TOP ACTIVATIONS
MAX = 4.985

with his family . The do ggy was very happy and
Alice decided to grab some pe aches from the tree in
sund ae with lots of topp ings . Tim my was
a big tree with many pe aches . The pe aches
The chip m unk was excited and realized
and Tommy always shared their pe aches and never got stubborn
he could spot some animals pe eking out from the banks
sure to always share his pe aches with his new friend
, Lily always kept a pe b ble in her pocket
back and find more pretty pe b bles and learn more
a big tree with many pe aches . They wanted to
They also see some fancy topp ings . They see sprink
I am a magical pe b ble and I â
I 'll put some o int ment on it ," said
. She loved to eat pe aches because they were so
play outside and collect pretty pe b bles . One day
all helped Beth put more pe b bles in the water
play too ! M ittens ' owner loved her
sund ae with lots of topp ings . They
together . It will be higher and nicer than our towers

MIN ACTIVATIONS
MIN = -5.564

away and disappeared in the alley . The t eddy bear
two cats fighting in the alley way . The cats were
shiny metal box in an alley way . She went over
she pe ered into the alley . There was a big
one she met in the alley that day . <|endoftext|> Once
she spotted it in an alley . But when
for tresp assing on his property , but when he heard
went back to the gloomy alley again . <|endoftext|> Once upon
She drove away to the charity shop . L
grandma 's house for safe keeping . When
noise coming from a nearby alley way . Ell a was
. She ran behind the laundry . She thought Ben could
. They belong to a charity that helps poor kids .
. It belongs to the charity . We provide clothes and
hole and found a dirty alley with trash and rats .
She was stuck behind the laundry . Ben laughed
Tom was walking down an alley and he heard a loud
to take them to the charity shop , where other children
of people coming onto my property . I suggest you leave
in circles and ran into alley ways , but nothing looked

INTERVAL 2.348 - 4.985
CONTAINS 0.953%

soft ! <|endoftext|> Once upon a time there was a little
the store and bought some fish . He took them home
happy . And so the tired root was sold
quiz !" <|endoftext|> Once upon a time there was a small
girl ! <|endoftext|> Once upon a time there was a curious

INTERVAL -0.290 - 2.348
CONTAINS 61.620%

groom them . One day while she was at the stable
owed in surprise . " How can I eat a zero
so happy and he played with his new toy truck and
was so excited to see what was in it ! She
the mask stopped running and became quiet . It turned around

INTERVAL -2.927 - -0.290
CONTAINS 37.413%

toys . You can share and take turns . That is
always shared their toys and had lots of fun playing together
, then disappeared just as quickly as it had appeared .
he remembered that Sam was his friend . He knew that
ia nods . She takes the ball . She throws it

INTERVAL -5.564 - -2.927
CONTAINS 0.015%

. The vet showed Wh isk ers the x - ray
As Daisy walked through the tunnel , she suddenly heard a
She had copied the wrong lecture ! It was a bad
she was feeling very gr umpy . She put the things
. He loved his cereal because it was always reliable to
Token with
Token ID351
Feature activation+4.54e-01

Pos logit contributions
keeping+0.050
handed+0.044
 unatt+0.044
taker+0.043
 rubble+0.039
Neg logit contributions
Sal-0.024
 ho-0.020
 hus-0.020
 chicken-0.020
 stretch-0.020
Token his
Token ID465
Feature activation-6.82e-01

Pos logit contributions
Sal+0.007
 ho+0.006
 chicken+0.006
oxy+0.006
 hus+0.006
Neg logit contributions
keeping-0.016
handed-0.014
 unatt-0.014
taker-0.013
 rubble-0.012
Token family
Token ID1641
Feature activation-1.65e+00

Pos logit contributions
keeping+0.021
handed+0.019
 unatt+0.018
ocular+0.017
taker+0.016
Neg logit contributions
Sal-0.012
 hus-0.011
ita-0.011
-0.010
oxy-0.010
Token.
Token ID13
Feature activation-8.84e-01

Pos logit contributions
keeping+0.057
handed+0.052
 unatt+0.051
taker+0.049
 rubble+0.045
Neg logit contributions
Sal-0.025
 ho-0.021
ita-0.021
oxy-0.021
 hus-0.021
Token The
Token ID383
Feature activation+1.73e+00

Pos logit contributions
keeping+0.024
handed+0.021
taker+0.019
 unatt+0.019
ocular+0.019
Neg logit contributions
Sal-0.017
ita-0.017
 ho-0.017
 hus-0.016
oxy-0.016
Token do
Token ID466
Feature activation+4.85e+00

Pos logit contributions
Sal+0.025
 ho+0.021
ita+0.021
oxy+0.021
 hus+0.021
Neg logit contributions
keeping-0.060
handed-0.055
 unatt-0.054
taker-0.052
 rubble-0.048
Tokenggy
Token ID19970
Feature activation+4.12e-01

Pos logit contributions
Sal+0.095
 hus+0.092
 ho+0.091
+0.088
 corn+0.087
Neg logit contributions
keeping-0.132
handed-0.123
taker-0.113
 unatt-0.111
eway-0.105
Token was
Token ID373
Feature activation-2.60e-01

Pos logit contributions
Sal+0.008
 ho+0.007
 hus+0.007
oxy+0.007
 stretch+0.007
Neg logit contributions
keeping-0.012
handed-0.011
taker-0.010
 unatt-0.010
ocular-0.009
Token very
Token ID845
Feature activation+4.93e-01

Pos logit contributions
keeping+0.008
handed+0.008
 unatt+0.008
taker+0.007
 rubble+0.007
Neg logit contributions
Sal-0.004
oxy-0.004
 ho-0.004
ita-0.004
 hus-0.004
Token happy
Token ID3772
Feature activation+5.34e-01

Pos logit contributions
Sal+0.010
oxy+0.009
ita+0.009
+0.009
 ho+0.009
Neg logit contributions
keeping-0.014
taker-0.013
handed-0.013
 unatt-0.012
 rubble-0.011
Token and
Token ID290
Feature activation+1.16e-01

Pos logit contributions
Sal+0.008
 chicken+0.007
 ho+0.007
 stretch+0.007
 each+0.007
Neg logit contributions
keeping-0.018
handed-0.017
 unatt-0.016
taker-0.016
 rubble-0.015
Token Alice
Token ID14862
Feature activation-8.83e-02

Pos logit contributions
keeping+0.079
handed+0.075
 unatt+0.074
taker+0.068
eway+0.064
Neg logit contributions
 long-0.032
 short-0.032
Sal-0.031
 be-0.031
 each-0.031
Token decided
Token ID3066
Feature activation-4.14e-01

Pos logit contributions
keeping+0.003
handed+0.003
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.002
ita-0.001
 ho-0.001
 hus-0.001
Le-0.001
Token to
Token ID284
Feature activation+7.93e-01

Pos logit contributions
keeping+0.014
handed+0.012
 unatt+0.012
taker+0.012
 rubble+0.011
Neg logit contributions
Sal-0.007
 ho-0.006
oxy-0.006
ita-0.006
 hus-0.006
Token grab
Token ID5552
Feature activation+1.16e+00

Pos logit contributions
Sal+0.013
 ho+0.011
ita+0.011
oxy+0.011
 hus+0.011
Neg logit contributions
keeping-0.026
handed-0.024
 unatt-0.023
taker-0.023
 rubble-0.021
Token some
Token ID617
Feature activation+8.57e-01

Pos logit contributions
Sal+0.015
 ho+0.012
ita+0.012
oxy+0.012
 hus+0.012
Neg logit contributions
keeping-0.043
handed-0.039
 unatt-0.039
taker-0.038
 rubble-0.035
Token pe
Token ID613
Feature activation+4.82e+00

Pos logit contributions
Sal+0.018
oxy+0.016
ita+0.016
Le+0.016
 ho+0.016
Neg logit contributions
keeping-0.024
handed-0.022
 unatt-0.021
taker-0.020
 rubble-0.019
Tokenaches
Token ID3694
Feature activation+1.93e+00

Pos logit contributions
Sal+0.113
 ho+0.102
oxy+0.102
ita+0.101
 chicken+0.101
Neg logit contributions
keeping-0.123
handed-0.109
 unatt-0.106
taker-0.103
ocular-0.089
Token from
Token ID422
Feature activation-1.63e+00

Pos logit contributions
Sal+0.033
 ho+0.028
ita+0.028
oxy+0.028
 hus+0.027
Neg logit contributions
keeping-0.063
handed-0.057
 unatt-0.056
taker-0.054
 rubble-0.049
Token the
Token ID262
Feature activation-7.06e-01

Pos logit contributions
keeping+0.061
handed+0.056
 unatt+0.056
taker+0.055
 rubble+0.050
Neg logit contributions
Sal-0.019
 ho-0.016
 chicken-0.016
oxy-0.015
ita-0.015
Token tree
Token ID5509
Feature activation+1.14e+00

Pos logit contributions
keeping+0.021
handed+0.018
 unatt+0.018
taker+0.018
 rubble+0.016
Neg logit contributions
Sal-0.014
 chicken-0.013
 ho-0.013
 hus-0.012
oxy-0.012
Token in
Token ID287
Feature activation-4.01e-01

Pos logit contributions
Sal+0.017
 ho+0.015
ita+0.014
oxy+0.014
 hus+0.014
Neg logit contributions
keeping-0.039
handed-0.035
 unatt-0.035
taker-0.034
 rubble-0.031
Token sund
Token ID37437
Feature activation+1.38e+00

Pos logit contributions
Sal+0.022
ita+0.020
oxy+0.019
Le+0.019
 hus+0.019
Neg logit contributions
keeping-0.034
 unatt-0.031
handed-0.031
taker-0.029
ocular-0.026
Tokenae
Token ID3609
Feature activation+1.37e+00

Pos logit contributions
Sal+0.024
 hus+0.024
 chicken+0.021
 corn+0.021
 ho+0.020
Neg logit contributions
keeping-0.044
handed-0.041
 unatt-0.041
taker-0.036
and-0.035
Token with
Token ID351
Feature activation-2.97e-02

Pos logit contributions
Sal+0.023
 ho+0.019
ita+0.019
oxy+0.019
 hus+0.019
Neg logit contributions
keeping-0.045
handed-0.041
 unatt-0.040
taker-0.039
 rubble-0.035
Token lots
Token ID6041
Feature activation-3.38e-01

Pos logit contributions
keeping+0.001
handed+0.001
taker+0.001
 unatt+0.001
 rubble+0.001
Neg logit contributions
Sal-0.001
ita-0.001
ira-0.001
oxy-0.001
 ho-0.000
Token of
Token ID286
Feature activation-3.68e-01

Pos logit contributions
keeping+0.009
handed+0.007
 unatt+0.007
taker+0.007
 rubble+0.006
Neg logit contributions
Sal-0.008
ita-0.007
 ho-0.007
oxy-0.007
 hus-0.007
Token topp
Token ID23126
Feature activation+4.80e+00

Pos logit contributions
keeping+0.010
handed+0.009
taker+0.009
 unatt+0.009
 rubble+0.008
Neg logit contributions
ita-0.007
Sal-0.007
-0.007
oxy-0.007
Le-0.007
Tokenings
Token ID654
Feature activation-8.39e-01

Pos logit contributions
ita+0.112
Sal+0.109
+0.107
Jump+0.106
Le+0.105
Neg logit contributions
keeping-0.120
handed-0.095
hell-0.089
ocular-0.087
taker-0.087
Token.
Token ID13
Feature activation-8.06e-01

Pos logit contributions
keeping+0.028
handed+0.026
 unatt+0.025
taker+0.025
 rubble+0.023
Neg logit contributions
Sal-0.013
 ho-0.011
 chicken-0.011
 hus-0.011
oxy-0.011
Token Tim
Token ID5045
Feature activation-4.76e-01

Pos logit contributions
keeping+0.021
handed+0.018
 unatt+0.018
taker+0.017
 rubble+0.015
Neg logit contributions
Sal-0.019
 ho-0.017
ita-0.017
oxy-0.017
 hus-0.017
Tokenmy
Token ID1820
Feature activation+5.75e-01

Pos logit contributions
keeping+0.011
handed+0.009
 unatt+0.009
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.013
 ho-0.012
ita-0.012
oxy-0.012
 hus-0.012
Token was
Token ID373
Feature activation-3.87e-01

Pos logit contributions
Sal+0.012
 ho+0.012
Le+0.012
ita+0.012
Hug+0.012
Neg logit contributions
keeping-0.012
handed-0.011
taker-0.011
 unatt-0.010
hell-0.009
Token a
Token ID257
Feature activation+9.38e-01

Pos logit contributions
Sal+0.010
ita+0.008
 ho+0.008
oxy+0.008
 hus+0.008
Neg logit contributions
keeping-0.024
handed-0.022
 unatt-0.021
taker-0.021
 rubble-0.019
Token big
Token ID1263
Feature activation-3.64e-01

Pos logit contributions
Sal+0.013
 ho+0.011
ita+0.010
oxy+0.010
 hus+0.010
Neg logit contributions
keeping-0.033
handed-0.030
 unatt-0.030
taker-0.029
 rubble-0.027
Token tree
Token ID5509
Feature activation+9.49e-01

Pos logit contributions
keeping+0.011
handed+0.010
 unatt+0.010
taker+0.009
 rubble+0.008
Neg logit contributions
Sal-0.007
 ho-0.006
ita-0.006
oxy-0.006
 hus-0.006
Token with
Token ID351
Feature activation+7.89e-01

Pos logit contributions
Sal+0.016
 ho+0.014
ita+0.014
 hus+0.014
 chicken+0.014
Neg logit contributions
keeping-0.031
handed-0.028
 unatt-0.027
taker-0.027
 rubble-0.024
Token many
Token ID867
Feature activation+4.26e-02

Pos logit contributions
Sal+0.013
 ho+0.011
 chicken+0.011
 hus+0.011
oxy+0.011
Neg logit contributions
keeping-0.026
handed-0.024
 unatt-0.023
taker-0.023
 rubble-0.020
Token pe
Token ID613
Feature activation+4.77e+00

Pos logit contributions
Sal+0.001
oxy+0.001
ita+0.001
 ho+0.001
 chicken+0.001
Neg logit contributions
keeping-0.001
handed-0.001
 unatt-0.001
taker-0.001
 rubble-0.001
Tokenaches
Token ID3694
Feature activation+1.97e+00

Pos logit contributions
Sal+0.120
 ho+0.109
 chicken+0.108
oxy+0.108
ita+0.108
Neg logit contributions
keeping-0.115
handed-0.100
 unatt-0.098
taker-0.094
 rubble-0.081
Token.
Token ID13
Feature activation-3.00e-02

Pos logit contributions
Sal+0.031
 ho+0.026
ita+0.026
oxy+0.025
 hus+0.025
Neg logit contributions
keeping-0.067
handed-0.060
 unatt-0.060
taker-0.058
 rubble-0.053
Token The
Token ID383
Feature activation+9.16e-01

Pos logit contributions
keeping+0.001
handed+0.001
taker+0.001
 unatt+0.001
 anymore+0.001
Neg logit contributions
ita-0.001
 ho-0.001
Sal-0.001
oxy-0.001
ira-0.001
Token pe
Token ID613
Feature activation+3.40e+00

Pos logit contributions
Sal+0.016
 ho+0.014
 hus+0.014
 chicken+0.014
oxy+0.014
Neg logit contributions
keeping-0.029
handed-0.026
 unatt-0.026
taker-0.025
 rubble-0.023
Tokenaches
Token ID3694
Feature activation+4.01e-01

Pos logit contributions
Sal+0.085
 ho+0.077
ita+0.076
oxy+0.076
 hus+0.076
Neg logit contributions
keeping-0.085
handed-0.073
 unatt-0.072
taker-0.069
 rubble-0.061
Token
Token ID220
Feature activation+1.22e+00

Pos logit contributions
Sal+0.003
 ho+0.003
oxy+0.003
ita+0.003
 chicken+0.003
Neg logit contributions
keeping-0.006
 unatt-0.006
handed-0.006
taker-0.005
 rubble-0.005
Token
Token ID198
Feature activation-9.52e-02

Pos logit contributions
Sal+0.017
 ho+0.014
oxy+0.014
ita+0.014
 hus+0.014
Neg logit contributions
keeping-0.043
handed-0.039
 unatt-0.039
taker-0.038
 rubble-0.035
Token
Token ID198
Feature activation-3.39e-01

Pos logit contributions
keeping+0.003
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.001
 ho-0.001
ita-0.001
oxy-0.001
 hus-0.001
TokenThe
Token ID464
Feature activation+2.15e+00

Pos logit contributions
keeping+0.011
handed+0.011
 unatt+0.010
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.005
 ho-0.005
ita-0.004
oxy-0.004
 chicken-0.004
Token chip
Token ID11594
Feature activation+1.28e+00

Pos logit contributions
Sal+0.039
ita+0.037
oxy+0.036
Le+0.035
 ho+0.035
Neg logit contributions
keeping-0.062
handed-0.058
 unatt-0.056
taker-0.052
 rubble-0.049
Tokenm
Token ID76
Feature activation+4.73e+00

Pos logit contributions
Sal+0.032
 ho+0.031
ita+0.031
 short+0.030
oxy+0.030
Neg logit contributions
keeping-0.028
handed-0.026
 unatt-0.024
taker-0.024
 rubble-0.020
Tokenunk
Token ID2954
Feature activation+2.60e-01

Pos logit contributions
Sal+0.105
 stretch+0.098
Le+0.097
 hus+0.097
+0.095
Neg logit contributions
keeping-0.125
handed-0.109
taker-0.104
 unatt-0.095
ocular-0.089
Token was
Token ID373
Feature activation-2.15e-01

Pos logit contributions
Sal+0.004
 ho+0.004
ita+0.004
oxy+0.004
 hus+0.004
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.007
 rubble-0.007
Token excited
Token ID6568
Feature activation-6.52e-03

Pos logit contributions
keeping+0.006
handed+0.005
 unatt+0.005
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.005
 ho-0.005
 short-0.005
 chicken-0.005
 stretch-0.005
Token and
Token ID290
Feature activation+3.47e-01

Pos logit contributions
keeping+0.000
handed+0.000
 unatt+0.000
taker+0.000
 rubble+0.000
Neg logit contributions
Sal-0.000
 ho-0.000
ita-0.000
 chicken-0.000
 stretch-0.000
Token realized
Token ID6939
Feature activation-1.59e+00

Pos logit contributions
Sal+0.007
oxy+0.006
ita+0.006
 ho+0.006
 hus+0.006
Neg logit contributions
keeping-0.011
handed-0.009
 unatt-0.009
taker-0.009
 rubble-0.008
Token and
Token ID290
Feature activation+1.29e+00

Pos logit contributions
keeping+0.007
handed+0.006
 unatt+0.006
taker+0.006
 rubble+0.005
Neg logit contributions
Sal-0.004
 hus-0.004
 ho-0.004
oxy-0.004
Le-0.003
Token Tommy
Token ID19919
Feature activation-4.68e-01

Pos logit contributions
Sal+0.033
 ho+0.030
ita+0.030
oxy+0.029
 hus+0.029
Neg logit contributions
keeping-0.031
handed-0.027
 unatt-0.027
taker-0.025
 rubble-0.022
Token always
Token ID1464
Feature activation-8.22e-01

Pos logit contributions
keeping+0.013
 unatt+0.012
handed+0.012
taker+0.012
 rubble+0.010
Neg logit contributions
Sal-0.009
oxy-0.008
ita-0.008
 hus-0.008
 short-0.008
Token shared
Token ID4888
Feature activation+1.28e+00

Pos logit contributions
keeping+0.025
 unatt+0.021
handed+0.021
taker+0.019
 rubble+0.018
Neg logit contributions
Sal-0.016
oxy-0.014
ita-0.014
Le-0.014
 hus-0.014
Token their
Token ID511
Feature activation-1.07e-01

Pos logit contributions
Sal+0.012
 chicken+0.009
 ho+0.009
 corn+0.009
 hus+0.009
Neg logit contributions
keeping-0.052
handed-0.047
 unatt-0.047
taker-0.047
 rubble-0.043
Token pe
Token ID613
Feature activation+4.66e+00

Pos logit contributions
keeping+0.003
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.002
Neg logit contributions
Sal-0.002
ita-0.002
 ho-0.002
oxy-0.002
 hus-0.002
Tokenaches
Token ID3694
Feature activation+2.20e+00

Pos logit contributions
Sal+0.127
 ho+0.117
ita+0.116
oxy+0.116
 hus+0.116
Neg logit contributions
keeping-0.101
handed-0.087
 unatt-0.084
taker-0.080
ocular-0.068
Token and
Token ID290
Feature activation-1.03e+00

Pos logit contributions
Sal+0.035
 ho+0.031
 chicken+0.031
 short+0.030
 corn+0.030
Neg logit contributions
keeping-0.073
handed-0.064
 unatt-0.063
taker-0.062
 rubble-0.058
Token never
Token ID1239
Feature activation-2.55e-01

Pos logit contributions
keeping+0.035
handed+0.031
 unatt+0.031
taker+0.030
 rubble+0.027
Neg logit contributions
Sal-0.016
 ho-0.014
 chicken-0.014
ita-0.014
oxy-0.014
Token got
Token ID1392
Feature activation+5.88e-01

Pos logit contributions
keeping+0.009
handed+0.008
taker+0.008
 unatt+0.008
eway+0.008
Neg logit contributions
 stretch-0.003
 ho-0.003
Sal-0.003
 short-0.003
 chicken-0.003
Token stubborn
Token ID24444
Feature activation-2.36e-01

Pos logit contributions
Sal+0.010
 chicken+0.009
 short+0.009
 ho+0.009
 hus+0.009
Neg logit contributions
keeping-0.019
handed-0.017
 unatt-0.016
taker-0.016
 rubble-0.015
Token he
Token ID339
Feature activation+9.68e-01

Pos logit contributions
keeping+0.001
handed+0.001
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.001
 ho-0.001
 chicken-0.001
 stretch-0.001
 hus-0.001
Token could
Token ID714
Feature activation+2.03e+00

Pos logit contributions
Sal+0.017
oxy+0.015
ita+0.014
 ho+0.014
 hus+0.013
Neg logit contributions
keeping-0.029
 unatt-0.027
handed-0.026
taker-0.026
 rubble-0.023
Token spot
Token ID4136
Feature activation+9.36e-01

Pos logit contributions
Sal+0.034
 ho+0.029
oxy+0.028
ita+0.028
 hus+0.028
Neg logit contributions
keeping-0.067
handed-0.060
 unatt-0.060
taker-0.058
 rubble-0.053
Token some
Token ID617
Feature activation+9.95e-01

Pos logit contributions
 ho+0.013
Sal+0.012
 corn+0.012
 stretch+0.012
 hus+0.012
Neg logit contributions
keeping-0.033
handed-0.030
taker-0.028
 unatt-0.028
 rubble-0.026
Token animals
Token ID4695
Feature activation+4.68e-01

Pos logit contributions
Sal+0.017
oxy+0.015
ita+0.015
 ho+0.014
 hus+0.014
Neg logit contributions
keeping-0.032
 unatt-0.029
handed-0.029
taker-0.028
 rubble-0.026
Token pe
Token ID613
Feature activation+4.61e+00

Pos logit contributions
Sal+0.007
 ho+0.007
 stretch+0.007
 chicken+0.007
 short+0.007
Neg logit contributions
keeping-0.015
handed-0.014
taker-0.013
 unatt-0.013
 rubble-0.012
Tokeneking
Token ID18754
Feature activation+3.90e-01

Pos logit contributions
Sal+0.121
oxy+0.108
 human+0.107
Le+0.107
 chicken+0.107
Neg logit contributions
keeping-0.105
handed-0.091
 unatt-0.088
taker-0.083
ocular-0.076
Token out
Token ID503
Feature activation+1.65e+00

Pos logit contributions
Sal+0.008
 ho+0.007
 hus+0.007
 chicken+0.007
ita+0.007
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.009
 rubble-0.009
Token from
Token ID422
Feature activation-1.60e+00

Pos logit contributions
Sal+0.030
 ho+0.030
 short+0.029
 stretch+0.028
 hus+0.028
Neg logit contributions
keeping-0.052
handed-0.046
 unatt-0.043
taker-0.042
 rubble-0.038
Token the
Token ID262
Feature activation-8.40e-01

Pos logit contributions
keeping+0.055
handed+0.050
 unatt+0.048
taker+0.048
ocular+0.043
Neg logit contributions
Sal-0.024
 ho-0.022
 hus-0.021
 chicken-0.021
 stretch-0.021
Token banks
Token ID6341
Feature activation-1.70e-01

Pos logit contributions
keeping+0.025
handed+0.022
 unatt+0.022
taker+0.021
 rubble+0.019
Neg logit contributions
Sal-0.017
oxy-0.015
 ho-0.015
ita-0.015
 hus-0.015
Token sure
Token ID1654
Feature activation-6.65e-01

Pos logit contributions
Sal+0.028
 ho+0.025
oxy+0.025
 hus+0.025
ita+0.025
Neg logit contributions
keeping-0.022
handed-0.019
 unatt-0.018
taker-0.018
 rubble-0.015
Token to
Token ID284
Feature activation+7.50e-01

Pos logit contributions
keeping+0.025
 unatt+0.023
handed+0.023
taker+0.022
 rubble+0.021
Neg logit contributions
Sal-0.007
 chicken-0.007
 ho-0.006
 each-0.006
 short-0.006
Token always
Token ID1464
Feature activation-3.72e-01

Pos logit contributions
Sal+0.010
oxy+0.008
ita+0.007
 ho+0.007
 hus+0.007
Neg logit contributions
keeping-0.028
handed-0.025
 unatt-0.025
taker-0.024
 rubble-0.022
Token share
Token ID2648
Feature activation+1.13e+00

Pos logit contributions
keeping+0.012
 unatt+0.011
handed+0.010
taker+0.009
 rubble+0.009
Neg logit contributions
Sal-0.007
oxy-0.006
Le-0.006
ira-0.005
ita-0.005
Token his
Token ID465
Feature activation-1.62e-01

Pos logit contributions
Sal+0.014
 ho+0.014
 chicken+0.013
 short+0.013
 hus+0.013
Neg logit contributions
keeping-0.041
handed-0.037
 unatt-0.036
taker-0.036
 rubble-0.033
Token pe
Token ID613
Feature activation+4.59e+00

Pos logit contributions
keeping+0.005
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.003
ita-0.003
oxy-0.003
 ho-0.003
 hus-0.003
Tokenaches
Token ID3694
Feature activation+2.16e+00

Pos logit contributions
Sal+0.111
 ho+0.099
oxy+0.099
 hus+0.099
ita+0.098
Neg logit contributions
keeping-0.116
handed-0.102
 unatt-0.100
taker-0.096
 rubble-0.083
Token with
Token ID351
Feature activation-2.52e-01

Pos logit contributions
Sal+0.037
 ho+0.035
 stretch+0.034
 chicken+0.034
 short+0.034
Neg logit contributions
keeping-0.068
handed-0.059
 unatt-0.058
taker-0.057
 rubble-0.053
Token his
Token ID465
Feature activation-1.61e+00

Pos logit contributions
keeping+0.008
handed+0.007
 unatt+0.007
taker+0.007
 rubble+0.006
Neg logit contributions
Sal-0.005
 ho-0.004
oxy-0.004
ita-0.004
 hus-0.004
Token new
Token ID649
Feature activation-4.34e-01

Pos logit contributions
keeping+0.048
 unatt+0.043
handed+0.042
taker+0.040
 rubble+0.037
Neg logit contributions
Sal-0.030
oxy-0.026
 hus-0.025
Le-0.025
ita-0.024
Token friend
Token ID1545
Feature activation-9.54e-01

Pos logit contributions
keeping+0.014
handed+0.012
 unatt+0.012
taker+0.011
ocular+0.011
Neg logit contributions
Sal-0.008
oxy-0.007
Le-0.006
 hus-0.006
ita-0.006
Token,
Token ID11
Feature activation-1.09e+00

Pos logit contributions
Sal+0.019
 ho+0.017
ita+0.017
oxy+0.017
 hus+0.017
Neg logit contributions
keeping-0.034
handed-0.030
 unatt-0.030
taker-0.029
 rubble-0.027
Token Lily
Token ID20037
Feature activation-4.52e-01

Pos logit contributions
keeping+0.034
handed+0.030
 unatt+0.028
taker+0.027
eway+0.025
Neg logit contributions
Sal-0.021
 chicken-0.020
 short-0.020
 each-0.019
 ho-0.019
Token always
Token ID1464
Feature activation-8.26e-01

Pos logit contributions
keeping+0.015
handed+0.013
 unatt+0.013
taker+0.013
 rubble+0.011
Neg logit contributions
Sal-0.008
oxy-0.007
 ho-0.007
 hus-0.007
 chicken-0.006
Token kept
Token ID4030
Feature activation+8.75e-02

Pos logit contributions
keeping+0.027
 unatt+0.024
handed+0.023
taker+0.022
 rubble+0.020
Neg logit contributions
Sal-0.014
oxy-0.013
 chicken-0.012
ita-0.012
Le-0.012
Token a
Token ID257
Feature activation-4.88e-01

Pos logit contributions
Sal+0.002
oxy+0.001
ita+0.001
ira+0.001
Le+0.001
Neg logit contributions
keeping-0.003
handed-0.002
 unatt-0.002
taker-0.002
 rubble-0.002
Token pe
Token ID613
Feature activation+4.59e+00

Pos logit contributions
keeping+0.016
handed+0.015
 unatt+0.015
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.008
 ho-0.007
ita-0.007
oxy-0.007
 hus-0.007
Tokenb
Token ID65
Feature activation+3.67e+00

Pos logit contributions
Sal+0.099
 ho+0.090
ita+0.088
 hus+0.087
oxy+0.087
Neg logit contributions
keeping-0.131
handed-0.113
 unatt-0.111
taker-0.107
 rubble-0.097
Tokenble
Token ID903
Feature activation+1.88e+00

Pos logit contributions
Sal+0.088
oxy+0.079
 hus+0.079
ita+0.079
 chicken+0.079
Neg logit contributions
keeping-0.092
handed-0.080
 unatt-0.079
taker-0.077
ocular-0.066
Token in
Token ID287
Feature activation+1.82e-01

Pos logit contributions
 ho+0.030
 stretch+0.029
 chicken+0.028
 short+0.028
Sal+0.028
Neg logit contributions
keeping-0.064
handed-0.055
taker-0.052
 unatt-0.051
 disl-0.049
Token her
Token ID607
Feature activation-1.11e+00

Pos logit contributions
Sal+0.002
 ho+0.001
 short+0.001
 stretch+0.001
 hus+0.001
Neg logit contributions
keeping-0.007
handed-0.007
 unatt-0.007
taker-0.007
ocular-0.006
Token pocket
Token ID10000
Feature activation+1.14e+00

Pos logit contributions
keeping+0.034
 unatt+0.031
handed+0.031
taker+0.028
 rubble+0.027
Neg logit contributions
Sal-0.021
oxy-0.018
ita-0.018
 hus-0.018
Le-0.017
Token back
Token ID736
Feature activation+2.26e+00

Pos logit contributions
Sal+0.030
 ho+0.027
ita+0.027
oxy+0.027
 hus+0.027
Neg logit contributions
keeping-0.035
handed-0.031
 unatt-0.030
taker-0.029
 rubble-0.026
Token and
Token ID290
Feature activation-4.42e-02

Pos logit contributions
Sal+0.042
 hus+0.035
oxy+0.035
 ho+0.035
ita+0.035
Neg logit contributions
keeping-0.072
handed-0.064
 unatt-0.063
taker-0.059
ocular-0.054
Token find
Token ID1064
Feature activation+1.69e-01

Pos logit contributions
keeping+0.002
handed+0.001
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.001
 ho-0.001
 chicken-0.001
 stretch-0.001
ita-0.001
Token more
Token ID517
Feature activation+5.82e-01

Pos logit contributions
Sal+0.003
 chicken+0.002
 ho+0.002
oxy+0.002
ita+0.002
Neg logit contributions
keeping-0.006
 unatt-0.005
handed-0.005
taker-0.005
 rubble-0.005
Token pretty
Token ID2495
Feature activation+1.38e+00

Pos logit contributions
 corn+0.008
 chicken+0.008
Sal+0.007
 ho+0.007
 short+0.007
Neg logit contributions
keeping-0.021
 unatt-0.019
handed-0.019
taker-0.019
 rubble-0.017
Token pe
Token ID613
Feature activation+4.57e+00

Pos logit contributions
Sal+0.019
 corn+0.019
 chicken+0.019
 ho+0.018
 hus+0.018
Neg logit contributions
keeping-0.047
handed-0.044
 unatt-0.044
taker-0.043
 rubble-0.038
Tokenb
Token ID65
Feature activation+3.16e+00

Pos logit contributions
Sal+0.097
 ho+0.087
ita+0.086
 hus+0.085
oxy+0.085
Neg logit contributions
keeping-0.130
handed-0.114
 unatt-0.113
taker-0.108
 rubble-0.098
Tokenbles
Token ID7689
Feature activation+2.52e+00

Pos logit contributions
Sal+0.070
 ho+0.062
ita+0.061
oxy+0.061
 hus+0.061
Neg logit contributions
keeping-0.088
handed-0.077
 unatt-0.076
taker-0.073
 rubble-0.065
Token and
Token ID290
Feature activation-5.29e-01

Pos logit contributions
Sal+0.037
 chicken+0.036
 ho+0.036
 stretch+0.035
 corn+0.034
Neg logit contributions
keeping-0.083
handed-0.075
 unatt-0.073
taker-0.071
 rubble-0.067
Token learn
Token ID2193
Feature activation-6.93e-01

Pos logit contributions
keeping+0.018
handed+0.016
 unatt+0.016
taker+0.016
 rubble+0.014
Neg logit contributions
Sal-0.008
 ho-0.008
 stretch-0.008
 chicken-0.007
 short-0.007
Token more
Token ID517
Feature activation+6.68e-01

Pos logit contributions
keeping+0.024
handed+0.021
 unatt+0.021
taker+0.020
 rubble+0.019
Neg logit contributions
 each-0.010
 corn-0.010
 chicken-0.010
 hus-0.010
 short-0.010
Token a
Token ID257
Feature activation+6.02e-01

Pos logit contributions
Sal+0.005
 chicken+0.004
 corn+0.004
 short+0.004
 hus+0.004
Neg logit contributions
keeping-0.020
 unatt-0.018
handed-0.018
taker-0.018
 rubble-0.016
Token big
Token ID1263
Feature activation-3.43e-01

Pos logit contributions
Sal+0.009
ita+0.007
oxy+0.007
 ho+0.007
 hus+0.007
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.019
taker-0.018
 rubble-0.017
Token tree
Token ID5509
Feature activation+6.66e-01

Pos logit contributions
keeping+0.010
 unatt+0.009
handed+0.009
taker+0.008
 rubble+0.008
Neg logit contributions
Sal-0.007
oxy-0.006
ita-0.006
 ho-0.006
Le-0.006
Token with
Token ID351
Feature activation+7.02e-01

Pos logit contributions
Sal+0.011
 ho+0.010
 hus+0.010
 stretch+0.010
ita+0.010
Neg logit contributions
keeping-0.022
handed-0.019
 unatt-0.019
taker-0.019
 rubble-0.017
Token many
Token ID867
Feature activation-9.77e-03

Pos logit contributions
Sal+0.012
 ho+0.011
 hus+0.011
ita+0.011
oxy+0.011
Neg logit contributions
keeping-0.022
handed-0.020
 unatt-0.020
taker-0.019
 rubble-0.017
Token pe
Token ID613
Feature activation+4.49e+00

Pos logit contributions
keeping+0.000
handed+0.000
 unatt+0.000
taker+0.000
 rubble+0.000
Neg logit contributions
Sal-0.000
oxy-0.000
ita-0.000
 ho-0.000
Le-0.000
Tokenaches
Token ID3694
Feature activation+1.72e+00

Pos logit contributions
Sal+0.117
 chicken+0.106
 ho+0.106
oxy+0.106
ita+0.105
Neg logit contributions
keeping-0.103
handed-0.089
 unatt-0.086
taker-0.084
ocular-0.073
Token.
Token ID13
Feature activation-2.65e-01

Pos logit contributions
Sal+0.027
 ho+0.023
ita+0.023
oxy+0.023
 hus+0.023
Neg logit contributions
keeping-0.058
handed-0.052
 unatt-0.052
taker-0.050
 rubble-0.046
Token They
Token ID1119
Feature activation+3.52e-01

Pos logit contributions
keeping+0.009
 unatt+0.008
handed+0.008
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.004
 ho-0.004
oxy-0.004
 hus-0.004
 chicken-0.004
Token wanted
Token ID2227
Feature activation+3.04e-01

Pos logit contributions
Sal+0.008
oxy+0.007
ita+0.007
 chicken+0.007
+0.007
Neg logit contributions
keeping-0.009
taker-0.008
handed-0.008
 unatt-0.007
ocular-0.007
Token to
Token ID284
Feature activation+3.19e-01

Pos logit contributions
Sal+0.004
 ho+0.004
 chicken+0.004
 hus+0.004
oxy+0.004
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.009
 rubble-0.009
Token They
Token ID1119
Feature activation+1.49e-01

Pos logit contributions
Sal+0.007
ita+0.007
 ho+0.006
 stretch+0.006
 hus+0.006
Neg logit contributions
keeping-0.013
handed-0.011
taker-0.011
 unatt-0.011
ocular-0.010
Token also
Token ID635
Feature activation+2.48e-02

Pos logit contributions
Sal+0.003
ita+0.003
oxy+0.003
Le+0.003
ira+0.003
Neg logit contributions
keeping-0.004
handed-0.003
taker-0.003
 unatt-0.003
 rubble-0.003
Token see
Token ID766
Feature activation-9.83e-02

Pos logit contributions
Sal+0.001
ita+0.001
oxy+0.001
Le+0.001
P+0.001
Neg logit contributions
keeping-0.001
 unatt-0.001
handed-0.001
taker-0.000
 rubble-0.000
Token some
Token ID617
Feature activation-2.06e-01

Pos logit contributions
keeping+0.003
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.002
ita-0.001
 ho-0.001
oxy-0.001
 stretch-0.001
Token fancy
Token ID14996
Feature activation+9.01e-01

Pos logit contributions
keeping+0.005
handed+0.005
 unatt+0.005
taker+0.005
ocular+0.004
Neg logit contributions
ita-0.004
Sal-0.004
oxy-0.004
Le-0.004
-0.004
Token topp
Token ID23126
Feature activation+4.48e+00

Pos logit contributions
Sal+0.019
ita+0.019
Le+0.018
oxy+0.018
Bon+0.017
Neg logit contributions
keeping-0.025
handed-0.022
 unatt-0.020
taker-0.019
ocular-0.019
Tokenings
Token ID654
Feature activation-3.04e-01

Pos logit contributions
ita+0.108
Sal+0.103
Scar+0.102
Bon+0.101
Le+0.101
Neg logit contributions
keeping-0.106
handed-0.086
ocular-0.085
taker-0.081
hell-0.080
Token.
Token ID13
Feature activation+5.61e-01

Pos logit contributions
keeping+0.009
handed+0.008
 unatt+0.008
taker+0.008
ocular+0.007
Neg logit contributions
Sal-0.006
ita-0.005
 ho-0.005
oxy-0.005
 stretch-0.005
Token They
Token ID1119
Feature activation+2.67e-01

Pos logit contributions
Sal+0.009
ita+0.009
 ho+0.008
 stretch+0.008
 hus+0.008
Neg logit contributions
keeping-0.018
handed-0.016
 unatt-0.015
taker-0.015
ocular-0.014
Token see
Token ID766
Feature activation+8.84e-02

Pos logit contributions
Sal+0.006
ita+0.005
oxy+0.005
Le+0.005
 chicken+0.005
Neg logit contributions
keeping-0.007
handed-0.006
 unatt-0.006
taker-0.006
 rubble-0.006
Token sprink
Token ID27562
Feature activation+1.31e+00

Pos logit contributions
Sal+0.001
ita+0.001
 ho+0.001
oxy+0.001
 hus+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.002
Token
Token ID241
Feature activation+8.36e-02

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.017
taker+0.017
 disl+0.015
Neg logit contributions
Sal-0.014
 chicken-0.013
 ho-0.013
 stretch-0.012
 corn-0.012
TokenI
Token ID40
Feature activation+1.02e+00

Pos logit contributions
Sal+0.002
 ho+0.001
 chicken+0.001
ita+0.001
 stretch+0.001
Neg logit contributions
keeping-0.003
handed-0.002
taker-0.002
 unatt-0.002
ocular-0.002
Token am
Token ID716
Feature activation+1.68e-01

Pos logit contributions
Sal+0.017
 ho+0.014
ita+0.014
oxy+0.014
 hus+0.014
Neg logit contributions
keeping-0.034
handed-0.031
 unatt-0.030
taker-0.029
 rubble-0.027
Token a
Token ID257
Feature activation+3.70e-01

Pos logit contributions
Sal+0.003
 ho+0.002
ita+0.002
oxy+0.002
 hus+0.002
Neg logit contributions
keeping-0.006
handed-0.005
 unatt-0.005
taker-0.005
 rubble-0.004
Token magical
Token ID10883
Feature activation+1.15e-02

Pos logit contributions
Sal+0.006
 ho+0.006
 chicken+0.006
 hus+0.005
 short+0.005
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.011
taker-0.010
 rubble-0.010
Token pe
Token ID613
Feature activation+4.45e+00

Pos logit contributions
Sal+0.000
 ho+0.000
 chicken+0.000
 hus+0.000
 stretch+0.000
Neg logit contributions
keeping-0.000
handed-0.000
 unatt-0.000
taker-0.000
 rubble-0.000
Tokenb
Token ID65
Feature activation+3.23e+00

Pos logit contributions
Sal+0.092
 ho+0.083
ita+0.081
 hus+0.081
oxy+0.080
Neg logit contributions
keeping-0.131
handed-0.114
 unatt-0.113
taker-0.108
 rubble-0.099
Tokenble
Token ID903
Feature activation+1.80e+00

Pos logit contributions
Sal+0.073
 ho+0.066
ita+0.065
oxy+0.065
 hus+0.065
Neg logit contributions
keeping-0.087
handed-0.076
 unatt-0.075
taker-0.072
 rubble-0.065
Token and
Token ID290
Feature activation-4.96e-02

Pos logit contributions
Sal+0.031
 ho+0.028
 chicken+0.028
ita+0.027
 stretch+0.027
Neg logit contributions
keeping-0.058
handed-0.051
 unatt-0.050
taker-0.048
 rubble-0.045
Token I
Token ID314
Feature activation+6.93e-01

Pos logit contributions
keeping+0.002
handed+0.001
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.001
 ho-0.001
 stretch-0.001
 chicken-0.001
 hus-0.001
Tokenâ
Token ID22940
Feature activation-5.70e-01

Pos logit contributions
Sal+0.011
 ho+0.009
 stretch+0.009
ita+0.009
 hus+0.009
Neg logit contributions
keeping-0.024
handed-0.021
 unatt-0.021
taker-0.020
 rubble-0.019
Token I
Token ID314
Feature activation+6.81e-01

Pos logit contributions
keeping+0.020
handed+0.019
 unatt+0.017
taker+0.015
eway+0.014
Neg logit contributions
 chicken-0.013
 ho-0.012
Sal-0.012
 stretch-0.012
 corn-0.012
Token'll
Token ID1183
Feature activation+1.36e+00

Pos logit contributions
Sal+0.009
 ho+0.008
 stretch+0.008
 chicken+0.008
ita+0.007
Neg logit contributions
keeping-0.025
handed-0.023
 unatt-0.023
taker-0.021
 rubble-0.020
Token put
Token ID1234
Feature activation+4.43e-01

Pos logit contributions
Sal+0.021
 ho+0.018
ita+0.018
oxy+0.018
 hus+0.018
Neg logit contributions
keeping-0.046
handed-0.042
 unatt-0.041
taker-0.040
 rubble-0.036
Token some
Token ID617
Feature activation+1.22e+00

Pos logit contributions
Sal+0.008
ita+0.007
oxy+0.007
 ho+0.007
 hus+0.007
Neg logit contributions
keeping-0.014
handed-0.012
 unatt-0.012
taker-0.012
 rubble-0.011
Token o
Token ID267
Feature activation+2.71e+00

Pos logit contributions
Sal+0.023
ita+0.020
oxy+0.020
+0.019
avi+0.019
Neg logit contributions
keeping-0.036
 unatt-0.033
handed-0.033
taker-0.033
 rubble-0.030
Tokenint
Token ID600
Feature activation+4.42e+00

Pos logit contributions
Sal+0.067
ita+0.064
 hus+0.063
 ho+0.062
 short+0.062
Neg logit contributions
keeping-0.062
 unatt-0.056
handed-0.056
taker-0.050
ocular-0.044
Tokenment
Token ID434
Feature activation+1.38e+00

Pos logit contributions
Sal+0.095
 ho+0.086
ita+0.084
 chicken+0.084
oxy+0.084
Neg logit contributions
keeping-0.124
handed-0.109
 unatt-0.107
taker-0.103
 rubble-0.093
Token on
Token ID319
Feature activation+6.55e-01

Pos logit contributions
Sal+0.025
 ho+0.023
 chicken+0.022
 short+0.022
 stretch+0.022
Neg logit contributions
keeping-0.043
handed-0.038
 unatt-0.037
taker-0.036
 rubble-0.033
Token it
Token ID340
Feature activation+1.03e-01

Pos logit contributions
Sal+0.012
ita+0.011
oxy+0.010
 ho+0.010
 hus+0.010
Neg logit contributions
keeping-0.020
handed-0.018
 unatt-0.018
taker-0.017
 rubble-0.016
Token,"
Token ID553
Feature activation+1.21e+00

Pos logit contributions
Sal+0.002
 short+0.002
 ho+0.002
 chicken+0.002
 stretch+0.002
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.002
 rubble-0.002
Token said
Token ID531
Feature activation-3.43e-02

Pos logit contributions
Sal+0.029
 chicken+0.026
 ho+0.026
 hus+0.025
 stretch+0.025
Neg logit contributions
keeping-0.032
handed-0.028
 unatt-0.028
taker-0.026
 rubble-0.023
Token.
Token ID13
Feature activation-1.29e+00

Pos logit contributions
keeping+0.002
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.001
Neg logit contributions
Sal-0.001
 ho-0.001
oxy-0.001
 hus-0.001
ita-0.001
Token She
Token ID1375
Feature activation+6.50e-01

Pos logit contributions
keeping+0.046
handed+0.043
 unatt+0.042
taker+0.040
 rubble+0.037
Neg logit contributions
Sal-0.018
 short-0.015
 stretch-0.015
 ho-0.015
 chicken-0.015
Token loved
Token ID6151
Feature activation+1.72e+00

Pos logit contributions
Sal+0.013
 ho+0.013
 hus+0.012
oxy+0.012
avi+0.012
Neg logit contributions
keeping-0.017
taker-0.015
 unatt-0.015
handed-0.014
 rubble-0.013
Token to
Token ID284
Feature activation+1.12e-02

Pos logit contributions
Sal+0.033
ita+0.030
oxy+0.029
Le+0.029
 ho+0.028
Neg logit contributions
keeping-0.053
 unatt-0.045
handed-0.044
taker-0.042
 rubble-0.041
Token eat
Token ID4483
Feature activation+1.49e+00

Pos logit contributions
Sal+0.000
ita+0.000
Le+0.000
oxy+0.000
+0.000
Neg logit contributions
keeping-0.000
handed-0.000
 unatt-0.000
taker-0.000
 disl-0.000
Token pe
Token ID613
Feature activation+4.42e+00

Pos logit contributions
Sal+0.033
ita+0.032
Le+0.030
oxy+0.030
+0.029
Neg logit contributions
keeping-0.038
 unatt-0.034
handed-0.034
taker-0.032
 rubble-0.029
Tokenaches
Token ID3694
Feature activation+1.85e+00

Pos logit contributions
Sal+0.113
oxy+0.105
ita+0.104
 ho+0.103
Le+0.103
Neg logit contributions
keeping-0.100
handed-0.088
 unatt-0.086
taker-0.082
ocular-0.072
Token because
Token ID780
Feature activation-2.50e+00

Pos logit contributions
Sal+0.038
oxy+0.034
ita+0.033
Le+0.033
 ho+0.033
Neg logit contributions
keeping-0.053
 unatt-0.047
handed-0.047
taker-0.046
ocular-0.042
Token they
Token ID484
Feature activation-4.12e-01

Pos logit contributions
keeping+0.080
handed+0.072
 unatt+0.071
taker+0.069
 rubble+0.063
Neg logit contributions
Sal-0.043
 ho-0.038
ita-0.037
oxy-0.037
 hus-0.037
Token were
Token ID547
Feature activation-7.78e-01

Pos logit contributions
keeping+0.011
 unatt+0.010
handed+0.010
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.009
ita-0.008
 ho-0.008
oxy-0.008
 hus-0.008
Token so
Token ID523
Feature activation+3.21e-01

Pos logit contributions
keeping+0.020
 unatt+0.018
taker+0.017
handed+0.017
ocular+0.017
Neg logit contributions
Sal-0.016
oxy-0.015
ita-0.015
Le-0.015
-0.015
Token play
Token ID711
Feature activation+8.58e-01

Pos logit contributions
Sal+0.001
ita+0.001
Le+0.001
oxy+0.001
+0.001
Neg logit contributions
keeping-0.001
handed-0.001
taker-0.001
 unatt-0.001
 disl-0.001
Token outside
Token ID2354
Feature activation+1.68e+00

Pos logit contributions
Sal+0.020
ita+0.019
 ho+0.019
oxy+0.019
 hus+0.019
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.019
taker-0.018
 rubble-0.016
Token and
Token ID290
Feature activation-5.73e-01

Pos logit contributions
Sal+0.042
oxy+0.039
ita+0.037
avi+0.037
ira+0.036
Neg logit contributions
keeping-0.041
 unatt-0.037
handed-0.034
taker-0.033
 rubble-0.030
Token collect
Token ID2824
Feature activation+1.71e+00

Pos logit contributions
keeping+0.018
 unatt+0.016
handed+0.016
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.011
oxy-0.009
ita-0.009
 ho-0.009
 hus-0.009
Token pretty
Token ID2495
Feature activation+1.94e+00

Pos logit contributions
Sal+0.028
 chicken+0.025
 hus+0.025
 ho+0.024
 corn+0.024
Neg logit contributions
keeping-0.058
handed-0.052
 unatt-0.051
taker-0.049
 rubble-0.045
Token pe
Token ID613
Feature activation+4.40e+00

Pos logit contributions
Sal+0.038
oxy+0.034
ita+0.033
 ho+0.033
Le+0.032
Neg logit contributions
keeping-0.058
handed-0.050
 unatt-0.050
taker-0.048
ocular-0.044
Tokenb
Token ID65
Feature activation+3.59e+00

Pos logit contributions
Sal+0.107
oxy+0.098
ita+0.098
 ho+0.097
 chicken+0.096
Neg logit contributions
keeping-0.107
handed-0.093
 unatt-0.091
taker-0.088
ocular-0.078
Tokenbles
Token ID7689
Feature activation+2.40e+00

Pos logit contributions
Sal+0.083
ita+0.076
 hus+0.076
oxy+0.076
 chicken+0.075
Neg logit contributions
keeping-0.092
handed-0.079
 unatt-0.077
taker-0.077
ocular-0.069
Token.
Token ID13
Feature activation-7.64e-01

Pos logit contributions
Sal+0.034
 ho+0.031
 chicken+0.030
 stretch+0.029
 short+0.029
Neg logit contributions
keeping-0.084
handed-0.076
 unatt-0.075
taker-0.072
 rubble-0.068
Token One
Token ID1881
Feature activation+1.59e+00

Pos logit contributions
keeping+0.021
handed+0.019
 unatt+0.019
taker+0.019
 anymore+0.018
Neg logit contributions
Sal-0.014
 ho-0.013
 hus-0.013
oxy-0.013
ita-0.013
Token day
Token ID1110
Feature activation+4.13e-01

Pos logit contributions
Sal+0.038
 hus+0.034
oxy+0.034
 ho+0.033
avi+0.032
Neg logit contributions
keeping-0.038
handed-0.035
 unatt-0.035
taker-0.033
 rubble-0.029
Token all
Token ID477
Feature activation-1.01e-01

Pos logit contributions
Sal+0.008
 ho+0.007
oxy+0.007
ita+0.007
 hus+0.007
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.011
taker-0.011
 rubble-0.010
Token helped
Token ID4193
Feature activation+9.46e-01

Pos logit contributions
keeping+0.004
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.001
 ho-0.001
ita-0.001
 hus-0.001
 stretch-0.001
Token Beth
Token ID19223
Feature activation-1.74e+00

Pos logit contributions
Sal+0.016
 ho+0.013
ita+0.013
oxy+0.013
 hus+0.013
Neg logit contributions
keeping-0.031
handed-0.028
 unatt-0.027
taker-0.027
 rubble-0.024
Token put
Token ID1234
Feature activation+8.93e-02

Pos logit contributions
keeping+0.057
handed+0.052
 unatt+0.051
taker+0.049
 rubble+0.045
Neg logit contributions
Sal-0.029
 ho-0.025
oxy-0.024
ita-0.024
 hus-0.024
Token more
Token ID517
Feature activation+1.47e+00

Pos logit contributions
Sal+0.002
 ho+0.001
ita+0.001
oxy+0.001
 hus+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.002
 rubble-0.002
Token pe
Token ID613
Feature activation+4.40e+00

Pos logit contributions
Sal+0.025
 ho+0.023
 chicken+0.022
 hus+0.022
 stretch+0.022
Neg logit contributions
keeping-0.047
handed-0.043
 unatt-0.042
taker-0.041
 rubble-0.037
Tokenb
Token ID65
Feature activation+3.01e+00

Pos logit contributions
Sal+0.108
oxy+0.097
ita+0.096
Le+0.095
 ho+0.095
Neg logit contributions
keeping-0.108
handed-0.094
 unatt-0.093
taker-0.089
ocular-0.078
Tokenbles
Token ID7689
Feature activation+1.94e+00

Pos logit contributions
Sal+0.070
 hus+0.063
 chicken+0.063
oxy+0.063
ita+0.062
Neg logit contributions
keeping-0.078
handed-0.067
taker-0.066
 unatt-0.066
ocular-0.059
Token in
Token ID287
Feature activation+9.87e-02

Pos logit contributions
Sal+0.036
 ho+0.034
 chicken+0.033
 stretch+0.032
 hus+0.032
Neg logit contributions
keeping-0.059
handed-0.052
 unatt-0.051
taker-0.049
 rubble-0.045
Token the
Token ID262
Feature activation-9.31e-01

Pos logit contributions
Sal+0.001
 ho+0.001
 hus+0.001
 chicken+0.001
 short+0.001
Neg logit contributions
keeping-0.004
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.003
Token water
Token ID1660
Feature activation-4.53e-01

Pos logit contributions
keeping+0.028
handed+0.026
 unatt+0.026
taker+0.024
 rubble+0.022
Neg logit contributions
Sal-0.018
ita-0.016
oxy-0.015
 ho-0.015
Le-0.015
Token play
Token ID711
Feature activation+1.66e-02

Pos logit contributions
keeping+0.029
handed+0.026
 unatt+0.025
taker+0.024
 rubble+0.022
Neg logit contributions
Sal-0.017
oxy-0.015
ita-0.015
-0.015
 ho-0.014
Token too
Token ID1165
Feature activation-1.13e+00

Pos logit contributions
Sal+0.000
 chicken+0.000
 stretch+0.000
 short+0.000
 ho+0.000
Neg logit contributions
keeping-0.001
handed-0.000
taker-0.000
 unatt-0.000
ocular-0.000
Token!
Token ID0
Feature activation+9.74e-02

Pos logit contributions
keeping+0.041
handed+0.037
 unatt+0.036
taker+0.036
 rubble+0.033
Neg logit contributions
Sal-0.015
 ho-0.013
 short-0.013
ita-0.013
 hus-0.013
Token
Token ID198
Feature activation+4.92e-01

Pos logit contributions
Sal+0.002
 ho+0.001
 hus+0.001
ita+0.001
oxy+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.002
Token
Token ID198
Feature activation-2.36e-01

Pos logit contributions
Sal+0.007
 ho+0.006
ita+0.006
 hus+0.006
oxy+0.006
Neg logit contributions
keeping-0.017
handed-0.015
 unatt-0.015
taker-0.015
 rubble-0.013
TokenM
Token ID44
Feature activation+4.39e+00

Pos logit contributions
keeping+0.006
handed+0.006
 unatt+0.006
taker+0.005
wheel+0.005
Neg logit contributions
ira-0.004
avi-0.004
 ho-0.004
Sal-0.004
 hus-0.004
Tokenittens
Token ID34978
Feature activation+1.29e-01

Pos logit contributions
Sal+0.078
 hus+0.077
 stretch+0.076
 each+0.072
Le+0.071
Neg logit contributions
keeping-0.114
handed-0.111
 unatt-0.106
taker-0.096
reating-0.093
Token'
Token ID6
Feature activation+2.11e+00

Pos logit contributions
Sal+0.002
 ho+0.002
ita+0.002
 hus+0.002
oxy+0.002
Neg logit contributions
keeping-0.004
handed-0.004
 unatt-0.004
taker-0.004
 rubble-0.003
Token owner
Token ID4870
Feature activation-9.17e-01

Pos logit contributions
Sal+0.033
 hus+0.029
ita+0.028
 stretch+0.028
 ho+0.027
Neg logit contributions
keeping-0.070
handed-0.063
 unatt-0.062
taker-0.058
ocular-0.056
Token loved
Token ID6151
Feature activation+1.01e+00

Pos logit contributions
keeping+0.033
handed+0.029
 unatt+0.029
taker+0.027
 rubble+0.025
Neg logit contributions
Sal-0.014
 ho-0.012
 chicken-0.012
 hus-0.012
ita-0.012
Token her
Token ID607
Feature activation-1.42e+00

Pos logit contributions
Sal+0.017
oxy+0.014
 ho+0.014
ita+0.014
 hus+0.014
Neg logit contributions
keeping-0.034
 unatt-0.030
handed-0.030
taker-0.029
 rubble-0.027
Token sund
Token ID37437
Feature activation+9.93e-01

Pos logit contributions
keeping+0.018
handed+0.017
 unatt+0.017
taker+0.017
 rubble+0.015
Neg logit contributions
Sal-0.011
ita-0.010
oxy-0.010
Le-0.010
avi-0.009
Tokenae
Token ID3609
Feature activation+8.24e-01

Pos logit contributions
Sal+0.017
 hus+0.016
 chicken+0.015
Le+0.015
 corn+0.014
Neg logit contributions
keeping-0.032
handed-0.030
 unatt-0.029
taker-0.027
ocular-0.025
Token with
Token ID351
Feature activation+1.34e-01

Pos logit contributions
Sal+0.012
 ho+0.011
 chicken+0.011
 stretch+0.011
 hus+0.011
Neg logit contributions
keeping-0.028
handed-0.025
 unatt-0.025
taker-0.024
 rubble-0.022
Token lots
Token ID6041
Feature activation-2.62e-01

Pos logit contributions
Sal+0.002
ita+0.002
oxy+0.002
Le+0.002
+0.002
Neg logit contributions
keeping-0.004
handed-0.004
 unatt-0.004
taker-0.004
 rubble-0.004
Token of
Token ID286
Feature activation+4.06e-02

Pos logit contributions
keeping+0.006
handed+0.006
 unatt+0.005
taker+0.005
 rubble+0.005
Neg logit contributions
Sal-0.007
ita-0.006
oxy-0.006
 ho-0.006
Le-0.006
Token topp
Token ID23126
Feature activation+4.36e+00

Pos logit contributions
ita+0.001
Sal+0.001
+0.001
oxy+0.001
+0.001
Neg logit contributions
keeping-0.001
handed-0.001
taker-0.001
 unatt-0.001
 rubble-0.001
Tokenings
Token ID654
Feature activation-6.90e-01

Pos logit contributions
Sal+0.103
+0.101
ita+0.099
Jump+0.099
Le+0.099
Neg logit contributions
keeping-0.109
handed-0.085
taker-0.082
ocular-0.080
hell-0.079
Token.
Token ID13
Feature activation-1.81e-01

Pos logit contributions
keeping+0.023
handed+0.020
 unatt+0.020
taker+0.020
 rubble+0.018
Neg logit contributions
Sal-0.011
 ho-0.010
ita-0.010
oxy-0.010
 hus-0.009
Token
Token ID198
Feature activation+3.96e-01

Pos logit contributions
keeping+0.006
handed+0.005
 unatt+0.005
taker+0.005
 rubble+0.004
Neg logit contributions
Sal-0.003
 ho-0.003
ita-0.003
 hus-0.003
oxy-0.003
Token
Token ID198
Feature activation-1.12e-01

Pos logit contributions
Sal+0.006
 ho+0.005
ita+0.005
 hus+0.005
 chicken+0.005
Neg logit contributions
keeping-0.014
handed-0.012
 unatt-0.012
taker-0.012
 rubble-0.011
TokenThey
Token ID2990
Feature activation+5.57e-01

Pos logit contributions
keeping+0.004
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.002
 ho-0.002
 chicken-0.002
 hus-0.002
ira-0.002
Token together
Token ID1978
Feature activation+9.34e-02

Pos logit contributions
Sal+0.013
 stretch+0.012
 ho+0.011
 short+0.011
 chicken+0.011
Neg logit contributions
keeping-0.034
handed-0.031
 unatt-0.030
taker-0.030
 rubble-0.027
Token.
Token ID13
Feature activation+5.48e-01

Pos logit contributions
Sal+0.002
 ho+0.001
ita+0.001
oxy+0.001
 hus+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.002
Token It
Token ID632
Feature activation+5.21e-01

Pos logit contributions
Sal+0.010
 ho+0.009
oxy+0.008
 chicken+0.008
 stretch+0.008
Neg logit contributions
keeping-0.018
handed-0.016
 unatt-0.016
taker-0.015
 rubble-0.014
Token will
Token ID481
Feature activation+5.66e-01

Pos logit contributions
Sal+0.007
 ho+0.006
ita+0.006
oxy+0.006
 hus+0.006
Neg logit contributions
keeping-0.019
handed-0.017
 unatt-0.017
taker-0.016
 rubble-0.015
Token be
Token ID307
Feature activation+1.46e+00

Pos logit contributions
Sal+0.006
 ho+0.005
 stretch+0.005
 short+0.004
 hus+0.004
Neg logit contributions
keeping-0.023
handed-0.021
 unatt-0.020
taker-0.020
 rubble-0.018
Token higher
Token ID2440
Feature activation+4.36e+00

Pos logit contributions
Sal+0.021
 short+0.020
 ho+0.020
 stretch+0.019
 chicken+0.018
Neg logit contributions
keeping-0.052
handed-0.047
 unatt-0.044
taker-0.044
 rubble-0.040
Token and
Token ID290
Feature activation-5.95e-01

Pos logit contributions
Sal+0.061
 ho+0.052
ita+0.051
 chicken+0.050
 hus+0.050
Neg logit contributions
keeping-0.155
handed-0.141
 unatt-0.138
taker-0.135
 rubble-0.124
Token nicer
Token ID36597
Feature activation-1.80e-01

Pos logit contributions
keeping+0.018
handed+0.016
 unatt+0.015
taker+0.015
 rubble+0.013
Neg logit contributions
Sal-0.012
 stretch-0.011
 short-0.011
 ho-0.011
 hus-0.011
Token than
Token ID621
Feature activation-2.07e+00

Pos logit contributions
keeping+0.006
handed+0.006
taker+0.005
 unatt+0.005
 rubble+0.005
Neg logit contributions
Sal-0.003
 ho-0.003
 each-0.003
 short-0.002
 chicken-0.002
Token our
Token ID674
Feature activation-2.07e+00

Pos logit contributions
keeping+0.064
handed+0.057
 unatt+0.056
taker+0.054
 rubble+0.049
Neg logit contributions
Sal-0.039
 ho-0.034
ita-0.034
oxy-0.034
 hus-0.034
Token towers
Token ID18028
Feature activation+1.37e+00

Pos logit contributions
keeping+0.056
handed+0.048
 unatt+0.046
taker+0.045
 rubble+0.038
Neg logit contributions
Sal-0.047
 ho-0.044
 short-0.043
 stretch-0.043
 chicken-0.043
Token away
Token ID1497
Feature activation-7.38e-01

Pos logit contributions
keeping+0.038
handed+0.036
 unatt+0.035
taker+0.033
 disl+0.033
Neg logit contributions
 each-0.018
 ho-0.018
 short-0.017
 long-0.017
 five-0.017
Token and
Token ID290
Feature activation-2.10e-01

Pos logit contributions
keeping+0.026
handed+0.025
 unatt+0.024
taker+0.023
reating+0.021
Neg logit contributions
 ho-0.010
 short-0.009
 hus-0.009
Sal-0.009
 stretch-0.009
Token disappeared
Token ID12120
Feature activation-6.53e-01

Pos logit contributions
keeping+0.007
handed+0.007
taker+0.007
reating+0.007
 unatt+0.007
Neg logit contributions
 ho-0.003
 stretch-0.002
 each-0.002
 be-0.002
 sne-0.002
Token in
Token ID287
Feature activation-1.05e+00

Pos logit contributions
keeping+0.023
handed+0.021
taker+0.020
 unatt+0.020
reating+0.018
Neg logit contributions
 ho-0.008
Sal-0.008
 hus-0.008
 short-0.008
 stretch-0.007
Token the
Token ID262
Feature activation-1.25e+00

Pos logit contributions
taker+0.040
keeping+0.040
 unatt+0.040
handed+0.039
ount+0.039
Neg logit contributions
 short-0.008
 each-0.007
 long-0.007
 human-0.007
 line-0.006
Token alley
Token ID23978
Feature activation-5.56e+00

Pos logit contributions
keeping+0.046
handed+0.039
taker+0.039
 unatt+0.037
ount+0.036
Neg logit contributions
 short-0.016
 stretch-0.016
 corn-0.016
 chicken-0.015
 human-0.015
Token.
Token ID13
Feature activation+7.09e-01

Pos logit contributions
keeping+0.218
 unatt+0.199
handed+0.194
taker+0.193
 disl+0.185
Neg logit contributions
 short-0.037
 line-0.035
 long-0.033
 each-0.032
Sal-0.032
Token The
Token ID383
Feature activation+2.19e-01

Pos logit contributions
Sal+0.009
oxy+0.007
 short+0.007
Le+0.007
Ben+0.007
Neg logit contributions
keeping-0.026
 unatt-0.024
handed-0.024
taker-0.022
 rubble-0.021
Token t
Token ID256
Feature activation-1.46e-01

Pos logit contributions
 human+0.005
 chicken+0.005
 short+0.004
 ho+0.004
 hus+0.004
Neg logit contributions
keeping-0.007
handed-0.005
ount-0.005
 unatt-0.005
rol-0.005
Tokeneddy
Token ID21874
Feature activation-1.22e+00

Pos logit contributions
keeping+0.004
 unatt+0.004
handed+0.004
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.003
ita-0.003
oxy-0.003
ira-0.003
 ho-0.003
Token bear
Token ID6842
Feature activation+8.29e-02

Pos logit contributions
keeping+0.038
handed+0.033
 unatt+0.033
taker+0.032
 rubble+0.029
Neg logit contributions
Sal-0.023
 chicken-0.021
 ho-0.021
ita-0.020
 hus-0.020
Token two
Token ID734
Feature activation+1.62e+00

Pos logit contributions
keeping+0.010
 unatt+0.009
handed+0.009
taker+0.008
reating+0.008
Neg logit contributions
 each-0.002
 chicken-0.002
 corn-0.002
 hus-0.002
Sal-0.002
Token cats
Token ID11875
Feature activation-7.10e-01

Pos logit contributions
Sal+0.030
oxy+0.025
ita+0.025
 ho+0.024
Le+0.024
Neg logit contributions
keeping-0.050
handed-0.045
 unatt-0.044
taker-0.043
 rubble-0.040
Token fighting
Token ID4330
Feature activation-3.63e-01

Pos logit contributions
keeping+0.023
handed+0.021
 unatt+0.021
taker+0.020
 rubble+0.018
Neg logit contributions
Sal-0.011
 ho-0.011
 hus-0.010
 chicken-0.010
ita-0.010
Token in
Token ID287
Feature activation-6.38e-02

Pos logit contributions
keeping+0.010
handed+0.009
 unatt+0.008
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.008
 ho-0.008
 hus-0.007
ita-0.007
 chicken-0.007
Token the
Token ID262
Feature activation-5.42e-01

Pos logit contributions
keeping+0.002
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.001
 ho-0.001
ita-0.001
oxy-0.001
 hus-0.001
Token alley
Token ID23978
Feature activation-4.97e+00

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.011
oxy-0.009
 ho-0.009
ita-0.009
 hus-0.009
Tokenway
Token ID1014
Feature activation-3.69e-01

Pos logit contributions
keeping+0.183
 unatt+0.171
handed+0.166
taker+0.162
 rubble+0.150
Neg logit contributions
Sal-0.050
 short-0.045
 hus-0.044
 each-0.043
ita-0.043
Token.
Token ID13
Feature activation-2.04e-01

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.011
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.006
 ho-0.006
ita-0.005
oxy-0.005
 hus-0.005
Token The
Token ID383
Feature activation+1.03e+00

Pos logit contributions
keeping+0.006
taker+0.005
handed+0.005
 unatt+0.005
ocular+0.004
Neg logit contributions
Sal-0.004
ita-0.004
oxy-0.004
 ho-0.004
P-0.004
Token cats
Token ID11875
Feature activation-2.03e+00

Pos logit contributions
Sal+0.020
oxy+0.018
 ho+0.017
ita+0.017
 hus+0.017
Neg logit contributions
keeping-0.030
handed-0.027
 unatt-0.027
taker-0.026
 rubble-0.024
Token were
Token ID547
Feature activation-3.83e-01

Pos logit contributions
keeping+0.066
handed+0.059
 unatt+0.058
taker+0.055
 rubble+0.051
Neg logit contributions
Sal-0.035
 ho-0.031
 hus-0.031
 chicken-0.030
 stretch-0.030
Token shiny
Token ID22441
Feature activation+2.17e-01

Pos logit contributions
keeping+0.004
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.002
ita-0.002
oxy-0.002
 ho-0.002
Le-0.002
Token metal
Token ID6147
Feature activation-5.27e-01

Pos logit contributions
Sal+0.005
oxy+0.004
 ho+0.004
ita+0.004
 hus+0.004
Neg logit contributions
keeping-0.006
handed-0.005
 unatt-0.005
taker-0.005
 rubble-0.005
Token box
Token ID3091
Feature activation-4.56e-01

Pos logit contributions
keeping+0.016
handed+0.015
 unatt+0.014
taker+0.014
 rubble+0.012
Neg logit contributions
Sal-0.010
 chicken-0.009
 ho-0.009
 hus-0.009
 stretch-0.009
Token in
Token ID287
Feature activation-7.94e-01

Pos logit contributions
keeping+0.016
handed+0.015
 unatt+0.014
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.006
 stretch-0.006
 ho-0.006
 short-0.005
 chicken-0.005
Token an
Token ID281
Feature activation-2.12e+00

Pos logit contributions
keeping+0.031
 unatt+0.030
taker+0.030
handed+0.029
onel+0.027
Neg logit contributions
Sal-0.006
 hus-0.006
 each-0.006
 short-0.006
 corn-0.006
Token alley
Token ID23978
Feature activation-4.89e+00

Pos logit contributions
keeping+0.063
handed+0.055
taker+0.054
 unatt+0.053
ocular+0.047
Neg logit contributions
Sal-0.041
 ho-0.037
 chicken-0.037
 corn-0.037
 short-0.037
Tokenway
Token ID1014
Feature activation-3.63e-01

Pos logit contributions
keeping+0.170
 unatt+0.151
handed+0.150
taker+0.145
 disl+0.133
Neg logit contributions
Sal-0.064
 short-0.059
 hus-0.057
 stretch-0.055
ita-0.054
Token.
Token ID13
Feature activation+5.83e-02

Pos logit contributions
keeping+0.012
handed+0.010
 unatt+0.010
taker+0.010
ocular+0.009
Neg logit contributions
Sal-0.007
oxy-0.006
 chicken-0.006
 ho-0.006
Le-0.005
Token She
Token ID1375
Feature activation+3.53e-01

Pos logit contributions
Sal+0.001
 ho+0.001
ita+0.001
 hus+0.001
oxy+0.001
Neg logit contributions
keeping-0.002
handed-0.002
 unatt-0.002
taker-0.001
ocular-0.001
Token went
Token ID1816
Feature activation-6.32e-01

Pos logit contributions
Sal+0.007
oxy+0.007
 chicken+0.007
 corn+0.007
 ho+0.007
Neg logit contributions
keeping-0.008
taker-0.007
handed-0.007
ocular-0.007
 unatt-0.006
Token over
Token ID625
Feature activation+8.46e-01

Pos logit contributions
keeping+0.022
handed+0.020
 unatt+0.019
taker+0.019
 rubble+0.017
Neg logit contributions
Sal-0.009
 ho-0.008
 short-0.008
 hus-0.008
 stretch-0.008
Token she
Token ID673
Feature activation-5.43e-02

Pos logit contributions
keeping+0.046
handed+0.041
 unatt+0.041
taker+0.039
 rubble+0.036
Neg logit contributions
Sal-0.026
 ho-0.023
oxy-0.022
ita-0.022
 chicken-0.022
Token pe
Token ID613
Feature activation+2.44e+00

Pos logit contributions
keeping+0.001
 unatt+0.001
handed+0.001
taker+0.001
ocular+0.001
Neg logit contributions
Sal-0.001
oxy-0.001
 chicken-0.001
 corn-0.001
 hus-0.001
Tokenered
Token ID1068
Feature activation-7.60e-01

Pos logit contributions
Sal+0.072
 chicken+0.068
 hus+0.066
 ho+0.066
 stretch+0.066
Neg logit contributions
keeping-0.047
handed-0.040
 unatt-0.038
taker-0.035
ocular-0.031
Token into
Token ID656
Feature activation+1.43e+00

Pos logit contributions
keeping+0.020
handed+0.017
 unatt+0.016
taker+0.016
 rubble+0.014
Neg logit contributions
Sal-0.018
 ho-0.017
 hus-0.017
ita-0.016
 stretch-0.016
Token the
Token ID262
Feature activation-3.83e-01

Pos logit contributions
Sal+0.020
 ho+0.017
 hus+0.016
ita+0.016
 chicken+0.016
Neg logit contributions
keeping-0.051
handed-0.047
 unatt-0.046
taker-0.045
 rubble-0.041
Token alley
Token ID23978
Feature activation-4.83e+00

Pos logit contributions
keeping+0.010
taker+0.008
 unatt+0.008
handed+0.008
eway+0.007
Neg logit contributions
Sal-0.009
 short-0.009
 human-0.009
 line-0.008
Be-0.008
Token.
Token ID13
Feature activation-2.02e-01

Pos logit contributions
keeping+0.187
 unatt+0.170
handed+0.167
taker+0.162
onel+0.155
Neg logit contributions
 short-0.038
Sal-0.035
-0.034
 hus-0.034
 long-0.034
Token There
Token ID1318
Feature activation-5.15e-01

Pos logit contributions
keeping+0.006
handed+0.006
 unatt+0.006
taker+0.006
 rubble+0.005
Neg logit contributions
Sal-0.004
 ho-0.003
ita-0.003
oxy-0.003
 chicken-0.003
Token was
Token ID373
Feature activation-1.66e-02

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.015
 rubble+0.013
Neg logit contributions
Sal-0.009
 ho-0.007
ita-0.007
oxy-0.007
 hus-0.007
Token a
Token ID257
Feature activation+3.16e-01

Pos logit contributions
keeping+0.001
handed+0.001
 unatt+0.001
taker+0.001
ocular+0.001
Neg logit contributions
Sal-0.000
 short-0.000
 chicken-0.000
 hus-0.000
 ho-0.000
Token big
Token ID1263
Feature activation-7.48e-01

Pos logit contributions
Sal+0.005
 ho+0.004
ita+0.004
oxy+0.004
 hus+0.004
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.009
 rubble-0.009
Token one
Token ID530
Feature activation-5.17e-01

Pos logit contributions
Sal+0.004
 short+0.004
 stretch+0.004
 chicken+0.004
 ho+0.004
Neg logit contributions
keeping-0.008
 unatt-0.007
handed-0.007
taker-0.007
ocular-0.006
Token she
Token ID673
Feature activation+7.75e-01

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.009
 ho-0.008
ita-0.008
oxy-0.008
 hus-0.008
Token met
Token ID1138
Feature activation-6.68e-01

Pos logit contributions
Sal+0.015
oxy+0.013
 ho+0.013
 hus+0.013
 chicken+0.013
Neg logit contributions
keeping-0.023
handed-0.021
 unatt-0.021
taker-0.020
 rubble-0.017
Token in
Token ID287
Feature activation-3.01e-02

Pos logit contributions
keeping+0.022
handed+0.020
 unatt+0.020
taker+0.019
 rubble+0.018
Neg logit contributions
Sal-0.011
 ho-0.010
ita-0.009
 hus-0.009
 chicken-0.009
Token the
Token ID262
Feature activation-9.15e-01

Pos logit contributions
keeping+0.001
handed+0.001
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.000
 ho-0.000
ita-0.000
oxy-0.000
 hus-0.000
Token alley
Token ID23978
Feature activation-4.82e+00

Pos logit contributions
keeping+0.029
handed+0.026
 unatt+0.025
taker+0.025
 rubble+0.022
Neg logit contributions
Sal-0.017
 ho-0.015
ita-0.014
 hus-0.014
 chicken-0.014
Token that
Token ID326
Feature activation-2.01e+00

Pos logit contributions
keeping+0.169
 unatt+0.150
handed+0.149
taker+0.146
 rubble+0.135
Neg logit contributions
Sal-0.060
ita-0.055
 short-0.055
 hus-0.054
 ho-0.052
Token day
Token ID1110
Feature activation-1.60e-01

Pos logit contributions
keeping+0.063
handed+0.056
 unatt+0.056
taker+0.054
 rubble+0.048
Neg logit contributions
Sal-0.037
oxy-0.032
 ho-0.032
ita-0.032
 hus-0.031
Token.
Token ID13
Feature activation+3.82e-01

Pos logit contributions
keeping+0.005
handed+0.005
 unatt+0.004
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.003
oxy-0.002
 ho-0.002
ita-0.002
 hus-0.002
Token<|endoftext|>
Token ID50256
Feature activation-2.06e+00

Pos logit contributions
Sal+0.006
 ho+0.005
ita+0.005
oxy+0.005
 hus+0.005
Neg logit contributions
keeping-0.013
handed-0.012
 unatt-0.011
taker-0.011
 rubble-0.010
TokenOnce
Token ID7454
Feature activation-1.73e+00

Pos logit contributions
keeping+0.074
handed+0.067
 unatt+0.067
taker+0.065
 rubble+0.060
Neg logit contributions
Sal-0.027
 ho-0.023
ita-0.022
oxy-0.022
 hus-0.022
Token she
Token ID673
Feature activation+9.15e-01

Pos logit contributions
keeping+0.015
handed+0.013
 unatt+0.013
taker+0.012
 rubble+0.012
Neg logit contributions
 short-0.004
Sal-0.004
 ho-0.004
 stretch-0.004
 long-0.004
Token spotted
Token ID13489
Feature activation-5.38e-01

Pos logit contributions
Sal+0.016
 ho+0.013
oxy+0.013
ita+0.013
 hus+0.013
Neg logit contributions
keeping-0.030
handed-0.027
 unatt-0.026
taker-0.025
 rubble-0.023
Token it
Token ID340
Feature activation+2.95e-02

Pos logit contributions
keeping+0.022
 unatt+0.021
taker+0.020
handed+0.020
makers+0.019
Neg logit contributions
 each-0.004
 chicken-0.004
 corn-0.003
 short-0.003
 hus-0.003
Token in
Token ID287
Feature activation+2.06e-01

Pos logit contributions
Sal+0.000
 ho+0.000
ita+0.000
 hus+0.000
oxy+0.000
Neg logit contributions
keeping-0.001
handed-0.001
 unatt-0.001
taker-0.001
 rubble-0.001
Token an
Token ID281
Feature activation-9.82e-01

Pos logit contributions
Sal+0.003
 hus+0.002
 ho+0.002
 chicken+0.002
 short+0.002
Neg logit contributions
keeping-0.008
 unatt-0.007
handed-0.007
taker-0.007
ocular-0.006
Token alley
Token ID23978
Feature activation-4.69e+00

Pos logit contributions
keeping+0.030
handed+0.027
taker+0.026
 unatt+0.026
ocular+0.023
Neg logit contributions
Sal-0.017
 corn-0.017
 chicken-0.016
ita-0.016
 ho-0.016
Token.
Token ID13
Feature activation+1.89e-01

Pos logit contributions
keeping+0.153
 unatt+0.135
handed+0.132
taker+0.131
 disl+0.122
Neg logit contributions
Sal-0.064
 short-0.063
 line-0.062
 hus-0.062
-0.060
Token
Token ID198
Feature activation-9.42e-02

Pos logit contributions
Sal+0.003
 short+0.002
ita+0.002
oxy+0.002
+0.002
Neg logit contributions
keeping-0.007
 unatt-0.006
handed-0.006
taker-0.006
 rubble-0.005
Token
Token ID198
Feature activation-4.32e-01

Pos logit contributions
keeping+0.004
 unatt+0.003
handed+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.001
 short-0.001
Le-0.001
 hus-0.001
ita-0.001
TokenBut
Token ID1537
Feature activation-8.69e-01

Pos logit contributions
keeping+0.013
handed+0.012
 unatt+0.012
taker+0.012
 rubble+0.010
Neg logit contributions
Sal-0.008
 ho-0.007
 chicken-0.007
oxy-0.007
ita-0.007
Token when
Token ID618
Feature activation+5.86e-01

Pos logit contributions
keeping+0.031
handed+0.028
 unatt+0.026
taker+0.026
 rubble+0.024
Neg logit contributions
 ho-0.012
Sal-0.012
 short-0.012
 stretch-0.012
ita-0.012
Token for
Token ID329
Feature activation-8.45e-01

Pos logit contributions
keeping+0.077
handed+0.069
 unatt+0.069
taker+0.068
 rubble+0.064
Neg logit contributions
 ho-0.019
ita-0.018
 stretch-0.017
 chicken-0.017
Sal-0.017
Token tresp
Token ID29723
Feature activation-2.88e+00

Pos logit contributions
keeping+0.028
handed+0.026
 unatt+0.026
taker+0.025
ocular+0.023
Neg logit contributions
Sal-0.013
 ho-0.012
 short-0.012
 hus-0.012
 stretch-0.011
Tokenassing
Token ID19696
Feature activation-2.81e+00

Pos logit contributions
keeping+0.067
taker+0.059
 unatt+0.058
handed+0.058
 rubble+0.048
Neg logit contributions
Sal-0.072
 ho-0.066
ita-0.066
oxy-0.066
 hus-0.065
Token on
Token ID319
Feature activation-2.47e-01

Pos logit contributions
keeping+0.098
handed+0.089
 unatt+0.088
taker+0.085
 rubble+0.080
Neg logit contributions
Sal-0.035
 ho-0.032
 short-0.031
 hus-0.031
ita-0.031
Token his
Token ID465
Feature activation-1.26e+00

Pos logit contributions
keeping+0.008
handed+0.008
 unatt+0.008
taker+0.008
ocular+0.007
Neg logit contributions
Sal-0.003
 ho-0.003
 hus-0.003
 short-0.003
 chicken-0.003
Token property
Token ID3119
Feature activation-4.67e+00

Pos logit contributions
keeping+0.039
handed+0.035
 unatt+0.034
taker+0.033
 rubble+0.030
Neg logit contributions
Sal-0.024
oxy-0.020
ita-0.020
 ho-0.020
 hus-0.020
Token,
Token ID11
Feature activation-5.49e-01

Pos logit contributions
keeping+0.177
handed+0.162
 unatt+0.158
taker+0.157
 disl+0.149
Neg logit contributions
Sal-0.040
 short-0.039
 ho-0.039
 each-0.037
 stretch-0.037
Token but
Token ID475
Feature activation-1.87e+00

Pos logit contributions
keeping+0.017
handed+0.016
 unatt+0.015
taker+0.015
 rubble+0.013
Neg logit contributions
Sal-0.009
 ho-0.009
 short-0.009
 hus-0.009
 chicken-0.009
Token when
Token ID618
Feature activation+5.84e-01

Pos logit contributions
keeping+0.069
handed+0.063
 unatt+0.059
ount+0.057
taker+0.055
Neg logit contributions
 short-0.022
 each-0.022
 be-0.021
 stretch-0.021
 ho-0.020
Token he
Token ID339
Feature activation-7.85e-02

Pos logit contributions
Sal+0.007
 short+0.007
 ho+0.007
 stretch+0.007
 long+0.007
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.019
taker-0.018
ocular-0.017
Token heard
Token ID2982
Feature activation-6.57e-01

Pos logit contributions
keeping+0.003
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.001
 ho-0.001
 hus-0.001
ita-0.001
 stretch-0.001
Token went
Token ID1816
Feature activation-9.73e-01

Pos logit contributions
keeping+0.003
handed+0.003
taker+0.003
 unatt+0.003
eway+0.003
Neg logit contributions
 stretch-0.001
 short-0.001
 ho-0.001
Sal-0.001
 sne-0.001
Token back
Token ID736
Feature activation+4.48e-01

Pos logit contributions
keeping+0.035
taker+0.031
handed+0.030
 unatt+0.029
eway+0.028
Neg logit contributions
 short-0.013
 chicken-0.013
Sal-0.013
 ho-0.012
 stretch-0.012
Token to
Token ID284
Feature activation-1.84e+00

Pos logit contributions
 each+0.003
 ho+0.002
 short+0.002
 chicken+0.002
 corn+0.002
Neg logit contributions
taker-0.018
keeping-0.018
handed-0.017
 unatt-0.016
hell-0.016
Token the
Token ID262
Feature activation-8.51e-01

Pos logit contributions
keeping+0.067
taker+0.063
handed+0.061
 unatt+0.060
ocular+0.058
Neg logit contributions
Sal-0.020
 chicken-0.020
 stretch-0.020
 short-0.019
 ho-0.019
Token gloomy
Token ID46400
Feature activation-1.21e+00

Pos logit contributions
keeping+0.026
handed+0.023
 unatt+0.023
taker+0.021
 rubble+0.020
Neg logit contributions
Sal-0.017
ita-0.015
 ho-0.015
oxy-0.014
ira-0.014
Token alley
Token ID23978
Feature activation-4.62e+00

Pos logit contributions
keeping+0.038
handed+0.033
 unatt+0.033
taker+0.031
 rubble+0.029
Neg logit contributions
Sal-0.024
 ho-0.020
oxy-0.020
ita-0.020
 hus-0.020
Token again
Token ID757
Feature activation+1.20e+00

Pos logit contributions
keeping+0.135
taker+0.114
handed+0.108
vell+0.107
eway+0.105
Neg logit contributions
 each-0.085
 short-0.080
 be-0.079
 line-0.078
 long-0.076
Token.
Token ID13
Feature activation-5.96e-01

Pos logit contributions
Sal+0.021
 ho+0.017
oxy+0.017
ita+0.017
 hus+0.017
Neg logit contributions
keeping-0.040
handed-0.036
 unatt-0.035
taker-0.034
 rubble-0.031
Token<|endoftext|>
Token ID50256
Feature activation-1.74e+00

Pos logit contributions
keeping+0.026
 unatt+0.024
handed+0.023
taker+0.023
 rubble+0.022
Neg logit contributions
Sal-0.004
 each-0.002
Le-0.002
-0.002
Ben-0.002
TokenOnce
Token ID7454
Feature activation-1.57e+00

Pos logit contributions
keeping+0.065
 unatt+0.060
handed+0.059
taker+0.058
 rubble+0.054
Neg logit contributions
Sal-0.021
oxy-0.016
Le-0.016
 hus-0.016
 chicken-0.016
Token upon
Token ID2402
Feature activation+7.20e-01

Pos logit contributions
keeping+0.059
handed+0.053
 unatt+0.051
taker+0.049
 rubble+0.046
Neg logit contributions
Sal-0.020
 short-0.018
 ho-0.017
ira-0.017
ita-0.016
Token She
Token ID1375
Feature activation+3.12e-01

Pos logit contributions
Sal+0.005
 short+0.004
ira+0.004
Le+0.004
Ben+0.004
Neg logit contributions
keeping-0.016
 unatt-0.015
handed-0.015
taker-0.014
 rubble-0.013
Token drove
Token ID10357
Feature activation-6.29e-01

Pos logit contributions
Sal+0.005
 ho+0.005
ita+0.005
oxy+0.005
 hus+0.005
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.009
 rubble-0.008
Token away
Token ID1497
Feature activation-6.34e-01

Pos logit contributions
keeping+0.022
handed+0.022
 unatt+0.021
 disl+0.020
taker+0.019
Neg logit contributions
 short-0.008
Sal-0.008
 each-0.007
 ho-0.007
 hus-0.007
Token to
Token ID284
Feature activation-1.66e+00

Pos logit contributions
keeping+0.023
handed+0.022
 unatt+0.021
taker+0.021
 disl+0.020
Neg logit contributions
 short-0.007
 ho-0.006
 long-0.006
 each-0.006
 hus-0.006
Token the
Token ID262
Feature activation-1.49e+00

Pos logit contributions
keeping+0.065
taker+0.061
 unatt+0.060
handed+0.059
ocular+0.056
Neg logit contributions
 each-0.015
 short-0.014
 stretch-0.014
Sal-0.013
 use-0.013
Token charity
Token ID11016
Feature activation-4.57e+00

Pos logit contributions
keeping+0.049
taker+0.042
handed+0.042
 unatt+0.040
ount+0.037
Neg logit contributions
Sal-0.024
 hus-0.023
 short-0.023
 human-0.023
 ho-0.023
Token shop
Token ID6128
Feature activation-8.34e-01

Pos logit contributions
keeping+0.150
handed+0.132
 unatt+0.131
taker+0.129
eway+0.117
Neg logit contributions
Sal-0.071
-0.063
 hus-0.063
 chicken-0.062
 each-0.061
Token.
Token ID13
Feature activation+6.86e-01

Pos logit contributions
keeping+0.031
taker+0.029
 unatt+0.028
handed+0.028
 disl+0.026
Neg logit contributions
 ho-0.008
Sal-0.008
 short-0.008
 each-0.007
 chicken-0.007
Token
Token ID198
Feature activation-1.10e-02

Pos logit contributions
Sal+0.008
 short+0.006
ira+0.006
Ben+0.006
Le+0.006
Neg logit contributions
keeping-0.027
 unatt-0.025
handed-0.025
taker-0.024
 rubble-0.023
Token
Token ID198
Feature activation-1.88e-01

Pos logit contributions
keeping+0.000
 unatt+0.000
handed+0.000
taker+0.000
 rubble+0.000
Neg logit contributions
Sal-0.000
Le-0.000
Ben-0.000
ira-0.000
 short-0.000
TokenL
Token ID43
Feature activation+9.28e-03

Pos logit contributions
keeping+0.007
handed+0.006
 unatt+0.006
taker+0.006
 rubble+0.006
Neg logit contributions
Sal-0.002
Le-0.002
 ho-0.002
 short-0.002
 hus-0.002
Token grandma
Token ID49890
Feature activation-8.28e-01

Pos logit contributions
keeping+0.044
handed+0.039
 unatt+0.039
taker+0.039
 rubble+0.035
Neg logit contributions
Sal-0.015
 ho-0.014
 chicken-0.013
 short-0.013
ita-0.013
Token's
Token ID338
Feature activation-1.58e+00

Pos logit contributions
keeping+0.031
handed+0.028
taker+0.028
 unatt+0.027
 rubble+0.026
Neg logit contributions
Sal-0.009
 chicken-0.009
 ho-0.008
ita-0.008
 stretch-0.008
Token house
Token ID2156
Feature activation-1.18e+00

Pos logit contributions
keeping+0.052
handed+0.047
taker+0.047
 unatt+0.047
eway+0.042
Neg logit contributions
 ho-0.023
Sal-0.023
 chicken-0.022
 human-0.022
 short-0.022
Token for
Token ID329
Feature activation-6.78e-01

Pos logit contributions
keeping+0.044
handed+0.040
taker+0.039
 unatt+0.039
 UFO+0.037
Neg logit contributions
 ho-0.011
 each-0.011
 short-0.011
Sal-0.010
 chicken-0.010
Token safe
Token ID3338
Feature activation-3.94e-01

Pos logit contributions
keeping+0.023
handed+0.021
 unatt+0.021
taker+0.020
ocular+0.018
Neg logit contributions
Sal-0.010
 ho-0.009
 chicken-0.009
 short-0.009
 hus-0.009
Tokenkeeping
Token ID19934
Feature activation-4.54e+00

Pos logit contributions
taker+0.010
handed+0.010
 unatt+0.009
keeping+0.009
ount+0.009
Neg logit contributions
 each-0.008
 ho-0.007
 chicken-0.007
 hus-0.007
 short-0.007
Token.
Token ID13
Feature activation-4.23e-01

Pos logit contributions
keeping+0.166
handed+0.161
taker+0.159
 unatt+0.158
 rubble+0.151
Neg logit contributions
 ho-0.036
Sal-0.034
 short-0.031
 stretch-0.030
 each-0.030
Token
Token ID220
Feature activation+1.24e+00

Pos logit contributions
keeping+0.014
 unatt+0.014
handed+0.012
taker+0.012
 rubble+0.011
Neg logit contributions
Sal-0.007
-0.006
 each-0.005
s-0.005
 short-0.005
Token
Token ID198
Feature activation-3.56e-01

Pos logit contributions
Sal+0.016
oxy+0.012
 ho+0.012
ita+0.012
 short+0.012
Neg logit contributions
keeping-0.046
 unatt-0.042
handed-0.041
taker-0.040
 rubble-0.037
Token
Token ID198
Feature activation-7.64e-01

Pos logit contributions
keeping+0.014
 unatt+0.013
handed+0.013
taker+0.012
 rubble+0.012
Neg logit contributions
Sal-0.004
Le-0.003
 short-0.003
oxy-0.003
Ben-0.003
TokenWhen
Token ID2215
Feature activation+5.29e-01

Pos logit contributions
keeping+0.027
 unatt+0.024
handed+0.024
taker+0.023
 rubble+0.021
Neg logit contributions
Sal-0.012
 short-0.009
oxy-0.009
 hus-0.009
Le-0.009
Token noise
Token ID7838
Feature activation-5.99e-01

Pos logit contributions
keeping+0.014
handed+0.012
 unatt+0.012
taker+0.012
 rubble+0.010
Neg logit contributions
Sal-0.016
oxy-0.014
ita-0.014
 ho-0.014
Le-0.014
Token coming
Token ID2406
Feature activation+4.72e-02

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.017
taker+0.017
 rubble+0.015
Neg logit contributions
Sal-0.010
 ho-0.009
 stretch-0.008
 chicken-0.008
 hus-0.008
Token from
Token ID422
Feature activation-1.71e+00

Pos logit contributions
Sal+0.001
 ho+0.001
 hus+0.001
 stretch+0.001
 chicken+0.001
Neg logit contributions
keeping-0.001
handed-0.001
 unatt-0.001
taker-0.001
 rubble-0.001
Token a
Token ID257
Feature activation-1.54e+00

Pos logit contributions
keeping+0.056
handed+0.051
 unatt+0.050
taker+0.049
ocular+0.044
Neg logit contributions
Sal-0.028
 hus-0.025
 ho-0.025
 chicken-0.024
 stretch-0.024
Token nearby
Token ID6716
Feature activation-1.78e+00

Pos logit contributions
keeping+0.051
handed+0.045
 unatt+0.044
taker+0.043
ocular+0.039
Neg logit contributions
Sal-0.026
 chicken-0.024
 hus-0.023
 short-0.023
 ho-0.023
Token alley
Token ID23978
Feature activation-4.50e+00

Pos logit contributions
keeping+0.054
handed+0.047
 unatt+0.047
taker+0.047
ocular+0.041
Neg logit contributions
Sal-0.033
 chicken-0.032
 hus-0.031
 ho-0.030
 stretch-0.030
Tokenway
Token ID1014
Feature activation-1.89e-01

Pos logit contributions
keeping+0.164
 unatt+0.148
handed+0.147
taker+0.141
 disl+0.131
Neg logit contributions
Sal-0.048
 hus-0.045
 short-0.044
 line-0.042
 stretch-0.042
Token.
Token ID13
Feature activation-8.77e-01

Pos logit contributions
keeping+0.006
handed+0.005
 unatt+0.005
taker+0.005
 rubble+0.005
Neg logit contributions
Sal-0.003
oxy-0.003
 ho-0.003
 chicken-0.003
ita-0.003
Token Ell
Token ID7122
Feature activation+1.01e+00

Pos logit contributions
keeping+0.026
handed+0.023
 unatt+0.023
taker+0.022
 rubble+0.020
Neg logit contributions
Sal-0.017
 ho-0.015
ita-0.015
oxy-0.015
 chicken-0.015
Tokena
Token ID64
Feature activation-2.88e-01

Pos logit contributions
Sal+0.013
 hus+0.011
 stretch+0.011
 ho+0.011
 corn+0.010
Neg logit contributions
keeping-0.036
handed-0.033
 unatt-0.032
taker-0.031
ocular-0.029
Token was
Token ID373
Feature activation-1.17e-01

Pos logit contributions
keeping+0.009
handed+0.008
 unatt+0.008
taker+0.008
ocular+0.007
Neg logit contributions
Sal-0.005
oxy-0.004
 chicken-0.004
 ho-0.004
ita-0.004
Token.
Token ID13
Feature activation+5.84e-01

Pos logit contributions
 ho+0.001
Sal+0.001
 short+0.001
 stretch+0.001
 hus+0.001
Neg logit contributions
keeping-0.003
handed-0.002
 unatt-0.002
taker-0.002
ocular-0.002
Token She
Token ID1375
Feature activation+3.30e-01

Pos logit contributions
Sal+0.009
 ho+0.008
oxy+0.008
ita+0.008
 chicken+0.008
Neg logit contributions
keeping-0.020
 unatt-0.018
handed-0.018
taker-0.017
 rubble-0.016
Token ran
Token ID4966
Feature activation-1.08e+00

Pos logit contributions
Sal+0.005
 ho+0.005
 hus+0.004
ita+0.004
 chicken+0.004
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.010
 rubble-0.009
Token behind
Token ID2157
Feature activation-1.57e-01

Pos logit contributions
keeping+0.036
handed+0.034
 unatt+0.033
taker+0.032
 disl+0.031
Neg logit contributions
 each-0.015
 short-0.015
 ho-0.014
 long-0.014
 five-0.013
Token the
Token ID262
Feature activation-1.34e+00

Pos logit contributions
keeping+0.006
handed+0.005
 unatt+0.005
taker+0.005
ocular+0.005
Neg logit contributions
Sal-0.002
 ho-0.002
 chicken-0.002
 short-0.002
 hus-0.002
Token laundry
Token ID25724
Feature activation-4.47e+00

Pos logit contributions
keeping+0.043
handed+0.040
 unatt+0.039
taker+0.037
 disl+0.034
Neg logit contributions
 chicken-0.022
 hus-0.021
 ho-0.021
 corn-0.021
Sal-0.020
Token.
Token ID13
Feature activation+3.91e-01

Pos logit contributions
keeping+0.135
handed+0.123
 unatt+0.121
taker+0.118
 disl+0.107
Neg logit contributions
Sal-0.075
 chicken-0.072
 ho-0.071
 hus-0.069
 stretch-0.069
Token She
Token ID1375
Feature activation+9.41e-02

Pos logit contributions
Sal+0.006
 ho+0.005
oxy+0.005
 hus+0.005
Le+0.005
Neg logit contributions
keeping-0.014
handed-0.013
 unatt-0.013
taker-0.012
 rubble-0.011
Token thought
Token ID1807
Feature activation-3.87e-01

Pos logit contributions
Sal+0.002
ita+0.001
oxy+0.001
 ho+0.001
 hus+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.002
Token Ben
Token ID3932
Feature activation-3.80e-01

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.011
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.007
 ho-0.006
 hus-0.006
 chicken-0.006
ita-0.006
Token could
Token ID714
Feature activation+1.29e+00

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.011
taker+0.011
ocular+0.010
Neg logit contributions
Sal-0.006
oxy-0.006
ita-0.005
 ho-0.005
 chicken-0.005
Token.
Token ID13
Feature activation+2.61e-01

Pos logit contributions
keeping+0.045
handed+0.041
taker+0.041
 unatt+0.041
 rubble+0.039
Neg logit contributions
Sal-0.013
 ho-0.013
 chicken-0.013
 each-0.012
 short-0.012
Token They
Token ID1119
Feature activation+3.98e-01

Pos logit contributions
Sal+0.005
oxy+0.004
 ho+0.004
 chicken+0.004
+0.004
Neg logit contributions
keeping-0.008
 unatt-0.008
handed-0.008
taker-0.007
 rubble-0.007
Token belong
Token ID5594
Feature activation-1.86e+00

Pos logit contributions
Sal+0.006
 ho+0.006
 stretch+0.005
 chicken+0.005
 each+0.005
Neg logit contributions
keeping-0.014
handed-0.012
 unatt-0.012
taker-0.012
eway-0.011
Token to
Token ID284
Feature activation-1.65e+00

Pos logit contributions
keeping+0.063
handed+0.058
taker+0.058
 unatt+0.056
 rubble+0.052
Neg logit contributions
Sal-0.025
 ho-0.023
 chicken-0.023
 each-0.022
 short-0.022
Token a
Token ID257
Feature activation-2.07e+00

Pos logit contributions
keeping+0.054
handed+0.049
 unatt+0.048
taker+0.046
 rubble+0.042
Neg logit contributions
Sal-0.028
ita-0.024
 ho-0.024
 hus-0.023
 stretch-0.023
Token charity
Token ID11016
Feature activation-4.44e+00

Pos logit contributions
keeping+0.073
handed+0.064
 unatt+0.064
taker+0.062
ocular+0.057
Neg logit contributions
Sal-0.031
 chicken-0.028
 short-0.027
 ho-0.027
 hus-0.026
Token that
Token ID326
Feature activation-1.70e+00

Pos logit contributions
keeping+0.144
handed+0.128
 unatt+0.124
taker+0.123
 rubble+0.114
Neg logit contributions
Sal-0.073
 chicken-0.065
-0.064
 ho-0.063
 hus-0.062
Token helps
Token ID5419
Feature activation-4.37e-01

Pos logit contributions
keeping+0.058
handed+0.052
 unatt+0.051
taker+0.050
 rubble+0.046
Neg logit contributions
Sal-0.025
 chicken-0.023
 ho-0.022
 stretch-0.022
 hus-0.021
Token poor
Token ID3595
Feature activation-1.44e+00

Pos logit contributions
keeping+0.015
handed+0.014
 unatt+0.014
taker+0.013
 rubble+0.012
Neg logit contributions
Sal-0.006
 ho-0.005
ita-0.005
oxy-0.005
 hus-0.005
Token kids
Token ID3988
Feature activation-5.34e-01

Pos logit contributions
keeping+0.047
 unatt+0.043
handed+0.043
taker+0.042
 rubble+0.038
Neg logit contributions
Sal-0.023
 chicken-0.021
 ho-0.021
 hus-0.021
 corn-0.020
Token.
Token ID13
Feature activation+2.95e-01

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.009
 ho-0.008
 chicken-0.008
 stretch-0.008
 short-0.008
Token.
Token ID13
Feature activation+1.09e+00

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.011
 short-0.010
 ho-0.010
 chicken-0.010
 stretch-0.010
Token It
Token ID632
Feature activation+8.10e-01

Pos logit contributions
Sal+0.019
 ho+0.017
oxy+0.017
ita+0.016
 chicken+0.016
Neg logit contributions
keeping-0.035
handed-0.032
 unatt-0.032
taker-0.030
 rubble-0.028
Token belongs
Token ID14448
Feature activation-7.18e-01

Pos logit contributions
Sal+0.012
 ho+0.010
oxy+0.010
ita+0.009
 hus+0.009
Neg logit contributions
keeping-0.029
handed-0.026
 unatt-0.026
taker-0.025
 rubble-0.023
Token to
Token ID284
Feature activation-1.52e+00

Pos logit contributions
keeping+0.024
handed+0.021
 unatt+0.021
taker+0.021
 rubble+0.019
Neg logit contributions
Sal-0.012
 ho-0.010
ita-0.010
oxy-0.010
 hus-0.010
Token the
Token ID262
Feature activation-1.26e+00

Pos logit contributions
keeping+0.050
handed+0.045
 unatt+0.044
taker+0.042
 rubble+0.039
Neg logit contributions
Sal-0.026
 ho-0.022
ita-0.021
 hus-0.021
oxy-0.021
Token charity
Token ID11016
Feature activation-4.43e+00

Pos logit contributions
keeping+0.043
handed+0.038
taker+0.037
 unatt+0.037
eway+0.033
Neg logit contributions
Sal-0.019
 chicken-0.018
 ho-0.018
 hus-0.018
 human-0.017
Token.
Token ID13
Feature activation+1.28e+00

Pos logit contributions
keeping+0.141
handed+0.125
 unatt+0.123
taker+0.117
 rubble+0.111
Neg logit contributions
Sal-0.076
 chicken-0.068
-0.067
 hus-0.066
 ho-0.065
Token We
Token ID775
Feature activation+3.61e-01

Pos logit contributions
Sal+0.023
oxy+0.020
 ho+0.020
ita+0.019
 chicken+0.019
Neg logit contributions
keeping-0.041
handed-0.037
 unatt-0.037
taker-0.035
 rubble-0.033
Token provide
Token ID2148
Feature activation-4.97e-01

Pos logit contributions
Sal+0.005
 ho+0.005
 stretch+0.005
 hus+0.005
 chicken+0.005
Neg logit contributions
keeping-0.013
handed-0.011
 unatt-0.011
taker-0.010
reating-0.010
Token clothes
Token ID8242
Feature activation+1.23e+00

Pos logit contributions
keeping+0.016
handed+0.015
 unatt+0.015
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.008
 ho-0.007
ita-0.007
oxy-0.007
 hus-0.007
Token and
Token ID290
Feature activation+4.26e-01

Pos logit contributions
Sal+0.016
 chicken+0.016
 ho+0.016
 each+0.016
 short+0.015
Neg logit contributions
keeping-0.041
handed-0.039
 unatt-0.037
taker-0.035
 rubble-0.035
Token hole
Token ID7604
Feature activation-1.70e+00

Pos logit contributions
keeping+0.013
handed+0.011
 unatt+0.011
taker+0.011
reating+0.009
Neg logit contributions
Sal-0.009
 ho-0.008
 hus-0.008
 short-0.008
 chicken-0.008
Token and
Token ID290
Feature activation-6.17e-01

Pos logit contributions
keeping+0.066
handed+0.061
 unatt+0.059
taker+0.057
 rubble+0.053
Neg logit contributions
Sal-0.018
 ho-0.016
 short-0.016
 stretch-0.015
ita-0.015
Token found
Token ID1043
Feature activation-4.54e-01

Pos logit contributions
keeping+0.024
handed+0.022
taker+0.021
 unatt+0.021
reating+0.020
Neg logit contributions
Sal-0.007
 ho-0.007
 stretch-0.007
 stretched-0.006
 short-0.006
Token a
Token ID257
Feature activation-5.69e-01

Pos logit contributions
keeping+0.020
 unatt+0.018
handed+0.018
taker+0.016
 disl+0.016
Neg logit contributions
Sal-0.003
 short-0.003
 chicken-0.003
 each-0.003
 ho-0.003
Token dirty
Token ID11841
Feature activation-1.50e+00

Pos logit contributions
keeping+0.018
handed+0.016
 unatt+0.016
taker+0.016
 rubble+0.014
Neg logit contributions
Sal-0.010
 ho-0.009
ita-0.008
oxy-0.008
 hus-0.008
Token alley
Token ID23978
Feature activation-4.42e+00

Pos logit contributions
keeping+0.042
handed+0.037
 unatt+0.037
taker+0.035
 rubble+0.032
Neg logit contributions
Sal-0.033
oxy-0.029
ita-0.029
 ho-0.028
Le-0.028
Token with
Token ID351
Feature activation+3.51e-01

Pos logit contributions
keeping+0.173
handed+0.156
 unatt+0.154
taker+0.150
 disl+0.146
Neg logit contributions
 short-0.034
 line-0.032
 long-0.031
 stretch-0.028
 each-0.028
Token trash
Token ID13913
Feature activation-2.11e+00

Pos logit contributions
Sal+0.004
 short+0.003
 chicken+0.003
 ho+0.003
 hus+0.003
Neg logit contributions
keeping-0.013
 unatt-0.012
handed-0.012
taker-0.012
makers-0.011
Token and
Token ID290
Feature activation-7.42e-01

Pos logit contributions
keeping+0.068
handed+0.063
 unatt+0.060
taker+0.059
eway+0.053
Neg logit contributions
Sal-0.035
 ho-0.033
 hus-0.032
 chicken-0.032
 stretch-0.030
Token rats
Token ID13623
Feature activation-9.75e-01

Pos logit contributions
keeping+0.024
handed+0.023
 unatt+0.022
taker+0.022
ount+0.020
Neg logit contributions
 chicken-0.011
 short-0.011
Sal-0.010
 corn-0.010
 ho-0.010
Token.
Token ID13
Feature activation+6.03e-02

Pos logit contributions
keeping+0.036
handed+0.035
 unatt+0.034
taker+0.033
reating+0.030
Neg logit contributions
 ho-0.011
Sal-0.010
 hus-0.009
 stretch-0.009
 short-0.009
Token She
Token ID1375
Feature activation+7.20e-02

Pos logit contributions
Sal+0.004
 ho+0.004
ita+0.004
 hus+0.004
 chicken+0.004
Neg logit contributions
keeping-0.008
handed-0.008
 unatt-0.007
taker-0.007
 rubble-0.007
Token was
Token ID373
Feature activation-4.55e-01

Pos logit contributions
Sal+0.001
+0.001
ita+0.001
+0.001
oxy+0.001
Neg logit contributions
keeping-0.002
 unatt-0.002
handed-0.002
taker-0.002
 rubble-0.002
Token stuck
Token ID7819
Feature activation-6.26e-01

Pos logit contributions
keeping+0.012
 unatt+0.011
handed+0.011
taker+0.011
 rubble+0.010
Neg logit contributions
Sal-0.010
oxy-0.009
ita-0.009
 ho-0.009
 hus-0.009
Token behind
Token ID2157
Feature activation-1.52e-01

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.018
taker+0.017
 rubble+0.016
Neg logit contributions
Sal-0.011
 ho-0.010
ita-0.009
oxy-0.009
 hus-0.009
Token the
Token ID262
Feature activation-1.08e+00

Pos logit contributions
keeping+0.005
handed+0.005
 unatt+0.005
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.002
ita-0.002
 ho-0.002
oxy-0.002
 hus-0.002
Token laundry
Token ID25724
Feature activation-4.39e+00

Pos logit contributions
keeping+0.031
 unatt+0.028
handed+0.027
taker+0.027
 rubble+0.025
Neg logit contributions
Sal-0.023
ita-0.020
oxy-0.020
 ho-0.019
Le-0.019
Token.
Token ID13
Feature activation+3.88e-01

Pos logit contributions
keeping+0.135
handed+0.122
 unatt+0.119
taker+0.116
 rubble+0.104
Neg logit contributions
Sal-0.078
 ho-0.071
 chicken-0.071
 hus-0.070
 stretch-0.069
Token
Token ID198
Feature activation-1.22e-01

Pos logit contributions
Sal+0.006
 ho+0.006
ita+0.006
 hus+0.005
 chicken+0.005
Neg logit contributions
keeping-0.013
handed-0.011
 unatt-0.011
taker-0.011
 rubble-0.010
Token
Token ID198
Feature activation-1.75e-01

Pos logit contributions
keeping+0.004
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.003
Neg logit contributions
Sal-0.002
 ho-0.002
ita-0.002
 chicken-0.002
 hus-0.002
TokenBen
Token ID11696
Feature activation-1.02e+00

Pos logit contributions
keeping+0.006
handed+0.006
 unatt+0.005
taker+0.005
makers+0.005
Neg logit contributions
Sal-0.002
 ho-0.002
 chicken-0.002
ita-0.002
 stretch-0.002
Token laughed
Token ID13818
Feature activation-4.32e-01

Pos logit contributions
keeping+0.028
 unatt+0.026
handed+0.025
taker+0.024
 rubble+0.022
Neg logit contributions
Sal-0.020
ita-0.018
oxy-0.018
-0.018
 corn-0.017
TokenTom
Token ID13787
Feature activation-3.56e-01

Pos logit contributions
keeping+0.031
handed+0.029
 unatt+0.028
taker+0.027
ocular+0.025
Neg logit contributions
Sal-0.013
 ho-0.012
 chicken-0.012
ita-0.012
oxy-0.011
Token was
Token ID373
Feature activation-6.60e-01

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.010
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.006
 ho-0.005
 hus-0.005
oxy-0.005
 chicken-0.005
Token walking
Token ID6155
Feature activation-7.88e-01

Pos logit contributions
keeping+0.019
handed+0.016
 unatt+0.016
taker+0.016
 rubble+0.014
Neg logit contributions
Sal-0.014
 ho-0.013
ita-0.012
oxy-0.012
 hus-0.012
Token down
Token ID866
Feature activation+8.90e-01

Pos logit contributions
keeping+0.025
handed+0.022
 unatt+0.022
taker+0.021
 rubble+0.019
Neg logit contributions
Sal-0.014
 ho-0.013
 short-0.013
 stretch-0.013
 hus-0.013
Token an
Token ID281
Feature activation-1.05e+00

Pos logit contributions
Sal+0.014
 ho+0.011
 hus+0.011
ita+0.011
 chicken+0.011
Neg logit contributions
keeping-0.031
handed-0.028
 unatt-0.027
taker-0.027
 rubble-0.024
Token alley
Token ID23978
Feature activation-4.34e+00

Pos logit contributions
keeping+0.032
handed+0.027
 unatt+0.027
taker+0.026
ocular+0.024
Neg logit contributions
Sal-0.020
ita-0.019
 chicken-0.018
 short-0.018
 ho-0.018
Token and
Token ID290
Feature activation-9.45e-01

Pos logit contributions
keeping+0.136
handed+0.117
 unatt+0.115
taker+0.112
 disl+0.100
Neg logit contributions
Sal-0.074
 short-0.072
 stretch-0.068
ita-0.067
 hus-0.066
Token he
Token ID339
Feature activation+3.32e-01

Pos logit contributions
keeping+0.037
handed+0.033
taker+0.032
 unatt+0.032
reating+0.030
Neg logit contributions
Sal-0.010
 short-0.009
 ho-0.008
 stretch-0.008
-0.008
Token heard
Token ID2982
Feature activation+5.45e-02

Pos logit contributions
Sal+0.005
oxy+0.005
 ho+0.005
ita+0.004
 hus+0.004
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.010
 rubble-0.009
Token a
Token ID257
Feature activation+3.27e-01

Pos logit contributions
 short+0.000
 long+0.000
 chicken+0.000
 five+0.000
 corn+0.000
Neg logit contributions
keeping-0.003
 unatt-0.002
handed-0.002
taker-0.002
ocular-0.002
Token loud
Token ID7812
Feature activation-9.01e-01

Pos logit contributions
Sal+0.006
ita+0.006
oxy+0.006
 ho+0.005
Le+0.005
Neg logit contributions
keeping-0.009
 unatt-0.008
handed-0.008
taker-0.008
 rubble-0.008
Token to
Token ID284
Feature activation-7.96e-01

Pos logit contributions
keeping+0.034
handed+0.031
 unatt+0.031
taker+0.031
 rubble+0.028
Neg logit contributions
Sal-0.010
 ho-0.009
 short-0.009
 chicken-0.009
 hus-0.008
Token take
Token ID1011
Feature activation+2.07e-02

Pos logit contributions
keeping+0.029
 unatt+0.027
handed+0.027
taker+0.026
ocular+0.024
Neg logit contributions
Sal-0.010
 stretch-0.009
 ho-0.009
 short-0.008
 chicken-0.008
Token them
Token ID606
Feature activation+8.08e-03

Pos logit contributions
Sal+0.000
 short+0.000
 chicken+0.000
 ho+0.000
 each+0.000
Neg logit contributions
keeping-0.001
 unatt-0.001
handed-0.001
taker-0.001
 rubble-0.001
Token to
Token ID284
Feature activation-1.31e+00

Pos logit contributions
 each+0.000
 ho+0.000
 short+0.000
 long+0.000
 chicken+0.000
Neg logit contributions
keeping-0.000
taker-0.000
handed-0.000
 disl-0.000
 UFO-0.000
Token the
Token ID262
Feature activation-1.12e+00

Pos logit contributions
keeping+0.045
handed+0.040
 unatt+0.040
taker+0.039
 rubble+0.035
Neg logit contributions
Sal-0.020
 ho-0.017
ita-0.017
oxy-0.017
 hus-0.017
Token charity
Token ID11016
Feature activation-4.29e+00

Pos logit contributions
keeping+0.034
handed+0.029
taker+0.029
 unatt+0.029
ocular+0.025
Neg logit contributions
Sal-0.022
 chicken-0.020
 ho-0.019
 hus-0.019
 short-0.019
Token shop
Token ID6128
Feature activation-3.39e-01

Pos logit contributions
keeping+0.136
handed+0.119
 unatt+0.117
taker+0.116
 rubble+0.105
Neg logit contributions
Sal-0.076
 chicken-0.066
 ho-0.065
 hus-0.065
ita-0.064
Token,
Token ID11
Feature activation+5.99e-01

Pos logit contributions
keeping+0.012
taker+0.011
handed+0.011
 unatt+0.011
 rubble+0.010
Neg logit contributions
Sal-0.004
 ho-0.004
 short-0.004
 chicken-0.004
ita-0.004
Token where
Token ID810
Feature activation+2.15e+00

Pos logit contributions
Sal+0.010
 ho+0.010
 chicken+0.009
 each+0.009
 stretch+0.009
Neg logit contributions
keeping-0.020
handed-0.018
taker-0.017
 unatt-0.016
 rubble-0.015
Token other
Token ID584
Feature activation+3.73e-01

Pos logit contributions
Sal+0.024
 chicken+0.023
 short+0.022
 each+0.022
ira+0.021
Neg logit contributions
keeping-0.080
handed-0.074
 unatt-0.073
taker-0.068
 rubble-0.065
Token children
Token ID1751
Feature activation-6.94e-01

Pos logit contributions
Sal+0.006
 chicken+0.005
 ho+0.005
 hus+0.005
 short+0.005
Neg logit contributions
keeping-0.013
handed-0.011
 unatt-0.011
taker-0.011
 rubble-0.010
Token of
Token ID286
Feature activation-9.43e-01

Pos logit contributions
Sal+0.006
 ho+0.006
 chicken+0.006
ita+0.006
 hus+0.006
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.008
 rubble-0.007
Token people
Token ID661
Feature activation+1.50e-01

Pos logit contributions
keeping+0.030
handed+0.027
 unatt+0.027
taker+0.026
 rubble+0.023
Neg logit contributions
Sal-0.017
 ho-0.014
ita-0.014
oxy-0.014
 hus-0.014
Token coming
Token ID2406
Feature activation+2.68e-01

Pos logit contributions
Sal+0.002
 ho+0.002
 chicken+0.002
ita+0.002
 hus+0.002
Neg logit contributions
keeping-0.005
handed-0.005
 unatt-0.005
taker-0.004
 rubble-0.004
Token onto
Token ID4291
Feature activation-1.07e-01

Pos logit contributions
 ho+0.004
Sal+0.004
 short+0.004
 hus+0.004
 each+0.004
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.008
 rubble-0.007
Token my
Token ID616
Feature activation-2.00e+00

Pos logit contributions
keeping+0.004
handed+0.003
 unatt+0.003
taker+0.003
ocular+0.003
Neg logit contributions
Sal-0.002
 chicken-0.001
 ho-0.001
 hus-0.001
 short-0.001
Token property
Token ID3119
Feature activation-4.27e+00

Pos logit contributions
keeping+0.061
handed+0.054
 unatt+0.054
taker+0.052
 rubble+0.047
Neg logit contributions
Sal-0.038
 ho-0.034
ita-0.033
oxy-0.033
 hus-0.033
Token.
Token ID13
Feature activation+3.63e-01

Pos logit contributions
keeping+0.147
handed+0.133
 unatt+0.129
taker+0.128
 disl+0.119
Neg logit contributions
Sal-0.055
 ho-0.051
 short-0.051
 hus-0.050
 stretch-0.049
Token I
Token ID314
Feature activation+6.29e-01

Pos logit contributions
Sal+0.007
 ho+0.006
oxy+0.006
ita+0.006
 chicken+0.006
Neg logit contributions
keeping-0.011
handed-0.010
 unatt-0.010
taker-0.009
 rubble-0.008
Token suggest
Token ID1950
Feature activation+9.67e-01

Pos logit contributions
Sal+0.011
 ho+0.009
ita+0.009
oxy+0.009
 hus+0.009
Neg logit contributions
keeping-0.021
handed-0.018
 unatt-0.018
taker-0.018
 rubble-0.016
Token you
Token ID345
Feature activation-3.34e-01

Pos logit contributions
Sal+0.018
 ho+0.016
ita+0.016
oxy+0.016
 hus+0.016
Neg logit contributions
keeping-0.030
handed-0.026
 unatt-0.026
taker-0.025
 rubble-0.023
Token leave
Token ID2666
Feature activation-7.67e-01

Pos logit contributions
keeping+0.011
handed+0.010
 unatt+0.010
taker+0.009
 rubble+0.009
Neg logit contributions
Sal-0.006
oxy-0.005
 ho-0.005
ita-0.005
 hus-0.005
Token in
Token ID287
Feature activation+5.67e-01

Pos logit contributions
keeping+0.033
handed+0.029
 unatt+0.028
taker+0.026
 disl+0.025
Neg logit contributions
 short-0.019
Sal-0.019
 ho-0.018
 long-0.018
 hus-0.018
Token circles
Token ID13332
Feature activation+8.70e-01

Pos logit contributions
 short+0.005
 long+0.005
 ho+0.005
 hus+0.005
 stretch+0.005
Neg logit contributions
keeping-0.022
handed-0.021
 unatt-0.021
taker-0.021
 disl-0.019
Token and
Token ID290
Feature activation-6.56e-02

Pos logit contributions
Sal+0.012
 ho+0.011
 short+0.011
 stretch+0.011
ita+0.011
Neg logit contributions
keeping-0.031
handed-0.027
 unatt-0.027
taker-0.026
 rubble-0.024
Token ran
Token ID4966
Feature activation-1.06e+00

Pos logit contributions
keeping+0.002
handed+0.002
taker+0.002
 unatt+0.002
ount+0.002
Neg logit contributions
 ho-0.001
 short-0.001
 long-0.001
 stretch-0.001
Sal-0.001
Token into
Token ID656
Feature activation+9.14e-01

Pos logit contributions
keeping+0.036
handed+0.034
 unatt+0.033
taker+0.031
 disl+0.030
Neg logit contributions
 short-0.015
 long-0.015
 ho-0.014
 each-0.014
 hus-0.013
Token alley
Token ID23978
Feature activation-4.16e+00

Pos logit contributions
Sal+0.010
 hus+0.009
 short+0.009
 chicken+0.009
 ho+0.009
Neg logit contributions
keeping-0.034
 unatt-0.032
handed-0.032
taker-0.031
ocular-0.028
Tokenways
Token ID1322
Feature activation+1.36e+00

Pos logit contributions
keeping+0.145
 unatt+0.131
handed+0.127
taker+0.125
 disl+0.112
Neg logit contributions
Sal-0.052
 short-0.051
 hus-0.048
 long-0.046
ita-0.045
Token,
Token ID11
Feature activation-4.13e-01

Pos logit contributions
Sal+0.018
 ho+0.017
 short+0.016
 hus+0.016
ita+0.016
Neg logit contributions
keeping-0.048
handed-0.043
 unatt-0.043
taker-0.042
 rubble-0.038
Token but
Token ID475
Feature activation-1.88e+00

Pos logit contributions
keeping+0.014
handed+0.012
taker+0.012
 unatt+0.011
reating+0.011
Neg logit contributions
 ho-0.007
Sal-0.007
 short-0.007
 stretch-0.006
 hus-0.006
Token nothing
Token ID2147
Feature activation-3.56e-01

Pos logit contributions
keeping+0.064
handed+0.057
 unatt+0.053
taker+0.053
 disl+0.049
Neg logit contributions
Sal-0.030
 ho-0.029
 short-0.028
 stretch-0.027
Le-0.026
Token looked
Token ID3114
Feature activation-6.92e-01

Pos logit contributions
keeping+0.011
handed+0.010
 unatt+0.010
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.007
 ho-0.006
ita-0.006
oxy-0.006
 hus-0.006
Token soft
Token ID2705
Feature activation-3.34e-01

Pos logit contributions
keeping+0.008
handed+0.007
 unatt+0.007
taker+0.007
 rubble+0.006
Neg logit contributions
Sal-0.004
 short-0.003
 stretch-0.003
 ho-0.003
 hus-0.003
Token!
Token ID0
Feature activation+1.39e-01

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.011
taker+0.011
 rubble+0.010
Neg logit contributions
Sal-0.004
 ho-0.003
 chicken-0.003
 stretch-0.003
 hus-0.003
Token<|endoftext|>
Token ID50256
Feature activation-2.30e+00

Pos logit contributions
Sal+0.003
 ho+0.003
 hus+0.002
 short+0.002
ita+0.002
Neg logit contributions
keeping-0.004
handed-0.004
taker-0.003
 unatt-0.003
ocular-0.003
TokenOnce
Token ID7454
Feature activation-1.63e+00

Pos logit contributions
keeping+0.084
 unatt+0.077
handed+0.077
taker+0.074
 rubble+0.069
Neg logit contributions
Sal-0.030
 ho-0.023
 hus-0.023
oxy-0.023
 chicken-0.023
Token upon
Token ID2402
Feature activation+2.91e-01

Pos logit contributions
keeping+0.060
handed+0.053
 unatt+0.052
taker+0.050
 rubble+0.047
Neg logit contributions
Sal-0.022
 ho-0.019
 short-0.019
ira-0.019
ita-0.019
Token a
Token ID257
Feature activation+2.79e+00

Pos logit contributions
Sal+0.005
 ho+0.004
ita+0.004
 hus+0.004
oxy+0.004
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.008
 rubble-0.008
Token time
Token ID640
Feature activation-1.21e-01

Pos logit contributions
Sal+0.056
 hus+0.050
+0.048
 chicken+0.047
+0.046
Neg logit contributions
 unatt-0.065
handed-0.064
hell-0.061
keeping-0.061
 anymore-0.060
Token there
Token ID612
Feature activation-1.31e+00

Pos logit contributions
keeping+0.005
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.002
 ho-0.001
 short-0.001
 stretch-0.001
ira-0.001
Token was
Token ID373
Feature activation+6.87e-01

Pos logit contributions
keeping+0.050
handed+0.045
 unatt+0.045
taker+0.043
 rubble+0.040
Neg logit contributions
Sal-0.016
 ho-0.013
 stretch-0.012
 chicken-0.012
ira-0.012
Token a
Token ID257
Feature activation+1.39e+00

Pos logit contributions
Sal+0.013
 hus+0.011
+0.011
 ho+0.011
+0.011
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.018
taker-0.018
 rubble-0.017
Token little
Token ID1310
Feature activation+2.55e-01

Pos logit contributions
Sal+0.031
ita+0.027
oxy+0.026
+0.026
 stretches+0.026
Neg logit contributions
 unatt-0.029
taker-0.028
handed-0.028
keeping-0.027
 wreck-0.026
Token the
Token ID262
Feature activation-1.86e-01

Pos logit contributions
Sal+0.007
ita+0.006
oxy+0.006
 ho+0.006
+0.006
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.011
taker-0.010
 rubble-0.009
Token store
Token ID3650
Feature activation-9.88e-01

Pos logit contributions
keeping+0.005
handed+0.005
 unatt+0.005
taker+0.005
 rubble+0.004
Neg logit contributions
Sal-0.004
ita-0.003
oxy-0.003
 ho-0.003
ira-0.003
Token and
Token ID290
Feature activation-5.70e-01

Pos logit contributions
keeping+0.033
handed+0.030
 unatt+0.029
taker+0.028
ocular+0.026
Neg logit contributions
Sal-0.016
oxy-0.014
ita-0.014
 ho-0.014
 hus-0.013
Token bought
Token ID5839
Feature activation+4.23e-01

Pos logit contributions
keeping+0.019
handed+0.017
 unatt+0.017
taker+0.016
 rubble+0.015
Neg logit contributions
Sal-0.010
ita-0.008
oxy-0.008
 ho-0.008
 hus-0.008
Token some
Token ID617
Feature activation+1.35e+00

Pos logit contributions
Sal+0.007
ita+0.006
 ho+0.006
oxy+0.006
 hus+0.006
Neg logit contributions
keeping-0.014
handed-0.012
 unatt-0.012
taker-0.012
 rubble-0.011
Token fish
Token ID5916
Feature activation+2.64e+00

Pos logit contributions
Sal+0.031
Le+0.029
+0.029
ita+0.029
oxy+0.028
Neg logit contributions
keeping-0.031
taker-0.031
handed-0.029
 unatt-0.028
 rubble-0.027
Token.
Token ID13
Feature activation-7.79e-01

Pos logit contributions
Sal+0.038
 ho+0.033
 chicken+0.033
ita+0.031
 hus+0.031
Neg logit contributions
keeping-0.093
handed-0.084
 unatt-0.084
taker-0.080
 rubble-0.075
Token He
Token ID679
Feature activation+4.55e-01

Pos logit contributions
keeping+0.024
handed+0.021
 unatt+0.020
taker+0.020
 rubble+0.018
Neg logit contributions
Sal-0.015
ita-0.013
 ho-0.013
oxy-0.013
 hus-0.013
Token took
Token ID1718
Feature activation+1.01e+00

Pos logit contributions
Sal+0.010
ita+0.010
Hug+0.010
oxy+0.010
+0.009
Neg logit contributions
keeping-0.010
taker-0.009
handed-0.009
 unatt-0.008
 rubble-0.008
Token them
Token ID606
Feature activation+1.37e-01

Pos logit contributions
Sal+0.015
 ho+0.013
 chicken+0.013
 hus+0.013
 short+0.013
Neg logit contributions
keeping-0.035
handed-0.032
 unatt-0.032
taker-0.030
 rubble-0.028
Token home
Token ID1363
Feature activation+6.30e-01

Pos logit contributions
Sal+0.002
 ho+0.002
 chicken+0.002
ita+0.002
 hus+0.002
Neg logit contributions
keeping-0.005
handed-0.004
 unatt-0.004
taker-0.004
 rubble-0.004
Token happy
Token ID3772
Feature activation-4.56e-01

Pos logit contributions
Sal+0.000
 ho+0.000
 short+0.000
 chicken+0.000
 hus+0.000
Neg logit contributions
keeping-0.000
handed-0.000
 unatt-0.000
taker-0.000
 rubble-0.000
Token.
Token ID13
Feature activation-7.59e-02

Pos logit contributions
keeping+0.016
handed+0.015
 unatt+0.015
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.005
 ho-0.005
 chicken-0.005
 hus-0.004
 stretch-0.004
Token
Token ID198
Feature activation+6.31e-02

Pos logit contributions
keeping+0.002
handed+0.002
taker+0.002
 unatt+0.002
ocular+0.002
Neg logit contributions
Sal-0.001
 ho-0.001
 hus-0.001
ita-0.001
 chicken-0.001
Token
Token ID198
Feature activation-8.92e-02

Pos logit contributions
Sal+0.001
 ho+0.001
ita+0.001
 chicken+0.001
 hus+0.001
Neg logit contributions
keeping-0.002
handed-0.002
taker-0.002
 unatt-0.002
 rubble-0.002
TokenAnd
Token ID1870
Feature activation+5.07e-01

Pos logit contributions
keeping+0.003
handed+0.002
taker+0.002
 unatt+0.002
eway+0.002
Neg logit contributions
 ho-0.002
Sal-0.002
ira-0.002
 chicken-0.001
ita-0.001
Token so
Token ID523
Feature activation+2.42e+00

Pos logit contributions
 each+0.010
 ho+0.009
 stretch+0.008
 long+0.008
 short+0.008
Neg logit contributions
keeping-0.016
handed-0.015
 unatt-0.015
 rubble-0.014
taker-0.014
Token the
Token ID262
Feature activation+2.25e+00

Pos logit contributions
Sal+0.028
 ho+0.027
 stretch+0.025
+0.025
 short+0.025
Neg logit contributions
keeping-0.094
handed-0.086
 unatt-0.084
taker-0.078
reating-0.075
Token tired
Token ID10032
Feature activation+1.22e+00

Pos logit contributions
Sal+0.039
 ho+0.036
 chicken+0.035
 short+0.035
 corn+0.035
Neg logit contributions
keeping-0.075
handed-0.066
 unatt-0.064
taker-0.062
reating-0.055
Token root
Token ID6808
Feature activation+1.14e+00

Pos logit contributions
Sal+0.019
 chicken+0.018
 ho+0.018
 corn+0.018
 hus+0.017
Neg logit contributions
keeping-0.041
handed-0.037
 unatt-0.037
taker-0.034
 rubble-0.031
Token was
Token ID373
Feature activation-3.61e-01

Pos logit contributions
Sal+0.018
 ho+0.016
 chicken+0.016
 stretch+0.016
ita+0.015
Neg logit contributions
keeping-0.038
handed-0.034
 unatt-0.034
taker-0.032
 rubble-0.030
Token sold
Token ID2702
Feature activation-3.45e-01

Pos logit contributions
keeping+0.011
handed+0.009
 unatt+0.009
taker+0.009
 rubble+0.008
Neg logit contributions
Sal-0.007
 ho-0.007
 short-0.006
 chicken-0.006
 stretch-0.006
Token quiz
Token ID38964
Feature activation-8.08e-01

Pos logit contributions
keeping+0.014
handed+0.012
 unatt+0.012
taker+0.012
 rubble+0.011
Neg logit contributions
Sal-0.007
 ho-0.006
 chicken-0.006
ita-0.006
 stretch-0.006
Token!"
Token ID2474
Feature activation+1.58e+00

Pos logit contributions
keeping+0.024
handed+0.021
 unatt+0.021
taker+0.020
 rubble+0.019
Neg logit contributions
Sal-0.016
 stretch-0.014
 chicken-0.014
 short-0.014
 ho-0.014
Token<|endoftext|>
Token ID50256
Feature activation-1.65e+00

Pos logit contributions
Sal+0.029
 ho+0.025
 hus+0.025
ita+0.024
oxy+0.024
Neg logit contributions
keeping-0.050
handed-0.045
 unatt-0.044
taker-0.043
 rubble-0.039
TokenOnce
Token ID7454
Feature activation-1.64e+00

Pos logit contributions
keeping+0.058
handed+0.052
 unatt+0.051
taker+0.049
 rubble+0.045
Neg logit contributions
Sal-0.024
 ho-0.021
ita-0.021
oxy-0.020
 chicken-0.020
Token upon
Token ID2402
Feature activation+8.48e-02

Pos logit contributions
keeping+0.060
handed+0.054
 unatt+0.053
taker+0.050
 rubble+0.048
Neg logit contributions
Sal-0.022
 ho-0.019
 short-0.019
ira-0.019
ita-0.018
Token a
Token ID257
Feature activation+2.57e+00

Pos logit contributions
Sal+0.001
 ho+0.001
ita+0.001
 hus+0.001
oxy+0.001
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.002
Token time
Token ID640
Feature activation-4.61e-01

Pos logit contributions
Sal+0.049
 hus+0.043
 chicken+0.041
+0.040
 pap+0.039
Neg logit contributions
 unatt-0.066
handed-0.066
keeping-0.064
 anymore-0.059
hell-0.059
Token there
Token ID612
Feature activation-1.65e+00

Pos logit contributions
keeping+0.018
handed+0.016
 unatt+0.015
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.005
 ho-0.005
 short-0.005
-0.005
ira-0.005
Token was
Token ID373
Feature activation+4.68e-01

Pos logit contributions
keeping+0.063
 unatt+0.056
handed+0.056
taker+0.055
 rubble+0.051
Neg logit contributions
Sal-0.020
 stretch-0.016
 ho-0.016
 chicken-0.015
 hus-0.015
Token a
Token ID257
Feature activation+1.35e+00

Pos logit contributions
Sal+0.008
 hus+0.007
 ho+0.007
ita+0.007
+0.007
Neg logit contributions
keeping-0.015
handed-0.014
 unatt-0.013
taker-0.013
 rubble-0.012
Token small
Token ID1402
Feature activation+6.41e-01

Pos logit contributions
Sal+0.029
ita+0.026
oxy+0.025
 ho+0.024
 stretches+0.024
Neg logit contributions
 unatt-0.030
taker-0.029
handed-0.028
keeping-0.028
 wreck-0.026
Token girl
Token ID2576
Feature activation-8.32e-01

Pos logit contributions
keeping+0.037
handed+0.033
 unatt+0.033
taker+0.032
 rubble+0.029
Neg logit contributions
Sal-0.021
oxy-0.018
 ho-0.017
ita-0.017
Le-0.017
Token!
Token ID0
Feature activation-1.70e-01

Pos logit contributions
keeping+0.029
handed+0.026
 unatt+0.026
taker+0.025
 rubble+0.024
Neg logit contributions
Sal-0.011
 ho-0.010
ita-0.010
 chicken-0.010
oxy-0.010
Token<|endoftext|>
Token ID50256
Feature activation-2.11e+00

Pos logit contributions
keeping+0.006
 unatt+0.006
handed+0.006
taker+0.005
 rubble+0.005
Neg logit contributions
Sal-0.002
oxy-0.002
ita-0.002
 ho-0.002
 chicken-0.002
TokenOnce
Token ID7454
Feature activation-1.78e+00

Pos logit contributions
keeping+0.075
handed+0.068
 unatt+0.067
taker+0.065
 rubble+0.060
Neg logit contributions
Sal-0.030
 ho-0.025
ita-0.024
oxy-0.024
 hus-0.024
Token upon
Token ID2402
Feature activation+3.43e-01

Pos logit contributions
keeping+0.069
handed+0.061
 unatt+0.060
taker+0.056
 rubble+0.054
Neg logit contributions
Sal-0.021
 short-0.019
ira-0.018
 ho-0.018
ita-0.017
Token a
Token ID257
Feature activation+2.63e+00

Pos logit contributions
Sal+0.005
 ho+0.005
ita+0.005
 hus+0.004
oxy+0.004
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.010
taker-0.010
 rubble-0.009
Token time
Token ID640
Feature activation-1.91e-01

Pos logit contributions
Sal+0.049
 hus+0.043
 chicken+0.040
+0.040
 pap+0.039
Neg logit contributions
handed-0.069
 unatt-0.068
keeping-0.067
hell-0.062
 anymore-0.061
Token there
Token ID612
Feature activation-1.32e+00

Pos logit contributions
keeping+0.008
handed+0.007
 unatt+0.006
taker+0.006
 rubble+0.006
Neg logit contributions
Sal-0.002
 ho-0.002
 short-0.002
-0.002
 stretch-0.002
Token was
Token ID373
Feature activation+5.90e-01

Pos logit contributions
keeping+0.050
handed+0.045
 unatt+0.045
taker+0.044
 rubble+0.041
Neg logit contributions
Sal-0.016
 stretch-0.012
 ho-0.012
 chicken-0.012
 hus-0.012
Token a
Token ID257
Feature activation+1.16e+00

Pos logit contributions
Sal+0.011
 hus+0.009
 ho+0.009
+0.009
ita+0.009
Neg logit contributions
keeping-0.018
handed-0.016
 unatt-0.016
taker-0.015
 rubble-0.014
Token curious
Token ID11040
Feature activation+1.07e+00

Pos logit contributions
Sal+0.025
ita+0.022
oxy+0.022
+0.021
 stretches+0.021
Neg logit contributions
 unatt-0.025
taker-0.024
handed-0.024
keeping-0.023
 wreck-0.022
Token groom
Token ID32443
Feature activation+9.62e-01

Pos logit contributions
keeping+0.005
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.004
Neg logit contributions
Sal-0.003
oxy-0.002
ita-0.002
 ho-0.002
Le-0.002
Token them
Token ID606
Feature activation+6.92e-01

Pos logit contributions
Sal+0.018
Le+0.017
 stretch+0.017
oxy+0.016
 hus+0.016
Neg logit contributions
keeping-0.028
handed-0.024
 unatt-0.023
taker-0.022
stick-0.020
Token.
Token ID13
Feature activation-7.12e-02

Pos logit contributions
Sal+0.013
oxy+0.011
ita+0.011
 ho+0.011
Le+0.011
Neg logit contributions
keeping-0.022
handed-0.019
 unatt-0.019
taker-0.019
ocular-0.017
Token One
Token ID1881
Feature activation+1.63e+00

Pos logit contributions
keeping+0.002
handed+0.002
taker+0.001
 unatt+0.001
 anymore+0.001
Neg logit contributions
Sal-0.002
 ho-0.002
 hus-0.002
oxy-0.001
avi-0.001
Token day
Token ID1110
Feature activation+7.01e-01

Pos logit contributions
Sal+0.047
oxy+0.043
Le+0.042
Scar+0.041
avi+0.041
Neg logit contributions
keeping-0.027
handed-0.025
 unatt-0.025
taker-0.025
lessness-0.019
Token while
Token ID981
Feature activation+2.24e+00

Pos logit contributions
Sal+0.018
oxy+0.016
avi+0.016
ita+0.016
 hus+0.016
Neg logit contributions
keeping-0.014
taker-0.013
 unatt-0.013
handed-0.012
 anymore-0.012
Token she
Token ID673
Feature activation+5.37e-01

Pos logit contributions
Sal+0.054
ita+0.047
avi+0.047
 hus+0.046
oxy+0.046
Neg logit contributions
keeping-0.056
 unatt-0.049
handed-0.046
taker-0.044
 anywhere-0.040
Token was
Token ID373
Feature activation+2.44e-01

Pos logit contributions
Sal+0.011
Le+0.010
oxy+0.010
ita+0.010
 ho+0.010
Neg logit contributions
keeping-0.015
handed-0.013
 unatt-0.012
taker-0.011
 rubble-0.010
Token at
Token ID379
Feature activation+2.83e-01

Pos logit contributions
Sal+0.006
oxy+0.005
Le+0.005
ita+0.005
+0.005
Neg logit contributions
keeping-0.006
 unatt-0.006
handed-0.006
taker-0.005
 rubble-0.005
Token the
Token ID262
Feature activation+6.29e-02

Pos logit contributions
Sal+0.006
oxy+0.005
ita+0.005
 ho+0.005
Le+0.005
Neg logit contributions
keeping-0.009
handed-0.007
 unatt-0.007
taker-0.007
 rubble-0.006
Token stable
Token ID8245
Feature activation-2.38e-01

Pos logit contributions
Sal+0.001
oxy+0.001
Le+0.001
aped+0.001
ira+0.001
Neg logit contributions
handed-0.001
 unatt-0.001
 anywhere-0.001
keeping-0.001
reating-0.001
Tokenowed
Token ID6972
Feature activation+6.59e-02

Pos logit contributions
Sal+0.005
 ho+0.005
 stretch+0.005
 short+0.005
ita+0.004
Neg logit contributions
keeping-0.016
 unatt-0.015
handed-0.015
taker-0.014
 rubble-0.013
Token in
Token ID287
Feature activation+1.51e-01

Pos logit contributions
Sal+0.001
 ho+0.001
 hus+0.001
 short+0.001
 chicken+0.001
Neg logit contributions
keeping-0.002
handed-0.002
 unatt-0.002
taker-0.002
 rubble-0.002
Token surprise
Token ID5975
Feature activation-4.76e-01

Pos logit contributions
 ho+0.002
 stretch+0.002
 short+0.002
Sal+0.002
 hus+0.002
Neg logit contributions
keeping-0.006
handed-0.005
taker-0.005
 unatt-0.005
ount-0.005
Token.
Token ID13
Feature activation-5.90e-01

Pos logit contributions
keeping+0.015
handed+0.013
 unatt+0.013
taker+0.013
 rubble+0.012
Neg logit contributions
Sal-0.009
ita-0.007
oxy-0.007
 ho-0.007
 hus-0.007
Token "
Token ID366
Feature activation+1.36e-01

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.018
taker+0.017
 rubble+0.016
Neg logit contributions
Sal-0.009
 ho-0.008
 chicken-0.008
 hus-0.008
 stretch-0.008
TokenHow
Token ID2437
Feature activation+2.33e+00

Pos logit contributions
Sal+0.003
Le+0.002
oxy+0.002
 ho+0.002
ita+0.002
Neg logit contributions
keeping-0.004
 unatt-0.004
handed-0.004
taker-0.004
 rubble-0.003
Token can
Token ID460
Feature activation+5.48e-01

Pos logit contributions
Sal+0.040
 ho+0.035
ita+0.034
 hus+0.034
 chicken+0.034
Neg logit contributions
keeping-0.077
handed-0.069
 unatt-0.068
taker-0.065
 rubble-0.060
Token I
Token ID314
Feature activation+6.55e-01

Pos logit contributions
Sal+0.011
ita+0.010
oxy+0.010
 hus+0.010
 ho+0.009
Neg logit contributions
keeping-0.016
handed-0.014
 unatt-0.014
taker-0.014
ocular-0.012
Token eat
Token ID4483
Feature activation+1.23e+00

Pos logit contributions
Sal+0.011
 ho+0.010
 stretch+0.009
 chicken+0.009
ita+0.009
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.019
taker-0.018
 rubble-0.017
Token a
Token ID257
Feature activation-1.04e+00

Pos logit contributions
Sal+0.026
 ho+0.023
ita+0.023
oxy+0.023
 hus+0.023
Neg logit contributions
keeping-0.035
handed-0.031
 unatt-0.030
taker-0.029
 rubble-0.026
Token zero
Token ID6632
Feature activation+6.87e-01

Pos logit contributions
keeping+0.033
handed+0.029
 unatt+0.028
taker+0.028
 rubble+0.025
Neg logit contributions
Sal-0.019
 ho-0.017
 chicken-0.017
 short-0.017
 stretch-0.017
Token so
Token ID523
Feature activation+9.56e-01

Pos logit contributions
keeping+0.029
handed+0.025
taker+0.024
 unatt+0.023
eway+0.023
Neg logit contributions
 short-0.011
Sal-0.010
 stretch-0.009
 each-0.009
 be-0.009
Token happy
Token ID3772
Feature activation+5.11e-01

Pos logit contributions
Sal+0.016
 short+0.015
 ho+0.014
 stretch+0.014
 chicken+0.013
Neg logit contributions
keeping-0.033
handed-0.028
 unatt-0.028
taker-0.027
 rubble-0.026
Token and
Token ID290
Feature activation+2.50e-01

Pos logit contributions
Sal+0.006
 each+0.006
 chicken+0.006
 ho+0.006
 stretch+0.005
Neg logit contributions
keeping-0.018
handed-0.017
 unatt-0.017
taker-0.017
 rubble-0.015
Token he
Token ID339
Feature activation+4.09e-01

Pos logit contributions
Sal+0.005
 ho+0.004
 chicken+0.004
 stretch+0.004
 short+0.004
Neg logit contributions
keeping-0.008
handed-0.007
 unatt-0.007
taker-0.007
 rubble-0.006
Token played
Token ID2826
Feature activation-6.30e-01

Pos logit contributions
Sal+0.008
 ho+0.007
 chicken+0.006
ita+0.006
 hus+0.006
Neg logit contributions
keeping-0.013
handed-0.011
 unatt-0.011
taker-0.011
 rubble-0.010
Token with
Token ID351
Feature activation-1.30e-01

Pos logit contributions
keeping+0.022
handed+0.019
 unatt+0.019
taker+0.019
ocular+0.018
Neg logit contributions
Sal-0.009
 stretch-0.008
 each-0.008
 short-0.008
 chicken-0.008
Token his
Token ID465
Feature activation-9.88e-01

Pos logit contributions
keeping+0.004
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.003
Neg logit contributions
Sal-0.002
 ho-0.002
 chicken-0.002
 hus-0.002
oxy-0.002
Token new
Token ID649
Feature activation-3.47e-01

Pos logit contributions
keeping+0.033
handed+0.029
 unatt+0.029
taker+0.028
 rubble+0.025
Neg logit contributions
Sal-0.017
 ho-0.015
ita-0.014
oxy-0.014
 chicken-0.014
Token toy
Token ID13373
Feature activation-5.66e-01

Pos logit contributions
keeping+0.011
handed+0.009
 unatt+0.009
taker+0.009
 rubble+0.008
Neg logit contributions
Sal-0.007
 ho-0.006
 chicken-0.006
ita-0.006
 hus-0.006
Token truck
Token ID7779
Feature activation+8.60e-01

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.015
 rubble+0.013
Neg logit contributions
Sal-0.011
 ho-0.010
ita-0.009
 chicken-0.009
oxy-0.009
Token and
Token ID290
Feature activation-1.11e+00

Pos logit contributions
 ho+0.011
 short+0.011
 each+0.011
 stretch+0.010
Sal+0.010
Neg logit contributions
keeping-0.030
handed-0.027
taker-0.026
 disl-0.026
 unatt-0.026
Token was
Token ID373
Feature activation-2.64e-01

Pos logit contributions
Sal+0.008
oxy+0.008
Le+0.008
avi+0.008
ita+0.008
Neg logit contributions
keeping-0.008
 unatt-0.007
taker-0.007
handed-0.007
ocular-0.006
Token so
Token ID523
Feature activation+8.41e-01

Pos logit contributions
keeping+0.009
handed+0.008
 unatt+0.007
taker+0.007
 rubble+0.007
Neg logit contributions
Sal-0.004
 short-0.004
 ho-0.004
 stretch-0.004
 chicken-0.004
Token excited
Token ID6568
Feature activation+4.31e-01

Pos logit contributions
Sal+0.016
ita+0.014
 ho+0.014
oxy+0.014
 chicken+0.013
Neg logit contributions
keeping-0.026
handed-0.023
 unatt-0.023
taker-0.023
 rubble-0.020
Token to
Token ID284
Feature activation+2.84e-01

Pos logit contributions
Sal+0.006
ita+0.006
 ho+0.006
 chicken+0.006
 short+0.006
Neg logit contributions
keeping-0.015
handed-0.013
 unatt-0.013
taker-0.013
 rubble-0.012
Token see
Token ID766
Feature activation-4.13e-01

Pos logit contributions
Sal+0.006
oxy+0.005
ita+0.005
 hus+0.005
Le+0.005
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.007
taker-0.007
 rubble-0.007
Token what
Token ID644
Feature activation+4.17e-01

Pos logit contributions
keeping+0.012
 unatt+0.010
handed+0.010
taker+0.010
 disl+0.009
Neg logit contributions
 each-0.009
Sal-0.009
 corn-0.008
 short-0.008
 ho-0.008
Token was
Token ID373
Feature activation-8.06e-01

Pos logit contributions
Sal+0.008
 ho+0.008
oxy+0.008
ita+0.008
avi+0.007
Neg logit contributions
keeping-0.012
handed-0.010
 unatt-0.010
taker-0.010
 rubble-0.009
Token in
Token ID287
Feature activation-5.03e-01

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.017
taker+0.016
 disl+0.015
Neg logit contributions
 short-0.018
Sal-0.018
 long-0.018
 chicken-0.018
 corn-0.018
Token it
Token ID340
Feature activation-7.62e-02

Pos logit contributions
keeping+0.017
handed+0.016
 unatt+0.016
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.007
 ho-0.006
 hus-0.006
 short-0.006
 chicken-0.006
Token!
Token ID0
Feature activation-3.93e-01

Pos logit contributions
keeping+0.003
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.001
 ho-0.001
ita-0.001
oxy-0.001
 hus-0.001
Token She
Token ID1375
Feature activation+5.27e-01

Pos logit contributions
keeping+0.014
 unatt+0.012
handed+0.012
taker+0.012
 rubble+0.011
Neg logit contributions
Sal-0.006
 ho-0.005
ita-0.005
 hus-0.005
oxy-0.005
Token the
Token ID262
Feature activation+1.37e+00

Pos logit contributions
keeping+0.017
handed+0.015
 unatt+0.015
taker+0.014
ocular+0.013
Neg logit contributions
Sal-0.006
 short-0.006
 stretch-0.006
 each-0.006
 ho-0.005
Token mask
Token ID9335
Feature activation-1.24e+00

Pos logit contributions
Sal+0.029
 ho+0.027
 short+0.026
 chicken+0.026
 stretch+0.026
Neg logit contributions
keeping-0.040
handed-0.034
 unatt-0.034
taker-0.032
 rubble-0.029
Token stopped
Token ID5025
Feature activation+1.68e+00

Pos logit contributions
keeping+0.043
handed+0.038
 unatt+0.038
taker+0.037
 rubble+0.033
Neg logit contributions
Sal-0.019
 ho-0.017
 stretch-0.016
 hus-0.016
 short-0.016
Token running
Token ID2491
Feature activation-6.10e-01

Pos logit contributions
Sal+0.043
oxy+0.036
ira+0.035
Le+0.035
Isa+0.034
Neg logit contributions
keeping-0.039
 unatt-0.037
 apart-0.032
 anymore-0.031
handed-0.030
Token and
Token ID290
Feature activation-1.08e+00

Pos logit contributions
keeping+0.020
handed+0.019
 unatt+0.018
taker+0.017
 disl+0.016
Neg logit contributions
 ho-0.009
 hus-0.008
 short-0.008
Sal-0.008
 each-0.008
Token became
Token ID2627
Feature activation+5.03e-01

Pos logit contributions
keeping+0.038
handed+0.035
 unatt+0.033
taker+0.032
 rubble+0.029
Neg logit contributions
Sal-0.016
 ho-0.015
 stretch-0.015
 hus-0.014
 chicken-0.014
Token quiet
Token ID5897
Feature activation-1.13e-01

Pos logit contributions
Sal+0.011
oxy+0.010
ita+0.010
ira+0.009
avi+0.009
Neg logit contributions
keeping-0.014
 unatt-0.013
handed-0.012
taker-0.012
 rubble-0.011
Token.
Token ID13
Feature activation-1.56e-01

Pos logit contributions
keeping+0.004
handed+0.004
taker+0.004
 unatt+0.004
 rubble+0.004
Neg logit contributions
 ho-0.001
 short-0.001
 stretch-0.001
 hus-0.001
Sal-0.001
Token It
Token ID632
Feature activation+2.95e-01

Pos logit contributions
keeping+0.004
 unatt+0.004
handed+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.004
 short-0.003
ita-0.003
-0.003
ira-0.003
Token turned
Token ID2900
Feature activation-1.76e-01

Pos logit contributions
Sal+0.005
 ho+0.004
oxy+0.004
ita+0.004
 chicken+0.004
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.008
 rubble-0.008
Token around
Token ID1088
Feature activation-4.44e-01

Pos logit contributions
keeping+0.005
handed+0.004
 unatt+0.004
taker+0.004
 rubble+0.003
Neg logit contributions
Sal-0.004
 ho-0.004
 hus-0.004
 short-0.004
 chicken-0.004
Token toys
Token ID14958
Feature activation-4.83e-01

Pos logit contributions
keeping+0.010
handed+0.008
 unatt+0.008
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.006
 ho-0.005
 hus-0.005
 chicken-0.005
ita-0.005
Token.
Token ID13
Feature activation-4.06e-01

Pos logit contributions
keeping+0.017
handed+0.015
taker+0.015
 unatt+0.015
 rubble+0.014
Neg logit contributions
Sal-0.006
 ho-0.006
 short-0.006
 chicken-0.006
 stretch-0.005
Token You
Token ID921
Feature activation-1.60e+00

Pos logit contributions
keeping+0.012
handed+0.011
 unatt+0.010
taker+0.010
 rubble+0.009
Neg logit contributions
Sal-0.008
 ho-0.007
oxy-0.007
 chicken-0.007
ita-0.007
Token can
Token ID460
Feature activation+2.50e-01

Pos logit contributions
keeping+0.057
handed+0.051
 unatt+0.050
taker+0.047
eway+0.045
Neg logit contributions
Sal-0.023
 ho-0.020
 stretch-0.020
 chicken-0.019
 short-0.019
Token share
Token ID2648
Feature activation+4.94e-01

Pos logit contributions
Sal+0.004
ita+0.004
 ho+0.004
oxy+0.004
 hus+0.004
Neg logit contributions
keeping-0.008
handed-0.007
 unatt-0.007
taker-0.007
 rubble-0.006
Token and
Token ID290
Feature activation-1.02e+00

Pos logit contributions
Sal+0.006
 each+0.006
 chicken+0.006
 short+0.006
 ho+0.006
Neg logit contributions
keeping-0.018
handed-0.016
taker-0.016
 unatt-0.016
 rubble-0.015
Token take
Token ID1011
Feature activation+2.62e-01

Pos logit contributions
keeping+0.039
handed+0.037
taker+0.036
 unatt+0.036
 rubble+0.032
Neg logit contributions
Sal-0.010
 stretch-0.010
 ho-0.009
 chicken-0.009
 short-0.009
Token turns
Token ID4962
Feature activation+1.17e+00

Pos logit contributions
Sal+0.003
 short+0.003
 ho+0.003
 each+0.003
 chicken+0.002
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.009
eway-0.008
Token.
Token ID13
Feature activation-8.76e-02

Pos logit contributions
Sal+0.017
 ho+0.015
 short+0.015
 chicken+0.014
 stretch+0.014
Neg logit contributions
keeping-0.040
handed-0.037
 unatt-0.036
taker-0.035
 rubble-0.033
Token That
Token ID1320
Feature activation-9.49e-02

Pos logit contributions
keeping+0.002
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.002
 ho-0.002
ita-0.002
 hus-0.002
oxy-0.002
Token is
Token ID318
Feature activation-3.96e-02

Pos logit contributions
keeping+0.003
handed+0.003
 unatt+0.003
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.001
oxy-0.001
 ho-0.001
ita-0.001
Le-0.001
Token always
Token ID1464
Feature activation-1.09e+00

Pos logit contributions
keeping+0.043
handed+0.038
 unatt+0.038
taker+0.037
 rubble+0.033
Neg logit contributions
Sal-0.022
 ho-0.019
ita-0.019
 hus-0.019
 chicken-0.019
Token shared
Token ID4888
Feature activation+8.54e-01

Pos logit contributions
keeping+0.035
handed+0.031
 unatt+0.031
taker+0.029
 rubble+0.027
Neg logit contributions
Sal-0.020
oxy-0.017
ita-0.017
 ho-0.017
 hus-0.017
Token their
Token ID511
Feature activation-4.86e-01

Pos logit contributions
Sal+0.007
 each+0.007
 chicken+0.007
 ho+0.007
 short+0.006
Neg logit contributions
keeping-0.035
handed-0.032
taker-0.032
 unatt-0.031
 rubble-0.028
Token toys
Token ID14958
Feature activation+6.38e-01

Pos logit contributions
keeping+0.015
handed+0.014
 unatt+0.013
taker+0.013
 rubble+0.012
Neg logit contributions
Sal-0.009
 ho-0.008
ita-0.008
oxy-0.008
 hus-0.008
Token and
Token ID290
Feature activation-1.08e+00

Pos logit contributions
Sal+0.007
 each+0.007
 ho+0.007
 short+0.006
 stretch+0.006
Neg logit contributions
keeping-0.024
handed-0.021
 unatt-0.021
taker-0.021
 rubble-0.019
Token had
Token ID550
Feature activation-1.17e+00

Pos logit contributions
keeping+0.037
handed+0.035
 unatt+0.034
taker+0.033
 rubble+0.029
Neg logit contributions
Sal-0.015
 ho-0.014
 stretch-0.014
 chicken-0.013
 each-0.013
Token lots
Token ID6041
Feature activation-6.19e-01

Pos logit contributions
keeping+0.028
handed+0.024
 unatt+0.023
taker+0.021
 rubble+0.020
Neg logit contributions
Sal-0.030
ira-0.027
ita-0.027
 ho-0.026
oxy-0.026
Token of
Token ID286
Feature activation-5.63e-01

Pos logit contributions
keeping+0.020
handed+0.017
 unatt+0.016
taker+0.016
ocular+0.015
Neg logit contributions
Sal-0.011
 short-0.010
 chicken-0.010
 stretch-0.010
 ho-0.010
Token fun
Token ID1257
Feature activation+8.03e-02

Pos logit contributions
keeping+0.017
handed+0.016
 unatt+0.015
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.011
ita-0.009
oxy-0.009
 ho-0.009
 hus-0.009
Token playing
Token ID2712
Feature activation-6.41e-02

Pos logit contributions
Sal+0.001
 ho+0.001
ita+0.001
 hus+0.001
oxy+0.001
Neg logit contributions
keeping-0.003
handed-0.002
 unatt-0.002
taker-0.002
 rubble-0.002
Token together
Token ID1978
Feature activation-6.28e-01

Pos logit contributions
keeping+0.002
handed+0.002
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.001
 ho-0.001
ita-0.001
oxy-0.001
 hus-0.001
Token,
Token ID11
Feature activation-9.71e-01

Pos logit contributions
Sal+0.015
oxy+0.013
 hus+0.012
 chicken+0.012
ita+0.012
Neg logit contributions
keeping-0.020
 unatt-0.018
handed-0.018
taker-0.017
 rubble-0.015
Token then
Token ID788
Feature activation-2.69e-01

Pos logit contributions
keeping+0.029
handed+0.026
 unatt+0.025
taker+0.025
 rubble+0.022
Neg logit contributions
Sal-0.019
 ho-0.017
 hus-0.017
 chicken-0.017
ita-0.017
Token disappeared
Token ID12120
Feature activation+2.88e-01

Pos logit contributions
keeping+0.008
handed+0.008
 unatt+0.008
taker+0.007
 rubble+0.007
Neg logit contributions
Sal-0.005
oxy-0.004
ita-0.004
 ho-0.004
 hus-0.004
Token just
Token ID655
Feature activation-2.26e-01

Pos logit contributions
Sal+0.005
 ho+0.005
 hus+0.005
ita+0.005
 short+0.005
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.008
 rubble-0.007
Token as
Token ID355
Feature activation+1.07e+00

Pos logit contributions
keeping+0.007
handed+0.007
taker+0.006
 unatt+0.006
 rubble+0.006
Neg logit contributions
 ho-0.004
 stretch-0.004
 short-0.003
 each-0.003
Sal-0.003
Token quickly
Token ID2952
Feature activation-3.85e-01

Pos logit contributions
Sal+0.017
 ho+0.015
 hus+0.015
 short+0.015
ita+0.014
Neg logit contributions
keeping-0.036
handed-0.032
 unatt-0.032
taker-0.030
 rubble-0.028
Token as
Token ID355
Feature activation+8.91e-01

Pos logit contributions
keeping+0.013
handed+0.012
 unatt+0.011
taker+0.011
 rubble+0.010
Neg logit contributions
Sal-0.006
 ho-0.006
 stretch-0.006
 hus-0.006
 chicken-0.006
Token it
Token ID340
Feature activation+4.90e-01

Pos logit contributions
Sal+0.013
 ho+0.012
 short+0.012
 hus+0.012
 chicken+0.012
Neg logit contributions
keeping-0.030
handed-0.028
 unatt-0.027
taker-0.026
 rubble-0.024
Token had
Token ID550
Feature activation+1.70e-01

Pos logit contributions
Sal+0.008
oxy+0.007
 ho+0.007
ita+0.007
 hus+0.007
Neg logit contributions
keeping-0.016
handed-0.015
 unatt-0.014
taker-0.014
 rubble-0.013
Token appeared
Token ID4120
Feature activation+3.48e-01

Pos logit contributions
Sal+0.003
ita+0.003
oxy+0.003
 ho+0.003
 chicken+0.003
Neg logit contributions
keeping-0.005
handed-0.005
 unatt-0.004
taker-0.004
 rubble-0.004
Token.
Token ID13
Feature activation+2.50e-01

Pos logit contributions
 ho+0.005
 short+0.004
Sal+0.004
 hus+0.004
 stretch+0.004
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.011
taker-0.011
 rubble-0.010
Token he
Token ID339
Feature activation+3.12e-01

Pos logit contributions
keeping+0.073
handed+0.066
 unatt+0.061
taker+0.058
 rubble+0.055
Neg logit contributions
Sal-0.032
 short-0.030
 ho-0.030
 stretch-0.029
Le-0.029
Token remembered
Token ID12086
Feature activation-4.92e-01

Pos logit contributions
Sal+0.006
oxy+0.005
 ho+0.005
ita+0.005
 hus+0.005
Neg logit contributions
keeping-0.009
 unatt-0.008
handed-0.008
taker-0.008
ocular-0.007
Token that
Token ID326
Feature activation-1.41e+00

Pos logit contributions
keeping+0.015
handed+0.013
taker+0.013
 unatt+0.012
makers+0.011
Neg logit contributions
 each-0.010
 chicken-0.009
 corn-0.009
 ho-0.009
 hus-0.009
Token Sam
Token ID3409
Feature activation-1.25e-01

Pos logit contributions
keeping+0.052
handed+0.048
 unatt+0.045
taker+0.043
ocular+0.040
Neg logit contributions
Sal-0.019
 short-0.018
 stretch-0.018
 chicken-0.018
 each-0.018
Token was
Token ID373
Feature activation-8.10e-01

Pos logit contributions
keeping+0.004
handed+0.004
 unatt+0.004
taker+0.003
 rubble+0.003
Neg logit contributions
Sal-0.002
 ho-0.002
oxy-0.002
ita-0.002
 hus-0.002
Token his
Token ID465
Feature activation-8.14e-01

Pos logit contributions
keeping+0.025
handed+0.022
 unatt+0.021
taker+0.021
 rubble+0.019
Neg logit contributions
Sal-0.016
 ho-0.014
 chicken-0.014
 hus-0.014
 short-0.014
Token friend
Token ID1545
Feature activation-6.53e-01

Pos logit contributions
keeping+0.029
handed+0.026
 unatt+0.025
taker+0.025
 rubble+0.023
Neg logit contributions
Sal-0.012
 ho-0.011
 chicken-0.010
ita-0.010
 hus-0.010
Token.
Token ID13
Feature activation+5.67e-01

Pos logit contributions
keeping+0.024
handed+0.022
taker+0.021
 unatt+0.021
 rubble+0.020
Neg logit contributions
Sal-0.008
 chicken-0.007
 ho-0.007
ita-0.006
 stretch-0.006
Token He
Token ID679
Feature activation+7.49e-01

Pos logit contributions
Sal+0.008
 ho+0.006
+0.006
 short+0.006
Le+0.006
Neg logit contributions
keeping-0.021
 unatt-0.019
handed-0.019
taker-0.018
 rubble-0.017
Token knew
Token ID2993
Feature activation-1.16e+00

Pos logit contributions
Sal+0.014
oxy+0.012
 ho+0.012
ita+0.012
 short+0.012
Neg logit contributions
keeping-0.022
handed-0.020
 unatt-0.020
taker-0.020
 rubble-0.018
Token that
Token ID326
Feature activation-7.47e-01

Pos logit contributions
keeping+0.038
handed+0.034
 unatt+0.033
taker+0.031
 rubble+0.028
Neg logit contributions
 each-0.020
Sal-0.019
 chicken-0.019
 ho-0.018
 corn-0.018
Tokenia
Token ID544
Feature activation-4.81e-01

Pos logit contributions
Sal+0.042
 ho+0.036
 chicken+0.036
 hus+0.036
ita+0.036
Neg logit contributions
keeping-0.082
handed-0.075
taker-0.075
 unatt-0.074
ocular-0.067
Token nods
Token ID34833
Feature activation+6.98e-01

Pos logit contributions
keeping+0.012
handed+0.012
 unatt+0.011
taker+0.011
ocular+0.011
Neg logit contributions
Sal-0.009
oxy-0.009
ita-0.008
Le-0.008
 short-0.008
Token.
Token ID13
Feature activation+3.62e-01

Pos logit contributions
Sal+0.014
oxy+0.012
ita+0.012
 ho+0.012
 chicken+0.012
Neg logit contributions
keeping-0.021
 unatt-0.018
handed-0.018
taker-0.017
ocular-0.015
Token She
Token ID1375
Feature activation-1.58e-01

Pos logit contributions
Sal+0.007
ita+0.006
oxy+0.006
 ho+0.006
 chicken+0.006
Neg logit contributions
keeping-0.011
handed-0.009
taker-0.009
 unatt-0.009
ocular-0.008
Token takes
Token ID2753
Feature activation+1.19e+00

Pos logit contributions
keeping+0.004
handed+0.004
taker+0.003
 unatt+0.003
ocular+0.003
Neg logit contributions
Sal-0.003
oxy-0.003
Le-0.003
ita-0.003
 short-0.003
Token the
Token ID262
Feature activation-6.96e-01

Pos logit contributions
Sal+0.024
ita+0.022
oxy+0.021
+0.021
 ho+0.021
Neg logit contributions
keeping-0.035
handed-0.031
 unatt-0.030
taker-0.030
 rubble-0.026
Token ball
Token ID2613
Feature activation-4.40e-01

Pos logit contributions
keeping+0.020
handed+0.018
 unatt+0.018
taker+0.018
 rubble+0.016
Neg logit contributions
Sal-0.014
ita-0.013
oxy-0.013
avi-0.012
 hus-0.012
Token.
Token ID13
Feature activation+3.24e-01

Pos logit contributions
keeping+0.014
handed+0.013
 unatt+0.013
taker+0.012
 rubble+0.011
Neg logit contributions
Sal-0.007
 ho-0.007
 chicken-0.006
 stretch-0.006
 hus-0.006
Token She
Token ID1375
Feature activation-1.48e-01

Pos logit contributions
Sal+0.006
ita+0.005
 ho+0.005
oxy+0.005
 hus+0.005
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.009
 rubble-0.008
Token throws
Token ID12542
Feature activation+1.21e-01

Pos logit contributions
keeping+0.003
handed+0.003
taker+0.003
 unatt+0.003
ocular+0.003
Neg logit contributions
oxy-0.003
Sal-0.003
ita-0.003
Le-0.003
 short-0.003
Token it
Token ID340
Feature activation-9.23e-01

Pos logit contributions
Sal+0.002
 ho+0.002
 chicken+0.002
 hus+0.002
 stretch+0.002
Neg logit contributions
keeping-0.004
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.003
Token.
Token ID13
Feature activation+3.03e-01

Pos logit contributions
keeping+0.015
handed+0.014
 unatt+0.014
taker+0.014
 rubble+0.013
Neg logit contributions
Sal-0.005
 chicken-0.005
 hus-0.005
 ho-0.005
 stretch-0.005
Token The
Token ID383
Feature activation+1.21e+00

Pos logit contributions
Sal+0.005
 ho+0.005
ita+0.004
oxy+0.004
 hus+0.004
Neg logit contributions
keeping-0.010
handed-0.009
 unatt-0.009
taker-0.008
 rubble-0.008
Token vet
Token ID20202
Feature activation-9.68e-01

Pos logit contributions
Sal+0.024
ita+0.021
oxy+0.021
 ho+0.020
Le+0.020
Neg logit contributions
keeping-0.036
handed-0.032
 unatt-0.032
taker-0.031
 rubble-0.028
Token showed
Token ID3751
Feature activation+1.55e-01

Pos logit contributions
keeping+0.032
handed+0.028
 unatt+0.028
taker+0.027
 rubble+0.025
Neg logit contributions
Sal-0.016
 ho-0.014
ita-0.014
 hus-0.014
 chicken-0.014
Token Wh
Token ID854
Feature activation-1.62e+00

Pos logit contributions
Sal+0.002
 chicken+0.002
 ho+0.002
 each+0.002
 hus+0.002
Neg logit contributions
keeping-0.006
 unatt-0.005
handed-0.005
taker-0.005
 rubble-0.004
Tokenisk
Token ID1984
Feature activation-3.77e+00

Pos logit contributions
keeping+0.031
handed+0.025
 unatt+0.025
taker+0.023
 rubble+0.019
Neg logit contributions
Sal-0.050
 ho-0.046
ita-0.045
oxy-0.045
 hus-0.045
Tokeners
Token ID364
Feature activation-1.49e-01

Pos logit contributions
keeping+0.098
 unatt+0.088
handed+0.084
taker+0.081
 rubble+0.074
Neg logit contributions
Sal-0.087
 hus-0.077
oxy-0.077
Le-0.077
ita-0.076
Token the
Token ID262
Feature activation-7.83e-02

Pos logit contributions
keeping+0.004
 unatt+0.004
handed+0.004
taker+0.004
 rubble+0.003
Neg logit contributions
Sal-0.003
 chicken-0.003
 ho-0.003
 hus-0.003
 corn-0.003
Token x
Token ID2124
Feature activation-4.86e-01

Pos logit contributions
keeping+0.002
handed+0.002
 unatt+0.002
taker+0.002
 rubble+0.002
Neg logit contributions
Sal-0.001
 ho-0.001
 chicken-0.001
 hus-0.001
 short-0.001
Token-
Token ID12
Feature activation-8.54e-01

Pos logit contributions
keeping+0.020
 unatt+0.019
handed+0.018
taker+0.018
 rubble+0.017
Neg logit contributions
Sal-0.004
 ho-0.003
 hus-0.003
 chicken-0.003
ira-0.003
Tokenray
Token ID2433
Feature activation-1.14e+00

Pos logit contributions
keeping+0.016
 unatt+0.013
handed+0.013
taker+0.012
 rubble+0.010
Neg logit contributions
Sal-0.027
oxy-0.025
 hus-0.025
 ho-0.025
ita-0.024
TokenAs
Token ID1722
Feature activation+1.58e+00

Pos logit contributions
keeping+0.000
handed+0.000
 unatt+0.000
taker+0.000
 rubble+0.000
Neg logit contributions
Sal-0.000
 ho-0.000
ira-0.000
 hus-0.000
 chicken-0.000
Token Daisy
Token ID40355
Feature activation+1.11e-01

Pos logit contributions
Sal+0.038
oxy+0.033
 hus+0.032
 ho+0.032
avi+0.032
Neg logit contributions
keeping-0.041
 unatt-0.035
handed-0.035
taker-0.034
 rubble-0.031
Token walked
Token ID6807
Feature activation-9.99e-01

Pos logit contributions
Sal+0.002
oxy+0.002
ita+0.002
 hus+0.002
 chicken+0.002
Neg logit contributions
keeping-0.003
handed-0.003
 unatt-0.003
taker-0.003
 rubble-0.003
Token through
Token ID832
Feature activation+2.01e-01

Pos logit contributions
keeping+0.032
handed+0.029
 unatt+0.029
taker+0.026
 disl+0.025
Neg logit contributions
Sal-0.017
 short-0.017
 ho-0.017
 stretch-0.016
 hus-0.016
Token the
Token ID262
Feature activation-4.78e-01

Pos logit contributions
Sal+0.002
 short+0.002
 ho+0.002
 hus+0.002
 stretch+0.002
Neg logit contributions
keeping-0.008
handed-0.007
 unatt-0.007
taker-0.007
ocular-0.006
Token tunnel
Token ID13275
Feature activation-3.19e+00

Pos logit contributions
keeping+0.013
 unatt+0.011
handed+0.011
taker+0.010
reating+0.009
Neg logit contributions
 long-0.011
 short-0.011
 line-0.011
 be-0.011
 human-0.011
Token,
Token ID11
Feature activation-1.76e+00

Pos logit contributions
keeping+0.121
handed+0.107
 unatt+0.105
taker+0.101
 disl+0.093
Neg logit contributions
Sal-0.036
 short-0.035
 stretch-0.033
 ho-0.033
ira-0.031
Token she
Token ID673
Feature activation-2.91e-02

Pos logit contributions
keeping+0.058
handed+0.052
 unatt+0.052
taker+0.050
 rubble+0.046
Neg logit contributions
Sal-0.029
 ho-0.025
ita-0.024
 chicken-0.024
 hus-0.024
Token suddenly
Token ID6451
Feature activation-3.11e-01

Pos logit contributions
keeping+0.001
handed+0.001
 unatt+0.001
taker+0.001
 rubble+0.001
Neg logit contributions
Sal-0.000
 ho-0.000
ita-0.000
oxy-0.000
 hus-0.000
Token heard
Token ID2982
Feature activation+2.44e-01

Pos logit contributions
keeping+0.009
handed+0.009
 unatt+0.008
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.006
oxy-0.005
 chicken-0.005
 ho-0.005
 hus-0.005
Token a
Token ID257
Feature activation+3.95e-01

Pos logit contributions
 short+0.002
Sal+0.001
 hus+0.001
 long+0.001
 chicken+0.001
Neg logit contributions
keeping-0.011
 unatt-0.010
handed-0.010
taker-0.009
makers-0.008
Token She
Token ID1375
Feature activation+6.26e-01

Pos logit contributions
Sal+0.005
 ho+0.005
ita+0.005
oxy+0.005
 hus+0.005
Neg logit contributions
keeping-0.008
handed-0.007
 unatt-0.007
taker-0.007
 rubble-0.006
Token had
Token ID550
Feature activation-1.07e+00

Pos logit contributions
Sal+0.012
 ho+0.010
oxy+0.010
Le+0.010
 hus+0.010
Neg logit contributions
keeping-0.019
handed-0.017
 unatt-0.017
taker-0.016
 rubble-0.015
Token copied
Token ID18984
Feature activation+9.54e-01

Pos logit contributions
keeping+0.033
handed+0.030
 unatt+0.029
taker+0.028
 rubble+0.026
Neg logit contributions
Sal-0.020
 ho-0.018
ita-0.017
oxy-0.017
 hus-0.017
Token the
Token ID262
Feature activation+3.22e-01

Pos logit contributions
 ho+0.008
Sal+0.008
 each+0.007
 stretch+0.007
 chicken+0.007
Neg logit contributions
keeping-0.039
handed-0.036
 unatt-0.035
taker-0.033
 rubble-0.032
Token wrong
Token ID2642
Feature activation+6.46e-01

Pos logit contributions
Sal+0.007
 ho+0.006
oxy+0.006
 hus+0.006
ita+0.006
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.008
 rubble-0.007
Token lecture
Token ID19143
Feature activation-3.29e+00

Pos logit contributions
Sal+0.012
 ho+0.011
oxy+0.011
ita+0.011
 hus+0.011
Neg logit contributions
keeping-0.020
handed-0.018
 unatt-0.017
taker-0.017
 rubble-0.015
Token!
Token ID0
Feature activation+2.37e-01

Pos logit contributions
keeping+0.114
handed+0.104
 unatt+0.102
taker+0.096
 rubble+0.094
Neg logit contributions
Sal-0.042
 short-0.041
 stretch-0.039
ita-0.039
 each-0.039
Token It
Token ID632
Feature activation+2.43e-01

Pos logit contributions
Sal+0.004
 ho+0.004
ita+0.004
 hus+0.004
oxy+0.004
Neg logit contributions
keeping-0.007
handed-0.007
 unatt-0.006
taker-0.006
 rubble-0.006
Token was
Token ID373
Feature activation+3.42e-01

Pos logit contributions
Sal+0.004
 ho+0.003
ita+0.003
 chicken+0.003
 hus+0.003
Neg logit contributions
keeping-0.009
handed-0.008
 unatt-0.008
taker-0.007
 rubble-0.007
Token a
Token ID257
Feature activation-5.88e-01

Pos logit contributions
Sal+0.005
 ho+0.005
 short+0.005
ita+0.005
 chicken+0.005
Neg logit contributions
keeping-0.012
handed-0.011
 unatt-0.010
taker-0.010
 rubble-0.009
Token bad
Token ID2089
Feature activation-1.21e+00

Pos logit contributions
keeping+0.020
handed+0.017
 unatt+0.017
taker+0.017
 rubble+0.015
Neg logit contributions
Sal-0.010
 ho-0.009
 chicken-0.008
 hus-0.008
 short-0.008
Token she
Token ID673
Feature activation+7.16e-01

Pos logit contributions
keeping+0.016
 unatt+0.014
handed+0.014
taker+0.014
ocular+0.013
Neg logit contributions
Sal-0.006
 hus-0.006
 chicken-0.006
 stretch-0.006
ita-0.006
Token was
Token ID373
Feature activation-7.00e-01

Pos logit contributions
Sal+0.014
 ho+0.012
oxy+0.012
Le+0.012
 chicken+0.012
Neg logit contributions
keeping-0.021
handed-0.019
 unatt-0.018
taker-0.018
 rubble-0.016
Token feeling
Token ID4203
Feature activation-1.93e+00

Pos logit contributions
keeping+0.021
handed+0.019
 unatt+0.018
taker+0.017
ocular+0.016
Neg logit contributions
Sal-0.013
 ho-0.012
 hus-0.012
 chicken-0.012
ita-0.012
Token very
Token ID845
Feature activation-2.63e-01

Pos logit contributions
keeping+0.059
handed+0.051
 unatt+0.050
taker+0.048
ocular+0.045
Neg logit contributions
Sal-0.037
 chicken-0.033
 hus-0.033
 short-0.033
 ho-0.033
Token gr
Token ID1036
Feature activation-9.69e-01

Pos logit contributions
keeping+0.007
handed+0.006
 unatt+0.006
taker+0.005
 rubble+0.005
Neg logit contributions
Sal-0.006
 ho-0.006
ita-0.006
 hus-0.006
oxy-0.006
Tokenumpy
Token ID32152
Feature activation-3.19e+00

Pos logit contributions
keeping+0.023
 unatt+0.019
handed+0.019
taker+0.017
 rubble+0.016
Neg logit contributions
Sal-0.025
ita-0.023
oxy-0.023
 ho-0.022
 hus-0.022
Token.
Token ID13
Feature activation+3.44e-01

Pos logit contributions
keeping+0.124
handed+0.116
 unatt+0.112
taker+0.107
 rubble+0.105
Neg logit contributions
 ho-0.025
 each-0.025
 short-0.024
 hus-0.023
ita-0.022
Token She
Token ID1375
Feature activation+7.86e-01

Pos logit contributions
Sal+0.005
ita+0.004
oxy+0.004
 ho+0.004
 hus+0.004
Neg logit contributions
keeping-0.012
 unatt-0.011
handed-0.011
taker-0.010
 rubble-0.010
Token put
Token ID1234
Feature activation-2.54e-01

Pos logit contributions
Sal+0.016
 ho+0.014
oxy+0.014
Le+0.014
 chicken+0.014
Neg logit contributions
keeping-0.022
handed-0.020
 unatt-0.019
taker-0.019
 rubble-0.017
Token the
Token ID262
Feature activation+4.94e-01

Pos logit contributions
keeping+0.009
handed+0.008
 unatt+0.008
taker+0.008
 rubble+0.007
Neg logit contributions
Sal-0.004
 ho-0.003
 chicken-0.003
 hus-0.003
 stretch-0.003
Token things
Token ID1243
Feature activation-8.00e-01

Pos logit contributions
Sal+0.008
ita+0.007
oxy+0.007
Le+0.007
 ho+0.007
Neg logit contributions
keeping-0.016
handed-0.015
 unatt-0.015
taker-0.014
 rubble-0.013
Token.
Token ID13
Feature activation-1.37e+00

Pos logit contributions
Sal+0.023
 ho+0.020
oxy+0.020
ita+0.020
 hus+0.019
Neg logit contributions
keeping-0.041
handed-0.037
 unatt-0.037
taker-0.035
 rubble-0.032
Token He
Token ID679
Feature activation+5.76e-01

Pos logit contributions
keeping+0.042
handed+0.038
 unatt+0.037
taker+0.036
 rubble+0.033
Neg logit contributions
Sal-0.026
 ho-0.023
ita-0.022
oxy-0.022
 hus-0.022
Token loved
Token ID6151
Feature activation+7.54e-01

Pos logit contributions
asha+0.013
oxy+0.013
avi+0.013
Le+0.012
ita+0.012
Neg logit contributions
keeping-0.012
taker-0.010
 unatt-0.010
handed-0.010
nob-0.009
Token his
Token ID465
Feature activation-8.72e-01

Pos logit contributions
Sal+0.015
ita+0.015
oxy+0.015
Le+0.014
avi+0.014
Neg logit contributions
keeping-0.022
 unatt-0.019
handed-0.018
taker-0.017
 rubble-0.017
Token cereal
Token ID33158
Feature activation+7.15e-01

Pos logit contributions
keeping+0.019
 unatt+0.017
handed+0.017
taker+0.017
 rubble+0.015
Neg logit contributions
oxy-0.022
ita-0.021
Sal-0.021
Le-0.021
avi-0.021
Token because
Token ID780
Feature activation-2.99e+00

Pos logit contributions
Sal+0.013
 ho+0.011
oxy+0.011
ita+0.011
 hus+0.011
Neg logit contributions
keeping-0.023
handed-0.020
 unatt-0.020
taker-0.019
 rubble-0.017
Token it
Token ID340
Feature activation-5.97e-01

Pos logit contributions
keeping+0.083
handed+0.073
 unatt+0.073
taker+0.072
 rubble+0.064
Neg logit contributions
Sal-0.064
ita-0.058
oxy-0.057
 ho-0.057
 hus-0.055
Token was
Token ID373
Feature activation-9.07e-01

Pos logit contributions
keeping+0.019
handed+0.017
 unatt+0.017
taker+0.017
 rubble+0.015
Neg logit contributions
Sal-0.010
 ho-0.009
oxy-0.009
ita-0.009
 hus-0.009
Token always
Token ID1464
Feature activation-1.04e+00

Pos logit contributions
keeping+0.028
 unatt+0.025
taker+0.024
handed+0.024
 rubble+0.022
Neg logit contributions
Sal-0.017
oxy-0.015
ita-0.015
Le-0.015
 ho-0.014
Token reliable
Token ID9314
Feature activation-4.63e-01

Pos logit contributions
keeping+0.034
handed+0.031
 unatt+0.029
taker+0.029
 rubble+0.026
Neg logit contributions
Sal-0.018
 ho-0.016
 short-0.016
 chicken-0.016
 hus-0.016
Token to
Token ID284
Feature activation-1.84e+00

Pos logit contributions
keeping+0.017
handed+0.016
 unatt+0.015
taker+0.015
 rubble+0.014
Neg logit contributions
Sal-0.005
 chicken-0.005
 ho-0.005
 short-0.005
 hus-0.005