TOP ACTIVATIONS
MAX = 7.103

They decided to have a boun cy ball party with the
time , there was a boun cy little bunny called Bob
time there was a f auc et , who wanted to
ball . It was so boun cy and fun to play
time , there was a boun cy rubber ball . It
the deer again . <|endoftext|> â Can I
there was a big , boun cy ball . It was
up and down like a boun cy ball .
round , and ever so boun cy . Every day ,
next time . And don â t be so
a shiny rock and a boun cy ball .
, â said Ali â s mom .
. He was a very boun cy bunny and all day
, there was a small boun cy thing . It was
he knew his mom wouldn â t let him
, â said Sarah â s mom .
Dad grabbed it first and boun ed it around the room
then , a big , boun cy ball floated by .
about it but Tim didn â t listen .
door , and to her amaz ement , she could see

MIN ACTIVATIONS
MIN = -5.886

park they found the biggest ear Tom had ever seen !
always looked like a tornado had gone through it . One
I like the w igg ly dance the most !" So
It was bright pink and had lots of spark les .
Finally , the ear had counted all the way
was a ranch . It had been used by many animals
big do ggy . The do ggy bark ed and Tommy
interesting things about the unknown ear . He still visits it
that . Whenever the chip m unk needed help , Sam
a big house . It had lots of surprises inside .
" Let 's replace the ear so your toy cat can
some are blue , some have flowers , some have dots
back and admired the spark ly space . The
some have flowers , some have dots . She likes to
box full of beautiful or nam ents .
other was sleepy . The alert human said , " Let
room . He saw the lonely machine and had an idea
counted that high ! The ear knew it was getting stronger
finally understood why the chip m unk had been so boss
had pictures , some books had stories , some books had

INTERVAL 3.856 - 7.103
CONTAINS 0.106%

replied , " Sure , but first you need to yield
never take something that doesn â t belong to
said , " Okay , but don 't forget to put
, â said Ali â s mom .
today , they have to un pack . Their family moved

INTERVAL 0.608 - 3.856
CONTAINS 26.346%

bed . The gem was shiny and blue . Kitty showed
<|endoftext|> Once upon a time , there was a rich man
M ummy and Daddy took Peter on a special adventure .
knew her dad was right . She still loved to run
and sw am to the pearl . It was hard ,

INTERVAL -2.639 - 0.608
CONTAINS 72.680%

were brother and sister . They liked to play in the
way home , they got into a car accident . Sarah
Mark think that is a good name . "
if we find it ." The boy smiled and
there was a pretty girl . She had a massage every

INTERVAL -5.886 - -2.639
CONTAINS 0.868%

much happier and all the little animals came to play around
trains and planes that he could zoom around the floor .
as fast as they could away from the weird thing drinking
by giving her a big hug and saying " w elcome
the street . The van has many colors and pictures on
TokenThey
Token ID2990
Feature activation-6.07e-01

Pos logit contributions
 Blood+0.027
 arrest+0.027
 robbers+0.026
 blown+0.025
growth+0.025
Neg logit contributions
Deal-0.033
 difference-0.032
ement-0.031
 decision-0.030
 isn-0.030
Token decided
Token ID3066
Feature activation+1.08e+00

Pos logit contributions
 Blood+0.012
 arrest+0.012
 robbers+0.012
 blown+0.012
growth+0.012
Neg logit contributions
Deal-0.018
 difference-0.017
ement-0.016
 decision-0.016
 isn-0.016
Token to
Token ID284
Feature activation+1.49e+00

Pos logit contributions
Deal+0.043
ement+0.040
 difference+0.039
 promise+0.036
 decision+0.036
Neg logit contributions
 arrest-0.014
 blown-0.014
 robbers-0.013
 Blood-0.013
 spilled-0.012
Token have
Token ID423
Feature activation-1.28e-01

Pos logit contributions
Deal+0.044
ement+0.039
 difference+0.037
 decision+0.036
 exactly+0.036
Neg logit contributions
 arrest-0.034
 robbers-0.032
growth-0.032
 blown-0.031
 Blood-0.031
Token a
Token ID257
Feature activation+1.68e+00

Pos logit contributions
 Blood+0.002
 arrest+0.002
 robbers+0.002
 blown+0.002
growth+0.002
Neg logit contributions
Deal-0.004
 difference-0.004
ement-0.004
 isn-0.003
 why-0.003
Token boun
Token ID31283
Feature activation+5.92e+00

Pos logit contributions
Deal+0.046
ement+0.042
uct+0.038
 difference+0.038
 isn+0.037
Neg logit contributions
 robbers-0.041
 arrest-0.039
 Blood-0.039
 spilled-0.039
 blown-0.038
Tokency
Token ID948
Feature activation-1.27e+00

Pos logit contributions
Deal+0.191
 difference+0.173
 why+0.168
 isn+0.167
 int+0.165
Neg logit contributions
 arrest-0.101
 Blood-0.096
growth-0.094
 robbers-0.093
 spilled-0.092
Token ball
Token ID2613
Feature activation+1.44e+00

Pos logit contributions
 arrest+0.027
 robbers+0.027
 spilled+0.026
 blown+0.025
 Blood+0.025
Neg logit contributions
Deal-0.039
 difference-0.033
ement-0.033
 why-0.033
 exactly-0.033
Token party
Token ID2151
Feature activation+3.30e-01

Pos logit contributions
Deal+0.039
 difference+0.034
ement+0.033
 isn+0.032
 decision+0.032
Neg logit contributions
 arrest-0.034
 robbers-0.034
 Blood-0.034
 blown-0.034
 mice-0.032
Token with
Token ID351
Feature activation+2.20e-02

Pos logit contributions
Deal+0.009
 difference+0.008
ement+0.008
 isn+0.008
 decision+0.008
Neg logit contributions
 robbers-0.007
 arrest-0.007
 blown-0.007
 mice-0.007
 spilled-0.007
Token the
Token ID262
Feature activation-3.57e-01

Pos logit contributions
Deal+0.001
 difference+0.001
 decision+0.001
ement+0.001
 promise+0.001
Neg logit contributions
 arrest-0.000
 Blood-0.000
 robbers-0.000
 blown-0.000
growth-0.000
Token time
Token ID640
Feature activation+2.40e-01

Pos logit contributions
Deal+0.028
 difference+0.026
ement+0.025
 isn+0.024
 why+0.024
Neg logit contributions
 Blood-0.026
 arrest-0.025
 spilled-0.024
 robbers-0.024
growth-0.024
Token,
Token ID11
Feature activation+1.07e+00

Pos logit contributions
Deal+0.006
 difference+0.005
 decision+0.005
ement+0.004
 promise+0.004
Neg logit contributions
 Blood-0.007
 arrest-0.006
 robbers-0.006
 spilled-0.006
 blown-0.006
Token there
Token ID612
Feature activation+2.83e-02

Pos logit contributions
Deal+0.030
 difference+0.026
ement+0.025
 decision+0.025
 isn+0.025
Neg logit contributions
 Blood-0.025
 arrest-0.024
 robbers-0.024
 blown-0.024
growth-0.023
Token was
Token ID373
Feature activation-2.12e+00

Pos logit contributions
Deal+0.001
 difference+0.001
izes+0.001
 int+0.001
ieved+0.001
Neg logit contributions
 spilled-0.001
 Blood-0.001
growth-0.001
 blown-0.001
 arrest-0.001
Token a
Token ID257
Feature activation+1.39e-01

Pos logit contributions
alks+0.054
 arrest+0.053
plets+0.052
 Blood+0.052
 robbers+0.052
Neg logit contributions
Deal-0.051
 exactly-0.045
 equally-0.043
 why-0.043
 isn-0.042
Token boun
Token ID31283
Feature activation+5.73e+00

Pos logit contributions
Deal+0.005
 difference+0.004
ement+0.004
 isn+0.004
 why+0.004
Neg logit contributions
 Blood-0.003
 arrest-0.002
 robbers-0.002
 blown-0.002
growth-0.002
Tokency
Token ID948
Feature activation-4.88e-01

Pos logit contributions
Deal+0.167
 difference+0.149
ement+0.143
 isn+0.142
 why+0.142
Neg logit contributions
 Blood-0.122
 arrest-0.121
 robbers-0.117
 blown-0.116
growth-0.115
Token little
Token ID1310
Feature activation-9.31e-01

Pos logit contributions
 Blood+0.010
 arrest+0.010
 blown+0.010
 robbers+0.010
growth+0.009
Neg logit contributions
Deal-0.015
 difference-0.013
ement-0.013
 isn-0.013
 decision-0.012
Token bunny
Token ID44915
Feature activation+1.25e+00

Pos logit contributions
 Blood+0.023
growth+0.023
 arrest+0.023
 blown+0.022
alks+0.022
Neg logit contributions
Deal-0.024
 difference-0.022
 promise-0.021
 isn-0.021
 decision-0.020
Token called
Token ID1444
Feature activation-1.87e+00

Pos logit contributions
Deal+0.033
 difference+0.029
 why+0.028
ement+0.028
 exactly+0.028
Neg logit contributions
 Blood-0.030
 arrest-0.030
 blown-0.029
 robbers-0.029
growth-0.029
Token Bob
Token ID5811
Feature activation+1.41e+00

Pos logit contributions
 Blood+0.044
 blown+0.044
 arrest+0.042
 spilled+0.041
alks+0.041
Neg logit contributions
Deal-0.049
 difference-0.043
 why-0.043
 isn-0.043
 exactly-0.042
Token time
Token ID640
Feature activation+5.62e-01

Pos logit contributions
Deal+0.035
 difference+0.032
ement+0.031
 isn+0.030
 why+0.030
Neg logit contributions
 Blood-0.030
 arrest-0.028
 spilled-0.027
growth-0.027
 robbers-0.026
Token there
Token ID612
Feature activation+1.41e-01

Pos logit contributions
Deal+0.014
 difference+0.012
 decision+0.011
 promise+0.011
ement+0.011
Neg logit contributions
 Blood-0.016
 arrest-0.015
 robbers-0.015
 spilled-0.014
 blown-0.014
Token was
Token ID373
Feature activation-2.03e+00

Pos logit contributions
Deal+0.004
 difference+0.004
izes+0.004
 int+0.004
 decision+0.003
Neg logit contributions
 spilled-0.003
 Blood-0.003
growth-0.003
 blown-0.003
ges-0.003
Token a
Token ID257
Feature activation+3.16e-01

Pos logit contributions
alks+0.053
 arrest+0.052
 robbers+0.051
plets+0.050
 Blood+0.050
Neg logit contributions
Deal-0.050
 exactly-0.042
 why-0.040
 equally-0.040
 isn-0.040
Token f
Token ID277
Feature activation+1.84e+00

Pos logit contributions
Deal+0.010
 difference+0.009
ement+0.009
 isn+0.009
 why+0.009
Neg logit contributions
 Blood-0.006
 arrest-0.006
 robbers-0.006
 blown-0.006
growth-0.005
Tokenauc
Token ID14272
Feature activation+5.71e+00

Pos logit contributions
Deal+0.052
 difference+0.046
ement+0.044
 why+0.043
 isn+0.043
Neg logit contributions
 Blood-0.042
 arrest-0.041
 robbers-0.040
 blown-0.040
growth-0.040
Tokenet
Token ID316
Feature activation+8.76e-01

Pos logit contributions
 difference+0.116
Deal+0.115
 exactly+0.111
ement+0.109
 isn+0.106
Neg logit contributions
 Blood-0.152
 arrest-0.150
 spilled-0.149
 robbers-0.148
 mice-0.145
Token,
Token ID11
Feature activation+1.32e-02

Pos logit contributions
Deal+0.023
 difference+0.020
 isn+0.019
ement+0.019
 decision+0.019
Neg logit contributions
 Blood-0.022
 arrest-0.022
 blown-0.021
 robbers-0.020
growth-0.020
Token who
Token ID508
Feature activation-5.13e-01

Pos logit contributions
Deal+0.000
 exactly+0.000
 why+0.000
 isn+0.000
 difference+0.000
Neg logit contributions
 arrest-0.000
growth-0.000
 blown-0.000
alks-0.000
 robbers-0.000
Token wanted
Token ID2227
Feature activation-2.39e+00

Pos logit contributions
 Blood+0.015
 arrest+0.014
 spilled+0.014
 blown+0.014
 robbers+0.014
Neg logit contributions
Deal-0.011
 difference-0.010
ement-0.010
 exactly-0.009
 decision-0.009
Token to
Token ID284
Feature activation-1.55e-01

Pos logit contributions
 arrest+0.058
 Blood+0.057
 robbers+0.055
plets+0.054
growth+0.054
Neg logit contributions
Deal-0.059
 why-0.056
 difference-0.056
 exactly-0.055
 isn-0.054
Token ball
Token ID2613
Feature activation+1.31e+00

Pos logit contributions
Deal+0.007
 difference+0.006
 decision+0.006
ement+0.006
 isn+0.006
Neg logit contributions
 Blood-0.005
 arrest-0.005
 blown-0.005
growth-0.005
 robbers-0.005
Token.
Token ID13
Feature activation-9.26e-01

Pos logit contributions
Deal+0.031
 difference+0.026
ement+0.025
 isn+0.025
 why+0.025
Neg logit contributions
 Blood-0.035
 arrest-0.035
 robbers-0.034
 blown-0.034
growth-0.034
Token It
Token ID632
Feature activation-1.47e+00

Pos logit contributions
 arrest+0.019
 Blood+0.018
 blown+0.018
 robbers+0.018
 spilled+0.017
Neg logit contributions
Deal-0.028
 difference-0.025
 isn-0.024
ement-0.024
 why-0.024
Token was
Token ID373
Feature activation-1.77e+00

Pos logit contributions
 arrest+0.045
 Blood+0.044
 robbers+0.044
alks+0.043
growth+0.043
Neg logit contributions
Deal-0.030
 isn-0.025
 difference-0.024
 why-0.023
 promise-0.022
Token so
Token ID523
Feature activation-5.98e-01

Pos logit contributions
 Blood+0.040
 arrest+0.040
 robbers+0.039
 blown+0.039
growth+0.038
Neg logit contributions
Deal-0.050
 difference-0.044
ement-0.042
 isn-0.042
 why-0.042
Token boun
Token ID31283
Feature activation+5.69e+00

Pos logit contributions
 arrest+0.013
 Blood+0.013
 robbers+0.013
growth+0.013
 blown+0.013
Neg logit contributions
Deal-0.017
 difference-0.015
 isn-0.015
ement-0.015
 exactly-0.014
Tokency
Token ID948
Feature activation-3.83e-02

Pos logit contributions
Deal+0.168
 difference+0.155
 decision+0.147
 why+0.145
ement+0.145
Neg logit contributions
 Blood-0.118
 arrest-0.114
 blown-0.110
growth-0.109
 robbers-0.108
Token and
Token ID290
Feature activation-7.42e-01

Pos logit contributions
 arrest+0.001
 Blood+0.001
 blown+0.001
 robbers+0.001
 spilled+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
ement-0.001
 decision-0.001
 why-0.001
Token fun
Token ID1257
Feature activation-3.82e-01

Pos logit contributions
 arrest+0.018
 Blood+0.018
growth+0.018
 robbers+0.018
 blown+0.017
Neg logit contributions
Deal-0.020
 difference-0.017
 isn-0.017
 why-0.016
 exactly-0.016
Token to
Token ID284
Feature activation+9.53e-01

Pos logit contributions
 Blood+0.009
 arrest+0.009
 robbers+0.009
 blown+0.009
 mice+0.009
Neg logit contributions
Deal-0.010
 difference-0.009
ement-0.009
 decision-0.008
 promise-0.008
Token play
Token ID711
Feature activation+2.01e-01

Pos logit contributions
Deal+0.026
 difference+0.023
ement+0.022
 isn+0.022
 exactly+0.022
Neg logit contributions
 Blood-0.023
 arrest-0.022
 robbers-0.022
 blown-0.022
growth-0.021
Token time
Token ID640
Feature activation+6.04e-01

Pos logit contributions
Deal+0.028
 difference+0.025
ement+0.024
 why+0.024
 isn+0.024
Neg logit contributions
 Blood-0.026
 arrest-0.024
 spilled-0.024
growth-0.023
 robbers-0.023
Token,
Token ID11
Feature activation+1.24e+00

Pos logit contributions
Deal+0.014
 difference+0.012
 decision+0.011
ement+0.011
 promise+0.011
Neg logit contributions
 Blood-0.018
 arrest-0.017
 robbers-0.016
 spilled-0.016
 blown-0.016
Token there
Token ID612
Feature activation+1.78e-01

Pos logit contributions
Deal+0.034
 difference+0.030
ement+0.029
 decision+0.028
 isn+0.028
Neg logit contributions
 Blood-0.029
 arrest-0.028
 robbers-0.028
 blown-0.027
growth-0.027
Token was
Token ID373
Feature activation-1.95e+00

Pos logit contributions
Deal+0.005
 difference+0.005
izes+0.005
 int+0.005
ieved+0.005
Neg logit contributions
 spilled-0.004
 Blood-0.004
growth-0.003
ges-0.003
 arrest-0.003
Token a
Token ID257
Feature activation+1.28e-01

Pos logit contributions
alks+0.049
 arrest+0.048
 robbers+0.047
 Blood+0.047
plets+0.047
Neg logit contributions
Deal-0.049
 exactly-0.042
 equally-0.041
 why-0.040
 difference-0.040
Token boun
Token ID31283
Feature activation+5.68e+00

Pos logit contributions
Deal+0.004
 difference+0.004
ement+0.004
 isn+0.004
 why+0.004
Neg logit contributions
 Blood-0.002
 arrest-0.002
 robbers-0.002
 blown-0.002
growth-0.002
Tokency
Token ID948
Feature activation-2.00e-01

Pos logit contributions
Deal+0.163
 difference+0.145
ement+0.139
 isn+0.138
 why+0.138
Neg logit contributions
 Blood-0.124
 arrest-0.123
 robbers-0.119
 blown-0.118
growth-0.117
Token rubber
Token ID14239
Feature activation+3.77e-01

Pos logit contributions
 Blood+0.004
 arrest+0.004
 blown+0.004
 robbers+0.004
growth+0.004
Neg logit contributions
Deal-0.006
 difference-0.006
ement-0.005
 isn-0.005
 why-0.005
Token ball
Token ID2613
Feature activation+1.32e+00

Pos logit contributions
Deal+0.011
 difference+0.010
 isn+0.009
 promise+0.009
 why+0.009
Neg logit contributions
 Blood-0.008
 arrest-0.008
growth-0.008
 blown-0.008
alks-0.008
Token.
Token ID13
Feature activation-6.93e-01

Pos logit contributions
Deal+0.031
 difference+0.026
ement+0.025
 isn+0.025
 why+0.025
Neg logit contributions
 Blood-0.036
 arrest-0.035
 robbers-0.035
 blown-0.034
growth-0.034
Token It
Token ID632
Feature activation-1.25e+00

Pos logit contributions
 Blood+0.014
 arrest+0.013
 robbers+0.013
 blown+0.013
growth+0.013
Neg logit contributions
Deal-0.022
 difference-0.020
ement-0.019
 isn-0.019
 decision-0.019
Token the
Token ID262
Feature activation-1.90e-01

Pos logit contributions
Deal+0.028
 difference+0.023
ement+0.023
 decision+0.023
 promise+0.023
Neg logit contributions
 arrest-0.021
 mice-0.020
 blown-0.020
growth-0.020
 robbers-0.020
Token deer
Token ID20096
Feature activation-6.87e-01

Pos logit contributions
 robbers+0.003
 mice+0.003
 arrest+0.003
 blown+0.003
 bushes+0.003
Neg logit contributions
Deal-0.007
ieved-0.006
ement-0.006
 why-0.006
 exactly-0.006
Token again
Token ID757
Feature activation-2.57e-02

Pos logit contributions
 arrest+0.016
 blown+0.016
 robbers+0.016
 mice+0.016
 spilled+0.015
Neg logit contributions
Deal-0.018
ement-0.015
 difference-0.014
 isn-0.014
uct-0.014
Token.
Token ID13
Feature activation+5.55e-01

Pos logit contributions
 arrest+0.001
 Blood+0.001
 robbers+0.001
growth+0.001
 blown+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
ement-0.001
 isn-0.001
 decision-0.001
Token<|endoftext|>
Token ID50256
Feature activation+5.01e-01

Pos logit contributions
Deal+0.016
 difference+0.014
 decision+0.014
ement+0.014
 promise+0.014
Neg logit contributions
 Blood-0.013
growth-0.012
 arrest-0.012
 robbers-0.012
alks-0.011
Tokenâ
Token ID22940
Feature activation+5.61e+00

Pos logit contributions
Deal+0.014
 difference+0.012
ement+0.012
 decision+0.012
 why+0.012
Neg logit contributions
 Blood-0.012
 arrest-0.011
 robbers-0.011
 blown-0.011
growth-0.011
Token
Token ID26391
Feature activation+4.01e-01

Pos logit contributions
Deal+0.138
 exactly+0.133
 difference+0.131
 decision+0.127
 why+0.125
Neg logit contributions
 Blood-0.134
growth-0.131
 arrest-0.129
 blown-0.126
 robbers-0.123
Token
Token ID129
Feature activation+3.08e+00

Pos logit contributions
Deal+0.012
 difference+0.011
 decision+0.011
ement+0.010
 int+0.010
Neg logit contributions
 Blood-0.009
 arrest-0.009
growth-0.008
 robbers-0.008
alks-0.008
Token
Token ID241
Feature activation+1.56e+00

Pos logit contributions
Deal+0.083
 difference+0.078
 decision+0.075
 choice+0.074
 exactly+0.071
Neg logit contributions
 arrest-0.080
 Blood-0.076
growth-0.072
 blown-0.068
 robbers-0.068
TokenCan
Token ID6090
Feature activation-1.17e-01

Pos logit contributions
Deal+0.043
 difference+0.041
ement+0.039
 decision+0.038
 promise+0.037
Neg logit contributions
 Blood-0.036
 arrest-0.033
growth-0.033
 blown-0.033
 robbers-0.032
Token I
Token ID314
Feature activation-9.58e-01

Pos logit contributions
 Blood+0.003
 arrest+0.003
 robbers+0.003
 blown+0.003
growth+0.002
Neg logit contributions
Deal-0.003
 difference-0.003
ement-0.003
 decision-0.003
 exactly-0.003
Token there
Token ID612
Feature activation+3.05e-01

Pos logit contributions
Deal+0.035
 difference+0.031
ement+0.030
 isn+0.029
 why+0.029
Neg logit contributions
 Blood-0.032
 arrest-0.031
 robbers-0.031
 blown-0.030
growth-0.030
Token was
Token ID373
Feature activation-1.76e+00

Pos logit contributions
Deal+0.009
 difference+0.008
izes+0.007
 int+0.007
 decision+0.007
Neg logit contributions
 spilled-0.007
 Blood-0.007
growth-0.006
 arrest-0.006
 blown-0.006
Token a
Token ID257
Feature activation+1.62e-01

Pos logit contributions
alks+0.043
 arrest+0.042
plets+0.042
 robbers+0.042
 Blood+0.042
Neg logit contributions
Deal-0.044
 exactly-0.039
 why-0.038
 difference-0.038
 equally-0.037
Token big
Token ID1263
Feature activation+1.84e+00

Pos logit contributions
Deal+0.005
 difference+0.005
ement+0.004
 isn+0.004
 why+0.004
Neg logit contributions
 Blood-0.003
 arrest-0.003
 blown-0.003
 robbers-0.003
growth-0.003
Token,
Token ID11
Feature activation-1.10e+00

Pos logit contributions
Deal+0.057
ement+0.050
ieved+0.049
 exactly+0.049
 difference+0.049
Neg logit contributions
 robbers-0.037
 arrest-0.036
 mice-0.036
 Blood-0.035
 spilled-0.034
Token boun
Token ID31283
Feature activation+5.60e+00

Pos logit contributions
 arrest+0.025
growth+0.025
 Blood+0.024
 blown+0.024
alks+0.024
Neg logit contributions
Deal-0.031
 difference-0.027
 isn-0.026
 promise-0.026
ement-0.026
Tokency
Token ID948
Feature activation-2.33e-01

Pos logit contributions
Deal+0.159
 difference+0.142
ement+0.136
 isn+0.135
 why+0.135
Neg logit contributions
 Blood-0.124
 arrest-0.123
 robbers-0.119
 blown-0.118
growth-0.116
Token ball
Token ID2613
Feature activation+1.25e+00

Pos logit contributions
 arrest+0.005
 Blood+0.005
 robbers+0.004
 blown+0.004
 mice+0.004
Neg logit contributions
Deal-0.007
 difference-0.006
ement-0.006
 why-0.006
 exactly-0.006
Token.
Token ID13
Feature activation-7.02e-01

Pos logit contributions
Deal+0.029
 difference+0.024
ement+0.023
 isn+0.023
 why+0.023
Neg logit contributions
 Blood-0.035
 arrest-0.034
 robbers-0.034
 blown-0.034
growth-0.033
Token It
Token ID632
Feature activation-1.10e+00

Pos logit contributions
 arrest+0.014
 Blood+0.014
 blown+0.014
 robbers+0.013
growth+0.013
Neg logit contributions
Deal-0.021
 difference-0.019
 isn-0.019
ement-0.018
 why-0.018
Token was
Token ID373
Feature activation-1.56e+00

Pos logit contributions
 arrest+0.032
 Blood+0.032
 robbers+0.032
 blown+0.031
growth+0.031
Neg logit contributions
Deal-0.023
 difference-0.019
 isn-0.019
ement-0.018
 why-0.018
Token up
Token ID510
Feature activation-1.00e+00

Pos logit contributions
 Blood+0.052
 arrest+0.051
 robbers+0.050
growth+0.050
 blown+0.050
Neg logit contributions
Deal-0.029
 difference-0.024
ement-0.022
 isn-0.022
 why-0.022
Token and
Token ID290
Feature activation+1.11e+00

Pos logit contributions
 Blood+0.024
 arrest+0.024
 robbers+0.024
growth+0.023
 blown+0.023
Neg logit contributions
Deal-0.026
 difference-0.023
ement-0.022
 exactly-0.022
 isn-0.022
Token down
Token ID866
Feature activation-1.62e+00

Pos logit contributions
Deal+0.033
 difference+0.030
ement+0.030
 decision+0.028
 promise+0.028
Neg logit contributions
 Blood-0.025
 blown-0.024
 spilled-0.024
 robbers-0.024
 arrest-0.023
Token like
Token ID588
Feature activation-2.02e-01

Pos logit contributions
 Blood+0.042
 arrest+0.042
 robbers+0.041
 blown+0.041
growth+0.040
Neg logit contributions
Deal-0.040
 difference-0.035
ement-0.033
 isn-0.033
 why-0.032
Token a
Token ID257
Feature activation+6.60e-01

Pos logit contributions
 Blood+0.005
 arrest+0.005
alks+0.005
growth+0.005
 blown+0.005
Neg logit contributions
Deal-0.005
 why-0.005
 exactly-0.004
 difference-0.004
 isn-0.004
Token boun
Token ID31283
Feature activation+5.60e+00

Pos logit contributions
Deal+0.021
ement+0.019
ieved+0.019
uct+0.019
 why+0.018
Neg logit contributions
 robbers-0.012
 spilled-0.012
 Blood-0.012
 arrest-0.012
 blown-0.012
Tokency
Token ID948
Feature activation-1.81e+00

Pos logit contributions
Deal+0.169
 difference+0.153
 why+0.147
ement+0.146
 isn+0.146
Neg logit contributions
 Blood-0.111
 arrest-0.110
 robbers-0.104
 blown-0.104
growth-0.104
Token ball
Token ID2613
Feature activation+5.29e-01

Pos logit contributions
 Blood+0.039
 arrest+0.039
 blown+0.038
 robbers+0.038
 spilled+0.037
Neg logit contributions
Deal-0.053
 difference-0.047
ement-0.045
 why-0.045
 exactly-0.045
Token.
Token ID13
Feature activation+4.60e-01

Pos logit contributions
Deal+0.012
 difference+0.011
 isn+0.010
ement+0.010
 why+0.010
Neg logit contributions
 Blood-0.014
 arrest-0.014
 robbers-0.014
growth-0.014
 blown-0.014
Token
Token ID220
Feature activation-1.05e+00

Pos logit contributions
Deal+0.014
 difference+0.013
 decision+0.012
 promise+0.012
ement+0.012
Neg logit contributions
 Blood-0.011
 arrest-0.009
 robbers-0.009
 spilled-0.009
alks-0.009
Token
Token ID198
Feature activation-1.65e-01

Pos logit contributions
 Blood+0.025
 arrest+0.022
 blown+0.022
alks+0.022
 robbers+0.022
Neg logit contributions
Deal-0.030
 difference-0.026
 decision-0.025
ement-0.025
 promise-0.024
Token round
Token ID2835
Feature activation-7.07e-01

Pos logit contributions
 Blood+0.021
 arrest+0.021
 robbers+0.020
 blown+0.020
growth+0.020
Neg logit contributions
Deal-0.028
 difference-0.025
ement-0.024
 isn-0.024
 why-0.024
Token,
Token ID11
Feature activation-1.09e+00

Pos logit contributions
 arrest+0.022
 Blood+0.021
 robbers+0.021
growth+0.021
 blown+0.021
Neg logit contributions
Deal-0.014
 isn-0.012
ement-0.012
 why-0.012
 difference-0.012
Token and
Token ID290
Feature activation-6.60e-01

Pos logit contributions
 Blood+0.027
 arrest+0.025
 blown+0.025
 robbers+0.025
 spilled+0.025
Neg logit contributions
Deal-0.029
 difference-0.026
ement-0.025
 decision-0.025
 promise-0.024
Token ever
Token ID1683
Feature activation-4.30e-01

Pos logit contributions
 Blood+0.017
 arrest+0.017
 robbers+0.016
 blown+0.016
growth+0.016
Neg logit contributions
Deal-0.016
 difference-0.014
 isn-0.014
ement-0.014
 why-0.014
Token so
Token ID523
Feature activation+5.84e-01

Pos logit contributions
 Blood+0.010
 blown+0.010
 robbers+0.010
 spilled+0.010
 arrest+0.010
Neg logit contributions
Deal-0.012
 difference-0.010
ement-0.010
 decision-0.010
 exactly-0.010
Token boun
Token ID31283
Feature activation+5.59e+00

Pos logit contributions
Deal+0.015
 difference+0.014
ement+0.013
 decision+0.013
 why+0.013
Neg logit contributions
 Blood-0.014
 arrest-0.014
 blown-0.014
 robbers-0.014
 mice-0.013
Tokency
Token ID948
Feature activation-2.80e-01

Pos logit contributions
Deal+0.169
 difference+0.156
 decision+0.148
 why+0.147
ement+0.146
Neg logit contributions
 Blood-0.112
 arrest-0.107
growth-0.104
 blown-0.104
 spilled-0.102
Token.
Token ID13
Feature activation+1.89e-02

Pos logit contributions
 Blood+0.007
 arrest+0.007
 robbers+0.007
 blown+0.007
growth+0.007
Neg logit contributions
Deal-0.007
 difference-0.006
ement-0.006
 isn-0.006
 why-0.006
Token Every
Token ID3887
Feature activation+1.47e+00

Pos logit contributions
Deal+0.001
 difference+0.001
 decision+0.001
 exactly+0.001
 promise+0.001
Neg logit contributions
 Blood-0.000
 arrest-0.000
growth-0.000
 robbers-0.000
 mice-0.000
Token day
Token ID1110
Feature activation-2.24e+00

Pos logit contributions
Deal+0.037
ement+0.032
 difference+0.032
 exactly+0.031
ieved+0.030
Neg logit contributions
 Blood-0.038
 spilled-0.036
growth-0.036
 blown-0.036
 arrest-0.035
Token,
Token ID11
Feature activation-5.67e-01

Pos logit contributions
 Blood+0.073
 arrest+0.072
 robbers+0.071
 blown+0.070
growth+0.070
Neg logit contributions
Deal-0.041
 difference-0.035
ement-0.032
 isn-0.032
 why-0.031
Token next
Token ID1306
Feature activation+2.14e+00

Pos logit contributions
Deal+0.017
 difference+0.015
ement+0.014
 exactly+0.013
ieved+0.013
Neg logit contributions
 arrest-0.014
 blown-0.014
 Blood-0.014
 robbers-0.014
 mice-0.014
Token time
Token ID640
Feature activation-1.74e+00

Pos logit contributions
Deal+0.062
ement+0.052
 difference+0.052
ieved+0.051
 exactly+0.050
Neg logit contributions
 Blood-0.046
 blown-0.046
 arrest-0.046
 mice-0.044
 robbers-0.044
Token.
Token ID13
Feature activation+7.62e-01

Pos logit contributions
 arrest+0.043
 Blood+0.042
 robbers+0.042
 blown+0.042
 spilled+0.040
Neg logit contributions
Deal-0.046
 difference-0.040
ement-0.039
 decision-0.037
 exactly-0.036
Token And
Token ID843
Feature activation+1.54e+00

Pos logit contributions
Deal+0.021
 decision+0.019
 difference+0.019
ement+0.018
 exactly+0.018
Neg logit contributions
 Blood-0.018
 robbers-0.017
 arrest-0.017
growth-0.017
 mice-0.016
Token don
Token ID836
Feature activation+3.64e-03

Pos logit contributions
ement+0.041
Deal+0.041
 difference+0.038
ieved+0.036
 decision+0.035
Neg logit contributions
 arrest-0.034
 robbers-0.033
 mice-0.033
 Blood-0.031
 blown-0.031
Tokenâ
Token ID22940
Feature activation+5.58e+00

Pos logit contributions
Deal+0.000
 difference+0.000
ement+0.000
 decision+0.000
 exactly+0.000
Neg logit contributions
 Blood-0.000
 arrest-0.000
 blown-0.000
 robbers-0.000
growth-0.000
Token
Token ID26391
Feature activation+1.13e-01

Pos logit contributions
 exactly+0.102
 amaz+0.095
 difference+0.095
 isn+0.093
 why+0.091
Neg logit contributions
 Blood-0.151
 arrest-0.149
 blown-0.146
growth-0.142
 robbers-0.141
Token
Token ID8151
Feature activation+1.07e+00

Pos logit contributions
Deal+0.003
 difference+0.003
 decision+0.003
ement+0.003
 int+0.003
Neg logit contributions
 Blood-0.003
 arrest-0.003
alks-0.002
 robbers-0.002
plets-0.002
Tokent
Token ID83
Feature activation+3.94e-01

Pos logit contributions
Deal+0.034
 difference+0.032
 decision+0.030
 why+0.029
ement+0.029
Neg logit contributions
 Blood-0.019
 arrest-0.018
 spilled-0.017
 blown-0.017
alks-0.017
Token be
Token ID307
Feature activation+7.30e-01

Pos logit contributions
Deal+0.011
ement+0.010
 difference+0.009
 exactly+0.009
 decision+0.009
Neg logit contributions
 Blood-0.009
 arrest-0.009
 robbers-0.009
 blown-0.009
 spilled-0.008
Token so
Token ID523
Feature activation+1.73e+00

Pos logit contributions
Deal+0.018
 difference+0.015
ement+0.015
 decision+0.014
 isn+0.014
Neg logit contributions
 Blood-0.020
 blown-0.020
 arrest-0.019
 robbers-0.019
 mice-0.019
Token a
Token ID257
Feature activation+6.16e-01

Pos logit contributions
 Blood+0.024
 arrest+0.024
 blown+0.023
 robbers+0.023
growth+0.022
Neg logit contributions
Deal-0.037
 difference-0.033
 isn-0.032
 why-0.032
 exactly-0.032
Token shiny
Token ID22441
Feature activation+8.26e-01

Pos logit contributions
Deal+0.020
uct+0.018
ieved+0.018
ement+0.018
 why+0.017
Neg logit contributions
 robbers-0.012
 spilled-0.012
 Blood-0.012
 arrest-0.011
 mice-0.011
Token rock
Token ID3881
Feature activation+7.41e-01

Pos logit contributions
Deal+0.025
uct+0.021
 difference+0.021
ieved+0.021
ement+0.021
Neg logit contributions
 robbers-0.017
 Blood-0.017
 spilled-0.017
 arrest-0.017
alks-0.016
Token and
Token ID290
Feature activation-9.74e-01

Pos logit contributions
Deal+0.017
 difference+0.015
ement+0.014
 isn+0.014
 why+0.014
Neg logit contributions
 Blood-0.020
 arrest-0.020
 robbers-0.020
 blown-0.019
growth-0.019
Token a
Token ID257
Feature activation-9.52e-02

Pos logit contributions
 Blood+0.022
 arrest+0.021
alks+0.021
 blown+0.020
growth+0.020
Neg logit contributions
Deal-0.027
 why-0.024
 isn-0.024
 exactly-0.024
 difference-0.024
Token boun
Token ID31283
Feature activation+5.58e+00

Pos logit contributions
 robbers+0.002
 Blood+0.002
 spilled+0.002
 arrest+0.002
 blown+0.002
Neg logit contributions
Deal-0.003
ement-0.003
ieved-0.003
uct-0.003
 why-0.003
Tokency
Token ID948
Feature activation-1.64e+00

Pos logit contributions
Deal+0.166
 difference+0.148
 why+0.142
ement+0.142
 exactly+0.140
Neg logit contributions
 Blood-0.115
 arrest-0.115
 robbers-0.110
growth-0.109
 blown-0.109
Token ball
Token ID2613
Feature activation+6.27e-01

Pos logit contributions
 Blood+0.034
 arrest+0.034
 robbers+0.033
 blown+0.032
 spilled+0.032
Neg logit contributions
Deal-0.050
 difference-0.043
ement-0.042
 why-0.042
 exactly-0.041
Token.
Token ID13
Feature activation+5.36e-01

Pos logit contributions
Deal+0.014
 difference+0.013
 isn+0.012
ement+0.012
 why+0.012
Neg logit contributions
 Blood-0.017
 arrest-0.017
 robbers-0.017
growth-0.017
 blown-0.016
Token
Token ID198
Feature activation-4.05e-02

Pos logit contributions
Deal+0.016
 difference+0.014
ement+0.014
 promise+0.014
 decision+0.014
Neg logit contributions
 Blood-0.012
 robbers-0.011
 arrest-0.011
alks-0.010
 spilled-0.010
Token
Token ID198
Feature activation-7.84e-01

Pos logit contributions
 Blood+0.001
 arrest+0.001
growth+0.001
 robbers+0.001
 blown+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
ement-0.001
 decision-0.001
 promise-0.001
Token,
Token ID11
Feature activation+7.10e-01

Pos logit contributions
 arrest+0.002
 mice+0.002
 blown+0.002
 spilled+0.002
alks+0.002
Neg logit contributions
Deal-0.004
 difference-0.003
ement-0.003
ieved-0.003
 int-0.003
Tokenâ
Token ID22940
Feature activation+5.11e+00

Pos logit contributions
Deal+0.021
ement+0.018
 difference+0.017
ieved+0.016
 decision+0.016
Neg logit contributions
 arrest-0.016
 robbers-0.016
 blown-0.015
 spilled-0.015
 mice-0.015
Token
Token ID26391
Feature activation+1.93e-01

Pos logit contributions
 exactly+0.097
 difference+0.094
Deal+0.090
 amaz+0.088
 why+0.087
Neg logit contributions
 Blood-0.145
 arrest-0.136
plets-0.135
 blown-0.132
growth-0.132
Token said
Token ID531
Feature activation+8.87e-01

Pos logit contributions
Deal+0.005
 difference+0.005
ement+0.004
 decision+0.004
 why+0.004
Neg logit contributions
 arrest-0.005
 Blood-0.005
 robbers-0.004
 blown-0.004
growth-0.004
Token Ali
Token ID12104
Feature activation+3.03e+00

Pos logit contributions
Deal+0.026
 difference+0.023
ieved+0.022
ement+0.022
uct+0.020
Neg logit contributions
 Blood-0.018
 arrest-0.017
 spilled-0.017
 blown-0.017
 Po-0.017
Tokenâ
Token ID22940
Feature activation+5.55e+00

Pos logit contributions
Deal+0.085
 difference+0.073
ement+0.071
ieved+0.066
 decision+0.065
Neg logit contributions
 arrest-0.069
 Blood-0.069
 robbers-0.069
 mice-0.067
 blown-0.067
Token
Token ID26391
Feature activation+6.29e-01

Pos logit contributions
 exactly+0.111
 difference+0.106
Deal+0.101
 why+0.099
 amaz+0.099
Neg logit contributions
 Blood-0.151
 robbers-0.144
 arrest-0.142
growth-0.139
 blown-0.137
Token
Token ID8151
Feature activation+2.17e+00

Pos logit contributions
Deal+0.018
 difference+0.017
ement+0.016
 decision+0.016
 why+0.016
Neg logit contributions
 arrest-0.014
 Blood-0.014
 robbers-0.013
growth-0.012
 blown-0.012
Tokens
Token ID82
Feature activation+2.07e+00

Pos logit contributions
Deal+0.070
 difference+0.063
ement+0.061
 isn+0.060
 why+0.060
Neg logit contributions
 Blood-0.040
 arrest-0.039
 robbers-0.037
 blown-0.037
growth-0.036
Token mom
Token ID1995
Feature activation+2.06e-01

Pos logit contributions
Deal+0.052
ement+0.043
ieved+0.042
 difference+0.041
uct+0.040
Neg logit contributions
 Blood-0.055
 arrest-0.055
alks-0.054
 robbers-0.053
 blown-0.053
Token.
Token ID13
Feature activation+1.03e+00

Pos logit contributions
Deal+0.006
 difference+0.005
ement+0.005
 decision+0.005
 deal+0.005
Neg logit contributions
 robbers-0.004
 Blood-0.004
 arrest-0.004
 spilled-0.004
 mice-0.004
Token.
Token ID13
Feature activation-7.05e-01

Pos logit contributions
Deal+0.042
 difference+0.037
 why+0.036
ement+0.036
 isn+0.036
Neg logit contributions
 Blood-0.037
 arrest-0.036
 robbers-0.035
 blown-0.035
growth-0.034
Token He
Token ID679
Feature activation-5.13e-01

Pos logit contributions
 blown+0.017
 arrest+0.017
 spilled+0.016
 robbers+0.016
 Blood+0.016
Neg logit contributions
Deal-0.018
 difference-0.016
 isn-0.016
 why-0.016
ement-0.015
Token was
Token ID373
Feature activation-4.84e-01

Pos logit contributions
 arrest+0.018
alks+0.018
 robbers+0.017
growth+0.017
 Blood+0.017
Neg logit contributions
Deal-0.008
 isn-0.006
 why-0.006
 exactly-0.005
 difference-0.005
Token a
Token ID257
Feature activation-3.61e-01

Pos logit contributions
 arrest+0.016
growth+0.016
 Blood+0.015
 robbers+0.015
alks+0.015
Neg logit contributions
Deal-0.009
 exactly-0.007
 why-0.007
 difference-0.007
 isn-0.006
Token very
Token ID845
Feature activation+9.19e-02

Pos logit contributions
 Blood+0.009
 arrest+0.009
 robbers+0.009
 blown+0.009
 mice+0.008
Neg logit contributions
Deal-0.009
 difference-0.008
ement-0.008
 isn-0.008
 why-0.008
Token boun
Token ID31283
Feature activation+5.53e+00

Pos logit contributions
Deal+0.002
 difference+0.002
 isn+0.002
ement+0.002
 promise+0.002
Neg logit contributions
 arrest-0.003
 Blood-0.003
growth-0.003
 blown-0.003
 robbers-0.002
Tokency
Token ID948
Feature activation-1.45e+00

Pos logit contributions
Deal+0.161
 difference+0.144
ement+0.138
 why+0.138
 isn+0.138
Neg logit contributions
 Blood-0.117
 arrest-0.116
 robbers-0.113
 blown-0.111
growth-0.110
Token bunny
Token ID44915
Feature activation+7.25e-01

Pos logit contributions
 Blood+0.024
 arrest+0.024
 robbers+0.023
 blown+0.023
growth+0.022
Neg logit contributions
Deal-0.049
 difference-0.045
ement-0.043
 why-0.043
 isn-0.043
Token and
Token ID290
Feature activation-5.01e-01

Pos logit contributions
Deal+0.020
 difference+0.018
ement+0.017
 decision+0.017
 isn+0.017
Neg logit contributions
 arrest-0.016
 mice-0.016
 Blood-0.016
 robbers-0.016
 blown-0.016
Token all
Token ID477
Feature activation-6.36e-01

Pos logit contributions
 arrest+0.015
 Blood+0.015
 blown+0.014
 robbers+0.014
growth+0.014
Neg logit contributions
Deal-0.011
 difference-0.009
 isn-0.009
 why-0.009
 exactly-0.008
Token day
Token ID1110
Feature activation-7.36e-01

Pos logit contributions
 Blood+0.017
 arrest+0.016
 blown+0.016
 robbers+0.016
 spilled+0.016
Neg logit contributions
Deal-0.015
 difference-0.014
ement-0.013
 decision-0.013
 exactly-0.013
Token,
Token ID11
Feature activation+1.17e+00

Pos logit contributions
Deal+0.010
 difference+0.009
 decision+0.008
 promise+0.008
ement+0.008
Neg logit contributions
 Blood-0.012
 robbers-0.011
 arrest-0.011
 spilled-0.011
 mice-0.010
Token there
Token ID612
Feature activation+1.28e-01

Pos logit contributions
Deal+0.033
 difference+0.030
ement+0.028
 decision+0.028
 isn+0.028
Neg logit contributions
 Blood-0.027
 arrest-0.026
 robbers-0.025
 blown-0.025
growth-0.025
Token was
Token ID373
Feature activation-1.96e+00

Pos logit contributions
Deal+0.004
 difference+0.003
izes+0.003
 int+0.003
ieved+0.003
Neg logit contributions
 spilled-0.003
 Blood-0.003
growth-0.003
ges-0.002
 blown-0.002
Token a
Token ID257
Feature activation+2.75e-01

Pos logit contributions
alks+0.050
 arrest+0.049
 Blood+0.048
 robbers+0.048
plets+0.048
Neg logit contributions
Deal-0.048
 exactly-0.042
 why-0.040
 equally-0.040
 difference-0.040
Token small
Token ID1402
Feature activation+5.20e-01

Pos logit contributions
Deal+0.009
 difference+0.008
ement+0.008
 isn+0.008
 why+0.008
Neg logit contributions
 Blood-0.005
 arrest-0.005
 robbers-0.005
 blown-0.005
growth-0.005
Token boun
Token ID31283
Feature activation+5.52e+00

Pos logit contributions
Deal+0.016
 difference+0.015
ement+0.014
 isn+0.014
 why+0.014
Neg logit contributions
 Blood-0.010
 arrest-0.010
 blown-0.010
growth-0.010
 robbers-0.010
Tokency
Token ID948
Feature activation-3.75e-01

Pos logit contributions
Deal+0.160
 difference+0.143
ement+0.137
 isn+0.136
 why+0.136
Neg logit contributions
 Blood-0.118
 arrest-0.117
 robbers-0.114
 blown-0.113
growth-0.111
Token thing
Token ID1517
Feature activation+1.77e+00

Pos logit contributions
 Blood+0.008
 arrest+0.007
 robbers+0.007
 blown+0.007
growth+0.007
Neg logit contributions
Deal-0.011
 difference-0.010
ement-0.010
 isn-0.010
 why-0.010
Token.
Token ID13
Feature activation-8.17e-01

Pos logit contributions
Deal+0.046
 difference+0.040
ement+0.038
 isn+0.038
 decision+0.037
Neg logit contributions
 Blood-0.044
 arrest-0.043
 robbers-0.043
 blown-0.042
growth-0.042
Token It
Token ID632
Feature activation-1.21e+00

Pos logit contributions
 Blood+0.018
 arrest+0.018
 blown+0.018
 robbers+0.017
growth+0.017
Neg logit contributions
Deal-0.023
 difference-0.021
 isn-0.020
ement-0.020
 why-0.019
Token was
Token ID373
Feature activation-1.61e+00

Pos logit contributions
 Blood+0.036
 arrest+0.035
 blown+0.034
 robbers+0.034
growth+0.034
Neg logit contributions
Deal-0.026
 difference-0.022
ement-0.021
 decision-0.021
 exactly-0.020
Token he
Token ID339
Feature activation+7.16e-01

Pos logit contributions
Deal+0.071
 difference+0.065
ement+0.061
 decision+0.058
ieved+0.057
Neg logit contributions
 robbers-0.048
 spilled-0.046
 Blood-0.046
 blown-0.045
 arrest-0.044
Token knew
Token ID2993
Feature activation+9.10e-01

Pos logit contributions
Deal+0.019
 difference+0.019
ement+0.017
 decision+0.016
 int+0.016
Neg logit contributions
 spilled-0.017
 Blood-0.016
 blown-0.016
 arrest-0.016
 robbers-0.015
Token his
Token ID465
Feature activation+2.68e+00

Pos logit contributions
Deal+0.025
 difference+0.022
ement+0.021
 promise+0.021
 decision+0.020
Neg logit contributions
 arrest-0.021
 Blood-0.021
 robbers-0.021
 blown-0.020
 spilled-0.020
Token mom
Token ID1995
Feature activation-5.22e-01

Pos logit contributions
Deal+0.084
ieved+0.076
uct+0.073
ement+0.073
izes+0.069
Neg logit contributions
 robbers-0.050
 arrest-0.048
 spilled-0.048
 mice-0.047
 blown-0.046
Token wouldn
Token ID3636
Feature activation+3.77e-01

Pos logit contributions
 spilled+0.013
 blown+0.012
 robbers+0.012
 mice+0.012
 Blood+0.012
Neg logit contributions
Deal-0.013
 difference-0.012
ement-0.011
 decision-0.011
 int-0.011
Tokenâ
Token ID22940
Feature activation+5.52e+00

Pos logit contributions
Deal+0.011
 difference+0.010
 isn+0.010
ement+0.010
 why+0.010
Neg logit contributions
 robbers-0.008
 Blood-0.008
 mice-0.007
 arrest-0.007
growth-0.007
Token
Token ID26391
Feature activation+4.76e-02

Pos logit contributions
Deal+0.098
 difference+0.096
 exactly+0.096
ement+0.089
 isn+0.088
Neg logit contributions
 Blood-0.163
 arrest-0.163
 robbers-0.159
 blown-0.157
growth-0.154
Token
Token ID8151
Feature activation+9.39e-01

Pos logit contributions
Deal+0.001
 difference+0.001
ement+0.001
 promise+0.001
 decision+0.001
Neg logit contributions
 arrest-0.001
 Blood-0.001
 robbers-0.001
 blown-0.001
growth-0.001
Tokent
Token ID83
Feature activation-3.50e-01

Pos logit contributions
Deal+0.032
 difference+0.029
ement+0.029
 isn+0.028
 why+0.028
Neg logit contributions
 Blood-0.015
 arrest-0.015
 robbers-0.014
 blown-0.014
growth-0.014
Token let
Token ID1309
Feature activation-7.86e-01

Pos logit contributions
 Blood+0.008
 arrest+0.008
 robbers+0.008
 blown+0.008
 spilled+0.007
Neg logit contributions
Deal-0.010
ement-0.009
 difference-0.008
 isn-0.008
 exactly-0.008
Token him
Token ID683
Feature activation-1.73e-01

Pos logit contributions
 arrest+0.019
 Blood+0.018
 robbers+0.018
 blown+0.018
 spilled+0.017
Neg logit contributions
Deal-0.022
ement-0.019
 difference-0.018
ieved-0.017
 exactly-0.017
Token,
Token ID11
Feature activation+7.25e-01

Pos logit contributions
Deal+0.014
 difference+0.012
ement+0.012
ieved+0.011
 decision+0.011
Neg logit contributions
 arrest-0.016
 Blood-0.016
 spilled-0.016
 blown-0.016
 robbers-0.015
Tokenâ
Token ID22940
Feature activation+4.87e+00

Pos logit contributions
Deal+0.021
 difference+0.019
ement+0.019
 decision+0.017
ieved+0.017
Neg logit contributions
 arrest-0.016
 robbers-0.015
 spilled-0.015
 Blood-0.015
 blown-0.015
Token
Token ID26391
Feature activation-7.88e-01

Pos logit contributions
 exactly+0.094
 difference+0.091
Deal+0.085
 why+0.083
 amaz+0.083
Neg logit contributions
 Blood-0.138
 arrest-0.133
plets-0.131
 blown-0.128
alks-0.127
Token said
Token ID531
Feature activation+7.42e-01

Pos logit contributions
 Blood+0.020
 arrest+0.019
alks+0.018
 robbers+0.018
 blown+0.018
Neg logit contributions
Deal-0.021
 difference-0.019
ement-0.018
 decision-0.018
 why-0.017
Token Sarah
Token ID10490
Feature activation+1.56e+00

Pos logit contributions
Deal+0.020
 difference+0.018
ement+0.017
 decision+0.017
ieved+0.016
Neg logit contributions
 Blood-0.018
 arrest-0.017
 spilled-0.017
 blown-0.016
 robbers-0.016
Tokenâ
Token ID22940
Feature activation+5.47e+00

Pos logit contributions
Deal+0.039
 difference+0.036
ement+0.033
 decision+0.032
 deal+0.032
Neg logit contributions
 Blood-0.039
 spilled-0.038
 arrest-0.038
 blown-0.037
 robbers-0.037
Token
Token ID26391
Feature activation-2.54e-01

Pos logit contributions
 exactly+0.110
 difference+0.105
 why+0.097
 amaz+0.097
Deal+0.094
Neg logit contributions
 Blood-0.149
plets-0.140
 arrest-0.140
 robbers-0.139
alks-0.137
Token
Token ID8151
Feature activation+1.52e+00

Pos logit contributions
 Blood+0.006
 arrest+0.006
 robbers+0.005
alks+0.005
plets+0.005
Neg logit contributions
Deal-0.007
 difference-0.007
ement-0.006
 decision-0.006
 exactly-0.006
Tokens
Token ID82
Feature activation+2.71e+00

Pos logit contributions
Deal+0.052
 difference+0.049
 decision+0.046
 why+0.046
 promise+0.046
Neg logit contributions
 Blood-0.025
 arrest-0.023
 spilled-0.022
alks-0.022
 blown-0.022
Token mom
Token ID1995
Feature activation-1.83e-01

Pos logit contributions
Deal+0.070
 difference+0.058
ement+0.057
ieved+0.056
uct+0.053
Neg logit contributions
 Blood-0.072
 arrest-0.069
alks-0.069
 spilled-0.067
 robbers-0.066
Token.
Token ID13
Feature activation+7.74e-01

Pos logit contributions
 spilled+0.004
 robbers+0.004
 Blood+0.004
 arrest+0.004
 mice+0.004
Neg logit contributions
Deal-0.005
 difference-0.004
ement-0.004
 decision-0.004
 int-0.004
Token Dad
Token ID17415
Feature activation+8.67e-01

Pos logit contributions
 arrest+0.005
 Blood+0.005
 robbers+0.005
 blown+0.005
growth+0.005
Neg logit contributions
Deal-0.006
 difference-0.006
 isn-0.005
ement-0.005
 why-0.005
Token grabbed
Token ID13176
Feature activation+2.39e-01

Pos logit contributions
Deal+0.026
 difference+0.024
 decision+0.022
ement+0.022
 exactly+0.021
Neg logit contributions
 spilled-0.018
 robbers-0.018
 Blood-0.018
 blown-0.018
 arrest-0.018
Token it
Token ID340
Feature activation+6.73e-01

Pos logit contributions
Deal+0.007
 difference+0.006
ement+0.006
 isn+0.006
 decision+0.005
Neg logit contributions
 Blood-0.006
 arrest-0.005
 robbers-0.005
 blown-0.005
growth-0.005
Token first
Token ID717
Feature activation+1.22e+00

Pos logit contributions
Deal+0.023
 difference+0.020
ement+0.019
 isn+0.019
 decision+0.018
Neg logit contributions
 blown-0.012
 robbers-0.012
 arrest-0.012
 spilled-0.011
 mice-0.011
Token and
Token ID290
Feature activation+7.60e-01

Pos logit contributions
Deal+0.040
 difference+0.035
ement+0.032
ieved+0.032
uct+0.032
Neg logit contributions
 spilled-0.023
 arrest-0.022
 blown-0.022
 Blood-0.022
alks-0.021
Token boun
Token ID31283
Feature activation+5.46e+00

Pos logit contributions
Deal+0.023
 difference+0.021
ement+0.020
 decision+0.020
 isn+0.019
Neg logit contributions
 blown-0.016
 arrest-0.015
 robbers-0.015
 Blood-0.015
 spilled-0.015
Tokened
Token ID276
Feature activation+7.98e-01

Pos logit contributions
Deal+0.179
 difference+0.168
 decision+0.157
 why+0.156
 exactly+0.154
Neg logit contributions
 Blood-0.094
 arrest-0.091
growth-0.088
 spilled-0.087
ges-0.087
Token it
Token ID340
Feature activation+5.18e-01

Pos logit contributions
Deal+0.026
 difference+0.023
 decision+0.021
 promise+0.021
ement+0.021
Neg logit contributions
 arrest-0.017
 mice-0.016
ges-0.016
 Blood-0.016
growth-0.016
Token around
Token ID1088
Feature activation+1.78e+00

Pos logit contributions
Deal+0.017
 difference+0.015
ement+0.014
 decision+0.014
 promise+0.014
Neg logit contributions
 blown-0.010
 spilled-0.010
 Blood-0.010
 arrest-0.009
 robbers-0.009
Token the
Token ID262
Feature activation-4.37e-01

Pos logit contributions
Deal+0.051
 difference+0.045
ement+0.043
 decision+0.042
 isn+0.042
Neg logit contributions
 Blood-0.041
 arrest-0.040
 robbers-0.039
 blown-0.039
growth-0.038
Token room
Token ID2119
Feature activation+6.64e-01

Pos logit contributions
 Blood+0.011
 arrest+0.011
 robbers+0.011
 blown+0.011
growth+0.011
Neg logit contributions
Deal-0.011
 difference-0.010
ement-0.009
 isn-0.009
 why-0.009
Token then
Token ID788
Feature activation-1.68e-01

Pos logit contributions
Deal+0.035
 difference+0.032
ement+0.030
 decision+0.029
 isn+0.029
Neg logit contributions
 Blood-0.043
 arrest-0.043
 robbers-0.043
 mice-0.042
 blown-0.042
Token,
Token ID11
Feature activation+1.33e+00

Pos logit contributions
 Blood+0.004
 arrest+0.004
 robbers+0.004
 blown+0.004
growth+0.004
Neg logit contributions
Deal-0.004
 difference-0.004
ement-0.003
 decision-0.003
 isn-0.003
Token a
Token ID257
Feature activation+1.08e+00

Pos logit contributions
Deal+0.038
 difference+0.034
 why+0.033
 exactly+0.033
 isn+0.033
Neg logit contributions
 Blood-0.029
 arrest-0.028
 blown-0.027
 robbers-0.027
alks-0.027
Token big
Token ID1263
Feature activation+2.36e+00

Pos logit contributions
Deal+0.031
 difference+0.028
ement+0.026
 isn+0.026
 why+0.026
Neg logit contributions
 Blood-0.024
 arrest-0.024
 blown-0.023
 robbers-0.023
growth-0.023
Token,
Token ID11
Feature activation+3.16e-01

Pos logit contributions
Deal+0.071
 difference+0.064
ement+0.062
 isn+0.061
 why+0.061
Neg logit contributions
 Blood-0.048
 arrest-0.047
 robbers-0.046
 blown-0.046
growth-0.045
Token boun
Token ID31283
Feature activation+5.43e+00

Pos logit contributions
Deal+0.008
 difference+0.008
 isn+0.007
 promise+0.007
 why+0.007
Neg logit contributions
 arrest-0.008
 Blood-0.008
growth-0.008
alks-0.007
 blown-0.007
Tokency
Token ID948
Feature activation-3.00e-01

Pos logit contributions
Deal+0.153
 difference+0.137
ement+0.131
 isn+0.130
 why+0.129
Neg logit contributions
 Blood-0.121
 arrest-0.120
 robbers-0.117
 blown-0.116
growth-0.114
Token ball
Token ID2613
Feature activation+1.10e+00

Pos logit contributions
 Blood+0.006
 arrest+0.006
 robbers+0.006
 blown+0.006
growth+0.006
Neg logit contributions
Deal-0.009
 difference-0.008
ement-0.008
 isn-0.008
 decision-0.008
Token floated
Token ID29742
Feature activation+1.26e-01

Pos logit contributions
Deal+0.028
 difference+0.024
 isn+0.024
 why+0.024
 exactly+0.024
Neg logit contributions
 Blood-0.027
growth-0.026
alks-0.026
 arrest-0.026
 robbers-0.025
Token by
Token ID416
Feature activation+8.79e-01

Pos logit contributions
Deal+0.003
 difference+0.003
ement+0.003
 decision+0.003
 promise+0.003
Neg logit contributions
 Blood-0.003
 blown-0.003
 robbers-0.003
 arrest-0.003
 spilled-0.003
Token.
Token ID13
Feature activation-1.08e-01

Pos logit contributions
Deal+0.022
 difference+0.019
ement+0.018
 isn+0.018
 why+0.018
Neg logit contributions
 Blood-0.023
 arrest-0.023
 robbers-0.022
 blown-0.022
growth-0.022
Token about
Token ID546
Feature activation+1.36e+00

Pos logit contributions
 Blood+0.015
 blown+0.014
 arrest+0.014
 spilled+0.014
 robbers+0.014
Neg logit contributions
Deal-0.013
 difference-0.011
 promise-0.010
 decision-0.010
ement-0.010
Token it
Token ID340
Feature activation+1.46e-01

Pos logit contributions
Deal+0.031
ement+0.025
 difference+0.025
ieved+0.024
 promise+0.024
Neg logit contributions
 arrest-0.039
 robbers-0.038
 Blood-0.038
 mice-0.037
 blown-0.036
Token but
Token ID475
Feature activation+3.26e+00

Pos logit contributions
Deal+0.005
 difference+0.004
ement+0.004
 decision+0.004
aman+0.004
Neg logit contributions
 robbers-0.003
 mice-0.003
 Blood-0.003
 spilled-0.003
 blown-0.003
Token Tim
Token ID5045
Feature activation+2.16e+00

Pos logit contributions
Deal+0.104
 difference+0.099
ement+0.096
 decision+0.089
 promise+0.088
Neg logit contributions
 Blood-0.061
 robbers-0.061
 blown-0.058
 spilled-0.057
 arrest-0.056
Token didn
Token ID1422
Feature activation-6.60e-01

Pos logit contributions
 difference+0.054
Deal+0.052
ement+0.048
 int+0.046
 amaz+0.046
Neg logit contributions
 spilled-0.054
 Blood-0.051
 blown-0.050
 bitten-0.049
 robbers-0.048
Tokenâ
Token ID22940
Feature activation+5.42e+00

Pos logit contributions
 Blood+0.011
 arrest+0.011
 robbers+0.010
 blown+0.010
growth+0.010
Neg logit contributions
Deal-0.022
 difference-0.020
ement-0.020
 isn-0.020
 why-0.020
Token
Token ID26391
Feature activation+1.02e-01

Pos logit contributions
Deal+0.087
 difference+0.080
 exactly+0.075
ement+0.073
 why+0.072
Neg logit contributions
 Blood-0.180
 arrest-0.176
 robbers-0.172
 blown-0.171
growth-0.171
Token
Token ID8151
Feature activation+1.04e+00

Pos logit contributions
Deal+0.003
 difference+0.003
ement+0.003
 decision+0.003
 promise+0.003
Neg logit contributions
 Blood-0.002
 arrest-0.002
 robbers-0.002
 blown-0.002
alks-0.002
Tokent
Token ID83
Feature activation+3.89e-01

Pos logit contributions
Deal+0.036
 difference+0.033
ement+0.032
 isn+0.032
 why+0.032
Neg logit contributions
 Blood-0.016
 arrest-0.016
 robbers-0.015
 blown-0.015
growth-0.015
Token listen
Token ID6004
Feature activation+2.01e+00

Pos logit contributions
Deal+0.010
ement+0.009
 difference+0.008
 isn+0.008
 exactly+0.008
Neg logit contributions
 Blood-0.010
 robbers-0.010
 arrest-0.010
 blown-0.010
 spilled-0.009
Token.
Token ID13
Feature activation+8.19e-01

Pos logit contributions
Deal+0.070
 difference+0.064
ement+0.058
 isn+0.058
ieved+0.057
Neg logit contributions
 Blood-0.032
 spilled-0.031
 blown-0.030
 arrest-0.029
 robbers-0.029
Token door
Token ID3420
Feature activation-3.74e-01

Pos logit contributions
Deal+0.013
 difference+0.011
ieved+0.011
 why+0.011
ement+0.011
Neg logit contributions
 arrest-0.006
 Blood-0.006
 spilled-0.006
 robbers-0.006
 blown-0.006
Token,
Token ID11
Feature activation+4.20e-01

Pos logit contributions
 Blood+0.007
 blown+0.007
 arrest+0.007
alks+0.007
growth+0.007
Neg logit contributions
Deal-0.012
 difference-0.011
 deal-0.010
 promise-0.010
 isn-0.010
Token and
Token ID290
Feature activation+2.68e-01

Pos logit contributions
Deal+0.011
 difference+0.010
ement+0.009
 isn+0.009
 why+0.009
Neg logit contributions
 Blood-0.010
 arrest-0.010
 robbers-0.010
 blown-0.010
growth-0.010
Token to
Token ID284
Feature activation+1.32e+00

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 decision+0.007
Neg logit contributions
 Blood-0.006
 arrest-0.006
 robbers-0.006
 blown-0.006
growth-0.006
Token her
Token ID607
Feature activation+1.05e+00

Pos logit contributions
Deal+0.037
 difference+0.033
ement+0.032
 isn+0.032
 why+0.031
Neg logit contributions
 Blood-0.030
 arrest-0.029
 robbers-0.028
 blown-0.028
growth-0.028
Token amaz
Token ID40642
Feature activation+5.42e+00

Pos logit contributions
Deal+0.028
ieved+0.021
uct+0.020
izes+0.020
 isn+0.019
Neg logit contributions
growth-0.028
 arrest-0.028
 mice-0.026
 robbers-0.026
alks-0.026
Tokenement
Token ID972
Feature activation+1.51e+00

Pos logit contributions
Deal+0.066
 difference+0.042
 promise+0.039
 isn+0.034
 why+0.033
Neg logit contributions
 Blood-0.207
 blown-0.203
growth-0.203
 arrest-0.201
alks-0.198
Token,
Token ID11
Feature activation+3.15e-01

Pos logit contributions
Deal+0.040
 difference+0.034
 isn+0.032
ement+0.031
 why+0.031
Neg logit contributions
 Blood-0.037
 blown-0.037
 arrest-0.036
alks-0.036
 spilled-0.035
Token she
Token ID673
Feature activation+3.81e-01

Pos logit contributions
Deal+0.010
 difference+0.009
ement+0.009
 decision+0.008
 promise+0.008
Neg logit contributions
 Blood-0.006
 blown-0.006
 arrest-0.006
 robbers-0.006
 spilled-0.006
Token could
Token ID714
Feature activation-1.30e+00

Pos logit contributions
Deal+0.012
 difference+0.011
 decision+0.010
ement+0.010
 promise+0.010
Neg logit contributions
 Blood-0.007
 blown-0.007
 arrest-0.007
 spilled-0.007
growth-0.007
Token see
Token ID766
Feature activation+1.10e+00

Pos logit contributions
 arrest+0.024
 Blood+0.023
 blown+0.022
growth+0.022
 robbers+0.022
Neg logit contributions
Deal-0.043
 difference-0.039
ement-0.037
 isn-0.037
 decision-0.036
Token park
Token ID3952
Feature activation-1.27e-01

Pos logit contributions
Deal+0.020
ement+0.019
 isn+0.019
 difference+0.019
 why+0.018
Neg logit contributions
 bushes-0.012
 arrest-0.012
 robbers-0.012
 spilled-0.011
 blown-0.011
Token they
Token ID484
Feature activation-3.43e-01

Pos logit contributions
 robbers+0.003
 Blood+0.003
 arrest+0.003
 spilled+0.003
 blown+0.003
Neg logit contributions
Deal-0.003
 difference-0.003
ement-0.003
ieved-0.003
uct-0.003
Token found
Token ID1043
Feature activation+3.35e-01

Pos logit contributions
 arrest+0.006
 Blood+0.006
 robbers+0.006
 blown+0.006
 spilled+0.006
Neg logit contributions
Deal-0.011
 difference-0.010
ement-0.010
 decision-0.010
 isn-0.010
Token the
Token ID262
Feature activation-1.75e-01

Pos logit contributions
Deal+0.012
 difference+0.011
 decision+0.010
 promise+0.010
ement+0.010
Neg logit contributions
 blown-0.005
 spilled-0.005
 arrest-0.005
 mice-0.005
growth-0.005
Token biggest
Token ID4094
Feature activation-8.73e-01

Pos logit contributions
 robbers+0.004
 blown+0.004
 bushes+0.004
 arrest+0.004
 spilled+0.004
Neg logit contributions
Deal-0.005
ement-0.004
 why-0.004
 difference-0.004
 isn-0.004
Token ear
Token ID1027
Feature activation-5.89e+00

Pos logit contributions
 arrest+0.019
 robbers+0.019
 spilled+0.019
 blown+0.018
 bushes+0.018
Neg logit contributions
Deal-0.025
uct-0.022
ieved-0.021
 int-0.021
 why-0.021
Token Tom
Token ID4186
Feature activation-2.51e+00

Pos logit contributions
 Blood+0.139
 arrest+0.137
 robbers+0.135
 blown+0.132
growth+0.132
Neg logit contributions
Deal-0.158
 difference-0.139
ement-0.135
 isn-0.134
 exactly-0.133
Token had
Token ID550
Feature activation-3.86e+00

Pos logit contributions
 Blood+0.049
 arrest+0.048
 robbers+0.047
 blown+0.047
growth+0.046
Neg logit contributions
Deal-0.078
 difference-0.070
ement-0.068
 isn-0.067
 why-0.067
Token ever
Token ID1683
Feature activation-3.40e+00

Pos logit contributions
 Blood+0.084
 arrest+0.080
 robbers+0.080
growth+0.078
 blown+0.078
Neg logit contributions
Deal-0.111
 difference-0.099
ement-0.096
 isn-0.096
 exactly-0.096
Token seen
Token ID1775
Feature activation-7.87e-01

Pos logit contributions
 Blood+0.072
 arrest+0.065
alks+0.065
growth+0.064
 robbers+0.064
Neg logit contributions
Deal-0.100
 difference-0.092
 why-0.091
 promise-0.091
 isn-0.090
Token!
Token ID0
Feature activation-4.17e-01

Pos logit contributions
 arrest+0.016
 blown+0.016
 mice+0.016
 spilled+0.016
 robbers+0.015
Neg logit contributions
Deal-0.024
 difference-0.021
 promise-0.020
 decision-0.020
 deal-0.019
Token always
Token ID1464
Feature activation-2.63e-01

Pos logit contributions
Deal+0.003
 difference+0.003
ement+0.002
 why+0.002
 exactly+0.002
Neg logit contributions
 Blood-0.005
 arrest-0.005
 blown-0.005
 robbers-0.005
growth-0.005
Token looked
Token ID3114
Feature activation-1.76e-01

Pos logit contributions
 blown+0.006
 spilled+0.006
 Blood+0.006
 robbers+0.005
 arrest+0.005
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.007
 decision-0.006
 int-0.006
Token like
Token ID588
Feature activation-9.54e-01

Pos logit contributions
 spilled+0.004
 blown+0.004
 Blood+0.004
 robbers+0.004
 mice+0.004
Neg logit contributions
Deal-0.005
 difference-0.004
ement-0.004
 decision-0.004
 promise-0.004
Token a
Token ID257
Feature activation-9.42e-01

Pos logit contributions
 Blood+0.021
 arrest+0.021
 robbers+0.020
growth+0.020
alks+0.020
Neg logit contributions
Deal-0.027
 difference-0.024
 why-0.024
 isn-0.024
 exactly-0.024
Token tornado
Token ID33718
Feature activation+9.49e-01

Pos logit contributions
 Blood+0.019
 robbers+0.019
 arrest+0.019
 spilled+0.018
 mice+0.018
Neg logit contributions
Deal-0.029
ement-0.026
 difference-0.025
ieved-0.025
 why-0.024
Token had
Token ID550
Feature activation-5.80e+00

Pos logit contributions
Deal+0.025
 difference+0.021
ement+0.021
 decision+0.020
 exactly+0.020
Neg logit contributions
 arrest-0.024
 blown-0.024
 spilled-0.024
 Blood-0.023
 robbers-0.023
Token gone
Token ID3750
Feature activation-1.64e+00

Pos logit contributions
 robbers+0.136
 Blood+0.134
alks+0.131
 mice+0.124
growth+0.123
Neg logit contributions
 isn-0.140
 exactly-0.134
Deal-0.131
 about-0.128
 why-0.127
Token through
Token ID832
Feature activation-8.87e-01

Pos logit contributions
 Blood+0.039
 arrest+0.039
 blown+0.039
 robbers+0.038
 spilled+0.038
Neg logit contributions
Deal-0.044
 difference-0.038
ement-0.036
 promise-0.035
 decision-0.035
Token it
Token ID340
Feature activation-3.54e-01

Pos logit contributions
 robbers+0.024
 Blood+0.024
alks+0.023
plets+0.023
 blown+0.022
Neg logit contributions
 difference-0.017
 isn-0.017
 why-0.017
Deal-0.017
ement-0.017
Token.
Token ID13
Feature activation+8.54e-02

Pos logit contributions
 Blood+0.008
 arrest+0.008
 blown+0.008
 spilled+0.008
 robbers+0.008
Neg logit contributions
Deal-0.010
 difference-0.008
ement-0.008
 decision-0.008
 why-0.008
Token One
Token ID1881
Feature activation+9.91e-02

Pos logit contributions
Deal+0.002
 difference+0.002
ement+0.002
 isn+0.002
 why+0.002
Neg logit contributions
 Blood-0.002
 arrest-0.002
 blown-0.002
 robbers-0.002
growth-0.002
TokenI
Token ID40
Feature activation+1.08e+00

Pos logit contributions
 difference+0.041
Deal+0.041
ement+0.039
 decision+0.039
 exactly+0.038
Neg logit contributions
 Blood-0.026
growth-0.023
 robbers-0.023
 arrest-0.023
 blown-0.022
Token like
Token ID588
Feature activation-1.38e-02

Pos logit contributions
Deal+0.030
ement+0.029
uct+0.027
 difference+0.026
ieved+0.025
Neg logit contributions
 spilled-0.025
 Blood-0.023
 blown-0.023
 arrest-0.022
 bitten-0.022
Token the
Token ID262
Feature activation-1.45e+00

Pos logit contributions
 Blood+0.000
 arrest+0.000
 robbers+0.000
 blown+0.000
growth+0.000
Neg logit contributions
Deal-0.000
 difference-0.000
ement-0.000
 decision-0.000
 exactly-0.000
Token w
Token ID266
Feature activation-2.07e+00

Pos logit contributions
 Blood+0.032
 blown+0.032
 arrest+0.031
 robbers+0.031
 mice+0.030
Neg logit contributions
Deal-0.041
 difference-0.036
ement-0.036
 exactly-0.035
 why-0.035
Tokenigg
Token ID6950
Feature activation-1.22e+00

Pos logit contributions
 robbers+0.054
 blown+0.053
 Blood+0.053
 arrest+0.053
growth+0.052
Neg logit contributions
Deal-0.050
ement-0.046
 isn-0.044
 difference-0.044
 decision-0.042
Tokenly
Token ID306
Feature activation-5.65e+00

Pos logit contributions
 Blood+0.021
 arrest+0.021
 blown+0.020
 robbers+0.020
alks+0.020
Neg logit contributions
Deal-0.041
 difference-0.037
ement-0.035
 exactly-0.035
 why-0.035
Token dance
Token ID9280
Feature activation-9.03e-01

Pos logit contributions
 Blood+0.132
 arrest+0.130
 robbers+0.127
 blown+0.126
growth+0.124
Neg logit contributions
Deal-0.154
 difference-0.137
ement-0.131
 isn-0.129
 why-0.129
Token the
Token ID262
Feature activation-1.27e+00

Pos logit contributions
 robbers+0.023
 Blood+0.022
 arrest+0.022
 mice+0.022
 blown+0.022
Neg logit contributions
Deal-0.022
 difference-0.020
ement-0.019
 exactly-0.019
uct-0.018
Token most
Token ID749
Feature activation+9.39e-02

Pos logit contributions
 Blood+0.029
 blown+0.029
 robbers+0.029
 arrest+0.029
 spilled+0.028
Neg logit contributions
Deal-0.034
ement-0.030
 difference-0.030
 exactly-0.030
 why-0.029
Token!"
Token ID2474
Feature activation-2.92e-01

Pos logit contributions
Deal+0.003
aman+0.002
ement+0.002
 difference+0.002
ieved+0.002
Neg logit contributions
 blown-0.002
 mice-0.002
 spilled-0.002
 arrest-0.002
 queens-0.002
Token So
Token ID1406
Feature activation+1.57e-01

Pos logit contributions
 Blood+0.007
 robbers+0.006
 arrest+0.006
alks+0.006
 spilled+0.006
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.007
 decision-0.007
 exactly-0.007
Token It
Token ID632
Feature activation-1.36e+00

Pos logit contributions
 Blood+0.027
growth+0.025
 arrest+0.024
alks+0.024
 robbers+0.024
Neg logit contributions
Deal-0.033
 difference-0.029
 decision-0.029
 promise-0.028
 exactly-0.028
Token was
Token ID373
Feature activation-2.88e+00

Pos logit contributions
 Blood+0.041
 blown+0.039
 spilled+0.039
 arrest+0.039
 robbers+0.039
Neg logit contributions
Deal-0.029
 difference-0.025
ement-0.023
 decision-0.023
 exactly-0.022
Token bright
Token ID6016
Feature activation-1.20e+00

Pos logit contributions
 Blood+0.054
 blown+0.052
 arrest+0.051
 robbers+0.050
 spilled+0.049
Neg logit contributions
Deal-0.094
 difference-0.084
ement-0.082
 decision-0.080
 isn-0.079
Token pink
Token ID11398
Feature activation-1.26e+00

Pos logit contributions
 Blood+0.026
 arrest+0.026
 blown+0.025
 robbers+0.025
growth+0.025
Neg logit contributions
Deal-0.035
 difference-0.031
ement-0.029
 decision-0.029
 why-0.029
Token and
Token ID290
Feature activation-1.83e+00

Pos logit contributions
 Blood+0.034
 arrest+0.033
 robbers+0.033
 blown+0.033
growth+0.032
Neg logit contributions
Deal-0.030
 difference-0.026
ement-0.025
 isn-0.025
 exactly-0.025
Token had
Token ID550
Feature activation-5.63e+00

Pos logit contributions
 Blood+0.035
 blown+0.033
 robbers+0.033
 spilled+0.032
 mice+0.031
Neg logit contributions
Deal-0.057
 difference-0.054
 decision-0.051
ement-0.051
 promise-0.050
Token lots
Token ID6041
Feature activation-2.99e+00

Pos logit contributions
 robbers+0.149
 Blood+0.144
 arrest+0.143
alks+0.142
plets+0.136
Neg logit contributions
 isn-0.118
 exactly-0.113
Deal-0.106
 why-0.104
 hairst-0.103
Token of
Token ID286
Feature activation-3.79e+00

Pos logit contributions
 Blood+0.067
 arrest+0.066
 robbers+0.064
 blown+0.063
growth+0.063
Neg logit contributions
Deal-0.083
 difference-0.076
ement-0.074
 isn-0.073
 why-0.072
Token spark
Token ID9009
Feature activation-1.12e+00

Pos logit contributions
 Blood+0.083
 arrest+0.082
 robbers+0.080
 blown+0.079
growth+0.078
Neg logit contributions
Deal-0.109
 difference-0.097
ement-0.094
 isn-0.094
 why-0.092
Tokenles
Token ID829
Feature activation-2.26e+00

Pos logit contributions
 Blood+0.028
 arrest+0.027
 robbers+0.027
growth+0.026
 blown+0.025
Neg logit contributions
Deal-0.028
ement-0.027
 isn-0.026
 difference-0.025
 exactly-0.025
Token.
Token ID13
Feature activation-4.92e-01

Pos logit contributions
 Blood+0.062
 arrest+0.059
 robbers+0.059
growth+0.058
 blown+0.057
Neg logit contributions
Deal-0.051
 difference-0.046
ement-0.045
 exactly-0.045
 isn-0.044
Token
Token ID198
Feature activation-7.99e-01

Pos logit contributions
 Blood+0.001
 arrest+0.001
 robbers+0.001
growth+0.001
 blown+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
 decision-0.001
ement-0.001
 promise-0.001
Token
Token ID198
Feature activation-1.19e+00

Pos logit contributions
 Blood+0.017
growth+0.016
 arrest+0.015
alks+0.015
 blown+0.015
Neg logit contributions
Deal-0.025
 difference-0.022
 decision-0.022
ement-0.021
 promise-0.021
TokenFinally
Token ID11158
Feature activation-4.60e-01

Pos logit contributions
 Blood+0.027
 arrest+0.026
 robbers+0.025
growth+0.025
 blown+0.024
Neg logit contributions
Deal-0.034
 difference-0.032
 decision-0.031
ement-0.030
 isn-0.030
Token,
Token ID11
Feature activation+8.18e-01

Pos logit contributions
 arrest+0.012
 Blood+0.012
 robbers+0.012
 blown+0.012
growth+0.012
Neg logit contributions
Deal-0.012
 difference-0.010
ement-0.009
 decision-0.009
 promise-0.009
Token the
Token ID262
Feature activation-7.36e-01

Pos logit contributions
Deal+0.025
 difference+0.022
ement+0.022
 decision+0.021
 isn+0.021
Neg logit contributions
 arrest-0.017
 Blood-0.017
 blown-0.016
growth-0.016
 robbers-0.016
Token ear
Token ID1027
Feature activation-5.58e+00

Pos logit contributions
 Blood+0.014
 arrest+0.014
 robbers+0.014
 blown+0.014
growth+0.013
Neg logit contributions
Deal-0.023
 difference-0.021
ement-0.020
 exactly-0.020
 why-0.020
Token had
Token ID550
Feature activation-2.17e+00

Pos logit contributions
 Blood+0.132
 arrest+0.130
 blown+0.128
growth+0.125
 robbers+0.124
Neg logit contributions
Deal-0.153
 difference-0.135
ement-0.127
 decision-0.125
 isn-0.124
Token counted
Token ID14789
Feature activation-5.32e-01

Pos logit contributions
 Blood+0.048
 robbers+0.047
 arrest+0.046
 mice+0.045
growth+0.045
Neg logit contributions
Deal-0.059
 difference-0.054
 exactly-0.053
 isn-0.053
 why-0.052
Token all
Token ID477
Feature activation-1.01e+00

Pos logit contributions
 Blood+0.012
 blown+0.012
alks+0.011
 arrest+0.011
 robbers+0.011
Neg logit contributions
Deal-0.016
 difference-0.014
 promise-0.013
 decision-0.013
ement-0.013
Token the
Token ID262
Feature activation-1.24e+00

Pos logit contributions
 blown+0.019
 arrest+0.018
alks+0.017
 Blood+0.017
 spilled+0.017
Neg logit contributions
Deal-0.035
 difference-0.028
 promise-0.028
 deal-0.028
olog-0.027
Token way
Token ID835
Feature activation-8.89e-01

Pos logit contributions
 robbers+0.029
 arrest+0.028
 blown+0.028
 mice+0.028
 Blood+0.027
Neg logit contributions
Deal-0.035
ement-0.029
 difference-0.029
 why-0.029
 isn-0.028
Token was
Token ID373
Feature activation-1.49e+00

Pos logit contributions
Deal+0.010
izes+0.009
 difference+0.009
 int+0.009
ieved+0.008
Neg logit contributions
 spilled-0.006
 Blood-0.006
ges-0.005
 blown-0.005
 bitten-0.005
Token a
Token ID257
Feature activation+2.41e-01

Pos logit contributions
 Blood+0.030
 arrest+0.030
 robbers+0.029
alks+0.029
 blown+0.029
Neg logit contributions
Deal-0.045
 difference-0.040
 exactly-0.039
ement-0.039
 why-0.039
Token ranch
Token ID28752
Feature activation+1.57e+00

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 why+0.007
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 blown-0.004
growth-0.004
Token.
Token ID13
Feature activation-1.29e+00

Pos logit contributions
Deal+0.041
 difference+0.037
ement+0.035
 decision+0.035
 isn+0.035
Neg logit contributions
 Blood-0.037
 arrest-0.037
 mice-0.036
 robbers-0.036
 blown-0.035
Token It
Token ID632
Feature activation-1.77e+00

Pos logit contributions
 arrest+0.031
 blown+0.031
 Blood+0.030
 spilled+0.029
 robbers+0.029
Neg logit contributions
Deal-0.033
 isn-0.029
 why-0.028
 difference-0.028
ement-0.028
Token had
Token ID550
Feature activation-5.52e+00

Pos logit contributions
 Blood+0.057
 arrest+0.056
 robbers+0.055
 blown+0.055
growth+0.054
Neg logit contributions
Deal-0.033
 difference-0.028
ement-0.026
 isn-0.026
 why-0.025
Token been
Token ID587
Feature activation-3.06e+00

Pos logit contributions
alks+0.158
 Blood+0.155
plets+0.151
 arrest+0.148
ges+0.148
Neg logit contributions
Deal-0.097
 isn-0.096
 about-0.093
 exactly-0.090
 wonderful-0.088
Token used
Token ID973
Feature activation-3.00e+00

Pos logit contributions
growth+0.096
 arrest+0.094
alks+0.088
 Blood+0.087
plets+0.086
Neg logit contributions
 exactly-0.056
Deal-0.056
 why-0.052
 equally-0.049
 about-0.048
Token by
Token ID416
Feature activation-4.28e-01

Pos logit contributions
 Blood+0.071
 arrest+0.071
growth+0.069
 robbers+0.068
 blown+0.066
Neg logit contributions
Deal-0.078
 exactly-0.070
 difference-0.069
 why-0.068
 isn-0.068
Token many
Token ID867
Feature activation-1.02e+00

Pos logit contributions
 Blood+0.010
 arrest+0.010
 robbers+0.010
 blown+0.010
growth+0.010
Neg logit contributions
Deal-0.011
 difference-0.010
 why-0.010
 exactly-0.010
 isn-0.010
Token animals
Token ID4695
Feature activation-1.18e-01

Pos logit contributions
 Blood+0.020
 arrest+0.020
 robbers+0.019
 blown+0.019
growth+0.019
Neg logit contributions
Deal-0.032
 difference-0.029
ement-0.028
 isn-0.028
 why-0.027
Token big
Token ID1263
Feature activation+1.73e+00

Pos logit contributions
Deal+0.017
 difference+0.015
ement+0.014
 why+0.014
 exactly+0.014
Neg logit contributions
 Blood-0.010
 robbers-0.009
 arrest-0.009
 blown-0.009
 spilled-0.009
Token do
Token ID466
Feature activation-4.02e+00

Pos logit contributions
Deal+0.052
 difference+0.043
ement+0.043
 exactly+0.043
ieved+0.042
Neg logit contributions
 arrest-0.037
 robbers-0.037
 Blood-0.036
 mice-0.035
 blown-0.035
Tokenggy
Token ID19970
Feature activation-3.69e-01

Pos logit contributions
 Blood+0.126
 arrest+0.124
 robbers+0.121
 blown+0.120
growth+0.119
Neg logit contributions
Deal-0.080
 difference-0.068
ement-0.062
 why-0.061
 isn-0.060
Token.
Token ID13
Feature activation-1.54e+00

Pos logit contributions
 arrest+0.008
 robbers+0.008
 spilled+0.008
 Blood+0.007
 blown+0.007
Neg logit contributions
Deal-0.011
 difference-0.010
ement-0.009
 decision-0.009
 isn-0.009
Token The
Token ID383
Feature activation-1.83e+00

Pos logit contributions
 Blood+0.038
 arrest+0.037
 robbers+0.036
 blown+0.035
growth+0.035
Neg logit contributions
Deal-0.042
 difference-0.036
ement-0.034
 exactly-0.034
 decision-0.034
Token do
Token ID466
Feature activation-5.51e+00

Pos logit contributions
 Blood+0.043
 arrest+0.043
 blown+0.042
 robbers+0.041
growth+0.041
Neg logit contributions
Deal-0.050
 difference-0.044
 isn-0.042
ement-0.042
 decision-0.041
Tokenggy
Token ID19970
Feature activation-1.94e+00

Pos logit contributions
 Blood+0.175
 arrest+0.171
 robbers+0.167
 blown+0.166
growth+0.163
Neg logit contributions
Deal-0.106
 difference-0.091
ement-0.083
 why-0.082
 isn-0.081
Token bark
Token ID21405
Feature activation-7.44e-01

Pos logit contributions
 Blood+0.050
 spilled+0.048
 robbers+0.048
 arrest+0.048
 blown+0.047
Neg logit contributions
Deal-0.050
 difference-0.045
ement-0.042
 decision-0.041
 promise-0.040
Tokened
Token ID276
Feature activation+3.01e-01

Pos logit contributions
 Blood+0.013
 arrest+0.013
 robbers+0.012
 blown+0.012
growth+0.012
Neg logit contributions
Deal-0.025
 difference-0.022
ement-0.022
 isn-0.022
 why-0.021
Token and
Token ID290
Feature activation-3.26e-01

Pos logit contributions
Deal+0.011
 difference+0.009
ieved+0.009
uct+0.008
ement+0.008
Neg logit contributions
 mice-0.006
 robbers-0.005
 Blood-0.005
 spilled-0.005
 blown-0.005
Token Tommy
Token ID19919
Feature activation-3.27e-01

Pos logit contributions
 Blood+0.007
 arrest+0.007
 robbers+0.007
 blown+0.007
growth+0.007
Neg logit contributions
Deal-0.009
 difference-0.008
ement-0.008
 isn-0.008
 decision-0.008
Token interesting
Token ID3499
Feature activation+2.16e-01

Pos logit contributions
Deal+0.031
ieved+0.024
uct+0.024
 difference+0.023
rats+0.022
Neg logit contributions
 mice-0.032
 blown-0.031
 robbers-0.030
 arrest-0.030
growth-0.030
Token things
Token ID1243
Feature activation+1.97e+00

Pos logit contributions
Deal+0.006
ieved+0.005
uct+0.005
 difference+0.005
ement+0.005
Neg logit contributions
 mice-0.005
 robbers-0.005
 arrest-0.005
growth-0.005
alks-0.005
Token about
Token ID546
Feature activation+3.70e-01

Pos logit contributions
Deal+0.059
 difference+0.051
ieved+0.049
 decision+0.049
ement+0.049
Neg logit contributions
 blown-0.040
 spilled-0.038
 robbers-0.038
 arrest-0.037
 Blood-0.036
Token the
Token ID262
Feature activation+5.17e-01

Pos logit contributions
Deal+0.010
 difference+0.009
ement+0.009
 decision+0.008
 promise+0.008
Neg logit contributions
 arrest-0.008
 Blood-0.008
 robbers-0.008
 blown-0.008
growth-0.008
Token unknown
Token ID6439
Feature activation-7.63e-01

Pos logit contributions
Deal+0.015
ieved+0.013
ement+0.012
uct+0.012
 exactly+0.012
Neg logit contributions
 arrest-0.012
 robbers-0.012
 mice-0.011
 blown-0.011
 spilled-0.011
Token ear
Token ID1027
Feature activation-5.46e+00

Pos logit contributions
 arrest+0.017
 blown+0.016
 Blood+0.016
 robbers+0.016
growth+0.016
Neg logit contributions
Deal-0.022
ieved-0.018
uct-0.018
 difference-0.018
ement-0.018
Token.
Token ID13
Feature activation+5.07e-01

Pos logit contributions
 Blood+0.144
 arrest+0.141
 robbers+0.139
 blown+0.137
growth+0.136
Neg logit contributions
Deal-0.131
 difference-0.116
ement-0.111
 isn-0.110
 why-0.109
Token He
Token ID679
Feature activation+3.30e-02

Pos logit contributions
Deal+0.014
 difference+0.012
 decision+0.011
 promise+0.011
ement+0.011
Neg logit contributions
 Blood-0.012
alks-0.012
growth-0.012
 arrest-0.012
 robbers-0.011
Token still
Token ID991
Feature activation+5.15e-01

Pos logit contributions
Deal+0.001
 difference+0.001
ement+0.001
 decision+0.001
 deal+0.001
Neg logit contributions
 blown-0.001
 arrest-0.001
 robbers-0.001
 Blood-0.001
growth-0.001
Token visits
Token ID11864
Feature activation+5.84e-01

Pos logit contributions
Deal+0.015
 difference+0.013
ement+0.013
 decision+0.013
 exactly+0.012
Neg logit contributions
 Blood-0.011
 arrest-0.011
 blown-0.011
 robbers-0.011
 spilled-0.010
Token it
Token ID340
Feature activation+7.82e-01

Pos logit contributions
Deal+0.018
 difference+0.015
ement+0.015
 promise+0.015
 decision+0.014
Neg logit contributions
 arrest-0.012
 Blood-0.012
 robbers-0.012
 blown-0.012
growth-0.012
Token that
Token ID326
Feature activation+2.51e+00

Pos logit contributions
Deal+0.031
 difference+0.028
 exactly+0.027
ement+0.027
 why+0.027
Neg logit contributions
 Blood-0.033
 arrest-0.032
 robbers-0.032
growth-0.032
 blown-0.031
Token.
Token ID13
Feature activation-2.44e-01

Pos logit contributions
Deal+0.072
ieved+0.062
 exactly+0.060
ement+0.059
uct+0.059
Neg logit contributions
 mice-0.055
 spilled-0.054
 blown-0.053
 arrest-0.053
 robbers-0.052
Token Whenever
Token ID21326
Feature activation-2.87e-01

Pos logit contributions
 Blood+0.005
 arrest+0.005
 blown+0.005
 robbers+0.005
growth+0.005
Neg logit contributions
Deal-0.007
 difference-0.007
 isn-0.006
ement-0.006
 why-0.006
Token the
Token ID262
Feature activation-1.77e+00

Pos logit contributions
 Blood+0.006
 arrest+0.006
 robbers+0.005
 blown+0.005
growth+0.005
Neg logit contributions
Deal-0.009
 difference-0.008
ement-0.008
 isn-0.008
 decision-0.008
Token chip
Token ID11594
Feature activation-2.85e+00

Pos logit contributions
 Blood+0.038
 arrest+0.037
growth+0.037
 robbers+0.036
 blown+0.036
Neg logit contributions
Deal-0.052
 difference-0.047
 decision-0.045
 promise-0.045
ement-0.044
Tokenm
Token ID76
Feature activation-5.46e+00

Pos logit contributions
 Blood+0.066
 arrest+0.065
 robbers+0.064
 blown+0.063
growth+0.062
Neg logit contributions
Deal-0.078
 difference-0.069
ement-0.066
 isn-0.066
 why-0.065
Tokenunk
Token ID2954
Feature activation-2.03e+00

Pos logit contributions
 robbers+0.107
growth+0.105
 arrest+0.104
 blown+0.096
 Blood+0.096
Neg logit contributions
Deal-0.160
 decision-0.155
 isn-0.151
 difference-0.151
ement-0.150
Token needed
Token ID2622
Feature activation-1.03e+00

Pos logit contributions
 Blood+0.049
 arrest+0.049
 blown+0.048
 robbers+0.048
 mice+0.047
Neg logit contributions
Deal-0.053
 difference-0.046
ement-0.045
 decision-0.044
 exactly-0.044
Token help
Token ID1037
Feature activation-5.61e-01

Pos logit contributions
 Blood+0.022
 arrest+0.022
 robbers+0.021
 blown+0.021
growth+0.021
Neg logit contributions
Deal-0.030
 difference-0.027
ement-0.026
 why-0.026
 isn-0.026
Token,
Token ID11
Feature activation+9.97e-02

Pos logit contributions
 Blood+0.015
 arrest+0.015
 mice+0.015
 robbers+0.015
 blown+0.014
Neg logit contributions
Deal-0.014
ement-0.011
 difference-0.011
ieved-0.011
uct-0.011
Token Sam
Token ID3409
Feature activation+5.09e-01

Pos logit contributions
Deal+0.003
 difference+0.003
ement+0.003
 isn+0.003
 decision+0.003
Neg logit contributions
 Blood-0.002
 arrest-0.002
 robbers-0.002
 blown-0.002
 mice-0.002
Token a
Token ID257
Feature activation+2.48e-01

Pos logit contributions
alks+0.058
 arrest+0.057
plets+0.056
 robbers+0.056
 Blood+0.055
Neg logit contributions
Deal-0.053
 exactly-0.046
 equally-0.043
 why-0.043
 difference-0.043
Token big
Token ID1263
Feature activation+1.89e+00

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 exactly+0.007
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 blown-0.004
 mice-0.004
Token house
Token ID2156
Feature activation+2.40e-01

Pos logit contributions
Deal+0.058
ieved+0.050
ement+0.049
 exactly+0.049
uct+0.049
Neg logit contributions
 robbers-0.039
 arrest-0.038
 mice-0.037
 Blood-0.037
 spilled-0.036
Token.
Token ID13
Feature activation-1.38e+00

Pos logit contributions
Deal+0.006
 difference+0.006
ement+0.005
 isn+0.005
 decision+0.005
Neg logit contributions
 Blood-0.006
 arrest-0.006
 blown-0.006
 robbers-0.006
 spilled-0.006
Token It
Token ID632
Feature activation-1.77e+00

Pos logit contributions
 blown+0.033
 arrest+0.033
 spilled+0.031
alks+0.031
 Blood+0.031
Neg logit contributions
Deal-0.036
 isn-0.032
 difference-0.031
 why-0.031
ement-0.030
Token had
Token ID550
Feature activation-5.45e+00

Pos logit contributions
 Blood+0.050
 spilled+0.048
 blown+0.048
 arrest+0.046
 robbers+0.046
Neg logit contributions
Deal-0.041
 difference-0.037
ement-0.035
 exactly-0.034
 decision-0.034
Token lots
Token ID6041
Feature activation-2.56e+00

Pos logit contributions
alks+0.141
plets+0.140
 robbers+0.140
 Blood+0.138
 arrest+0.136
Neg logit contributions
Deal-0.112
 isn-0.109
 exactly-0.107
 why-0.103
 difference-0.102
Token of
Token ID286
Feature activation-3.93e+00

Pos logit contributions
 Blood+0.063
 arrest+0.061
 robbers+0.060
growth+0.060
 blown+0.059
Neg logit contributions
Deal-0.065
 difference-0.060
ement-0.058
 isn-0.058
 why-0.056
Token surprises
Token ID24072
Feature activation+1.41e+00

Pos logit contributions
 Blood+0.074
 arrest+0.072
 robbers+0.071
 blown+0.070
growth+0.068
Neg logit contributions
Deal-0.126
 difference-0.113
ement-0.110
 isn-0.108
 why-0.108
Token inside
Token ID2641
Feature activation-6.70e-01

Pos logit contributions
Deal+0.040
 difference+0.036
ement+0.034
ieved+0.033
 decision+0.033
Neg logit contributions
 blown-0.031
 robbers-0.030
 spilled-0.030
 Blood-0.030
 arrest-0.030
Token.
Token ID13
Feature activation-7.62e-01

Pos logit contributions
 Blood+0.014
 arrest+0.013
 mice+0.013
 blown+0.013
 robbers+0.013
Neg logit contributions
Deal-0.020
 difference-0.018
ement-0.017
 decision-0.017
 promise-0.017
Token "
Token ID366
Feature activation+1.55e+00

Pos logit contributions
Deal+0.004
 difference+0.003
 why+0.003
ement+0.003
 isn+0.003
Neg logit contributions
 arrest-0.005
 Blood-0.005
 robbers-0.005
 blown-0.005
growth-0.004
TokenLet
Token ID5756
Feature activation+1.83e-01

Pos logit contributions
Deal+0.044
 difference+0.041
ement+0.039
 decision+0.038
 isn+0.038
Neg logit contributions
 Blood-0.034
 arrest-0.033
 blown-0.032
 robbers-0.032
growth-0.032
Token's
Token ID338
Feature activation+2.86e+00

Pos logit contributions
Deal+0.004
ement+0.003
 difference+0.003
 exactly+0.003
 decision+0.003
Neg logit contributions
 arrest-0.006
 Blood-0.006
 blown-0.005
 spilled-0.005
 robbers-0.005
Token replace
Token ID6330
Feature activation+1.44e+00

Pos logit contributions
Deal+0.087
ement+0.078
 difference+0.073
 isn+0.071
 decision+0.071
Neg logit contributions
 arrest-0.061
 robbers-0.058
 Blood-0.058
 blown-0.058
growth-0.057
Token the
Token ID262
Feature activation-1.40e+00

Pos logit contributions
Deal+0.038
 difference+0.032
ement+0.032
 decision+0.031
 promise+0.030
Neg logit contributions
 arrest-0.036
 Blood-0.035
 mice-0.035
 robbers-0.035
 blown-0.034
Token ear
Token ID1027
Feature activation-5.43e+00

Pos logit contributions
 Blood+0.030
 arrest+0.030
 robbers+0.029
 blown+0.029
 mice+0.028
Neg logit contributions
Deal-0.041
 difference-0.036
ement-0.036
 why-0.035
 exactly-0.035
Token so
Token ID523
Feature activation+6.22e-01

Pos logit contributions
 Blood+0.136
 arrest+0.134
 blown+0.130
 robbers+0.130
growth+0.127
Neg logit contributions
Deal-0.140
 difference-0.123
ement-0.117
 isn-0.114
 decision-0.114
Token your
Token ID534
Feature activation+2.95e-01

Pos logit contributions
Deal+0.015
 difference+0.013
ement+0.013
 decision+0.012
 isn+0.012
Neg logit contributions
 Blood-0.017
 arrest-0.016
 blown-0.016
 robbers-0.016
 mice-0.016
Token toy
Token ID13373
Feature activation-6.87e-01

Pos logit contributions
Deal+0.009
ement+0.008
 difference+0.008
 why+0.007
ieved+0.007
Neg logit contributions
 Blood-0.006
 arrest-0.006
 blown-0.006
 robbers-0.006
 mice-0.006
Token cat
Token ID3797
Feature activation-5.55e-01

Pos logit contributions
 arrest+0.018
 Blood+0.017
 robbers+0.017
 blown+0.017
 mice+0.017
Neg logit contributions
Deal-0.017
 difference-0.015
ement-0.015
 decision-0.014
 why-0.014
Token can
Token ID460
Feature activation-1.42e+00

Pos logit contributions
 arrest+0.015
 Blood+0.015
 blown+0.015
 robbers+0.015
 spilled+0.015
Neg logit contributions
Deal-0.013
 difference-0.011
ement-0.011
 decision-0.010
 int-0.010
Token some
Token ID617
Feature activation-2.00e+00

Pos logit contributions
 arrest+0.021
 Blood+0.021
alks+0.020
 robbers+0.020
 blown+0.020
Neg logit contributions
Deal-0.020
 why-0.018
 isn-0.018
 exactly-0.017
ement-0.017
Token are
Token ID389
Feature activation-2.22e+00

Pos logit contributions
 Blood+0.053
 arrest+0.052
 robbers+0.050
growth+0.050
 blown+0.049
Neg logit contributions
Deal-0.048
 difference-0.043
 isn-0.042
 why-0.041
 promise-0.041
Token blue
Token ID4171
Feature activation-5.29e-01

Pos logit contributions
 arrest+0.050
 Blood+0.050
growth+0.048
 robbers+0.048
 blown+0.047
Neg logit contributions
Deal-0.061
 difference-0.055
 why-0.054
 exactly-0.054
 isn-0.053
Token,
Token ID11
Feature activation-1.30e+00

Pos logit contributions
 Blood+0.019
 arrest+0.018
 robbers+0.018
growth+0.018
 blown+0.018
Neg logit contributions
Deal-0.008
 isn-0.007
 difference-0.007
ement-0.007
 why-0.006
Token some
Token ID617
Feature activation-2.26e+00

Pos logit contributions
 arrest+0.033
 Blood+0.032
 robbers+0.032
alks+0.031
growth+0.031
Neg logit contributions
Deal-0.033
 isn-0.030
 why-0.029
 exactly-0.029
ement-0.029
Token have
Token ID423
Feature activation-5.42e+00

Pos logit contributions
 Blood+0.057
 arrest+0.056
 robbers+0.055
growth+0.054
 blown+0.053
Neg logit contributions
Deal-0.057
 difference-0.051
 isn-0.049
 why-0.048
 promise-0.048
Token flowers
Token ID12734
Feature activation-5.82e-01

Pos logit contributions
 Blood+0.136
 arrest+0.131
 robbers+0.130
alks+0.125
growth+0.121
Neg logit contributions
 isn-0.121
 exactly-0.114
 why-0.113
Deal-0.111
 hairst-0.109
Token,
Token ID11
Feature activation-1.62e+00

Pos logit contributions
 Blood+0.019
 arrest+0.018
 robbers+0.018
growth+0.018
 mice+0.017
Neg logit contributions
Deal-0.010
 isn-0.009
 why-0.009
 difference-0.009
 exactly-0.009
Token some
Token ID617
Feature activation-2.02e+00

Pos logit contributions
 Blood+0.040
 arrest+0.039
alks+0.038
 robbers+0.038
 blown+0.038
Neg logit contributions
Deal-0.038
 isn-0.038
 why-0.037
 exactly-0.035
ement-0.034
Token have
Token ID423
Feature activation-5.39e+00

Pos logit contributions
 Blood+0.045
 arrest+0.044
 robbers+0.043
growth+0.043
 blown+0.041
Neg logit contributions
Deal-0.056
 difference-0.051
 isn-0.051
 why-0.049
ement-0.049
Token dots
Token ID22969
Feature activation-2.91e-01

Pos logit contributions
 Blood+0.130
 robbers+0.125
 arrest+0.122
alks+0.117
plets+0.113
Neg logit contributions
 isn-0.126
 exactly-0.117
 why-0.117
Deal-0.116
 hairst-0.115
Token back
Token ID736
Feature activation-1.21e+00

Pos logit contributions
Deal+0.004
ement+0.004
 exactly+0.004
 difference+0.004
 why+0.004
Neg logit contributions
 arrest-0.004
 Blood-0.004
 robbers-0.004
growth-0.004
alks-0.004
Token and
Token ID290
Feature activation+9.15e-01

Pos logit contributions
 Blood+0.027
 arrest+0.026
 blown+0.026
 robbers+0.026
growth+0.025
Neg logit contributions
Deal-0.035
 difference-0.031
ement-0.030
 isn-0.029
 decision-0.029
Token admired
Token ID29382
Feature activation-4.60e-01

Pos logit contributions
Deal+0.023
 difference+0.020
ement+0.020
 decision+0.019
 isn+0.018
Neg logit contributions
 blown-0.024
 spilled-0.024
 Blood-0.024
 arrest-0.023
 robbers-0.023
Token the
Token ID262
Feature activation-1.23e+00

Pos logit contributions
 Blood+0.011
 arrest+0.011
 blown+0.011
 robbers+0.011
growth+0.011
Neg logit contributions
Deal-0.012
 difference-0.011
ement-0.010
 decision-0.010
 promise-0.010
Token spark
Token ID9009
Feature activation-1.13e+00

Pos logit contributions
 robbers+0.023
 blown+0.023
 arrest+0.023
 mice+0.022
 spilled+0.022
Neg logit contributions
Deal-0.040
ement-0.035
ieved-0.035
uct-0.035
 why-0.034
Tokenly
Token ID306
Feature activation-5.41e+00

Pos logit contributions
 arrest+0.024
 Blood+0.023
 blown+0.023
 mice+0.023
 spilled+0.023
Neg logit contributions
Deal-0.033
 difference-0.028
 exactly-0.027
 why-0.027
ement-0.027
Token space
Token ID2272
Feature activation-3.95e-01

Pos logit contributions
 Blood+0.126
 arrest+0.122
 robbers+0.120
 blown+0.119
growth+0.117
Neg logit contributions
Deal-0.146
 difference-0.134
 isn-0.128
ement-0.126
 decision-0.126
Token.
Token ID13
Feature activation+2.67e-02

Pos logit contributions
 blown+0.008
 arrest+0.008
 mice+0.008
alks+0.008
 spilled+0.008
Neg logit contributions
Deal-0.012
 difference-0.010
ieved-0.010
uct-0.009
ement-0.009
Token
Token ID198
Feature activation-6.48e-01

Pos logit contributions
Deal+0.001
 difference+0.001
 isn+0.001
ement+0.001
 why+0.001
Neg logit contributions
 arrest-0.001
 robbers-0.001
 blown-0.001
 Blood-0.001
 spilled-0.001
Token
Token ID198
Feature activation-1.20e+00

Pos logit contributions
 Blood+0.015
 arrest+0.014
growth+0.014
 blown+0.014
alks+0.014
Neg logit contributions
Deal-0.019
 difference-0.017
 decision-0.016
ement-0.016
 isn-0.015
TokenThe
Token ID464
Feature activation-9.66e-01

Pos logit contributions
 Blood+0.031
 arrest+0.030
 blown+0.030
 robbers+0.030
growth+0.029
Neg logit contributions
Deal-0.030
 difference-0.027
ement-0.026
 decision-0.025
 isn-0.025
Token some
Token ID617
Feature activation-2.26e+00

Pos logit contributions
 arrest+0.033
 Blood+0.032
 robbers+0.032
alks+0.031
growth+0.031
Neg logit contributions
Deal-0.033
 isn-0.030
 why-0.029
 exactly-0.029
ement-0.029
Token have
Token ID423
Feature activation-5.42e+00

Pos logit contributions
 Blood+0.057
 arrest+0.056
 robbers+0.055
growth+0.054
 blown+0.053
Neg logit contributions
Deal-0.057
 difference-0.051
 isn-0.049
 why-0.048
 promise-0.048
Token flowers
Token ID12734
Feature activation-5.82e-01

Pos logit contributions
 Blood+0.136
 arrest+0.131
 robbers+0.130
alks+0.125
growth+0.121
Neg logit contributions
 isn-0.121
 exactly-0.114
 why-0.113
Deal-0.111
 hairst-0.109
Token,
Token ID11
Feature activation-1.62e+00

Pos logit contributions
 Blood+0.019
 arrest+0.018
 robbers+0.018
growth+0.018
 mice+0.017
Neg logit contributions
Deal-0.010
 isn-0.009
 why-0.009
 difference-0.009
 exactly-0.009
Token some
Token ID617
Feature activation-2.02e+00

Pos logit contributions
 Blood+0.040
 arrest+0.039
alks+0.038
 robbers+0.038
 blown+0.038
Neg logit contributions
Deal-0.038
 isn-0.038
 why-0.037
 exactly-0.035
ement-0.034
Token have
Token ID423
Feature activation-5.39e+00

Pos logit contributions
 Blood+0.045
 arrest+0.044
 robbers+0.043
growth+0.043
 blown+0.041
Neg logit contributions
Deal-0.056
 difference-0.051
 isn-0.051
 why-0.049
ement-0.049
Token dots
Token ID22969
Feature activation-2.91e-01

Pos logit contributions
 Blood+0.130
 robbers+0.125
 arrest+0.122
alks+0.117
plets+0.113
Neg logit contributions
 isn-0.126
 exactly-0.117
 why-0.117
Deal-0.116
 hairst-0.115
Token.
Token ID13
Feature activation+2.96e-01

Pos logit contributions
 arrest+0.010
 robbers+0.009
 Blood+0.009
growth+0.009
 blown+0.009
Neg logit contributions
Deal-0.005
 why-0.004
 isn-0.004
 exactly-0.004
ement-0.004
Token She
Token ID1375
Feature activation-9.33e-01

Pos logit contributions
Deal+0.009
 difference+0.008
 decision+0.008
ement+0.008
 exactly+0.008
Neg logit contributions
 Blood-0.006
 arrest-0.006
 robbers-0.006
 blown-0.006
growth-0.006
Token likes
Token ID7832
Feature activation-6.28e-01

Pos logit contributions
 Blood+0.023
 arrest+0.023
 blown+0.022
 robbers+0.022
growth+0.022
Neg logit contributions
Deal-0.024
 difference-0.021
ement-0.020
 decision-0.020
 exactly-0.020
Token to
Token ID284
Feature activation-5.07e-02

Pos logit contributions
 Blood+0.015
 arrest+0.015
growth+0.014
 robbers+0.014
plets+0.014
Neg logit contributions
Deal-0.016
 why-0.015
 isn-0.015
 difference-0.015
 exactly-0.014
Token box
Token ID3091
Feature activation+7.80e-01

Pos logit contributions
Deal+0.030
ieved+0.025
uct+0.025
ement+0.025
 exactly+0.024
Neg logit contributions
 robbers-0.020
 arrest-0.019
 spilled-0.019
 mice-0.018
 Blood-0.018
Token full
Token ID1336
Feature activation-3.55e+00

Pos logit contributions
Deal+0.020
 difference+0.017
ement+0.017
 why+0.017
 exactly+0.017
Neg logit contributions
 Blood-0.020
 arrest-0.019
 robbers-0.019
 blown-0.019
growth-0.019
Token of
Token ID286
Feature activation-3.72e+00

Pos logit contributions
 Blood+0.062
 arrest+0.061
 robbers+0.061
 blown+0.060
 mice+0.060
Neg logit contributions
Deal-0.118
 difference-0.103
ement-0.100
 isn-0.098
 decision-0.098
Token beautiful
Token ID4950
Feature activation-2.44e+00

Pos logit contributions
 robbers+0.070
 mice+0.070
 arrest+0.069
 Blood+0.068
 blown+0.067
Neg logit contributions
Deal-0.119
 difference-0.103
ement-0.102
ieved-0.100
 decision-0.099
Token or
Token ID393
Feature activation-3.99e-01

Pos logit contributions
 arrest+0.052
 Blood+0.052
 robbers+0.051
 blown+0.050
 mice+0.050
Neg logit contributions
Deal-0.072
 difference-0.063
ement-0.061
 why-0.060
 exactly-0.060
Tokennam
Token ID7402
Feature activation-5.38e+00

Pos logit contributions
 Blood+0.008
 robbers+0.008
 arrest+0.008
 blown+0.008
 mice+0.007
Neg logit contributions
Deal-0.012
 difference-0.011
ement-0.011
 isn-0.010
 decision-0.010
Tokenents
Token ID658
Feature activation-1.78e+00

Pos logit contributions
 robbers+0.119
growth+0.113
 Blood+0.112
 arrest+0.112
 spilled+0.109
Neg logit contributions
Deal-0.140
 equally-0.136
 exactly-0.135
ement-0.134
 difference-0.132
Token.
Token ID13
Feature activation-8.55e-01

Pos logit contributions
 Blood+0.049
 arrest+0.047
 robbers+0.047
growth+0.046
alks+0.045
Neg logit contributions
Deal-0.041
 difference-0.036
 exactly-0.035
ement-0.035
 why-0.035
Token
Token ID220
Feature activation-2.06e+00

Pos logit contributions
 arrest+0.019
 Blood+0.019
 blown+0.019
 robbers+0.019
alks+0.018
Neg logit contributions
Deal-0.024
 difference-0.021
 isn-0.020
ement-0.020
 why-0.020
Token
Token ID198
Feature activation-1.12e+00

Pos logit contributions
 Blood+0.053
 arrest+0.051
 robbers+0.050
 blown+0.050
growth+0.049
Neg logit contributions
Deal-0.052
 difference-0.045
ement-0.043
 isn-0.043
 decision-0.042
Token
Token ID198
Feature activation-1.52e+00

Pos logit contributions
 Blood+0.026
 arrest+0.025
 robbers+0.025
growth+0.024
 blown+0.024
Neg logit contributions
Deal-0.031
 difference-0.027
 decision-0.026
ement-0.026
 isn-0.026
Token other
Token ID584
Feature activation-4.28e+00

Pos logit contributions
 arrest+0.057
 Blood+0.057
growth+0.056
 blown+0.054
alks+0.052
Neg logit contributions
Deal-0.073
 difference-0.068
 decision-0.064
 promise-0.064
 why-0.061
Token was
Token ID373
Feature activation-9.09e-01

Pos logit contributions
 Blood+0.095
 arrest+0.094
 blown+0.090
alks+0.089
growth+0.087
Neg logit contributions
Deal-0.125
 difference-0.113
 isn-0.112
 why-0.109
 decision-0.108
Token sleepy
Token ID42023
Feature activation-9.27e-01

Pos logit contributions
 Blood+0.024
 arrest+0.024
 robbers+0.024
growth+0.024
 blown+0.023
Neg logit contributions
Deal-0.022
 difference-0.019
 why-0.018
ement-0.018
 exactly-0.018
Token.
Token ID13
Feature activation-1.10e+00

Pos logit contributions
 Blood+0.022
 robbers+0.022
 arrest+0.022
 mice+0.022
 blown+0.021
Neg logit contributions
Deal-0.025
 difference-0.021
ement-0.021
 exactly-0.020
ieved-0.020
Token The
Token ID383
Feature activation-1.47e+00

Pos logit contributions
 blown+0.026
 arrest+0.026
 Blood+0.024
 spilled+0.024
growth+0.024
Neg logit contributions
Deal-0.030
 difference-0.026
 isn-0.026
 why-0.026
ement-0.025
Token alert
Token ID7995
Feature activation-5.37e+00

Pos logit contributions
 arrest+0.033
 Blood+0.033
growth+0.032
 blown+0.031
alks+0.031
Neg logit contributions
Deal-0.042
 difference-0.039
 promise-0.037
 decision-0.037
 why-0.036
Token human
Token ID1692
Feature activation-1.46e+00

Pos logit contributions
 Blood+0.110
 arrest+0.105
growth+0.103
 blown+0.103
 bitten+0.097
Neg logit contributions
 decision-0.149
 difference-0.146
 why-0.142
 isn-0.140
Deal-0.140
Token said
Token ID531
Feature activation-8.83e-01

Pos logit contributions
 Blood+0.033
 mice+0.033
 robbers+0.033
 spilled+0.033
 arrest+0.032
Neg logit contributions
Deal-0.040
 difference-0.036
ement-0.035
 decision-0.034
 exactly-0.033
Token,
Token ID11
Feature activation-3.43e-01

Pos logit contributions
 Blood+0.023
 mice+0.022
 robbers+0.021
 arrest+0.021
 spilled+0.021
Neg logit contributions
Deal-0.024
 difference-0.021
ement-0.020
ieved-0.019
 decision-0.019
Token "
Token ID366
Feature activation+3.53e-01

Pos logit contributions
 Blood+0.008
 robbers+0.008
 arrest+0.008
 mice+0.008
 blown+0.008
Neg logit contributions
Deal-0.009
 difference-0.009
 decision-0.008
ement-0.008
 promise-0.008
TokenLet
Token ID5756
Feature activation-5.50e-01

Pos logit contributions
Deal+0.010
 difference+0.009
 isn+0.009
ement+0.009
 why+0.009
Neg logit contributions
 Blood-0.008
 arrest-0.008
 robbers-0.007
 blown-0.007
alks-0.007
Token room
Token ID2119
Feature activation+5.75e-01

Pos logit contributions
 Blood+0.001
 arrest+0.001
 robbers+0.001
 blown+0.001
growth+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
ement-0.001
 isn-0.001
 why-0.001
Token.
Token ID13
Feature activation-1.64e+00

Pos logit contributions
Deal+0.014
 difference+0.013
ement+0.012
 isn+0.012
 why+0.012
Neg logit contributions
 Blood-0.015
 arrest-0.015
 robbers-0.014
 blown-0.014
growth-0.014
Token He
Token ID679
Feature activation-4.35e-01

Pos logit contributions
 arrest+0.036
 Blood+0.036
 blown+0.035
 robbers+0.034
growth+0.034
Neg logit contributions
Deal-0.047
 difference-0.042
 isn-0.040
 why-0.040
ement-0.040
Token saw
Token ID2497
Feature activation+4.46e-01

Pos logit contributions
 Blood+0.009
 robbers+0.009
 arrest+0.009
 blown+0.009
 spilled+0.008
Neg logit contributions
Deal-0.013
 difference-0.012
ement-0.012
 decision-0.011
 exactly-0.011
Token the
Token ID262
Feature activation-9.50e-01

Pos logit contributions
 why+0.009
Deal+0.009
 exactly+0.008
 difference+0.008
 isn+0.008
Neg logit contributions
plets-0.012
alks-0.012
 Blood-0.012
 arrest-0.012
growth-0.012
Token lonely
Token ID21757
Feature activation-5.36e+00

Pos logit contributions
 Blood+0.025
 arrest+0.025
growth+0.025
 robbers+0.024
 blown+0.024
Neg logit contributions
Deal-0.023
 difference-0.020
 decision-0.019
 isn-0.019
ement-0.019
Token machine
Token ID4572
Feature activation+1.15e+00

Pos logit contributions
growth+0.169
 arrest+0.166
 blown+0.164
 Blood+0.162
 robbers+0.160
Neg logit contributions
Deal-0.094
 decision-0.089
 difference-0.088
 promise-0.083
 isn-0.083
Token and
Token ID290
Feature activation+1.52e-01

Pos logit contributions
Deal+0.037
 difference+0.031
ement+0.030
 decision+0.030
ieved+0.030
Neg logit contributions
 spilled-0.023
 blown-0.022
 Blood-0.022
 arrest-0.022
 robbers-0.022
Token had
Token ID550
Feature activation-1.50e+00

Pos logit contributions
Deal+0.004
 difference+0.004
ement+0.003
 isn+0.003
 why+0.003
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 blown-0.004
growth-0.003
Token an
Token ID281
Feature activation+1.77e+00

Pos logit contributions
 Blood+0.031
 arrest+0.030
 robbers+0.029
growth+0.029
 blown+0.029
Neg logit contributions
Deal-0.045
 difference-0.041
 isn-0.039
 exactly-0.039
 why-0.039
Token idea
Token ID2126
Feature activation-9.52e-02

Pos logit contributions
Deal+0.027
 difference+0.020
 isn+0.019
ement+0.018
 why+0.017
Neg logit contributions
 Blood-0.062
 arrest-0.061
 robbers-0.061
 blown-0.061
 mice-0.060
Token counted
Token ID14789
Feature activation-1.20e+00

Pos logit contributions
 Blood+0.057
 arrest+0.057
 blown+0.055
 robbers+0.055
growth+0.054
Neg logit contributions
Deal-0.078
 difference-0.070
ement-0.067
 isn-0.066
 why-0.066
Token that
Token ID326
Feature activation+6.33e-01

Pos logit contributions
 blown+0.026
alks+0.026
 Blood+0.025
 arrest+0.024
 robbers+0.024
Neg logit contributions
Deal-0.036
 difference-0.031
 promise-0.029
 decision-0.029
ement-0.028
Token high
Token ID1029
Feature activation+2.03e-01

Pos logit contributions
Deal+0.018
ement+0.015
 difference+0.015
ieved+0.015
 decision+0.014
Neg logit contributions
 blown-0.014
 spilled-0.013
 Blood-0.013
 mice-0.013
 arrest-0.013
Token!
Token ID0
Feature activation+4.65e-01

Pos logit contributions
Deal+0.005
 difference+0.005
ement+0.004
 promise+0.004
 exactly+0.004
Neg logit contributions
 blown-0.005
 arrest-0.005
 Blood-0.005
growth-0.005
 spilled-0.005
Token The
Token ID383
Feature activation-3.16e-01

Pos logit contributions
Deal+0.012
 difference+0.011
 decision+0.010
ement+0.010
 why+0.010
Neg logit contributions
 Blood-0.011
 arrest-0.011
growth-0.011
 robbers-0.011
 blown-0.011
Token ear
Token ID1027
Feature activation-5.35e+00

Pos logit contributions
 arrest+0.006
 Blood+0.006
 blown+0.006
 robbers+0.006
growth+0.006
Neg logit contributions
Deal-0.010
 difference-0.008
ement-0.008
 exactly-0.008
 isn-0.008
Token knew
Token ID2993
Feature activation+1.07e+00

Pos logit contributions
 Blood+0.139
 arrest+0.137
 blown+0.136
 spilled+0.131
growth+0.131
Neg logit contributions
Deal-0.134
 difference-0.118
ement-0.109
 decision-0.108
 why-0.105
Token it
Token ID340
Feature activation+2.52e-01

Pos logit contributions
Deal+0.030
 difference+0.025
ement+0.023
 decision+0.023
 Amen+0.023
Neg logit contributions
 blown-0.025
 arrest-0.023
 Blood-0.022
 spilled-0.022
 robbers-0.022
Token was
Token ID373
Feature activation+1.52e+00

Pos logit contributions
Deal+0.007
 difference+0.007
ement+0.006
 decision+0.006
 deal+0.006
Neg logit contributions
 blown-0.005
 spilled-0.005
 Blood-0.005
 robbers-0.005
 mice-0.005
Token getting
Token ID1972
Feature activation-7.57e-01

Pos logit contributions
Deal+0.048
ement+0.041
 difference+0.040
olog+0.037
oc+0.037
Neg logit contributions
 blown-0.031
 Blood-0.028
 bitten-0.027
 spilled-0.027
alks-0.026
Token stronger
Token ID7387
Feature activation+1.09e+00

Pos logit contributions
 blown+0.019
 Blood+0.018
 robbers+0.017
alks+0.017
 spilled+0.017
Neg logit contributions
Deal-0.020
 difference-0.018
ement-0.017
 decision-0.017
 isn-0.017
Token finally
Token ID3443
Feature activation-1.41e+00

Pos logit contributions
 arrest+0.015
 blown+0.015
 robbers+0.014
 spilled+0.014
 Blood+0.014
Neg logit contributions
Deal-0.020
 difference-0.017
ement-0.017
 decision-0.016
 isn-0.015
Token understood
Token ID7247
Feature activation+1.87e+00

Pos logit contributions
 arrest+0.029
 blown+0.028
 Blood+0.028
 robbers+0.027
growth+0.027
Neg logit contributions
Deal-0.043
 difference-0.038
ement-0.037
 isn-0.036
 decision-0.036
Token why
Token ID1521
Feature activation+1.23e+00

Pos logit contributions
Deal+0.057
ement+0.047
 difference+0.046
ieved+0.046
 isn+0.044
Neg logit contributions
 arrest-0.038
 blown-0.037
 robbers-0.037
 mice-0.036
growth-0.036
Token the
Token ID262
Feature activation-1.25e+00

Pos logit contributions
Deal+0.033
 difference+0.029
ement+0.028
 why+0.028
 isn+0.028
Neg logit contributions
 Blood-0.029
 arrest-0.029
 robbers-0.028
 blown-0.028
growth-0.028
Token chip
Token ID11594
Feature activation-2.32e+00

Pos logit contributions
 Blood+0.028
 arrest+0.028
 robbers+0.027
growth+0.027
 blown+0.027
Neg logit contributions
Deal-0.035
 difference-0.032
 decision-0.030
ement-0.030
 isn-0.030
Tokenm
Token ID76
Feature activation-5.35e+00

Pos logit contributions
 Blood+0.056
 arrest+0.055
 robbers+0.054
 blown+0.054
growth+0.053
Neg logit contributions
Deal-0.062
 difference-0.055
ement-0.052
 isn-0.052
 why-0.051
Tokenunk
Token ID2954
Feature activation-1.78e+00

Pos logit contributions
 robbers+0.112
growth+0.111
 arrest+0.109
 blown+0.099
 bandits+0.098
Neg logit contributions
 decision-0.149
Deal-0.148
 isn-0.144
 promise-0.142
ement-0.142
Token had
Token ID550
Feature activation-1.19e+00

Pos logit contributions
 Blood+0.047
 arrest+0.047
 blown+0.046
 robbers+0.046
 spilled+0.045
Neg logit contributions
Deal-0.042
 difference-0.037
ement-0.035
 decision-0.034
 exactly-0.033
Token been
Token ID587
Feature activation-5.64e-01

Pos logit contributions
 Blood+0.026
 arrest+0.025
growth+0.025
 robbers+0.025
alks+0.024
Neg logit contributions
Deal-0.034
 difference-0.031
 exactly-0.030
 isn-0.030
 why-0.030
Token so
Token ID523
Feature activation+1.41e+00

Pos logit contributions
 Blood+0.014
 arrest+0.014
 blown+0.014
 robbers+0.014
 mice+0.013
Neg logit contributions
Deal-0.015
 difference-0.013
ement-0.012
 isn-0.012
 decision-0.012
Token boss
Token ID6478
Feature activation-2.28e-01

Pos logit contributions
Deal+0.035
 difference+0.031
ement+0.029
 why+0.029
 exactly+0.029
Neg logit contributions
 Blood-0.036
 arrest-0.036
 robbers-0.035
growth-0.035
 blown-0.034
Token had
Token ID550
Feature activation-4.03e+00

Pos logit contributions
Deal+0.011
 difference+0.010
 decision+0.010
ement+0.009
 exactly+0.009
Neg logit contributions
 Blood-0.010
 arrest-0.009
 blown-0.009
 robbers-0.009
 spilled-0.009
Token pictures
Token ID5986
Feature activation-1.18e+00

Pos logit contributions
 Blood+0.100
plets+0.095
 robbers+0.094
 arrest+0.093
alks+0.093
Neg logit contributions
Deal-0.095
 why-0.089
 isn-0.089
 exactly-0.089
 difference-0.086
Token,
Token ID11
Feature activation-8.31e-01

Pos logit contributions
 Blood+0.029
 arrest+0.028
 robbers+0.028
 blown+0.028
growth+0.027
Neg logit contributions
Deal-0.031
 difference-0.027
ement-0.026
 isn-0.026
 why-0.026
Token some
Token ID617
Feature activation-1.59e+00

Pos logit contributions
 Blood+0.019
 arrest+0.019
 blown+0.019
alks+0.018
growth+0.018
Neg logit contributions
Deal-0.023
 why-0.020
 difference-0.020
 isn-0.020
 exactly-0.020
Token books
Token ID3835
Feature activation+4.47e-01

Pos logit contributions
 Blood+0.037
 arrest+0.037
 blown+0.036
 robbers+0.035
growth+0.035
Neg logit contributions
Deal-0.043
 difference-0.039
 isn-0.037
 why-0.037
ement-0.037
Token had
Token ID550
Feature activation-5.34e+00

Pos logit contributions
Deal+0.009
 difference+0.008
ement+0.007
 isn+0.007
 why+0.007
Neg logit contributions
 Blood-0.014
 arrest-0.013
 robbers-0.013
 blown-0.013
growth-0.013
Token stories
Token ID3923
Feature activation-1.60e+00

Pos logit contributions
 Blood+0.134
 robbers+0.126
 arrest+0.124
 blown+0.122
alks+0.122
Neg logit contributions
Deal-0.126
 isn-0.121
 exactly-0.117
 why-0.116
 difference-0.113
Token,
Token ID11
Feature activation-1.61e+00

Pos logit contributions
 Blood+0.043
 arrest+0.042
 robbers+0.041
 blown+0.041
alks+0.040
Neg logit contributions
Deal-0.039
 difference-0.033
 decision-0.031
ement-0.031
 promise-0.031
Token some
Token ID617
Feature activation-1.51e+00

Pos logit contributions
 Blood+0.040
 arrest+0.039
 blown+0.038
 robbers+0.038
growth+0.037
Neg logit contributions
Deal-0.042
 difference-0.037
 isn-0.036
 why-0.036
ement-0.035
Token books
Token ID3835
Feature activation+4.65e-01

Pos logit contributions
 Blood+0.035
 arrest+0.035
 robbers+0.034
 blown+0.034
growth+0.033
Neg logit contributions
Deal-0.041
 difference-0.037
ement-0.035
 isn-0.035
 why-0.035
Token had
Token ID550
Feature activation-5.34e+00

Pos logit contributions
Deal+0.010
 difference+0.008
ement+0.008
 isn+0.008
 why+0.008
Neg logit contributions
 Blood-0.014
 arrest-0.014
 robbers-0.014
 blown-0.013
growth-0.013
Token replied
Token ID8712
Feature activation-4.09e-01

Pos logit contributions
 difference+0.016
roads+0.014
Deal+0.014
 decision+0.014
 destinations+0.014
Neg logit contributions
 spilled-0.013
 Blood-0.012
 bitten-0.011
 mice-0.011
 blown-0.011
Token,
Token ID11
Feature activation+4.48e-01

Pos logit contributions
 Blood+0.013
 arrest+0.012
 blown+0.012
 robbers+0.012
 spilled+0.012
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.006
 decision-0.006
 isn-0.006
Token "
Token ID366
Feature activation+1.62e+00

Pos logit contributions
Deal+0.011
 difference+0.010
ement+0.009
 isn+0.009
 decision+0.009
Neg logit contributions
 Blood-0.012
 arrest-0.012
 robbers-0.011
 blown-0.011
growth-0.011
TokenSure
Token ID19457
Feature activation+2.05e+00

Pos logit contributions
 difference+0.030
ement+0.030
Deal+0.029
 decision+0.029
 exactly+0.028
Neg logit contributions
 Blood-0.048
 arrest-0.044
 blown-0.044
 mice-0.044
 robbers-0.044
Token,
Token ID11
Feature activation+2.74e-01

Pos logit contributions
Deal+0.050
 difference+0.043
 int+0.042
ement+0.042
ieved+0.040
Neg logit contributions
 arrest-0.054
 Blood-0.054
 robbers-0.052
 mice-0.051
 spilled-0.049
Token but
Token ID475
Feature activation+4.37e+00

Pos logit contributions
Deal+0.008
ement+0.007
 difference+0.007
 decision+0.007
 int+0.007
Neg logit contributions
 Blood-0.006
 robbers-0.006
 arrest-0.006
 mice-0.006
 spilled-0.005
Token first
Token ID717
Feature activation+1.93e+00

Pos logit contributions
Deal+0.111
ement+0.105
 difference+0.101
 int+0.095
ieved+0.094
Neg logit contributions
 Blood-0.104
 robbers-0.098
 blown-0.098
 arrest-0.097
 mice-0.096
Token you
Token ID345
Feature activation+1.10e+00

Pos logit contributions
Deal+0.049
ement+0.046
 int+0.045
 difference+0.043
ieved+0.042
Neg logit contributions
 mice-0.045
 robbers-0.044
 arrest-0.043
 Blood-0.043
 spilled-0.043
Token need
Token ID761
Feature activation+5.29e-01

Pos logit contributions
Deal+0.033
 difference+0.029
ement+0.028
 isn+0.028
 why+0.028
Neg logit contributions
 Blood-0.023
 arrest-0.023
 robbers-0.022
 blown-0.022
growth-0.022
Token to
Token ID284
Feature activation+9.05e-01

Pos logit contributions
Deal+0.013
 why+0.012
 difference+0.012
 isn+0.011
 exactly+0.011
Neg logit contributions
 Blood-0.013
 arrest-0.013
growth-0.013
plets-0.013
 robbers-0.013
Token yield
Token ID7800
Feature activation+4.96e-01

Pos logit contributions
Deal+0.026
ement+0.024
 difference+0.023
 isn+0.022
 exactly+0.022
Neg logit contributions
 arrest-0.020
 Blood-0.019
 robbers-0.019
 blown-0.018
 mice-0.018
Token never
Token ID1239
Feature activation-1.30e+00

Pos logit contributions
Deal+0.003
 difference+0.002
ement+0.002
 decision+0.002
ieved+0.002
Neg logit contributions
 arrest-0.002
 Blood-0.002
 robbers-0.002
 blown-0.002
 spilled-0.002
Token take
Token ID1011
Feature activation-7.94e-01

Pos logit contributions
 Blood+0.027
 arrest+0.027
 blown+0.027
 robbers+0.027
 spilled+0.026
Neg logit contributions
Deal-0.039
 difference-0.034
ement-0.034
 exactly-0.032
 decision-0.032
Token something
Token ID1223
Feature activation+2.07e+00

Pos logit contributions
 Blood+0.019
 arrest+0.019
 robbers+0.018
 blown+0.018
growth+0.018
Neg logit contributions
Deal-0.021
 difference-0.019
ement-0.018
 isn-0.018
 why-0.018
Token that
Token ID326
Feature activation+6.97e-01

Pos logit contributions
Deal+0.055
 difference+0.045
ement+0.045
 isn+0.042
 decision+0.042
Neg logit contributions
 blown-0.051
 arrest-0.051
 Blood-0.051
 robbers-0.050
 spilled-0.050
Token doesn
Token ID1595
Feature activation-1.19e+00

Pos logit contributions
Deal+0.015
ement+0.013
 difference+0.012
ieved+0.012
 exactly+0.012
Neg logit contributions
 spilled-0.020
 Blood-0.019
 blown-0.019
 arrest-0.019
 robbers-0.018
Tokenâ
Token ID22940
Feature activation+4.16e+00

Pos logit contributions
 Blood+0.022
 robbers+0.022
 arrest+0.021
growth+0.021
 blown+0.021
Neg logit contributions
Deal-0.037
 difference-0.034
 isn-0.034
 why-0.033
ement-0.033
Token
Token ID26391
Feature activation-1.61e+00

Pos logit contributions
Deal+0.071
 difference+0.065
 exactly+0.063
ement+0.060
 why+0.058
Neg logit contributions
 Blood-0.133
 arrest-0.132
 robbers-0.128
 blown-0.128
growth-0.126
Token
Token ID8151
Feature activation-1.26e-01

Pos logit contributions
 Blood+0.038
 arrest+0.037
 robbers+0.036
 blown+0.036
growth+0.035
Neg logit contributions
Deal-0.044
 difference-0.039
ement-0.037
 why-0.037
 decision-0.037
Tokent
Token ID83
Feature activation-4.25e-01

Pos logit contributions
 Blood+0.002
 arrest+0.002
 robbers+0.002
 blown+0.002
growth+0.002
Neg logit contributions
Deal-0.004
 difference-0.004
ement-0.004
 isn-0.004
 why-0.004
Token belong
Token ID5594
Feature activation+6.62e-01

Pos logit contributions
 Blood+0.012
 arrest+0.012
growth+0.012
 robbers+0.012
alks+0.012
Neg logit contributions
Deal-0.009
 difference-0.008
 why-0.008
 isn-0.008
 promise-0.008
Token to
Token ID284
Feature activation+3.21e-01

Pos logit contributions
Deal+0.021
 difference+0.020
ement+0.019
 decision+0.018
 int+0.018
Neg logit contributions
 blown-0.012
 arrest-0.012
 Blood-0.012
 robbers-0.011
 spilled-0.011
Token said
Token ID531
Feature activation+1.09e-01

Pos logit contributions
Deal+0.023
 difference+0.020
ement+0.019
 isn+0.019
 decision+0.019
Neg logit contributions
 Blood-0.028
 arrest-0.028
 robbers-0.028
 blown-0.028
growth-0.027
Token,
Token ID11
Feature activation+6.34e-01

Pos logit contributions
Deal+0.003
 difference+0.002
 decision+0.002
ement+0.002
ieved+0.002
Neg logit contributions
 Blood-0.003
 spilled-0.003
 arrest-0.003
 blown-0.003
 mice-0.003
Token "
Token ID366
Feature activation+1.90e+00

Pos logit contributions
Deal+0.016
 difference+0.014
ement+0.013
 isn+0.013
 decision+0.013
Neg logit contributions
 Blood-0.017
 arrest-0.016
 robbers-0.016
 blown-0.016
growth-0.016
TokenOkay
Token ID16454
Feature activation+8.96e-01

Pos logit contributions
Deal+0.036
 difference+0.033
ement+0.031
 decision+0.030
 why+0.030
Neg logit contributions
 Blood-0.059
 arrest-0.057
 robbers-0.057
 blown-0.057
growth-0.056
Token,
Token ID11
Feature activation-3.79e-01

Pos logit contributions
Deal+0.024
 difference+0.020
ement+0.020
roads+0.019
 int+0.019
Neg logit contributions
 Blood-0.021
 robbers-0.021
 arrest-0.021
 mice-0.021
 spilled-0.020
Token but
Token ID475
Feature activation+3.97e+00

Pos logit contributions
 robbers+0.008
 arrest+0.007
 mice+0.007
 Blood+0.007
 blown+0.007
Neg logit contributions
Deal-0.011
ement-0.011
 difference-0.010
roads-0.010
 decision-0.010
Token don
Token ID836
Feature activation-4.37e-01

Pos logit contributions
ement+0.109
Deal+0.108
 difference+0.098
ieved+0.097
uct+0.093
Neg logit contributions
 robbers-0.084
 Blood-0.084
 arrest-0.081
 blown-0.080
 mice-0.079
Token't
Token ID470
Feature activation+1.62e-01

Pos logit contributions
 Blood+0.008
 robbers+0.007
 arrest+0.007
growth+0.007
 blown+0.007
Neg logit contributions
Deal-0.014
 difference-0.013
 isn-0.013
ement-0.013
 why-0.013
Token forget
Token ID6044
Feature activation+6.89e-01

Pos logit contributions
Deal+0.004
 difference+0.004
ement+0.003
 isn+0.003
 why+0.003
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 blown-0.004
growth-0.004
Token to
Token ID284
Feature activation+1.24e+00

Pos logit contributions
Deal+0.024
ement+0.021
 difference+0.021
 decision+0.020
ieved+0.020
Neg logit contributions
 arrest-0.011
 Blood-0.011
 mice-0.011
 robbers-0.011
 spilled-0.010
Token put
Token ID1234
Feature activation+1.07e+00

Pos logit contributions
Deal+0.034
ement+0.032
 difference+0.030
 isn+0.029
 exactly+0.028
Neg logit contributions
 arrest-0.029
 robbers-0.028
 Blood-0.027
 blown-0.027
growth-0.026
Token,
Token ID11
Feature activation+7.10e-01

Pos logit contributions
 arrest+0.002
 mice+0.002
 blown+0.002
 spilled+0.002
alks+0.002
Neg logit contributions
Deal-0.004
 difference-0.003
ement-0.003
ieved-0.003
 int-0.003
Tokenâ
Token ID22940
Feature activation+5.11e+00

Pos logit contributions
Deal+0.021
ement+0.018
 difference+0.017
ieved+0.016
 decision+0.016
Neg logit contributions
 arrest-0.016
 robbers-0.016
 blown-0.015
 spilled-0.015
 mice-0.015
Token
Token ID26391
Feature activation+1.93e-01

Pos logit contributions
 exactly+0.097
 difference+0.094
Deal+0.090
 amaz+0.088
 why+0.087
Neg logit contributions
 Blood-0.145
 arrest-0.136
plets-0.135
 blown-0.132
growth-0.132
Token said
Token ID531
Feature activation+8.87e-01

Pos logit contributions
Deal+0.005
 difference+0.005
ement+0.004
 decision+0.004
 why+0.004
Neg logit contributions
 arrest-0.005
 Blood-0.005
 robbers-0.004
 blown-0.004
growth-0.004
Token Ali
Token ID12104
Feature activation+3.03e+00

Pos logit contributions
Deal+0.026
 difference+0.023
ieved+0.022
ement+0.022
uct+0.020
Neg logit contributions
 Blood-0.018
 arrest-0.017
 spilled-0.017
 blown-0.017
 Po-0.017
Tokenâ
Token ID22940
Feature activation+5.55e+00

Pos logit contributions
Deal+0.085
 difference+0.073
ement+0.071
ieved+0.066
 decision+0.065
Neg logit contributions
 arrest-0.069
 Blood-0.069
 robbers-0.069
 mice-0.067
 blown-0.067
Token
Token ID26391
Feature activation+6.29e-01

Pos logit contributions
 exactly+0.111
 difference+0.106
Deal+0.101
 why+0.099
 amaz+0.099
Neg logit contributions
 Blood-0.151
 robbers-0.144
 arrest-0.142
growth-0.139
 blown-0.137
Token
Token ID8151
Feature activation+2.17e+00

Pos logit contributions
Deal+0.018
 difference+0.017
ement+0.016
 decision+0.016
 why+0.016
Neg logit contributions
 arrest-0.014
 Blood-0.014
 robbers-0.013
growth-0.012
 blown-0.012
Tokens
Token ID82
Feature activation+2.07e+00

Pos logit contributions
Deal+0.070
 difference+0.063
ement+0.061
 isn+0.060
 why+0.060
Neg logit contributions
 Blood-0.040
 arrest-0.039
 robbers-0.037
 blown-0.037
growth-0.036
Token mom
Token ID1995
Feature activation+2.06e-01

Pos logit contributions
Deal+0.052
ement+0.043
ieved+0.042
 difference+0.041
uct+0.040
Neg logit contributions
 Blood-0.055
 arrest-0.055
alks-0.054
 robbers-0.053
 blown-0.053
Token.
Token ID13
Feature activation+1.03e+00

Pos logit contributions
Deal+0.006
 difference+0.005
ement+0.005
 decision+0.005
 deal+0.005
Neg logit contributions
 robbers-0.004
 Blood-0.004
 arrest-0.004
 spilled-0.004
 mice-0.004
Token today
Token ID1909
Feature activation+4.19e-01

Pos logit contributions
Deal+0.061
 difference+0.058
ement+0.056
 decision+0.054
ieved+0.051
Neg logit contributions
 Blood-0.043
 robbers-0.042
 spilled-0.040
 mice-0.038
 arrest-0.038
Token,
Token ID11
Feature activation+3.94e-01

Pos logit contributions
Deal+0.012
ieved+0.010
ement+0.010
uct+0.010
 int+0.010
Neg logit contributions
 spilled-0.010
 Blood-0.010
 arrest-0.009
 mice-0.009
 robbers-0.009
Token they
Token ID484
Feature activation-3.63e-01

Pos logit contributions
Deal+0.013
 difference+0.012
ement+0.011
 decision+0.011
 promise+0.011
Neg logit contributions
 robbers-0.007
 Blood-0.007
 arrest-0.007
 mice-0.007
 blown-0.007
Token have
Token ID423
Feature activation-1.63e+00

Pos logit contributions
 arrest+0.008
 robbers+0.008
 Blood+0.008
 spilled+0.008
 blown+0.008
Neg logit contributions
Deal-0.010
ement-0.009
 difference-0.009
 exactly-0.008
 decision-0.008
Token to
Token ID284
Feature activation+5.06e-01

Pos logit contributions
 Blood+0.039
plets+0.038
alks+0.038
growth+0.038
 arrest+0.038
Neg logit contributions
Deal-0.041
 why-0.037
 isn-0.037
 exactly-0.036
 difference-0.036
Token un
Token ID555
Feature activation+3.93e+00

Pos logit contributions
Deal+0.014
ement+0.013
 difference+0.012
 isn+0.012
 exactly+0.011
Neg logit contributions
 arrest-0.012
 robbers-0.012
 Blood-0.012
growth-0.011
 blown-0.011
Tokenpack
Token ID8002
Feature activation+1.48e+00

Pos logit contributions
 difference+0.119
Deal+0.117
 isn+0.113
ement+0.111
 decision+0.109
Neg logit contributions
 robbers-0.076
 Blood-0.074
 arrest-0.074
 blown-0.068
 mice-0.064
Token.
Token ID13
Feature activation-5.69e-01

Pos logit contributions
Deal+0.051
 difference+0.046
ieved+0.043
ement+0.043
uct+0.042
Neg logit contributions
 spilled-0.024
 Blood-0.023
 mice-0.023
 robbers-0.023
alks-0.022
Token Their
Token ID5334
Feature activation+9.10e-01

Pos logit contributions
 arrest+0.013
 Blood+0.013
 blown+0.012
 robbers+0.012
growth+0.012
Neg logit contributions
Deal-0.016
 difference-0.014
 isn-0.014
ement-0.014
 why-0.013
Token family
Token ID1641
Feature activation-5.49e-01

Pos logit contributions
Deal+0.030
ieved+0.028
ement+0.027
 difference+0.027
uct+0.026
Neg logit contributions
 robbers-0.017
 Blood-0.017
 spilled-0.016
 mice-0.016
 arrest-0.015
Token moved
Token ID3888
Feature activation+3.97e-01

Pos logit contributions
 spilled+0.015
 robbers+0.014
 arrest+0.014
 mice+0.013
 blown+0.013
Neg logit contributions
Deal-0.013
ement-0.012
 difference-0.011
ieved-0.011
uct-0.011
Token bed
Token ID3996
Feature activation+4.28e-01

Pos logit contributions
 difference+0.023
 promise+0.021
 decision+0.020
Deal+0.019
 wish+0.019
Neg logit contributions
plets-0.029
 Blood-0.028
 blown-0.028
 robbers-0.027
 bitten-0.026
Token.
Token ID13
Feature activation-1.69e+00

Pos logit contributions
Deal+0.008
ement+0.007
 exactly+0.007
 difference+0.007
 decision+0.006
Neg logit contributions
 robbers-0.013
 arrest-0.013
 Blood-0.013
growth-0.013
 blown-0.013
Token The
Token ID383
Feature activation-5.31e-01

Pos logit contributions
 arrest+0.037
 blown+0.037
 robbers+0.036
growth+0.035
 Blood+0.035
Neg logit contributions
Deal-0.048
 isn-0.042
 difference-0.041
 why-0.041
 decision-0.039
Token gem
Token ID16840
Feature activation-1.09e+00

Pos logit contributions
 arrest+0.013
 Blood+0.013
growth+0.013
 robbers+0.012
 blown+0.012
Neg logit contributions
Deal-0.014
 difference-0.013
 isn-0.012
ement-0.012
 decision-0.012
Token was
Token ID373
Feature activation-1.07e+00

Pos logit contributions
 Blood+0.029
 arrest+0.029
 robbers+0.028
 blown+0.028
growth+0.028
Neg logit contributions
Deal-0.026
 difference-0.023
ement-0.022
 isn-0.022
 why-0.021
Token shiny
Token ID22441
Feature activation+8.34e-01

Pos logit contributions
 Blood+0.027
 blown+0.026
 arrest+0.026
 robbers+0.026
 spilled+0.025
Neg logit contributions
Deal-0.028
 difference-0.024
ement-0.023
 isn-0.023
 decision-0.022
Token and
Token ID290
Feature activation-8.62e-01

Pos logit contributions
Deal+0.023
 difference+0.021
ement+0.020
 decision+0.020
 why+0.019
Neg logit contributions
 Blood-0.019
 arrest-0.018
 blown-0.018
 robbers-0.018
 mice-0.018
Token blue
Token ID4171
Feature activation+6.10e-02

Pos logit contributions
 Blood+0.017
 blown+0.016
 robbers+0.016
 spilled+0.015
 mice+0.015
Neg logit contributions
Deal-0.027
 difference-0.025
ement-0.024
 decision-0.024
ieved-0.023
Token.
Token ID13
Feature activation-1.37e+00

Pos logit contributions
Deal+0.001
 isn+0.001
ement+0.001
 difference+0.001
 why+0.001
Neg logit contributions
 arrest-0.002
 Blood-0.002
 robbers-0.002
growth-0.002
alks-0.002
Token Kitty
Token ID21168
Feature activation-2.86e-01

Pos logit contributions
 arrest+0.031
 blown+0.031
growth+0.030
 robbers+0.030
 Blood+0.030
Neg logit contributions
Deal-0.038
 isn-0.033
 difference-0.032
 why-0.032
ement-0.031
Token showed
Token ID3751
Feature activation+3.13e-01

Pos logit contributions
 Blood+0.007
 arrest+0.007
 robbers+0.007
 blown+0.007
 spilled+0.007
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.007
 decision-0.006
 exactly-0.006
Token<|endoftext|>
Token ID50256
Feature activation+8.72e-01

Pos logit contributions
Deal+0.044
 difference+0.038
ement+0.038
 decision+0.037
 promise+0.037
Neg logit contributions
 Blood-0.038
 arrest-0.037
growth-0.036
 robbers-0.036
alks-0.035
TokenOnce
Token ID7454
Feature activation+1.74e+00

Pos logit contributions
Deal+0.021
 difference+0.019
ement+0.018
 isn+0.018
 why+0.018
Neg logit contributions
 Blood-0.023
 arrest-0.022
 robbers-0.022
 blown-0.022
growth-0.021
Token upon
Token ID2402
Feature activation+1.28e-01

Pos logit contributions
Deal+0.051
 difference+0.046
ement+0.044
 isn+0.044
 why+0.043
Neg logit contributions
 Blood-0.037
 arrest-0.036
 robbers-0.035
 blown-0.035
growth-0.034
Token a
Token ID257
Feature activation+1.23e+00

Pos logit contributions
Deal+0.004
 why+0.003
 difference+0.003
 exactly+0.003
 isn+0.003
Neg logit contributions
alks-0.003
 arrest-0.003
 blown-0.003
 robbers-0.003
 Blood-0.003
Token time
Token ID640
Feature activation+1.52e-01

Pos logit contributions
Deal+0.032
 difference+0.029
ement+0.029
 why+0.028
 isn+0.028
Neg logit contributions
 Blood-0.029
 arrest-0.027
 spilled-0.027
growth-0.026
 robbers-0.026
Token,
Token ID11
Feature activation+1.03e+00

Pos logit contributions
Deal+0.003
 difference+0.003
 decision+0.003
ement+0.003
 promise+0.003
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 spilled-0.004
 blown-0.004
Token there
Token ID612
Feature activation-4.76e-02

Pos logit contributions
Deal+0.028
 difference+0.025
ement+0.024
 isn+0.024
 decision+0.024
Neg logit contributions
 Blood-0.024
 arrest-0.023
 robbers-0.023
 blown-0.023
growth-0.022
Token was
Token ID373
Feature activation-1.70e+00

Pos logit contributions
 Blood+0.001
 spilled+0.001
 arrest+0.001
growth+0.001
 blown+0.001
Neg logit contributions
Deal-0.001
 difference-0.001
 int-0.001
ieved-0.001
izes-0.001
Token a
Token ID257
Feature activation+1.11e-01

Pos logit contributions
alks+0.042
plets+0.041
 arrest+0.041
 robbers+0.040
 blown+0.040
Neg logit contributions
Deal-0.043
 exactly-0.037
 why-0.036
 equally-0.035
 difference-0.035
Token rich
Token ID5527
Feature activation+6.61e-01

Pos logit contributions
Deal+0.003
 difference+0.003
ement+0.003
 isn+0.003
 why+0.003
Neg logit contributions
 Blood-0.002
 arrest-0.002
 blown-0.002
 robbers-0.002
growth-0.002
Token man
Token ID582
Feature activation-4.92e-02

Pos logit contributions
Deal+0.016
 difference+0.015
 decision+0.014
 promise+0.013
ement+0.013
Neg logit contributions
 Blood-0.018
growth-0.017
 blown-0.017
 arrest-0.017
alks-0.016
TokenM
Token ID44
Feature activation+2.93e+00

Pos logit contributions
Deal+0.018
 difference+0.016
ement+0.015
 isn+0.015
 why+0.015
Neg logit contributions
 Blood-0.012
 arrest-0.011
 robbers-0.011
 blown-0.011
growth-0.011
Tokenummy
Token ID13513
Feature activation+2.89e+00

Pos logit contributions
Deal+0.079
 difference+0.068
 isn+0.064
 why+0.063
 decision+0.062
Neg logit contributions
 Blood-0.069
alks-0.068
 arrest-0.067
 robbers-0.065
 spilled-0.064
Token and
Token ID290
Feature activation-1.48e-01

Pos logit contributions
Deal+0.078
 difference+0.069
ement+0.066
 decision+0.065
 isn+0.064
Neg logit contributions
 Blood-0.069
 arrest-0.068
 robbers-0.066
 blown-0.066
growth-0.065
Token Daddy
Token ID26926
Feature activation+1.68e+00

Pos logit contributions
 Blood+0.003
 arrest+0.003
 blown+0.003
 robbers+0.003
growth+0.003
Neg logit contributions
Deal-0.004
 difference-0.004
 why-0.004
 isn-0.003
ement-0.003
Token took
Token ID1718
Feature activation-6.13e-01

Pos logit contributions
Deal+0.050
 difference+0.044
ement+0.043
 decision+0.042
 isn+0.042
Neg logit contributions
 Blood-0.035
 arrest-0.035
 robbers-0.034
 blown-0.034
growth-0.033
Token Peter
Token ID5613
Feature activation+1.18e+00

Pos logit contributions
 Blood+0.014
 arrest+0.014
 blown+0.014
 robbers+0.014
growth+0.013
Neg logit contributions
Deal-0.016
 difference-0.015
ement-0.014
 isn-0.014
 why-0.014
Token on
Token ID319
Feature activation-2.30e-01

Pos logit contributions
Deal+0.034
 difference+0.030
ement+0.029
 why+0.029
 isn+0.029
Neg logit contributions
 Blood-0.026
 arrest-0.025
 robbers-0.025
 blown-0.024
growth-0.024
Token a
Token ID257
Feature activation+1.09e+00

Pos logit contributions
 arrest+0.004
 Blood+0.004
 robbers+0.004
growth+0.004
 blown+0.004
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.007
 isn-0.007
 promise-0.007
Token special
Token ID2041
Feature activation+3.54e+00

Pos logit contributions
Deal+0.034
 difference+0.029
ement+0.029
 isn+0.028
 exactly+0.028
Neg logit contributions
 Blood-0.022
 arrest-0.022
 robbers-0.021
 blown-0.021
growth-0.021
Token adventure
Token ID8855
Feature activation+1.79e+00

Pos logit contributions
Deal+0.100
ement+0.084
 exactly+0.082
 difference+0.082
 isn+0.081
Neg logit contributions
 robbers-0.081
 arrest-0.081
 Blood-0.079
 spilled-0.078
growth-0.077
Token.
Token ID13
Feature activation-9.11e-01

Pos logit contributions
Deal+0.053
 difference+0.048
ement+0.045
 isn+0.045
 promise+0.044
Neg logit contributions
 arrest-0.037
 robbers-0.036
 blown-0.036
 Blood-0.035
alks-0.035
Token knew
Token ID2993
Feature activation+4.56e-01

Pos logit contributions
 spilled+0.015
 Blood+0.015
 arrest+0.015
 blown+0.015
 robbers+0.014
Neg logit contributions
Deal-0.019
 difference-0.018
ement-0.017
 decision-0.017
 why-0.016
Token her
Token ID607
Feature activation+1.62e+00

Pos logit contributions
Deal+0.013
 difference+0.012
 decision+0.012
ement+0.011
 promise+0.011
Neg logit contributions
 arrest-0.009
 robbers-0.009
 Blood-0.009
 blown-0.009
 mice-0.009
Token dad
Token ID9955
Feature activation-8.92e-01

Pos logit contributions
Deal+0.055
 difference+0.048
 exactly+0.047
ieved+0.047
ement+0.047
Neg logit contributions
 robbers-0.027
 arrest-0.025
 Blood-0.024
 spilled-0.024
 blown-0.023
Token was
Token ID373
Feature activation+1.71e+00

Pos logit contributions
 spilled+0.021
 mice+0.020
 blown+0.019
 Blood+0.019
 arrest+0.019
Neg logit contributions
Deal-0.022
 difference-0.021
 int-0.020
ement-0.019
 decision-0.019
Token right
Token ID826
Feature activation+3.49e+00

Pos logit contributions
Deal+0.039
 difference+0.033
ement+0.032
 isn+0.032
 decision+0.032
Neg logit contributions
 blown-0.048
 Blood-0.047
 robbers-0.047
 arrest-0.046
 mice-0.045
Token.
Token ID13
Feature activation+7.05e-01

Pos logit contributions
Deal+0.117
 difference+0.109
ieved+0.102
 isn+0.102
olog+0.101
Neg logit contributions
 mice-0.049
 robbers-0.048
 spilled-0.047
 bushes-0.046
 blown-0.044
Token She
Token ID1375
Feature activation-2.59e-01

Pos logit contributions
Deal+0.023
 difference+0.020
ement+0.020
 decision+0.020
 promise+0.019
Neg logit contributions
 Blood-0.013
 arrest-0.013
 robbers-0.013
growth-0.012
 blown-0.012
Token still
Token ID991
Feature activation+3.14e-01

Pos logit contributions
 spilled+0.006
 blown+0.006
 Blood+0.006
 arrest+0.006
 robbers+0.006
Neg logit contributions
Deal-0.007
 difference-0.007
ement-0.006
 decision-0.006
 isn-0.006
Token loved
Token ID6151
Feature activation-5.60e-01

Pos logit contributions
Deal+0.009
ement+0.009
 difference+0.009
 decision+0.008
 int+0.008
Neg logit contributions
 Blood-0.006
 spilled-0.006
 blown-0.006
 robbers-0.006
 arrest-0.006
Token to
Token ID284
Feature activation+8.98e-02

Pos logit contributions
 Blood+0.012
 arrest+0.012
 robbers+0.012
 blown+0.011
 mice+0.011
Neg logit contributions
Deal-0.016
 difference-0.015
ement-0.014
 isn-0.014
 decision-0.014
Token run
Token ID1057
Feature activation-9.74e-01

Pos logit contributions
Deal+0.003
ement+0.002
 difference+0.002
 exactly+0.002
 isn+0.002
Neg logit contributions
 arrest-0.002
 robbers-0.002
 Blood-0.002
 blown-0.002
growth-0.002
Token and
Token ID290
Feature activation+4.35e-01

Pos logit contributions
Deal+0.024
 difference+0.021
 isn+0.020
 decision+0.019
 promise+0.019
Neg logit contributions
 Blood-0.018
 blown-0.017
 arrest-0.017
growth-0.017
alks-0.017
Token sw
Token ID1509
Feature activation+1.88e+00

Pos logit contributions
Deal+0.013
 difference+0.011
ement+0.011
 decision+0.011
 isn+0.011
Neg logit contributions
 Blood-0.009
 arrest-0.009
 blown-0.009
 robbers-0.009
growth-0.009
Tokenam
Token ID321
Feature activation+3.50e-01

Pos logit contributions
Deal+0.049
 difference+0.044
 why+0.040
ement+0.039
 decision+0.039
Neg logit contributions
 Blood-0.048
alks-0.047
 arrest-0.046
 blown-0.045
growth-0.045
Token to
Token ID284
Feature activation+1.53e+00

Pos logit contributions
Deal+0.012
 difference+0.010
ement+0.010
uct+0.009
aman+0.009
Neg logit contributions
 Blood-0.007
 blown-0.006
 spilled-0.006
 robbers-0.006
alks-0.006
Token the
Token ID262
Feature activation+1.19e+00

Pos logit contributions
Deal+0.040
 difference+0.036
 why+0.035
 exactly+0.034
 isn+0.034
Neg logit contributions
 Blood-0.037
 arrest-0.036
 blown-0.036
 robbers-0.035
alks-0.035
Token pearl
Token ID43836
Feature activation+1.60e+00

Pos logit contributions
Deal+0.036
ement+0.031
 difference+0.030
 isn+0.030
 exactly+0.029
Neg logit contributions
 arrest-0.025
 robbers-0.025
 Blood-0.025
 blown-0.024
growth-0.024
Token.
Token ID13
Feature activation-7.06e-01

Pos logit contributions
Deal+0.050
 difference+0.042
 isn+0.040
ieved+0.040
ement+0.040
Neg logit contributions
alks-0.031
 blown-0.031
 Blood-0.031
 arrest-0.031
 spilled-0.030
Token It
Token ID632
Feature activation-1.67e-01

Pos logit contributions
 Blood+0.016
 arrest+0.016
 blown+0.016
 robbers+0.016
growth+0.015
Neg logit contributions
Deal-0.019
 difference-0.017
 isn-0.017
ement-0.017
 why-0.016
Token was
Token ID373
Feature activation-3.71e-01

Pos logit contributions
 blown+0.004
 spilled+0.004
 Blood+0.004
 arrest+0.003
growth+0.003
Neg logit contributions
Deal-0.005
 difference-0.004
ement-0.004
 decision-0.004
 isn-0.004
Token hard
Token ID1327
Feature activation+1.12e+00

Pos logit contributions
 blown+0.009
 Blood+0.009
 spilled+0.009
 robbers+0.008
 mice+0.008
Neg logit contributions
Deal-0.011
 difference-0.009
ement-0.009
 decision-0.008
 promise-0.008
Token,
Token ID11
Feature activation-6.71e-02

Pos logit contributions
Deal+0.027
 difference+0.025
 isn+0.024
 exactly+0.023
ement+0.023
Neg logit contributions
 Blood-0.029
 arrest-0.028
 robbers-0.028
growth-0.028
 blown-0.027
Token were
Token ID547
Feature activation+8.12e-01

Pos logit contributions
Deal+0.036
 difference+0.033
ement+0.032
 decision+0.031
 exactly+0.031
Neg logit contributions
 arrest-0.023
 Blood-0.022
 robbers-0.022
 blown-0.022
 spilled-0.022
Token brother
Token ID3956
Feature activation-1.83e+00

Pos logit contributions
Deal+0.023
 difference+0.020
ement+0.019
 isn+0.019
 why+0.019
Neg logit contributions
 Blood-0.019
 arrest-0.018
 robbers-0.018
 blown-0.018
growth-0.017
Token and
Token ID290
Feature activation-1.94e+00

Pos logit contributions
 robbers+0.039
 arrest+0.038
 mice+0.038
 Blood+0.038
 spilled+0.036
Neg logit contributions
Deal-0.053
 difference-0.047
 decision-0.045
ement-0.044
 isn-0.044
Token sister
Token ID6621
Feature activation-1.38e+00

Pos logit contributions
growth+0.053
plets+0.052
 Blood+0.052
alks+0.052
 blown+0.051
Neg logit contributions
Deal-0.047
 why-0.041
 promise-0.038
 exactly-0.037
 replied-0.036
Token.
Token ID13
Feature activation-1.05e+00

Pos logit contributions
 robbers+0.027
 arrest+0.026
 spilled+0.026
 mice+0.026
 Blood+0.025
Neg logit contributions
Deal-0.042
 difference-0.037
ement-0.036
 decision-0.036
ieved-0.035
Token They
Token ID1119
Feature activation-1.03e+00

Pos logit contributions
 arrest+0.020
 Blood+0.020
 blown+0.020
 robbers+0.019
growth+0.019
Neg logit contributions
Deal-0.033
 difference-0.029
ement-0.028
 isn-0.028
 why-0.028
Token liked
Token ID8288
Feature activation-1.18e+00

Pos logit contributions
 Blood+0.028
 arrest+0.028
 robbers+0.027
 blown+0.027
growth+0.027
Neg logit contributions
Deal-0.024
 difference-0.021
ement-0.020
 isn-0.020
 why-0.020
Token to
Token ID284
Feature activation-3.53e-01

Pos logit contributions
 Blood+0.027
 arrest+0.026
growth+0.026
 blown+0.025
plets+0.025
Neg logit contributions
Deal-0.032
 difference-0.029
 why-0.029
 isn-0.028
 exactly-0.028
Token play
Token ID711
Feature activation-6.54e-02

Pos logit contributions
 Blood+0.009
 arrest+0.009
 blown+0.008
 robbers+0.008
growth+0.008
Neg logit contributions
Deal-0.009
 difference-0.008
ement-0.008
 isn-0.008
 why-0.008
Token in
Token ID287
Feature activation+6.77e-02

Pos logit contributions
 Blood+0.002
 mice+0.002
 robbers+0.002
 arrest+0.001
 spilled+0.001
Neg logit contributions
Deal-0.002
 difference-0.002
 decision-0.002
ement-0.002
uct-0.002
Token the
Token ID262
Feature activation-2.81e-01

Pos logit contributions
Deal+0.001
 difference+0.001
 why+0.001
 isn+0.001
 decision+0.001
Neg logit contributions
plets-0.002
 Blood-0.002
 blown-0.002
 arrest-0.002
alks-0.002
Token way
Token ID835
Feature activation+3.78e-01

Pos logit contributions
Deal+0.037
ement+0.031
 difference+0.030
 exactly+0.030
 isn+0.030
Neg logit contributions
 arrest-0.041
 robbers-0.039
 Blood-0.039
 blown-0.039
 mice-0.038
Token home
Token ID1363
Feature activation-1.21e+00

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 why+0.007
Neg logit contributions
 Blood-0.011
 arrest-0.011
 robbers-0.011
 blown-0.011
growth-0.010
Token,
Token ID11
Feature activation+6.15e-01

Pos logit contributions
 Blood+0.035
 arrest+0.034
 spilled+0.034
 blown+0.034
 robbers+0.033
Neg logit contributions
Deal-0.027
 difference-0.023
ement-0.021
 isn-0.020
 decision-0.020
Token they
Token ID484
Feature activation-2.04e-01

Pos logit contributions
Deal+0.017
 difference+0.016
ement+0.015
 isn+0.014
 decision+0.014
Neg logit contributions
 Blood-0.014
 arrest-0.014
 robbers-0.013
 blown-0.013
 spilled-0.013
Token got
Token ID1392
Feature activation-8.81e-01

Pos logit contributions
 Blood+0.005
 arrest+0.005
 robbers+0.004
 blown+0.004
growth+0.004
Neg logit contributions
Deal-0.006
 difference-0.005
ement-0.005
 isn-0.005
 why-0.005
Token into
Token ID656
Feature activation+4.84e-01

Pos logit contributions
growth+0.023
 arrest+0.022
 Blood+0.022
 robbers+0.022
alks+0.021
Neg logit contributions
Deal-0.021
 exactly-0.020
 difference-0.019
 why-0.019
 promise-0.019
Token a
Token ID257
Feature activation+7.82e-01

Pos logit contributions
Deal+0.011
 difference+0.011
 decision+0.011
 promise+0.010
 exactly+0.010
Neg logit contributions
 Blood-0.012
 robbers-0.012
plets-0.012
 blown-0.012
 arrest-0.012
Token car
Token ID1097
Feature activation-1.52e-01

Pos logit contributions
Deal+0.026
ement+0.023
ieved+0.022
uct+0.021
 isn+0.021
Neg logit contributions
 spilled-0.015
 robbers-0.015
 arrest-0.014
 Blood-0.014
 mice-0.014
Token accident
Token ID5778
Feature activation+3.76e-01

Pos logit contributions
 arrest+0.003
 blown+0.003
 robbers+0.003
 Blood+0.003
 spilled+0.003
Neg logit contributions
Deal-0.005
 difference-0.004
ement-0.004
 isn-0.004
 decision-0.004
Token.
Token ID13
Feature activation-2.94e-01

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 exactly+0.007
Neg logit contributions
 Blood-0.010
 arrest-0.010
 robbers-0.010
growth-0.010
alks-0.010
Token Sarah
Token ID10490
Feature activation-4.12e-01

Pos logit contributions
 arrest+0.008
 blown+0.008
 robbers+0.008
growth+0.008
alks+0.008
Neg logit contributions
Deal-0.006
 isn-0.006
 difference-0.006
ement-0.006
 why-0.005
Token Mark
Token ID2940
Feature activation-1.59e-01

Pos logit contributions
 arrest+0.047
 Blood+0.047
 robbers+0.046
 mice+0.044
growth+0.044
Neg logit contributions
Deal-0.047
 difference-0.042
ement-0.041
 decision-0.039
 exactly-0.039
Token think
Token ID892
Feature activation-6.29e-01

Pos logit contributions
 arrest+0.004
 robbers+0.003
 blown+0.003
 spilled+0.003
growth+0.003
Neg logit contributions
Deal-0.005
ement-0.004
 exactly-0.004
 decision-0.004
 difference-0.004
Token that
Token ID326
Feature activation+1.67e+00

Pos logit contributions
 Blood+0.015
 arrest+0.015
 blown+0.015
 robbers+0.015
 mice+0.015
Neg logit contributions
Deal-0.017
 difference-0.014
ement-0.014
 decision-0.014
 exactly-0.013
Token is
Token ID318
Feature activation+1.52e+00

Pos logit contributions
Deal+0.039
ement+0.036
ieved+0.036
uct+0.033
 difference+0.032
Neg logit contributions
 mice-0.047
 Blood-0.044
 spilled-0.044
 robbers-0.044
 blown-0.044
Token a
Token ID257
Feature activation+1.47e+00

Pos logit contributions
Deal+0.052
ement+0.047
 difference+0.047
 decision+0.046
ieved+0.045
Neg logit contributions
 mice-0.025
 Blood-0.025
 robbers-0.024
 blown-0.024
 spilled-0.023
Token good
Token ID922
Feature activation+3.93e-01

Pos logit contributions
ement+0.039
Deal+0.039
uct+0.038
ieved+0.037
 correctly+0.035
Neg logit contributions
 robbers-0.035
 blown-0.033
 mice-0.033
 Blood-0.032
 spilled-0.032
Token name
Token ID1438
Feature activation-2.12e+00

Pos logit contributions
Deal+0.010
uct+0.009
ement+0.009
ieved+0.009
 exactly+0.009
Neg logit contributions
 robbers-0.009
 blown-0.009
 spilled-0.009
 arrest-0.009
 Blood-0.009
Token.
Token ID13
Feature activation+4.01e-01

Pos logit contributions
 mice+0.047
 Blood+0.046
 robbers+0.046
 arrest+0.046
 blown+0.045
Neg logit contributions
Deal-0.059
 difference-0.054
 decision-0.051
ement-0.050
 exactly-0.049
Token
Token ID198
Feature activation+2.70e-03

Pos logit contributions
Deal+0.014
 difference+0.013
 decision+0.012
ement+0.012
 exactly+0.012
Neg logit contributions
 Blood-0.007
 robbers-0.006
 arrest-0.006
alks-0.006
 mice-0.006
Token
Token ID198
Feature activation-7.42e-01

Pos logit contributions
Deal+0.000
 difference+0.000
 decision+0.000
ement+0.000
 exactly+0.000
Neg logit contributions
 Blood-0.000
growth-0.000
 arrest-0.000
alks-0.000
 blown-0.000
Token"
Token ID1
Feature activation-3.42e-01

Pos logit contributions
 Blood+0.017
 arrest+0.016
 robbers+0.016
growth+0.015
 mice+0.015
Neg logit contributions
Deal-0.020
 difference-0.020
ement-0.019
 decision-0.019
 exactly-0.018
Token if
Token ID611
Feature activation+9.35e-01

Pos logit contributions
Deal+0.012
 difference+0.011
ement+0.010
 decision+0.010
 why+0.010
Neg logit contributions
 Blood-0.012
 arrest-0.011
 blown-0.011
 robbers-0.011
 mice-0.011
Token we
Token ID356
Feature activation-2.63e-01

Pos logit contributions
Deal+0.022
 difference+0.019
 isn+0.019
 why+0.019
 exactly+0.018
Neg logit contributions
 Blood-0.025
 arrest-0.024
growth-0.024
 robbers-0.024
 blown-0.024
Token find
Token ID1064
Feature activation+2.43e-02

Pos logit contributions
 Blood+0.008
 arrest+0.007
growth+0.007
alks+0.007
 robbers+0.007
Neg logit contributions
Deal-0.006
 promise-0.005
 difference-0.005
 why-0.005
 isn-0.005
Token it
Token ID340
Feature activation-2.97e-01

Pos logit contributions
Deal+0.000
 difference+0.000
ement+0.000
 isn+0.000
 why+0.000
Neg logit contributions
 Blood-0.001
 arrest-0.001
 robbers-0.001
 blown-0.001
growth-0.001
Token."
Token ID526
Feature activation+5.82e-01

Pos logit contributions
 Blood+0.008
 arrest+0.008
 robbers+0.008
 blown+0.008
growth+0.008
Neg logit contributions
Deal-0.007
 difference-0.006
ement-0.006
 isn-0.006
 why-0.006
Token
Token ID198
Feature activation+1.83e-01

Pos logit contributions
Deal+0.013
 difference+0.011
 isn+0.011
ement+0.010
 why+0.010
Neg logit contributions
 blown-0.016
 arrest-0.016
 robbers-0.016
 Blood-0.016
growth-0.016
Token
Token ID198
Feature activation-2.28e-01

Pos logit contributions
Deal+0.005
 difference+0.004
ement+0.004
 decision+0.004
 isn+0.004
Neg logit contributions
 Blood-0.005
 arrest-0.004
 robbers-0.004
 blown-0.004
growth-0.004
TokenThe
Token ID464
Feature activation+2.35e-01

Pos logit contributions
 Blood+0.005
 arrest+0.005
 robbers+0.005
 blown+0.005
 mice+0.005
Neg logit contributions
Deal-0.006
 difference-0.006
ement-0.006
 decision-0.005
 isn-0.005
Token boy
Token ID2933
Feature activation-7.69e-01

Pos logit contributions
Deal+0.006
 difference+0.005
ement+0.005
 isn+0.005
 why+0.005
Neg logit contributions
 Blood-0.006
 arrest-0.006
 blown-0.006
 robbers-0.006
growth-0.006
Token smiled
Token ID13541
Feature activation-1.54e-01

Pos logit contributions
 Blood+0.017
 spilled+0.017
 mice+0.016
 arrest+0.016
 blown+0.016
Neg logit contributions
Deal-0.022
 difference-0.020
ement-0.019
 decision-0.018
 isn-0.018
Token and
Token ID290
Feature activation+5.67e-01

Pos logit contributions
 Blood+0.003
 robbers+0.003
 blown+0.003
 arrest+0.003
 mice+0.003
Neg logit contributions
Deal-0.005
 difference-0.004
 isn-0.004
ement-0.004
 decision-0.004
Token there
Token ID612
Feature activation+1.51e-01

Pos logit contributions
Deal+0.035
 difference+0.031
ement+0.030
 decision+0.029
 isn+0.029
Neg logit contributions
 Blood-0.029
 arrest-0.028
 robbers-0.028
 blown-0.028
growth-0.027
Token was
Token ID373
Feature activation-1.97e+00

Pos logit contributions
Deal+0.004
 difference+0.004
izes+0.004
 int+0.004
 decision+0.004
Neg logit contributions
 spilled-0.004
 Blood-0.003
growth-0.003
 blown-0.003
 arrest-0.003
Token a
Token ID257
Feature activation+2.33e-01

Pos logit contributions
alks+0.054
 arrest+0.054
 robbers+0.052
plets+0.052
 Blood+0.051
Neg logit contributions
Deal-0.044
 exactly-0.039
 why-0.037
 equally-0.037
 �-0.036
Token pretty
Token ID2495
Feature activation+6.59e-03

Pos logit contributions
Deal+0.008
 difference+0.007
ement+0.007
 isn+0.007
 exactly+0.007
Neg logit contributions
 Blood-0.004
 arrest-0.004
 robbers-0.004
 blown-0.004
 mice-0.004
Token girl
Token ID2576
Feature activation-4.02e-01

Pos logit contributions
Deal+0.000
 difference+0.000
ement+0.000
 isn+0.000
 why+0.000
Neg logit contributions
 Blood-0.000
 arrest-0.000
 robbers-0.000
 blown-0.000
growth-0.000
Token.
Token ID13
Feature activation-8.94e-01

Pos logit contributions
 Blood+0.009
 arrest+0.009
 robbers+0.009
 blown+0.009
growth+0.009
Neg logit contributions
Deal-0.011
 difference-0.010
 why-0.009
ement-0.009
 isn-0.009
Token She
Token ID1375
Feature activation-1.09e+00

Pos logit contributions
 arrest+0.021
 blown+0.020
 robbers+0.020
 Blood+0.019
alks+0.019
Neg logit contributions
Deal-0.025
 difference-0.023
 isn-0.022
 why-0.022
ement-0.021
Token had
Token ID550
Feature activation-3.27e+00

Pos logit contributions
 arrest+0.033
 robbers+0.032
alks+0.032
 Blood+0.032
growth+0.032
Neg logit contributions
Deal-0.023
 isn-0.018
 difference-0.018
 why-0.017
 promise-0.017
Token a
Token ID257
Feature activation-4.31e-01

Pos logit contributions
alks+0.086
 robbers+0.084
plets+0.084
 arrest+0.082
 Blood+0.080
Neg logit contributions
Deal-0.075
 exactly-0.067
 isn-0.065
 why-0.063
 wonderful-0.063
Token massage
Token ID26900
Feature activation+1.26e+00

Pos logit contributions
 Blood+0.009
 robbers+0.009
 arrest+0.008
 spilled+0.008
 blown+0.008
Neg logit contributions
Deal-0.013
 difference-0.012
ement-0.011
 why-0.011
 isn-0.011
Token every
Token ID790
Feature activation+5.61e-01

Pos logit contributions
Deal+0.033
 difference+0.029
ement+0.028
 isn+0.027
 why+0.027
Neg logit contributions
 Blood-0.031
 arrest-0.031
 robbers-0.030
 blown-0.030
growth-0.029
Token much
Token ID881
Feature activation-1.10e+00

Pos logit contributions
Deal+0.011
 difference+0.010
ement+0.010
 decision+0.009
 isn+0.009
Neg logit contributions
 blown-0.009
 Blood-0.009
 arrest-0.008
 robbers-0.008
 mice-0.008
Token happier
Token ID23030
Feature activation-6.82e-01

Pos logit contributions
 blown+0.030
 mice+0.030
 robbers+0.029
 Blood+0.028
 spilled+0.028
Neg logit contributions
Deal-0.026
ement-0.022
 isn-0.021
ieved-0.020
uct-0.020
Token and
Token ID290
Feature activation-2.61e-02

Pos logit contributions
 arrest+0.016
 robbers+0.016
 blown+0.016
 Blood+0.015
 mice+0.015
Neg logit contributions
Deal-0.019
ement-0.016
ieved-0.015
 difference-0.015
uct-0.015
Token all
Token ID477
Feature activation-3.99e-01

Pos logit contributions
 blown+0.001
 arrest+0.001
 Blood+0.001
 robbers+0.001
 mice+0.001
Neg logit contributions
Deal-0.001
ement-0.001
 difference-0.001
 decision-0.001
ieved-0.001
Token the
Token ID262
Feature activation-1.39e+00

Pos logit contributions
 Blood+0.008
 arrest+0.008
 blown+0.008
 robbers+0.008
 mice+0.008
Neg logit contributions
Deal-0.012
 difference-0.011
ement-0.010
 decision-0.010
 exactly-0.010
Token little
Token ID1310
Feature activation-3.13e+00

Pos logit contributions
 Blood+0.028
 arrest+0.027
 robbers+0.027
 blown+0.026
growth+0.026
Neg logit contributions
Deal-0.043
 difference-0.039
ement-0.038
 isn-0.037
 why-0.037
Token animals
Token ID4695
Feature activation-4.22e-01

Pos logit contributions
 Blood+0.066
 arrest+0.063
growth+0.063
alks+0.061
 blown+0.060
Neg logit contributions
Deal-0.092
 difference-0.088
 isn-0.083
 decision-0.081
ement-0.081
Token came
Token ID1625
Feature activation-2.77e+00

Pos logit contributions
 arrest+0.009
 Blood+0.009
 blown+0.009
 robbers+0.009
 spilled+0.009
Neg logit contributions
Deal-0.012
 difference-0.010
ement-0.010
 decision-0.010
 isn-0.010
Token to
Token ID284
Feature activation+1.90e+00

Pos logit contributions
 Blood+0.056
 arrest+0.056
 blown+0.055
 robbers+0.054
alks+0.052
Neg logit contributions
Deal-0.085
 difference-0.075
ement-0.073
 isn-0.071
 decision-0.071
Token play
Token ID711
Feature activation+3.36e-01

Pos logit contributions
Deal+0.055
ement+0.048
 difference+0.047
 isn+0.046
 decision+0.045
Neg logit contributions
 arrest-0.043
 Blood-0.041
 robbers-0.041
growth-0.040
 blown-0.040
Token around
Token ID1088
Feature activation+8.17e-01

Pos logit contributions
Deal+0.009
 difference+0.008
aman+0.008
 decision+0.008
uct+0.007
Neg logit contributions
 mice-0.008
 arrest-0.008
 robbers-0.008
 Blood-0.008
growth-0.008
Token trains
Token ID13404
Feature activation+7.55e-01

Pos logit contributions
alks+0.063
 Blood+0.062
growth+0.060
 arrest+0.060
plets+0.059
Neg logit contributions
Deal-0.047
 why-0.044
 isn-0.042
 exactly-0.039
 difference-0.038
Token and
Token ID290
Feature activation-3.12e+00

Pos logit contributions
Deal+0.014
 difference+0.012
ement+0.011
 isn+0.011
 why+0.011
Neg logit contributions
 Blood-0.024
 arrest-0.024
 robbers-0.023
 blown-0.023
growth-0.023
Token planes
Token ID13016
Feature activation+8.70e-01

Pos logit contributions
 Blood+0.073
alks+0.070
growth+0.066
 blown+0.066
 arrest+0.066
Neg logit contributions
Deal-0.085
 why-0.079
 difference-0.077
 isn-0.077
 deal-0.073
Token that
Token ID326
Feature activation-6.58e-01

Pos logit contributions
Deal+0.022
 difference+0.020
ement+0.019
 decision+0.018
 isn+0.018
Neg logit contributions
 Blood-0.022
 arrest-0.022
 robbers-0.021
 blown-0.021
growth-0.020
Token he
Token ID339
Feature activation-1.12e+00

Pos logit contributions
 Blood+0.015
 arrest+0.015
 robbers+0.015
 blown+0.014
growth+0.014
Neg logit contributions
Deal-0.018
 difference-0.016
 isn-0.015
 why-0.015
ement-0.015
Token could
Token ID714
Feature activation-2.65e+00

Pos logit contributions
 Blood+0.030
 arrest+0.029
growth+0.029
alks+0.029
 robbers+0.028
Neg logit contributions
Deal-0.027
 isn-0.023
 why-0.022
 difference-0.022
 exactly-0.021
Token zoom
Token ID19792
Feature activation-5.94e-01

Pos logit contributions
 Blood+0.065
 arrest+0.063
 robbers+0.062
 blown+0.061
growth+0.061
Neg logit contributions
Deal-0.069
 difference-0.061
 why-0.058
 isn-0.057
 promise-0.057
Token around
Token ID1088
Feature activation-1.07e-01

Pos logit contributions
 Blood+0.016
 arrest+0.015
 robbers+0.015
 blown+0.015
growth+0.015
Neg logit contributions
Deal-0.015
 difference-0.013
ement-0.012
 decision-0.012
 promise-0.012
Token the
Token ID262
Feature activation-8.64e-01

Pos logit contributions
 arrest+0.003
 Blood+0.003
 robbers+0.002
growth+0.002
 mice+0.002
Neg logit contributions
Deal-0.003
 difference-0.003
ement-0.002
 promise-0.002
 decision-0.002
Token floor
Token ID4314
Feature activation+6.03e-02

Pos logit contributions
 arrest+0.017
 robbers+0.017
 mice+0.016
 Blood+0.016
 spilled+0.016
Neg logit contributions
Deal-0.027
ement-0.024
 exactly-0.024
ieved-0.023
 why-0.023
Token.
Token ID13
Feature activation-3.46e-01

Pos logit contributions
Deal+0.002
 difference+0.002
ement+0.001
 isn+0.001
 decision+0.001
Neg logit contributions
 Blood-0.001
 arrest-0.001
 blown-0.001
 robbers-0.001
 spilled-0.001
Token as
Token ID355
Feature activation-1.34e+00

Pos logit contributions
 robbers+0.013
 arrest+0.013
 blown+0.012
 mice+0.012
 Blood+0.012
Neg logit contributions
Deal-0.014
 difference-0.012
ement-0.012
 isn-0.011
 decision-0.011
Token fast
Token ID3049
Feature activation-1.04e+00

Pos logit contributions
growth+0.035
 arrest+0.035
 Blood+0.035
alks+0.033
 robbers+0.033
Neg logit contributions
Deal-0.033
 exactly-0.028
 equally-0.026
 why-0.026
 difference-0.026
Token as
Token ID355
Feature activation-1.80e+00

Pos logit contributions
 Blood+0.031
 robbers+0.028
alks+0.028
umbs+0.028
growth+0.028
Neg logit contributions
Deal-0.020
 why-0.019
 isn-0.018
ement-0.017
 exactly-0.017
Token they
Token ID484
Feature activation-2.49e+00

Pos logit contributions
 Blood+0.042
 arrest+0.042
 robbers+0.040
growth+0.040
 blown+0.040
Neg logit contributions
Deal-0.048
 difference-0.043
 exactly-0.042
 why-0.042
 isn-0.042
Token could
Token ID714
Feature activation-3.04e+00

Pos logit contributions
 Blood+0.067
 arrest+0.067
 robbers+0.065
alks+0.063
growth+0.063
Neg logit contributions
Deal-0.058
 isn-0.049
 difference-0.049
 why-0.049
 exactly-0.048
Token away
Token ID1497
Feature activation-2.81e+00

Pos logit contributions
 Blood+0.082
 arrest+0.081
 robbers+0.080
 blown+0.078
growth+0.078
Neg logit contributions
Deal-0.071
 difference-0.063
ement-0.059
 why-0.059
 isn-0.059
Token from
Token ID422
Feature activation+2.08e+00

Pos logit contributions
 Blood+0.076
 arrest+0.073
 robbers+0.072
 blown+0.071
growth+0.070
Neg logit contributions
Deal-0.067
 difference-0.060
ement-0.056
 why-0.056
 isn-0.056
Token the
Token ID262
Feature activation-2.85e-01

Pos logit contributions
Deal+0.051
 difference+0.047
 why+0.046
 exactly+0.045
ement+0.044
Neg logit contributions
 Blood-0.054
 arrest-0.051
 robbers-0.050
 blown-0.049
alks-0.049
Token weird
Token ID7650
Feature activation-8.84e-01

Pos logit contributions
 Blood+0.006
 arrest+0.006
 robbers+0.006
 blown+0.006
growth+0.006
Neg logit contributions
Deal-0.008
 difference-0.007
ement-0.007
 isn-0.007
 decision-0.007
Token thing
Token ID1517
Feature activation+1.12e+00

Pos logit contributions
 Blood+0.020
 arrest+0.020
 robbers+0.019
 blown+0.019
growth+0.019
Neg logit contributions
Deal-0.025
 difference-0.022
ement-0.021
 isn-0.021
 why-0.021
Token drinking
Token ID7722
Feature activation-1.11e+00

Pos logit contributions
Deal+0.032
 difference+0.028
 decision+0.027
ement+0.026
 promise+0.026
Neg logit contributions
 arrest-0.025
 blown-0.025
 Blood-0.025
 spilled-0.024
 robbers-0.024
Token by
Token ID416
Feature activation+1.01e+00

Pos logit contributions
 arrest+0.013
 Blood+0.013
 blown+0.013
 robbers+0.013
 spilled+0.013
Neg logit contributions
Deal-0.019
 difference-0.017
 isn-0.017
ement-0.016
ieved-0.016
Token giving
Token ID3501
Feature activation-1.09e-01

Pos logit contributions
Deal+0.025
 why+0.024
 difference+0.024
 exactly+0.023
 decision+0.022
Neg logit contributions
 Blood-0.024
 arrest-0.023
growth-0.023
 robbers-0.023
plets-0.023
Token her
Token ID607
Feature activation+4.08e-01

Pos logit contributions
 mice+0.002
 Blood+0.002
 arrest+0.002
 blown+0.002
 spilled+0.002
Neg logit contributions
Deal-0.004
 difference-0.003
ement-0.003
 decision-0.003
 isn-0.003
Token a
Token ID257
Feature activation+9.82e-01

Pos logit contributions
Deal+0.014
ement+0.012
 difference+0.011
ieved+0.011
 isn+0.011
Neg logit contributions
 mice-0.007
 arrest-0.007
 blown-0.007
 spilled-0.007
growth-0.007
Token big
Token ID1263
Feature activation+2.04e+00

Pos logit contributions
Deal+0.023
ement+0.019
 difference+0.018
uct+0.018
 isn+0.018
Neg logit contributions
 Blood-0.027
 arrest-0.026
 blown-0.026
 spilled-0.026
 robbers-0.026
Token hug
Token ID16225
Feature activation-2.94e+00

Pos logit contributions
Deal+0.060
uct+0.050
 exactly+0.050
ieved+0.049
ement+0.049
Neg logit contributions
 arrest-0.045
 robbers-0.044
 Blood-0.044
 spilled-0.044
 mice-0.043
Token and
Token ID290
Feature activation+9.09e-02

Pos logit contributions
 spilled+0.058
 blown+0.058
 mice+0.058
 Blood+0.058
 robbers+0.057
Neg logit contributions
Deal-0.087
 difference-0.078
ement-0.074
 decision-0.073
ieved-0.072
Token saying
Token ID2282
Feature activation+9.39e-02

Pos logit contributions
Deal+0.003
 difference+0.002
ement+0.002
 isn+0.002
 decision+0.002
Neg logit contributions
 Blood-0.002
 arrest-0.002
 robbers-0.002
 blown-0.002
growth-0.002
Token "
Token ID366
Feature activation+1.33e+00

Pos logit contributions
Deal+0.002
 difference+0.002
ieved+0.002
 decision+0.002
 promise+0.002
Neg logit contributions
 Blood-0.002
 mice-0.002
 spilled-0.002
alks-0.002
 arrest-0.002
Tokenw
Token ID86
Feature activation-1.46e-01

Pos logit contributions
Deal+0.035
 difference+0.030
ement+0.029
 isn+0.029
 why+0.029
Neg logit contributions
 arrest-0.033
 Blood-0.032
 robbers-0.032
growth-0.031
 blown-0.031
Tokenelcome
Token ID9571
Feature activation-7.47e-01

Pos logit contributions
 Blood+0.004
 arrest+0.004
 spilled+0.004
alks+0.004
 blown+0.004
Neg logit contributions
Deal-0.003
 difference-0.003
ement-0.003
 decision-0.002
 promise-0.002
Token the
Token ID262
Feature activation-3.31e-01

Pos logit contributions
 Blood+0.034
 robbers+0.032
 blown+0.032
plets+0.032
 arrest+0.032
Neg logit contributions
Deal-0.027
 difference-0.026
 why-0.026
 exactly-0.024
ement-0.024
Token street
Token ID4675
Feature activation-5.81e-01

Pos logit contributions
 Blood+0.008
 blown+0.007
 arrest+0.007
alks+0.007
plets+0.007
Neg logit contributions
Deal-0.009
 difference-0.008
 decision-0.008
 promise-0.008
 isn-0.007
Token.
Token ID13
Feature activation-1.21e+00

Pos logit contributions
 arrest+0.012
 Blood+0.012
 blown+0.011
 spilled+0.011
 robbers+0.011
Neg logit contributions
Deal-0.018
 difference-0.016
ement-0.015
ieved-0.015
 isn-0.015
Token The
Token ID383
Feature activation-7.16e-01

Pos logit contributions
 arrest+0.033
 blown+0.031
 robbers+0.030
growth+0.029
 Blood+0.029
Neg logit contributions
Deal-0.027
 isn-0.026
 difference-0.025
 �-0.025
 why-0.025
Token van
Token ID5719
Feature activation-9.20e-01

Pos logit contributions
 Blood+0.020
 arrest+0.020
 robbers+0.020
 blown+0.020
growth+0.019
Neg logit contributions
Deal-0.016
 difference-0.014
ement-0.013
 isn-0.013
 why-0.013
Token has
Token ID468
Feature activation-3.71e+00

Pos logit contributions
 Blood+0.022
 arrest+0.022
 spilled+0.021
 blown+0.021
 robbers+0.020
Neg logit contributions
Deal-0.026
 difference-0.023
 why-0.021
ieved-0.021
 decision-0.021
Token many
Token ID867
Feature activation-2.27e+00

Pos logit contributions
 Blood+0.091
plets+0.089
alks+0.088
ges+0.085
growth+0.085
Neg logit contributions
 exactly-0.080
Deal-0.079
 isn-0.078
 why-0.075
 difference-0.073
Token colors
Token ID7577
Feature activation-7.73e-01

Pos logit contributions
 Blood+0.045
 arrest+0.044
 robbers+0.043
 blown+0.042
growth+0.041
Neg logit contributions
Deal-0.071
 difference-0.064
ement-0.061
 why-0.061
 isn-0.061
Token and
Token ID290
Feature activation-1.99e+00

Pos logit contributions
 arrest+0.021
 Blood+0.021
 robbers+0.021
growth+0.020
 mice+0.020
Neg logit contributions
Deal-0.017
ement-0.015
 difference-0.015
 isn-0.015
 exactly-0.015
Token pictures
Token ID5986
Feature activation-6.60e-01

Pos logit contributions
 Blood+0.047
 arrest+0.047
 robbers+0.046
 blown+0.045
growth+0.045
Neg logit contributions
Deal-0.054
 difference-0.047
ement-0.045
 isn-0.044
 why-0.044
Token on
Token ID319
Feature activation-1.54e+00

Pos logit contributions
 Blood+0.015
 blown+0.015
 arrest+0.015
 robbers+0.015
alks+0.014
Neg logit contributions
Deal-0.019
 difference-0.016
 decision-0.015
ement-0.015
 promise-0.015