TOP ACTIVATIONS
MAX = 0.085

the zipper and gasped with amaz ement at all the treasures
was a little girl named Is la . She loved basketball
and M ummy gasped with amaz ement . "
children watched from outside with amaz ement . They wanted to
her eyes grew wide with amaz ement . She wanted to
. M illy gasped with amaz ement ! "
her mom my watched in amaz ement as the machine flew
animals were watching him with amaz ement .
ered in , blinking in amaz ement . But
both laughed and smiled in amaz ement . Bob had printed
berry bush and stared in amaz ement at all the juicy
my 's eyes widened with amaz ement . Grand ma then
was filled with wonder and amaz ement . He was so
they both looked around in amaz ement . The lake was
were filled with wonder and amaz ement . They thanked the
, her eyes wide with amaz ement . "
. Tim my watched with amaz ement as he saw stars
made him sh iver with amaz ement . He went around
Her eye spark led with amaz ement . Everything was so
gasped and looked up in amaz ement . The flower smiled

MIN ACTIVATIONS
MIN = -0.097

Maybe they could grow more av oc ados and share them
every day to supply the av oc ados . And the
and try to grab the av oc ados . But the
Everyday he would take the av oc ados and scatter them
" I will buy more av oc ados tomorrow . You
his new knowledge he made gu ac am ole .
weather before he picked his av oc ados . <|endoftext|> Once
happy that it produced more av oc ados than ever before
something new ! She thought av oc ados were the most
whistle . It made the nic est sound , like a
. So Bill picked the nic est one he could find
something new ! She thought av oc ados were the most
He quickly hung colorful or nam ents and tw ink ling
. Did you know that av oc ados are very healthy
ksh akes . The mil ksh akes were blue and pink
the store to get more av oc ados . Gary looked
if she could request some av oc ados from the store
they had never seen the av oc ados . They wished
It turned out that the av oc ados were bad and
store and forgot to buy av oc ados . Lily was

INTERVAL 0.040 - 0.085
CONTAINS 0.222%

He couldn 't believe it was all for him
was very frustrated with him because she had told him many
the store . Maggie was feeling excited and happy . She
and white and had a sign that said " Circ us
dragon was pleased and took Maria up into the sky .

INTERVAL -0.006 - 0.040
CONTAINS 68.638%

, than it was to be selfish and keep things only
match met a bright , shiny coin . The coin told
-- mem or izing the answers and trying to remember them
She picked it up and went to her mom to show
because he had never been to the bar ber before .

INTERVAL -0.051 - -0.006
CONTAINS 31.112%

ives . One day , Benny received a message from his
surfaces to make sure everything was sparkling clean . After that
Jack nodded and realized he should always have
One day she was walking in the forest , searching for
boy named Tim my . Tim my loved to eat candy

INTERVAL -0.097 - -0.051
CONTAINS 0.027%

They share and help each other . They are good friends
, I can make your t eddy bear disappear and then
Friends also listen to each other . That is how we
and try to grab the av oc ados . But the
agreed . They gave each other a hug and said ,
Token the
Token ID262
Feature activation-5.95e-03

Pos logit contributions
iley+0.007
zed+0.006
 seeing+0.006
 tending+0.006
 inspection+0.006
Neg logit contributions
ags-0.003
 off-0.003
 fart-0.003
ac-0.003
 brothers-0.003
Token zipper
Token ID48992
Feature activation-1.98e-03

Pos logit contributions
ags+0.009
ac+0.008
 fart+0.008
 brothers+0.007
 teams+0.007
Neg logit contributions
iley-0.009
 inspection-0.009
joy-0.009
 seeing-0.009
ya-0.009
Token and
Token ID290
Feature activation-2.37e-02

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 brothers+0.002
 ends+0.002
Neg logit contributions
iley-0.004
 seeing-0.004
 inspection-0.004
 tending-0.004
joy-0.004
Token gasped
Token ID45236
Feature activation+2.41e-02

Pos logit contributions
ags+0.036
 fart+0.033
ac+0.033
 brothers+0.030
 ends+0.030
Neg logit contributions
iley-0.036
 seeing-0.035
 inspection-0.035
 tending-0.034
zed-0.034
Token with
Token ID351
Feature activation+1.71e-02

Pos logit contributions
iley+0.046
 inspection+0.046
zed+0.045
anne+0.045
aster+0.045
Neg logit contributions
 off-0.024
 back-0.023
ac-0.023
 brothers-0.023
ags-0.023
Token amaz
Token ID40642
Feature activation+8.54e-02

Pos logit contributions
iley+0.030
 inspection+0.029
zed+0.029
 seeing+0.029
anne+0.028
Neg logit contributions
ags-0.021
 off-0.020
 brothers-0.019
ac-0.019
 back-0.019
Tokenement
Token ID972
Feature activation+4.25e-02

Pos logit contributions
iley+0.107
 seeing+0.105
aster+0.105
 tending+0.103
zed+0.102
Neg logit contributions
ags-0.151
ac-0.139
 off-0.136
ay-0.135
 back-0.133
Token at
Token ID379
Feature activation+1.44e-02

Pos logit contributions
iley+0.079
aster+0.079
zed+0.077
handed+0.076
ited+0.074
Neg logit contributions
 off-0.051
 back-0.046
 brothers-0.046
 both-0.045
 down-0.044
Token all
Token ID477
Feature activation+1.57e-05

Pos logit contributions
ited+0.031
anne+0.030
looking+0.030
iley+0.030
 inspection+0.029
Neg logit contributions
 brothers-0.013
 off-0.013
 both-0.012
 back-0.012
ags-0.012
Token the
Token ID262
Feature activation+1.30e-02

Pos logit contributions
iley+0.000
zed+0.000
joy+0.000
cious+0.000
 seeing+0.000
Neg logit contributions
 off-0.000
 brothers-0.000
ags-0.000
ac-0.000
 back-0.000
Token treasures
Token ID29561
Feature activation+1.70e-02

Pos logit contributions
aster+0.026
zed+0.025
iley+0.025
anne+0.025
 inspection+0.024
Neg logit contributions
 brothers-0.012
 cars-0.012
 fart-0.012
 friends-0.012
 ends-0.012
Token was
Token ID373
Feature activation+2.57e-03

Pos logit contributions
 are+0.010
 will+0.008
 can+0.008
'll+0.008
ac+0.007
Neg logit contributions
iley-0.036
aster-0.034
ya-0.034
 tending-0.034
zed-0.033
Token a
Token ID257
Feature activation-8.92e-04

Pos logit contributions
iley+0.005
 seeing+0.005
 inspection+0.005
 tending+0.005
joy+0.005
Neg logit contributions
ags-0.003
 fart-0.003
ac-0.003
 off-0.002
 brothers-0.002
Token little
Token ID1310
Feature activation-1.93e-03

Pos logit contributions
 brothers+0.001
 fart+0.001
ac+0.001
 back+0.001
 cars+0.001
Neg logit contributions
 inspection-0.002
zed-0.002
iley-0.002
handed-0.002
 seeing-0.002
Token girl
Token ID2576
Feature activation+1.55e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 brothers+0.002
 off+0.001
Neg logit contributions
zed-0.004
iley-0.004
 inspection-0.004
 seeing-0.004
 tending-0.004
Token named
Token ID3706
Feature activation+2.21e-02

Pos logit contributions
 inspection+0.004
zed+0.003
joy+0.003
 hesitation+0.003
handed+0.003
Neg logit contributions
 back-0.001
 both-0.001
 friends-0.001
 brothers-0.001
ac-0.001
Token Is
Token ID1148
Feature activation+8.34e-02

Pos logit contributions
iley+0.037
 seeing+0.035
 inspection+0.035
 tending+0.034
zed+0.034
Neg logit contributions
ags-0.030
 fart-0.028
ac-0.028
 off-0.026
 brothers-0.025
Tokenla
Token ID5031
Feature activation+8.93e-03

Pos logit contributions
iley+0.115
 inspection+0.110
joy+0.110
 tending+0.109
 seeing+0.109
Neg logit contributions
ags-0.126
ac-0.125
 off-0.124
 fart-0.120
 back-0.120
Token.
Token ID13
Feature activation+1.21e-02

Pos logit contributions
 inspection+0.015
iley+0.014
joy+0.014
 Heidi+0.013
 Samantha+0.013
Neg logit contributions
ac-0.013
 both-0.012
 back-0.012
 are-0.012
 brothers-0.012
Token She
Token ID1375
Feature activation+5.58e-03

Pos logit contributions
iley+0.025
handed+0.024
 tending+0.024
zed+0.024
joy+0.024
Neg logit contributions
ags-0.011
 are-0.010
 off-0.010
ac-0.010
 ends-0.009
Token loved
Token ID6151
Feature activation+2.66e-02

Pos logit contributions
 inspection+0.013
 hesitation+0.012
joy+0.012
iley+0.012
handed+0.012
Neg logit contributions
 are-0.004
 can-0.004
 will-0.004
 ends-0.004
 brothers-0.004
Token basketball
Token ID9669
Feature activation+1.46e-03

Pos logit contributions
iley+0.052
 inspection+0.050
zed+0.050
 seeing+0.050
joy+0.049
Neg logit contributions
ags-0.027
 fart-0.025
ac-0.024
 off-0.023
 brothers-0.023
Token and
Token ID290
Feature activation-8.22e-04

Pos logit contributions
ac+0.007
 fart+0.007
ags+0.007
 off+0.007
 brothers+0.007
Neg logit contributions
iley-0.011
 inspection-0.011
joy-0.011
 seeing-0.011
zed-0.011
Token M
Token ID337
Feature activation+1.36e-02

Pos logit contributions
 off+0.001
ac+0.001
 brothers+0.001
 both+0.001
 back+0.001
Neg logit contributions
zed-0.002
iley-0.002
handed-0.002
 seeing-0.002
 inspection-0.002
Tokenummy
Token ID13513
Feature activation-2.13e-03

Pos logit contributions
iley+0.023
zed+0.022
 seeing+0.022
 inspection+0.022
joy+0.022
Neg logit contributions
ags-0.018
ac-0.016
 fart-0.016
 brothers-0.016
 ends-0.015
Token gasped
Token ID45236
Feature activation+2.35e-02

Pos logit contributions
ags+0.002
ac+0.002
 fart+0.002
 brothers+0.002
 off+0.002
Neg logit contributions
iley-0.004
 seeing-0.004
joy-0.004
 inspection-0.004
zed-0.004
Token with
Token ID351
Feature activation+1.82e-02

Pos logit contributions
iley+0.049
zed+0.048
 inspection+0.048
aster+0.047
anne+0.046
Neg logit contributions
 off-0.019
ags-0.019
ac-0.018
 back-0.018
 fart-0.017
Token amaz
Token ID40642
Feature activation+8.27e-02

Pos logit contributions
iley+0.029
zed+0.028
 inspection+0.027
 seeing+0.027
aster+0.027
Neg logit contributions
ags-0.025
 off-0.023
 fart-0.023
 are-0.023
ac-0.023
Tokenement
Token ID972
Feature activation+3.89e-02

Pos logit contributions
iley+0.100
aster+0.096
zed+0.095
 seeing+0.095
 tending+0.093
Neg logit contributions
ags-0.152
ac-0.141
 off-0.136
 back-0.133
ay-0.133
Token.
Token ID13
Feature activation-5.78e-03

Pos logit contributions
aster+0.075
zed+0.074
iley+0.074
handed+0.073
 occasion+0.069
Neg logit contributions
 off-0.045
 both-0.042
 back-0.040
 brothers-0.040
 friends-0.038
Token
Token ID198
Feature activation+1.88e-03

Pos logit contributions
ags+0.007
 fart+0.006
ac+0.006
 off+0.006
 brothers+0.006
Neg logit contributions
iley-0.011
 seeing-0.010
 inspection-0.010
joy-0.010
 tending-0.010
Token
Token ID198
Feature activation+3.51e-03

Pos logit contributions
iley+0.003
 seeing+0.003
 inspection+0.003
 tending+0.003
zed+0.003
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 off-0.002
 brothers-0.002
Token"
Token ID1
Feature activation+7.23e-03

Pos logit contributions
iley+0.006
 seeing+0.006
 inspection+0.006
zed+0.006
 tending+0.006
Neg logit contributions
ags-0.005
 fart-0.004
ac-0.004
 off-0.004
 brothers-0.004
Token children
Token ID1751
Feature activation-3.16e-03

Pos logit contributions
iley+0.015
 seeing+0.015
 inspection+0.015
 tending+0.015
zed+0.015
Neg logit contributions
ac-0.010
ags-0.010
 fart-0.010
 brothers-0.010
 ends-0.010
Token watched
Token ID7342
Feature activation+6.59e-03

Pos logit contributions
ags+0.003
ac+0.003
 fart+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.006
joy-0.006
 inspection-0.006
zed-0.006
 seeing-0.006
Token from
Token ID422
Feature activation-2.50e-02

Pos logit contributions
iley+0.013
zed+0.012
handed+0.012
 inspection+0.012
joy+0.012
Neg logit contributions
ags-0.007
 off-0.007
ac-0.006
 fart-0.006
 back-0.006
Token outside
Token ID2354
Feature activation-2.02e-03

Pos logit contributions
ags+0.023
 back+0.023
 cars+0.022
 off+0.022
 both+0.022
Neg logit contributions
iley-0.051
zed-0.050
anne-0.048
handed-0.047
 inspection-0.046
Token with
Token ID351
Feature activation+2.72e-03

Pos logit contributions
ags+0.002
 off+0.002
ac+0.002
 back+0.002
 fart+0.002
Neg logit contributions
iley-0.004
zed-0.004
ya-0.004
handed-0.004
 inspection-0.004
Token amaz
Token ID40642
Feature activation+8.00e-02

Pos logit contributions
iley+0.005
zed+0.005
 inspection+0.005
ya+0.005
 seeing+0.005
Neg logit contributions
ags-0.003
 brothers-0.003
 both-0.003
 off-0.003
 are-0.003
Tokenement
Token ID972
Feature activation+3.44e-02

Pos logit contributions
iley+0.089
 tending+0.084
 seeing+0.084
zed+0.083
 gazing+0.081
Neg logit contributions
ags-0.154
ac-0.141
 off-0.138
 brothers-0.134
 ends-0.133
Token.
Token ID13
Feature activation-2.64e-03

Pos logit contributions
iley+0.065
handed+0.063
zed+0.063
aster+0.061
joy+0.061
Neg logit contributions
 off-0.038
 brothers-0.035
 both-0.034
 back-0.034
ags-0.033
Token They
Token ID1119
Feature activation+1.46e-02

Pos logit contributions
ags+0.003
 fart+0.002
ac+0.002
 off+0.002
 brothers+0.002
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
zed-0.005
 tending-0.005
Token wanted
Token ID2227
Feature activation-4.64e-03

Pos logit contributions
iley+0.026
 seeing+0.025
 inspection+0.025
 tending+0.025
zed+0.025
Neg logit contributions
ags-0.017
 fart-0.016
ac-0.016
 off-0.014
 brothers-0.014
Token to
Token ID284
Feature activation+1.69e-02

Pos logit contributions
 off+0.003
 fart+0.003
 back+0.003
ags+0.003
ac+0.003
Neg logit contributions
iley-0.010
zed-0.010
ited-0.010
joy-0.009
handed-0.009
Token her
Token ID607
Feature activation-1.13e-04

Pos logit contributions
 off+0.002
 are+0.002
 cars+0.002
 brothers+0.002
 back+0.002
Neg logit contributions
iley-0.002
anne-0.001
ited-0.001
zed-0.001
ya-0.001
Token eyes
Token ID2951
Feature activation+1.06e-02

Pos logit contributions
ags+0.000
 fart+0.000
ac+0.000
 brothers+0.000
 ends+0.000
Neg logit contributions
iley-0.000
 inspection-0.000
zed-0.000
 seeing-0.000
 tending-0.000
Token grew
Token ID6348
Feature activation-5.36e-03

Pos logit contributions
iley+0.021
ya+0.019
joy+0.019
 inspection+0.018
 tending+0.018
Neg logit contributions
 off-0.012
 fart-0.011
 are-0.011
 ends-0.011
ags-0.011
Token wide
Token ID3094
Feature activation-6.92e-03

Pos logit contributions
ags+0.007
ac+0.006
 fart+0.006
Bang+0.005
 brothers+0.005
Neg logit contributions
 seeing-0.009
joy-0.009
 tending-0.009
 inspection-0.009
 gazing-0.009
Token with
Token ID351
Feature activation+7.08e-03

Pos logit contributions
 off+0.007
 back+0.007
 fart+0.007
ac+0.006
 ends+0.006
Neg logit contributions
iley-0.014
zed-0.014
aster-0.013
 inspection-0.013
anne-0.013
Token amaz
Token ID40642
Feature activation+7.79e-02

Pos logit contributions
iley+0.014
zed+0.013
 inspection+0.013
aster+0.013
ya+0.013
Neg logit contributions
ags-0.007
 off-0.007
 fart-0.007
ac-0.007
 ends-0.007
Tokenement
Token ID972
Feature activation+3.62e-02

Pos logit contributions
iley+0.096
 seeing+0.092
zed+0.091
 tending+0.090
aster+0.089
Neg logit contributions
ags-0.141
ac-0.131
 off-0.127
 back-0.126
ay-0.125
Token.
Token ID13
Feature activation-2.75e-05

Pos logit contributions
iley+0.078
zed+0.074
aster+0.074
ited+0.074
handed+0.074
Neg logit contributions
 off-0.039
 back-0.034
 down-0.032
 into-0.031
 from-0.031
Token She
Token ID1375
Feature activation+6.27e-03

Pos logit contributions
ags+0.000
 fart+0.000
 off+0.000
ac+0.000
 brothers+0.000
Neg logit contributions
iley-0.000
 seeing-0.000
 tending-0.000
zed-0.000
 inspection-0.000
Token wanted
Token ID2227
Feature activation-1.78e-03

Pos logit contributions
iley+0.009
 seeing+0.009
 inspection+0.009
zed+0.009
 tending+0.009
Neg logit contributions
ags-0.010
 fart-0.009
ac-0.009
 off-0.008
 brothers-0.008
Token to
Token ID284
Feature activation+1.57e-02

Pos logit contributions
 off+0.001
 back+0.001
ags+0.001
 fart+0.001
 cars+0.001
Neg logit contributions
iley-0.004
zed-0.004
ited-0.004
anne-0.004
joy-0.004
Token.
Token ID13
Feature activation-2.46e-03

Pos logit contributions
iley+0.028
 inspection+0.025
zed+0.025
joy+0.025
aster+0.025
Neg logit contributions
 off-0.019
 back-0.018
 fart-0.017
ags-0.017
 are-0.017
Token M
Token ID337
Feature activation-1.14e-03

Pos logit contributions
ags+0.003
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.004
 seeing-0.004
 inspection-0.004
joy-0.004
zed-0.004
Tokenilly
Token ID6548
Feature activation-4.16e-03

Pos logit contributions
ags+0.002
ac+0.002
 brothers+0.002
 fart+0.002
 off+0.002
Neg logit contributions
 inspection-0.001
 seeing-0.001
zed-0.001
 tending-0.001
aster-0.001
Token gasped
Token ID45236
Feature activation+1.23e-02

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.007
 seeing-0.007
zed-0.007
 inspection-0.006
 tending-0.006
Token with
Token ID351
Feature activation+1.63e-02

Pos logit contributions
zed+0.028
 inspection+0.028
iley+0.028
aster+0.028
anne+0.028
Neg logit contributions
 off-0.008
 back-0.008
 both-0.007
ags-0.007
 down-0.007
Token amaz
Token ID40642
Feature activation+7.59e-02

Pos logit contributions
iley+0.026
 seeing+0.025
 inspection+0.024
 tending+0.024
joy+0.024
Neg logit contributions
ags-0.023
 fart-0.022
ac-0.022
 off-0.020
 brothers-0.020
Tokenement
Token ID972
Feature activation+3.66e-02

Pos logit contributions
iley+0.092
aster+0.090
 seeing+0.088
 tending+0.088
handed+0.084
Neg logit contributions
ags-0.140
ac-0.126
 off-0.124
 back-0.120
ay-0.119
Token!
Token ID0
Feature activation+5.28e-03

Pos logit contributions
aster+0.078
handed+0.077
iley+0.075
zed+0.075
 occasion+0.073
Neg logit contributions
 off-0.037
 both-0.035
 back-0.034
 down-0.031
 into-0.030
Token
Token ID198
Feature activation+3.98e-03

Pos logit contributions
iley+0.009
 seeing+0.009
 inspection+0.009
zed+0.009
 tending+0.009
Neg logit contributions
ags-0.006
 fart-0.006
ac-0.006
 off-0.005
 brothers-0.005
Token
Token ID198
Feature activation+6.22e-03

Pos logit contributions
iley+0.007
 seeing+0.007
 inspection+0.007
zed+0.007
 tending+0.007
Neg logit contributions
ags-0.005
 fart-0.004
ac-0.004
 off-0.004
 brothers-0.004
Token"
Token ID1
Feature activation+6.76e-03

Pos logit contributions
iley+0.011
 inspection+0.011
joy+0.011
 seeing+0.011
 tending+0.010
Neg logit contributions
ags-0.007
ac-0.007
 fart-0.007
 off-0.007
 brothers-0.006
Token her
Token ID607
Feature activation+1.05e-02

Pos logit contributions
ac+0.010
 fart+0.010
 off+0.010
ags+0.010
 brothers+0.009
Neg logit contributions
iley-0.012
 inspection-0.012
zed-0.012
joy-0.012
 tending-0.011
Token mom
Token ID1995
Feature activation-8.52e-03

Pos logit contributions
iley+0.020
 inspection+0.020
 seeing+0.020
zed+0.019
 tending+0.019
Neg logit contributions
ags-0.011
 fart-0.010
ac-0.010
 off-0.009
 brothers-0.009
Tokenmy
Token ID1820
Feature activation-1.48e-02

Pos logit contributions
ags+0.009
ac+0.009
 fart+0.008
 off+0.008
 brothers+0.008
Neg logit contributions
iley-0.016
 inspection-0.016
joy-0.015
zed-0.015
 seeing-0.015
Token watched
Token ID7342
Feature activation+1.37e-02

Pos logit contributions
ags+0.016
ac+0.015
 off+0.015
 brothers+0.015
 fart+0.015
Neg logit contributions
iley-0.027
 inspection-0.026
joy-0.026
zed-0.025
 seeing-0.025
Token in
Token ID287
Feature activation-2.60e-02

Pos logit contributions
iley+0.024
 inspection+0.023
joy+0.023
zed+0.023
 seeing+0.023
Neg logit contributions
ags-0.016
 off-0.016
ac-0.015
 fart-0.015
 back-0.014
Token amaz
Token ID40642
Feature activation+7.58e-02

Pos logit contributions
ags+0.023
 off+0.023
 back+0.023
 are+0.022
 both+0.021
Neg logit contributions
iley-0.053
zed-0.053
anne-0.053
handed-0.051
looking-0.051
Tokenement
Token ID972
Feature activation+1.93e-02

Pos logit contributions
iley+0.082
zed+0.077
 seeing+0.077
 tending+0.077
 inspection+0.074
Neg logit contributions
ags-0.146
ac-0.134
 off-0.132
 back-0.129
ay-0.127
Token as
Token ID355
Feature activation+6.36e-03

Pos logit contributions
iley+0.034
joy+0.033
zed+0.033
 inspection+0.032
handed+0.032
Neg logit contributions
 off-0.023
ags-0.021
ac-0.021
 back-0.021
 brothers-0.021
Token the
Token ID262
Feature activation-3.72e-03

Pos logit contributions
iley+0.009
 inspection+0.009
zed+0.009
joy+0.009
ited+0.008
Neg logit contributions
 off-0.009
 brothers-0.009
ac-0.009
 both-0.009
 are-0.009
Token machine
Token ID4572
Feature activation+3.09e-03

Pos logit contributions
 brothers+0.004
 fart+0.004
 ends+0.004
 cars+0.004
ags+0.004
Neg logit contributions
 inspection-0.007
zed-0.007
iley-0.007
 seeing-0.007
joy-0.006
Token flew
Token ID13112
Feature activation+4.41e-04

Pos logit contributions
iley+0.006
joy+0.006
 inspection+0.005
 tending+0.005
 seeing+0.005
Neg logit contributions
 off-0.003
 fart-0.003
ags-0.003
ac-0.003
 ends-0.003
Token animals
Token ID4695
Feature activation+1.15e-02

Pos logit contributions
ags+0.015
 fart+0.014
ac+0.014
 off+0.012
 brothers+0.012
Neg logit contributions
iley-0.037
 seeing-0.036
 inspection-0.036
zed-0.036
 tending-0.036
Token were
Token ID547
Feature activation+1.29e-02

Pos logit contributions
iley+0.024
joy+0.023
zed+0.023
 inspection+0.023
 seeing+0.022
Neg logit contributions
ags-0.010
 fart-0.009
ac-0.009
 off-0.009
 are-0.009
Token watching
Token ID4964
Feature activation-1.47e-03

Pos logit contributions
iley+0.024
 inspection+0.023
zed+0.023
aster+0.022
joy+0.022
Neg logit contributions
ags-0.015
 fart-0.014
 off-0.014
ac-0.013
 brothers-0.013
Token him
Token ID683
Feature activation-1.51e-02

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.002
 brothers+0.002
Neg logit contributions
iley-0.003
 seeing-0.003
 inspection-0.002
 tending-0.002
zed-0.002
Token with
Token ID351
Feature activation+7.50e-03

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.017
 off+0.015
 brothers+0.015
Neg logit contributions
iley-0.027
 seeing-0.026
 inspection-0.026
 tending-0.025
zed-0.025
Token amaz
Token ID40642
Feature activation+7.50e-02

Pos logit contributions
iley+0.016
ya+0.015
zed+0.015
 inspection+0.015
anne+0.014
Neg logit contributions
ags-0.007
 off-0.007
 brothers-0.006
 fart-0.006
 are-0.006
Tokenement
Token ID972
Feature activation+2.57e-02

Pos logit contributions
iley+0.073
 seeing+0.068
zed+0.066
 tending+0.066
 inspection+0.065
Neg logit contributions
ags-0.153
ac-0.145
 fart-0.142
 off-0.140
 brothers-0.137
Token.
Token ID13
Feature activation-1.93e-03

Pos logit contributions
iley+0.046
zed+0.043
joy+0.043
aster+0.042
handed+0.041
Neg logit contributions
 off-0.032
ags-0.030
 brothers-0.029
 fart-0.029
ac-0.029
Token
Token ID220
Feature activation-8.37e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.002
 ends+0.002
Neg logit contributions
iley-0.004
 seeing-0.003
 inspection-0.003
joy-0.003
 tending-0.003
Token
Token ID198
Feature activation+4.21e-03

Pos logit contributions
ags+0.011
 fart+0.011
ac+0.010
 off+0.010
 ends+0.010
Neg logit contributions
iley-0.013
 inspection-0.013
 seeing-0.013
zed-0.013
joy-0.013
Token
Token ID198
Feature activation+6.77e-03

Pos logit contributions
iley+0.008
 seeing+0.008
 inspection+0.007
 tending+0.007
zed+0.007
Neg logit contributions
ags-0.005
 fart-0.004
ac-0.004
 off-0.004
 brothers-0.004
Tokenered
Token ID1068
Feature activation+2.44e-02

Pos logit contributions
iley+0.038
joy+0.036
 seeing+0.036
 tending+0.036
 inspection+0.035
Neg logit contributions
ags-0.034
ac-0.034
 fart-0.032
 off-0.031
 brothers-0.030
Token in
Token ID287
Feature activation-2.04e-02

Pos logit contributions
iley+0.056
zed+0.054
 seeing+0.053
 inspection+0.053
aster+0.053
Neg logit contributions
ags-0.017
 off-0.015
ac-0.015
 fart-0.014
 back-0.014
Token,
Token ID11
Feature activation+7.28e-03

Pos logit contributions
 off+0.017
 back+0.016
ags+0.015
ac+0.014
 ends+0.014
Neg logit contributions
zed-0.044
iley-0.044
anne-0.043
storm-0.042
handed-0.042
Token blinking
Token ID43196
Feature activation+1.22e-02

Pos logit contributions
iley+0.015
handed+0.013
aster+0.013
zed+0.013
 occasion+0.013
Neg logit contributions
 back-0.008
 off-0.007
 both-0.007
 down-0.007
 fart-0.007
Token in
Token ID287
Feature activation-1.02e-02

Pos logit contributions
iley+0.024
 seeing+0.023
zed+0.023
 tending+0.022
 inspection+0.022
Neg logit contributions
ags-0.012
ac-0.012
 fart-0.012
 off-0.011
 brothers-0.010
Token amaz
Token ID40642
Feature activation+7.47e-02

Pos logit contributions
 off+0.008
 back+0.007
ags+0.007
 ends+0.007
ac+0.007
Neg logit contributions
zed-0.022
iley-0.022
anne-0.022
ya-0.021
handed-0.021
Tokenement
Token ID972
Feature activation+2.98e-02

Pos logit contributions
iley+0.082
zed+0.080
 tending+0.079
 seeing+0.079
aster+0.078
Neg logit contributions
ags-0.141
ac-0.131
 off-0.128
 back-0.124
 ends-0.123
Token.
Token ID13
Feature activation-2.44e-03

Pos logit contributions
iley+0.055
zed+0.055
handed+0.054
aster+0.054
ited+0.054
Neg logit contributions
 off-0.035
 back-0.031
 into-0.030
 down-0.029
 both-0.029
Token
Token ID198
Feature activation+4.47e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 brothers+0.002
 off+0.002
Neg logit contributions
iley-0.005
 seeing-0.005
 tending-0.005
zed-0.005
 inspection-0.005
Token
Token ID198
Feature activation+6.39e-03

Pos logit contributions
iley+0.008
 seeing+0.008
 inspection+0.008
 tending+0.008
zed+0.008
Neg logit contributions
ags-0.005
 fart-0.005
ac-0.005
 off-0.004
 brothers-0.004
TokenBut
Token ID1537
Feature activation+2.10e-02

Pos logit contributions
iley+0.010
 seeing+0.010
 inspection+0.010
 tending+0.010
joy+0.010
Neg logit contributions
ags-0.008
ac-0.008
 fart-0.008
 off-0.007
 brothers-0.007
Token both
Token ID1111
Feature activation-2.28e-02

Pos logit contributions
 seeing+0.028
iley+0.028
 tending+0.027
 inspection+0.027
 gazing+0.026
Neg logit contributions
ags-0.021
 fart-0.018
ac-0.018
 off-0.016
 ends-0.016
Token laughed
Token ID13818
Feature activation-4.85e-03

Pos logit contributions
ac+0.027
ags+0.027
 fart+0.027
 off+0.025
 ends+0.025
Neg logit contributions
iley-0.039
zed-0.038
 inspection-0.038
 tending-0.037
joy-0.037
Token and
Token ID290
Feature activation+1.39e-04

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.004
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.010
 seeing-0.009
 inspection-0.009
 tending-0.009
joy-0.009
Token smiled
Token ID13541
Feature activation+3.84e-03

Pos logit contributions
iley+0.000
 inspection+0.000
zed+0.000
 tending+0.000
joy+0.000
Neg logit contributions
ags-0.000
 off-0.000
 fart-0.000
ac-0.000
 brothers-0.000
Token in
Token ID287
Feature activation-3.29e-03

Pos logit contributions
iley+0.007
zed+0.006
 inspection+0.006
aster+0.006
 seeing+0.006
Neg logit contributions
ags-0.005
 off-0.005
ac-0.005
 fart-0.005
 back-0.004
Token amaz
Token ID40642
Feature activation+7.40e-02

Pos logit contributions
 both+0.004
 off+0.004
ags+0.004
 back+0.003
 are+0.003
Neg logit contributions
zed-0.006
anne-0.006
iley-0.006
handed-0.006
 inspection-0.006
Tokenement
Token ID972
Feature activation+4.02e-02

Pos logit contributions
aster+0.096
iley+0.095
 tending+0.093
 seeing+0.091
zed+0.089
Neg logit contributions
ags-0.131
 off-0.116
ac-0.115
ay-0.113
 back-0.112
Token.
Token ID13
Feature activation-4.37e-03

Pos logit contributions
aster+0.077
iley+0.076
handed+0.074
zed+0.073
 tending+0.071
Neg logit contributions
 off-0.046
 both-0.043
 back-0.041
 brothers-0.040
 friends-0.040
Token Bob
Token ID5811
Feature activation+3.27e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.009
 seeing-0.008
 tending-0.008
zed-0.008
 inspection-0.008
Token had
Token ID550
Feature activation-3.10e-03

Pos logit contributions
iley+0.006
 seeing+0.005
zed+0.005
 tending+0.005
 inspection+0.005
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 off-0.003
 brothers-0.003
Token printed
Token ID10398
Feature activation-3.05e-03

Pos logit contributions
ags+0.004
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.006
 inspection-0.005
 tending-0.005
zed-0.005
 seeing-0.005
Tokenberry
Token ID8396
Feature activation-2.40e-02

Pos logit contributions
iley+0.008
 tending+0.008
 seeing+0.008
aster+0.008
joy+0.008
Neg logit contributions
ags-0.008
ac-0.007
 fart-0.007
 off-0.007
amed-0.007
Token bush
Token ID24484
Feature activation+1.64e-02

Pos logit contributions
ags+0.035
ac+0.032
 fart+0.031
Bang+0.029
 brothers+0.029
Neg logit contributions
iley-0.036
 seeing-0.036
 tending-0.036
aster-0.035
zed-0.035
Token and
Token ID290
Feature activation+3.05e-03

Pos logit contributions
iley+0.033
 inspection+0.031
joy+0.031
zed+0.031
 seeing+0.030
Neg logit contributions
ags-0.017
 fart-0.016
 off-0.016
ac-0.015
 back-0.015
Token stared
Token ID18484
Feature activation+2.86e-02

Pos logit contributions
iley+0.005
joy+0.005
 inspection+0.005
 tending+0.005
anne+0.005
Neg logit contributions
 fart-0.004
 off-0.004
ags-0.004
ac-0.004
 ends-0.003
Token in
Token ID287
Feature activation-1.23e-02

Pos logit contributions
iley+0.057
 seeing+0.055
joy+0.055
 inspection+0.055
 tending+0.055
Neg logit contributions
ags-0.028
 fart-0.027
ac-0.025
 ends-0.023
 brothers-0.023
Token amaz
Token ID40642
Feature activation+7.39e-02

Pos logit contributions
 back+0.010
 off+0.010
 both+0.009
ags+0.008
 are+0.008
Neg logit contributions
anne-0.029
iley-0.027
zed-0.027
iya-0.026
storm-0.026
Tokenement
Token ID972
Feature activation+4.19e-02

Pos logit contributions
iley+0.075
 seeing+0.072
 tending+0.071
 inspection+0.069
zed+0.069
Neg logit contributions
ags-0.149
ac-0.136
 off-0.134
 fart-0.130
 brothers-0.129
Token at
Token ID379
Feature activation+1.44e-02

Pos logit contributions
iley+0.073
handed+0.071
zed+0.071
 tending+0.070
 inspection+0.069
Neg logit contributions
ags-0.051
 off-0.051
 brothers-0.046
 back-0.046
ac-0.046
Token all
Token ID477
Feature activation-2.22e-03

Pos logit contributions
iley+0.032
anne+0.031
joy+0.030
ited+0.030
looking+0.030
Neg logit contributions
 both-0.012
 back-0.012
 off-0.011
ags-0.011
 are-0.010
Token the
Token ID262
Feature activation+1.03e-02

Pos logit contributions
ags+0.002
 off+0.002
ac+0.002
 brothers+0.002
 ends+0.002
Neg logit contributions
iley-0.005
joy-0.005
zed-0.005
 seeing-0.005
 tending-0.004
Token juicy
Token ID35075
Feature activation+8.96e-03

Pos logit contributions
iley+0.021
zed+0.021
 inspection+0.020
 seeing+0.020
joy+0.020
Neg logit contributions
ags-0.009
 fart-0.009
ac-0.009
 brothers-0.008
 ends-0.008
Tokenmy
Token ID1820
Feature activation-1.90e-02

Pos logit contributions
ags+0.022
 fart+0.021
 teams+0.018
ac+0.017
 cars+0.017
Neg logit contributions
iley-0.039
anne-0.037
ya-0.036
 seeing-0.036
zed-0.036
Token's
Token ID338
Feature activation-1.27e-02

Pos logit contributions
ags+0.028
 fart+0.026
ac+0.025
 brothers+0.023
 off+0.023
Neg logit contributions
iley-0.030
zed-0.029
 seeing-0.028
 inspection-0.028
 tending-0.028
Token eyes
Token ID2951
Feature activation+3.31e-03

Pos logit contributions
ags+0.013
ac+0.012
 fart+0.012
 brothers+0.011
 ends+0.011
Neg logit contributions
iley-0.025
 inspection-0.024
 seeing-0.024
 tending-0.024
zed-0.023
Token widened
Token ID36405
Feature activation+2.19e-02

Pos logit contributions
iley+0.005
 seeing+0.005
 inspection+0.005
joy+0.004
zed+0.004
Neg logit contributions
ags-0.005
 fart-0.005
ac-0.005
 off-0.005
 ends-0.004
Token with
Token ID351
Feature activation+1.32e-02

Pos logit contributions
iley+0.047
 inspection+0.046
zed+0.045
aster+0.044
anne+0.044
Neg logit contributions
 off-0.017
 fart-0.017
ac-0.016
ags-0.016
 back-0.016
Token amaz
Token ID40642
Feature activation+7.35e-02

Pos logit contributions
iley+0.025
 inspection+0.024
zed+0.024
anne+0.024
ya+0.023
Neg logit contributions
ags-0.014
 off-0.014
 brothers-0.014
 are-0.013
 back-0.013
Tokenement
Token ID972
Feature activation+3.25e-02

Pos logit contributions
iley+0.092
aster+0.087
 seeing+0.086
 tending+0.084
 inspection+0.083
Neg logit contributions
ags-0.133
ac-0.121
 off-0.118
 back-0.118
ay-0.116
Token.
Token ID13
Feature activation-2.73e-03

Pos logit contributions
iley+0.066
zed+0.065
aster+0.065
handed+0.063
ited+0.063
Neg logit contributions
 off-0.033
 back-0.031
 brothers-0.030
 both-0.029
 down-0.028
Token Grand
Token ID5675
Feature activation+1.61e-02

Pos logit contributions
ags+0.003
 fart+0.003
ac+0.003
 off+0.002
 ends+0.002
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
joy-0.005
zed-0.005
Tokenma
Token ID2611
Feature activation-7.11e-03

Pos logit contributions
iley+0.024
 seeing+0.023
 inspection+0.023
zed+0.022
 tending+0.022
Neg logit contributions
ags-0.024
ac-0.023
 fart-0.023
 brothers-0.022
 off-0.022
Token then
Token ID788
Feature activation-6.57e-03

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.009
 off+0.009
 ends+0.008
Neg logit contributions
iley-0.011
 seeing-0.011
 tending-0.011
 inspection-0.011
zed-0.011
Token was
Token ID373
Feature activation+6.52e-03

Pos logit contributions
iley+0.018
 inspection+0.018
joy+0.017
 seeing+0.017
 tending+0.017
Neg logit contributions
ags-0.013
ac-0.013
 fart-0.013
 off-0.012
 ends-0.012
Token filled
Token ID5901
Feature activation-1.52e-02

Pos logit contributions
iley+0.013
 inspection+0.013
zed+0.013
aster+0.012
joy+0.012
Neg logit contributions
ags-0.006
 fart-0.006
ac-0.006
 off-0.006
 brothers-0.005
Token with
Token ID351
Feature activation-5.07e-03

Pos logit contributions
 off+0.010
ags+0.010
 fart+0.010
ac+0.010
 back+0.009
Neg logit contributions
iley-0.034
 inspection-0.034
joy-0.033
 hesitation-0.032
anne-0.032
Token wonder
Token ID4240
Feature activation+2.50e-02

Pos logit contributions
ags+0.007
 fart+0.006
ac+0.006
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.009
 seeing-0.008
 inspection-0.008
joy-0.008
 tending-0.008
Token and
Token ID290
Feature activation+4.33e-03

Pos logit contributions
iley+0.046
zed+0.043
aster+0.043
handed+0.043
anne+0.042
Neg logit contributions
 off-0.030
 back-0.028
ags-0.027
 are-0.026
ac-0.026
Token amaz
Token ID40642
Feature activation+7.31e-02

Pos logit contributions
iley+0.007
aster+0.007
joy+0.007
 inspection+0.007
handed+0.006
Neg logit contributions
 off-0.006
 fart-0.005
 brothers-0.005
ags-0.005
ac-0.005
Tokenement
Token ID972
Feature activation+3.82e-02

Pos logit contributions
iley+0.088
aster+0.087
 tending+0.082
 seeing+0.082
zed+0.080
Neg logit contributions
ags-0.132
ac-0.121
 off-0.120
 back-0.118
ay-0.115
Token.
Token ID13
Feature activation+4.60e-03

Pos logit contributions
iley+0.070
aster+0.069
zed+0.066
 tending+0.066
handed+0.065
Neg logit contributions
 off-0.046
 back-0.044
 both-0.041
am-0.041
 down-0.040
Token He
Token ID679
Feature activation+7.61e-03

Pos logit contributions
iley+0.009
 seeing+0.009
joy+0.009
 inspection+0.009
zed+0.009
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 off-0.003
 ends-0.003
Token was
Token ID373
Feature activation+6.48e-03

Pos logit contributions
iley+0.012
 seeing+0.011
 inspection+0.011
zed+0.011
 tending+0.011
Neg logit contributions
ags-0.011
 fart-0.010
ac-0.010
 off-0.010
 brothers-0.009
Token so
Token ID523
Feature activation-5.05e-03

Pos logit contributions
 inspection+0.013
iley+0.013
aster+0.013
zed+0.013
handed+0.013
Neg logit contributions
 off-0.006
ags-0.005
 back-0.005
ac-0.005
 both-0.005
Token they
Token ID484
Feature activation-2.97e-03

Pos logit contributions
 off+0.022
 fart+0.022
ags+0.021
ac+0.021
 both+0.021
Neg logit contributions
iley-0.034
zed-0.033
 inspection-0.032
 tending-0.031
handed-0.031
Token both
Token ID1111
Feature activation-2.31e-02

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
 tending-0.005
zed-0.005
Token looked
Token ID3114
Feature activation+1.24e-02

Pos logit contributions
ac+0.028
ags+0.028
 fart+0.028
 off+0.027
 brothers+0.026
Neg logit contributions
iley-0.039
 inspection-0.037
joy-0.037
 tending-0.037
 seeing-0.037
Token around
Token ID1088
Feature activation-1.45e-03

Pos logit contributions
iley+0.025
 seeing+0.024
 inspection+0.024
zed+0.024
 tending+0.024
Neg logit contributions
ags-0.012
 fart-0.011
ac-0.010
 off-0.009
 brothers-0.009
Token in
Token ID287
Feature activation-1.57e-02

Pos logit contributions
 back+0.001
 off+0.001
 both+0.001
ags+0.001
ac+0.001
Neg logit contributions
handed-0.003
iley-0.003
zed-0.003
aster-0.003
anne-0.003
Token amaz
Token ID40642
Feature activation+7.22e-02

Pos logit contributions
 back+0.013
 are+0.013
 off+0.013
 both+0.012
ags+0.012
Neg logit contributions
zed-0.035
anne-0.034
iley-0.034
storm-0.033
looking-0.033
Tokenement
Token ID972
Feature activation+2.49e-02

Pos logit contributions
iley+0.082
zed+0.078
aster+0.077
 tending+0.077
 seeing+0.076
Neg logit contributions
ags-0.137
ac-0.125
 off-0.123
 back-0.119
ay-0.118
Token.
Token ID13
Feature activation-6.89e-03

Pos logit contributions
iley+0.048
zed+0.047
aster+0.047
handed+0.047
ited+0.046
Neg logit contributions
 off-0.027
 back-0.024
 both-0.024
 into-0.022
 friends-0.022
Token The
Token ID383
Feature activation+4.25e-04

Pos logit contributions
ags+0.007
 fart+0.006
ac+0.006
 off+0.006
 brothers+0.006
Neg logit contributions
iley-0.014
 seeing-0.013
zed-0.013
 tending-0.013
 inspection-0.013
Token lake
Token ID13546
Feature activation+2.41e-03

Pos logit contributions
iley+0.001
 seeing+0.001
 inspection+0.001
zed+0.001
 tending+0.001
Neg logit contributions
ags-0.001
 fart-0.000
ac-0.000
 off-0.000
 brothers-0.000
Token was
Token ID373
Feature activation+1.86e-03

Pos logit contributions
iley+0.005
joy+0.004
zed+0.004
aster+0.004
 inspection+0.004
Neg logit contributions
 fart-0.002
ac-0.002
ags-0.002
 ends-0.002
 off-0.002
Token were
Token ID547
Feature activation+9.39e-03

Pos logit contributions
iley+0.021
 inspection+0.020
 seeing+0.020
joy+0.020
zed+0.019
Neg logit contributions
ags-0.012
 fart-0.012
ac-0.012
 off-0.011
 brothers-0.011
Token filled
Token ID5901
Feature activation-1.54e-02

Pos logit contributions
iley+0.020
 inspection+0.019
zed+0.019
joy+0.018
aster+0.018
Neg logit contributions
ags-0.008
ac-0.008
 off-0.008
 fart-0.008
 brothers-0.007
Token with
Token ID351
Feature activation-9.09e-03

Pos logit contributions
ags+0.016
 fart+0.015
ac+0.014
 off+0.013
 brothers+0.013
Neg logit contributions
iley-0.030
 seeing-0.029
 inspection-0.029
zed-0.029
 tending-0.029
Token wonder
Token ID4240
Feature activation+2.13e-02

Pos logit contributions
ags+0.011
 fart+0.010
ac+0.010
 off+0.009
 brothers+0.009
Neg logit contributions
iley-0.016
 seeing-0.016
 inspection-0.015
 tending-0.015
joy-0.015
Token and
Token ID290
Feature activation+2.56e-03

Pos logit contributions
iley+0.042
zed+0.039
aster+0.039
 inspection+0.039
joy+0.039
Neg logit contributions
ags-0.021
 off-0.021
ac-0.020
 back-0.019
 fart-0.019
Token amaz
Token ID40642
Feature activation+7.14e-02

Pos logit contributions
iley+0.004
 inspection+0.004
joy+0.004
zed+0.004
aster+0.004
Neg logit contributions
 off-0.003
ags-0.003
ac-0.003
 fart-0.003
 brothers-0.003
Tokenement
Token ID972
Feature activation+3.69e-02

Pos logit contributions
iley+0.079
aster+0.073
 seeing+0.071
 tending+0.070
zed+0.070
Neg logit contributions
ags-0.137
ac-0.130
 off-0.126
 back-0.123
 ends-0.123
Token.
Token ID13
Feature activation+1.14e-03

Pos logit contributions
iley+0.064
aster+0.062
zed+0.059
 tending+0.058
joy+0.058
Neg logit contributions
 off-0.046
 back-0.044
ac-0.043
 brothers-0.043
 both-0.043
Token They
Token ID1119
Feature activation+1.97e-02

Pos logit contributions
iley+0.002
 seeing+0.002
 inspection+0.002
zed+0.002
 tending+0.002
Neg logit contributions
ags-0.001
 fart-0.001
ac-0.001
 off-0.001
 brothers-0.001
Token thanked
Token ID26280
Feature activation-1.74e-02

Pos logit contributions
iley+0.034
 seeing+0.033
 inspection+0.032
 tending+0.032
zed+0.032
Neg logit contributions
ags-0.025
 fart-0.023
ac-0.023
 off-0.021
 brothers-0.021
Token the
Token ID262
Feature activation-1.37e-02

Pos logit contributions
ags+0.020
 fart+0.018
ac+0.018
 off+0.017
 brothers+0.017
Neg logit contributions
iley-0.032
 seeing-0.031
 inspection-0.031
zed-0.030
 tending-0.030
Token,
Token ID11
Feature activation-2.02e-03

Pos logit contributions
iley+0.001
ited+0.001
 inspection+0.001
ya+0.001
joy+0.001
Neg logit contributions
 off-0.001
 back-0.001
 down-0.001
 both-0.000
 into-0.000
Token her
Token ID607
Feature activation-1.18e-02

Pos logit contributions
 off+0.002
 back+0.002
 both+0.002
 down+0.002
 into+0.002
Neg logit contributions
iley-0.004
ited-0.004
handed-0.004
 inspection-0.004
anne-0.003
Token eyes
Token ID2951
Feature activation+6.97e-03

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 off+0.011
 ends+0.011
Neg logit contributions
iley-0.022
 seeing-0.021
joy-0.021
 inspection-0.021
 tending-0.021
Token wide
Token ID3094
Feature activation-4.65e-04

Pos logit contributions
iley+0.013
ya+0.012
 inspection+0.012
aster+0.011
zed+0.011
Neg logit contributions
 off-0.009
 back-0.008
 both-0.008
ags-0.008
 fart-0.008
Token with
Token ID351
Feature activation+3.73e-03

Pos logit contributions
 off+0.001
 back+0.001
ags+0.000
 down+0.000
ac+0.000
Neg logit contributions
iley-0.001
aster-0.001
zed-0.001
 inspection-0.001
handed-0.001
Token amaz
Token ID40642
Feature activation+7.12e-02

Pos logit contributions
iley+0.007
zed+0.006
 inspection+0.006
 seeing+0.006
ya+0.006
Neg logit contributions
ags-0.005
 off-0.004
 fart-0.004
ac-0.004
 brothers-0.004
Tokenement
Token ID972
Feature activation+2.98e-02

Pos logit contributions
iley+0.092
 seeing+0.086
zed+0.085
 tending+0.084
aster+0.084
Neg logit contributions
ags-0.129
ac-0.117
 back-0.114
 off-0.113
ay-0.111
Token.
Token ID13
Feature activation-4.98e-03

Pos logit contributions
iley+0.065
handed+0.064
zed+0.063
aster+0.063
Charl+0.061
Neg logit contributions
 off-0.030
 back-0.027
 both-0.025
 down-0.024
 into-0.023
Token
Token ID198
Feature activation+3.49e-03

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.009
zed-0.009
 seeing-0.009
 tending-0.009
 inspection-0.009
Token
Token ID198
Feature activation+5.45e-03

Pos logit contributions
iley+0.007
 seeing+0.006
 inspection+0.006
 tending+0.006
joy+0.006
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.003
 off-0.003
 brothers-0.003
Token"
Token ID1
Feature activation+7.65e-03

Pos logit contributions
iley+0.009
 inspection+0.009
 seeing+0.008
joy+0.008
 tending+0.008
Neg logit contributions
ags-0.007
ac-0.007
 fart-0.007
 off-0.006
Bang-0.006
Token.
Token ID13
Feature activation+3.66e-03

Pos logit contributions
 inspection+0.023
iley+0.022
zed+0.022
anne+0.022
ya+0.021
Neg logit contributions
 fart-0.010
 off-0.010
 back-0.010
ac-0.010
ags-0.010
Token Tim
Token ID5045
Feature activation-1.97e-02

Pos logit contributions
iley+0.007
 seeing+0.007
 inspection+0.007
 tending+0.007
zed+0.007
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.003
 off-0.003
 brothers-0.003
Tokenmy
Token ID1820
Feature activation-1.79e-02

Pos logit contributions
ags+0.021
 fart+0.019
ac+0.018
 off+0.016
 ends+0.015
Neg logit contributions
iley-0.039
 seeing-0.038
zed-0.037
 inspection-0.037
ya-0.037
Token watched
Token ID7342
Feature activation+7.49e-03

Pos logit contributions
ags+0.021
 fart+0.020
ac+0.020
 off+0.019
 brothers+0.018
Neg logit contributions
iley-0.032
 inspection-0.031
joy-0.031
 seeing-0.031
 tending-0.030
Token with
Token ID351
Feature activation-9.57e-04

Pos logit contributions
iley+0.013
 inspection+0.013
joy+0.012
 seeing+0.012
 tending+0.012
Neg logit contributions
ags-0.009
 fart-0.009
 off-0.009
ac-0.008
 brothers-0.008
Token amaz
Token ID40642
Feature activation+7.07e-02

Pos logit contributions
ags+0.001
 off+0.001
 back+0.001
 brothers+0.001
 cars+0.001
Neg logit contributions
iley-0.002
zed-0.002
ya-0.002
anne-0.002
 inspection-0.002
Tokenement
Token ID972
Feature activation+1.97e-02

Pos logit contributions
iley+0.068
 seeing+0.062
 tending+0.062
zed+0.061
 inspection+0.061
Neg logit contributions
ags-0.145
ac-0.137
 fart-0.134
 off-0.132
 brothers-0.130
Token as
Token ID355
Feature activation+5.08e-03

Pos logit contributions
iley+0.037
zed+0.036
handed+0.036
ited+0.035
joy+0.035
Neg logit contributions
 off-0.022
 back-0.020
 brothers-0.019
ags-0.019
 cars-0.019
Token he
Token ID339
Feature activation-1.58e-02

Pos logit contributions
iley+0.008
zed+0.008
 inspection+0.008
ited+0.008
 tending+0.008
Neg logit contributions
 brothers-0.007
 off-0.007
 cars-0.006
ags-0.006
 both-0.006
Token saw
Token ID2497
Feature activation+2.13e-03

Pos logit contributions
ags+0.023
 fart+0.022
ac+0.022
 off+0.020
 brothers+0.020
Neg logit contributions
iley-0.024
 seeing-0.023
 inspection-0.023
zed-0.023
 tending-0.022
Token stars
Token ID5788
Feature activation-1.16e-02

Pos logit contributions
iley+0.004
zed+0.004
 inspection+0.004
 tending+0.004
joy+0.004
Neg logit contributions
ags-0.002
 off-0.002
 fart-0.002
 brothers-0.002
ac-0.002
Token made
Token ID925
Feature activation-7.88e-03

Pos logit contributions
ags+0.023
ac+0.020
 fart+0.019
 brothers+0.018
 off+0.018
Neg logit contributions
 seeing-0.024
 inspection-0.024
zed-0.024
iley-0.024
 expected-0.023
Token him
Token ID683
Feature activation-1.32e-02

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.010
 off+0.009
 brothers+0.009
Neg logit contributions
iley-0.013
 seeing-0.013
 inspection-0.013
zed-0.012
 tending-0.012
Token sh
Token ID427
Feature activation-1.12e-02

Pos logit contributions
ags+0.016
ac+0.014
 fart+0.014
Bang+0.012
 brothers+0.012
Neg logit contributions
 seeing-0.023
 inspection-0.023
iley-0.023
zed-0.023
 tending-0.023
Tokeniver
Token ID1428
Feature activation-3.53e-03

Pos logit contributions
ags+0.016
 fart+0.015
ac+0.015
 off+0.014
 ends+0.014
Neg logit contributions
iley-0.017
 seeing-0.017
zed-0.016
 inspection-0.016
 tending-0.016
Token with
Token ID351
Feature activation+5.65e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 brothers+0.003
Neg logit contributions
iley-0.006
 inspection-0.006
 seeing-0.006
zed-0.006
joy-0.006
Token amaz
Token ID40642
Feature activation+7.06e-02

Pos logit contributions
iley+0.011
 seeing+0.010
 inspection+0.010
ya+0.010
 tending+0.010
Neg logit contributions
ags-0.006
 off-0.006
 fart-0.006
ac-0.006
 are-0.005
Tokenement
Token ID972
Feature activation+3.05e-02

Pos logit contributions
iley+0.083
 seeing+0.078
 tending+0.078
aster+0.077
 inspection+0.075
Neg logit contributions
ags-0.133
ac-0.122
 off-0.120
 back-0.118
 ends-0.116
Token.
Token ID13
Feature activation-8.20e-04

Pos logit contributions
iley+0.055
aster+0.052
handed+0.051
ited+0.051
zed+0.051
Neg logit contributions
 off-0.039
 back-0.036
ags-0.035
 brothers-0.034
 both-0.033
Token He
Token ID679
Feature activation+1.74e-03

Pos logit contributions
ags+0.001
ac+0.001
 fart+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.002
 inspection-0.002
 seeing-0.002
joy-0.002
zed-0.002
Token went
Token ID1816
Feature activation-2.45e-03

Pos logit contributions
 seeing+0.002
iley+0.002
zed+0.002
 inspection+0.002
 expected+0.002
Neg logit contributions
ags-0.003
 fart-0.003
ac-0.003
ay-0.003
 brothers-0.003
Token around
Token ID1088
Feature activation+8.28e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 brothers+0.001
 ends+0.001
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
 tending-0.005
joy-0.005
Token Her
Token ID2332
Feature activation+1.27e-03

Pos logit contributions
ags+0.004
 off+0.003
 are+0.003
 ends+0.003
 fart+0.003
Neg logit contributions
iley-0.009
zed-0.008
ya-0.008
ited-0.008
 tending-0.008
Token eye
Token ID4151
Feature activation+1.93e-02

Pos logit contributions
iley+0.002
 seeing+0.002
joy+0.002
 gazing+0.002
 tending+0.002
Neg logit contributions
 fart-0.001
ags-0.001
ac-0.001
Bang-0.001
 off-0.001
Token spark
Token ID9009
Feature activation-1.28e-03

Pos logit contributions
iley+0.037
joy+0.035
 inspection+0.035
 seeing+0.035
 tending+0.034
Neg logit contributions
ags-0.020
ac-0.019
 fart-0.019
 off-0.019
 ends-0.018
Tokenled
Token ID992
Feature activation-2.64e-02

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.002
 seeing-0.002
 inspection-0.002
joy-0.002
 tending-0.002
Token with
Token ID351
Feature activation+5.74e-03

Pos logit contributions
ags+0.023
 fart+0.021
ac+0.021
 off+0.020
 back+0.018
Neg logit contributions
iley-0.056
zed-0.053
 inspection-0.053
 seeing-0.053
 tending-0.052
Token amaz
Token ID40642
Feature activation+7.06e-02

Pos logit contributions
iley+0.010
 seeing+0.010
 inspection+0.010
 tending+0.009
joy+0.009
Neg logit contributions
ags-0.007
 fart-0.007
ac-0.007
 off-0.006
 brothers-0.006
Tokenement
Token ID972
Feature activation+2.71e-02

Pos logit contributions
iley+0.078
 seeing+0.074
 tending+0.073
aster+0.072
zed+0.071
Neg logit contributions
ags-0.138
ac-0.127
 off-0.123
 ends-0.121
 back-0.120
Token.
Token ID13
Feature activation-2.97e-03

Pos logit contributions
iley+0.057
handed+0.056
aster+0.055
ited+0.054
zed+0.052
Neg logit contributions
 off-0.030
 back-0.027
 down-0.027
 both-0.024
 into-0.024
Token Everything
Token ID11391
Feature activation+2.69e-02

Pos logit contributions
ags+0.003
 off+0.002
 fart+0.002
 ends+0.002
ac+0.002
Neg logit contributions
iley-0.006
zed-0.006
 tending-0.006
 seeing-0.006
ited-0.006
Token was
Token ID373
Feature activation-1.74e-02

Pos logit contributions
iley+0.056
joy+0.053
aster+0.051
handed+0.051
 seeing+0.050
Neg logit contributions
 off-0.026
 fart-0.025
 ends-0.025
ags-0.024
 back-0.024
Token so
Token ID523
Feature activation-1.75e-02

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.017
 off+0.016
 brothers+0.015
Neg logit contributions
iley-0.034
 inspection-0.032
 seeing-0.032
zed-0.032
joy-0.032
Token gasped
Token ID45236
Feature activation+8.07e-03

Pos logit contributions
iley+0.047
aster+0.045
zed+0.044
joy+0.044
handed+0.043
Neg logit contributions
 off-0.021
 are-0.021
ac-0.020
 back-0.020
 ends-0.019
Token and
Token ID290
Feature activation-6.12e-04

Pos logit contributions
iley+0.017
zed+0.016
 inspection+0.016
aster+0.016
handed+0.016
Neg logit contributions
ags-0.007
 off-0.007
ac-0.006
 fart-0.006
 back-0.006
Token looked
Token ID3114
Feature activation+2.03e-02

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.001
 inspection-0.001
 seeing-0.001
zed-0.001
joy-0.001
Token up
Token ID510
Feature activation+1.63e-02

Pos logit contributions
iley+0.043
 seeing+0.042
 inspection+0.042
 tending+0.042
joy+0.041
Neg logit contributions
ags-0.017
 fart-0.016
ac-0.015
 off-0.014
 brothers-0.013
Token in
Token ID287
Feature activation-1.25e-02

Pos logit contributions
iley+0.034
zed+0.032
aster+0.031
handed+0.031
anne+0.031
Neg logit contributions
 off-0.015
ags-0.015
 back-0.015
 fart-0.014
ac-0.014
Token amaz
Token ID40642
Feature activation+7.06e-02

Pos logit contributions
 back+0.010
 off+0.010
 are+0.009
ags+0.009
 both+0.009
Neg logit contributions
zed-0.029
iley-0.029
handed-0.028
ya-0.027
anne-0.027
Tokenement
Token ID972
Feature activation+3.79e-02

Pos logit contributions
iley+0.085
 tending+0.083
aster+0.082
 seeing+0.081
zed+0.080
Neg logit contributions
ags-0.131
 off-0.116
ac-0.115
 back-0.111
ay-0.111
Token.
Token ID13
Feature activation-5.85e-04

Pos logit contributions
iley+0.072
aster+0.070
zed+0.070
handed+0.068
 tending+0.066
Neg logit contributions
 off-0.045
 both-0.041
 back-0.041
ags-0.038
 down-0.037
Token The
Token ID383
Feature activation+5.37e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.001
 inspection-0.001
 seeing-0.001
joy-0.001
 tending-0.001
Token flower
Token ID15061
Feature activation+8.14e-03

Pos logit contributions
iley+0.010
 inspection+0.010
zed+0.010
 seeing+0.010
joy+0.010
Neg logit contributions
ags-0.006
 fart-0.006
 ends-0.005
ac-0.005
 back-0.005
Token smiled
Token ID13541
Feature activation-9.04e-03

Pos logit contributions
iley+0.015
joy+0.014
 inspection+0.014
aster+0.014
zed+0.014
Neg logit contributions
ac-0.009
 fart-0.009
ags-0.008
 off-0.008
 ends-0.008
Token Maybe
Token ID6674
Feature activation+1.21e-02

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.004
Neg logit contributions
iley-0.019
 seeing-0.019
 inspection-0.019
 tending-0.019
zed-0.019
Token they
Token ID484
Feature activation-6.58e-03

Pos logit contributions
iley+0.021
zed+0.020
anne+0.020
joy+0.019
handed+0.019
Neg logit contributions
 off-0.015
ac-0.015
 cars-0.014
 back-0.014
 fart-0.014
Token could
Token ID714
Feature activation-1.58e-02

Pos logit contributions
ags+0.007
 fart+0.006
ac+0.006
 off+0.005
 brothers+0.005
Neg logit contributions
 seeing-0.013
iley-0.013
 tending-0.012
 inspection-0.012
zed-0.012
Token grow
Token ID1663
Feature activation+7.54e-03

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.017
 off+0.016
 brothers+0.015
Neg logit contributions
iley-0.029
 inspection-0.028
joy-0.028
zed-0.028
 seeing-0.028
Token more
Token ID517
Feature activation-2.21e-03

Pos logit contributions
 seeing+0.013
 inspection+0.012
 tending+0.012
 watching+0.011
joy+0.011
Neg logit contributions
ags-0.009
 fart-0.009
Bang-0.009
ac-0.009
ay-0.008
Token av
Token ID1196
Feature activation-9.24e-02

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.002
 brothers+0.002
Neg logit contributions
iley-0.004
 seeing-0.004
 inspection-0.004
 tending-0.004
joy-0.004
Tokenoc
Token ID420
Feature activation-3.45e-02

Pos logit contributions
ags+0.161
 fart+0.146
ac+0.143
 off+0.140
 ends+0.139
Neg logit contributions
iley-0.116
 seeing-0.111
 tending-0.111
 inspection-0.108
joy-0.105
Tokenados
Token ID22484
Feature activation-2.27e-02

Pos logit contributions
ags+0.044
 fart+0.042
ac+0.041
 off+0.038
 brothers+0.038
Neg logit contributions
iley-0.058
 seeing-0.056
 inspection-0.056
zed-0.055
 tending-0.055
Token and
Token ID290
Feature activation-6.86e-03

Pos logit contributions
ags+0.028
 fart+0.026
ac+0.026
 off+0.024
 brothers+0.024
Neg logit contributions
iley-0.040
 seeing-0.038
 inspection-0.038
 tending-0.038
zed-0.037
Token share
Token ID2648
Feature activation-5.98e-03

Pos logit contributions
ags+0.008
 fart+0.008
ac+0.008
 off+0.007
 brothers+0.007
Neg logit contributions
iley-0.012
 seeing-0.012
 inspection-0.012
zed-0.011
 tending-0.011
Token them
Token ID606
Feature activation+1.92e-02

Pos logit contributions
ags+0.009
 fart+0.008
ac+0.008
 off+0.008
 brothers+0.008
Neg logit contributions
iley-0.009
zed-0.009
 inspection-0.009
 seeing-0.009
joy-0.009
Token every
Token ID790
Feature activation-2.63e-03

Pos logit contributions
iley+0.000
 inspection+0.000
 seeing+0.000
zed+0.000
joy+0.000
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
 off-0.000
 brothers-0.000
Token day
Token ID1110
Feature activation+1.07e-02

Pos logit contributions
 back+0.003
 fart+0.003
 off+0.003
ac+0.003
 ends+0.003
Neg logit contributions
iley-0.005
 seeing-0.004
zed-0.004
 inspection-0.004
joy-0.004
Token to
Token ID284
Feature activation+5.29e-03

Pos logit contributions
iley+0.022
 inspection+0.021
zed+0.021
aster+0.021
handed+0.021
Neg logit contributions
 back-0.010
 off-0.009
 both-0.009
 brothers-0.009
ac-0.008
Token supply
Token ID5127
Feature activation-2.73e-02

Pos logit contributions
iley+0.010
joy+0.010
 inspection+0.009
anne+0.009
ya+0.009
Neg logit contributions
ags-0.006
 fart-0.005
 off-0.005
 are-0.005
ac-0.005
Token the
Token ID262
Feature activation-2.34e-02

Pos logit contributions
ags+0.032
 fart+0.031
ac+0.030
 ends+0.027
 brothers+0.027
Neg logit contributions
iley-0.047
 seeing-0.047
 tending-0.046
 inspection-0.046
joy-0.045
Token av
Token ID1196
Feature activation-9.14e-02

Pos logit contributions
ags+0.032
 fart+0.029
ac+0.029
 off+0.028
 brothers+0.027
Neg logit contributions
iley-0.038
 seeing-0.037
joy-0.036
 tending-0.036
 inspection-0.036
Tokenoc
Token ID420
Feature activation-4.44e-02

Pos logit contributions
ags+0.155
 fart+0.142
ac+0.134
 off+0.132
 brothers+0.131
Neg logit contributions
iley-0.120
 tending-0.119
 seeing-0.116
 inspection-0.115
ya-0.112
Tokenados
Token ID22484
Feature activation-2.27e-02

Pos logit contributions
ags+0.061
 fart+0.059
ac+0.057
 off+0.055
 ends+0.053
Neg logit contributions
iley-0.068
joy-0.067
 seeing-0.066
 inspection-0.066
 tending-0.066
Token.
Token ID13
Feature activation-5.85e-03

Pos logit contributions
ags+0.023
 off+0.022
ac+0.022
 fart+0.022
 back+0.022
Neg logit contributions
iley-0.045
zed-0.042
joy-0.042
 inspection-0.042
aster-0.041
Token And
Token ID843
Feature activation+9.59e-03

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.012
 seeing-0.011
 inspection-0.011
 tending-0.011
zed-0.011
Token the
Token ID262
Feature activation-3.49e-03

Pos logit contributions
iley+0.015
joy+0.015
 inspection+0.015
zed+0.014
 tending+0.014
Neg logit contributions
ac-0.013
 fart-0.012
ags-0.012
 off-0.012
 brothers-0.012
Token and
Token ID290
Feature activation-9.06e-03

Pos logit contributions
ags+0.034
ac+0.033
 fart+0.033
 brothers+0.031
 ends+0.029
Neg logit contributions
 seeing-0.043
 inspection-0.043
 tending-0.042
 gazing-0.042
iley-0.041
Token try
Token ID1949
Feature activation+9.51e-05

Pos logit contributions
ags+0.012
ac+0.012
 fart+0.011
 brothers+0.010
amed+0.010
Neg logit contributions
iley-0.015
 seeing-0.014
 inspection-0.014
 tending-0.014
zed-0.014
Token to
Token ID284
Feature activation-5.41e-03

Pos logit contributions
iley+0.000
handed+0.000
zed+0.000
ited+0.000
anne+0.000
Neg logit contributions
ags-0.000
 off-0.000
 fart-0.000
 back-0.000
ac-0.000
Token grab
Token ID5552
Feature activation+7.16e-03

Pos logit contributions
ags+0.008
ac+0.008
 fart+0.008
 brothers+0.007
 off+0.007
Neg logit contributions
iley-0.008
 seeing-0.008
 tending-0.007
 inspection-0.007
aster-0.007
Token the
Token ID262
Feature activation-9.51e-03

Pos logit contributions
iley+0.012
 inspection+0.011
 seeing+0.011
joy+0.011
 tending+0.011
Neg logit contributions
ags-0.009
 fart-0.009
ac-0.009
 brothers-0.008
 ends-0.008
Token av
Token ID1196
Feature activation-9.06e-02

Pos logit contributions
ags+0.014
ac+0.013
amed+0.012
apped+0.012
 off+0.012
Neg logit contributions
 inspection-0.014
joy-0.014
 tending-0.013
 tasting-0.013
 seeing-0.013
Tokenoc
Token ID420
Feature activation-3.14e-02

Pos logit contributions
ags+0.170
 brothers+0.150
 fart+0.146
apped+0.145
 cars+0.144
Neg logit contributions
iley-0.103
 inspection-0.097
aster-0.095
 tending-0.094
ya-0.093
Tokenados
Token ID22484
Feature activation-1.82e-02

Pos logit contributions
ags+0.044
 fart+0.043
ac+0.042
 off+0.040
Bang+0.039
Neg logit contributions
iley-0.046
joy-0.045
 inspection-0.045
 tending-0.044
 seeing-0.044
Token.
Token ID13
Feature activation-6.31e-03

Pos logit contributions
ags+0.025
 fart+0.023
ac+0.023
 brothers+0.021
 off+0.021
Neg logit contributions
iley-0.030
 seeing-0.029
 inspection-0.028
 tending-0.028
zed-0.028
Token But
Token ID887
Feature activation+2.55e-02

Pos logit contributions
ags+0.005
 fart+0.004
ac+0.004
 off+0.004
 ends+0.004
Neg logit contributions
iley-0.014
 tending-0.014
 seeing-0.013
zed-0.013
 inspection-0.013
Token the
Token ID262
Feature activation-3.88e-03

Pos logit contributions
iley+0.035
 seeing+0.034
 inspection+0.034
 tending+0.033
zed+0.033
Neg logit contributions
ags-0.042
 fart-0.039
ac-0.039
 brothers-0.036
 ends-0.036
Token Everyday
Token ID48154
Feature activation+1.46e-02

Pos logit contributions
iley+0.011
 inspection+0.011
joy+0.011
 seeing+0.011
zed+0.011
Neg logit contributions
ags-0.006
 fart-0.005
ac-0.005
 off-0.005
 brothers-0.005
Token he
Token ID339
Feature activation-2.75e-03

Pos logit contributions
iley+0.020
joy+0.020
handed+0.020
zed+0.020
ited+0.019
Neg logit contributions
 off-0.021
 brothers-0.021
 cars-0.021
 friends-0.020
 are-0.020
Token would
Token ID561
Feature activation-3.59e-03

Pos logit contributions
ags+0.003
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
zed-0.004
 tending-0.004
Token take
Token ID1011
Feature activation+9.35e-03

Pos logit contributions
ags+0.004
 fart+0.003
ac+0.003
 off+0.003
 are+0.003
Neg logit contributions
iley-0.007
joy-0.007
 inspection-0.007
 seeing-0.006
zed-0.006
Token the
Token ID262
Feature activation+2.62e-03

Pos logit contributions
iley+0.018
 inspection+0.018
 seeing+0.017
joy+0.017
 tending+0.017
Neg logit contributions
ags-0.010
 off-0.010
 fart-0.009
ac-0.009
 brothers-0.008
Token av
Token ID1196
Feature activation-8.48e-02

Pos logit contributions
iley+0.004
 seeing+0.004
 inspection+0.004
 tending+0.004
joy+0.004
Neg logit contributions
ags-0.003
ac-0.003
 fart-0.003
 off-0.003
 brothers-0.003
Tokenoc
Token ID420
Feature activation-3.53e-02

Pos logit contributions
ags+0.161
 fart+0.142
 are+0.142
 brothers+0.142
 cars+0.139
Neg logit contributions
iley-0.093
ited-0.088
 tending-0.087
ya-0.087
aster-0.086
Tokenados
Token ID22484
Feature activation-1.76e-02

Pos logit contributions
ags+0.052
 fart+0.051
ac+0.047
Bang+0.047
 off+0.047
Neg logit contributions
iley-0.052
joy-0.049
 seeing-0.049
ya-0.049
 tending-0.048
Token and
Token ID290
Feature activation-5.42e-03

Pos logit contributions
ags+0.021
 fart+0.019
ac+0.019
 brothers+0.017
Bang+0.017
Neg logit contributions
iley-0.031
 seeing-0.031
 inspection-0.030
 tending-0.030
zed-0.029
Token scatter
Token ID41058
Feature activation+1.46e-02

Pos logit contributions
ags+0.007
 fart+0.007
ac+0.007
 off+0.006
 brothers+0.006
Neg logit contributions
iley-0.009
 seeing-0.009
 tending-0.009
 inspection-0.009
zed-0.008
Token them
Token ID606
Feature activation+1.21e-02

Pos logit contributions
iley+0.025
 inspection+0.024
 seeing+0.024
 tending+0.024
joy+0.024
Neg logit contributions
ags-0.018
 fart-0.017
ac-0.017
 brothers-0.015
 ends-0.015
Token "
Token ID366
Feature activation+1.34e-02

Pos logit contributions
iley+0.012
handed+0.012
ited+0.012
aster+0.011
anne+0.011
Neg logit contributions
 back-0.004
 off-0.004
 can-0.003
 both-0.003
 brothers-0.003
TokenI
Token ID40
Feature activation+1.15e-02

Pos logit contributions
zed+0.023
iley+0.023
 seeing+0.022
 tending+0.022
 gazing+0.022
Neg logit contributions
ac-0.016
 fart-0.016
ags-0.016
Bang-0.015
 will-0.015
Token will
Token ID481
Feature activation-1.48e-02

Pos logit contributions
iley+0.022
zed+0.021
 inspection+0.021
 seeing+0.021
 tending+0.021
Neg logit contributions
ac-0.012
ags-0.012
 fart-0.011
'll-0.011
 will-0.011
Token buy
Token ID2822
Feature activation-1.04e-02

Pos logit contributions
 are+0.013
 both+0.012
 off+0.012
 will+0.012
 back+0.011
Neg logit contributions
iley-0.033
 inspection-0.033
joy-0.033
anne-0.032
ya-0.031
Token more
Token ID517
Feature activation-3.50e-03

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 off+0.012
 brothers+0.012
Neg logit contributions
iley-0.017
 seeing-0.017
 inspection-0.017
zed-0.016
 tending-0.016
Token av
Token ID1196
Feature activation-8.43e-02

Pos logit contributions
ags+0.004
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.007
 seeing-0.007
 inspection-0.006
zed-0.006
 tending-0.006
Tokenoc
Token ID420
Feature activation-2.73e-02

Pos logit contributions
ags+0.153
 fart+0.140
 off+0.134
 brothers+0.133
 ends+0.131
Neg logit contributions
iley-0.101
 tending-0.094
 inspection-0.093
aster-0.093
 seeing-0.091
Tokenados
Token ID22484
Feature activation-2.31e-02

Pos logit contributions
ags+0.037
 fart+0.037
ac+0.035
 off+0.033
 ends+0.033
Neg logit contributions
iley-0.043
 seeing-0.041
 inspection-0.041
joy-0.041
 tending-0.041
Token tomorrow
Token ID9439
Feature activation-5.64e-03

Pos logit contributions
ags+0.028
 fart+0.026
ac+0.026
 off+0.024
 brothers+0.024
Neg logit contributions
iley-0.041
 seeing-0.039
 inspection-0.039
 tending-0.038
zed-0.038
Token.
Token ID13
Feature activation-1.02e-03

Pos logit contributions
ags+0.007
 fart+0.007
ac+0.007
 off+0.006
 ends+0.006
Neg logit contributions
iley-0.010
 inspection-0.009
 seeing-0.009
joy-0.009
 tending-0.009
Token You
Token ID921
Feature activation+3.48e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.002
 seeing-0.002
 inspection-0.001
 tending-0.001
joy-0.001
Token his
Token ID465
Feature activation+3.48e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.006
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.010
 seeing-0.010
 inspection-0.010
zed-0.010
joy-0.010
Token new
Token ID649
Feature activation+1.52e-02

Pos logit contributions
iley+0.006
 seeing+0.006
 inspection+0.006
zed+0.006
 tending+0.006
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 off-0.003
 brothers-0.003
Token knowledge
Token ID3725
Feature activation+1.27e-02

Pos logit contributions
iley+0.026
 inspection+0.025
zed+0.025
 seeing+0.025
joy+0.024
Neg logit contributions
ags-0.019
ac-0.018
 fart-0.018
 brothers-0.017
 off-0.017
Token he
Token ID339
Feature activation-8.44e-03

Pos logit contributions
iley+0.025
joy+0.024
zed+0.023
 inspection+0.023
 seeing+0.023
Neg logit contributions
ags-0.013
ac-0.012
 off-0.012
 fart-0.012
 ends-0.012
Token made
Token ID925
Feature activation-1.21e-02

Pos logit contributions
ags+0.010
 fart+0.009
ac+0.009
 off+0.008
 brothers+0.008
Neg logit contributions
iley-0.015
 seeing-0.015
 inspection-0.015
zed-0.014
 tending-0.014
Token gu
Token ID915
Feature activation-8.34e-02

Pos logit contributions
ags+0.013
 off+0.013
 brothers+0.013
ac+0.013
 fart+0.012
Neg logit contributions
iley-0.022
 inspection-0.022
joy-0.021
 seeing-0.021
zed-0.021
Tokenac
Token ID330
Feature activation-5.93e-02

Pos logit contributions
 fart+0.047
ags+0.045
Bang+0.040
 cars+0.039
 off+0.039
Neg logit contributions
iley-0.201
 tending-0.176
joy-0.175
handed-0.173
ya-0.173
Tokenam
Token ID321
Feature activation-1.85e-02

Pos logit contributions
 fart+0.073
ags+0.072
Bang+0.068
'll+0.062
 are+0.060
Neg logit contributions
iley-0.098
handed-0.094
anne-0.094
zed-0.094
ya-0.093
Tokenole
Token ID2305
Feature activation-2.55e-02

Pos logit contributions
ags+0.022
 fart+0.021
ac+0.020
 off+0.019
 brothers+0.019
Neg logit contributions
iley-0.033
 seeing-0.032
 inspection-0.032
 tending-0.031
zed-0.031
Token.
Token ID13
Feature activation-1.88e-03

Pos logit contributions
 fart+0.026
ac+0.026
 off+0.026
ags+0.026
 back+0.025
Neg logit contributions
iley-0.048
zed-0.046
 seeing-0.046
joy-0.046
 inspection-0.046
Token
Token ID198
Feature activation+3.67e-03

Pos logit contributions
ags+0.002
ac+0.002
 fart+0.002
 brothers+0.002
 off+0.002
Neg logit contributions
iley-0.004
 seeing-0.003
 inspection-0.003
 tending-0.003
zed-0.003
Token weather
Token ID6193
Feature activation+5.00e-03

Pos logit contributions
ags+0.016
ac+0.015
 fart+0.015
 off+0.014
apped+0.014
Neg logit contributions
 seeing-0.018
joy-0.018
 tending-0.018
iley-0.018
 inspection-0.017
Token before
Token ID878
Feature activation+1.10e-02

Pos logit contributions
iley+0.010
 inspection+0.009
joy+0.009
zed+0.009
ya+0.009
Neg logit contributions
 off-0.005
 fart-0.005
 back-0.005
ags-0.005
ac-0.005
Token he
Token ID339
Feature activation-1.37e-02

Pos logit contributions
anne+0.019
iley+0.018
ya+0.018
ited+0.017
handed+0.017
Neg logit contributions
 off-0.016
 cars-0.015
 brothers-0.015
 fart-0.014
 both-0.014
Token picked
Token ID6497
Feature activation+5.53e-03

Pos logit contributions
ags+0.016
 fart+0.015
ac+0.015
 ends+0.014
 off+0.014
Neg logit contributions
iley-0.025
 inspection-0.024
 seeing-0.024
joy-0.023
 tending-0.023
Token his
Token ID465
Feature activation-9.21e-03

Pos logit contributions
iley+0.011
 seeing+0.011
 inspection+0.011
 tending+0.011
joy+0.011
Neg logit contributions
ags-0.005
 fart-0.005
ac-0.005
 off-0.004
 brothers-0.004
Token av
Token ID1196
Feature activation-8.25e-02

Pos logit contributions
ags+0.014
ac+0.013
Bang+0.013
 fart+0.013
 teams+0.012
Neg logit contributions
 seeing-0.013
 tending-0.013
 tasting-0.013
joy-0.013
 hearing-0.012
Tokenoc
Token ID420
Feature activation-1.97e-02

Pos logit contributions
ags+0.145
 fart+0.132
 teams+0.127
 brothers+0.126
ac+0.126
Neg logit contributions
iley-0.101
 tending-0.098
 inspection-0.096
joy-0.095
 seeing-0.095
Tokenados
Token ID22484
Feature activation-1.79e-02

Pos logit contributions
ags+0.026
 fart+0.025
ac+0.024
 off+0.023
 ends+0.023
Neg logit contributions
iley-0.032
joy-0.031
 seeing-0.031
 inspection-0.031
 tending-0.030
Token.
Token ID13
Feature activation-4.51e-03

Pos logit contributions
ags+0.026
 fart+0.025
ac+0.025
 off+0.024
 back+0.023
Neg logit contributions
iley-0.027
 inspection-0.026
zed-0.026
 seeing-0.025
joy-0.025
Token<|endoftext|>
Token ID50256
Feature activation+3.10e-03

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.004
 off+0.004
 ends+0.004
Neg logit contributions
iley-0.008
 seeing-0.008
 inspection-0.008
zed-0.008
joy-0.008
TokenOnce
Token ID7454
Feature activation+5.23e-03

Pos logit contributions
iley+0.006
 inspection+0.006
 seeing+0.006
 tending+0.006
zed+0.006
Neg logit contributions
ags-0.003
 fart-0.003
ac-0.003
 off-0.003
 brothers-0.003
Token happy
Token ID3772
Feature activation-6.16e-03

Pos logit contributions
ags+0.013
 fart+0.012
ac+0.012
 off+0.011
 ends+0.011
Neg logit contributions
iley-0.014
 seeing-0.014
 tending-0.014
 inspection-0.013
zed-0.013
Token that
Token ID326
Feature activation-1.31e-02

Pos logit contributions
 off+0.005
 back+0.005
 friends+0.005
 from+0.004
 by+0.004
Neg logit contributions
ited-0.015
aster-0.014
handed-0.014
iley-0.014
anne-0.013
Token it
Token ID340
Feature activation-1.12e-02

Pos logit contributions
 fart+0.017
ac+0.017
ags+0.017
 off+0.016
 brothers+0.016
Neg logit contributions
iley-0.022
 inspection-0.020
joy-0.020
zed-0.020
 seeing-0.020
Token produced
Token ID4635
Feature activation+1.34e-04

Pos logit contributions
ac+0.012
 fart+0.012
ags+0.011
 ends+0.011
 will+0.011
Neg logit contributions
iley-0.021
 inspection-0.020
joy-0.020
 seeing-0.020
aster-0.019
Token more
Token ID517
Feature activation+6.49e-03

Pos logit contributions
iley+0.000
 seeing+0.000
 tending+0.000
 inspection+0.000
joy+0.000
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
 off-0.000
 brothers-0.000
Token av
Token ID1196
Feature activation-8.20e-02

Pos logit contributions
iley+0.011
zed+0.011
 seeing+0.011
 inspection+0.011
joy+0.011
Neg logit contributions
ags-0.008
ac-0.007
 fart-0.007
 off-0.007
 brothers-0.007
Tokenoc
Token ID420
Feature activation-2.62e-02

Pos logit contributions
ags+0.138
 fart+0.128
ac+0.122
 off+0.119
 ends+0.118
Neg logit contributions
iley-0.107
 tending-0.105
 seeing-0.103
 inspection-0.103
joy-0.100
Tokenados
Token ID22484
Feature activation-1.36e-02

Pos logit contributions
ags+0.035
 fart+0.034
ac+0.033
 off+0.031
 ends+0.030
Neg logit contributions
iley-0.042
 seeing-0.041
 inspection-0.041
 tending-0.040
joy-0.040
Token than
Token ID621
Feature activation+3.80e-02

Pos logit contributions
 off+0.014
 back+0.014
 are+0.014
 fart+0.014
 will+0.013
Neg logit contributions
iley-0.027
zed-0.025
aster-0.025
joy-0.025
 inspection-0.025
Token ever
Token ID1683
Feature activation-9.21e-03

Pos logit contributions
iley+0.067
zed+0.062
 inspection+0.061
anne+0.060
joy+0.059
Neg logit contributions
 fart-0.047
ags-0.046
ac-0.045
 off-0.045
 brothers-0.043
Token before
Token ID878
Feature activation+1.56e-02

Pos logit contributions
 off+0.011
 back+0.011
ac+0.011
 will+0.011
 are+0.011
Neg logit contributions
zed-0.016
iley-0.016
Jo-0.015
anne-0.015
becca-0.015
Token something
Token ID1223
Feature activation+2.38e-02

Pos logit contributions
 off+0.014
ags+0.013
 back+0.012
 fart+0.012
ac+0.012
Neg logit contributions
iley-0.019
joy-0.018
 inspection-0.018
anne-0.018
ited-0.018
Token new
Token ID649
Feature activation-1.06e-02

Pos logit contributions
iley+0.052
aster+0.052
joy+0.051
 inspection+0.050
zed+0.050
Neg logit contributions
 off-0.019
 best-0.018
 back-0.018
 will-0.017
ac-0.017
Token!
Token ID0
Feature activation+3.88e-03

Pos logit contributions
 off+0.011
 back+0.011
ac+0.010
 will+0.010
ags+0.010
Neg logit contributions
aster-0.021
iley-0.021
ited-0.020
 inspection-0.020
zed-0.020
Token She
Token ID1375
Feature activation+3.20e-03

Pos logit contributions
iley+0.008
 seeing+0.007
 tending+0.007
 inspection+0.007
joy+0.007
Neg logit contributions
ags-0.004
ac-0.004
 fart-0.003
 off-0.003
 brothers-0.003
Token thought
Token ID1807
Feature activation-1.34e-04

Pos logit contributions
iley+0.005
 seeing+0.005
 inspection+0.005
zed+0.005
 tending+0.005
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 off-0.004
 brothers-0.004
Token av
Token ID1196
Feature activation-8.18e-02

Pos logit contributions
 off+0.000
ags+0.000
ac+0.000
 ends+0.000
 brothers+0.000
Neg logit contributions
iley-0.000
handed-0.000
ited-0.000
joy-0.000
anne-0.000
Tokenoc
Token ID420
Feature activation-2.63e-02

Pos logit contributions
ags+0.140
 fart+0.129
 off+0.123
 brothers+0.121
ac+0.121
Neg logit contributions
iley-0.107
 inspection-0.101
 tending-0.098
 seeing-0.098
joy-0.096
Tokenados
Token ID22484
Feature activation-8.30e-03

Pos logit contributions
ags+0.034
 fart+0.033
ac+0.032
 off+0.030
 ends+0.030
Neg logit contributions
iley-0.044
 seeing-0.042
 inspection-0.042
zed-0.041
joy-0.041
Token were
Token ID547
Feature activation-6.84e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.006
 off+0.005
 ends+0.005
Neg logit contributions
iley-0.018
joy-0.018
 seeing-0.017
 inspection-0.017
zed-0.017
Token the
Token ID262
Feature activation+1.12e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.006
 off+0.006
 brothers+0.005
Neg logit contributions
iley-0.015
 inspection-0.014
joy-0.014
zed-0.014
anne-0.014
Token most
Token ID749
Feature activation-1.82e-02

Pos logit contributions
iley+0.003
 seeing+0.003
 tending+0.003
joy+0.003
 inspection+0.003
Neg logit contributions
ags-0.001
ac-0.000
 fart-0.000
 off-0.000
 ends-0.000
Token whistle
Token ID16121
Feature activation-4.73e-03

Pos logit contributions
iley+0.005
 inspection+0.004
zed+0.004
 seeing+0.004
 tending+0.004
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 cars-0.002
 off-0.002
Token.
Token ID13
Feature activation-1.27e-03

Pos logit contributions
 off+0.006
ags+0.006
 fart+0.006
ac+0.006
 back+0.006
Neg logit contributions
iley-0.008
 inspection-0.008
joy-0.008
ya-0.007
 tending-0.007
Token It
Token ID632
Feature activation+8.12e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.002
 seeing-0.002
 inspection-0.002
joy-0.002
zed-0.002
Token made
Token ID925
Feature activation-1.16e-02

Pos logit contributions
iley+0.015
joy+0.015
 inspection+0.015
 seeing+0.014
 tending+0.014
Neg logit contributions
 fart-0.009
ac-0.008
ags-0.008
 off-0.008
 ends-0.008
Token the
Token ID262
Feature activation-7.88e-03

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 brothers+0.012
 off+0.012
Neg logit contributions
iley-0.020
 seeing-0.020
 inspection-0.020
zed-0.019
 tending-0.019
Token nic
Token ID9200
Feature activation-8.14e-02

Pos logit contributions
ags+0.009
ac+0.008
 fart+0.008
 off+0.007
 ends+0.007
Neg logit contributions
iley-0.014
 seeing-0.014
joy-0.014
 tending-0.014
 inspection-0.014
Tokenest
Token ID395
Feature activation-9.19e-03

Pos logit contributions
ags+0.090
 fart+0.084
ac+0.082
 off+0.076
 brothers+0.075
Neg logit contributions
iley-0.153
 seeing-0.147
 inspection-0.147
zed-0.145
 tending-0.145
Token sound
Token ID2128
Feature activation-3.66e-03

Pos logit contributions
ags+0.008
 fart+0.008
 brothers+0.008
 off+0.008
 cars+0.008
Neg logit contributions
iley-0.019
zed-0.018
 inspection-0.018
 tending-0.018
handed-0.018
Token,
Token ID11
Feature activation+3.16e-03

Pos logit contributions
 off+0.003
 are+0.003
 back+0.003
 ends+0.003
ac+0.003
Neg logit contributions
iley-0.008
joy-0.008
handed-0.008
 inspection-0.008
ya-0.007
Token like
Token ID588
Feature activation+1.78e-02

Pos logit contributions
iley+0.006
 inspection+0.006
zed+0.006
 seeing+0.005
 tending+0.005
Neg logit contributions
ags-0.004
 fart-0.003
ac-0.003
 off-0.003
 brothers-0.003
Token a
Token ID257
Feature activation-5.08e-03

Pos logit contributions
iley+0.034
ited+0.032
 inspection+0.032
zed+0.032
anne+0.032
Neg logit contributions
 off-0.018
 cars-0.018
 brothers-0.018
 fart-0.018
ags-0.017
Token.
Token ID13
Feature activation-5.00e-03

Pos logit contributions
ags+0.012
 fart+0.012
ac+0.011
 off+0.011
 brothers+0.011
Neg logit contributions
iley-0.017
 inspection-0.016
 seeing-0.016
 tending-0.016
joy-0.016
Token So
Token ID1406
Feature activation-9.34e-03

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.004
Neg logit contributions
iley-0.009
 seeing-0.009
 inspection-0.009
joy-0.009
zed-0.009
Token Bill
Token ID3941
Feature activation+9.61e-03

Pos logit contributions
 off+0.013
ags+0.013
 fart+0.013
ac+0.013
 brothers+0.012
Neg logit contributions
iley-0.015
 tending-0.013
zed-0.013
joy-0.013
 inspection-0.013
Token picked
Token ID6497
Feature activation-3.62e-03

Pos logit contributions
iley+0.016
 seeing+0.016
 inspection+0.015
zed+0.015
 tending+0.015
Neg logit contributions
ags-0.013
 fart-0.012
ac-0.011
 off-0.011
 brothers-0.010
Token the
Token ID262
Feature activation-8.77e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 ends+0.003
Neg logit contributions
iley-0.007
 seeing-0.006
 inspection-0.006
 tending-0.006
zed-0.006
Token nic
Token ID9200
Feature activation-8.05e-02

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.010
 off+0.009
 brothers+0.009
Neg logit contributions
iley-0.016
 seeing-0.015
 inspection-0.015
zed-0.015
 tending-0.015
Tokenest
Token ID395
Feature activation-4.16e-03

Pos logit contributions
ags+0.075
ac+0.075
 ends+0.064
 off+0.063
 fart+0.062
Neg logit contributions
iley-0.169
 seeing-0.167
 tending-0.163
joy-0.162
 gazing-0.160
Token one
Token ID530
Feature activation-3.21e-02

Pos logit contributions
 off+0.004
 back+0.004
ac+0.004
 cars+0.003
ags+0.003
Neg logit contributions
iley-0.009
anne-0.008
zed-0.008
 inspection-0.008
 tending-0.008
Token he
Token ID339
Feature activation-1.30e-02

Pos logit contributions
 back+0.029
ac+0.029
 will+0.027
 ends+0.027
 off+0.026
Neg logit contributions
iley-0.067
joy-0.064
zed-0.064
 seeing-0.064
 inspection-0.064
Token could
Token ID714
Feature activation-1.21e-02

Pos logit contributions
ac+0.014
ags+0.014
 fart+0.013
 brothers+0.013
 will+0.013
Neg logit contributions
iley-0.024
 inspection-0.023
joy-0.023
 seeing-0.023
 tending-0.022
Token find
Token ID1064
Feature activation-7.27e-03

Pos logit contributions
 off+0.012
ags+0.012
 are+0.012
 both+0.012
 back+0.012
Neg logit contributions
iley-0.024
joy-0.023
 inspection-0.023
ya-0.023
anne-0.022
Token something
Token ID1223
Feature activation+2.29e-02

Pos logit contributions
 off+0.013
ags+0.012
 back+0.012
 fart+0.011
ac+0.011
Neg logit contributions
iley-0.018
joy-0.017
 inspection-0.017
 tending-0.017
anne-0.017
Token new
Token ID649
Feature activation-1.12e-02

Pos logit contributions
iley+0.050
aster+0.050
joy+0.049
 inspection+0.048
zed+0.048
Neg logit contributions
 off-0.018
 best-0.017
 back-0.017
 will-0.016
ac-0.016
Token!
Token ID0
Feature activation+4.25e-03

Pos logit contributions
 off+0.011
 back+0.011
 will+0.011
ac+0.011
ags+0.011
Neg logit contributions
aster-0.022
iley-0.022
ited-0.021
 inspection-0.021
zed-0.020
Token She
Token ID1375
Feature activation+4.30e-03

Pos logit contributions
iley+0.008
 seeing+0.008
 tending+0.008
 inspection+0.008
zed+0.008
Neg logit contributions
ags-0.004
ac-0.004
 fart-0.004
 off-0.004
 brothers-0.004
Token thought
Token ID1807
Feature activation+1.00e-03

Pos logit contributions
iley+0.007
 seeing+0.007
 inspection+0.007
zed+0.007
 tending+0.007
Neg logit contributions
ags-0.006
 fart-0.006
ac-0.005
 off-0.005
 brothers-0.005
Token av
Token ID1196
Feature activation-8.03e-02

Pos logit contributions
iley+0.002
handed+0.001
joy+0.001
ited+0.001
 tending+0.001
Neg logit contributions
 off-0.001
ags-0.001
ac-0.001
 ends-0.001
 brothers-0.001
Tokenoc
Token ID420
Feature activation-2.65e-02

Pos logit contributions
ags+0.138
 fart+0.127
 off+0.122
 brothers+0.120
ac+0.119
Neg logit contributions
iley-0.105
 inspection-0.098
 tending-0.096
 seeing-0.095
joy-0.093
Tokenados
Token ID22484
Feature activation-7.08e-03

Pos logit contributions
ags+0.035
 fart+0.034
ac+0.033
 off+0.031
 ends+0.030
Neg logit contributions
iley-0.044
 seeing-0.042
 inspection-0.042
joy-0.041
zed-0.041
Token were
Token ID547
Feature activation-4.70e-03

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 off+0.005
 ends+0.004
Neg logit contributions
iley-0.015
joy-0.015
 seeing-0.015
 inspection-0.015
zed-0.015
Token the
Token ID262
Feature activation+1.57e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.010
 inspection-0.010
joy-0.010
zed-0.009
aster-0.009
Token most
Token ID749
Feature activation-1.69e-02

Pos logit contributions
iley+0.004
 seeing+0.004
 inspection+0.004
 tending+0.004
joy+0.004
Neg logit contributions
ags-0.001
 fart-0.001
ac-0.001
 off-0.000
 brothers-0.000
Token He
Token ID679
Feature activation-3.33e-03

Pos logit contributions
ags+0.012
 fart+0.011
ac+0.011
 off+0.010
 ends+0.010
Neg logit contributions
iley-0.019
 inspection-0.019
 seeing-0.019
joy-0.019
zed-0.019
Token quickly
Token ID2952
Feature activation-7.40e-03

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.004
Neg logit contributions
iley-0.005
 seeing-0.005
zed-0.004
 inspection-0.004
 tending-0.004
Token hung
Token ID9174
Feature activation-7.00e-03

Pos logit contributions
 off+0.008
 back+0.007
 ends+0.007
ac+0.007
 fart+0.007
Neg logit contributions
iley-0.014
joy-0.013
 tending-0.013
 inspection-0.013
handed-0.013
Token colorful
Token ID20239
Feature activation-2.13e-02

Pos logit contributions
ags+0.009
 fart+0.009
ac+0.009
 brothers+0.008
 ends+0.008
Neg logit contributions
iley-0.011
 seeing-0.011
 inspection-0.011
joy-0.011
 tending-0.011
Token or
Token ID393
Feature activation+1.79e-03

Pos logit contributions
ags+0.023
 fart+0.022
ac+0.021
 off+0.020
 ends+0.019
Neg logit contributions
iley-0.040
 seeing-0.039
 inspection-0.039
zed-0.038
joy-0.038
Tokennam
Token ID7402
Feature activation-7.97e-02

Pos logit contributions
 seeing+0.003
iley+0.003
 tending+0.003
 inspection+0.003
 gazing+0.003
Neg logit contributions
ags-0.003
ac-0.002
 fart-0.002
 brothers-0.002
ay-0.002
Tokenents
Token ID658
Feature activation-2.29e-02

Pos logit contributions
ags+0.131
 fart+0.116
 will+0.114
 can+0.110
Bang+0.110
Neg logit contributions
iley-0.108
 tending-0.100
Kat-0.099
 seeing-0.099
handed-0.098
Token and
Token ID290
Feature activation+3.23e-03

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.016
 off+0.016
 ends+0.015
Neg logit contributions
iley-0.051
joy-0.048
 seeing-0.048
 inspection-0.048
zed-0.048
Token tw
Token ID665
Feature activation+1.21e-02

Pos logit contributions
iley+0.005
 inspection+0.005
 seeing+0.005
joy+0.005
zed+0.005
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 off-0.004
 brothers-0.004
Tokenink
Token ID676
Feature activation-1.29e-02

Pos logit contributions
iley+0.022
 seeing+0.022
joy+0.022
 tending+0.021
 hesitation+0.021
Neg logit contributions
ags-0.015
ac-0.013
 off-0.012
am-0.012
 ends-0.011
Tokenling
Token ID1359
Feature activation+8.32e-03

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 off+0.012
 ends+0.012
Neg logit contributions
iley-0.025
 seeing-0.024
 inspection-0.024
joy-0.023
 tending-0.023
Token.
Token ID13
Feature activation-5.14e-04

Pos logit contributions
 will+0.016
 brothers+0.016
ac+0.016
 off+0.016
ags+0.016
Neg logit contributions
 inspection-0.021
iley-0.021
zed-0.020
 tending-0.020
joy-0.019
Token Did
Token ID7731
Feature activation-6.11e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.001
 seeing-0.001
 inspection-0.001
zed-0.001
 tending-0.001
Token you
Token ID345
Feature activation-1.66e-02

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.005
 off+0.005
 brothers+0.005
Neg logit contributions
joy-0.012
iley-0.012
 tending-0.012
 inspection-0.012
zed-0.012
Token know
Token ID760
Feature activation-8.87e-03

Pos logit contributions
ags+0.023
ac+0.021
 fart+0.021
 brothers+0.019
 ends+0.019
Neg logit contributions
 seeing-0.027
iley-0.026
aster-0.025
 tending-0.025
 inspection-0.025
Token that
Token ID326
Feature activation-1.96e-02

Pos logit contributions
ags+0.016
 fart+0.016
ac+0.015
 off+0.015
 brothers+0.015
Neg logit contributions
iley-0.010
 seeing-0.010
 inspection-0.010
zed-0.009
 tending-0.009
Token av
Token ID1196
Feature activation-7.95e-02

Pos logit contributions
ags+0.028
ac+0.026
 fart+0.025
 off+0.023
 brothers+0.023
Neg logit contributions
iley-0.031
 seeing-0.030
 inspection-0.030
 tending-0.030
zed-0.029
Tokenoc
Token ID420
Feature activation-2.89e-02

Pos logit contributions
ags+0.142
 fart+0.127
 off+0.126
 teams+0.123
 brothers+0.123
Neg logit contributions
iley-0.095
 seeing-0.091
 tending-0.090
 inspection-0.089
ited-0.088
Tokenados
Token ID22484
Feature activation-1.29e-02

Pos logit contributions
ags+0.041
 fart+0.041
ac+0.038
 off+0.037
 ends+0.037
Neg logit contributions
iley-0.043
 seeing-0.042
joy-0.041
 tending-0.041
 inspection-0.041
Token are
Token ID389
Feature activation-1.26e-02

Pos logit contributions
ags+0.011
 fart+0.010
ac+0.009
 off+0.008
 brothers+0.008
Neg logit contributions
iley-0.027
 seeing-0.027
 inspection-0.027
 tending-0.026
zed-0.026
Token very
Token ID845
Feature activation-1.46e-02

Pos logit contributions
ags+0.016
 fart+0.015
ac+0.015
 off+0.014
 ends+0.014
Neg logit contributions
iley-0.021
 seeing-0.021
 tending-0.020
 inspection-0.020
zed-0.020
Token healthy
Token ID5448
Feature activation-1.60e-02

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.017
 off+0.016
 brothers+0.016
Neg logit contributions
iley-0.025
 seeing-0.024
 inspection-0.024
joy-0.024
zed-0.024
Tokenksh
Token ID50133
Feature activation-6.17e-02

Pos logit contributions
iley+0.013
joy+0.013
 tending+0.013
 seeing+0.013
ited+0.012
Neg logit contributions
ags-0.010
 fart-0.009
 brothers-0.009
ac-0.009
 ends-0.009
Tokenakes
Token ID1124
Feature activation+7.50e-03

Pos logit contributions
ags+0.085
 fart+0.084
 off+0.079
ac+0.078
 back+0.076
Neg logit contributions
iley-0.093
zed-0.089
anne-0.088
handed-0.088
aster-0.086
Token.
Token ID13
Feature activation-4.54e-03

Pos logit contributions
iley+0.014
joy+0.013
 inspection+0.013
zed+0.013
 seeing+0.012
Neg logit contributions
 off-0.009
ags-0.009
 fart-0.008
ac-0.008
 back-0.008
Token The
Token ID383
Feature activation-8.83e-04

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.009
 seeing-0.009
 inspection-0.009
zed-0.009
 tending-0.009
Token mil
Token ID1465
Feature activation-1.48e-02

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.002
 seeing-0.001
 inspection-0.001
zed-0.001
joy-0.001
Tokenksh
Token ID50133
Feature activation-7.94e-02

Pos logit contributions
ags+0.016
 fart+0.015
ac+0.015
 brothers+0.014
 ends+0.014
Neg logit contributions
iley-0.026
 tending-0.025
anne-0.025
zed-0.025
 seeing-0.025
Tokenakes
Token ID1124
Feature activation-1.21e-02

Pos logit contributions
ags+0.111
 fart+0.108
 off+0.103
ac+0.101
 back+0.098
Neg logit contributions
iley-0.121
zed-0.119
anne-0.113
aster-0.113
 tending-0.113
Token were
Token ID547
Feature activation+2.43e-03

Pos logit contributions
 off+0.013
 fart+0.013
ags+0.013
 are+0.013
ac+0.013
Neg logit contributions
iley-0.023
joy-0.022
 inspection-0.021
 seeing-0.021
ya-0.020
Token blue
Token ID4171
Feature activation+4.65e-03

Pos logit contributions
iley+0.005
 inspection+0.005
joy+0.005
zed+0.004
 seeing+0.004
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 off-0.002
 brothers-0.002
Token and
Token ID290
Feature activation+1.09e-02

Pos logit contributions
iley+0.008
 inspection+0.008
 seeing+0.008
zed+0.008
joy+0.008
Neg logit contributions
ags-0.005
 fart-0.005
ac-0.005
 off-0.005
 brothers-0.005
Token pink
Token ID11398
Feature activation+1.34e-02

Pos logit contributions
iley+0.019
joy+0.018
 inspection+0.018
 seeing+0.018
zed+0.018
Neg logit contributions
 fart-0.013
ags-0.013
ac-0.012
 off-0.012
 brothers-0.012
Token the
Token ID262
Feature activation-1.51e-02

Pos logit contributions
ags+0.000
 fart+0.000
ac+0.000
 off+0.000
 brothers+0.000
Neg logit contributions
iley-0.001
 inspection-0.001
 seeing-0.001
zed-0.001
 tending-0.001
Token store
Token ID3650
Feature activation-2.76e-03

Pos logit contributions
ags+0.023
 fart+0.023
ac+0.022
 off+0.021
 brothers+0.021
Neg logit contributions
iley-0.021
 inspection-0.020
 seeing-0.020
zed-0.020
joy-0.020
Token to
Token ID284
Feature activation-1.14e-03

Pos logit contributions
ags+0.002
ac+0.002
 off+0.002
 back+0.002
 fart+0.002
Neg logit contributions
iley-0.006
zed-0.006
 inspection-0.006
aster-0.006
 tending-0.006
Token get
Token ID651
Feature activation+6.08e-03

Pos logit contributions
ags+0.001
 fart+0.001
 off+0.001
ac+0.001
 back+0.001
Neg logit contributions
iley-0.002
 inspection-0.002
joy-0.002
zed-0.002
anne-0.002
Token more
Token ID517
Feature activation+4.51e-03

Pos logit contributions
iley+0.011
 seeing+0.011
 tending+0.010
joy+0.010
 inspection+0.010
Neg logit contributions
ags-0.007
 fart-0.007
ac-0.007
 ends-0.006
 brothers-0.006
Token av
Token ID1196
Feature activation-7.91e-02

Pos logit contributions
iley+0.008
zed+0.007
 seeing+0.007
 inspection+0.007
joy+0.007
Neg logit contributions
ags-0.006
 fart-0.006
ac-0.005
 off-0.005
 brothers-0.005
Tokenoc
Token ID420
Feature activation-2.74e-02

Pos logit contributions
ags+0.134
 fart+0.122
ac+0.119
 off+0.114
 ends+0.113
Neg logit contributions
iley-0.105
 tending-0.100
 seeing-0.098
 inspection-0.098
joy-0.096
Tokenados
Token ID22484
Feature activation-1.74e-02

Pos logit contributions
ags+0.036
 fart+0.034
ac+0.034
 off+0.031
 ends+0.031
Neg logit contributions
iley-0.045
 seeing-0.043
 inspection-0.043
zed-0.043
 tending-0.043
Token.
Token ID13
Feature activation-8.03e-03

Pos logit contributions
ags+0.020
 fart+0.020
 off+0.019
ac+0.019
 back+0.018
Neg logit contributions
iley-0.032
zed-0.031
 inspection-0.030
joy-0.030
 seeing-0.030
Token Gary
Token ID10936
Feature activation+2.05e-02

Pos logit contributions
ac+0.011
ags+0.011
 fart+0.011
eps+0.010
 ends+0.010
Neg logit contributions
iley-0.012
zed-0.012
 tending-0.012
 inspection-0.012
joy-0.012
Token looked
Token ID3114
Feature activation+4.33e-03

Pos logit contributions
iley+0.037
 inspection+0.036
 seeing+0.035
joy+0.035
 tending+0.035
Neg logit contributions
ags-0.023
 fart-0.022
ac-0.022
 off-0.020
 brothers-0.020
Token if
Token ID611
Feature activation+2.94e-03

Pos logit contributions
ags+0.007
ac+0.007
 fart+0.007
 off+0.006
 brothers+0.006
Neg logit contributions
iley-0.013
zed-0.013
 inspection-0.013
 seeing-0.013
joy-0.013
Token she
Token ID673
Feature activation-4.43e-03

Pos logit contributions
iley+0.004
anne+0.004
ited+0.003
 inspection+0.003
joy+0.003
Neg logit contributions
 will-0.005
 off-0.005
 are-0.005
 back-0.005
 brothers-0.005
Token could
Token ID714
Feature activation-9.87e-03

Pos logit contributions
ags+0.003
ac+0.003
 fart+0.003
 brothers+0.003
 ends+0.003
Neg logit contributions
iley-0.010
 seeing-0.009
 inspection-0.009
 tending-0.009
zed-0.009
Token request
Token ID2581
Feature activation-7.68e-03

Pos logit contributions
ags+0.013
 fart+0.012
ac+0.012
 off+0.011
 brothers+0.011
Neg logit contributions
iley-0.017
 seeing-0.016
 inspection-0.016
joy-0.016
zed-0.016
Token some
Token ID617
Feature activation-1.87e-02

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.010
 ends+0.009
 brothers+0.009
Neg logit contributions
iley-0.012
 seeing-0.012
 inspection-0.012
 tending-0.012
joy-0.012
Token av
Token ID1196
Feature activation-7.88e-02

Pos logit contributions
ags+0.023
 fart+0.021
ac+0.021
 off+0.019
 brothers+0.019
Neg logit contributions
iley-0.033
 seeing-0.032
 inspection-0.032
 tending-0.031
joy-0.031
Tokenoc
Token ID420
Feature activation-2.30e-02

Pos logit contributions
ags+0.136
 fart+0.126
 off+0.118
 ends+0.118
ac+0.116
Neg logit contributions
iley-0.101
 tending-0.097
 inspection-0.094
 seeing-0.093
joy-0.091
Tokenados
Token ID22484
Feature activation-1.23e-02

Pos logit contributions
ags+0.032
 fart+0.032
ac+0.030
 off+0.029
 ends+0.029
Neg logit contributions
iley-0.035
 seeing-0.034
 tending-0.034
 inspection-0.034
joy-0.033
Token from
Token ID422
Feature activation-2.31e-02

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 off+0.013
 brothers+0.012
Neg logit contributions
iley-0.023
zed-0.022
 inspection-0.021
 seeing-0.021
joy-0.021
Token the
Token ID262
Feature activation+1.00e-02

Pos logit contributions
ags+0.024
 both+0.022
 off+0.021
 back+0.021
 cars+0.021
Neg logit contributions
 inspection-0.047
anne-0.047
zed-0.045
iley-0.044
joy-0.043
Token store
Token ID3650
Feature activation+5.45e-03

Pos logit contributions
iley+0.015
zed+0.015
 inspection+0.015
anne+0.014
 seeing+0.014
Neg logit contributions
ags-0.015
 fart-0.015
ac-0.014
 back-0.014
 brothers-0.014
Token they
Token ID484
Feature activation-1.18e-03

Pos logit contributions
ited+0.040
iley+0.039
handed+0.037
ya+0.036
anne+0.035
Neg logit contributions
 cars-0.031
 brothers-0.030
 back-0.029
 off-0.029
 both-0.029
Token had
Token ID550
Feature activation-9.78e-03

Pos logit contributions
ags+0.002
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.002
 seeing-0.002
 inspection-0.002
joy-0.002
 tending-0.002
Token never
Token ID1239
Feature activation+1.01e-03

Pos logit contributions
ags+0.012
 fart+0.012
 off+0.012
 back+0.012
ac+0.011
Neg logit contributions
iley-0.017
 inspection-0.016
joy-0.016
ya-0.016
anne-0.016
Token seen
Token ID1775
Feature activation+9.69e-03

Pos logit contributions
iley+0.002
 inspection+0.002
ya+0.002
zed+0.002
anne+0.001
Neg logit contributions
 fart-0.001
ags-0.001
ac-0.001
 ends-0.001
 off-0.001
Token the
Token ID262
Feature activation-2.63e-03

Pos logit contributions
iley+0.019
zed+0.018
 inspection+0.018
anne+0.018
joy+0.017
Neg logit contributions
ags-0.010
 off-0.010
 fart-0.010
 brothers-0.009
 back-0.009
Token av
Token ID1196
Feature activation-7.86e-02

Pos logit contributions
 fart+0.004
ags+0.003
 back+0.003
 brothers+0.003
 cars+0.003
Neg logit contributions
iley-0.004
zed-0.004
 inspection-0.004
anne-0.004
 seeing-0.004
Tokenoc
Token ID420
Feature activation-1.91e-02

Pos logit contributions
ags+0.136
ac+0.122
 fart+0.122
 off+0.119
 brothers+0.115
Neg logit contributions
iley-0.100
 tending-0.098
 seeing-0.096
joy-0.095
 inspection-0.094
Tokenados
Token ID22484
Feature activation-1.59e-02

Pos logit contributions
ags+0.024
 fart+0.022
ac+0.022
 off+0.021
 brothers+0.020
Neg logit contributions
iley-0.033
 seeing-0.031
 inspection-0.031
zed-0.031
 tending-0.031
Token.
Token ID13
Feature activation+5.11e-03

Pos logit contributions
 fart+0.020
 back+0.019
 off+0.019
ags+0.019
 are+0.018
Neg logit contributions
iley-0.028
 inspection-0.027
anne-0.026
joy-0.026
ya-0.025
Token They
Token ID1119
Feature activation+1.35e-02

Pos logit contributions
iley+0.013
 tending+0.012
 seeing+0.012
zed+0.012
 inspection+0.012
Neg logit contributions
ags-0.003
 fart-0.002
ac-0.002
 off-0.002
 brothers-0.002
Token wished
Token ID16555
Feature activation+1.24e-02

Pos logit contributions
iley+0.011
 seeing+0.011
 inspection+0.010
zed+0.010
 expected+0.010
Neg logit contributions
ags-0.030
ac-0.028
 fart-0.028
ay-0.026
 off-0.026
Token It
Token ID632
Feature activation+1.06e-02

Pos logit contributions
iley+0.008
 tending+0.008
 seeing+0.008
 inspection+0.008
zed+0.008
Neg logit contributions
ags-0.003
 fart-0.003
 brothers-0.003
ac-0.003
 off-0.003
Token turned
Token ID2900
Feature activation+1.22e-02

Pos logit contributions
zed+0.016
 seeing+0.015
iley+0.015
 inspection+0.015
 tending+0.015
Neg logit contributions
ags-0.017
ac-0.015
 fart-0.015
 teams-0.013
ay-0.013
Token out
Token ID503
Feature activation-2.45e-02

Pos logit contributions
 inspection+0.023
 leaving+0.022
 tending+0.022
joy+0.022
 seeing+0.022
Neg logit contributions
ags-0.013
ac-0.013
 fart-0.011
rr-0.011
 brothers-0.010
Token that
Token ID326
Feature activation-7.44e-03

Pos logit contributions
ags+0.029
 fart+0.027
ac+0.026
Bang+0.023
 brothers+0.023
Neg logit contributions
 inspection-0.044
 seeing-0.044
iley-0.044
joy-0.043
 tending-0.042
Token the
Token ID262
Feature activation-1.82e-03

Pos logit contributions
ags+0.012
ac+0.011
 fart+0.010
 brothers+0.010
 off+0.009
Neg logit contributions
iley-0.011
 seeing-0.011
 inspection-0.011
 tending-0.010
zed-0.010
Token av
Token ID1196
Feature activation-7.83e-02

Pos logit contributions
ags+0.002
ac+0.002
 fart+0.002
 off+0.002
 ends+0.002
Neg logit contributions
iley-0.003
 seeing-0.003
joy-0.003
 tending-0.003
 inspection-0.003
Tokenoc
Token ID420
Feature activation-3.10e-02

Pos logit contributions
ags+0.154
 fart+0.135
 brothers+0.134
 teams+0.133
ac+0.132
Neg logit contributions
iley-0.084
ited-0.081
 inspection-0.080
aster-0.079
 tending-0.078
Tokenados
Token ID22484
Feature activation-1.16e-02

Pos logit contributions
ags+0.044
 fart+0.043
ac+0.041
 off+0.039
 ends+0.039
Neg logit contributions
iley-0.047
joy-0.046
 inspection-0.045
 seeing-0.045
 tending-0.045
Token were
Token ID547
Feature activation-4.93e-04

Pos logit contributions
ags+0.013
 fart+0.012
ac+0.011
 off+0.010
 brothers+0.010
Neg logit contributions
iley-0.022
 seeing-0.021
 inspection-0.021
zed-0.021
 tending-0.021
Token bad
Token ID2089
Feature activation+6.78e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.001
 inspection-0.001
 seeing-0.001
zed-0.001
joy-0.001
Token and
Token ID290
Feature activation+4.73e-03

Pos logit contributions
iley+0.014
aster+0.014
zed+0.013
handed+0.013
 inspection+0.013
Neg logit contributions
 off-0.006
 brothers-0.006
ags-0.006
 fart-0.005
 back-0.005
Token store
Token ID3650
Feature activation-1.66e-04

Pos logit contributions
 fart+0.016
ags+0.016
 off+0.015
 back+0.015
ac+0.015
Neg logit contributions
iley-0.019
 inspection-0.018
zed-0.018
joy-0.018
anne-0.017
Token and
Token ID290
Feature activation-1.21e-02

Pos logit contributions
 back+0.000
 off+0.000
ags+0.000
 cars+0.000
 both+0.000
Neg logit contributions
iley-0.000
zed-0.000
 inspection-0.000
anne-0.000
 hesitation-0.000
Token forgot
Token ID16453
Feature activation+2.30e-02

Pos logit contributions
ags+0.016
 fart+0.016
 off+0.016
ac+0.016
 back+0.015
Neg logit contributions
iley-0.020
joy-0.018
 inspection-0.018
zed-0.018
 seeing-0.017
Token to
Token ID284
Feature activation+2.76e-02

Pos logit contributions
iley+0.047
joy+0.044
zed+0.044
 inspection+0.043
anne+0.043
Neg logit contributions
 off-0.023
ags-0.020
 back-0.020
 are-0.019
 both-0.019
Token buy
Token ID2822
Feature activation+1.34e-03

Pos logit contributions
joy+0.057
iley+0.057
 inspection+0.056
anne+0.053
zed+0.052
Neg logit contributions
ags-0.027
 off-0.025
 are-0.024
ac-0.024
 both-0.023
Token av
Token ID1196
Feature activation-7.77e-02

Pos logit contributions
iley+0.003
zed+0.003
 inspection+0.002
 seeing+0.002
anne+0.002
Neg logit contributions
ags-0.001
 off-0.001
 fart-0.001
ac-0.001
 back-0.001
Tokenoc
Token ID420
Feature activation-1.78e-02

Pos logit contributions
ags+0.134
 fart+0.124
 ends+0.116
ac+0.116
 off+0.115
Neg logit contributions
iley-0.098
 tending-0.096
 seeing-0.095
 inspection-0.093
aster-0.090
Tokenados
Token ID22484
Feature activation-1.28e-02

Pos logit contributions
ags+0.025
 fart+0.025
ac+0.024
 off+0.022
 ends+0.022
Neg logit contributions
iley-0.027
 seeing-0.026
 tending-0.026
 inspection-0.026
joy-0.026
Token.
Token ID13
Feature activation-1.31e-03

Pos logit contributions
ags+0.015
 fart+0.014
ac+0.014
 off+0.014
 back+0.013
Neg logit contributions
iley-0.023
zed-0.023
 inspection-0.022
joy-0.022
 seeing-0.022
Token Lily
Token ID20037
Feature activation-2.19e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.002
 seeing-0.002
 inspection-0.002
zed-0.002
 tending-0.002
Token was
Token ID373
Feature activation+4.20e-03

Pos logit contributions
ags+0.003
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.004
 seeing-0.003
 inspection-0.003
 tending-0.003
zed-0.003
Token
Token ID220
Feature activation-5.23e-03

Pos logit contributions
iley+0.001
 seeing+0.001
 inspection+0.001
 tending+0.001
zed+0.001
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
 brothers-0.000
 off-0.000
Token
Token ID198
Feature activation+6.50e-03

Pos logit contributions
ags+0.007
 fart+0.007
ac+0.006
 off+0.006
 ends+0.006
Neg logit contributions
iley-0.009
 seeing-0.009
 inspection-0.008
zed-0.008
joy-0.008
TokenHe
Token ID1544
Feature activation-4.27e-03

Pos logit contributions
iley+0.012
 seeing+0.011
 inspection+0.011
zed+0.011
 tending+0.011
Neg logit contributions
ags-0.008
 fart-0.007
ac-0.007
 off-0.007
 ends-0.007
Token couldn
Token ID3521
Feature activation-2.18e-02

Pos logit contributions
ags+0.007
 fart+0.006
ac+0.006
 ends+0.005
 brothers+0.005
Neg logit contributions
iley-0.006
zed-0.006
 seeing-0.006
 tending-0.006
ya-0.006
Token't
Token ID470
Feature activation+7.04e-03

Pos logit contributions
ags+0.036
ac+0.035
 fart+0.034
 brothers+0.031
 teams+0.030
Neg logit contributions
zed-0.029
iley-0.029
aster-0.028
 seeing-0.028
 tending-0.027
Token believe
Token ID1975
Feature activation+4.11e-02

Pos logit contributions
joy+0.012
iley+0.011
 inspection+0.011
zed+0.010
 seeing+0.010
Neg logit contributions
ags-0.009
 fart-0.009
 are-0.009
 both-0.009
 back-0.009
Token it
Token ID340
Feature activation-9.18e-03

Pos logit contributions
ited+0.098
handed+0.093
anne+0.092
iley+0.088
looking+0.088
Neg logit contributions
 off-0.041
 both-0.033
 back-0.033
 our-0.031
 down-0.031
Token was
Token ID373
Feature activation-6.60e-03

Pos logit contributions
 off+0.012
 back+0.011
 both+0.011
 fart+0.011
 brothers+0.011
Neg logit contributions
iley-0.015
joy-0.015
aster-0.015
 inspection-0.015
handed-0.014
Token all
Token ID477
Feature activation-1.78e-02

Pos logit contributions
 off+0.007
 back+0.006
 both+0.006
ags+0.005
 fart+0.005
Neg logit contributions
iley-0.015
 inspection-0.013
ya-0.013
aster-0.013
handed-0.013
Token for
Token ID329
Feature activation-4.62e-03

Pos logit contributions
 off+0.016
 back+0.015
ags+0.014
 fart+0.014
 brothers+0.013
Neg logit contributions
iley-0.040
zed-0.037
 inspection-0.037
iya-0.037
aster-0.036
Token him
Token ID683
Feature activation+1.62e-02

Pos logit contributions
 brothers+0.004
 friends+0.004
 both+0.004
 cars+0.003
 off+0.003
Neg logit contributions
anne-0.011
ited-0.010
looking-0.010
iley-0.010
 inspection-0.010
Token was
Token ID373
Feature activation+1.23e-02

Pos logit contributions
iley+0.004
 inspection+0.004
 seeing+0.004
joy+0.004
zed+0.004
Neg logit contributions
ags-0.003
 fart-0.003
ac-0.003
 off-0.003
 brothers-0.003
Token very
Token ID845
Feature activation+2.88e-03

Pos logit contributions
iley+0.024
 inspection+0.024
joy+0.023
zed+0.022
ya+0.022
Neg logit contributions
ags-0.013
 fart-0.012
 off-0.012
ac-0.012
 back-0.011
Token frustrated
Token ID14718
Feature activation+1.09e-02

Pos logit contributions
iley+0.005
 seeing+0.005
 tending+0.005
zed+0.005
 inspection+0.005
Neg logit contributions
ags-0.004
 fart-0.004
ac-0.004
 ends-0.003
 off-0.003
Token with
Token ID351
Feature activation+2.59e-02

Pos logit contributions
iya+0.027
handed+0.026
joy+0.026
iley+0.026
 Sandra+0.026
Neg logit contributions
 off-0.007
 by-0.006
 from-0.006
 back-0.006
 both-0.006
Token him
Token ID683
Feature activation+1.27e-02

Pos logit contributions
iley+0.062
anne+0.061
joy+0.060
ya+0.058
ited+0.057
Neg logit contributions
 both-0.020
 off-0.019
ags-0.019
 back-0.017
 are-0.017
Token because
Token ID780
Feature activation+4.04e-02

Pos logit contributions
iley+0.025
 inspection+0.024
joy+0.023
zed+0.023
 tending+0.023
Neg logit contributions
 off-0.013
ags-0.013
 fart-0.012
 back-0.012
ac-0.012
Token she
Token ID673
Feature activation+8.72e-03

Pos logit contributions
iley+0.044
joy+0.040
 inspection+0.039
zed+0.039
ya+0.038
Neg logit contributions
ags-0.075
 brothers-0.074
 off-0.073
 fart-0.073
 cars-0.072
Token had
Token ID550
Feature activation+1.37e-03

Pos logit contributions
iley+0.014
 seeing+0.013
 inspection+0.013
zed+0.013
 tending+0.013
Neg logit contributions
ags-0.012
 fart-0.012
ac-0.011
 off-0.011
 ends-0.010
Token told
Token ID1297
Feature activation+1.00e-02

Pos logit contributions
iley+0.002
 inspection+0.002
ya+0.002
joy+0.002
 tending+0.002
Neg logit contributions
ags-0.002
 fart-0.002
 off-0.002
ac-0.002
 ends-0.002
Token him
Token ID683
Feature activation+8.29e-03

Pos logit contributions
iley+0.022
joy+0.021
anne+0.020
 tending+0.020
ya+0.020
Neg logit contributions
 off-0.009
ags-0.009
 back-0.008
 both-0.008
 cars-0.008
Token many
Token ID867
Feature activation+1.14e-02

Pos logit contributions
iley+0.016
joy+0.015
ited+0.015
ya+0.015
 tending+0.015
Neg logit contributions
 off-0.009
 back-0.009
 brothers-0.009
 are-0.008
 will-0.008
Token the
Token ID262
Feature activation-8.03e-03

Pos logit contributions
ags+0.000
 fart+0.000
ac+0.000
 off+0.000
 brothers+0.000
Neg logit contributions
iley-0.000
 seeing-0.000
 inspection-0.000
zed-0.000
joy-0.000
Token store
Token ID3650
Feature activation+7.29e-03

Pos logit contributions
 fart+0.011
ags+0.011
 off+0.010
 back+0.010
 cars+0.010
Neg logit contributions
iley-0.013
 inspection-0.012
zed-0.012
joy-0.012
 seeing-0.012
Token.
Token ID13
Feature activation-5.84e-04

Pos logit contributions
iley+0.016
zed+0.015
 inspection+0.015
anne+0.015
aster+0.014
Neg logit contributions
 off-0.007
 back-0.007
 both-0.006
ags-0.006
 brothers-0.006
Token Maggie
Token ID33014
Feature activation+1.01e-02

Pos logit contributions
ac+0.000
ags+0.000
 off+0.000
 are+0.000
'll+0.000
Neg logit contributions
iley-0.001
handed-0.001
joy-0.001
aster-0.001
 tending-0.001
Token was
Token ID373
Feature activation+7.38e-03

Pos logit contributions
iley+0.018
 inspection+0.017
joy+0.017
 seeing+0.016
zed+0.016
Neg logit contributions
ac-0.012
ags-0.012
 fart-0.011
 brothers-0.011
 ends-0.011
Token feeling
Token ID4203
Feature activation+4.95e-02

Pos logit contributions
iley+0.015
 seeing+0.014
 inspection+0.014
zed+0.014
 tending+0.014
Neg logit contributions
ags-0.007
 fart-0.007
ac-0.007
 off-0.006
 brothers-0.006
Token excited
Token ID6568
Feature activation+1.03e-05

Pos logit contributions
iley+0.103
 inspection+0.102
anne+0.098
 seeing+0.098
zed+0.097
Neg logit contributions
ags-0.046
 fart-0.041
ac-0.041
 brothers-0.039
 off-0.039
Token and
Token ID290
Feature activation+1.66e-02

Pos logit contributions
zed+0.000
plet+0.000
iley+0.000
aster+0.000
handed+0.000
Neg logit contributions
 for-0.000
 back-0.000
 off-0.000
 by-0.000
 from-0.000
Token happy
Token ID3772
Feature activation+9.95e-03

Pos logit contributions
iley+0.034
 inspection+0.032
aster+0.032
joy+0.031
anne+0.030
Neg logit contributions
 both-0.016
 off-0.016
 back-0.015
ags-0.014
 down-0.014
Token.
Token ID13
Feature activation+7.66e-03

Pos logit contributions
aster+0.025
ited+0.025
anne+0.024
zed+0.024
iley+0.024
Neg logit contributions
 off-0.008
 friends-0.008
 for-0.007
 from-0.007
 back-0.007
Token She
Token ID1375
Feature activation+9.03e-03

Pos logit contributions
iley+0.017
aster+0.016
handed+0.016
 tending+0.016
zed+0.016
Neg logit contributions
ags-0.006
 off-0.006
'll-0.005
ac-0.005
 back-0.005
Token and
Token ID290
Feature activation+1.39e-02

Pos logit contributions
iley+0.004
 seeing+0.004
joy+0.004
 inspection+0.004
 tending+0.004
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 off-0.002
Bang-0.002
Token white
Token ID2330
Feature activation+5.87e-03

Pos logit contributions
iley+0.026
 inspection+0.024
 seeing+0.024
joy+0.024
zed+0.024
Neg logit contributions
ags-0.016
 fart-0.016
ac-0.015
 off-0.014
 ends-0.014
Token and
Token ID290
Feature activation+1.75e-02

Pos logit contributions
iley+0.014
zed+0.014
ya+0.014
 inspection+0.013
aster+0.013
Neg logit contributions
ags-0.004
 back-0.003
 off-0.003
 cars-0.003
 fart-0.003
Token had
Token ID550
Feature activation+1.54e-03

Pos logit contributions
iley+0.030
 inspection+0.028
 seeing+0.028
joy+0.028
zed+0.027
Neg logit contributions
ags-0.022
 fart-0.022
ac-0.021
 off-0.020
 ends-0.020
Token a
Token ID257
Feature activation+1.07e-02

Pos logit contributions
iley+0.003
 seeing+0.003
 inspection+0.003
zed+0.003
 tending+0.003
Neg logit contributions
ags-0.002
ac-0.001
 fart-0.001
 off-0.001
 brothers-0.001
Token sign
Token ID1051
Feature activation+4.37e-02

Pos logit contributions
iley+0.019
 seeing+0.019
 inspection+0.019
joy+0.018
 tending+0.018
Neg logit contributions
ags-0.013
ac-0.012
 fart-0.011
 off-0.010
 brothers-0.010
Token that
Token ID326
Feature activation+4.11e-04

Pos logit contributions
iley+0.093
joy+0.086
ya+0.085
 inspection+0.084
zed+0.084
Neg logit contributions
 off-0.048
 back-0.046
 cars-0.045
 ends-0.043
 are-0.041
Token said
Token ID531
Feature activation+1.27e-02

Pos logit contributions
iley+0.001
joy+0.001
 inspection+0.001
 seeing+0.001
 tending+0.001
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
 off-0.000
 brothers-0.000
Token "
Token ID366
Feature activation+9.95e-03

Pos logit contributions
iley+0.027
ited+0.027
ya+0.025
handed+0.025
 tending+0.024
Neg logit contributions
 are-0.012
ags-0.011
 off-0.011
ac-0.011
 back-0.011
TokenCirc
Token ID31560
Feature activation+1.60e-02

Pos logit contributions
iley+0.015
 seeing+0.014
 inspection+0.014
joy+0.014
zed+0.014
Neg logit contributions
ags-0.014
 fart-0.013
ac-0.013
 off-0.013
 brothers-0.013
Tokenus
Token ID385
Feature activation-5.12e-02

Pos logit contributions
joy+0.030
 gazing+0.028
 seeing+0.028
 inspection+0.028
 leaving+0.028
Neg logit contributions
ac-0.020
ags-0.017
 ends-0.017
 will-0.017
 off-0.016
Token dragon
Token ID10441
Feature activation-1.46e-03

Pos logit contributions
ags+0.004
ac+0.004
 fart+0.004
 off+0.004
'll+0.004
Neg logit contributions
iley-0.005
 seeing-0.005
 tending-0.005
zed-0.004
joy-0.004
Token was
Token ID373
Feature activation-2.48e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.002
 brothers+0.002
Neg logit contributions
iley-0.003
 inspection-0.002
joy-0.002
 seeing-0.002
 tending-0.002
Token pleased
Token ID10607
Feature activation-6.84e-03

Pos logit contributions
 off+0.002
ags+0.002
ac+0.002
 back+0.002
 fart+0.002
Neg logit contributions
 inspection-0.005
iley-0.005
joy-0.005
handed-0.005
 tending-0.005
Token and
Token ID290
Feature activation-3.95e-03

Pos logit contributions
 off+0.005
 back+0.005
 friends+0.004
 brothers+0.004
 from+0.004
Neg logit contributions
aster-0.016
ited-0.016
handed-0.016
zed-0.016
anne-0.015
Token took
Token ID1718
Feature activation+7.87e-03

Pos logit contributions
ags+0.005
 fart+0.005
 off+0.005
ac+0.005
 brothers+0.005
Neg logit contributions
iley-0.006
 inspection-0.006
joy-0.006
 tending-0.006
 seeing-0.006
Token Maria
Token ID14200
Feature activation+4.17e-02

Pos logit contributions
iley+0.014
 seeing+0.014
 inspection+0.013
 tending+0.013
joy+0.013
Neg logit contributions
ags-0.009
 fart-0.009
ac-0.009
 off-0.008
 brothers-0.008
Token up
Token ID510
Feature activation+2.67e-02

Pos logit contributions
iley+0.097
 inspection+0.095
 seeing+0.094
joy+0.093
zed+0.093
Neg logit contributions
ags-0.025
 fart-0.024
 off-0.024
ac-0.024
 back-0.021
Token into
Token ID656
Feature activation-1.38e-02

Pos logit contributions
iley+0.055
 inspection+0.053
zed+0.053
 tending+0.053
anne+0.052
Neg logit contributions
ags-0.024
 off-0.024
ac-0.022
 fart-0.021
 back-0.021
Token the
Token ID262
Feature activation+3.79e-03

Pos logit contributions
 off+0.010
ags+0.009
 brothers+0.009
 both+0.009
 cars+0.008
Neg logit contributions
anne-0.031
zed-0.030
ited-0.030
 inspection-0.029
iley-0.029
Token sky
Token ID6766
Feature activation+1.10e-02

Pos logit contributions
iley+0.006
joy+0.006
 seeing+0.006
 tending+0.006
 gazing+0.006
Neg logit contributions
ags-0.005
ac-0.005
 fart-0.005
 off-0.004
 brothers-0.004
Token.
Token ID13
Feature activation-6.34e-03

Pos logit contributions
iley+0.020
 inspection+0.019
zed+0.019
aster+0.018
joy+0.018
Neg logit contributions
 off-0.013
 back-0.013
ags-0.012
ac-0.012
 fart-0.012
Token,
Token ID11
Feature activation-1.56e-03

Pos logit contributions
 inspection+0.014
iley+0.014
zed+0.014
 seeing+0.014
aster+0.014
Neg logit contributions
ac-0.011
ags-0.011
 brothers-0.011
 fart-0.011
 off-0.010
Token than
Token ID621
Feature activation+3.17e-02

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 ends+0.002
 off+0.002
Neg logit contributions
 seeing-0.003
iley-0.003
 inspection-0.003
 tending-0.003
joy-0.003
Token it
Token ID340
Feature activation-9.48e-04

Pos logit contributions
iley+0.055
 inspection+0.053
zed+0.053
 seeing+0.053
joy+0.052
Neg logit contributions
ags-0.039
 fart-0.037
ac-0.036
 off-0.034
 brothers-0.034
Token was
Token ID373
Feature activation-9.15e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 are+0.001
Neg logit contributions
iley-0.002
 inspection-0.002
joy-0.002
 seeing-0.002
aster-0.002
Token to
Token ID284
Feature activation+7.97e-03

Pos logit contributions
 off+0.010
 back+0.010
ac+0.010
ags+0.009
 fart+0.009
Neg logit contributions
iley-0.018
 inspection-0.017
zed-0.017
anne-0.017
aster-0.016
Token be
Token ID307
Feature activation+7.91e-04

Pos logit contributions
joy+0.017
iley+0.017
 inspection+0.017
zed+0.017
 tending+0.016
Neg logit contributions
ags-0.007
 fart-0.006
ac-0.006
 are-0.006
 off-0.006
Token selfish
Token ID20363
Feature activation-2.83e-02

Pos logit contributions
iley+0.001
 inspection+0.001
 seeing+0.001
zed+0.001
 tending+0.001
Neg logit contributions
ags-0.001
 fart-0.001
ac-0.001
 off-0.001
 brothers-0.001
Token and
Token ID290
Feature activation+1.32e-02

Pos logit contributions
ags+0.034
 fart+0.032
ac+0.032
 off+0.029
 brothers+0.029
Neg logit contributions
iley-0.050
 seeing-0.048
 inspection-0.048
 tending-0.047
zed-0.047
Token keep
Token ID1394
Feature activation+2.09e-02

Pos logit contributions
iley+0.023
 inspection+0.022
 seeing+0.022
zed+0.021
joy+0.021
Neg logit contributions
ags-0.016
 fart-0.016
ac-0.015
 off-0.014
 brothers-0.014
Token things
Token ID1243
Feature activation-5.62e-04

Pos logit contributions
 inspection+0.042
iley+0.041
anne+0.040
zed+0.039
joy+0.039
Neg logit contributions
ags-0.022
 off-0.021
 brothers-0.020
 back-0.020
 both-0.019
Token only
Token ID691
Feature activation-1.75e-02

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 brothers+0.001
Neg logit contributions
iley-0.001
 inspection-0.001
 seeing-0.001
joy-0.001
zed-0.001
Token match
Token ID2872
Feature activation-1.90e-03

Pos logit contributions
ags+0.005
 fart+0.004
ac+0.004
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.007
 seeing-0.007
 tending-0.007
 inspection-0.007
joy-0.007
Token met
Token ID1138
Feature activation-7.21e-03

Pos logit contributions
ags+0.002
 fart+0.002
ac+0.002
 off+0.002
 brothers+0.002
Neg logit contributions
iley-0.003
 seeing-0.003
 inspection-0.003
joy-0.003
 tending-0.003
Token a
Token ID257
Feature activation-1.76e-02

Pos logit contributions
 fart+0.010
ags+0.010
ac+0.009
'll+0.008
 brothers+0.008
Neg logit contributions
 seeing-0.012
joy-0.011
iley-0.011
 tending-0.011
zed-0.011
Token bright
Token ID6016
Feature activation-1.44e-02

Pos logit contributions
ags+0.019
 fart+0.018
ac+0.018
 brothers+0.016
 off+0.016
Neg logit contributions
iley-0.033
zed-0.032
 inspection-0.032
 seeing-0.031
 tending-0.031
Token,
Token ID11
Feature activation-7.19e-03

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 brothers+0.013
 ends+0.013
Neg logit contributions
iley-0.029
zed-0.029
 inspection-0.028
 seeing-0.027
 tending-0.027
Token shiny
Token ID22441
Feature activation+5.92e-03

Pos logit contributions
ags+0.011
ac+0.010
 fart+0.009
am+0.009
 off+0.009
Neg logit contributions
 seeing-0.011
joy-0.011
 tending-0.011
iley-0.011
 gazing-0.011
Token coin
Token ID10752
Feature activation+8.06e-04

Pos logit contributions
iley+0.010
zed+0.010
 inspection+0.010
 seeing+0.010
 tending+0.010
Neg logit contributions
ags-0.007
 fart-0.007
ac-0.007
 off-0.006
 ends-0.006
Token.
Token ID13
Feature activation-9.26e-03

Pos logit contributions
iley+0.002
 inspection+0.002
zed+0.002
 Samantha+0.002
joy+0.002
Neg logit contributions
 back-0.001
 off-0.001
 cars-0.001
 will-0.001
 brothers-0.001
Token The
Token ID383
Feature activation+1.43e-03

Pos logit contributions
ac+0.014
 fart+0.014
ags+0.014
eps+0.012
 will+0.012
Neg logit contributions
joy-0.013
iley-0.013
 Katie-0.013
 seeing-0.012
zed-0.012
Token coin
Token ID10752
Feature activation-2.87e-03

Pos logit contributions
iley+0.003
 inspection+0.003
 seeing+0.003
zed+0.003
joy+0.002
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 brothers-0.001
 ends-0.001
Token told
Token ID1297
Feature activation-5.67e-03

Pos logit contributions
ags+0.004
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
zed-0.005
 tending-0.005
Token--
Token ID438
Feature activation+7.80e-03

Pos logit contributions
iley+0.041
joy+0.038
anne+0.038
 Samantha+0.038
 gazing+0.038
Neg logit contributions
 off-0.023
ags-0.021
 fart-0.021
 ends-0.019
ac-0.019
Tokenmem
Token ID11883
Feature activation+1.36e-02

Pos logit contributions
iley+0.012
 seeing+0.012
 inspection+0.012
joy+0.011
 leaving+0.011
Neg logit contributions
ac-0.011
ags-0.011
 fart-0.010
 off-0.010
 will-0.010
Tokenor
Token ID273
Feature activation-1.30e-03

Pos logit contributions
iley+0.021
 seeing+0.020
 inspection+0.020
 tending+0.020
joy+0.020
Neg logit contributions
ags-0.020
ac-0.018
 fart-0.018
 off-0.017
 ends-0.017
Tokenizing
Token ID2890
Feature activation+1.10e-02

Pos logit contributions
ac+0.002
ags+0.002
 off+0.002
 fart+0.002
 brothers+0.002
Neg logit contributions
iley-0.002
 inspection-0.002
joy-0.002
 seeing-0.002
 tending-0.002
Token the
Token ID262
Feature activation+1.59e-03

Pos logit contributions
iley+0.020
joy+0.019
handed+0.018
aster+0.018
 inspection+0.018
Neg logit contributions
 off-0.013
ags-0.013
ac-0.013
 fart-0.013
 ends-0.012
Token answers
Token ID7429
Feature activation+1.05e-02

Pos logit contributions
iley+0.003
 inspection+0.003
 seeing+0.003
zed+0.003
 tending+0.003
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.002
 brothers-0.002
 ends-0.002
Token and
Token ID290
Feature activation-1.12e-02

Pos logit contributions
iley+0.018
joy+0.017
 inspection+0.017
handed+0.017
 seeing+0.017
Neg logit contributions
 off-0.013
ac-0.013
ags-0.012
 ends-0.012
 back-0.012
Token trying
Token ID2111
Feature activation-1.21e-02

Pos logit contributions
 fart+0.014
ags+0.014
 off+0.014
ac+0.014
 ends+0.013
Neg logit contributions
iley-0.019
 inspection-0.018
joy-0.017
zed-0.017
 tending-0.017
Token to
Token ID284
Feature activation+7.56e-03

Pos logit contributions
 off+0.011
ags+0.010
 fart+0.010
 back+0.009
ac+0.009
Neg logit contributions
iley-0.026
handed-0.024
aster-0.024
joy-0.024
ited-0.024
Token remember
Token ID3505
Feature activation+2.06e-02

Pos logit contributions
iley+0.014
 seeing+0.013
 inspection+0.013
zed+0.013
 tending+0.013
Neg logit contributions
ags-0.009
 fart-0.008
ac-0.008
 off-0.007
 brothers-0.007
Token them
Token ID606
Feature activation+2.39e-02

Pos logit contributions
iley+0.038
ited+0.037
handed+0.036
joy+0.036
anne+0.036
Neg logit contributions
 off-0.025
 will-0.022
 are-0.022
 back-0.022
ags-0.022
Token She
Token ID1375
Feature activation+1.52e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.003
 brothers+0.003
Neg logit contributions
iley-0.006
 seeing-0.005
 inspection-0.005
zed-0.005
 tending-0.005
Token picked
Token ID6497
Feature activation-3.36e-03

Pos logit contributions
iley+0.002
 seeing+0.002
 inspection+0.002
zed+0.002
 tending+0.002
Neg logit contributions
ags-0.003
 fart-0.003
ac-0.002
 off-0.002
ay-0.002
Token it
Token ID340
Feature activation-1.51e-02

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.005
 off+0.004
 ends+0.004
Neg logit contributions
iley-0.005
 seeing-0.005
 inspection-0.005
zed-0.005
 tending-0.005
Token up
Token ID510
Feature activation-3.10e-03

Pos logit contributions
ags+0.019
ac+0.016
 brothers+0.016
 teams+0.015
 fart+0.015
Neg logit contributions
 inspection-0.027
 seeing-0.027
 tending-0.026
zed-0.026
looking-0.026
Token and
Token ID290
Feature activation-2.58e-02

Pos logit contributions
ags+0.003
 fart+0.002
ac+0.002
 off+0.002
 back+0.002
Neg logit contributions
iley-0.007
zed-0.006
 inspection-0.006
 seeing-0.006
 tending-0.006
Token went
Token ID1816
Feature activation+3.40e-03

Pos logit contributions
ags+0.034
 fart+0.033
ac+0.032
 off+0.032
 ends+0.030
Neg logit contributions
iley-0.042
 inspection-0.039
joy-0.039
zed-0.039
 seeing-0.039
Token to
Token ID284
Feature activation-5.69e-03

Pos logit contributions
iley+0.007
 inspection+0.007
zed+0.007
joy+0.007
ya+0.007
Neg logit contributions
ags-0.003
 off-0.003
 back-0.003
 fart-0.003
ac-0.002
Token her
Token ID607
Feature activation-1.45e-03

Pos logit contributions
 off+0.006
ags+0.006
 back+0.005
 fart+0.005
 both+0.005
Neg logit contributions
iley-0.011
joy-0.010
anne-0.010
 inspection-0.010
ya-0.010
Token mom
Token ID1995
Feature activation+1.89e-03

Pos logit contributions
ags+0.001
 fart+0.001
ac+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.003
 seeing-0.003
 tending-0.003
 inspection-0.003
joy-0.003
Token to
Token ID284
Feature activation+1.69e-03

Pos logit contributions
iley+0.004
zed+0.004
 inspection+0.004
joy+0.004
ya+0.004
Neg logit contributions
 off-0.001
ags-0.001
ac-0.001
 back-0.001
 fart-0.001
Token show
Token ID905
Feature activation-4.80e-03

Pos logit contributions
iley+0.003
 inspection+0.003
joy+0.003
zed+0.003
anne+0.003
Neg logit contributions
ags-0.002
 off-0.002
 fart-0.002
 back-0.002
ac-0.002
Token because
Token ID780
Feature activation+3.92e-02

Pos logit contributions
zed+0.014
iley+0.013
 tending+0.013
 inspection+0.013
anne+0.013
Neg logit contributions
 off-0.006
ags-0.006
 back-0.006
 brothers-0.006
 fart-0.005
Token he
Token ID339
Feature activation+6.09e-03

Pos logit contributions
iley+0.046
 inspection+0.043
joy+0.043
 seeing+0.042
zed+0.042
Neg logit contributions
ags-0.068
 brothers-0.067
 off-0.067
 fart-0.066
ac-0.066
Token had
Token ID550
Feature activation-1.35e-03

Pos logit contributions
iley+0.009
 seeing+0.009
zed+0.009
 tending+0.008
 inspection+0.008
Neg logit contributions
ags-0.010
 fart-0.009
ac-0.009
 off-0.008
 teams-0.008
Token never
Token ID1239
Feature activation+7.54e-03

Pos logit contributions
 off+0.001
 fart+0.001
ags+0.001
 back+0.001
ac+0.001
Neg logit contributions
iley-0.003
 inspection-0.003
ya-0.003
anne-0.003
joy-0.002
Token been
Token ID587
Feature activation+1.48e-02

Pos logit contributions
iley+0.010
 inspection+0.010
joy+0.009
zed+0.009
ya+0.009
Neg logit contributions
ags-0.012
 fart-0.012
ac-0.012
 off-0.012
 ends-0.011
Token to
Token ID284
Feature activation+1.33e-02

Pos logit contributions
 inspection+0.030
iley+0.029
anne+0.029
zed+0.029
aster+0.028
Neg logit contributions
 off-0.014
 back-0.014
ags-0.013
 fart-0.013
ac-0.013
Token the
Token ID262
Feature activation-7.03e-03

Pos logit contributions
 inspection+0.026
iley+0.026
joy+0.026
anne+0.025
zed+0.025
Neg logit contributions
 off-0.014
 fart-0.013
ags-0.013
 back-0.013
 ends-0.012
Token bar
Token ID2318
Feature activation-7.92e-04

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.010
 off+0.009
 ends+0.009
Neg logit contributions
iley-0.011
 inspection-0.010
zed-0.010
 seeing-0.010
anne-0.010
Tokenber
Token ID527
Feature activation+3.27e-03

Pos logit contributions
ac+0.001
 brothers+0.001
 back+0.001
 off+0.001
 fart+0.001
Neg logit contributions
 inspection-0.001
zed-0.001
iley-0.001
joy-0.001
 tending-0.001
Token before
Token ID878
Feature activation+1.85e-02

Pos logit contributions
iley+0.005
 inspection+0.005
joy+0.005
zed+0.005
anne+0.005
Neg logit contributions
 fart-0.004
ac-0.004
ags-0.004
 off-0.004
 back-0.004
Token.
Token ID13
Feature activation-4.43e-04

Pos logit contributions
iley+0.032
ya+0.030
anne+0.030
 inspection+0.030
joy+0.029
Neg logit contributions
 off-0.024
 brothers-0.023
 cars-0.023
ags-0.022
 fart-0.022
Tokenives
Token ID1083
Feature activation-3.16e-03

Pos logit contributions
ags+0.011
 fart+0.011
ac+0.010
 off+0.010
 ends+0.010
Neg logit contributions
iley-0.015
joy-0.014
 seeing-0.014
 tending-0.014
zed-0.013
Token.
Token ID13
Feature activation+1.17e-03

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.005
 off+0.005
 brothers+0.005
Neg logit contributions
iley-0.004
 inspection-0.004
 seeing-0.004
joy-0.004
zed-0.004
Token One
Token ID1881
Feature activation-1.57e-02

Pos logit contributions
iley+0.002
 seeing+0.002
 inspection+0.002
joy+0.002
 tending+0.002
Neg logit contributions
ags-0.002
ac-0.002
 fart-0.002
 off-0.002
 ends-0.002
Token day
Token ID1110
Feature activation+5.25e-03

Pos logit contributions
ags+0.021
ac+0.021
 fart+0.021
 off+0.019
 brothers+0.019
Neg logit contributions
iley-0.025
joy-0.024
 inspection-0.024
 seeing-0.024
zed-0.023
Token,
Token ID11
Feature activation-7.53e-03

Pos logit contributions
iley+0.009
 seeing+0.009
 inspection+0.008
 tending+0.008
zed+0.008
Neg logit contributions
ags-0.007
 fart-0.006
ac-0.006
 off-0.006
 brothers-0.006
Token Benny
Token ID44275
Feature activation-9.42e-03

Pos logit contributions
ags+0.010
 fart+0.010
ac+0.010
 off+0.009
 brothers+0.009
Neg logit contributions
iley-0.012
 inspection-0.011
 seeing-0.011
zed-0.011
joy-0.011
Token received
Token ID2722
Feature activation+4.60e-03

Pos logit contributions
ags+0.013
 fart+0.012
ac+0.012
 off+0.011
 brothers+0.011
Neg logit contributions
iley-0.015
 seeing-0.015
 inspection-0.015
zed-0.014
 tending-0.014
Token a
Token ID257
Feature activation-6.70e-03

Pos logit contributions
iley+0.007
 seeing+0.007
joy+0.007
 tending+0.007
aster+0.007
Neg logit contributions
ags-0.006
 fart-0.006
ac-0.006
 ends-0.006
'll-0.006
Token message
Token ID3275
Feature activation+8.56e-03

Pos logit contributions
ags+0.008
 fart+0.007
ac+0.007
 off+0.007
 brothers+0.006
Neg logit contributions
iley-0.012
 seeing-0.012
 inspection-0.012
zed-0.012
 tending-0.012
Token from
Token ID422
Feature activation-2.70e-02

Pos logit contributions
iley+0.020
 inspection+0.019
joy+0.019
zed+0.018
ya+0.018
Neg logit contributions
 off-0.007
 back-0.006
ags-0.006
ac-0.005
 fart-0.005
Token his
Token ID465
Feature activation+1.17e-02

Pos logit contributions
ags+0.025
 off+0.023
 fart+0.021
 back+0.021
 brothers+0.020
Neg logit contributions
iley-0.056
 inspection-0.055
joy-0.055
anne-0.054
 tending-0.053
Token surfaces
Token ID16649
Feature activation+1.65e-02

Pos logit contributions
ags+0.005
ac+0.004
 fart+0.004
apped+0.004
Bang+0.004
Neg logit contributions
 tending-0.007
 seeing-0.007
 inspection-0.007
ya-0.006
 smelling-0.006
Token to
Token ID284
Feature activation+1.60e-02

Pos logit contributions
iley+0.033
 inspection+0.030
joy+0.030
zed+0.030
 seeing+0.030
Neg logit contributions
ags-0.016
 off-0.016
 fart-0.016
ac-0.015
 back-0.015
Token make
Token ID787
Feature activation-1.07e-02

Pos logit contributions
iley+0.034
joy+0.032
 inspection+0.032
 seeing+0.031
 tending+0.031
Neg logit contributions
ags-0.015
 fart-0.013
ac-0.013
 off-0.013
 are-0.012
Token sure
Token ID1654
Feature activation-2.42e-02

Pos logit contributions
ags+0.012
 off+0.011
 fart+0.011
ac+0.011
 brothers+0.010
Neg logit contributions
iley-0.021
 inspection-0.019
joy-0.019
 seeing-0.019
 tending-0.019
Token everything
Token ID2279
Feature activation+2.57e-03

Pos logit contributions
 off+0.032
ags+0.031
ac+0.030
 fart+0.030
 ends+0.030
Neg logit contributions
iley-0.042
joy-0.038
zed-0.037
 inspection-0.037
ya-0.037
Token was
Token ID373
Feature activation-1.87e-02

Pos logit contributions
iley+0.005
joy+0.005
 seeing+0.005
 inspection+0.005
aster+0.005
Neg logit contributions
 ends-0.002
ags-0.002
 fart-0.002
 off-0.002
 will-0.002
Token sparkling
Token ID39072
Feature activation-2.04e-02

Pos logit contributions
ags+0.017
 fart+0.015
ac+0.015
 off+0.015
 back+0.014
Neg logit contributions
iley-0.041
 inspection-0.038
joy-0.038
ya-0.037
zed-0.037
Token clean
Token ID3424
Feature activation-9.67e-03

Pos logit contributions
ags+0.023
 fart+0.023
 off+0.023
ac+0.022
 back+0.022
Neg logit contributions
iley-0.038
zed-0.036
 inspection-0.036
 seeing-0.035
aster-0.035
Token.
Token ID13
Feature activation-1.06e-02

Pos logit contributions
ags+0.014
 fart+0.014
ac+0.014
 off+0.013
 brothers+0.013
Neg logit contributions
iley-0.014
 seeing-0.014
 inspection-0.014
zed-0.013
joy-0.013
Token After
Token ID2293
Feature activation+1.03e-02

Pos logit contributions
ags+0.011
 fart+0.010
ac+0.010
 off+0.009
 brothers+0.009
Neg logit contributions
iley-0.021
 seeing-0.020
 inspection-0.020
joy-0.020
zed-0.020
Token that
Token ID326
Feature activation-1.75e-02

Pos logit contributions
anne+0.023
ited+0.022
iley+0.022
ya+0.021
handed+0.021
Neg logit contributions
 both-0.010
 brothers-0.010
 cars-0.009
 friends-0.009
 off-0.009
Token
Token ID220
Feature activation-6.90e-03

Pos logit contributions
iley+0.001
 tending+0.001
 seeing+0.001
zed+0.001
 inspection+0.001
Neg logit contributions
ags-0.001
 fart-0.001
 off-0.001
 brothers-0.001
ac-0.001
Token
Token ID198
Feature activation+8.80e-03

Pos logit contributions
ags+0.009
 fart+0.009
ac+0.009
 off+0.008
 ends+0.008
Neg logit contributions
iley-0.011
 inspection-0.011
 seeing-0.011
zed-0.011
joy-0.011
Token
Token ID198
Feature activation+1.16e-02

Pos logit contributions
iley+0.017
 seeing+0.017
 tending+0.017
 inspection+0.016
zed+0.016
Neg logit contributions
ags-0.009
 fart-0.008
ac-0.008
 off-0.007
 brothers-0.007
TokenJack
Token ID14295
Feature activation-2.42e-03

Pos logit contributions
 seeing+0.020
 inspection+0.020
iley+0.020
 tending+0.020
joy+0.020
Neg logit contributions
ags-0.013
ac-0.013
 fart-0.013
Bang-0.012
 brothers-0.012
Token nodded
Token ID14464
Feature activation+3.48e-03

Pos logit contributions
ags+0.003
 fart+0.003
ac+0.003
 off+0.003
 brothers+0.002
Neg logit contributions
iley-0.004
 seeing-0.004
 inspection-0.004
 tending-0.004
joy-0.004
Token and
Token ID290
Feature activation-1.80e-02

Pos logit contributions
iley+0.007
 inspection+0.007
zed+0.007
handed+0.007
 seeing+0.007
Neg logit contributions
ags-0.003
 off-0.003
 back-0.003
ac-0.003
 brothers-0.003
Token realized
Token ID6939
Feature activation+2.24e-02

Pos logit contributions
 off+0.019
ags+0.019
 fart+0.018
ac+0.018
 brothers+0.018
Neg logit contributions
iley-0.033
joy-0.032
 inspection-0.032
 tending-0.031
 seeing-0.031
Token he
Token ID339
Feature activation-7.52e-03

Pos logit contributions
ited+0.036
iley+0.034
anne+0.034
 tending+0.031
cious+0.031
Neg logit contributions
 off-0.035
 brothers-0.032
 back-0.031
 friends-0.031
 cars-0.031
Token should
Token ID815
Feature activation+6.99e-03

Pos logit contributions
ags+0.008
ac+0.008
 fart+0.008
 ends+0.008
 off+0.007
Neg logit contributions
iley-0.014
 inspection-0.013
 seeing-0.013
joy-0.013
 tending-0.013
Token always
Token ID1464
Feature activation+4.06e-04

Pos logit contributions
joy+0.015
iley+0.015
 inspection+0.014
ya+0.014
 tending+0.014
Neg logit contributions
 are-0.007
ags-0.006
 will-0.006
 off-0.006
 back-0.006
Token have
Token ID423
Feature activation+4.40e-03

Pos logit contributions
joy+0.001
iley+0.001
 inspection+0.001
anne+0.001
ya+0.001
Neg logit contributions
ags-0.000
 will-0.000
 are-0.000
 off-0.000
ac-0.000
Token One
Token ID1881
Feature activation-1.33e-02

Pos logit contributions
iley+0.016
zed+0.016
 seeing+0.015
 inspection+0.015
 tending+0.015
Neg logit contributions
ags-0.008
 fart-0.007
ac-0.007
 ends-0.007
 are-0.007
Token day
Token ID1110
Feature activation+6.83e-03

Pos logit contributions
ags+0.016
ac+0.016
 fart+0.015
 back+0.015
 off+0.015
Neg logit contributions
iley-0.023
joy-0.022
 inspection-0.022
 seeing-0.022
zed-0.021
Token she
Token ID673
Feature activation-6.92e-03

Pos logit contributions
handed+0.018
iley+0.017
anne+0.016
ya+0.016
ited+0.016
Neg logit contributions
 off-0.004
 are-0.003
 down-0.003
 both-0.003
 can-0.003
Token was
Token ID373
Feature activation-2.54e-03

Pos logit contributions
ags+0.010
 fart+0.009
ac+0.009
 off+0.008
 brothers+0.008
Neg logit contributions
iley-0.011
 seeing-0.011
zed-0.011
 inspection-0.011
 tending-0.011
Token walking
Token ID6155
Feature activation-5.12e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 ends+0.004
Neg logit contributions
iley-0.003
 seeing-0.003
 inspection-0.003
 tending-0.003
zed-0.003
Token in
Token ID287
Feature activation-1.55e-02

Pos logit contributions
ags+0.006
 fart+0.005
ac+0.005
 ends+0.005
 teams+0.005
Neg logit contributions
 seeing-0.010
 tending-0.009
iley-0.009
 inspection-0.009
joy-0.009
Token the
Token ID262
Feature activation-8.75e-04

Pos logit contributions
 cars+0.008
 back+0.008
ags+0.008
 off+0.008
 are+0.007
Neg logit contributions
anne-0.037
zed-0.037
iley-0.036
ya-0.036
cious-0.035
Token forest
Token ID8222
Feature activation+2.53e-02

Pos logit contributions
ags+0.001
ac+0.001
 fart+0.001
 off+0.001
 ends+0.001
Neg logit contributions
iley-0.001
 seeing-0.001
 tending-0.001
joy-0.001
 inspection-0.001
Token,
Token ID11
Feature activation+4.71e-03

Pos logit contributions
iley+0.059
ya+0.054
iya+0.054
anne+0.054
zed+0.053
Neg logit contributions
 back-0.019
 both-0.018
 down-0.018
 by-0.018
 cars-0.017
Token searching
Token ID10342
Feature activation-6.36e-04

Pos logit contributions
iley+0.009
anne+0.008
 product+0.008
handed+0.008
iya+0.008
Neg logit contributions
 down-0.006
 back-0.006
 both-0.006
 up-0.006
 fart-0.005
Token for
Token ID329
Feature activation-9.63e-03

Pos logit contributions
ags+0.000
 fart+0.000
ac+0.000
 off+0.000
 brothers+0.000
Neg logit contributions
iley-0.001
 seeing-0.001
 inspection-0.001
 tending-0.001
zed-0.001
Token boy
Token ID2933
Feature activation-7.81e-03

Pos logit contributions
ags+0.004
 fart+0.004
ac+0.004
 off+0.004
 ends+0.003
Neg logit contributions
iley-0.006
 seeing-0.006
joy-0.006
 tending-0.006
 inspection-0.006
Token named
Token ID3706
Feature activation+1.76e-02

Pos logit contributions
 brothers+0.008
ac+0.008
 friends+0.008
 cars+0.007
 both+0.007
Neg logit contributions
 inspection-0.017
joy-0.016
 hesitation-0.016
zed-0.015
aman-0.015
Token Tim
Token ID5045
Feature activation-4.10e-03

Pos logit contributions
 seeing+0.026
iley+0.025
 inspection+0.025
 tending+0.024
aster+0.024
Neg logit contributions
ags-0.027
ac-0.026
 fart-0.025
amed-0.024
 ends-0.023
Tokenmy
Token ID1820
Feature activation-9.77e-03

Pos logit contributions
ags+0.003
ac+0.002
 brothers+0.002
 fart+0.002
 back+0.002
Neg logit contributions
iley-0.010
zed-0.010
 inspection-0.010
joy-0.009
 tending-0.009
Token.
Token ID13
Feature activation+8.75e-03

Pos logit contributions
 brothers+0.011
 back+0.011
 both+0.011
ags+0.010
ac+0.010
Neg logit contributions
 inspection-0.019
iley-0.018
joy-0.018
zed-0.018
 hesitation-0.018
Token Tim
Token ID5045
Feature activation-5.73e-03

Pos logit contributions
 seeing+0.014
 inspection+0.014
iley+0.014
 tending+0.013
joy+0.013
Neg logit contributions
ags-0.012
ac-0.012
 fart-0.011
 off-0.011
 ends-0.011
Tokenmy
Token ID1820
Feature activation-1.03e-02

Pos logit contributions
ags+0.005
 fart+0.004
ac+0.004
 off+0.004
 teams+0.004
Neg logit contributions
iley-0.012
 seeing-0.012
 tending-0.012
zed-0.012
 inspection-0.012
Token loved
Token ID6151
Feature activation+2.62e-02

Pos logit contributions
ags+0.011
ac+0.010
 fart+0.010
 brothers+0.010
 ends+0.009
Neg logit contributions
iley-0.019
 inspection-0.019
joy-0.019
 seeing-0.018
 hesitation-0.018
Token to
Token ID284
Feature activation+1.98e-02

Pos logit contributions
iley+0.048
 seeing+0.047
 inspection+0.046
 tending+0.046
zed+0.046
Neg logit contributions
ags-0.030
 fart-0.028
ac-0.028
 off-0.025
 ends-0.025
Token eat
Token ID4483
Feature activation+7.14e-03

Pos logit contributions
iley+0.042
joy+0.040
 inspection+0.040
ya+0.038
 seeing+0.038
Neg logit contributions
ags-0.018
 fart-0.016
 off-0.016
 brothers-0.016
 are-0.015
Token candy
Token ID18550
Feature activation+3.89e-03

Pos logit contributions
iley+0.011
 seeing+0.011
 tending+0.010
 inspection+0.010
joy+0.010
Neg logit contributions
ags-0.010
ac-0.010
 fart-0.010
 ends-0.009
'll-0.009
Token They
Token ID1119
Feature activation+5.12e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.006
 off+0.005
 ends+0.005
Neg logit contributions
iley-0.013
 seeing-0.012
 inspection-0.012
joy-0.012
 tending-0.012
Token share
Token ID2648
Feature activation-1.34e-02

Pos logit contributions
 seeing+0.009
aster+0.008
 tending+0.008
 expected+0.008
 inspection+0.008
Neg logit contributions
ags-0.007
 fart-0.006
ac-0.006
 off-0.005
ay-0.005
Token and
Token ID290
Feature activation+1.15e-03

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.016
'll+0.015
 brothers+0.014
Neg logit contributions
 seeing-0.023
 inspection-0.022
 tending-0.022
joy-0.021
iley-0.021
Token help
Token ID1037
Feature activation-3.27e-02

Pos logit contributions
 seeing+0.002
 tending+0.002
aster+0.002
 inspection+0.002
 tasting+0.002
Neg logit contributions
ags-0.002
ac-0.001
 fart-0.001
ay-0.001
'll-0.001
Token each
Token ID1123
Feature activation-2.57e-02

Pos logit contributions
 fart+0.045
ags+0.045
ac+0.044
amed+0.042
'll+0.041
Neg logit contributions
 seeing-0.049
aster-0.048
iley-0.047
 tending-0.046
 Katie-0.045
Token other
Token ID584
Feature activation-5.58e-02

Pos logit contributions
ags+0.023
 fart+0.019
ac+0.016
 off+0.015
amed+0.015
Neg logit contributions
aster-0.056
 seeing-0.055
 tending-0.054
 inspection-0.052
iley-0.052
Token.
Token ID13
Feature activation-6.36e-03

Pos logit contributions
ags+0.095
 fart+0.090
ac+0.089
 off+0.082
 brothers+0.082
Neg logit contributions
iley-0.070
 seeing-0.069
 tending-0.068
 inspection-0.067
zed-0.067
Token They
Token ID1119
Feature activation+5.57e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.006
 off+0.005
 ends+0.005
Neg logit contributions
iley-0.012
 seeing-0.012
 inspection-0.012
joy-0.012
 tending-0.012
Token are
Token ID389
Feature activation+7.65e-03

Pos logit contributions
 seeing+0.010
aster+0.010
 tending+0.010
 expected+0.010
iley+0.009
Neg logit contributions
ags-0.006
ac-0.006
 fart-0.006
 off-0.005
amed-0.005
Token good
Token ID922
Feature activation+1.07e-02

Pos logit contributions
 seeing+0.012
 tending+0.012
 watching+0.012
 leaving+0.011
 gazing+0.011
Neg logit contributions
 fart-0.010
ags-0.010
ac-0.010
Bang-0.009
ay-0.009
Token friends
Token ID2460
Feature activation-2.52e-02

Pos logit contributions
 seeing+0.022
joy+0.022
iley+0.022
 tending+0.021
 watching+0.021
Neg logit contributions
ags-0.009
 fart-0.009
Bang-0.008
ac-0.008
'll-0.008
Token,
Token ID11
Feature activation-3.58e-02

Pos logit contributions
ags+0.006
ac+0.006
 fart+0.006
 off+0.005
 are+0.005
Neg logit contributions
iley-0.011
zed-0.010
 tending-0.010
 inspection-0.010
 seeing-0.010
Token I
Token ID314
Feature activation+2.27e-03

Pos logit contributions
ags+0.049
 fart+0.047
ac+0.046
 off+0.043
 brothers+0.043
Neg logit contributions
iley-0.057
 seeing-0.055
 inspection-0.054
 tending-0.053
zed-0.053
Token can
Token ID460
Feature activation-1.68e-02

Pos logit contributions
iley+0.005
 seeing+0.005
 inspection+0.005
zed+0.005
joy+0.005
Neg logit contributions
ags-0.002
 fart-0.002
ac-0.001
 off-0.001
 cars-0.001
Token make
Token ID787
Feature activation-1.16e-02

Pos logit contributions
ags+0.019
 fart+0.017
ac+0.017
 off+0.016
 brothers+0.016
Neg logit contributions
iley-0.032
 inspection-0.030
joy-0.030
 seeing-0.030
zed-0.030
Token your
Token ID534
Feature activation-1.47e-02

Pos logit contributions
ags+0.014
 fart+0.013
ac+0.013
 off+0.012
 brothers+0.012
Neg logit contributions
iley-0.021
 seeing-0.020
 inspection-0.020
zed-0.020
 tending-0.020
Token t
Token ID256
Feature activation-5.15e-02

Pos logit contributions
ags+0.017
ac+0.016
 fart+0.015
 off+0.015
amed+0.014
Neg logit contributions
iley-0.026
 seeing-0.025
joy-0.025
 tending-0.024
 gazing-0.024
Tokeneddy
Token ID21874
Feature activation-5.06e-02

Pos logit contributions
ags+0.079
 fart+0.074
Bang+0.072
 teams+0.068
 sticks+0.065
Neg logit contributions
iley-0.076
handed-0.073
 tending-0.068
 seeing-0.066
aster-0.066
Token bear
Token ID6842
Feature activation-2.59e-02

Pos logit contributions
ags+0.070
 fart+0.066
ac+0.064
Bang+0.060
 teams+0.060
Neg logit contributions
iley-0.076
 seeing-0.074
 tending-0.073
 dreaming-0.073
zed-0.073
Token disappear
Token ID10921
Feature activation-4.66e-03

Pos logit contributions
ags+0.032
 fart+0.029
ac+0.028
Bang+0.026
 ends+0.026
Neg logit contributions
 seeing-0.045
iley-0.045
 tending-0.044
zed-0.043
aster-0.043
Token and
Token ID290
Feature activation-9.02e-03

Pos logit contributions
ags+0.006
 fart+0.006
ac+0.005
Bang+0.005
 brothers+0.005
Neg logit contributions
iley-0.008
 seeing-0.007
 inspection-0.007
joy-0.007
ya-0.007
Token then
Token ID788
Feature activation-1.31e-02

Pos logit contributions
ags+0.012
ac+0.011
 fart+0.011
 brothers+0.010
 ends+0.010
Neg logit contributions
 seeing-0.015
iley-0.015
 tending-0.014
 inspection-0.014
zed-0.014
Token Friends
Token ID14213
Feature activation-2.81e-02

Pos logit contributions
 fart+0.011
ags+0.011
 off+0.010
ac+0.010
 ends+0.009
Neg logit contributions
iley-0.013
 seeing-0.012
joy-0.011
 tending-0.011
aster-0.011
Token also
Token ID635
Feature activation-2.28e-02

Pos logit contributions
 fart+0.032
ac+0.032
ags+0.031
 teams+0.029
am+0.029
Neg logit contributions
 seeing-0.046
 tending-0.044
 watching-0.043
 expected-0.043
 hearing-0.043
Token listen
Token ID6004
Feature activation+3.24e-04

Pos logit contributions
ags+0.029
ac+0.027
 fart+0.027
 off+0.024
amed+0.024
Neg logit contributions
 seeing-0.038
 tending-0.036
iley-0.036
 inspection-0.035
 hearing-0.035
Token to
Token ID284
Feature activation-1.11e-02

Pos logit contributions
 seeing+0.001
iley+0.001
zed+0.001
 tending+0.001
 inspection+0.001
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
Bang-0.000
 brothers-0.000
Token each
Token ID1123
Feature activation-3.18e-02

Pos logit contributions
ags+0.011
 fart+0.011
ac+0.011
 off+0.010
 ends+0.009
Neg logit contributions
iley-0.022
 seeing-0.021
 tending-0.020
 inspection-0.020
zed-0.020
Token other
Token ID584
Feature activation-6.14e-02

Pos logit contributions
ags+0.028
 fart+0.027
ac+0.024
apped+0.023
amed+0.022
Neg logit contributions
aster-0.065
iley-0.064
 seeing-0.064
 tending-0.062
ya-0.062
Token.
Token ID13
Feature activation-5.48e-03

Pos logit contributions
ags+0.095
 fart+0.090
ac+0.088
 brothers+0.082
Bang+0.081
Neg logit contributions
iley-0.086
 seeing-0.084
zed-0.082
 tending-0.082
 inspection-0.081
Token That
Token ID1320
Feature activation-1.98e-03

Pos logit contributions
ags+0.007
 fart+0.007
 off+0.007
ac+0.007
amed+0.006
Neg logit contributions
iley-0.009
 seeing-0.008
joy-0.008
aster-0.008
zed-0.008
Token is
Token ID318
Feature activation-7.72e-03

Pos logit contributions
ags+0.002
ac+0.002
 fart+0.002
amed+0.002
 off+0.002
Neg logit contributions
 seeing-0.004
iley-0.004
 tending-0.004
zed-0.004
 hearing-0.004
Token how
Token ID703
Feature activation+2.26e-04

Pos logit contributions
ags+0.011
 fart+0.010
ac+0.010
 ends+0.009
ay+0.009
Neg logit contributions
 seeing-0.012
iley-0.012
 tending-0.012
zed-0.011
 watching-0.011
Token we
Token ID356
Feature activation-1.98e-03

Pos logit contributions
iley+0.000
 seeing+0.000
 inspection+0.000
joy+0.000
zed+0.000
Neg logit contributions
ags-0.000
 fart-0.000
ac-0.000
amed-0.000
 off-0.000
Token and
Token ID290
Feature activation-9.06e-03

Pos logit contributions
ags+0.034
ac+0.033
 fart+0.033
 brothers+0.031
 ends+0.029
Neg logit contributions
 seeing-0.043
 inspection-0.043
 tending-0.042
 gazing-0.042
iley-0.041
Token try
Token ID1949
Feature activation+9.51e-05

Pos logit contributions
ags+0.012
ac+0.012
 fart+0.011
 brothers+0.010
amed+0.010
Neg logit contributions
iley-0.015
 seeing-0.014
 inspection-0.014
 tending-0.014
zed-0.014
Token to
Token ID284
Feature activation-5.41e-03

Pos logit contributions
iley+0.000
handed+0.000
zed+0.000
ited+0.000
anne+0.000
Neg logit contributions
ags-0.000
 off-0.000
 fart-0.000
 back-0.000
ac-0.000
Token grab
Token ID5552
Feature activation+7.16e-03

Pos logit contributions
ags+0.008
ac+0.008
 fart+0.008
 brothers+0.007
 off+0.007
Neg logit contributions
iley-0.008
 seeing-0.008
 tending-0.007
 inspection-0.007
aster-0.007
Token the
Token ID262
Feature activation-9.51e-03

Pos logit contributions
iley+0.012
 inspection+0.011
 seeing+0.011
joy+0.011
 tending+0.011
Neg logit contributions
ags-0.009
 fart-0.009
ac-0.009
 brothers-0.008
 ends-0.008
Token av
Token ID1196
Feature activation-9.06e-02

Pos logit contributions
ags+0.014
ac+0.013
amed+0.012
apped+0.012
 off+0.012
Neg logit contributions
 inspection-0.014
joy-0.014
 tending-0.013
 tasting-0.013
 seeing-0.013
Tokenoc
Token ID420
Feature activation-3.14e-02

Pos logit contributions
ags+0.170
 brothers+0.150
 fart+0.146
apped+0.145
 cars+0.144
Neg logit contributions
iley-0.103
 inspection-0.097
aster-0.095
 tending-0.094
ya-0.093
Tokenados
Token ID22484
Feature activation-1.82e-02

Pos logit contributions
ags+0.044
 fart+0.043
ac+0.042
 off+0.040
Bang+0.039
Neg logit contributions
iley-0.046
joy-0.045
 inspection-0.045
 tending-0.044
 seeing-0.044
Token.
Token ID13
Feature activation-6.31e-03

Pos logit contributions
ags+0.025
 fart+0.023
ac+0.023
 brothers+0.021
 off+0.021
Neg logit contributions
iley-0.030
 seeing-0.029
 inspection-0.028
 tending-0.028
zed-0.028
Token But
Token ID887
Feature activation+2.55e-02

Pos logit contributions
ags+0.005
 fart+0.004
ac+0.004
 off+0.004
 ends+0.004
Neg logit contributions
iley-0.014
 tending-0.014
 seeing-0.013
zed-0.013
 inspection-0.013
Token the
Token ID262
Feature activation-3.88e-03

Pos logit contributions
iley+0.035
 seeing+0.034
 inspection+0.034
 tending+0.033
zed+0.033
Neg logit contributions
ags-0.042
 fart-0.039
ac-0.039
 brothers-0.036
 ends-0.036
Token agreed
Token ID4987
Feature activation-6.16e-03

Pos logit contributions
iley+0.042
 seeing+0.041
 inspection+0.040
 tending+0.040
zed+0.039
Neg logit contributions
ags-0.039
 fart-0.038
ac-0.037
 off-0.034
 brothers-0.034
Token.
Token ID13
Feature activation-1.05e-02

Pos logit contributions
 off+0.006
 back+0.006
 both+0.005
 are+0.004
 from+0.004
Neg logit contributions
iley-0.015
ited-0.014
ya-0.013
handed-0.013
isa-0.013
Token They
Token ID1119
Feature activation+5.02e-03

Pos logit contributions
ags+0.009
 fart+0.009
ac+0.008
 ends+0.007
 off+0.007
Neg logit contributions
iley-0.022
 inspection-0.022
 seeing-0.022
joy-0.022
 tending-0.021
Token gave
Token ID2921
Feature activation-2.02e-02

Pos logit contributions
iley+0.007
 seeing+0.007
 expected+0.007
ya+0.006
 tending+0.006
Neg logit contributions
ags-0.008
 fart-0.008
ac-0.007
 Lions-0.007
 off-0.007
Token each
Token ID1123
Feature activation-1.72e-02

Pos logit contributions
 fart+0.030
ags+0.029
ac+0.028
'll+0.026
rr+0.026
Neg logit contributions
iley-0.028
 Samantha-0.027
aster-0.027
 seeing-0.027
joy-0.026
Token other
Token ID584
Feature activation-5.16e-02

Pos logit contributions
ags+0.018
 fart+0.017
ac+0.014
am+0.014
apped+0.013
Neg logit contributions
aster-0.033
looking-0.032
 seeing-0.032
 tending-0.031
iley-0.031
Token a
Token ID257
Feature activation-1.83e-02

Pos logit contributions
ags+0.060
 fart+0.058
ac+0.054
 off+0.048
Bang+0.048
Neg logit contributions
iley-0.092
 seeing-0.089
joy-0.089
 tending-0.088
 inspection-0.088
Token hug
Token ID16225
Feature activation-4.54e-03

Pos logit contributions
ags+0.026
ac+0.023
 fart+0.022
efully+0.021
am+0.020
Neg logit contributions
 inspection-0.029
aster-0.029
joy-0.029
looking-0.028
 tending-0.027
Token and
Token ID290
Feature activation-1.72e-02

Pos logit contributions
ags+0.005
 fart+0.005
ac+0.005
 off+0.004
 brothers+0.004
Neg logit contributions
iley-0.008
 seeing-0.008
 inspection-0.008
 tending-0.008
joy-0.008
Token said
Token ID531
Feature activation+1.14e-02

Pos logit contributions
ags+0.022
 fart+0.021
ac+0.020
 off+0.019
 brothers+0.018
Neg logit contributions
iley-0.029
 seeing-0.028
 inspection-0.028
zed-0.028
 tending-0.028
Token,
Token ID11
Feature activation+6.47e-03

Pos logit contributions
iley+0.020
 seeing+0.019
 inspection+0.019
 tending+0.019
joy+0.019
Neg logit contributions
ags-0.014
ac-0.013
 fart-0.013
 off-0.012
 brothers-0.012