
Feature activation+8.54e-01

Pos logit contributions
rob+0.062
robat+0.048
 seek+0.045
thank+0.039
please+0.038
Neg logit contributions
aying-0.128
 wipes-0.119
adoes-0.117
 cleans-0.116
 fears-0.115
Token said
Token ID531
Feature activation+5.24e-01

Pos logit contributions
 couple+0.032
rob+0.031
 seek+0.028
 gentleman+0.026
 lady+0.026
Neg logit contributions
aying-0.047
arse-0.044
 fears-0.044
uning-0.043
 Ow-0.043
Token,
Token ID11
Feature activation-2.62e-01

Pos logit contributions
 couple+0.015
 lady+0.012
 seek+0.012
 man+0.012
 gentleman+0.012
Neg logit contributions
aying-0.034
arse-0.033
 fears-0.033
iqu-0.032
ains-0.032
Token "
Token ID366
Feature activation-2.63e-01

Pos logit contributions
aying+0.016
 fears+0.016
arse+0.016
uning+0.015
iqu+0.015
Neg logit contributions
 couple-0.008
rob-0.007
 lady-0.007
 man-0.007
 gentleman-0.006
TokenYou
Token ID1639
Feature activation-3.32e-02

Pos logit contributions
aying+0.013
arse+0.012
 Ow+0.011
 fears+0.011
omed+0.011
Neg logit contributions
 couple-0.011
rob-0.011
 seek-0.010
 gentleman-0.010
 trainer-0.010
Token and
Token ID290
Feature activation+6.47e-02

Pos logit contributions
 couple+0.029
rob+0.027
 seek+0.025
 gentleman+0.024
 lady+0.024
Neg logit contributions
aying-0.050
 fears-0.046
arse-0.046
uning-0.046
 Ow-0.046
Token looked
Token ID3114
Feature activation-6.85e-02

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.002
 gentleman+0.002
 lady+0.002
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
uning-0.003
 Ow-0.003
Token back
Token ID736
Feature activation-9.65e-03

Pos logit contributions
aying+0.004
uning+0.004
 Ow+0.004
arse+0.004
 fears+0.004
Neg logit contributions
 couple-0.002
rob-0.002
 gentleman-0.002
 seek-0.002
 lady-0.002
Token.
Token ID13
Feature activation+8.31e-02

Pos logit contributions
aying+0.001
arse+0.000
 fears+0.000
uning+0.000
 Ow+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 seek-0.000
 gentleman-0.000
 lady-0.000
Token The
Token ID383
Feature activation+9.83e-02

Pos logit contributions
 couple+0.003
 gentleman+0.002
 seek+0.002
 lady+0.002
 man+0.002
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.005
uning-0.005
adoes-0.005
Token elderly
Token ID13830
Feature activation+2.11e+00

Pos logit contributions
rob+0.003
 seek+0.003
robat+0.002
seek+0.002
 couple+0.002
Neg logit contributions
 fears-0.006
aying-0.006
 fists-0.006
arse-0.006
 wipes-0.006
Token woman
Token ID2415
Feature activation+7.72e-01

Pos logit contributions
rob+0.078
 seek+0.072
robat+0.068
 yak+0.061
Ar+0.060
Neg logit contributions
aying-0.105
 wipes-0.102
 cleans-0.101
adoes-0.094
omed-0.092
Token was
Token ID373
Feature activation+1.82e-02

Pos logit contributions
 couple+0.029
rob+0.026
 seek+0.025
 lady+0.024
 gentleman+0.024
Neg logit contributions
aying-0.043
arse-0.041
uning-0.040
 fears-0.040
 Ow-0.040
Token not
Token ID407
Feature activation-5.28e-03

Pos logit contributions
rob+0.001
 couple+0.001
 seek+0.001
robat+0.001
 gentleman+0.001
Neg logit contributions
aying-0.001
 fears-0.001
arse-0.001
omed-0.001
 fists-0.001
Token angry
Token ID7954
Feature activation+6.66e-01

Pos logit contributions
arse+0.000
aying+0.000
 ok+0.000
 fears+0.000
 streaming+0.000
Neg logit contributions
rob-0.000
 couple-0.000
robat-0.000
 trainer-0.000
 gentleman-0.000
Token.
Token ID13
Feature activation-1.18e-01

Pos logit contributions
 couple+0.033
rob+0.032
 seek+0.029
 gentleman+0.028
 trainer+0.028
Neg logit contributions
aying-0.029
 fears-0.026
arse-0.026
uning-0.026
omed-0.025
Token okay
Token ID8788
Feature activation-7.94e-01

Pos logit contributions
 couple+0.004
rob+0.003
 seek+0.003
 lady+0.003
 gentleman+0.003
Neg logit contributions
aying-0.002
 fears-0.001
arse-0.001
uning-0.001
omed-0.001
Token.
Token ID13
Feature activation-2.73e-01

Pos logit contributions
aying+0.049
 fears+0.047
arse+0.047
uning+0.046
 Ow+0.045
Neg logit contributions
 couple-0.024
rob-0.021
 seek-0.020
 lady-0.019
 gentleman-0.018
Token
Token ID198
Feature activation+6.93e-02

Pos logit contributions
aying+0.014
 fears+0.013
arse+0.013
uning+0.012
omed+0.012
Neg logit contributions
 couple-0.011
rob-0.011
 seek-0.010
 gentleman-0.010
 lady-0.010
Token
Token ID198
Feature activation-2.87e-02

Pos logit contributions
 couple+0.003
rob+0.002
 seek+0.002
 gentleman+0.002
 trainer+0.002
Neg logit contributions
aying-0.004
arse-0.004
 fears-0.004
 Ow-0.004
omed-0.004
TokenThe
Token ID464
Feature activation+5.53e-02

Pos logit contributions
aying+0.002
arse+0.001
 fears+0.001
omed+0.001
uning+0.001
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 gentleman-0.001
 trainer-0.001
Token elderly
Token ID13830
Feature activation+2.10e+00

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.001
 gentleman+0.001
thank+0.001
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
uning-0.003
 Ow-0.003
Token man
Token ID582
Feature activation+6.99e-01

Pos logit contributions
rob+0.047
 seek+0.042
robat+0.036
thank+0.032
seek+0.032
Neg logit contributions
aying-0.130
 wipes-0.125
uning-0.123
 cleans-0.123
ours-0.120
Token smiled
Token ID13541
Feature activation+5.10e-01

Pos logit contributions
 couple+0.021
rob+0.018
 lady+0.017
 man+0.016
 trainer+0.016
Neg logit contributions
aying-0.045
arse-0.043
 fears-0.042
 Ow-0.042
adoes-0.042
Token weak
Token ID4939
Feature activation-3.91e-01

Pos logit contributions
rob+0.020
 couple+0.020
 seek+0.017
 gentleman+0.016
 trainer+0.016
Neg logit contributions
aying-0.027
 fears-0.026
arse-0.025
uning-0.025
omed-0.025
Tokenly
Token ID306
Feature activation-8.00e-01

Pos logit contributions
aying+0.021
arse+0.020
 fears+0.020
 Ow+0.019
uning+0.019
Neg logit contributions
 couple-0.015
rob-0.014
 lady-0.013
 gentleman-0.013
 man-0.013
Token and
Token ID290
Feature activation-3.12e-01

Pos logit contributions
aying+0.048
arse+0.046
 fears+0.045
uning+0.045
 Ow+0.045
Neg logit contributions
 couple-0.026
rob-0.023
 gentleman-0.021
 seek-0.021
 lady-0.021
Token feel
Token ID1254
Feature activation+1.71e-02

Pos logit contributions
 couple+0.004
 seek+0.004
rob+0.004
 lady+0.003
thank+0.003
Neg logit contributions
aying-0.005
omed-0.005
uning-0.005
 fears-0.005
arse-0.005
Token excited
Token ID6568
Feature activation+9.74e-02

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 lady+0.000
 gentleman+0.000
Neg logit contributions
aying-0.001
arse-0.001
uning-0.001
 Ow-0.001
omed-0.001
Token.
Token ID13
Feature activation-3.87e-01

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.003
 gentleman+0.003
 lady+0.003
Neg logit contributions
aying-0.006
 fears-0.005
arse-0.005
uning-0.005
 Ow-0.005
Token When
Token ID1649
Feature activation+2.27e-02

Pos logit contributions
aying+0.020
 fears+0.019
oken+0.019
uning+0.019
arse+0.019
Neg logit contributions
 couple-0.015
 gentleman-0.012
rob-0.012
 lady-0.012
 seek-0.012
Token the
Token ID262
Feature activation+3.54e-01

Pos logit contributions
 couple+0.001
 seek+0.001
rob+0.001
 lady+0.001
 man+0.001
Neg logit contributions
aying-0.001
 fears-0.001
uning-0.001
arse-0.001
outed-0.001
Token haircut
Token ID45548
Feature activation+2.09e+00

Pos logit contributions
rob+0.016
 couple+0.014
 seek+0.013
 trainer+0.013
robat+0.013
Neg logit contributions
aying-0.016
 fears-0.016
 wipes-0.015
 fists-0.015
arse-0.015
Token was
Token ID373
Feature activation+4.91e-01

Pos logit contributions
 couple+0.079
rob+0.076
 seek+0.068
 gentleman+0.065
 lady+0.063
Neg logit contributions
aying-0.114
 fears-0.107
arse-0.107
uning-0.105
omed-0.103
Token done
Token ID1760
Feature activation+3.79e-01

Pos logit contributions
 couple+0.016
 seek+0.014
thank+0.013
 lady+0.013
rob+0.013
Neg logit contributions
aying-0.028
 fears-0.028
 Ow-0.028
uning-0.028
arse-0.027
Token,
Token ID11
Feature activation+9.32e-03

Pos logit contributions
 couple+0.011
 seek+0.009
rob+0.009
 lady+0.008
 gentleman+0.008
Neg logit contributions
 fears-0.024
aying-0.024
 Ow-0.023
arse-0.023
outed-0.023
Token Little
Token ID7703
Feature activation-7.62e-01

Pos logit contributions
 couple+0.000
 seek+0.000
rob+0.000
 lady+0.000
 man+0.000
Neg logit contributions
aying-0.001
arse-0.000
 fears-0.000
uning-0.000
omed-0.000
Token one
Token ID530
Feature activation+3.53e-01

Pos logit contributions
aying+0.041
 fears+0.039
ither+0.038
oken+0.038
arse+0.037
Neg logit contributions
 couple-0.031
 man-0.028
 lady-0.028
 seek-0.028
rob-0.027
Token home
Token ID1363
Feature activation+6.34e-01

Pos logit contributions
aying+0.006
 fears+0.006
uning+0.005
arse+0.005
omed+0.005
Neg logit contributions
rob-0.004
 couple-0.003
 seek-0.003
 gentleman-0.003
 trainer-0.003
Token.
Token ID13
Feature activation+1.60e-01

Pos logit contributions
 couple+0.022
rob+0.021
 seek+0.018
 gentleman+0.017
 lady+0.017
Neg logit contributions
aying-0.037
 fears-0.035
arse-0.034
uning-0.034
 Ow-0.033
Token
Token ID198
Feature activation+1.57e-01

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.006
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.008
 fears-0.008
arse-0.008
uning-0.008
 Ow-0.008
Token
Token ID198
Feature activation+1.76e-01

Pos logit contributions
 couple+0.008
rob+0.008
 seek+0.007
thank+0.007
 gentleman+0.007
Neg logit contributions
aying-0.006
 Ow-0.006
 fears-0.006
omed-0.006
arse-0.006
TokenThe
Token ID464
Feature activation-6.89e-02

Pos logit contributions
 couple+0.008
rob+0.008
 seek+0.007
 gentleman+0.007
 trainer+0.007
Neg logit contributions
aying-0.008
arse-0.007
omed-0.007
 fears-0.007
 Ow-0.007
Token elderly
Token ID13830
Feature activation+2.09e+00

Pos logit contributions
aying+0.004
 fears+0.004
arse+0.004
uning+0.004
omed+0.004
Neg logit contributions
rob-0.002
 couple-0.002
 seek-0.002
 gentleman-0.002
thank-0.002
Token lady
Token ID10846
Feature activation+1.10e+00

Pos logit contributions
rob+0.048
 seek+0.042
robat+0.035
 couple+0.032
seek+0.032
Neg logit contributions
aying-0.130
 wipes-0.124
 cleans-0.121
adoes-0.119
 fears-0.116
Token was
Token ID373
Feature activation+6.78e-02

Pos logit contributions
 couple+0.041
rob+0.040
 seek+0.035
 gentleman+0.034
 trainer+0.033
Neg logit contributions
aying-0.059
 fears-0.056
arse-0.056
uning-0.055
omed-0.054
Token very
Token ID845
Feature activation+7.42e-01

Pos logit contributions
rob+0.003
 couple+0.003
robat+0.002
 seek+0.002
 gentleman+0.002
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
omed-0.003
 ok-0.003
Token happy
Token ID3772
Feature activation-3.35e-02

Pos logit contributions
rob+0.032
 couple+0.027
robat+0.026
 seek+0.024
 gentleman+0.023
Neg logit contributions
aying-0.038
 fears-0.037
outed-0.035
omed-0.035
arse-0.035
Token to
Token ID284
Feature activation+2.32e-01

Pos logit contributions
aying+0.002
 fears+0.002
uning+0.002
 wipes+0.001
omed+0.001
Neg logit contributions
rob-0.001
 couple-0.001
 seek-0.001
robat-0.001
 trainer-0.001
Token She
Token ID1375
Feature activation+1.11e-01

Pos logit contributions
 couple+0.014
rob+0.013
 lady+0.012
 gentleman+0.012
 man+0.011
Neg logit contributions
aying-0.024
 fears-0.023
arse-0.023
uning-0.022
omed-0.022
Token takes
Token ID2753
Feature activation-1.96e-01

Pos logit contributions
 couple+0.005
rob+0.005
 man+0.005
 lady+0.005
 trainer+0.005
Neg logit contributions
arse-0.005
aying-0.005
 Ow-0.004
uning-0.004
arding-0.004
Token out
Token ID503
Feature activation-2.44e-01

Pos logit contributions
aying+0.010
 fears+0.009
arse+0.009
omed+0.009
uning+0.009
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 lady-0.007
 gentleman-0.007
Token a
Token ID257
Feature activation+7.28e-02

Pos logit contributions
aying+0.016
arse+0.016
 fears+0.015
arding+0.015
uning+0.015
Neg logit contributions
 couple-0.007
 lady-0.006
 man-0.005
rob-0.005
 seek-0.005
Token magn
Token ID7842
Feature activation-2.45e-01

Pos logit contributions
 couple+0.003
 seek+0.003
rob+0.003
 lady+0.003
 man+0.003
Neg logit contributions
aying-0.004
uning-0.003
 fears-0.003
arse-0.003
adoes-0.003
Tokenifying
Token ID4035
Feature activation-2.40e+00

Pos logit contributions
 fears+0.006
 cleans+0.006
 wipes+0.006
 Acc+0.006
adoes+0.006
Neg logit contributions
 couple-0.015
rob-0.013
 trainer-0.013
amer-0.013
 seek-0.013
Token glass
Token ID5405
Feature activation-7.67e-02

Pos logit contributions
aying+0.083
 fears+0.076
 Ow+0.074
arse+0.073
uning+0.073
Neg logit contributions
 couple-0.136
rob-0.134
 seek-0.123
 gentleman-0.121
 trainer-0.119
Token.
Token ID13
Feature activation+5.92e-01

Pos logit contributions
aying+0.004
 fears+0.004
arse+0.004
uning+0.004
omed+0.004
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 gentleman-0.003
 trainer-0.002
Token She
Token ID1375
Feature activation+1.00e-01

Pos logit contributions
 couple+0.019
 lady+0.018
 man+0.016
 gentleman+0.015
rob+0.015
Neg logit contributions
 fears-0.035
arse-0.035
aying-0.034
oken-0.032
iqu-0.032
Token looks
Token ID3073
Feature activation+2.52e-02

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.004
 lady+0.004
 trainer+0.004
Neg logit contributions
arse-0.004
aying-0.004
 Ow-0.004
uning-0.004
arding-0.004
Token at
Token ID379
Feature activation+3.93e-02

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 lady+0.001
 man+0.001
Neg logit contributions
aying-0.001
arse-0.001
uning-0.001
 fears-0.001
 Ow-0.001
Token a
Token ID257
Feature activation+3.75e-01

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.005
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.007
 fears-0.006
arse-0.006
 Ow-0.006
uning-0.006
Token flashlight
Token ID38371
Feature activation-4.79e-01

Pos logit contributions
 couple+0.016
rob+0.015
 seek+0.014
 gentleman+0.013
 trainer+0.013
Neg logit contributions
aying-0.019
 fears-0.018
arse-0.018
uning-0.017
omed-0.017
Token and
Token ID290
Feature activation+3.40e-02

Pos logit contributions
aying+0.028
 fears+0.027
arse+0.027
 Ow+0.027
uning+0.026
Neg logit contributions
 couple-0.016
rob-0.015
 seek-0.013
 lady-0.013
 trainer-0.013
Token a
Token ID257
Feature activation-3.08e-01

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 gentleman+0.001
 trainer+0.001
Neg logit contributions
aying-0.002
arse-0.002
 fears-0.002
uning-0.002
omed-0.002
Token magn
Token ID7842
Feature activation-5.86e-01

Pos logit contributions
aying+0.015
arse+0.014
 fears+0.014
uning+0.014
omed+0.014
Neg logit contributions
 couple-0.013
rob-0.013
 seek-0.012
 gentleman-0.011
 lady-0.011
Tokenifying
Token ID4035
Feature activation-2.38e+00

Pos logit contributions
 fears+0.015
 cleans+0.014
 wipes+0.014
adoes+0.013
 Acc+0.013
Neg logit contributions
 couple-0.037
rob-0.036
 trainer-0.034
robat-0.033
 king-0.033
Token glass
Token ID5405
Feature activation-7.18e-02

Pos logit contributions
aying+0.079
 Ow+0.073
 fears+0.072
uning+0.069
arse+0.068
Neg logit contributions
 couple-0.138
rob-0.136
 gentleman-0.125
 seek-0.123
 trainer-0.121
Token to
Token ID284
Feature activation+4.28e-01

Pos logit contributions
aying+0.004
 fears+0.004
arse+0.004
uning+0.004
omed+0.004
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.002
 gentleman-0.002
thank-0.002
Token examine
Token ID10716
Feature activation+2.24e-01

Pos logit contributions
 couple+0.021
rob+0.021
 gentleman+0.018
 seek+0.018
thank+0.017
Neg logit contributions
aying-0.019
 fears-0.017
uning-0.017
arse-0.017
 wipes-0.016
Token their
Token ID511
Feature activation+2.59e-01

Pos logit contributions
 couple+0.007
rob+0.007
 seek+0.006
 lady+0.006
 man+0.006
Neg logit contributions
arse-0.013
aying-0.013
 fears-0.013
uning-0.013
omed-0.012
Token finds
Token ID7228
Feature activation+8.15e-02

Pos logit contributions
 couple+0.011
rob+0.010
 seek+0.010
 lady+0.009
 trainer+0.009
Neg logit contributions
aying-0.013
arse-0.012
omed-0.012
uning-0.011
 fears-0.011
Token He
Token ID679
Feature activation-1.78e-01

Pos logit contributions
 couple+0.023
rob+0.022
 seek+0.021
 gentleman+0.020
 lady+0.020
Neg logit contributions
aying-0.027
arse-0.025
 fears-0.025
uning-0.024
omed-0.024
Token took
Token ID1718
Feature activation-8.16e-03

Pos logit contributions
aying+0.008
 fears+0.007
 cleans+0.007
 wipes+0.007
outed+0.007
Neg logit contributions
rob-0.008
 couple-0.008
robat-0.007
 gentleman-0.007
thank-0.007
Token out
Token ID503
Feature activation+4.92e-02

Pos logit contributions
aying+0.000
 fears+0.000
arse+0.000
uning+0.000
 Ow+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 seek-0.000
 gentleman-0.000
 trainer-0.000
Token a
Token ID257
Feature activation+4.32e-01

Pos logit contributions
 couple+0.001
rob+0.001
 lady+0.001
 man+0.001
 seek+0.001
Neg logit contributions
aying-0.003
arse-0.003
uning-0.003
arding-0.003
ours-0.003
Token magn
Token ID7842
Feature activation-1.74e-01

Pos logit contributions
 couple+0.018
rob+0.018
 seek+0.016
 gentleman+0.016
 lady+0.015
Neg logit contributions
aying-0.022
 fears-0.020
arse-0.020
uning-0.020
 Ow-0.019
Tokenifying
Token ID4035
Feature activation-2.20e+00

Pos logit contributions
 fears+0.004
aying+0.004
 cleans+0.004
 wipes+0.003
adoes+0.003
Neg logit contributions
 couple-0.012
rob-0.011
 trainer-0.010
 gentleman-0.010
 seek-0.010
Token glass
Token ID5405
Feature activation+1.50e-01

Pos logit contributions
aying+0.074
 Ow+0.068
 fears+0.068
omed+0.065
uning+0.065
Neg logit contributions
 couple-0.125
rob-0.125
 seek-0.113
 gentleman-0.112
 trainer-0.110
Token and
Token ID290
Feature activation+3.26e-01

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.005
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.008
 fears-0.007
arse-0.007
uning-0.007
omed-0.007
Token a
Token ID257
Feature activation-8.26e-02

Pos logit contributions
rob+0.014
 couple+0.013
 gentleman+0.012
robat+0.011
 trainer+0.011
Neg logit contributions
aying-0.015
 fears-0.015
 wipes-0.015
 cleans-0.015
uning-0.014
Token notebook
Token ID20922
Feature activation+1.61e-01

Pos logit contributions
aying+0.004
 fears+0.004
arse+0.004
uning+0.004
omed+0.004
Neg logit contributions
 couple-0.004
rob-0.003
 seek-0.003
 gentleman-0.003
 trainer-0.003
Token and
Token ID290
Feature activation+2.53e-01

Pos logit contributions
rob+0.007
 couple+0.007
 gentleman+0.006
 seek+0.006
robat+0.006
Neg logit contributions
aying-0.008
omed-0.007
ows-0.007
 wipes-0.007
uning-0.007
Token the
Token ID262
Feature activation-4.81e-02

Pos logit contributions
aying+0.014
arse+0.014
arding+0.013
uning+0.013
 fears+0.013
Neg logit contributions
 couple-0.004
 lady-0.003
 l-0.003
 seek-0.002
 man-0.002
Token necklace
Token ID38987
Feature activation+8.12e-01

Pos logit contributions
aying+0.003
arse+0.002
 fears+0.002
uning+0.002
omed+0.002
Neg logit contributions
 couple-0.002
rob-0.002
 seek-0.002
 lady-0.002
 gentleman-0.002
Token and
Token ID290
Feature activation-1.10e-01

Pos logit contributions
rob+0.041
 couple+0.038
 gentleman+0.034
 seek+0.034
 businessman+0.033
Neg logit contributions
aying-0.035
ows-0.032
icking-0.032
uning-0.032
adoes-0.031
Token the
Token ID262
Feature activation-3.92e-01

Pos logit contributions
aying+0.005
 wipes+0.005
 fears+0.005
 cleans+0.005
uning+0.005
Neg logit contributions
rob-0.005
 couple-0.005
 gentleman-0.004
 seek-0.004
robat-0.004
Token magn
Token ID7842
Feature activation-2.80e-01

Pos logit contributions
aying+0.021
 fears+0.020
arse+0.019
uning+0.019
 Ow+0.019
Neg logit contributions
 couple-0.015
rob-0.015
 seek-0.013
 gentleman-0.013
 trainer-0.012
Tokenifying
Token ID4035
Feature activation-2.20e+00

Pos logit contributions
aying+0.007
 fears+0.007
 cleans+0.006
 wipes+0.006
adoes+0.006
Neg logit contributions
 couple-0.018
rob-0.018
 seek-0.016
 trainer-0.016
 gentleman-0.016
Token glass
Token ID5405
Feature activation+2.47e-03

Pos logit contributions
 Ow+0.057
aying+0.056
 wipes+0.050
 fears+0.050
 cleans+0.049
Neg logit contributions
 couple-0.142
rob-0.142
 gentleman-0.132
 seek-0.130
robat-0.128
Token and
Token ID290
Feature activation-7.37e-03

Pos logit contributions
 couple+0.000
rob+0.000
 seek+0.000
 gentleman+0.000
 trainer+0.000
Neg logit contributions
aying-0.000
 fears-0.000
arse-0.000
uning-0.000
omed-0.000
Token started
Token ID2067
Feature activation+2.65e-01

Pos logit contributions
 wipes+0.000
 cleans+0.000
aying+0.000
 fears+0.000
 fists+0.000
Neg logit contributions
rob-0.000
 couple-0.000
 gentleman-0.000
robat-0.000
 trainer-0.000
Token to
Token ID284
Feature activation+3.19e-01

Pos logit contributions
 couple+0.010
rob+0.010
 seek+0.009
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.014
 fears-0.014
arse-0.013
uning-0.013
omed-0.013
Token examine
Token ID10716
Feature activation+2.58e-01

Pos logit contributions
 couple+0.013
 seek+0.013
rob+0.013
thank+0.011
 lady+0.011
Neg logit contributions
aying-0.016
omed-0.015
arse-0.015
 fears-0.015
uning-0.015
Token the
Token ID262
Feature activation-1.67e-01

Pos logit contributions
arse+0.013
aying+0.012
omed+0.012
 fears+0.012
arding+0.011
Neg logit contributions
 couple-0.006
 lady-0.005
rob-0.005
 trainer-0.005
 l-0.005
Token necklace
Token ID38987
Feature activation+8.16e-01

Pos logit contributions
aying+0.008
 fears+0.008
arse+0.008
uning+0.008
omed+0.007
Neg logit contributions
 couple-0.007
rob-0.007
 seek-0.006
 gentleman-0.006
 trainer-0.006
Token and
Token ID290
Feature activation+9.57e-02

Pos logit contributions
rob+0.036
 couple+0.035
 seek+0.030
 gentleman+0.029
thank+0.029
Neg logit contributions
aying-0.040
 fears-0.036
uning-0.036
omed-0.036
 cleans-0.035
Token the
Token ID262
Feature activation-7.03e-01

Pos logit contributions
rob+0.004
 couple+0.004
robat+0.004
thank+0.003
 seek+0.003
Neg logit contributions
 cleans-0.005
 wipes-0.005
ours-0.004
 fists-0.004
 fears-0.004
Token magn
Token ID7842
Feature activation-4.35e-01

Pos logit contributions
aying+0.035
 fears+0.033
arse+0.033
uning+0.032
 Ow+0.032
Neg logit contributions
 couple-0.029
rob-0.029
 seek-0.025
 gentleman-0.025
 trainer-0.024
Tokenifying
Token ID4035
Feature activation-2.19e+00

Pos logit contributions
 fears+0.011
aying+0.011
 wipes+0.010
 cleans+0.010
adoes+0.010
Neg logit contributions
 couple-0.028
rob-0.027
 trainer-0.024
 seek-0.024
robat-0.024
Token glass
Token ID5405
Feature activation-2.14e-02

Pos logit contributions
aying+0.053
 Ow+0.052
 wipes+0.047
 fears+0.047
 cleans+0.047
Neg logit contributions
rob-0.146
 couple-0.144
 gentleman-0.135
 seek-0.133
robat-0.132
Token and
Token ID290
Feature activation+1.39e-01

Pos logit contributions
aying+0.001
 fears+0.001
arse+0.001
uning+0.001
omed+0.001
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 gentleman-0.001
 trainer-0.001
Token told
Token ID1297
Feature activation-1.69e-01

Pos logit contributions
rob+0.006
 couple+0.006
robat+0.005
 gentleman+0.005
 seek+0.005
Neg logit contributions
 cleans-0.007
 wipes-0.006
aying-0.006
 fears-0.006
 fists-0.006
Token him
Token ID683
Feature activation+2.61e-02

Pos logit contributions
aying+0.009
arse+0.008
 fears+0.008
uning+0.008
omed+0.008
Neg logit contributions
 couple-0.007
rob-0.007
 seek-0.006
 gentleman-0.006
 trainer-0.006
Token that
Token ID326
Feature activation-4.74e-01

Pos logit contributions
 couple+0.001
 lady+0.001
 ranger+0.001
 man+0.001
 trainer+0.001
Neg logit contributions
arse-0.002
aying-0.002
 Ow-0.001
uning-0.001
akes-0.001
Token She
Token ID1375
Feature activation+2.38e-01

Pos logit contributions
 couple+0.027
rob+0.025
 seek+0.023
 gentleman+0.023
 lady+0.023
Neg logit contributions
aying-0.029
arse-0.027
 fears-0.027
omed-0.026
 Ow-0.026
Token took
Token ID1718
Feature activation-8.02e-02

Pos logit contributions
 couple+0.010
rob+0.010
 gentleman+0.009
 seek+0.009
 trainer+0.008
Neg logit contributions
aying-0.012
 fears-0.011
uning-0.011
 cleans-0.011
omed-0.011
Token out
Token ID503
Feature activation-2.03e-01

Pos logit contributions
aying+0.004
arse+0.004
 fears+0.004
uning+0.004
omed+0.004
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 gentleman-0.003
 lady-0.003
Token a
Token ID257
Feature activation+3.31e-01

Pos logit contributions
aying+0.013
arse+0.013
 Ow+0.012
 fears+0.012
arding+0.012
Neg logit contributions
 couple-0.006
 lady-0.004
rob-0.004
 seek-0.004
 l-0.004
Token magn
Token ID7842
Feature activation-2.55e-01

Pos logit contributions
 couple+0.013
 seek+0.012
 lady+0.011
 man+0.011
rob+0.011
Neg logit contributions
aying-0.019
uning-0.018
adoes-0.017
outed-0.017
 fears-0.017
Tokenifying
Token ID4035
Feature activation-2.19e+00

Pos logit contributions
 cleans+0.007
 fears+0.007
 wipes+0.006
ours+0.006
 Acc+0.006
Neg logit contributions
 couple-0.015
rob-0.014
 trainer-0.014
amer-0.013
 seek-0.013
Token glass
Token ID5405
Feature activation+2.71e-04

Pos logit contributions
aying+0.074
 fears+0.067
 Ow+0.067
uning+0.065
omed+0.064
Neg logit contributions
 couple-0.126
rob-0.125
 seek-0.113
 gentleman-0.113
 trainer-0.110
Token and
Token ID290
Feature activation+3.33e-01

Pos logit contributions
 couple+0.000
rob+0.000
 seek+0.000
 gentleman+0.000
 trainer+0.000
Neg logit contributions
aying-0.000
 fears-0.000
arse-0.000
uning-0.000
omed-0.000
Token looked
Token ID3114
Feature activation-5.32e-01

Pos logit contributions
rob+0.014
 couple+0.014
 gentleman+0.012
 seek+0.012
 trainer+0.012
Neg logit contributions
aying-0.016
 fears-0.015
uning-0.015
 wipes-0.015
arse-0.015
Token very
Token ID845
Feature activation+4.71e-02

Pos logit contributions
aying+0.030
arse+0.029
 Ow+0.028
uning+0.028
 fears+0.027
Neg logit contributions
 couple-0.019
rob-0.016
 seek-0.016
 lady-0.015
 gentleman-0.015
Token carefully
Token ID7773
Feature activation-5.00e-01

Pos logit contributions
 couple+0.002
 seek+0.001
 man+0.001
 lady+0.001
 king+0.001
Neg logit contributions
aying-0.003
arse-0.003
 Ow-0.003
uning-0.003
adoes-0.003
Token flashlight
Token ID38371
Feature activation-4.96e-01

Pos logit contributions
aying+0.014
 fears+0.013
arse+0.013
uning+0.013
omed+0.013
Neg logit contributions
 couple-0.011
rob-0.010
 seek-0.010
 gentleman-0.009
 lady-0.009
Token,
Token ID11
Feature activation-3.77e-02

Pos logit contributions
aying+0.029
 fears+0.027
arse+0.027
 Ow+0.027
uning+0.026
Neg logit contributions
 couple-0.017
rob-0.016
 seek-0.015
 lady-0.014
 gentleman-0.014
Token and
Token ID290
Feature activation-2.72e-02

Pos logit contributions
aying+0.002
arse+0.002
 fears+0.002
uning+0.002
 Ow+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 lady-0.001
 gentleman-0.001
Token a
Token ID257
Feature activation-4.97e-01

Pos logit contributions
aying+0.001
 fears+0.001
uning+0.001
arse+0.001
 wipes+0.001
Neg logit contributions
rob-0.001
 couple-0.001
 seek-0.001
 gentleman-0.001
 trainer-0.001
Token magn
Token ID7842
Feature activation-3.47e-01

Pos logit contributions
aying+0.026
arse+0.024
 fears+0.024
uning+0.024
omed+0.023
Neg logit contributions
 couple-0.021
rob-0.019
 seek-0.018
 gentleman-0.017
 lady-0.017
Tokenifying
Token ID4035
Feature activation-2.19e+00

Pos logit contributions
 fears+0.010
 cleans+0.009
 wipes+0.009
adoes+0.009
 Acc+0.009
Neg logit contributions
 couple-0.021
rob-0.020
 king-0.018
 trainer-0.018
 gentleman-0.018
Token glass
Token ID5405
Feature activation+1.73e-01

Pos logit contributions
aying+0.083
 fears+0.076
arse+0.075
uning+0.074
 Ow+0.074
Neg logit contributions
 couple-0.117
rob-0.114
 seek-0.105
 gentleman-0.103
 trainer-0.102
Token.
Token ID13
Feature activation+2.19e-01

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.005
 gentleman+0.005
 lady+0.005
Neg logit contributions
aying-0.010
 fears-0.009
arse-0.009
uning-0.009
 Ow-0.009
Token
Token ID198
Feature activation+3.87e-01

Pos logit contributions
 couple+0.009
rob+0.008
 seek+0.008
 lady+0.008
 gentleman+0.008
Neg logit contributions
aying-0.011
arse-0.011
 fears-0.011
uning-0.010
 fists-0.010
Token
Token ID198
Feature activation+3.12e-01

Pos logit contributions
 couple+0.014
rob+0.013
 seek+0.012
 gentleman+0.011
 trainer+0.011
Neg logit contributions
aying-0.022
 fears-0.021
arse-0.020
omed-0.020
 Ow-0.020
TokenAs
Token ID1722
Feature activation+7.58e-01

Pos logit contributions
 couple+0.011
rob+0.011
 seek+0.009
 gentleman+0.009
 trainer+0.009
Neg logit contributions
aying-0.018
 fears-0.016
arse-0.016
omed-0.016
uning-0.016
Token When
Token ID1649
Feature activation+3.60e-01

Pos logit contributions
 couple+0.016
rob+0.014
 seek+0.014
 gentleman+0.013
 lady+0.013
Neg logit contributions
aying-0.019
arse-0.018
 fears-0.018
uning-0.017
omed-0.017
Token Daisy
Token ID40355
Feature activation+2.08e-01

Pos logit contributions
 couple+0.011
rob+0.010
 seek+0.009
 lady+0.009
 gentleman+0.009
Neg logit contributions
aying-0.022
 fears-0.021
arse-0.021
uning-0.021
omed-0.021
Token put
Token ID1234
Feature activation+4.58e-01

Pos logit contributions
 couple+0.008
rob+0.007
 seek+0.006
 gentleman+0.006
 lady+0.006
Neg logit contributions
aying-0.012
arse-0.011
 fears-0.011
adoes-0.011
uning-0.011
Token the
Token ID262
Feature activation+3.36e-01

Pos logit contributions
 couple+0.017
rob+0.016
 seek+0.014
 gentleman+0.014
 lady+0.013
Neg logit contributions
aying-0.026
 fears-0.024
arse-0.024
uning-0.024
omed-0.023
Token magn
Token ID7842
Feature activation+2.83e-01

Pos logit contributions
 couple+0.014
rob+0.012
 lady+0.012
 seek+0.011
 man+0.011
Neg logit contributions
aying-0.018
arse-0.017
 fears-0.017
uning-0.017
omed-0.017
Tokenifying
Token ID4035
Feature activation-2.17e+00

Pos logit contributions
 couple+0.017
rob+0.015
 seek+0.015
 trainer+0.014
 lady+0.014
Neg logit contributions
 fears-0.009
 cleans-0.008
 wipes-0.008
 Acc-0.008
aying-0.008
Token glass
Token ID5405
Feature activation+2.80e-01

Pos logit contributions
aying+0.081
 Ow+0.074
 fears+0.073
uning+0.071
omed+0.071
Neg logit contributions
 couple-0.118
rob-0.116
 seek-0.106
 gentleman-0.105
 trainer-0.103
Token down
Token ID866
Feature activation+6.27e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.009
 lady+0.008
 man+0.008
Neg logit contributions
aying-0.016
arse-0.015
 fears-0.015
uning-0.015
 Ow-0.015
Token,
Token ID11
Feature activation+6.22e-01

Pos logit contributions
 couple+0.019
 seek+0.017
 lady+0.016
rob+0.016
 man+0.015
Neg logit contributions
aying-0.039
arse-0.038
 fears-0.037
uning-0.036
iqu-0.036
Token the
Token ID262
Feature activation+9.07e-02

Pos logit contributions
 couple+0.014
rob+0.012
 seek+0.012
 lady+0.010
 gentleman+0.010
Neg logit contributions
aying-0.044
arse-0.042
 fears-0.042
uning-0.041
ours-0.041
Token cater
Token ID20825
Feature activation-3.11e-01

Pos logit contributions
 couple+0.003
 seek+0.003
 lady+0.003
rob+0.003
 man+0.003
Neg logit contributions
aying-0.005
arse-0.005
uning-0.005
 fears-0.005
omed-0.005
Token learned
Token ID4499
Feature activation-6.90e-02

Pos logit contributions
rob+0.002
 couple+0.002
 gentleman+0.002
 trainer+0.002
robat+0.002
Neg logit contributions
 cleans-0.003
 fears-0.003
aying-0.003
 wipes-0.003
adoes-0.003
Token even
Token ID772
Feature activation-4.20e-01

Pos logit contributions
aying+0.004
arse+0.004
uning+0.004
 fears+0.004
omed+0.004
Neg logit contributions
 couple-0.002
rob-0.002
 lady-0.002
 seek-0.002
 man-0.002
Token more
Token ID517
Feature activation+8.25e-02

Pos logit contributions
aying+0.020
 fears+0.020
 wipes+0.018
uning+0.018
 cleans+0.018
Neg logit contributions
rob-0.019
 couple-0.018
 seek-0.016
 gentleman-0.016
robat-0.016
Token about
Token ID546
Feature activation-3.83e-01

Pos logit contributions
 couple+0.004
rob+0.003
 seek+0.003
 lady+0.003
 gentleman+0.003
Neg logit contributions
aying-0.004
arse-0.004
 Ow-0.004
 fears-0.004
uning-0.004
Token how
Token ID703
Feature activation-9.48e-01

Pos logit contributions
aying+0.021
 fears+0.020
arse+0.020
uning+0.019
omed+0.019
Neg logit contributions
 couple-0.014
rob-0.014
 seek-0.012
 gentleman-0.012
 trainer-0.011
Token volcan
Token ID17516
Feature activation-2.12e+00

Pos logit contributions
aying+0.056
arse+0.054
 Ow+0.052
uning+0.051
omed+0.051
Neg logit contributions
 couple-0.033
 seek-0.027
 lady-0.026
 man-0.025
rob-0.025
Tokenoes
Token ID3028
Feature activation-3.57e-01

Pos logit contributions
 wipes+0.091
arse+0.091
 fists+0.088
 fears+0.086
 cleans+0.085
Neg logit contributions
rob-0.092
 couple-0.088
thank-0.082
 king-0.081
 trainer-0.078
Token work
Token ID670
Feature activation+4.31e-01

Pos logit contributions
arse+0.019
aying+0.019
 fears+0.018
omed+0.018
outed+0.018
Neg logit contributions
 couple-0.013
rob-0.012
 seek-0.012
 man-0.011
 lady-0.011
Token.
Token ID13
Feature activation+3.75e-02

Pos logit contributions
 couple+0.012
rob+0.011
 seek+0.011
 lady+0.010
thank+0.010
Neg logit contributions
aying-0.027
arse-0.026
 Ow-0.026
 fears-0.026
uning-0.025
Token She
Token ID1375
Feature activation+2.04e-02

Pos logit contributions
 couple+0.002
rob+0.001
 seek+0.001
 gentleman+0.001
 trainer+0.001
Neg logit contributions
aying-0.002
 fears-0.002
arse-0.002
uning-0.002
omed-0.002
Token thought
Token ID1807
Feature activation+1.35e-01

Pos logit contributions
 couple+0.001
rob+0.001
 gentleman+0.001
 seek+0.001
robat+0.001
Neg logit contributions
aying-0.001
 fears-0.001
 cleans-0.001
 wipes-0.001
uning-0.001
Token job
Token ID1693
Feature activation+7.73e-01

Pos logit contributions
aying+0.004
 fears+0.004
arse+0.004
uning+0.004
 Ow+0.004
Neg logit contributions
 couple-0.004
rob-0.004
 seek-0.003
 gentleman-0.003
 trainer-0.003
Token cleaning
Token ID12724
Feature activation+5.48e-01

Pos logit contributions
 couple+0.027
rob+0.027
 seek+0.023
 gentleman+0.022
 trainer+0.022
Neg logit contributions
aying-0.043
 fears-0.041
arse-0.040
uning-0.040
omed-0.040
Token the
Token ID262
Feature activation+7.52e-02

Pos logit contributions
 couple+0.022
rob+0.021
 seek+0.019
 gentleman+0.018
 lady+0.018
Neg logit contributions
aying-0.028
 fears-0.027
arse-0.027
uning-0.026
omed-0.026
Token garage
Token ID15591
Feature activation+5.26e-01

Pos logit contributions
 couple+0.004
rob+0.004
 seek+0.003
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
uning-0.003
omed-0.003
Token."
Token ID526
Feature activation-5.17e-02

Pos logit contributions
 couple+0.020
rob+0.020
 seek+0.017
 gentleman+0.017
 trainer+0.017
Neg logit contributions
aying-0.028
 fears-0.026
arse-0.026
uning-0.025
omed-0.025
Token Little
Token ID7703
Feature activation-2.10e+00

Pos logit contributions
 fears+0.003
aying+0.003
uning+0.003
oken+0.003
arding+0.003
Neg logit contributions
 couple-0.002
rob-0.002
 gentleman-0.001
 seek-0.001
 lady-0.001
Token Jimmy
Token ID12963
Feature activation-3.13e-01

Pos logit contributions
 fears+0.119
aying+0.118
uning+0.118
oken+0.114
ither+0.114
Neg logit contributions
 couple-0.077
 man-0.075
rob-0.074
 lady-0.069
 gentleman-0.069
Token smiled
Token ID13541
Feature activation+5.14e-01

Pos logit contributions
aying+0.017
arse+0.016
uning+0.016
 fears+0.016
omed+0.015
Neg logit contributions
 couple-0.012
rob-0.011
 seek-0.010
 gentleman-0.010
 lady-0.010
Token.
Token ID13
Feature activation-8.75e-02

Pos logit contributions
 couple+0.019
rob+0.018
 seek+0.016
 gentleman+0.015
 trainer+0.015
Neg logit contributions
aying-0.029
 fears-0.027
arse-0.027
uning-0.026
omed-0.026
Token He
Token ID679
Feature activation-1.66e-01

Pos logit contributions
aying+0.005
arse+0.005
 fears+0.005
uning+0.005
omed+0.005
Neg logit contributions
 couple-0.003
rob-0.003
 gentleman-0.003
 seek-0.003
 lady-0.003
Token had
Token ID550
Feature activation-1.16e-01

Pos logit contributions
aying+0.007
 wipes+0.007
 fears+0.007
 cleans+0.007
omed+0.007
Neg logit contributions
rob-0.008
 couple-0.007
 seek-0.007
 gentleman-0.006
 trainer-0.006
Token.
Token ID13
Feature activation-5.31e-01

Pos logit contributions
aying+0.053
 fears+0.050
uning+0.047
ours+0.047
arse+0.047
Neg logit contributions
rob-0.044
 couple-0.042
 gentleman-0.037
 seek-0.036
robat-0.035
Token They
Token ID1119
Feature activation+3.23e-01

Pos logit contributions
arse+0.022
aying+0.022
 fears+0.021
iqu+0.021
omed+0.020
Neg logit contributions
 couple-0.026
rob-0.024
 seek-0.023
 trainer-0.023
 lady-0.023
Token wear
Token ID5806
Feature activation-2.03e-01

Pos logit contributions
 couple+0.012
rob+0.011
 seek+0.011
 trainer+0.011
 king+0.010
Neg logit contributions
aying-0.017
arse-0.017
adoes-0.017
uning-0.017
omed-0.017
Token masks
Token ID20680
Feature activation-6.72e-01

Pos logit contributions
aying+0.011
arse+0.011
 fears+0.010
uning+0.010
omed+0.010
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 trainer-0.007
 par-0.006
Token and
Token ID290
Feature activation+1.02e-01

Pos logit contributions
aying+0.038
 fears+0.037
arse+0.037
 Ow+0.036
uning+0.036
Neg logit contributions
 couple-0.024
rob-0.022
 seek-0.020
 lady-0.019
 trainer-0.019
Token fins
Token ID42080
Feature activation-2.09e+00

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.004
 gentleman+0.004
 trainer+0.004
Neg logit contributions
aying-0.005
 fears-0.004
arse-0.004
uning-0.004
omed-0.004
Token to
Token ID284
Feature activation+2.87e-02

Pos logit contributions
arse+0.138
aying+0.136
 fears+0.132
 Ow+0.130
uning+0.130
Neg logit contributions
 couple-0.054
 lady-0.043
 man-0.042
 trainer-0.042
rob-0.042
Token see
Token ID766
Feature activation-3.30e-01

Pos logit contributions
 couple+0.001
rob+0.001
 gentleman+0.001
 seek+0.001
thank+0.001
Neg logit contributions
aying-0.001
 fears-0.001
arse-0.001
 wipes-0.001
uning-0.001
Token the
Token ID262
Feature activation+6.61e-02

Pos logit contributions
aying+0.019
arse+0.019
omed+0.018
 fears+0.018
uning+0.018
Neg logit contributions
 couple-0.011
rob-0.010
 lady-0.009
 trainer-0.009
 seek-0.009
Token fish
Token ID5916
Feature activation-8.36e-01

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 trainer+0.002
 lady+0.002
Neg logit contributions
aying-0.004
arse-0.003
 fears-0.003
uning-0.003
omed-0.003
Token and
Token ID290
Feature activation+1.18e-01

Pos logit contributions
aying+0.045
arse+0.043
 fears+0.043
uning+0.042
 Ow+0.042
Neg logit contributions
 couple-0.031
rob-0.029
 seek-0.027
 trainer-0.026
 lady-0.025
Token said
Token ID531
Feature activation-5.54e-02

Pos logit contributions
rob+0.010
 couple+0.009
robat+0.008
thank+0.008
 gentleman+0.008
Neg logit contributions
 wipes-0.010
outed-0.010
aying-0.010
 cleans-0.010
 fears-0.010
Token,
Token ID11
Feature activation+9.06e-01

Pos logit contributions
aying+0.003
 fears+0.003
arse+0.003
 Ow+0.003
uning+0.003
Neg logit contributions
 couple-0.002
rob-0.001
 seek-0.001
 lady-0.001
 gentleman-0.001
Token holding
Token ID4769
Feature activation-5.97e-02

Pos logit contributions
 couple+0.034
rob+0.033
 seek+0.029
 gentleman+0.028
 trainer+0.027
Neg logit contributions
aying-0.049
 fears-0.046
arse-0.046
uning-0.045
omed-0.045
Token the
Token ID262
Feature activation-5.74e-02

Pos logit contributions
aying+0.003
arse+0.003
 fears+0.003
omed+0.003
uning+0.003
Neg logit contributions
 couple-0.002
rob-0.002
 lady-0.002
 seek-0.002
 man-0.002
Token magn
Token ID7842
Feature activation+2.47e-01

Pos logit contributions
aying+0.003
omed+0.003
arse+0.003
 fears+0.003
ours+0.003
Neg logit contributions
 couple-0.002
 seek-0.002
rob-0.002
 lady-0.002
 man-0.002
Tokenifying
Token ID4035
Feature activation-2.08e+00

Pos logit contributions
 couple+0.015
rob+0.014
 seek+0.014
 trainer+0.014
robat+0.014
Neg logit contributions
 cleans-0.006
 fears-0.006
 wipes-0.006
aying-0.006
ours-0.005
Token glass
Token ID5405
Feature activation+3.52e-01

Pos logit contributions
aying+0.060
 Ow+0.057
 fears+0.054
 wipes+0.052
uning+0.052
Neg logit contributions
rob-0.129
 couple-0.129
 gentleman-0.117
 seek-0.117
 trainer-0.114
Token over
Token ID625
Feature activation+4.40e-01

Pos logit contributions
 couple+0.012
rob+0.010
 seek+0.010
 lady+0.010
 trainer+0.009
Neg logit contributions
aying-0.021
arse-0.020
 fears-0.020
uning-0.019
 Ow-0.019
Token a
Token ID257
Feature activation+4.31e-01

Pos logit contributions
 couple+0.016
rob+0.016
 seek+0.014
 gentleman+0.013
 trainer+0.013
Neg logit contributions
aying-0.024
 fears-0.023
arse-0.023
uning-0.023
omed-0.022
Token black
Token ID2042
Feature activation+3.58e-01

Pos logit contributions
rob+0.019
 couple+0.018
 seek+0.015
 gentleman+0.015
thank+0.015
Neg logit contributions
aying-0.020
 fears-0.020
arse-0.020
 Ow-0.019
omed-0.019
Token ant
Token ID1885
Feature activation-4.52e-01

Pos logit contributions
 couple+0.015
rob+0.014
 seek+0.013
 gentleman+0.012
 trainer+0.012
Neg logit contributions
aying-0.018
 fears-0.017
arse-0.017
uning-0.017
omed-0.017
Token examine
Token ID10716
Feature activation+3.60e-01

Pos logit contributions
 couple+0.041
rob+0.039
 seek+0.036
 gentleman+0.034
 trainer+0.034
Neg logit contributions
aying-0.045
 fears-0.042
arse-0.042
uning-0.041
omed-0.041
Token it
Token ID340
Feature activation-2.25e-01

Pos logit contributions
 couple+0.013
rob+0.012
 seek+0.011
 lady+0.010
 trainer+0.010
Neg logit contributions
aying-0.020
arse-0.020
 fears-0.019
uning-0.019
omed-0.019
Token with
Token ID351
Feature activation+5.66e-01

Pos logit contributions
aying+0.009
arse+0.008
uning+0.008
 fears+0.008
 Ow+0.008
Neg logit contributions
 couple-0.011
rob-0.011
 gentleman-0.010
 seek-0.010
 trainer-0.010
Token the
Token ID262
Feature activation+3.92e-01

Pos logit contributions
 couple+0.021
rob+0.019
 seek+0.018
 lady+0.017
 trainer+0.017
Neg logit contributions
aying-0.032
arse-0.030
 fears-0.030
uning-0.029
omed-0.029
Token magn
Token ID7842
Feature activation-6.64e-03

Pos logit contributions
 couple+0.017
rob+0.016
 seek+0.015
 lady+0.015
 trainer+0.014
Neg logit contributions
aying-0.019
arse-0.018
 fears-0.018
uning-0.018
omed-0.017
Tokenifying
Token ID4035
Feature activation-2.04e+00

Pos logit contributions
 fears+0.000
 cleans+0.000
 wipes+0.000
aying+0.000
 fists+0.000
Neg logit contributions
 couple-0.000
rob-0.000
robat-0.000
 trainer-0.000
 seek-0.000
Token glass
Token ID5405
Feature activation+1.35e-01

Pos logit contributions
aying+0.062
 Ow+0.055
 fears+0.054
 wipes+0.053
uning+0.053
Neg logit contributions
 couple-0.126
rob-0.124
 seek-0.114
 gentleman-0.114
 trainer-0.111
Token?"
Token ID1701
Feature activation+2.96e-02

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.006
 gentleman+0.006
 trainer+0.005
Neg logit contributions
aying-0.006
 fears-0.005
uning-0.005
omed-0.005
arse-0.005
Token
Token ID198
Feature activation-1.40e-01

Pos logit contributions
rob+0.001
 couple+0.001
robat+0.001
 seek+0.001
thank+0.001
Neg logit contributions
aying-0.001
 Ow-0.001
 wipes-0.001
omed-0.001
 fears-0.001
Token
Token ID198
Feature activation-7.06e-02

Pos logit contributions
aying+0.008
omed+0.007
 fears+0.007
 Ow+0.007
arse+0.007
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.004
 gentleman-0.004
 trainer-0.004
Token"
Token ID1
Feature activation+1.30e+00

Pos logit contributions
aying+0.004
omed+0.004
arse+0.004
 fears+0.004
 wipes+0.004
Neg logit contributions
rob-0.003
 couple-0.003
 gentleman-0.002
robat-0.002
 trainer-0.002
Token big
Token ID1263
Feature activation-1.43e-01

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 lady+0.002
 gentleman+0.002
Neg logit contributions
aying-0.003
 fears-0.003
arse-0.003
uning-0.003
omed-0.003
Token book
Token ID1492
Feature activation-4.71e-01

Pos logit contributions
aying+0.008
arse+0.007
 fears+0.007
uning+0.007
omed+0.007
Neg logit contributions
 couple-0.006
rob-0.006
 seek-0.005
 lady-0.005
 gentleman-0.005
Token with
Token ID351
Feature activation+7.09e-01

Pos logit contributions
aying+0.022
omed+0.020
uning+0.019
arse+0.019
 fears+0.019
Neg logit contributions
 couple-0.022
rob-0.021
 seek-0.019
 gentleman-0.018
 trainer-0.018
Token pictures
Token ID5986
Feature activation-2.49e-01

Pos logit contributions
rob+0.036
 couple+0.034
 seek+0.033
 gentleman+0.032
robat+0.032
Neg logit contributions
aying-0.027
 fears-0.027
 Ow-0.027
 fists-0.026
 cleans-0.026
Token of
Token ID286
Feature activation-6.36e-01

Pos logit contributions
aying+0.013
omed+0.011
uning+0.011
 cleans+0.011
ours+0.011
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.009
 trainer-0.009
Token volcan
Token ID17516
Feature activation-2.03e+00

Pos logit contributions
aying+0.032
 fears+0.031
 Ow+0.030
arse+0.030
 wipes+0.029
Neg logit contributions
 couple-0.025
rob-0.025
 seek-0.022
 gentleman-0.022
thank-0.021
Tokenoes
Token ID3028
Feature activation-6.53e-01

Pos logit contributions
aying+0.074
arse+0.071
 fears+0.071
 wipes+0.070
 fists+0.067
Neg logit contributions
rob-0.107
 couple-0.105
 seek-0.094
 trainer-0.093
thank-0.093
Token.
Token ID13
Feature activation-5.18e-01

Pos logit contributions
aying+0.037
 fears+0.035
arse+0.035
uning+0.035
omed+0.034
Neg logit contributions
 couple-0.022
rob-0.022
 seek-0.019
 gentleman-0.018
 trainer-0.018
Token Vol
Token ID4709
Feature activation-5.16e-01

Pos logit contributions
arse+0.024
aying+0.024
 fears+0.024
uning+0.022
iqu+0.022
Neg logit contributions
 couple-0.025
rob-0.022
 lady-0.022
 gentleman-0.021
 man-0.021
Tokencan
Token ID5171
Feature activation-7.65e-01

Pos logit contributions
 fears+0.027
aying+0.026
 wipes+0.025
 fists+0.025
uning+0.025
Neg logit contributions
rob-0.021
 couple-0.019
 seek-0.018
 gentleman-0.017
 trainer-0.017
Tokenoes
Token ID3028
Feature activation-7.65e-01

Pos logit contributions
arse+0.040
 fears+0.039
aying+0.039
 fists+0.038
iqu+0.037
Neg logit contributions
 couple-0.027
 seek-0.026
 lady-0.025
rob-0.025
 ranger-0.025
Token a
Token ID257
Feature activation-4.08e-01

Pos logit contributions
aying+0.008
 fears+0.008
arse+0.008
uning+0.007
omed+0.007
Neg logit contributions
 couple-0.006
rob-0.005
 seek-0.005
 gentleman-0.005
 trainer-0.004
Token furry
Token ID40857
Feature activation-2.00e-01

Pos logit contributions
 fears+0.018
aying+0.018
arse+0.017
 Ow+0.017
 wipes+0.017
Neg logit contributions
rob-0.018
 couple-0.016
robat-0.015
 seek-0.014
seek-0.014
Token,
Token ID11
Feature activation-2.02e-01

Pos logit contributions
aying+0.011
 fears+0.010
uning+0.010
arse+0.010
 Ow+0.010
Neg logit contributions
rob-0.007
 couple-0.007
 seek-0.006
 gentleman-0.006
thank-0.006
Token brown
Token ID7586
Feature activation-4.32e-01

Pos logit contributions
aying+0.012
arse+0.011
uning+0.011
 fears+0.011
ours+0.011
Neg logit contributions
 couple-0.007
 seek-0.007
rob-0.007
 lady-0.006
 man-0.006
Token bear
Token ID6842
Feature activation-7.11e-01

Pos logit contributions
aying+0.024
 fears+0.023
uning+0.023
arse+0.023
 Ow+0.023
Neg logit contributions
 couple-0.015
rob-0.015
 seek-0.013
 gentleman-0.012
thank-0.012
Token cub
Token ID13617
Feature activation-2.03e+00

Pos logit contributions
aying+0.030
 fears+0.028
uning+0.028
ist+0.027
aves+0.026
Neg logit contributions
rob-0.033
 couple-0.032
 seek-0.029
thank-0.028
 gentleman-0.028
Token.
Token ID13
Feature activation-3.01e-01

Pos logit contributions
aying+0.112
arse+0.106
 fears+0.105
uning+0.103
omed+0.102
Neg logit contributions
 couple-0.076
rob-0.072
 seek-0.064
 trainer-0.062
 gentleman-0.062
Token
Token ID198
Feature activation+7.54e-02

Pos logit contributions
aying+0.018
 fears+0.017
arse+0.017
uning+0.016
omed+0.016
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 lady-0.008
 gentleman-0.008
Token
Token ID198
Feature activation+6.22e-02

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.003
thank+0.003
 gentleman+0.003
Neg logit contributions
aying-0.004
 Ow-0.003
omed-0.003
 fears-0.003
arse-0.003
TokenOne
Token ID3198
Feature activation+6.15e-01

Pos logit contributions
 couple+0.002
rob+0.002
 seek+0.002
 lady+0.002
 gentleman+0.002
Neg logit contributions
aying-0.003
 fears-0.003
uning-0.003
arse-0.003
 fists-0.003
Token day
Token ID1110
Feature activation-3.02e-01

Pos logit contributions
rob+0.018
 couple+0.016
 gentleman+0.013
thank+0.013
 seek+0.013
Neg logit contributions
aying-0.039
 fears-0.037
 Ow-0.036
uning-0.035
omed-0.035
Token fox
Token ID21831
Feature activation-6.32e-01

Pos logit contributions
aying+0.044
arse+0.042
 fears+0.041
uning+0.041
omed+0.040
Neg logit contributions
 couple-0.021
rob-0.018
 lady-0.017
 seek-0.016
 man-0.016
Token living
Token ID2877
Feature activation-1.02e+00

Pos logit contributions
aying+0.032
 fears+0.031
uning+0.030
 Ow+0.029
omed+0.029
Neg logit contributions
 couple-0.025
rob-0.024
 seek-0.021
 gentleman-0.021
thank-0.020
Token in
Token ID287
Feature activation+2.25e-01

Pos logit contributions
 fears+0.042
aying+0.042
 Ow+0.041
adoes+0.040
 fists+0.039
Neg logit contributions
rob-0.050
 couple-0.050
 gentleman-0.045
 trainer-0.043
 seek-0.042
Token a
Token ID257
Feature activation+6.92e-01

Pos logit contributions
 couple+0.009
rob+0.009
 seek+0.008
 gentleman+0.008
thank+0.008
Neg logit contributions
aying-0.011
 fears-0.011
 Ow-0.010
arse-0.010
 fists-0.010
Token big
Token ID1263
Feature activation+2.42e-01

Pos logit contributions
rob+0.035
 couple+0.033
thank+0.031
 gentleman+0.030
robat+0.029
Neg logit contributions
 fears-0.028
aying-0.027
arse-0.027
 Ow-0.026
 wipes-0.026
Token chim
Token ID18205
Feature activation-2.02e+00

Pos logit contributions
rob+0.009
 couple+0.009
thank+0.008
 seek+0.007
 gentleman+0.007
Neg logit contributions
 fears-0.013
aying-0.013
arse-0.012
 wipes-0.012
 Ow-0.012
Tokenney
Token ID1681
Feature activation-1.34e+00

Pos logit contributions
aying+0.071
outed+0.065
icker+0.064
 fears+0.064
uning+0.063
Neg logit contributions
 couple-0.114
rob-0.109
 gentleman-0.102
 seek-0.099
 trainer-0.098
Token.
Token ID13
Feature activation-2.31e-01

Pos logit contributions
aying+0.080
arse+0.076
 fears+0.076
uning+0.075
 Ow+0.074
Neg logit contributions
 couple-0.043
rob-0.041
 seek-0.036
 gentleman-0.034
 lady-0.033
Token Every
Token ID3887
Feature activation+3.25e-01

Pos logit contributions
arse+0.012
aying+0.012
 fears+0.011
iqu+0.011
arding+0.011
Neg logit contributions
 couple-0.009
rob-0.008
 seek-0.008
 lady-0.008
 gentleman-0.008
Token day
Token ID1110
Feature activation-2.27e-01

Pos logit contributions
 couple+0.007
rob+0.006
 seek+0.005
 gentleman+0.004
 trainer+0.004
Neg logit contributions
aying-0.023
 fears-0.022
arse-0.022
uning-0.022
omed-0.022
Token,
Token ID11
Feature activation+9.86e-02

Pos logit contributions
aying+0.014
arse+0.014
 fears+0.014
uning+0.013
iqu+0.013
Neg logit contributions
 couple-0.007
rob-0.006
 seek-0.006
 lady-0.006
 man-0.006
Token big
Token ID1263
Feature activation-1.24e-01

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.003
 lady+0.002
 gentleman+0.002
Neg logit contributions
aying-0.004
 fears-0.004
arse-0.004
uning-0.004
omed-0.004
Token book
Token ID1492
Feature activation-4.35e-01

Pos logit contributions
aying+0.006
arse+0.006
 fears+0.006
uning+0.006
omed+0.006
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.005
 lady-0.004
 gentleman-0.004
Token with
Token ID351
Feature activation+6.85e-01

Pos logit contributions
aying+0.020
omed+0.019
uning+0.018
arse+0.018
ours+0.018
Neg logit contributions
 couple-0.020
rob-0.020
 seek-0.017
 gentleman-0.017
 trainer-0.016
Token pictures
Token ID5986
Feature activation-2.33e-01

Pos logit contributions
rob+0.035
 couple+0.034
 seek+0.032
 gentleman+0.031
robat+0.031
Neg logit contributions
aying-0.026
 fears-0.026
 Ow-0.026
 fists-0.025
 cleans-0.025
Token of
Token ID286
Feature activation-6.53e-01

Pos logit contributions
aying+0.012
omed+0.011
uning+0.010
 cleans+0.010
ours+0.010
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.008
 trainer-0.008
Token volcan
Token ID17516
Feature activation-2.02e+00

Pos logit contributions
aying+0.033
 fears+0.031
 Ow+0.030
 wipes+0.030
arse+0.030
Neg logit contributions
 couple-0.026
rob-0.026
 seek-0.023
 gentleman-0.023
thank-0.022
Tokenoes
Token ID3028
Feature activation-6.09e-01

Pos logit contributions
aying+0.075
arse+0.072
 fears+0.072
 wipes+0.071
 fists+0.069
Neg logit contributions
rob-0.105
 couple-0.103
 seek-0.092
 trainer-0.092
thank-0.091
Token.
Token ID13
Feature activation-5.16e-01

Pos logit contributions
aying+0.034
 fears+0.032
arse+0.032
uning+0.032
omed+0.031
Neg logit contributions
 couple-0.022
rob-0.021
 seek-0.018
 gentleman-0.017
 trainer-0.017
Token Vol
Token ID4709
Feature activation-4.82e-01

Pos logit contributions
arse+0.023
aying+0.023
 fears+0.023
uning+0.021
iqu+0.021
Neg logit contributions
 couple-0.025
rob-0.022
 lady-0.022
 gentleman-0.021
 man-0.021
Tokencan
Token ID5171
Feature activation-7.48e-01

Pos logit contributions
 fears+0.025
aying+0.025
 wipes+0.024
 fists+0.023
arse+0.023
Neg logit contributions
rob-0.019
 couple-0.018
 seek-0.016
 gentleman-0.016
 trainer-0.015
Tokenoes
Token ID3028
Feature activation-7.01e-01

Pos logit contributions
arse+0.037
 fears+0.037
aying+0.036
 fists+0.035
 wipes+0.035
Neg logit contributions
 couple-0.028
rob-0.027
 seek-0.027
 lady-0.026
 ranger-0.025
Token bear
Token ID6842
Feature activation-1.02e+00

Pos logit contributions
aying+0.030
 cleans+0.028
ist+0.028
uning+0.028
icking+0.027
Neg logit contributions
rob-0.041
 couple-0.039
 seek-0.039
 gentleman-0.036
robat-0.036
Token who
Token ID508
Feature activation-3.34e-01

Pos logit contributions
aying+0.052
uning+0.050
 fears+0.048
 cleans+0.048
ours+0.048
Neg logit contributions
rob-0.039
 couple-0.038
 seek-0.034
 gentleman-0.032
robat-0.032
Token had
Token ID550
Feature activation-5.66e-01

Pos logit contributions
aying+0.016
 fears+0.015
arse+0.015
uning+0.014
 Ow+0.014
Neg logit contributions
 couple-0.015
rob-0.014
 seek-0.013
 gentleman-0.013
 trainer-0.012
Token a
Token ID257
Feature activation-2.47e-02

Pos logit contributions
aying+0.028
 fears+0.027
arse+0.026
uning+0.026
 Ow+0.026
Neg logit contributions
 couple-0.023
rob-0.023
 seek-0.020
 gentleman-0.020
 trainer-0.019
Token little
Token ID1310
Feature activation-7.66e-01

Pos logit contributions
aying+0.001
omed+0.001
arse+0.001
uning+0.001
adoes+0.001
Neg logit contributions
 couple-0.001
 seek-0.001
 lady-0.001
rob-0.001
 man-0.001
Token cub
Token ID13617
Feature activation-2.01e+00

Pos logit contributions
aying+0.048
arse+0.045
omed+0.045
uning+0.044
 Ow+0.043
Neg logit contributions
 couple-0.025
rob-0.021
 lady-0.020
 seek-0.020
 gentleman-0.019
Token.
Token ID13
Feature activation-5.14e-01

Pos logit contributions
aying+0.114
arse+0.110
 fears+0.107
uning+0.105
 Ow+0.105
Neg logit contributions
 couple-0.074
rob-0.068
 seek-0.060
 trainer-0.059
 gentleman-0.059
Token One
Token ID1881
Feature activation+3.62e-01

Pos logit contributions
aying+0.026
 fears+0.026
arse+0.025
iqu+0.024
 fists+0.024
Neg logit contributions
 couple-0.023
rob-0.020
 lady-0.020
 seek-0.019
 gentleman-0.019
Token day
Token ID1110
Feature activation-3.03e-01

Pos logit contributions
rob+0.010
 couple+0.009
 seek+0.007
 gentleman+0.007
 trainer+0.007
Neg logit contributions
aying-0.024
 fears-0.023
 Ow-0.022
arse-0.022
uning-0.022
Token,
Token ID11
Feature activation-1.31e-01

Pos logit contributions
aying+0.021
 fears+0.020
arse+0.019
iqu+0.019
oken+0.019
Neg logit contributions
 couple-0.008
rob-0.006
 seek-0.006
 lady-0.006
 man-0.006
Token little
Token ID1310
Feature activation-2.53e-01

Pos logit contributions
aying+0.009
 fears+0.008
arse+0.008
uning+0.008
omed+0.008
Neg logit contributions
 couple-0.004
rob-0.003
 seek-0.003
 lady-0.003
 man-0.003
Token a
Token ID257
Feature activation+1.63e-01

Pos logit contributions
rob+0.000
 couple+0.000
 gentleman+0.000
 trainer+0.000
robat+0.000
Neg logit contributions
aying-0.000
 Ow-0.000
 fears-0.000
 cleans-0.000
 wipes-0.000
Token house
Token ID2156
Feature activation-8.36e-01

Pos logit contributions
rob+0.009
 couple+0.008
thank+0.008
 gentleman+0.008
robat+0.008
Neg logit contributions
 fears-0.006
aying-0.006
 Ow-0.006
 wipes-0.006
arse-0.006
Token with
Token ID351
Feature activation+6.49e-01

Pos logit contributions
aying+0.041
ours+0.037
uning+0.037
 fears+0.037
 wipes+0.036
Neg logit contributions
rob-0.036
 couple-0.034
robat-0.030
 gentleman-0.030
 trainer-0.029
Token a
Token ID257
Feature activation-2.90e-01

Pos logit contributions
rob+0.027
 couple+0.025
 gentleman+0.023
robat+0.022
 trainer+0.021
Neg logit contributions
aying-0.033
 Ow-0.032
 fears-0.031
 cleans-0.031
 wipes-0.031
Token big
Token ID1263
Feature activation-3.78e-01

Pos logit contributions
aying+0.016
arse+0.015
uning+0.015
 fears+0.015
omed+0.015
Neg logit contributions
 couple-0.012
rob-0.010
 lady-0.010
 seek-0.010
 man-0.010
Token chim
Token ID18205
Feature activation-2.01e+00

Pos logit contributions
aying+0.022
arse+0.020
omed+0.020
 fears+0.020
uning+0.020
Neg logit contributions
 couple-0.014
rob-0.013
 lady-0.012
 seek-0.012
 man-0.012
Tokenney
Token ID1681
Feature activation-1.51e+00

Pos logit contributions
aying+0.068
icker+0.063
uning+0.061
outed+0.060
akes+0.060
Neg logit contributions
 couple-0.115
rob-0.113
 gentleman-0.103
 seek-0.101
 trainer-0.101
Token.
Token ID13
Feature activation-4.46e-01

Pos logit contributions
aying+0.086
 fears+0.082
arse+0.081
uning+0.080
omed+0.079
Neg logit contributions
 couple-0.052
rob-0.050
 seek-0.044
 gentleman-0.042
 trainer-0.041
Token The
Token ID383
Feature activation-5.59e-01

Pos logit contributions
aying+0.022
arse+0.021
 fears+0.021
uning+0.020
 fists+0.020
Neg logit contributions
 couple-0.019
rob-0.018
 seek-0.017
 lady-0.017
 gentleman-0.016
Token chim
Token ID18205
Feature activation-1.47e+00

Pos logit contributions
aying+0.026
arse+0.025
 fears+0.025
omed+0.024
uning+0.024
Neg logit contributions
 couple-0.025
rob-0.024
 lady-0.022
 seek-0.022
 man-0.021
Tokenney
Token ID1681
Feature activation-9.51e-01

Pos logit contributions
aying+0.063
 fears+0.056
uning+0.056
omed+0.055
 Ow+0.055
Neg logit contributions
 couple-0.074
rob-0.071
 seek-0.065
 gentleman-0.064
 trainer-0.063
Token.
Token ID13
Feature activation-5.44e-01

Pos logit contributions
aying+0.025
uning+0.023
 fears+0.023
omed+0.022
arse+0.022
Neg logit contributions
 couple-0.019
rob-0.018
 seek-0.016
 gentleman-0.016
thank-0.015
Token They
Token ID1119
Feature activation+2.36e-01

Pos logit contributions
arse+0.032
aying+0.031
 fears+0.031
 fists+0.029
uning+0.029
Neg logit contributions
 couple-0.017
rob-0.015
 ranger-0.015
 lady-0.015
 seek-0.015
Token had
Token ID550
Feature activation-6.11e-03

Pos logit contributions
 couple+0.009
rob+0.009
 seek+0.008
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.013
 fears-0.012
arse-0.012
uning-0.011
omed-0.011
Token a
Token ID257
Feature activation+1.85e-01

Pos logit contributions
aying+0.000
arse+0.000
 fears+0.000
uning+0.000
omed+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 seek-0.000
 trainer-0.000
 lady-0.000
Token magn
Token ID7842
Feature activation-1.52e-01

Pos logit contributions
 couple+0.008
rob+0.008
 seek+0.007
 lady+0.007
 trainer+0.007
Neg logit contributions
aying-0.009
uning-0.008
arse-0.008
 fears-0.008
omed-0.008
Tokenifying
Token ID4035
Feature activation-2.01e+00

Pos logit contributions
 cleans+0.004
 fears+0.004
 wipes+0.004
 Acc+0.004
adoes+0.004
Neg logit contributions
 couple-0.009
rob-0.009
 trainer-0.008
robat-0.008
 king-0.008
Token glass
Token ID5405
Feature activation+1.81e-01

Pos logit contributions
aying+0.071
 fears+0.064
 Ow+0.064
uning+0.061
omed+0.061
Neg logit contributions
 couple-0.114
rob-0.112
 gentleman-0.102
 seek-0.101
 trainer-0.099
Token and
Token ID290
Feature activation+2.83e-01

Pos logit contributions
 couple+0.007
rob+0.007
 seek+0.006
 gentleman+0.006
thank+0.006
Neg logit contributions
aying-0.009
 fears-0.009
uning-0.009
arse-0.009
omed-0.009
Token a
Token ID257
Feature activation-5.81e-02

Pos logit contributions
 couple+0.011
rob+0.011
 seek+0.010
 gentleman+0.009
 lady+0.009
Neg logit contributions
aying-0.015
arse-0.014
 fears-0.014
uning-0.014
omed-0.014
Token small
Token ID1402
Feature activation-3.55e-02

Pos logit contributions
aying+0.003
arse+0.003
 fears+0.003
uning+0.003
omed+0.003
Neg logit contributions
 couple-0.002
rob-0.002
 seek-0.002
 gentleman-0.002
 lady-0.002
Token box
Token ID3091
Feature activation-7.82e-01

Pos logit contributions
aying+0.002
arse+0.002
 fears+0.002
uning+0.002
omed+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 lady-0.001
 gentleman-0.001
Token He
Token ID679
Feature activation+2.20e-01

Pos logit contributions
 couple+0.005
 gentleman+0.004
 seek+0.004
rob+0.004
 ranger+0.004
Neg logit contributions
arse-0.008
aying-0.008
 fears-0.008
arding-0.008
 cleans-0.008
Token found
Token ID1043
Feature activation-6.84e-02

Pos logit contributions
 couple+0.008
rob+0.008
 seek+0.007
 gentleman+0.007
 lady+0.006
Neg logit contributions
aying-0.012
arse-0.012
 fears-0.011
uning-0.011
 Ow-0.011
Token a
Token ID257
Feature activation+6.40e-01

Pos logit contributions
aying+0.004
arse+0.003
 fears+0.003
uning+0.003
omed+0.003
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.002
 gentleman-0.002
 trainer-0.002
Token big
Token ID1263
Feature activation+3.82e-01

Pos logit contributions
 couple+0.025
rob+0.024
 seek+0.022
 gentleman+0.021
 lady+0.021
Neg logit contributions
aying-0.034
 fears-0.032
arse-0.032
uning-0.031
omed-0.031
Token tree
Token ID5509
Feature activation-3.05e-01

Pos logit contributions
 couple+0.014
rob+0.013
 seek+0.012
 lady+0.012
 gentleman+0.012
Neg logit contributions
aying-0.022
arse-0.020
 fears-0.020
uning-0.020
 Ow-0.020
Token with
Token ID351
Feature activation+1.25e+00

Pos logit contributions
aying+0.016
 fears+0.015
arse+0.015
uning+0.015
omed+0.015
Neg logit contributions
 couple-0.012
rob-0.011
 seek-0.010
 gentleman-0.009
 trainer-0.009
Token an
Token ID281
Feature activation+2.16e-01

Pos logit contributions
 couple+0.052
rob+0.049
 seek+0.044
 gentleman+0.043
 lady+0.042
Neg logit contributions
aying-0.064
arse-0.060
 fears-0.060
uning-0.059
omed-0.058
Token old
Token ID1468
Feature activation+7.19e-01

Pos logit contributions
rob+0.010
 couple+0.009
 seek+0.008
 trainer+0.008
 gentleman+0.008
Neg logit contributions
aying-0.011
 fears-0.010
uning-0.010
arse-0.009
 fists-0.009
Token branch
Token ID8478
Feature activation+3.14e-02

Pos logit contributions
 couple+0.025
rob+0.024
 seek+0.022
 gentleman+0.021
 lady+0.020
Neg logit contributions
aying-0.041
 fears-0.038
arse-0.038
uning-0.038
omed-0.037
Token reaching
Token ID8978
Feature activation+3.11e-01

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 gentleman+0.001
 lady+0.001
Neg logit contributions
aying-0.002
 fears-0.002
arse-0.002
 Ow-0.002
uning-0.002
Token down
Token ID866
Feature activation-1.86e-01

Pos logit contributions
 couple+0.009
rob+0.008
 man+0.008
 seek+0.008
 lady+0.008
Neg logit contributions
aying-0.018
arse-0.017
 fears-0.017
 wipes-0.017
 cleans-0.017
Tokenâ
Token ID22940
Feature activation-2.31e-01

Pos logit contributions
aying+0.005
 fears+0.005
arse+0.005
uning+0.005
 fists+0.005
Neg logit contributions
 couple-0.004
rob-0.004
 seek-0.004
 gentleman-0.004
 lady-0.004
Token€
Token ID26391
Feature activation+7.57e-01

Pos logit contributions
aying+0.012
 fears+0.012
arse+0.012
uning+0.012
 Ow+0.012
Neg logit contributions
rob-0.009
 couple-0.009
 seek-0.008
 gentleman-0.007
thank-0.007
Token
Token ID198
Feature activation+1.13e-01

Pos logit contributions
 couple+0.029
rob+0.028
 seek+0.026
 gentleman+0.024
 lady+0.024
Neg logit contributions
aying-0.041
arse-0.039
 fears-0.039
uning-0.038
 fists-0.037
Token
Token ID198
Feature activation+2.40e-01

Pos logit contributions
 couple+0.004
rob+0.004
 seek+0.003
 gentleman+0.003
 lady+0.003
Neg logit contributions
aying-0.007
 fears-0.006
arse-0.006
uning-0.006
adoes-0.006
TokenThe
Token ID464
Feature activation+1.67e-02

Pos logit contributions
 couple+0.009
rob+0.009
 seek+0.008
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.013
 fears-0.012
arse-0.012
uning-0.012
omed-0.012
Token old
Token ID1468
Feature activation+1.34e+00

Pos logit contributions
rob+0.000
 couple+0.000
 seek+0.000
robat+0.000
 gentleman+0.000
Neg logit contributions
aying-0.001
 fears-0.001
 wipes-0.001
uning-0.001
 cleans-0.001
Token lady
Token ID10846
Feature activation+1.19e+00

Pos logit contributions
rob+0.034
 couple+0.023
robat+0.022
amer+0.020
cest+0.019
Neg logit contributions
 wipes-0.088
 cleans-0.086
aying-0.084
ours-0.081
uning-0.081
Token was
Token ID373
Feature activation-2.49e-02

Pos logit contributions
 couple+0.039
rob+0.034
 seek+0.032
 lady+0.032
 trainer+0.031
Neg logit contributions
aying-0.072
arse-0.070
 Ow-0.068
 fears-0.068
uning-0.068
Token astonished
Token ID40962
Feature activation+2.66e-01

Pos logit contributions
aying+0.001
 fears+0.001
arse+0.001
omed+0.001
 fists+0.001
Neg logit contributions
rob-0.001
 couple-0.001
 seek-0.001
robat-0.001
 gentleman-0.001
Token at
Token ID379
Feature activation-2.25e-01

Pos logit contributions
 couple+0.009
rob+0.007
 seek+0.007
 lady+0.007
 gentleman+0.007
Neg logit contributions
aying-0.016
arse-0.015
 fears-0.015
 Ow-0.015
uning-0.015
Token Daisy
Token ID40355
Feature activation-5.88e-02

Pos logit contributions
aying+0.012
arse+0.011
omed+0.011
 fears+0.011
uning+0.010
Neg logit contributions
 couple-0.009
rob-0.008
 lady-0.008
 seek-0.008
 gentleman-0.008
Token mom
Token ID1995
Feature activation+5.69e-01

Pos logit contributions
 couple+0.009
rob+0.009
 seek+0.008
 gentleman+0.008
 lady+0.008
Neg logit contributions
aying-0.011
arse-0.010
 fears-0.010
uning-0.010
omed-0.010
Token holding
Token ID4769
Feature activation-9.41e-02

Pos logit contributions
 couple+0.019
rob+0.017
 seek+0.016
 lady+0.015
 gentleman+0.015
Neg logit contributions
aying-0.034
 fears-0.032
 Ow-0.032
uning-0.031
arse-0.031
Token a
Token ID257
Feature activation+2.44e-01

Pos logit contributions
aying+0.006
arse+0.006
uning+0.005
 fears+0.005
omed+0.005
Neg logit contributions
 couple-0.003
rob-0.003
 lady-0.002
 seek-0.002
 gentleman-0.002
Token big
Token ID1263
Feature activation+1.29e-01

Pos logit contributions
 couple+0.009
rob+0.008
 lady+0.008
 seek+0.008
 man+0.008
Neg logit contributions
aying-0.014
uning-0.013
arse-0.013
 fears-0.013
 Ow-0.012
Token file
Token ID2393
Feature activation+1.63e-01

Pos logit contributions
 couple+0.005
rob+0.004
 seek+0.004
 lady+0.004
 gentleman+0.004
Neg logit contributions
aying-0.007
arse-0.007
 fears-0.007
uning-0.007
 Ow-0.007
Token with
Token ID351
Feature activation+1.20e+00

Pos logit contributions
rob+0.007
 couple+0.006
 seek+0.006
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.008
arse-0.008
 fears-0.008
omed-0.008
uning-0.008
Token many
Token ID867
Feature activation+6.10e-01

Pos logit contributions
 couple+0.047
rob+0.043
 seek+0.040
 lady+0.038
 gentleman+0.038
Neg logit contributions
aying-0.065
arse-0.061
 fears-0.060
uning-0.059
omed-0.058
Token papers
Token ID9473
Feature activation+7.09e-01

Pos logit contributions
 couple+0.026
rob+0.024
 seek+0.022
 lady+0.022
 gentleman+0.022
Neg logit contributions
aying-0.031
arse-0.029
uning-0.028
 Ow-0.028
 fears-0.028
Token.
Token ID13
Feature activation-9.86e-02

Pos logit contributions
rob+0.030
 couple+0.028
 seek+0.025
 gentleman+0.025
 trainer+0.024
Neg logit contributions
aying-0.036
omed-0.033
uning-0.032
arse-0.032
 fears-0.032
Token
Token ID198
Feature activation+2.78e-01

Pos logit contributions
aying+0.006
arse+0.006
 fears+0.006
uning+0.005
ist+0.005
Neg logit contributions
 couple-0.003
 lady-0.003
rob-0.003
 gentleman-0.003
 man-0.003
Token
Token ID198
Feature activation+2.09e-01

Pos logit contributions
 couple+0.010
rob+0.010
 seek+0.008
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.016
arse-0.015
 fears-0.014
omed-0.014
 Ow-0.014
Token.
Token ID13
Feature activation+2.35e-01

Pos logit contributions
rob+0.008
 couple+0.008
 seek+0.007
 gentleman+0.007
 trainer+0.007
Neg logit contributions
aying-0.011
 fears-0.011
arse-0.011
omed-0.011
uning-0.010
Token
Token ID198
Feature activation-4.33e-02

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.008
 gentleman+0.008
 trainer+0.008
Neg logit contributions
aying-0.012
 fears-0.011
arse-0.011
uning-0.011
omed-0.011
Token
Token ID198
Feature activation+1.61e-02

Pos logit contributions
aying+0.002
 fears+0.002
arse+0.002
 Ow+0.002
omed+0.002
Neg logit contributions
 couple-0.002
rob-0.002
 seek-0.002
 gentleman-0.002
 trainer-0.002
TokenAfter
Token ID3260
Feature activation+9.26e-01

Pos logit contributions
rob+0.001
 couple+0.001
 gentleman+0.001
 seek+0.001
 trainer+0.001
Neg logit contributions
aying-0.001
arse-0.001
 fears-0.001
omed-0.001
 Ow-0.001
Token a
Token ID257
Feature activation+1.18e+00

Pos logit contributions
rob+0.038
 couple+0.037
 seek+0.033
 gentleman+0.033
thank+0.032
Neg logit contributions
aying-0.046
 fears-0.043
omed-0.042
arse-0.042
uning-0.042
Token few
Token ID1178
Feature activation+1.46e+00

Pos logit contributions
rob+0.055
 couple+0.048
 gentleman+0.045
thank+0.045
robat+0.044
Neg logit contributions
arse-0.052
 fears-0.050
aying-0.050
omed-0.050
 Ow-0.047
Token hours
Token ID2250
Feature activation+1.04e+00

Pos logit contributions
 couple+0.047
rob+0.045
 seek+0.039
 gentleman+0.039
thank+0.037
Neg logit contributions
aying-0.086
 fears-0.083
omed-0.080
arse-0.080
 wipes-0.080
Token,
Token ID11
Feature activation+4.89e-01

Pos logit contributions
rob+0.044
 couple+0.042
 seek+0.040
 gentleman+0.038
 trainer+0.038
Neg logit contributions
aying-0.050
omed-0.047
uning-0.046
arse-0.045
 cleans-0.045
Token the
Token ID262
Feature activation+1.52e-01

Pos logit contributions
 couple+0.017
rob+0.017
 seek+0.015
 gentleman+0.014
 trainer+0.014
Neg logit contributions
aying-0.027
 fears-0.026
arse-0.025
uning-0.025
 Ow-0.025
Token witch
Token ID16365
Feature activation+8.34e-01

Pos logit contributions
rob+0.006
 couple+0.004
robat+0.004
 gentleman+0.004
 seek+0.004
Neg logit contributions
 fears-0.008
aying-0.008
 wipes-0.008
 cleans-0.008
oken-0.008
Token stirred
Token ID33091
Feature activation+1.30e-02

Pos logit contributions
rob+0.036
 couple+0.034
 seek+0.030
 gentleman+0.030
robat+0.029
Neg logit contributions
aying-0.040
 fears-0.038
omed-0.037
 cleans-0.037
arse-0.037
Token there
Token ID612
Feature activation-2.93e-01

Pos logit contributions
aying+0.023
 fears+0.023
arse+0.022
uning+0.022
adoes+0.022
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.010
 lady-0.008
 gentleman-0.008
Token was
Token ID373
Feature activation+4.26e-03

Pos logit contributions
aying+0.017
 fears+0.016
arse+0.016
uning+0.016
omed+0.016
Neg logit contributions
 couple-0.010
rob-0.009
 seek-0.008
 gentleman-0.008
 lady-0.008
Token a
Token ID257
Feature activation-1.97e-01

Pos logit contributions
 couple+0.000
rob+0.000
 seek+0.000
 gentleman+0.000
 trainer+0.000
Neg logit contributions
aying-0.000
 Ow-0.000
 fears-0.000
arse-0.000
uning-0.000
Token city
Token ID1748
Feature activation-8.10e-01

Pos logit contributions
aying+0.008
 fears+0.008
arse+0.008
uning+0.008
 Ow+0.008
Neg logit contributions
 couple-0.009
rob-0.009
 seek-0.008
 gentleman-0.008
thank-0.008
Token filled
Token ID5901
Feature activation+1.49e-01

Pos logit contributions
aying+0.039
omed+0.036
uning+0.035
ours+0.035
 fears+0.035
Neg logit contributions
rob-0.033
 couple-0.031
 seek-0.028
robat-0.028
 gentleman-0.027
Token with
Token ID351
Feature activation+1.43e+00

Pos logit contributions
rob+0.007
 couple+0.007
 gentleman+0.006
robat+0.006
 trainer+0.006
Neg logit contributions
aying-0.006
adoes-0.006
 Ow-0.005
ours-0.005
icker-0.005
Token joy
Token ID8716
Feature activation-4.74e-01

Pos logit contributions
rob+0.062
 couple+0.059
 seek+0.055
 gentleman+0.053
robat+0.052
Neg logit contributions
aying-0.067
 fears-0.064
 Ow-0.061
uning-0.061
arse-0.061
Token.
Token ID13
Feature activation-3.73e-01

Pos logit contributions
aying+0.028
 fears+0.027
arse+0.027
uning+0.026
 Ow+0.026
Neg logit contributions
 couple-0.016
rob-0.014
 seek-0.013
 lady-0.012
 gentleman-0.012
Token Everyone
Token ID11075
Feature activation-5.26e-01

Pos logit contributions
arse+0.019
 fears+0.019
aying+0.018
iqu+0.018
icking+0.017
Neg logit contributions
 couple-0.017
 seek-0.015
 lady-0.015
rob-0.014
 gentleman-0.014
Token was
Token ID373
Feature activation-6.04e-01

Pos logit contributions
aying+0.028
arse+0.027
 fears+0.026
uning+0.026
omed+0.026
Neg logit contributions
 couple-0.021
rob-0.020
 seek-0.018
 gentleman-0.017
 trainer-0.017
Token happy
Token ID3772
Feature activation-5.95e-01

Pos logit contributions
aying+0.029
 fears+0.026
uning+0.026
ours+0.025
ains+0.025
Neg logit contributions
rob-0.028
robat-0.024
 couple-0.023
 seek-0.022
 gentleman-0.022
Token her
Token ID607
Feature activation+8.75e-02

Pos logit contributions
 couple+0.041
rob+0.039
 seek+0.036
 gentleman+0.034
 trainer+0.034
Neg logit contributions
aying-0.047
 fears-0.044
arse-0.043
uning-0.043
omed-0.042
Token paper
Token ID3348
Feature activation+5.31e-01

Pos logit contributions
 couple+0.004
rob+0.004
 seek+0.003
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.004
 fears-0.004
arse-0.004
uning-0.004
 Ow-0.004
Token and
Token ID290
Feature activation+1.95e-01

Pos logit contributions
 couple+0.020
rob+0.020
 seek+0.017
 gentleman+0.017
 trainer+0.016
Neg logit contributions
aying-0.029
 fears-0.027
arse-0.027
uning-0.026
omed-0.026
Token stamped
Token ID28305
Feature activation+9.79e-02

Pos logit contributions
rob+0.010
 couple+0.009
 gentleman+0.009
robat+0.009
 seek+0.009
Neg logit contributions
 cleans-0.007
 wipes-0.007
 fears-0.007
 fists-0.007
aying-0.007
Token them
Token ID606
Feature activation-8.04e-02

Pos logit contributions
rob+0.004
 couple+0.004
 seek+0.004
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.005
 fears-0.005
arse-0.004
uning-0.004
 Ow-0.004
Token with
Token ID351
Feature activation+3.78e-01

Pos logit contributions
aying+0.005
omed+0.005
 Ow+0.005
ours+0.005
uning+0.005
Neg logit contributions
 couple-0.002
 seek-0.002
rob-0.002
 lady-0.002
 man-0.002
Token the
Token ID262
Feature activation-1.53e-01

Pos logit contributions
 couple+0.015
rob+0.014
 seek+0.013
 lady+0.013
 gentleman+0.013
Neg logit contributions
aying-0.020
arse-0.018
 fears-0.018
uning-0.018
omed-0.018
Token heart
Token ID2612
Feature activation+3.20e-01

Pos logit contributions
aying+0.008
 fears+0.007
arse+0.007
uning+0.007
omed+0.007
Neg logit contributions
 couple-0.007
rob-0.006
 seek-0.006
 gentleman-0.006
 lady-0.006
Token.
Token ID13
Feature activation+2.17e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.008
 lady+0.008
 trainer+0.008
Neg logit contributions
aying-0.019
arse-0.019
 fears-0.018
 Ow-0.018
uning-0.018
Token She
Token ID1375
Feature activation-2.62e-01

Pos logit contributions
 couple+0.008
 lady+0.007
rob+0.007
 gentleman+0.007
thank+0.007
Neg logit contributions
arse-0.012
aying-0.012
 fears-0.011
omed-0.011
ist-0.011
Token wrote
Token ID2630
Feature activation-1.73e-02

Pos logit contributions
aying+0.013
 fears+0.013
 wipes+0.012
 cleans+0.012
uning+0.012
Neg logit contributions
 couple-0.011
rob-0.010
 seek-0.009
 gentleman-0.009
 trainer-0.009
Token He
Token ID679
Feature activation+1.53e-01

Pos logit contributions
 fears+0.046
aying+0.046
arse+0.046
arding+0.043
iqu+0.043
Neg logit contributions
rob-0.027
 seek-0.026
 couple-0.026
 ranger-0.025
thank-0.024
Token loved
Token ID6151
Feature activation-7.76e-02

Pos logit contributions
 couple+0.007
rob+0.007
 seek+0.006
 gentleman+0.006
 trainer+0.006
Neg logit contributions
aying-0.007
 fears-0.007
arse-0.007
uning-0.007
 Ow-0.007
Token to
Token ID284
Feature activation-1.29e-01

Pos logit contributions
aying+0.004
 fears+0.004
uning+0.004
arse+0.004
 Ow+0.004
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 gentleman-0.003
 trainer-0.002
Token go
Token ID467
Feature activation-5.91e-02

Pos logit contributions
aying+0.008
omed+0.008
arse+0.008
outed+0.008
 fears+0.008
Neg logit contributions
 couple-0.004
 seek-0.004
rob-0.003
 lady-0.003
 par-0.003
Token for
Token ID329
Feature activation+5.00e-01

Pos logit contributions
aying+0.003
uning+0.003
arse+0.003
 fears+0.003
omed+0.003
Neg logit contributions
 couple-0.003
rob-0.002
 gentleman-0.002
 seek-0.002
 trainer-0.002
Token walks
Token ID11114
Feature activation+4.01e-01

Pos logit contributions
 couple+0.023
rob+0.022
 seek+0.021
 gentleman+0.020
 trainer+0.020
Neg logit contributions
aying-0.022
 fears-0.021
arse-0.021
uning-0.020
 Ow-0.020
Token.
Token ID13
Feature activation-3.65e-01

Pos logit contributions
 couple+0.015
rob+0.015
 gentleman+0.013
 seek+0.013
 trainer+0.012
Neg logit contributions
aying-0.021
 fears-0.020
arse-0.020
uning-0.020
omed-0.019
Token One
Token ID1881
Feature activation+3.17e-01

Pos logit contributions
aying+0.018
 fears+0.017
arse+0.017
adoes+0.017
omed+0.017
Neg logit contributions
 couple-0.015
rob-0.015
 seek-0.013
 gentleman-0.013
 lady-0.013
Token day
Token ID1110
Feature activation-6.17e-01

Pos logit contributions
rob+0.009
 couple+0.008
 seek+0.007
 trainer+0.007
 gentleman+0.007
Neg logit contributions
aying-0.020
 fears-0.019
 Ow-0.018
uning-0.018
arse-0.018
Token,
Token ID11
Feature activation-3.04e-01

Pos logit contributions
aying+0.037
 fears+0.036
arse+0.035
uning+0.034
omed+0.034
Neg logit contributions
 couple-0.020
rob-0.019
 seek-0.017
 gentleman-0.016
 lady-0.016
Token Tim
Token ID5045
Feature activation-1.46e-01

Pos logit contributions
aying+0.018
 fears+0.017
arse+0.017
uning+0.016
omed+0.016
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.008
 lady-0.008
Token
Token ID198
Feature activation-1.28e-01

Pos logit contributions
aying+0.005
 Ow+0.005
 fears+0.004
omed+0.004
arse+0.004
Neg logit contributions
 couple-0.004
rob-0.004
 gentleman-0.004
 seek-0.004
 businessman-0.004
TokenAfter
Token ID3260
Feature activation+9.81e-01

Pos logit contributions
aying+0.006
 fears+0.005
 Ow+0.005
arse+0.005
omed+0.005
Neg logit contributions
 couple-0.006
rob-0.006
 gentleman-0.005
 trainer-0.005
 seek-0.005
Token playing
Token ID2712
Feature activation+1.09e+00

Pos logit contributions
 couple+0.043
rob+0.043
 gentleman+0.039
 seek+0.038
thank+0.038
Neg logit contributions
aying-0.044
omed-0.041
uning-0.041
 Ow-0.041
 fears-0.041
Token for
Token ID329
Feature activation+1.19e+00

Pos logit contributions
 couple+0.046
rob+0.044
 gentleman+0.039
 seek+0.038
 trainer+0.038
Neg logit contributions
aying-0.055
 fears-0.051
uning-0.050
adoes-0.050
 wipes-0.050
Token a
Token ID257
Feature activation+9.94e-01

Pos logit contributions
 couple+0.057
rob+0.055
 gentleman+0.053
 trainer+0.050
thank+0.050
Neg logit contributions
aying-0.048
 Ow-0.046
 fears-0.046
 cleans-0.045
oken-0.044
Token while
Token ID981
Feature activation+1.38e-01

Pos logit contributions
rob+0.041
 couple+0.040
 seek+0.035
 gentleman+0.035
thank+0.034
Neg logit contributions
aying-0.048
arse-0.047
 fears-0.046
omed-0.045
 Ow-0.044
Token,
Token ID11
Feature activation-3.42e-02

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.005
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.007
 fears-0.007
arse-0.006
uning-0.006
omed-0.006
Token Tim
Token ID5045
Feature activation-5.19e-02

Pos logit contributions
aying+0.002
 fears+0.002
arse+0.002
omed+0.002
uning+0.002
Neg logit contributions
 couple-0.001
rob-0.001
 seek-0.001
 lady-0.001
 man-0.001
Token started
Token ID2067
Feature activation+5.00e-01

Pos logit contributions
aying+0.002
 fears+0.002
arse+0.002
 cleans+0.002
omed+0.002
Neg logit contributions
 couple-0.002
rob-0.002
 gentleman-0.002
 seek-0.002
robat-0.002
Token to
Token ID284
Feature activation+2.19e-01

Pos logit contributions
 couple+0.020
rob+0.020
 seek+0.017
 gentleman+0.016
 trainer+0.016
Neg logit contributions
aying-0.026
 fears-0.025
arse-0.024
uning-0.024
 Ow-0.024
Token feel
Token ID1254
Feature activation+3.50e-01

Pos logit contributions
 couple+0.010
 seek+0.010
rob+0.010
 king+0.009
 lady+0.009
Neg logit contributions
aying-0.010
omed-0.009
ours-0.009
uning-0.009
outed-0.009
Token.
Token ID13
Feature activation+1.27e-01

Pos logit contributions
aying+0.017
arse+0.017
 Ow+0.016
omed+0.016
uning+0.016
Neg logit contributions
 couple-0.010
 seek-0.009
 lady-0.009
rob-0.008
 man-0.008
Token The
Token ID383
Feature activation+1.56e-01

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.004
 gentleman+0.004
 lady+0.004
Neg logit contributions
aying-0.007
arse-0.006
 fears-0.006
uning-0.006
omed-0.006
Token end
Token ID886
Feature activation+6.26e-01

Pos logit contributions
rob+0.007
 couple+0.006
 seek+0.006
robat+0.005
 gentleman+0.005
Neg logit contributions
aying-0.008
 fears-0.008
 cleans-0.007
 wipes-0.007
arse-0.007
Token.
Token ID13
Feature activation+2.31e-01

Pos logit contributions
 couple+0.025
rob+0.024
 seek+0.020
thank+0.020
 gentleman+0.020
Neg logit contributions
aying-0.033
 fears-0.031
omed-0.031
arse-0.030
uning-0.030
Token<|endoftext|>
Token ID50256
Feature activation-2.74e-01

Pos logit contributions
 couple+0.009
rob+0.008
 seek+0.008
 trainer+0.007
 gentleman+0.007
Neg logit contributions
aying-0.013
 fears-0.012
arse-0.012
 Ow-0.012
omed-0.011
TokenOnce
Token ID7454
Feature activation+4.18e-01

Pos logit contributions
aying+0.014
 Ow+0.013
omed+0.013
arse+0.013
 fears+0.013
Neg logit contributions
 couple-0.011
rob-0.010
 seek-0.009
 gentleman-0.009
 trainer-0.009
Token upon
Token ID2402
Feature activation-1.14e-01

Pos logit contributions
 couple+0.021
rob+0.020
 seek+0.018
 gentleman+0.018
 lady+0.018
Neg logit contributions
aying-0.017
 fears-0.016
arse-0.016
uning-0.016
omed-0.015
Token a
Token ID257
Feature activation+4.42e-01

Pos logit contributions
aying+0.005
 Ow+0.005
 fears+0.005
arse+0.005
 wipes+0.005
Neg logit contributions
rob-0.005
 couple-0.005
 gentleman-0.005
 seek-0.004
 trainer-0.004
Token time
Token ID640
Feature activation-3.93e-01

Pos logit contributions
 couple+0.019
rob+0.018
 seek+0.017
 lady+0.016
 gentleman+0.016
Neg logit contributions
uning-0.021
aying-0.021
adoes-0.020
arse-0.020
ours-0.019
Token there
Token ID612
Feature activation-3.84e-01

Pos logit contributions
aying+0.024
 fears+0.024
arse+0.023
uning+0.023
adoes+0.022
Neg logit contributions
 couple-0.011
rob-0.011
 seek-0.010
 lady-0.009
 gentleman-0.009
Token were
Token ID547
Feature activation+3.37e-01

Pos logit contributions
aying+0.022
 fears+0.021
arse+0.021
uning+0.021
omed+0.020
Neg logit contributions
 couple-0.013
rob-0.013
 seek-0.011
 gentleman-0.011
 lady-0.010
Token away
Token ID1497
Feature activation+4.18e-02

Pos logit contributions
aying+0.002
uning+0.002
 fears+0.002
omed+0.002
adoes+0.002
Neg logit contributions
 couple-0.002
rob-0.002
 gentleman-0.002
 trainer-0.002
 seek-0.002
Token.
Token ID13
Feature activation-2.38e-01

Pos logit contributions
 couple+0.002
rob+0.001
 seek+0.001
 gentleman+0.001
 trainer+0.001
Neg logit contributions
aying-0.002
 fears-0.002
arse-0.002
uning-0.002
omed-0.002
Token One
Token ID1881
Feature activation+1.44e-01

Pos logit contributions
iqu+0.013
arse+0.013
 fears+0.012
icking+0.012
arding+0.012
Neg logit contributions
 couple-0.009
rob-0.008
 lady-0.008
 seek-0.008
thank-0.008
Token day
Token ID1110
Feature activation-3.03e-01

Pos logit contributions
rob+0.004
 couple+0.003
 seek+0.003
 gentleman+0.003
 trainer+0.003
Neg logit contributions
aying-0.010
 fears-0.009
uning-0.009
 Ow-0.009
arse-0.009
Token it
Token ID340
Feature activation-1.80e-01

Pos logit contributions
aying+0.023
 fears+0.022
arse+0.022
iqu+0.022
oken+0.021
Neg logit contributions
 couple-0.006
 man-0.005
 seek-0.005
 lady-0.005
rob-0.004
Token stored
Token ID8574
Feature activation+2.12e-01

Pos logit contributions
aying+0.008
 cleans+0.008
 wipes+0.007
ours+0.007
outed+0.007
Neg logit contributions
rob-0.009
 couple-0.008
robat-0.007
 gentleman-0.007
 trainer-0.007
Token a
Token ID257
Feature activation+3.18e-01

Pos logit contributions
 couple+0.008
rob+0.008
 seek+0.007
 gentleman+0.007
 lady+0.007
Neg logit contributions
aying-0.011
arse-0.010
 fears-0.010
uning-0.010
omed-0.010
Token toy
Token ID13373
Feature activation+5.28e-01

Pos logit contributions
 couple+0.013
rob+0.013
 seek+0.012
 gentleman+0.011
 trainer+0.011
Neg logit contributions
aying-0.016
 fears-0.015
arse-0.015
uning-0.014
omed-0.014
Token plane
Token ID6614
Feature activation-2.05e-01

Pos logit contributions
rob+0.022
 couple+0.022
 seek+0.020
 gentleman+0.019
thank+0.018
Neg logit contributions
aying-0.025
uning-0.024
 fears-0.024
 Ow-0.023
adoes-0.023
Token,
Token ID11
Feature activation+2.78e-01

Pos logit contributions
aying+0.010
 fears+0.010
arse+0.009
uning+0.009
omed+0.009
Neg logit contributions
 couple-0.009
rob-0.008
 seek-0.007
 gentleman-0.007
 trainer-0.007
Token a
Token ID257
Feature activation-3.44e-01

Pos logit contributions
 couple+0.008
 seek+0.007
 man+0.007
 lady+0.007
 par+0.006
Neg logit contributions
aying-0.018
arse-0.018
omed-0.018
 fears-0.018
uning-0.017
Token small
Token ID1402
Feature activation-3.97e-01

Pos logit contributions
rob+0.025
 couple+0.024
thank+0.022
 gentleman+0.022
 seek+0.021
Neg logit contributions
aying-0.019
 fears-0.019
arse-0.018
 Ow-0.018
 wipes-0.018
Token house
Token ID2156
Feature activation-8.04e-01

Pos logit contributions
aying+0.019
 fears+0.019
arse+0.018
 Ow+0.018
 wipes+0.018
Neg logit contributions
rob-0.016
 couple-0.016
 seek-0.014
thank-0.014
 gentleman-0.014
Token with
Token ID351
Feature activation+7.56e-01

Pos logit contributions
aying+0.045
 fears+0.041
uning+0.041
arse+0.041
ours+0.041
Neg logit contributions
 couple-0.029
rob-0.028
 gentleman-0.024
 seek-0.023
 trainer-0.023
Token his
Token ID465
Feature activation+6.72e-02

Pos logit contributions
 couple+0.030
rob+0.030
 gentleman+0.025
 seek+0.025
thank+0.024
Neg logit contributions
aying-0.039
 Ow-0.038
 fears-0.037
uning-0.036
 wipes-0.036
Token best
Token ID1266
Feature activation-1.27e+00

Pos logit contributions
rob+0.002
 couple+0.002
 seek+0.002
 gentleman+0.002
thank+0.002
Neg logit contributions
aying-0.004
 fears-0.004
uning-0.003
arse-0.003
 Ow-0.003
Token friend
Token ID1545
Feature activation-2.44e-01

Pos logit contributions
aying+0.078
arse+0.074
omed+0.073
uning+0.071
 fears+0.071
Neg logit contributions
 couple-0.042
rob-0.034
 seek-0.034
 lady-0.033
 trainer-0.032
Token,
Token ID11
Feature activation+1.93e-01

Pos logit contributions
arse+0.017
aying+0.016
 fears+0.016
iqu+0.016
isel+0.016
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.005
 lady-0.005
 king-0.004
Token a
Token ID257
Feature activation-5.29e-01

Pos logit contributions
 couple+0.006
 seek+0.006
rob+0.006
 lady+0.005
 ranger+0.005
Neg logit contributions
aying-0.012
arse-0.011
 fears-0.011
uning-0.011
arding-0.011
Token little
Token ID1310
Feature activation-1.53e+00

Pos logit contributions
aying+0.028
 fears+0.026
arse+0.026
uning+0.026
omed+0.025
Neg logit contributions
 couple-0.022
rob-0.020
 seek-0.019
 lady-0.019
 gentleman-0.018
Token boy
Token ID2933
Feature activation-1.63e-01

Pos logit contributions
aying+0.107
arse+0.102
omed+0.100
 fears+0.098
uning+0.098
Neg logit contributions
 couple-0.040
rob-0.032
 lady-0.032
 seek-0.030
 man-0.030
Token named
Token ID3706
Feature activation-6.66e-01

Pos logit contributions
aying+0.009
 fears+0.008
uning+0.008
 cleans+0.008
 Ow+0.008
Neg logit contributions
 couple-0.006
rob-0.006
 gentleman-0.005
 seek-0.005
thank-0.005
Token.
Token ID13
Feature activation+3.86e-02

Pos logit contributions
 couple+0.004
rob+0.003
 seek+0.003
 lady+0.003
thank+0.003
Neg logit contributions
aying-0.009
arse-0.009
 fears-0.009
 Ow-0.008
uning-0.008
Token They
Token ID1119
Feature activation-1.80e-01

Pos logit contributions
 couple+0.002
rob+0.002
 lady+0.002
 seek+0.002
 gentleman+0.002
Neg logit contributions
arse-0.002
aying-0.002
 fears-0.002
uning-0.002
iqu-0.002
Token mixed
Token ID7668
Feature activation+8.39e-02

Pos logit contributions
aying+0.009
 fears+0.008
arse+0.008
uning+0.008
omed+0.008
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 gentleman-0.007
 trainer-0.006
Token them
Token ID606
Feature activation-6.82e-01

Pos logit contributions
 couple+0.003
rob+0.003
 seek+0.003
thank+0.003
 lady+0.003
Neg logit contributions
aying-0.004
arse-0.004
 fears-0.004
uning-0.004
omed-0.004
Token with
Token ID351
Feature activation+5.44e-01

Pos logit contributions
aying+0.041
arse+0.041
 fears+0.040
omed+0.038
iqu+0.038
Neg logit contributions
 couple-0.022
 seek-0.019
rob-0.018
thank-0.018
 l-0.018
Token a
Token ID257
Feature activation-6.98e-02

Pos logit contributions
 couple+0.019
 l+0.016
thank+0.015
rob+0.015
 par+0.015
Neg logit contributions
aying-0.032
arse-0.032
uning-0.030
 fears-0.030
omed-0.030
Token spoon
Token ID24556
Feature activation-3.25e-01

Pos logit contributions
aying+0.004
uning+0.003
adoes+0.003
 fears+0.003
arse+0.003
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 lady-0.003
 man-0.003
Token.
Token ID13
Feature activation+8.83e-02

Pos logit contributions
aying+0.018
 fears+0.017
arse+0.017
uning+0.016
omed+0.016
Neg logit contributions
 couple-0.012
rob-0.012
 seek-0.010
 gentleman-0.010
 trainer-0.010
Token They
Token ID1119
Feature activation-3.15e-01

Pos logit contributions
 couple+0.003
rob+0.003
 lady+0.003
 seek+0.003
 ranger+0.003
Neg logit contributions
arse-0.005
 fears-0.005
aying-0.005
iqu-0.005
 fists-0.005
Token said
Token ID531
Feature activation+2.43e-01

Pos logit contributions
aying+0.016
 fears+0.015
arse+0.015
uning+0.014
omed+0.014
Neg logit contributions
 couple-0.013
rob-0.013
 seek-0.012
 gentleman-0.011
 trainer-0.011
Token,
Token ID11
Feature activation+2.03e-01

Pos logit contributions
 couple+0.007
 lady+0.005
 man+0.005
 trainer+0.005
 par+0.005
Neg logit contributions
arse-0.017
aying-0.016
iqu-0.016
 fears-0.016
ist-0.015
Token Sarah
Token ID10490
Feature activation+2.61e-01

Pos logit contributions
 couple+0.012
rob+0.012
 seek+0.010
 gentleman+0.010
 trainer+0.010
Neg logit contributions
aying-0.019
 fears-0.018
arse-0.017
uning-0.017
 Ow-0.017
Token,
Token ID11
Feature activation-2.91e-01

Pos logit contributions
 couple+0.007
 lady+0.006
 man+0.006
 gentleman+0.005
 ranger+0.005
Neg logit contributions
aying-0.017
arse-0.017
 fears-0.016
 Ow-0.016
oken-0.016
Token "
Token ID366
Feature activation-3.96e-01

Pos logit contributions
aying+0.017
 fears+0.017
arse+0.016
uning+0.016
 Ow+0.016
Neg logit contributions
 couple-0.010
rob-0.008
 lady-0.008
 gentleman-0.008
 seek-0.008
TokenIf
Token ID1532
Feature activation+4.41e-01

Pos logit contributions
aying+0.019
arse+0.017
 fears+0.017
uning+0.017
omed+0.017
Neg logit contributions
 couple-0.018
rob-0.017
 seek-0.015
 gentleman-0.015
 trainer-0.015
Token you
Token ID345
Feature activation-1.01e-01

Pos logit contributions
 couple+0.012
 seek+0.011
 lady+0.010
 man+0.009
rob+0.009
Neg logit contributions
aying-0.029
arse-0.028
omed-0.027
uning-0.027
 fears-0.027
Token can
Token ID460
Feature activation-1.89e-01

Pos logit contributions
adoes+0.005
arse+0.005
ours+0.005
aying+0.005
 fists+0.005
Neg logit contributions
 seek-0.004
 couple-0.004
 trainer-0.003
 king-0.003
thank-0.003
Token guess
Token ID4724
Feature activation+1.25e-01

Pos logit contributions
aying+0.010
 fears+0.009
adoes+0.009
arse+0.009
omed+0.009
Neg logit contributions
 couple-0.008
 seek-0.007
rob-0.007
 trainer-0.006
 lady-0.006
Token what
Token ID644
Feature activation-1.30e-01

Pos logit contributions
 couple+0.005
 l+0.005
 seek+0.005
rob+0.005
 trainer+0.004
Neg logit contributions
arse-0.006
aying-0.006
outed-0.006
omed-0.006
oken-0.006
Token's
Token ID338
Feature activation-4.97e-01

Pos logit contributions
arse+0.007
aying+0.007
omed+0.006
 fears+0.006
uning+0.006
Neg logit contributions
 couple-0.005
rob-0.005
 seek-0.005
 lady-0.004
 man-0.004
Token inside
Token ID2641
Feature activation+1.75e-01

Pos logit contributions
aying+0.027
arse+0.026
omed+0.026
 fears+0.025
 Ow+0.025
Neg logit contributions
 couple-0.018
rob-0.017
 seek-0.017
 lady-0.016
 trainer-0.016
Token the
Token ID262
Feature activation-2.10e-01

Pos logit contributions
 couple+0.006
rob+0.005
 seek+0.005
 lady+0.005
 man+0.005
Neg logit contributions
aying-0.010
arse-0.010
 fears-0.010
uning-0.010
omed-0.009
Token time
Token ID640
Feature activation-4.78e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.009
 lady+0.008
 trainer+0.008
Neg logit contributions
aying-0.010
uning-0.010
adoes-0.010
arse-0.009
ours-0.009
Token,
Token ID11
Feature activation+1.65e-01

Pos logit contributions
aying+0.027
 fears+0.025
arse+0.025
uning+0.025
 wipes+0.024
Neg logit contributions
 couple-0.017
rob-0.016
 seek-0.015
 gentleman-0.014
 lady-0.014
Token there
Token ID612
Feature activation-7.60e-02

Pos logit contributions
 couple+0.006
rob+0.006
 seek+0.006
 lady+0.005
 man+0.005
Neg logit contributions
aying-0.009
 fears-0.008
uning-0.008
 wipes-0.008
arse-0.008
Token was
Token ID373
Feature activation-7.22e-02

Pos logit contributions
aying+0.004
 Ow+0.004
 fears+0.003
arse+0.003
 cleans+0.003
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 gentleman-0.003
 trainer-0.003
Token a
Token ID257
Feature activation-3.02e-01

Pos logit contributions
aying+0.003
 Ow+0.003
 fears+0.003
arse+0.003
 wipes+0.003
Neg logit contributions
rob-0.003
 couple-0.003
 gentleman-0.003
 seek-0.003
 trainer-0.003
Token little
Token ID1310
Feature activation-8.86e-01

Pos logit contributions
aying+0.011
 fears+0.011
arse+0.011
 Ow+0.010
 wipes+0.010
Neg logit contributions
rob-0.016
 couple-0.015
 seek-0.014
thank-0.014
 gentleman-0.014
Token boy
Token ID2933
Feature activation-1.96e-01

Pos logit contributions
aying+0.056
 fears+0.053
arse+0.053
uning+0.052
 Ow+0.051
Neg logit contributions
 couple-0.025
rob-0.024
 seek-0.021
 gentleman-0.019
 trainer-0.019
Token named
Token ID3706
Feature activation-3.47e-01

Pos logit contributions
aying+0.010
 fears+0.009
 cleans+0.009
 wipes+0.009
uning+0.009
Neg logit contributions
 couple-0.008
rob-0.008
 gentleman-0.007
 seek-0.007
 trainer-0.006
Token Tim
Token ID5045
Feature activation-1.52e-01

Pos logit contributions
aying+0.016
 fears+0.015
uning+0.014
arse+0.014
omed+0.014
Neg logit contributions
 couple-0.016
rob-0.016
 seek-0.015
 lady-0.014
 trainer-0.014
Tokenmy
Token ID1820
Feature activation-2.83e-01

Pos logit contributions
aying+0.006
 fears+0.006
arse+0.006
 Ow+0.005
uning+0.005
Neg logit contributions
 couple-0.008
rob-0.008
 seek-0.007
 gentleman-0.007
 trainer-0.007
Token.
Token ID13
Feature activation-6.20e-01

Pos logit contributions
aying+0.013
 fears+0.012
 Ow+0.012
arse+0.012
 cleans+0.012
Neg logit contributions
 couple-0.013
rob-0.013
 seek-0.012
 gentleman-0.011
 trainer-0.011
Token a
Token ID257
Feature activation+4.37e-01

Pos logit contributions
 couple+0.008
 seek+0.007
rob+0.007
 lady+0.007
 gentleman+0.006
Neg logit contributions
aying-0.015
uning-0.015
arse-0.014
 fears-0.014
omed-0.014
Token great
Token ID1049
Feature activation+8.84e-02

Pos logit contributions
 couple+0.020
rob+0.019
 seek+0.018
 lady+0.017
 gentleman+0.017
Neg logit contributions
aying-0.020
uning-0.019
arse-0.019
 fears-0.019
omed-0.018
Token job
Token ID1693
Feature activation+8.37e-01

Pos logit contributions
rob+0.004
 couple+0.004
 seek+0.004
 trainer+0.003
 gentleman+0.003
Neg logit contributions
aying-0.004
 fears-0.004
uning-0.004
 Ow-0.004
arse-0.004
Token,
Token ID11
Feature activation-9.07e-01

Pos logit contributions
 couple+0.030
rob+0.029
 seek+0.026
 gentleman+0.024
 lady+0.024
Neg logit contributions
aying-0.047
 fears-0.045
arse-0.044
uning-0.044
 Ow-0.043
Token Lily
Token ID20037
Feature activation+4.69e-01

Pos logit contributions
aying+0.050
arse+0.048
uning+0.047
 fears+0.047
omed+0.046
Neg logit contributions
 couple-0.034
rob-0.031
 seek-0.030
 lady-0.029
 gentleman-0.027
Token.
Token ID13
Feature activation-3.06e-01

Pos logit contributions
 couple+0.019
rob+0.019
 seek+0.016
 gentleman+0.016
 trainer+0.016
Neg logit contributions
aying-0.024
 fears-0.022
arse-0.022
 cleans-0.021
 wipes-0.021
Token I
Token ID314
Feature activation+1.21e-01

Pos logit contributions
 fears+0.015
aying+0.015
arse+0.015
 fists+0.014
omed+0.014
Neg logit contributions
 couple-0.012
rob-0.012
 seek-0.011
 gentleman-0.011
thank-0.010
Token'm
Token ID1101
Feature activation-9.23e-02

Pos logit contributions
 couple+0.006
rob+0.006
 gentleman+0.005
 trainer+0.005
 seek+0.005
Neg logit contributions
aying-0.005
 fears-0.005
 cleans-0.005
 wipes-0.005
omed-0.005
Token happy
Token ID3772
Feature activation-1.65e-01

Pos logit contributions
aying+0.005
 ok+0.004
ours+0.004
arse+0.004
 cleans+0.004
Neg logit contributions
rob-0.004
 couple-0.004
robat-0.004
 seek-0.004
 trainer-0.003
Token you
Token ID345
Feature activation-9.57e-03

Pos logit contributions
aying+0.009
 fears+0.008
arse+0.008
uning+0.008
omed+0.008
Neg logit contributions
 couple-0.006
rob-0.006
 seek-0.005
 gentleman-0.005
 trainer-0.005
Token're
Token ID821
Feature activation-1.41e-01

Pos logit contributions
aying+0.000
 fears+0.000
arse+0.000
omed+0.000
uning+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 gentleman-0.000
 seek-0.000
 trainer-0.000
Token it
Token ID340
Feature activation+1.33e-01

Pos logit contributions
aying+0.006
uning+0.006
arse+0.006
omed+0.006
 fears+0.005
Neg logit contributions
 couple-0.004
rob-0.003
 seek-0.003
 lady-0.003
 trainer-0.003
Token!
Token ID0
Feature activation-3.56e-01

Pos logit contributions
 couple+0.006
rob+0.005
 gentleman+0.005
 seek+0.004
 trainer+0.004
Neg logit contributions
aying-0.007
arse-0.006
 fears-0.006
 Ow-0.006
uning-0.006
Token We
Token ID775
Feature activation+3.19e-01

Pos logit contributions
arse+0.017
aying+0.017
 fears+0.017
omed+0.016
uning+0.016
Neg logit contributions
 couple-0.015
rob-0.014
 seek-0.013
 trainer-0.013
 gentleman-0.013
Token have
Token ID423
Feature activation-4.44e-01

Pos logit contributions
 couple+0.012
rob+0.012
 seek+0.012
thank+0.011
 trainer+0.011
Neg logit contributions
aying-0.016
arse-0.016
adoes-0.016
uning-0.015
 fears-0.015
Token the
Token ID262
Feature activation-4.54e-01

Pos logit contributions
aying+0.025
arse+0.024
 fears+0.023
uning+0.023
omed+0.023
Neg logit contributions
 couple-0.017
rob-0.015
 seek-0.014
 lady-0.013
thank-0.013
Token best
Token ID1266
Feature activation-1.55e+00

Pos logit contributions
aying+0.025
omed+0.024
arse+0.024
ours+0.023
arding+0.023
Neg logit contributions
 couple-0.016
 king-0.015
 lady-0.015
rob-0.014
 seek-0.014
Token tower
Token ID10580
Feature activation-2.65e-01

Pos logit contributions
aying+0.072
 fears+0.067
arse+0.066
uning+0.065
omed+0.064
Neg logit contributions
 couple-0.071
rob-0.069
 seek-0.063
 gentleman-0.061
 trainer-0.060
Token!"
Token ID2474
Feature activation+3.54e-02

Pos logit contributions
aying+0.016
 fears+0.015
 cleans+0.015
adoes+0.014
omed+0.014
Neg logit contributions
 couple-0.008
rob-0.008
 gentleman-0.007
 seek-0.006
 trainer-0.006
Token They
Token ID1119
Feature activation-4.55e-01

Pos logit contributions
 couple+0.001
rob+0.001
 lady+0.001
 man+0.001
 trainer+0.001
Neg logit contributions
arse-0.002
aying-0.002
uning-0.002
 fears-0.002
iqu-0.002
Token see
Token ID766
Feature activation-1.92e-01

Pos logit contributions
aying+0.024
adoes+0.023
arse+0.023
uning+0.023
 fears+0.023
Neg logit contributions
 couple-0.017
 lady-0.016
 seek-0.016
 trainer-0.015
rob-0.015
Token Lily
Token ID20037
Feature activation-2.34e-01

Pos logit contributions
aying+0.012
arse+0.012
omed+0.011
uning+0.011
 fears+0.011
Neg logit contributions
 couple-0.006
rob-0.006
 lady-0.005
 man-0.005
 trainer-0.005
Token always
Token ID1464
Feature activation+2.85e-01

Pos logit contributions
 couple+0.021
 seek+0.019
 lady+0.018
thank+0.018
 king+0.018
Neg logit contributions
omed-0.039
aying-0.037
 fists-0.037
ours-0.037
arse-0.037
Token be
Token ID307
Feature activation-6.67e-01

Pos logit contributions
 couple+0.011
rob+0.011
 seek+0.010
 gentleman+0.010
 trainer+0.009
Neg logit contributions
aying-0.015
arse-0.014
 fears-0.014
omed-0.014
uning-0.014
Token kind
Token ID1611
Feature activation-3.10e-01

Pos logit contributions
aying+0.039
arse+0.037
 fears+0.036
uning+0.036
 Ow+0.036
Neg logit contributions
 couple-0.023
rob-0.021
 seek-0.020
 lady-0.019
 gentleman-0.018
Token and
Token ID290
Feature activation-2.40e-01

Pos logit contributions
aying+0.018
arse+0.018
omed+0.017
 Ow+0.017
 fears+0.017
Neg logit contributions
 couple-0.010
rob-0.008
 seek-0.008
 lady-0.008
 gentleman-0.008
Token think
Token ID892
Feature activation-6.00e-01

Pos logit contributions
aying+0.012
 fears+0.011
arse+0.011
uning+0.011
omed+0.011
Neg logit contributions
 couple-0.010
rob-0.010
 seek-0.009
 gentleman-0.009
 trainer-0.009
Token of
Token ID286
Feature activation-1.44e+00

Pos logit contributions
arse+0.037
 Ow+0.036
 fears+0.034
uning+0.034
aying+0.034
Neg logit contributions
 couple-0.018
 man-0.015
 lady-0.015
 knowing-0.014
 king-0.014
Token others
Token ID1854
Feature activation-1.11e-01

Pos logit contributions
omed+0.079
aying+0.077
arse+0.076
outed+0.074
uning+0.074
Neg logit contributions
 couple-0.050
rob-0.045
 man-0.044
 lady-0.043
 knowing-0.042
Token before
Token ID878
Feature activation+2.06e-01

Pos logit contributions
arse+0.006
 Ow+0.006
aying+0.006
uning+0.006
omed+0.006
Neg logit contributions
 couple-0.003
 man-0.003
rob-0.003
 seek-0.003
 lady-0.003
Token we
Token ID356
Feature activation-6.86e-02

Pos logit contributions
 couple+0.008
 seek+0.006
rob+0.006
 man+0.006
 lady+0.006
Neg logit contributions
aying-0.011
uning-0.011
omed-0.011
arse-0.011
 fears-0.011
Token think
Token ID892
Feature activation-6.27e-01

Pos logit contributions
aying+0.003
arse+0.003
omed+0.003
adoes+0.003
ours+0.003
Neg logit contributions
 couple-0.003
 seek-0.003
 trainer-0.003
thank-0.003
rob-0.003
Token of
Token ID286
Feature activation-1.02e+00

Pos logit contributions
arse+0.037
 Ow+0.036
 fears+0.036
aying+0.036
uning+0.036
Neg logit contributions
 couple-0.020
 man-0.016
 lady-0.016
rob-0.016
 seek-0.016
Token appetite
Token ID20788
Feature activation-3.74e-01

Pos logit contributions
aying+0.004
arse+0.004
 fears+0.004
omed+0.004
uning+0.004
Neg logit contributions
 couple-0.003
rob-0.003
 seek-0.003
 gentleman-0.003
 lady-0.003
Token and
Token ID290
Feature activation+1.12e-01

Pos logit contributions
aying+0.019
uning+0.018
 fears+0.017
ows+0.017
icking+0.017
Neg logit contributions
rob-0.016
 couple-0.015
 seek-0.013
 gentleman-0.013
thank-0.012
Token lived
Token ID5615
Feature activation-5.08e-02

Pos logit contributions
 couple+0.005
rob+0.005
robat+0.005
 gentleman+0.005
thank+0.005
Neg logit contributions
aying-0.004
 fears-0.004
 cleans-0.004
 wipes-0.004
adoes-0.004
Token in
Token ID287
Feature activation+1.89e-02

Pos logit contributions
aying+0.002
 fears+0.002
uning+0.002
 Ow+0.002
outed+0.002
Neg logit contributions
 couple-0.003
rob-0.003
 gentleman-0.002
robat-0.002
 businessman-0.002
Token a
Token ID257
Feature activation+1.99e-01

Pos logit contributions
 couple+0.001
rob+0.001
 gentleman+0.001
 trainer+0.001
thank+0.001
Neg logit contributions
aying-0.001
 fears-0.001
 Ow-0.001
 fists-0.001
adoes-0.001
Token swamp
Token ID26837
Feature activation-1.41e+00

Pos logit contributions
rob+0.010
 couple+0.010
thank+0.009
 gentleman+0.009
robat+0.008
Neg logit contributions
 fears-0.008
aying-0.008
arse-0.007
 Ow-0.007
 wipes-0.007
Token in
Token ID287
Feature activation-1.69e-04

Pos logit contributions
aying+0.081
 fears+0.077
arse+0.077
uning+0.075
omed+0.075
Neg logit contributions
 couple-0.048
rob-0.046
 seek-0.041
 gentleman-0.039
 trainer-0.038
Token the
Token ID262
Feature activation+3.68e-01

Pos logit contributions
aying+0.000
 Ow+0.000
 fears+0.000
adoes+0.000
 cleans+0.000
Neg logit contributions
 couple-0.000
rob-0.000
 gentleman-0.000
 trainer-0.000
thank-0.000
Token middle
Token ID3504
Feature activation+2.03e-02

Pos logit contributions
rob+0.017
 couple+0.016
 gentleman+0.014
robat+0.014
thank+0.013
Neg logit contributions
aying-0.017
 fears-0.017
 Ow-0.016
arse-0.015
 wipes-0.015
Token of
Token ID286
Feature activation+3.62e-01

Pos logit contributions
rob+0.001
 couple+0.001
 gentleman+0.001
thank+0.001
 trainer+0.001
Neg logit contributions
aying-0.001
 fears-0.001
omed-0.001
uning-0.001
ours-0.001
Token the
Token ID262
Feature activation+3.36e-01

Pos logit contributions
 couple+0.014
rob+0.013
 seek+0.012
 gentleman+0.012
 trainer+0.011
Neg logit contributions
aying-0.019
 fears-0.018
 Ow-0.018
arse-0.018
uning-0.018
Token his
Token ID465
Feature activation-2.12e-01

Pos logit contributions
aying+0.016
arse+0.015
 fears+0.015
uning+0.015
omed+0.014
Neg logit contributions
 couple-0.006
 seek-0.005
rob-0.005
 l-0.004
 lady-0.004
Token best
Token ID1266
Feature activation-9.48e-01

Pos logit contributions
aying+0.012
omed+0.011
arse+0.011
uning+0.011
 Ow+0.011
Neg logit contributions
 couple-0.008
 seek-0.007
rob-0.007
 man-0.006
 lady-0.006
Token to
Token ID284
Feature activation-4.92e-01

Pos logit contributions
aying+0.054
arse+0.051
 fears+0.050
uning+0.050
omed+0.049
Neg logit contributions
 couple-0.034
rob-0.032
 seek-0.029
 gentleman-0.028
 lady-0.027
Token get
Token ID651
Feature activation-5.87e-01

Pos logit contributions
aying+0.028
omed+0.027
uning+0.027
ains+0.027
ours+0.027
Neg logit contributions
 seek-0.018
 couple-0.017
 lady-0.015
thank-0.014
 l-0.013
Token rid
Token ID5755
Feature activation-3.57e-01

Pos logit contributions
aying+0.036
arse+0.035
 fears+0.034
uning+0.033
 Ow+0.033
Neg logit contributions
 couple-0.018
rob-0.016
 seek-0.016
 lady-0.015
 l-0.015
Token of
Token ID286
Feature activation-1.20e+00

Pos logit contributions
arse+0.024
 fears+0.024
iqu+0.024
ours+0.024
uning+0.023
Neg logit contributions
 couple-0.006
rob-0.005
 seek-0.005
 lady-0.005
thank-0.005
Token the
Token ID262
Feature activation-3.56e-01

Pos logit contributions
aying+0.074
arse+0.071
uning+0.069
 fears+0.069
omed+0.069
Neg logit contributions
 couple-0.037
rob-0.034
 seek-0.031
 lady-0.031
 gentleman-0.029
Token big
Token ID1263
Feature activation-1.15e-01

Pos logit contributions
aying+0.021
arse+0.020
uning+0.019
 Ow+0.019
 fears+0.018
Neg logit contributions
 couple-0.014
 lady-0.012
 man-0.012
 seek-0.012
rob-0.012
Token spot
Token ID4136
Feature activation+4.42e-02

Pos logit contributions
aying+0.007
arse+0.006
uning+0.006
 fears+0.006
 Ow+0.006
Neg logit contributions
 couple-0.004
rob-0.004
 seek-0.004
 lady-0.004
 gentleman-0.004
Token.
Token ID13
Feature activation-1.40e-01

Pos logit contributions
 couple+0.001
rob+0.001
 seek+0.001
 lady+0.001
 gentleman+0.001
Neg logit contributions
aying-0.003
arse-0.003
 fears-0.003
 Ow-0.002
uning-0.002
Token Lucy
Token ID22162
Feature activation-2.86e-01

Pos logit contributions
aying+0.007
arse+0.007
 fears+0.007
uning+0.006
arding+0.006
Neg logit contributions
 couple-0.006
rob-0.005
 lady-0.005
 seek-0.005
 gentleman-0.005
Token colorful
Token ID20239
Feature activation+5.31e-01

Pos logit contributions
rob+0.034
 couple+0.034
 seek+0.031
thank+0.030
 gentleman+0.030
Neg logit contributions
aying-0.042
 fears-0.041
omed-0.039
 wipes-0.039
arse-0.039
Token cand
Token ID2658
Feature activation+9.81e-02

Pos logit contributions
 couple+0.021
rob+0.021
 seek+0.019
 gentleman+0.018
thank+0.018
Neg logit contributions
aying-0.027
 fears-0.025
uning-0.025
omed-0.025
arse-0.025
Tokenies
Token ID444
Feature activation-4.86e-01

Pos logit contributions
 couple+0.005
rob+0.005
 seek+0.005
 gentleman+0.005
 trainer+0.005
Neg logit contributions
aying-0.004
 fears-0.003
arse-0.003
uning-0.003
omed-0.003
Token and
Token ID290
Feature activation+5.34e-02

Pos logit contributions
aying+0.028
 fears+0.026
arse+0.026
uning+0.026
 Ow+0.026
Neg logit contributions
 couple-0.017
rob-0.016
 seek-0.014
 gentleman-0.014
 trainer-0.013
Token ch
Token ID442
Feature activation-6.30e-01

Pos logit contributions
rob+0.002
robat+0.002
 couple+0.002
 gentleman+0.002
 seek+0.002
Neg logit contributions
 wipes-0.002
 fears-0.002
 fists-0.002
 cleans-0.002
aying-0.002
Tokenocol
Token ID4668
Feature activation-1.30e+00

Pos logit contributions
arse+0.031
aying+0.029
 cleans+0.028
 wipes+0.028
 Ow+0.027
Neg logit contributions
 couple-0.026
rob-0.022
 trainer-0.022
thank-0.020
 par-0.020
Tokenates
Token ID689
Feature activation-8.17e-01

Pos logit contributions
aying+0.057
uning+0.054
 Ow+0.051
omed+0.051
arse+0.050
Neg logit contributions
 couple-0.064
rob-0.058
 seek-0.058
 gentleman-0.057
 lady-0.056
Token.
Token ID13
Feature activation+1.76e-01

Pos logit contributions
aying+0.049
 fears+0.047
arse+0.046
 Ow+0.045
uning+0.045
Neg logit contributions
 couple-0.027
rob-0.024
 seek-0.022
 lady-0.021
 trainer-0.021
Token She
Token ID1375
Feature activation-1.97e-01

Pos logit contributions
 couple+0.007
 lady+0.006
 gentleman+0.006
rob+0.005
 man+0.005
Neg logit contributions
aying-0.010
arse-0.010
 fears-0.010
omed-0.009
oken-0.009
Token said
Token ID531
Feature activation+2.99e-01

Pos logit contributions
aying+0.009
 fears+0.009
outed+0.009
 wipes+0.009
ours+0.009
Neg logit contributions
rob-0.008
 couple-0.008
robat-0.007
 seek-0.007
 gentleman-0.007
Token to
Token ID284
Feature activation+2.56e-01

Pos logit contributions
 couple+0.010
rob+0.009
 seek+0.008
 lady+0.008
 gentleman+0.008
Neg logit contributions
aying-0.018
arse-0.017
 fears-0.017
uning-0.016
omed-0.016