TOP ACTIVATIONS
MAX = 0.213

. They were called ost ric hes . Lily and Ben
cried . " The ost ric hes are mean ! They
ostr ich pen . Ost ric hes are big birds with
ostr ich pen . Ost ric hes are big birds with
but be careful . Ost ric hes can be gr umpy
ie . Unlike other ost ric hes , his feathers gl
too close . The ost ric hes might pe ck you
shouted . " The ost ric hes are so funny !"
Look , mom , ost ric hes !" "
Yes , they are ost ric hes ," mom said .
" O st ric hes do not like people
that said " O st ric hes ". Sara and Tom
and looked at the ost ric hes . They saw one
But the ost ric hes did not like that
should not tease the ost ric hes . They are not
and run to the ost ric hes . "
Did you know that cater pill ars turn into beautiful butterflies
They do not pack ost ric hes anymore . They pack
did not know what ost ric hes were . They looked
. It sees other ost ric hes . It makes new

MIN ACTIVATIONS
MIN = -0.124

to the hill and climbed to the top . They looked
tall tree and climbed up to the top . From there
aded in the water delic ately with his fishing rod .
girl called Daisy went outside to wander . She picked up
big slide and climbed up to the top . He slid
girl called Lucy went outside to play in the park .
the stairs and climbed up to the attic . They opened
a chair and climbed up to get the ball down .
slide and Max climbed up to the top . Tim watched
and managed to slide down to the bottom .
he decided to slide down to get a better look .
to climb over a fence to get it . Tim my
<|endoftext|> Jimmy wanted to climb to the top of the tree
and his Mom went outside to play in the snow .
cat named Tim went outside to play on the lawn .
would often climb the ladder to peek over the fence and
to it and climbed up to the top . From the
and boots and went outside to jump in p udd les
to jump over a fence to get it . But as
to climb onto the counter to reach the top shelf of

INTERVAL 0.129 - 0.213
CONTAINS 0.015%

sick person . â € Kevin was so
vegetables in the back . One day ,
picking up the phone . " Hi ,
and healthy once more . Max 's owner
was bright and clear . Jimmy shouted in

INTERVAL 0.045 - 0.129
CONTAINS 12.267%

, " You are good twins . I am glad you
I 'm OK . It was a bit of a silly
happily ever after . <|endoftext|> Once upon a time , there
the bee on their flower necklace . They felt sorry for
happy and hungry . But then , a loud

INTERVAL -0.040 - 0.045
CONTAINS 75.204%

He felt proud to be visiting his grandparents for the very
you trust your mom 's suggestions , you can be very
. They share their toys and have fun .
should play the drum together , and make a happy sound
gave him a pe b ble . He had never seen

INTERVAL -0.124 - -0.040
CONTAINS 12.514%

a better look . He noticed the body had all different
Look , it 's coming towards us !"
years old and liked to explore outside with his parents .
, Anna and Emma , let 's go have dinner .
with his pencil and at last , he was done !
Token.
Token ID13
Feature activation-1.51e-02

Pos logit contributions
istered+0.135
hes+0.128
aj+0.128
oll+0.127
ittens+0.124
Neg logit contributions
 home-0.034
 itself-0.034
 towards-0.033
 Sus-0.032
 Mill-0.032
Token They
Token ID1119
Feature activation+2.69e-02

Pos logit contributions
away+0.019
 angle+0.019
riage+0.018
 inspect+0.017
keeping+0.017
Neg logit contributions
hes-0.030
uts-0.029
istered-0.029
as-0.028
aj-0.027
Token were
Token ID547
Feature activation+9.87e-02

Pos logit contributions
istered+0.058
hes+0.057
uts+0.055
aj+0.055
oll+0.054
Neg logit contributions
 Sus-0.030
 Mill-0.029
 Joey-0.029
 inspect-0.029
 investigate-0.029
Token called
Token ID1444
Feature activation+4.72e-02

Pos logit contributions
istered+0.198
uts+0.183
arest+0.181
uch+0.180
aj+0.180
Neg logit contributions
 home-0.147
 Mill-0.139
 Sus-0.132
 Joey-0.132
 Ab-0.127
Token ost
Token ID23619
Feature activation+7.66e-02

Pos logit contributions
hes+0.092
istered+0.091
as+0.087
uts+0.086
aj+0.084
Neg logit contributions
 angle-0.060
 Sus-0.059
 Mill-0.059
away-0.058
 Joey-0.058
Tokenric
Token ID1173
Feature activation+2.13e-01

Pos logit contributions
hes+0.145
istered+0.143
oll+0.140
uts+0.127
br+0.126
Neg logit contributions
 angle-0.104
riage-0.103
 Mill-0.103
 Joey-0.102
 side-0.099
Tokenhes
Token ID956
Feature activation+4.19e-02

Pos logit contributions
istered+0.195
aj+0.179
ittens+0.167
ucker+0.142
oll+0.124
Neg logit contributions
 itself-0.523
 home-0.468
 over-0.468
 him-0.468
 the-0.467
Token.
Token ID13
Feature activation-1.31e-02

Pos logit contributions
istered+0.086
hes+0.083
aj+0.081
uts+0.080
as+0.079
Neg logit contributions
 angle-0.048
away-0.048
 Sus-0.048
 Mill-0.047
 Joey-0.047
Token Lily
Token ID20037
Feature activation-1.87e-03

Pos logit contributions
 angle+0.015
away+0.015
riage+0.014
 inspect+0.014
keeping+0.014
Neg logit contributions
hes-0.027
istered-0.026
as-0.026
uts-0.026
aj-0.025
Token and
Token ID290
Feature activation+1.75e-02

Pos logit contributions
 Sus+0.002
 Mill+0.002
 angle+0.002
 Joey+0.002
away+0.002
Neg logit contributions
istered-0.004
hes-0.004
uts-0.004
agging-0.004
oll-0.004
Token Ben
Token ID3932
Feature activation+7.87e-03

Pos logit contributions
hes+0.032
istered+0.032
as+0.030
uts+0.030
oll+0.029
Neg logit contributions
 angle-0.024
away-0.024
riage-0.023
 Sus-0.023
 Mill-0.023
Token cried
Token ID16896
Feature activation+3.09e-02

Pos logit contributions
istered+0.017
hes+0.017
aj+0.017
as+0.016
oll+0.016
Neg logit contributions
 investigate-0.010
 Sus-0.010
 away-0.010
 Joey-0.010
 side-0.009
Token.
Token ID13
Feature activation-1.75e-02

Pos logit contributions
hes+0.061
istered+0.060
as+0.058
uts+0.058
aj+0.055
Neg logit contributions
 angle-0.039
 Sus-0.037
away-0.037
 Mill-0.037
riage-0.037
Token "
Token ID366
Feature activation+2.36e-02

Pos logit contributions
 angle+0.020
away+0.020
riage+0.019
 Sus+0.019
 Mill+0.019
Neg logit contributions
hes-0.037
istered-0.036
uts-0.035
as-0.035
aj-0.033
TokenThe
Token ID464
Feature activation+3.78e-02

Pos logit contributions
aj+0.045
ows+0.044
istered+0.043
as+0.043
oll+0.043
Neg logit contributions
away-0.029
 angle-0.028
 itself-0.028
keeping-0.027
 Joey-0.027
Token ost
Token ID23619
Feature activation+6.63e-02

Pos logit contributions
istered+0.073
hes+0.071
uts+0.071
as+0.071
ows+0.069
Neg logit contributions
 angle-0.050
 home-0.047
away-0.047
 side-0.046
riage-0.046
Tokenric
Token ID1173
Feature activation+2.09e-01

Pos logit contributions
istered+0.128
uts+0.120
oll+0.119
hes+0.117
cases+0.114
Neg logit contributions
riage-0.085
 angle-0.083
 side-0.080
 itself-0.079
away-0.076
Tokenhes
Token ID956
Feature activation+2.55e-02

Pos logit contributions
istered+0.209
aj+0.192
ittens+0.171
cases+0.147
uts+0.135
Neg logit contributions
 itself-0.479
 his-0.435
crow-0.435
 the-0.433
 its-0.433
Token are
Token ID389
Feature activation+7.26e-02

Pos logit contributions
istered+0.043
hes+0.042
uts+0.040
as+0.040
aj+0.039
Neg logit contributions
 angle-0.039
away-0.039
 Sus-0.038
 Mill-0.037
riage-0.037
Token mean
Token ID1612
Feature activation-2.34e-02

Pos logit contributions
cases+0.128
aj+0.120
istered+0.118
oll+0.117
uts+0.117
Neg logit contributions
 away-0.111
 itself-0.110
 home-0.107
 its-0.106
 the-0.106
Token!
Token ID0
Feature activation-2.95e-02

Pos logit contributions
away+0.028
 home+0.025
 angle+0.025
 towards+0.025
keeping+0.025
Neg logit contributions
istered-0.050
hes-0.049
aj-0.046
oll-0.045
zel-0.045
Token They
Token ID1119
Feature activation+2.14e-02

Pos logit contributions
 angle+0.026
 Sus+0.026
 Mill+0.025
away+0.025
riage+0.025
Neg logit contributions
hes-0.068
istered-0.067
as-0.065
uts-0.064
aj-0.063
Token ostr
Token ID42387
Feature activation+6.78e-02

Pos logit contributions
hes+0.040
istered+0.039
as+0.038
uts+0.037
aj+0.036
Neg logit contributions
 angle-0.024
 Sus-0.023
away-0.023
 Mill-0.023
riage-0.023
Tokenich
Token ID488
Feature activation+1.44e-03

Pos logit contributions
hes+0.098
istered+0.087
oll+0.082
ows+0.076
umbs+0.074
Neg logit contributions
 side-0.123
riage-0.123
 home-0.123
away-0.122
 angle-0.121
Token pen
Token ID3112
Feature activation+3.04e-02

Pos logit contributions
hes+0.002
istered+0.002
as+0.002
uts+0.002
ows+0.002
Neg logit contributions
 Sus-0.002
 angle-0.002
 Mill-0.002
away-0.002
riage-0.002
Token.
Token ID13
Feature activation-2.75e-02

Pos logit contributions
istered+0.069
hes+0.065
as+0.063
uts+0.062
aj+0.062
Neg logit contributions
 angle-0.031
away-0.030
 home-0.029
 side-0.029
 itself-0.029
Token Ost
Token ID38919
Feature activation+2.71e-02

Pos logit contributions
 angle+0.037
away+0.037
keeping+0.033
riage+0.033
 state+0.032
Neg logit contributions
hes-0.053
uts-0.052
istered-0.050
as-0.048
ows-0.048
Tokenric
Token ID1173
Feature activation+2.08e-01

Pos logit contributions
hes+0.036
istered+0.035
as+0.033
uts+0.033
aj+0.031
Neg logit contributions
 angle-0.051
away-0.050
 Sus-0.050
 Mill-0.050
riage-0.049
Tokenhes
Token ID956
Feature activation+3.72e-02

Pos logit contributions
istered+0.195
aj+0.172
ittens+0.153
ider+0.124
ucker+0.124
Neg logit contributions
 itself-0.508
 behold-0.459
 him-0.457
 his-0.455
 the-0.455
Token are
Token ID389
Feature activation+8.35e-02

Pos logit contributions
istered+0.066
hes+0.066
as+0.063
uts+0.062
aj+0.060
Neg logit contributions
 angle-0.053
away-0.052
 Sus-0.052
 Mill-0.051
riage-0.051
Token big
Token ID1263
Feature activation+6.32e-03

Pos logit contributions
istered+0.139
uts+0.134
plets+0.133
arest+0.132
oll+0.129
Neg logit contributions
 home-0.130
 itself-0.123
 Mill-0.121
 Joey-0.119
 Sus-0.118
Token birds
Token ID10087
Feature activation+6.74e-02

Pos logit contributions
hes+0.012
istered+0.012
uts+0.011
as+0.011
aj+0.011
Neg logit contributions
 angle-0.008
 Sus-0.008
away-0.008
 Mill-0.008
 Joey-0.008
Token with
Token ID351
Feature activation+2.95e-02

Pos logit contributions
aj+0.169
istered+0.168
hes+0.166
oll+0.162
ittens+0.157
Neg logit contributions
 side-0.046
 home-0.044
 itself-0.042
 Mill-0.041
away-0.041
Token ostr
Token ID42387
Feature activation+6.70e-02

Pos logit contributions
hes+0.045
istered+0.044
as+0.043
uts+0.041
aj+0.041
Neg logit contributions
 angle-0.030
 Sus-0.030
 Mill-0.030
away-0.029
 Joey-0.029
Tokenich
Token ID488
Feature activation+2.11e-03

Pos logit contributions
hes+0.093
istered+0.084
oll+0.078
ows+0.071
uts+0.071
Neg logit contributions
riage-0.126
 side-0.124
away-0.124
 home-0.122
 angle-0.122
Token pen
Token ID3112
Feature activation+3.27e-02

Pos logit contributions
hes+0.004
istered+0.004
as+0.003
uts+0.003
ows+0.003
Neg logit contributions
 Sus-0.003
 angle-0.003
 Mill-0.003
riage-0.003
 Joey-0.003
Token.
Token ID13
Feature activation-2.31e-02

Pos logit contributions
istered+0.074
hes+0.070
as+0.067
aj+0.067
uts+0.066
Neg logit contributions
away-0.032
 angle-0.032
 home-0.032
 itself-0.031
 side-0.031
Token Ost
Token ID38919
Feature activation+2.47e-02

Pos logit contributions
 angle+0.035
away+0.033
keeping+0.031
 state+0.030
riage+0.030
Neg logit contributions
hes-0.042
uts-0.040
ends-0.038
istered-0.038
as-0.038
Tokenric
Token ID1173
Feature activation+2.08e-01

Pos logit contributions
hes+0.031
istered+0.030
as+0.029
uts+0.028
aj+0.027
Neg logit contributions
 angle-0.048
 Sus-0.047
away-0.047
 Mill-0.047
riage-0.046
Tokenhes
Token ID956
Feature activation+3.75e-02

Pos logit contributions
istered+0.202
aj+0.174
ittens+0.152
ider+0.133
ucker+0.123
Neg logit contributions
 itself-0.497
 the-0.453
 him-0.453
 his-0.451
 behold-0.450
Token are
Token ID389
Feature activation+8.36e-02

Pos logit contributions
istered+0.065
hes+0.064
uts+0.061
as+0.061
aj+0.060
Neg logit contributions
 angle-0.055
away-0.054
 Sus-0.053
riage-0.053
 Mill-0.053
Token big
Token ID1263
Feature activation+7.98e-03

Pos logit contributions
istered+0.134
oll+0.129
uts+0.127
plets+0.126
arest+0.124
Neg logit contributions
 home-0.135
 itself-0.129
 angle-0.124
 Mill-0.124
 side-0.123
Token birds
Token ID10087
Feature activation+6.72e-02

Pos logit contributions
hes+0.015
istered+0.014
uts+0.014
as+0.014
aj+0.013
Neg logit contributions
 angle-0.011
away-0.011
 Sus-0.011
 Mill-0.011
riage-0.010
Token with
Token ID351
Feature activation+2.91e-02

Pos logit contributions
istered+0.171
aj+0.170
oll+0.164
hes+0.163
ittens+0.158
Neg logit contributions
 side-0.044
 itself-0.042
 home-0.042
away-0.042
 away-0.038
Token but
Token ID475
Feature activation-2.64e-02

Pos logit contributions
istered+0.064
hes+0.062
aj+0.057
uts+0.057
as+0.055
Neg logit contributions
 Joey-0.035
 Sus-0.033
 Tim-0.033
 him-0.032
 side-0.032
Token be
Token ID307
Feature activation+5.24e-02

Pos logit contributions
 angle+0.044
away+0.043
 Sus+0.042
keeping+0.042
riage+0.042
Neg logit contributions
hes-0.039
istered-0.039
as-0.037
uts-0.037
aj-0.035
Token careful
Token ID8161
Feature activation-1.28e-02

Pos logit contributions
uts+0.075
istered+0.072
cases+0.070
oll+0.067
plets+0.067
Neg logit contributions
 home-0.093
 itself-0.092
away-0.089
 directions-0.087
 Joey-0.087
Token.
Token ID13
Feature activation-2.68e-02

Pos logit contributions
 itself+0.012
 side+0.012
 Joey+0.011
 angle+0.011
 home+0.011
Neg logit contributions
istered-0.031
hes-0.029
oll-0.028
aj-0.028
br-0.027
Token Ost
Token ID38919
Feature activation+1.86e-02

Pos logit contributions
 angle+0.027
away+0.026
 Sus+0.025
 Mill+0.024
riage+0.024
Neg logit contributions
hes-0.060
istered-0.059
uts-0.057
as-0.057
aj-0.055
Tokenric
Token ID1173
Feature activation+2.02e-01

Pos logit contributions
hes+0.025
istered+0.024
as+0.023
uts+0.023
aj+0.021
Neg logit contributions
 angle-0.035
 Sus-0.034
away-0.034
 Mill-0.034
riage-0.034
Tokenhes
Token ID956
Feature activation+2.56e-02

Pos logit contributions
istered+0.189
aj+0.157
ittens+0.131
ider+0.118
acas+0.105
Neg logit contributions
 itself-0.504
 the-0.456
 him-0.454
 behold-0.450
 its-0.449
Token can
Token ID460
Feature activation+2.28e-02

Pos logit contributions
istered+0.045
hes+0.043
uts+0.042
as+0.042
aj+0.041
Neg logit contributions
 angle-0.037
away-0.037
riage-0.036
 Sus-0.036
 Mill-0.036
Token be
Token ID307
Feature activation+4.82e-02

Pos logit contributions
istered+0.050
aj+0.049
oll+0.048
as+0.047
hes+0.046
Neg logit contributions
 investigate-0.024
away-0.024
 Mill-0.023
 Joey-0.023
 side-0.023
Token gr
Token ID1036
Feature activation+4.68e-02

Pos logit contributions
istered+0.076
oll+0.075
uts+0.073
cases+0.071
aj+0.069
Neg logit contributions
 home-0.078
away-0.076
 itself-0.074
 away-0.074
 Mill-0.072
Tokenumpy
Token ID32152
Feature activation-1.80e-02

Pos logit contributions
istered+0.072
hes+0.065
aj+0.065
ends+0.063
cases+0.062
Neg logit contributions
away-0.079
 angle-0.078
keeping-0.075
 itself-0.073
gent-0.072
Tokenie
Token ID494
Feature activation-1.16e-02

Pos logit contributions
hes+0.071
istered+0.069
oll+0.063
umbs+0.062
ends+0.062
Neg logit contributions
 Mill-0.027
 Sus-0.027
away-0.027
ie-0.027
 Joey-0.026
Token.
Token ID13
Feature activation-3.22e-02

Pos logit contributions
 angle+0.015
 Sus+0.014
away+0.014
 Mill+0.014
riage+0.014
Neg logit contributions
hes-0.022
istered-0.022
as-0.022
uts-0.021
aj-0.021
Token Unlike
Token ID12101
Feature activation-2.56e-02

Pos logit contributions
away+0.040
keeping+0.036
 angle+0.035
riage+0.034
 generation+0.032
Neg logit contributions
hes-0.067
uts-0.065
oll-0.065
ittens-0.064
umbs-0.064
Token other
Token ID584
Feature activation-3.85e-02

Pos logit contributions
 angle+0.031
keeping+0.030
away+0.030
 Mill+0.029
riage+0.029
Neg logit contributions
hes-0.051
as-0.049
istered-0.048
uts-0.047
oll-0.047
Token ost
Token ID23619
Feature activation+6.16e-02

Pos logit contributions
 Sus+0.056
 angle+0.056
 Joey+0.055
riage+0.055
 Mill+0.055
Neg logit contributions
as-0.064
istered-0.062
hes-0.061
uts-0.059
agging-0.057
Tokenric
Token ID1173
Feature activation+1.96e-01

Pos logit contributions
istered+0.113
hes+0.111
oll+0.101
br+0.098
uts+0.098
Neg logit contributions
 angle-0.091
 Mill-0.085
riage-0.084
 Joey-0.084
 Sus-0.083
Tokenhes
Token ID956
Feature activation+2.30e-02

Pos logit contributions
istered+0.159
aj+0.128
ittens+0.116
ucker+0.102
uts+0.090
Neg logit contributions
 itself-0.495
 home-0.459
away-0.456
 angle-0.456
 behold-0.450
Token,
Token ID11
Feature activation+7.21e-03

Pos logit contributions
istered+0.046
hes+0.045
as+0.043
aj+0.043
uts+0.043
Neg logit contributions
 angle-0.028
away-0.027
 Sus-0.027
 Mill-0.027
 Joey-0.027
Token his
Token ID465
Feature activation-1.90e-03

Pos logit contributions
istered+0.016
hes+0.016
as+0.015
uts+0.015
aj+0.015
Neg logit contributions
 Sus-0.007
 Mill-0.007
 angle-0.007
 Joey-0.007
away-0.007
Token feathers
Token ID27400
Feature activation+7.29e-02

Pos logit contributions
 angle+0.003
 Mill+0.002
away+0.002
 Sus+0.002
 home+0.002
Neg logit contributions
hes-0.004
istered-0.004
uts-0.003
as-0.003
oll-0.003
Token gl
Token ID1278
Feature activation+6.47e-02

Pos logit contributions
istered+0.168
hes+0.163
aj+0.156
oll+0.154
as+0.154
Neg logit contributions
 Joey-0.068
 angle-0.068
 Mill-0.067
 Sus-0.067
away-0.066
Token too
Token ID1165
Feature activation+3.22e-02

Pos logit contributions
hes+0.013
istered+0.013
as+0.012
uts+0.012
aj+0.012
Neg logit contributions
 angle-0.009
 Sus-0.008
 Mill-0.008
away-0.008
riage-0.008
Token close
Token ID1969
Feature activation-3.80e-02

Pos logit contributions
istered+0.060
hes+0.060
zel+0.056
oll+0.056
aj+0.056
Neg logit contributions
away-0.050
 angle-0.047
 home-0.046
 investigate-0.045
 Joey-0.045
Token.
Token ID13
Feature activation-2.86e-02

Pos logit contributions
 angle+0.041
away+0.039
 Sus+0.039
 Mill+0.039
riage+0.038
Neg logit contributions
hes-0.081
istered-0.081
as-0.078
uts-0.077
aj-0.075
Token The
Token ID383
Feature activation+2.51e-02

Pos logit contributions
 angle+0.035
away+0.034
 Sus+0.032
 Mill+0.032
 inspect+0.031
Neg logit contributions
hes-0.058
istered-0.057
as-0.056
uts-0.055
aj-0.053
Token ost
Token ID23619
Feature activation+5.81e-02

Pos logit contributions
istered+0.051
hes+0.050
uts+0.049
as+0.048
oll+0.047
Neg logit contributions
 angle-0.031
riage-0.029
away-0.029
 Mill-0.029
 Sus-0.029
Tokenric
Token ID1173
Feature activation+1.92e-01

Pos logit contributions
istered+0.114
oll+0.106
hes+0.106
uts+0.099
br+0.098
Neg logit contributions
riage-0.080
 angle-0.077
 side-0.076
 itself-0.072
 Mill-0.071
Tokenhes
Token ID956
Feature activation+2.04e-02

Pos logit contributions
istered+0.170
aj+0.139
ittens+0.119
ucker+0.101
uts+0.093
Neg logit contributions
 itself-0.469
 behold-0.445
away-0.444
riage-0.441
crow-0.439
Token might
Token ID1244
Feature activation+1.92e-02

Pos logit contributions
istered+0.035
hes+0.034
uts+0.032
as+0.032
aj+0.032
Neg logit contributions
 angle-0.031
away-0.030
 Sus-0.030
 Mill-0.030
riage-0.029
Token pe
Token ID613
Feature activation+5.55e-02

Pos logit contributions
istered+0.046
oll+0.045
hes+0.044
aj+0.044
as+0.042
Neg logit contributions
away-0.020
 side-0.019
 home-0.019
 investigate-0.018
 angle-0.018
Tokenck
Token ID694
Feature activation+3.52e-03

Pos logit contributions
istered+0.097
oll+0.087
uts+0.087
umbs+0.084
aj+0.081
Neg logit contributions
away-0.084
 itself-0.083
ered-0.080
 inspect-0.079
 home-0.077
Token you
Token ID345
Feature activation-1.03e-02

Pos logit contributions
istered+0.007
hes+0.007
uts+0.006
aj+0.006
as+0.006
Neg logit contributions
away-0.005
 angle-0.004
 Sus-0.004
 Mill-0.004
 Joey-0.004
Token shouted
Token ID17293
Feature activation+2.57e-02

Pos logit contributions
 Sus+0.016
 Mill+0.016
 angle+0.016
 Joey+0.015
 investigate+0.015
Neg logit contributions
istered-0.025
hes-0.025
agging-0.023
as-0.023
oll-0.022
Token.
Token ID13
Feature activation-3.06e-02

Pos logit contributions
istered+0.045
hes+0.044
uts+0.044
as+0.042
ows+0.041
Neg logit contributions
 angle-0.039
riage-0.038
keeping-0.036
away-0.036
 inspect-0.035
Token "
Token ID366
Feature activation+1.35e-02

Pos logit contributions
 angle+0.039
riage+0.039
away+0.039
 Sus+0.037
 Mill+0.037
Neg logit contributions
hes-0.060
istered-0.058
uts-0.057
as-0.056
ends-0.053
TokenThe
Token ID464
Feature activation+3.00e-02

Pos logit contributions
hes+0.022
istered+0.022
as+0.021
uts+0.021
aj+0.020
Neg logit contributions
 angle-0.021
 Sus-0.020
away-0.020
 Mill-0.020
riage-0.020
Token ost
Token ID23619
Feature activation+5.98e-02

Pos logit contributions
hes+0.058
istered+0.058
as+0.055
uts+0.055
ows+0.053
Neg logit contributions
 angle-0.041
riage-0.038
 Mill-0.038
away-0.037
 Sus-0.037
Tokenric
Token ID1173
Feature activation+1.91e-01

Pos logit contributions
istered+0.116
hes+0.110
oll+0.108
uts+0.103
br+0.100
Neg logit contributions
riage-0.080
 angle-0.078
 itself-0.074
 side-0.074
 Mill-0.072
Tokenhes
Token ID956
Feature activation+1.53e-02

Pos logit contributions
istered+0.178
aj+0.146
ittens+0.128
ucker+0.103
itching+0.093
Neg logit contributions
 itself-0.466
 behold-0.435
crow-0.430
away-0.420
 its-0.418
Token are
Token ID389
Feature activation+6.46e-02

Pos logit contributions
istered+0.026
hes+0.025
uts+0.024
as+0.024
aj+0.024
Neg logit contributions
 angle-0.023
away-0.023
 Sus-0.023
riage-0.023
 Mill-0.022
Token so
Token ID523
Feature activation+3.81e-02

Pos logit contributions
istered+0.110
oll+0.107
aj+0.102
uts+0.102
umbs+0.102
Neg logit contributions
 home-0.098
 itself-0.096
 Mill-0.096
 away-0.092
 Sus-0.092
Token funny
Token ID8258
Feature activation-1.47e-02

Pos logit contributions
istered+0.064
oll+0.064
aj+0.062
agging+0.058
ittens+0.057
Neg logit contributions
away-0.060
 Mill-0.056
 itself-0.056
 Sus-0.055
 home-0.055
Token!"
Token ID2474
Feature activation-7.60e-03

Pos logit contributions
away+0.018
 itself+0.017
 Sus+0.017
 Mill+0.017
 Joey+0.017
Neg logit contributions
istered-0.031
hes-0.029
aj-0.027
oll-0.027
agging-0.026
TokenLook
Token ID8567
Feature activation+2.44e-02

Pos logit contributions
hes+0.009
as+0.009
uts+0.008
istered+0.008
umbs+0.008
Neg logit contributions
 angle-0.003
 Sus-0.003
away-0.003
 Mill-0.003
 Joey-0.003
Token,
Token ID11
Feature activation+1.46e-02

Pos logit contributions
istered+0.048
hes+0.047
as+0.047
uts+0.046
aj+0.045
Neg logit contributions
 angle-0.030
keeping-0.029
away-0.028
riage-0.028
 Mill-0.028
Token mom
Token ID1995
Feature activation+1.13e-02

Pos logit contributions
hes+0.029
istered+0.028
as+0.028
uts+0.027
aj+0.026
Neg logit contributions
 angle-0.018
away-0.018
 Sus-0.017
 Mill-0.017
riage-0.017
Token,
Token ID11
Feature activation+2.12e-02

Pos logit contributions
istered+0.027
hes+0.027
uts+0.025
as+0.025
oll+0.025
Neg logit contributions
 Sus-0.010
 Mill-0.010
 angle-0.010
riage-0.010
 Joey-0.010
Token ost
Token ID23619
Feature activation+6.06e-02

Pos logit contributions
hes+0.039
istered+0.038
as+0.037
uts+0.037
oll+0.035
Neg logit contributions
 angle-0.029
away-0.028
 Sus-0.028
 Mill-0.028
riage-0.027
Tokenric
Token ID1173
Feature activation+1.88e-01

Pos logit contributions
istered+0.111
hes+0.108
oll+0.104
uts+0.098
aj+0.094
Neg logit contributions
riage-0.089
 angle-0.086
 Mill-0.085
 itself-0.083
 Joey-0.082
Tokenhes
Token ID956
Feature activation+1.29e-02

Pos logit contributions
istered+0.165
aj+0.135
ittens+0.133
ucker+0.104
oll+0.097
Neg logit contributions
 itself-0.471
 behold-0.430
riage-0.429
 Where-0.427
 investigate-0.426
Token!"
Token ID2474
Feature activation-1.61e-02

Pos logit contributions
istered+0.025
hes+0.023
aj+0.023
uts+0.023
oll+0.022
Neg logit contributions
away-0.017
 angle-0.017
 Sus-0.017
riage-0.017
 itself-0.017
Token
Token ID198
Feature activation+7.65e-02

Pos logit contributions
 angle+0.021
away+0.021
riage+0.020
keeping+0.020
 Sus+0.019
Neg logit contributions
hes-0.033
uts-0.031
istered-0.030
as-0.029
ows-0.029
Token
Token ID198
Feature activation+5.80e-02

Pos logit contributions
hes+0.147
istered+0.141
as+0.140
uts+0.139
aj+0.132
Neg logit contributions
 angle-0.100
riage-0.097
away-0.096
 Sus-0.096
 Mill-0.095
Token"
Token ID1
Feature activation+7.39e-03

Pos logit contributions
hes+0.095
as+0.091
istered+0.090
uts+0.088
keys+0.085
Neg logit contributions
 Sus-0.094
 angle-0.093
keeping-0.092
 Joey-0.092
riage-0.091
TokenYes
Token ID5297
Feature activation+1.63e-03

Pos logit contributions
hes+0.019
istered+0.018
as+0.018
uts+0.018
zel+0.017
Neg logit contributions
 angle-0.005
away-0.005
 Sus-0.005
 Mill-0.005
riage-0.005
Token,
Token ID11
Feature activation+1.87e-02

Pos logit contributions
hes+0.003
istered+0.003
as+0.003
uts+0.003
aj+0.003
Neg logit contributions
 angle-0.002
away-0.002
riage-0.002
 Sus-0.002
 Mill-0.002
Token they
Token ID484
Feature activation+2.49e-02

Pos logit contributions
istered+0.039
hes+0.037
uts+0.036
aj+0.036
as+0.036
Neg logit contributions
 Sus-0.022
 Mill-0.022
 angle-0.021
 Joey-0.021
away-0.021
Token are
Token ID389
Feature activation+6.59e-02

Pos logit contributions
uts+0.046
istered+0.045
aj+0.045
hes+0.045
ittens+0.044
Neg logit contributions
 investigate-0.035
 inspect-0.033
riage-0.033
 angle-0.032
 Sus-0.032
Token ost
Token ID23619
Feature activation+5.80e-02

Pos logit contributions
istered+0.128
hes+0.126
aj+0.118
oll+0.118
as+0.118
Neg logit contributions
 Mill-0.088
 Sus-0.086
 Joey-0.083
 angle-0.081
 home-0.080
Tokenric
Token ID1173
Feature activation+1.86e-01

Pos logit contributions
hes+0.107
istered+0.106
oll+0.100
uts+0.092
br+0.092
Neg logit contributions
riage-0.082
 angle-0.077
 Mill-0.077
 itself-0.075
 Joey-0.074
Tokenhes
Token ID956
Feature activation+4.75e-03

Pos logit contributions
istered+0.154
aj+0.124
ittens+0.111
ucker+0.093
uts+0.082
Neg logit contributions
 itself-0.472
 behold-0.436
 home-0.432
away-0.430
crow-0.426
Token,"
Token ID553
Feature activation-9.23e-03

Pos logit contributions
istered+0.010
hes+0.010
uts+0.009
aj+0.009
oll+0.009
Neg logit contributions
away-0.005
 Sus-0.005
 Mill-0.005
 Joey-0.005
 itself-0.005
Token mom
Token ID1995
Feature activation-9.81e-05

Pos logit contributions
 angle+0.014
away+0.014
keeping+0.014
riage+0.013
 generation+0.013
Neg logit contributions
hes-0.017
as-0.015
istered-0.014
keys-0.014
uts-0.014
Token said
Token ID531
Feature activation+3.23e-04

Pos logit contributions
 angle+0.000
 Sus+0.000
 Mill+0.000
away+0.000
 Joey+0.000
Neg logit contributions
istered-0.000
hes-0.000
as-0.000
uts-0.000
aj-0.000
Token.
Token ID13
Feature activation-4.07e-02

Pos logit contributions
hes+0.001
istered+0.001
uts+0.001
ows+0.001
as+0.001
Neg logit contributions
 angle-0.000
riage-0.000
away-0.000
 inspect-0.000
 generation-0.000
Token
Token ID198
Feature activation+8.98e-02

Pos logit contributions
 angle+0.026
 Sus+0.026
away+0.026
 Mill+0.025
riage+0.025
Neg logit contributions
hes-0.052
istered-0.052
as-0.050
uts-0.050
aj-0.048
Token
Token ID198
Feature activation+6.94e-02

Pos logit contributions
hes+0.169
istered+0.163
as+0.162
uts+0.161
aj+0.153
Neg logit contributions
 angle-0.120
riage-0.118
away-0.117
 Sus-0.114
 Mill-0.114
Token"
Token ID1
Feature activation+1.21e-02

Pos logit contributions
hes+0.117
istered+0.116
as+0.111
uts+0.110
aj+0.105
Neg logit contributions
 angle-0.106
away-0.104
 Sus-0.104
 Mill-0.103
riage-0.103
TokenO
Token ID46
Feature activation-4.66e-03

Pos logit contributions
istered+0.025
hes+0.025
as+0.024
uts+0.024
aj+0.023
Neg logit contributions
 angle-0.014
away-0.013
 Sus-0.013
 Mill-0.013
riage-0.013
Tokenst
Token ID301
Feature activation+1.62e-02

Pos logit contributions
riage+0.010
away+0.010
 angle+0.009
keeping+0.009
 Sus+0.009
Neg logit contributions
hes-0.005
istered-0.005
as-0.004
aj-0.004
acas-0.004
Tokenric
Token ID1173
Feature activation+1.83e-01

Pos logit contributions
istered+0.025
hes+0.024
uts+0.022
oll+0.021
aj+0.021
Neg logit contributions
away-0.027
 angle-0.027
 Sus-0.027
 investigate-0.027
 itself-0.027
Tokenhes
Token ID956
Feature activation+1.69e-02

Pos logit contributions
istered+0.149
ittens+0.117
aj+0.111
ucker+0.093
uts+0.087
Neg logit contributions
 itself-0.459
 investigate-0.416
 Where-0.414
riage-0.411
 behold-0.411
Token do
Token ID466
Feature activation-9.28e-03

Pos logit contributions
istered+0.028
hes+0.027
uts+0.026
as+0.026
aj+0.025
Neg logit contributions
 angle-0.026
 Sus-0.026
away-0.026
riage-0.026
 Mill-0.025
Token not
Token ID407
Feature activation+8.98e-03

Pos logit contributions
 itself+0.017
riage+0.017
away+0.017
 side+0.016
 investigate+0.016
Neg logit contributions
istered-0.013
umbs-0.012
oll-0.012
hes-0.012
aj-0.012
Token like
Token ID588
Feature activation-1.14e-02

Pos logit contributions
istered+0.017
hes+0.017
uts+0.016
as+0.016
aj+0.016
Neg logit contributions
away-0.012
 angle-0.012
 Sus-0.011
 Mill-0.011
riage-0.011
Token people
Token ID661
Feature activation+2.39e-02

Pos logit contributions
 angle+0.013
 Sus+0.013
 Mill+0.013
away+0.012
 Joey+0.012
Neg logit contributions
istered-0.024
hes-0.023
as-0.022
uts-0.022
aj-0.022
Token that
Token ID326
Feature activation-3.63e-02

Pos logit contributions
istered+0.050
hes+0.049
as+0.047
uts+0.047
aj+0.046
Neg logit contributions
 angle-0.032
 Sus-0.032
away-0.032
 Mill-0.031
riage-0.031
Token said
Token ID531
Feature activation+3.78e-03

Pos logit contributions
 Sus+0.050
 Mill+0.050
 Joey+0.048
riage+0.047
 angle+0.046
Neg logit contributions
istered-0.070
agging-0.066
as-0.064
hes-0.064
aj-0.064
Token "
Token ID366
Feature activation-1.44e-02

Pos logit contributions
hes+0.008
istered+0.008
as+0.008
uts+0.008
aj+0.007
Neg logit contributions
 angle-0.004
away-0.004
 Sus-0.004
 Mill-0.004
riage-0.004
TokenO
Token ID46
Feature activation-7.83e-03

Pos logit contributions
 angle+0.019
 Mill+0.019
 Sus+0.018
 Joey+0.018
riage+0.018
Neg logit contributions
hes-0.026
as-0.026
uts-0.026
istered-0.025
zel-0.024
Tokenst
Token ID301
Feature activation+1.18e-02

Pos logit contributions
away+0.014
riage+0.014
 Mill+0.013
 Joey+0.013
 Sus+0.013
Neg logit contributions
hes-0.011
istered-0.011
aj-0.010
oll-0.010
agging-0.010
Tokenric
Token ID1173
Feature activation+1.82e-01

Pos logit contributions
istered+0.019
hes+0.019
aj+0.017
uts+0.017
as+0.017
Neg logit contributions
away-0.019
 angle-0.019
 investigate-0.018
 Mill-0.018
 Sus-0.018
Tokenhes
Token ID956
Feature activation+8.16e-03

Pos logit contributions
istered+0.164
aj+0.138
ittens+0.127
ucker+0.114
uts+0.103
Neg logit contributions
 itself-0.432
riage-0.423
 Where-0.412
 angle-0.411
away-0.409
Token".
Token ID1911
Feature activation-2.94e-02

Pos logit contributions
istered+0.015
hes+0.015
aj+0.014
as+0.014
uts+0.014
Neg logit contributions
 angle-0.011
riage-0.011
 Sus-0.011
away-0.011
 Mill-0.011
Token Sara
Token ID24799
Feature activation-9.23e-03

Pos logit contributions
 angle+0.043
away+0.043
riage+0.040
 Mill+0.039
keeping+0.039
Neg logit contributions
hes-0.053
uts-0.051
istered-0.049
ows-0.048
aj-0.048
Token and
Token ID290
Feature activation+2.47e-03

Pos logit contributions
 Sus+0.010
 angle+0.010
 Mill+0.010
 Joey+0.010
 inspect+0.009
Neg logit contributions
istered-0.020
hes-0.019
agging-0.018
oll-0.018
uts-0.018
Token Tom
Token ID4186
Feature activation-2.06e-02

Pos logit contributions
hes+0.005
istered+0.004
as+0.004
uts+0.004
aj+0.004
Neg logit contributions
 angle-0.003
 Sus-0.003
 Mill-0.003
away-0.003
riage-0.003
Token and
Token ID290
Feature activation-2.53e-02

Pos logit contributions
hes+0.049
istered+0.049
as+0.047
uts+0.046
agging+0.044
Neg logit contributions
 angle-0.044
riage-0.043
away-0.042
 Sus-0.042
 Mill-0.042
Token looked
Token ID3114
Feature activation+3.16e-02

Pos logit contributions
 angle+0.031
away+0.030
 Sus+0.030
 Mill+0.030
riage+0.029
Neg logit contributions
hes-0.050
istered-0.050
as-0.048
uts-0.048
aj-0.046
Token at
Token ID379
Feature activation-3.76e-02

Pos logit contributions
hes+0.057
istered+0.056
uts+0.055
as+0.054
oll+0.052
Neg logit contributions
away-0.044
 angle-0.043
 Sus-0.043
 Mill-0.043
 Joey-0.042
Token the
Token ID262
Feature activation+1.16e-02

Pos logit contributions
 angle+0.042
away+0.041
riage+0.039
keeping+0.038
 inspect+0.038
Neg logit contributions
hes-0.079
istered-0.073
as-0.073
uts-0.070
ows-0.068
Token ost
Token ID23619
Feature activation+4.30e-02

Pos logit contributions
istered+0.023
uts+0.022
hes+0.022
oll+0.021
aj+0.021
Neg logit contributions
 angle-0.016
away-0.015
 Sus-0.015
 side-0.015
 Mill-0.015
Tokenric
Token ID1173
Feature activation+1.79e-01

Pos logit contributions
istered+0.079
hes+0.076
oll+0.074
uts+0.069
br+0.069
Neg logit contributions
 angle-0.061
riage-0.061
 side-0.059
 Mill-0.057
 itself-0.056
Tokenhes
Token ID956
Feature activation+8.04e-03

Pos logit contributions
istered+0.134
aj+0.104
ittens+0.088
uts+0.072
ucker+0.070
Neg logit contributions
 itself-0.459
 behold-0.428
 angle-0.425
away-0.425
riage-0.423
Token.
Token ID13
Feature activation-4.39e-02

Pos logit contributions
istered+0.016
hes+0.016
as+0.016
uts+0.016
aj+0.015
Neg logit contributions
 angle-0.009
 Sus-0.009
away-0.009
 Mill-0.009
riage-0.009
Token They
Token ID1119
Feature activation+1.04e-02

Pos logit contributions
away+0.073
 angle+0.071
riage+0.064
 inspect+0.064
ie+0.063
Neg logit contributions
hes-0.076
istered-0.071
uts-0.070
ows-0.069
aj-0.067
Token saw
Token ID2497
Feature activation-1.59e-02

Pos logit contributions
istered+0.025
uts+0.024
aj+0.024
hes+0.024
agging+0.024
Neg logit contributions
 Sus-0.009
 investigate-0.009
 Mill-0.009
 Joey-0.009
 inspect-0.009
Token one
Token ID530
Feature activation+5.27e-02

Pos logit contributions
 Sus+0.020
 Mill+0.019
 angle+0.019
 Joey+0.019
away+0.018
Neg logit contributions
istered-0.032
hes-0.031
uts-0.030
as-0.030
aj-0.030
Token
Token ID198
Feature activation+7.85e-02

Pos logit contributions
 angle+0.047
away+0.045
riage+0.042
keeping+0.042
 Mill+0.041
Neg logit contributions
hes-0.066
uts-0.062
istered-0.062
as-0.060
ends-0.059
Token
Token ID198
Feature activation+5.68e-02

Pos logit contributions
hes+0.150
istered+0.144
as+0.142
uts+0.141
aj+0.134
Neg logit contributions
 angle-0.106
away-0.103
riage-0.102
 Sus-0.102
 Mill-0.101
TokenBut
Token ID1537
Feature activation-4.67e-02

Pos logit contributions
istered+0.110
hes+0.108
uts+0.104
as+0.103
oll+0.103
Neg logit contributions
away-0.071
 angle-0.069
riage-0.069
 Sus-0.067
 Mill-0.067
Token the
Token ID262
Feature activation+2.06e-02

Pos logit contributions
 angle+0.060
away+0.058
keeping+0.056
 Sus+0.055
 Mill+0.055
Neg logit contributions
hes-0.094
istered-0.089
as-0.086
uts-0.085
ends-0.084
Token ost
Token ID23619
Feature activation+4.94e-02

Pos logit contributions
istered+0.040
hes+0.039
uts+0.038
as+0.038
oll+0.037
Neg logit contributions
 angle-0.027
away-0.026
 Sus-0.026
riage-0.026
 Mill-0.026
Tokenric
Token ID1173
Feature activation+1.77e-01

Pos logit contributions
istered+0.093
hes+0.090
oll+0.087
br+0.085
uts+0.083
Neg logit contributions
 angle-0.067
riage-0.065
 Mill-0.060
 side-0.060
 itself-0.060
Tokenhes
Token ID956
Feature activation+6.88e-03

Pos logit contributions
istered+0.136
aj+0.107
ittens+0.091
ucker+0.076
uts+0.076
Neg logit contributions
 itself-0.448
 behold-0.417
 angle-0.408
crow-0.403
away-0.403
Token did
Token ID750
Feature activation+2.62e-02

Pos logit contributions
hes+0.014
istered+0.014
as+0.014
uts+0.014
aj+0.013
Neg logit contributions
 angle-0.008
 Sus-0.007
away-0.007
 Mill-0.007
riage-0.007
Token not
Token ID407
Feature activation+1.43e-03

Pos logit contributions
oll+0.033
umbs+0.032
as+0.032
istered+0.032
aj+0.031
Neg logit contributions
 itself-0.051
away-0.050
riage-0.050
 side-0.049
 away-0.049
Token like
Token ID588
Feature activation-8.52e-03

Pos logit contributions
istered+0.003
hes+0.003
aj+0.003
uts+0.003
as+0.003
Neg logit contributions
away-0.002
 angle-0.002
 Sus-0.002
 investigate-0.002
 Mill-0.002
Token that
Token ID326
Feature activation-4.87e-02

Pos logit contributions
 angle+0.008
away+0.007
riage+0.007
 Sus+0.007
 Mill+0.007
Neg logit contributions
hes-0.020
istered-0.019
as-0.019
uts-0.019
aj-0.018
Token should
Token ID815
Feature activation-2.67e-02

Pos logit contributions
istered+0.018
agging+0.018
aj+0.018
hes+0.017
uts+0.017
Neg logit contributions
 Sus-0.011
 Mill-0.011
 investigate-0.011
 Joey-0.011
 inspect-0.010
Token not
Token ID407
Feature activation-7.06e-03

Pos logit contributions
 Joey+0.038
 Mill+0.038
 Sus+0.038
 investigate+0.038
away+0.037
Neg logit contributions
istered-0.049
hes-0.048
as-0.047
uts-0.045
aj-0.045
Token tease
Token ID34691
Feature activation-2.77e-02

Pos logit contributions
 angle+0.009
 Sus+0.008
 Mill+0.008
riage+0.008
away+0.008
Neg logit contributions
hes-0.014
istered-0.014
as-0.013
uts-0.013
aj-0.012
Token the
Token ID262
Feature activation+1.67e-02

Pos logit contributions
away+0.032
 angle+0.032
 Sus+0.031
 Joey+0.031
 Mill+0.031
Neg logit contributions
istered-0.058
hes-0.057
uts-0.053
as-0.053
aj-0.053
Token ost
Token ID23619
Feature activation+3.60e-02

Pos logit contributions
istered+0.031
hes+0.031
uts+0.031
as+0.029
oll+0.029
Neg logit contributions
 angle-0.022
away-0.021
 Sus-0.021
 Mill-0.021
riage-0.021
Tokenric
Token ID1173
Feature activation+1.74e-01

Pos logit contributions
istered+0.068
hes+0.065
oll+0.064
acas+0.059
br+0.059
Neg logit contributions
riage-0.048
 angle-0.048
 side-0.047
 itself-0.045
 Mill-0.044
Tokenhes
Token ID956
Feature activation+8.28e-04

Pos logit contributions
istered+0.140
aj+0.110
ittens+0.087
oll+0.073
acas+0.072
Neg logit contributions
 itself-0.435
away-0.422
riage-0.405
crow-0.405
 home-0.403
Token.
Token ID13
Feature activation-3.41e-02

Pos logit contributions
hes+0.002
istered+0.002
as+0.002
uts+0.002
aj+0.001
Neg logit contributions
 angle-0.001
 Sus-0.001
 Mill-0.001
away-0.001
riage-0.001
Token They
Token ID1119
Feature activation+1.79e-02

Pos logit contributions
 angle+0.039
away+0.037
 Sus+0.036
 Mill+0.035
riage+0.035
Neg logit contributions
hes-0.073
istered-0.071
uts-0.069
as-0.069
aj-0.066
Token are
Token ID389
Feature activation+8.76e-02

Pos logit contributions
aj+0.036
istered+0.035
hes+0.034
agging+0.033
oll+0.033
Neg logit contributions
 investigate-0.023
 Mill-0.021
 Joey-0.021
 inspect-0.021
 Sus-0.021
Token not
Token ID407
Feature activation+5.62e-02

Pos logit contributions
umbs+0.152
cases+0.144
uts+0.139
plets+0.139
istered+0.138
Neg logit contributions
 home-0.141
 itself-0.139
 the-0.133
 Tim-0.131
 its-0.131
Token and
Token ID290
Feature activation-7.16e-03

Pos logit contributions
hes+0.009
istered+0.008
as+0.008
uts+0.008
ows+0.008
Neg logit contributions
 Sus-0.007
 angle-0.007
 Mill-0.007
riage-0.007
away-0.007
Token run
Token ID1057
Feature activation-9.58e-03

Pos logit contributions
 angle+0.008
 inspect+0.008
away+0.008
riage+0.008
 Sus+0.008
Neg logit contributions
istered-0.015
hes-0.015
as-0.015
uts-0.014
aj-0.014
Token to
Token ID284
Feature activation-8.77e-02

Pos logit contributions
riage+0.013
 angle+0.013
 Mill+0.013
 Sus+0.012
 Joey+0.012
Neg logit contributions
as-0.018
uts-0.018
hes-0.018
ends-0.017
istered-0.017
Token the
Token ID262
Feature activation+9.97e-03

Pos logit contributions
 angle+0.094
riage+0.092
keeping+0.091
away+0.089
 zone+0.087
Neg logit contributions
hes-0.181
istered-0.168
ends-0.168
as-0.164
ows-0.163
Token ost
Token ID23619
Feature activation+3.67e-02

Pos logit contributions
hes+0.019
istered+0.018
as+0.018
uts+0.017
aj+0.017
Neg logit contributions
 Sus-0.013
 angle-0.013
 Mill-0.013
away-0.013
 Joey-0.013
Tokenric
Token ID1173
Feature activation+1.73e-01

Pos logit contributions
istered+0.065
hes+0.063
oll+0.062
uts+0.057
acas+0.056
Neg logit contributions
 angle-0.054
riage-0.054
 side-0.052
away-0.050
keeping-0.050
Tokenhes
Token ID956
Feature activation+1.18e-02

Pos logit contributions
istered+0.135
aj+0.100
ittens+0.084
acas+0.069
ucker+0.069
Neg logit contributions
 itself-0.443
away-0.415
 behold-0.415
 home-0.413
 angle-0.412
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
hes+0.023
istered+0.023
as+0.022
uts+0.022
aj+0.021
Neg logit contributions
 angle-0.015
 Sus-0.014
 Mill-0.014
away-0.014
riage-0.014
Token
Token ID198
Feature activation+8.93e-02

Pos logit contributions
 angle+0.033
away+0.032
riage+0.030
keeping+0.030
 Mill+0.029
Neg logit contributions
hes-0.045
uts-0.042
as-0.041
istered-0.041
ends-0.041
Token
Token ID198
Feature activation+6.63e-02

Pos logit contributions
hes+0.173
istered+0.165
as+0.164
uts+0.162
aj+0.154
Neg logit contributions
 angle-0.116
away-0.113
riage-0.113
 Sus-0.112
 Mill-0.112
Token"
Token ID1
Feature activation+9.21e-03

Pos logit contributions
istered+0.126
hes+0.122
oll+0.119
uts+0.118
aj+0.116
Neg logit contributions
away-0.086
 angle-0.085
riage-0.083
 Sus-0.082
 Mill-0.082
Token Did
Token ID7731
Feature activation+1.39e-03

Pos logit contributions
 angle+0.026
away+0.025
keeping+0.023
 Sus+0.023
 Mill+0.023
Neg logit contributions
hes-0.044
istered-0.042
as-0.042
uts-0.041
aj-0.040
Token you
Token ID345
Feature activation+7.13e-03

Pos logit contributions
istered+0.003
hes+0.003
as+0.002
umbs+0.002
agging+0.002
Neg logit contributions
 Sus-0.002
 Mill-0.002
 Joey-0.002
 itself-0.002
riage-0.002
Token know
Token ID760
Feature activation-1.27e-02

Pos logit contributions
istered+0.016
hes+0.016
aj+0.016
as+0.015
oll+0.015
Neg logit contributions
 investigate-0.007
 Sus-0.007
 Joey-0.007
 Mill-0.007
away-0.007
Token that
Token ID326
Feature activation-2.29e-02

Pos logit contributions
 Sus+0.010
 Mill+0.010
riage+0.009
 angle+0.009
 Joey+0.009
Neg logit contributions
istered-0.033
hes-0.031
uts-0.030
umbs-0.030
as-0.029
Token cater
Token ID20825
Feature activation+6.42e-02

Pos logit contributions
 angle+0.027
 Sus+0.027
 Mill+0.027
away+0.026
riage+0.026
Neg logit contributions
istered-0.047
hes-0.046
as-0.045
uts-0.044
aj-0.043
Tokenpill
Token ID27215
Feature activation+1.71e-01

Pos logit contributions
istered+0.092
hes+0.085
aj+0.084
cases+0.078
ends+0.078
Neg logit contributions
 Mill-0.118
 angle-0.118
away-0.116
 Joey-0.116
 Sus-0.114
Tokenars
Token ID945
Feature activation+5.78e-02

Pos logit contributions
aj+0.224
istered+0.187
ucker+0.167
ittens+0.158
cases+0.157
Neg logit contributions
away-0.362
 behold-0.357
 itself-0.347
 investigate-0.335
crow-0.331
Token turn
Token ID1210
Feature activation+3.72e-02

Pos logit contributions
istered+0.116
hes+0.110
aj+0.108
oll+0.107
uts+0.104
Neg logit contributions
away-0.073
 Joey-0.070
 angle-0.070
 Mill-0.069
 Sus-0.068
Token into
Token ID656
Feature activation-1.97e-02

Pos logit contributions
istered+0.080
hes+0.077
uts+0.074
oll+0.073
as+0.073
Neg logit contributions
away-0.042
 Joey-0.041
 Sus-0.041
 Mill-0.041
 angle-0.041
Token beautiful
Token ID4950
Feature activation-3.37e-02

Pos logit contributions
 angle+0.023
 Sus+0.023
 Mill+0.023
away+0.023
 Joey+0.022
Neg logit contributions
istered-0.040
hes-0.040
as-0.039
uts-0.039
aj-0.037
Token butterflies
Token ID44571
Feature activation+2.45e-02

Pos logit contributions
 angle+0.048
riage+0.047
 Sus+0.047
away+0.046
 Mill+0.046
Neg logit contributions
hes-0.059
istered-0.058
as-0.057
uts-0.056
aj-0.054
Token They
Token ID1119
Feature activation+8.10e-03

Pos logit contributions
 angle+0.068
away+0.064
 Mill+0.062
 inspect+0.061
 Sus+0.059
Neg logit contributions
hes-0.071
uts-0.070
istered-0.067
ends-0.064
as-0.062
Token do
Token ID466
Feature activation-2.26e-03

Pos logit contributions
hes+0.016
istered+0.015
uts+0.015
as+0.015
aj+0.014
Neg logit contributions
 angle-0.010
away-0.010
 Sus-0.010
 Mill-0.010
riage-0.010
Token not
Token ID407
Feature activation+1.12e-02

Pos logit contributions
riage+0.005
 angle+0.005
away+0.005
 Sus+0.005
 itself+0.005
Neg logit contributions
istered-0.002
as-0.002
hes-0.002
uts-0.002
oll-0.002
Token pack
Token ID2353
Feature activation-1.56e-02

Pos logit contributions
hes+0.020
istered+0.020
as+0.019
uts+0.019
oll+0.018
Neg logit contributions
 angle-0.016
 Sus-0.015
 Mill-0.015
riage-0.015
 Joey-0.015
Token ost
Token ID23619
Feature activation+2.65e-02

Pos logit contributions
 angle+0.017
away+0.016
 Sus+0.016
 Mill+0.016
riage+0.016
Neg logit contributions
hes-0.033
istered-0.033
as-0.032
uts-0.032
aj-0.030
Tokenric
Token ID1173
Feature activation+1.69e-01

Pos logit contributions
istered+0.049
hes+0.049
oll+0.046
br+0.045
arest+0.045
Neg logit contributions
riage-0.033
keeping-0.033
wood-0.032
 home-0.031
en-0.031
Tokenhes
Token ID956
Feature activation-2.71e-03

Pos logit contributions
istered+0.134
aj+0.101
ittens+0.090
ucker+0.083
uts+0.077
Neg logit contributions
 itself-0.420
 home-0.397
away-0.396
 him-0.393
riage-0.387
Token anymore
Token ID7471
Feature activation-3.06e-02

Pos logit contributions
 angle+0.003
 Sus+0.003
 Mill+0.003
away+0.003
riage+0.003
Neg logit contributions
hes-0.006
istered-0.006
as-0.006
uts-0.006
aj-0.005
Token.
Token ID13
Feature activation-3.23e-02

Pos logit contributions
 angle+0.035
away+0.035
 Sus+0.035
 Mill+0.034
riage+0.034
Neg logit contributions
hes-0.062
istered-0.062
as-0.059
uts-0.058
aj-0.057
Token They
Token ID1119
Feature activation+9.78e-03

Pos logit contributions
 angle+0.045
away+0.043
 inspect+0.040
 Mill+0.040
 Sus+0.039
Neg logit contributions
hes-0.063
uts-0.061
istered-0.060
as-0.058
ends-0.056
Token pack
Token ID2353
Feature activation-1.57e-02

Pos logit contributions
hes+0.018
istered+0.018
uts+0.017
as+0.017
aj+0.017
Neg logit contributions
 angle-0.013
away-0.013
 investigate-0.013
riage-0.012
 Sus-0.012
Token did
Token ID750
Feature activation+2.86e-02

Pos logit contributions
 angle+0.028
away+0.027
 Sus+0.027
 Mill+0.026
riage+0.026
Neg logit contributions
hes-0.038
istered-0.038
as-0.037
uts-0.036
aj-0.035
Token not
Token ID407
Feature activation+3.11e-03

Pos logit contributions
hes+0.023
istered+0.023
as+0.021
uts+0.020
aj+0.019
Neg logit contributions
 angle-0.068
 Sus-0.068
away-0.068
 Mill-0.068
riage-0.067
Token know
Token ID760
Feature activation-1.52e-02

Pos logit contributions
hes+0.006
istered+0.006
as+0.006
uts+0.005
aj+0.005
Neg logit contributions
 angle-0.004
 Sus-0.004
away-0.004
 Mill-0.004
riage-0.004
Token what
Token ID644
Feature activation-1.54e-02

Pos logit contributions
 Sus+0.015
 Mill+0.015
riage+0.015
 angle+0.015
 Joey+0.014
Neg logit contributions
istered-0.034
hes-0.033
uts-0.032
as-0.032
aj-0.031
Token ost
Token ID23619
Feature activation+3.75e-02

Pos logit contributions
 angle+0.016
away+0.015
keeping+0.015
 Sus+0.015
riage+0.015
Neg logit contributions
hes-0.035
as-0.032
uts-0.032
keys-0.031
istered-0.031
Tokenric
Token ID1173
Feature activation+1.68e-01

Pos logit contributions
istered+0.070
hes+0.068
oll+0.063
zel+0.060
br+0.060
Neg logit contributions
riage-0.053
 angle-0.050
 Mill-0.050
 Joey-0.049
 Sus-0.048
Tokenhes
Token ID956
Feature activation-2.97e-03

Pos logit contributions
istered+0.142
aj+0.117
ittens+0.093
oll+0.082
ucker+0.081
Neg logit contributions
 itself-0.406
riage-0.404
away-0.393
 angle-0.392
 Joey-0.386
Token were
Token ID547
Feature activation+2.37e-02

Pos logit contributions
 angle+0.003
 Sus+0.003
 Mill+0.003
riage+0.003
away+0.003
Neg logit contributions
istered-0.007
hes-0.006
as-0.006
uts-0.006
aj-0.006
Token.
Token ID13
Feature activation-4.45e-02

Pos logit contributions
istered+0.050
hes+0.048
oll+0.046
as+0.045
aj+0.045
Neg logit contributions
 Mill-0.026
 angle-0.026
 Sus-0.026
riage-0.026
 Joey-0.025
Token They
Token ID1119
Feature activation+4.73e-03

Pos logit contributions
away+0.058
 angle+0.057
riage+0.053
 Mill+0.053
keeping+0.053
Neg logit contributions
hes-0.088
uts-0.084
istered-0.083
as-0.081
ows-0.080
Token looked
Token ID3114
Feature activation+4.28e-02

Pos logit contributions
hes+0.010
istered+0.010
uts+0.009
as+0.009
aj+0.009
Neg logit contributions
 angle-0.005
 Sus-0.005
 Mill-0.005
away-0.005
 Joey-0.005
Token.
Token ID13
Feature activation-4.90e-02

Pos logit contributions
away+0.006
 home+0.006
 angle+0.006
 side+0.005
keeping+0.005
Neg logit contributions
hes-0.015
istered-0.014
as-0.014
uts-0.014
keys-0.013
Token It
Token ID632
Feature activation+2.28e-02

Pos logit contributions
 angle+0.061
away+0.057
keeping+0.053
 Sus+0.051
 Mill+0.051
Neg logit contributions
hes-0.103
uts-0.097
istered-0.096
oll-0.093
ows-0.093
Token sees
Token ID7224
Feature activation-3.04e-02

Pos logit contributions
istered+0.041
uts+0.040
hes+0.039
as+0.038
ittens+0.038
Neg logit contributions
 inspect-0.032
 angle-0.031
 investigate-0.031
riage-0.031
away-0.031
Token other
Token ID584
Feature activation-3.02e-02

Pos logit contributions
 Sus+0.030
away+0.029
 angle+0.029
 Joey+0.029
 Mill+0.029
Neg logit contributions
istered-0.068
hes-0.065
uts-0.064
aj-0.064
as-0.064
Token ost
Token ID23619
Feature activation+3.49e-02

Pos logit contributions
 angle+0.042
away+0.040
 Joey+0.040
riage+0.040
keeping+0.040
Neg logit contributions
as-0.053
hes-0.053
istered-0.052
uts-0.049
aj-0.048
Tokenric
Token ID1173
Feature activation+1.68e-01

Pos logit contributions
istered+0.067
hes+0.065
oll+0.063
br+0.061
uts+0.060
Neg logit contributions
riage-0.047
 angle-0.044
keeping-0.043
 side-0.043
 Mill-0.042
Tokenhes
Token ID956
Feature activation-2.65e-03

Pos logit contributions
istered+0.133
ittens+0.105
aj+0.100
ucker+0.089
uts+0.077
Neg logit contributions
 itself-0.428
 home-0.394
 behold-0.392
 him-0.391
 his-0.383
Token.
Token ID13
Feature activation-4.02e-02

Pos logit contributions
 angle+0.002
 Sus+0.002
away+0.002
 Mill+0.002
riage+0.002
Neg logit contributions
hes-0.006
istered-0.006
as-0.006
uts-0.006
aj-0.006
Token It
Token ID632
Feature activation+1.60e-02

Pos logit contributions
 angle+0.038
away+0.035
riage+0.033
 Sus+0.033
 Mill+0.033
Neg logit contributions
hes-0.095
istered-0.091
uts-0.090
as-0.088
oll-0.086
Token makes
Token ID1838
Feature activation+9.39e-04

Pos logit contributions
istered+0.029
uts+0.029
ittens+0.028
hes+0.028
as+0.028
Neg logit contributions
 inspect-0.021
 investigate-0.021
 angle-0.021
away-0.020
riage-0.020
Token new
Token ID649
Feature activation-2.69e-02

Pos logit contributions
hes+0.002
istered+0.002
as+0.002
uts+0.002
aj+0.002
Neg logit contributions
 angle-0.001
 Sus-0.001
away-0.001
 Mill-0.001
 Joey-0.001
Token to
Token ID284
Feature activation-9.71e-02

Pos logit contributions
 angle+0.034
riage+0.034
 Mill+0.031
 Sus+0.031
keeping+0.030
Neg logit contributions
hes-0.055
uts-0.054
as-0.054
ends-0.054
istered-0.054
Token the
Token ID262
Feature activation-9.57e-03

Pos logit contributions
keeping+0.105
riage+0.102
 angle+0.102
away+0.101
 Mill+0.094
Neg logit contributions
hes-0.202
ends-0.189
istered-0.186
as-0.184
ows-0.182
Token hill
Token ID12788
Feature activation-1.80e-03

Pos logit contributions
 angle+0.009
away+0.008
 Sus+0.008
 Mill+0.008
riage+0.008
Neg logit contributions
istered-0.022
hes-0.022
as-0.022
uts-0.022
aj-0.021
Token and
Token ID290
Feature activation-3.66e-02

Pos logit contributions
riage+0.003
 angle+0.003
 Mill+0.002
away+0.002
 Sus+0.002
Neg logit contributions
hes-0.003
istered-0.003
as-0.003
uts-0.003
ows-0.003
Token climbed
Token ID19952
Feature activation-5.10e-02

Pos logit contributions
 angle+0.051
away+0.049
riage+0.049
 Sus+0.049
 Mill+0.049
Neg logit contributions
hes-0.067
istered-0.066
as-0.064
uts-0.063
aj-0.060
Token to
Token ID284
Feature activation-1.24e-01

Pos logit contributions
 angle+0.068
riage+0.065
away+0.065
 Mill+0.064
keeping+0.063
Neg logit contributions
hes-0.096
ends-0.089
uts-0.089
as-0.088
ows-0.088
Token the
Token ID262
Feature activation-1.79e-02

Pos logit contributions
keeping+0.132
riage+0.127
 angle+0.127
 Mill+0.122
away+0.122
Neg logit contributions
hes-0.256
istered-0.235
as-0.234
ends-0.231
ows-0.229
Token top
Token ID1353
Feature activation-1.35e-02

Pos logit contributions
 angle+0.018
away+0.017
 Sus+0.017
 Mill+0.017
riage+0.017
Neg logit contributions
hes-0.040
istered-0.040
as-0.039
uts-0.038
aj-0.038
Token.
Token ID13
Feature activation-5.19e-02

Pos logit contributions
 angle+0.019
away+0.018
 Sus+0.018
 Mill+0.018
riage+0.018
Neg logit contributions
istered-0.025
hes-0.025
as-0.024
uts-0.023
aj-0.023
Token They
Token ID1119
Feature activation-5.81e-03

Pos logit contributions
away+0.065
 angle+0.063
riage+0.062
keeping+0.059
 Mill+0.058
Neg logit contributions
hes-0.107
uts-0.102
istered-0.101
as-0.099
ends-0.097
Token looked
Token ID3114
Feature activation+3.63e-02

Pos logit contributions
 Sus+0.007
 angle+0.007
 Joey+0.007
 Mill+0.007
 inspect+0.007
Neg logit contributions
istered-0.012
hes-0.011
uts-0.011
aj-0.011
agging-0.011
Token tall
Token ID7331
Feature activation-3.17e-02

Pos logit contributions
hes+0.033
istered+0.032
as+0.031
uts+0.031
aj+0.029
Neg logit contributions
 angle-0.040
 Sus-0.039
away-0.039
 Mill-0.039
riage-0.038
Token tree
Token ID5509
Feature activation+4.60e-03

Pos logit contributions
riage+0.039
 Sus+0.038
keeping+0.038
 Joey+0.038
 Mill+0.038
Neg logit contributions
hes-0.059
istered-0.058
as-0.057
uts-0.056
zel-0.055
Token and
Token ID290
Feature activation-2.25e-02

Pos logit contributions
istered+0.009
hes+0.009
as+0.009
uts+0.009
aj+0.008
Neg logit contributions
 angle-0.006
away-0.006
 Sus-0.005
 Mill-0.005
 Joey-0.005
Token climbed
Token ID19952
Feature activation-4.71e-02

Pos logit contributions
 angle+0.030
away+0.029
riage+0.029
 Sus+0.029
 Mill+0.028
Neg logit contributions
hes-0.043
istered-0.041
as-0.040
uts-0.039
aj-0.038
Token up
Token ID510
Feature activation-5.77e-02

Pos logit contributions
 angle+0.068
riage+0.068
 basis+0.065
away+0.065
 zone+0.063
Neg logit contributions
hes-0.085
ends-0.077
uts-0.074
ows-0.073
br-0.070
Token to
Token ID284
Feature activation-1.22e-01

Pos logit contributions
riage+0.090
 zone+0.083
 angle+0.082
 basis+0.078
 state+0.078
Neg logit contributions
hes-0.097
uts-0.092
ows-0.085
ends-0.084
 long-0.082
Token the
Token ID262
Feature activation-1.82e-02

Pos logit contributions
keeping+0.149
 angle+0.145
 zone+0.144
riage+0.141
 Mill+0.137
Neg logit contributions
hes-0.229
ends-0.191
istered-0.191
 allergic-0.190
eps-0.182
Token top
Token ID1353
Feature activation-1.24e-02

Pos logit contributions
 angle+0.018
 Sus+0.018
away+0.018
 Mill+0.017
riage+0.017
Neg logit contributions
hes-0.040
istered-0.040
as-0.039
uts-0.038
aj-0.037
Token.
Token ID13
Feature activation-5.48e-02

Pos logit contributions
 angle+0.020
 Sus+0.019
away+0.019
 Mill+0.019
riage+0.019
Neg logit contributions
hes-0.020
istered-0.020
as-0.019
uts-0.019
aj-0.018
Token From
Token ID3574
Feature activation-2.96e-02

Pos logit contributions
away+0.077
 angle+0.076
riage+0.068
keeping+0.066
 behold+0.064
Neg logit contributions
hes-0.105
uts-0.099
ows-0.097
cases-0.094
aj-0.092
Token there
Token ID612
Feature activation-1.12e-03

Pos logit contributions
 angle+0.045
keeping+0.045
 Sus+0.044
 Mill+0.043
riage+0.043
Neg logit contributions
hes-0.049
as-0.046
istered-0.046
uts-0.045
aj-0.044
Tokenaded
Token ID5286
Feature activation-1.63e-02

Pos logit contributions
istered+0.099
aj+0.089
hes+0.089
zel+0.087
uts+0.086
Neg logit contributions
 Joey-0.094
away-0.094
 angle-0.090
 home-0.087
 itself-0.086
Token in
Token ID287
Feature activation-4.39e-02

Pos logit contributions
riage+0.021
 angle+0.021
 Sus+0.020
 Mill+0.020
away+0.020
Neg logit contributions
hes-0.032
as-0.030
istered-0.030
uts-0.029
aj-0.028
Token the
Token ID262
Feature activation-4.38e-04

Pos logit contributions
riage+0.052
 Sus+0.046
 Mill+0.045
away+0.045
 angle+0.043
Neg logit contributions
hes-0.090
istered-0.083
as-0.083
aj-0.080
umbs-0.080
Token water
Token ID1660
Feature activation+1.86e-02

Pos logit contributions
 angle+0.001
away+0.001
 Sus+0.001
 Mill+0.001
riage+0.001
Neg logit contributions
hes-0.001
istered-0.001
as-0.001
uts-0.001
aj-0.001
Token delic
Token ID8675
Feature activation-6.19e-03

Pos logit contributions
istered+0.042
hes+0.040
uts+0.039
oll+0.039
aj+0.039
Neg logit contributions
 angle-0.019
away-0.019
 home-0.019
 side-0.019
 Joey-0.018
Tokenately
Token ID1286
Feature activation-1.21e-01

Pos logit contributions
 angle+0.006
 Joey+0.006
away+0.006
 itself+0.006
ened+0.006
Neg logit contributions
hes-0.014
istered-0.014
zel-0.013
as-0.013
ends-0.013
Token with
Token ID351
Feature activation-1.90e-02

Pos logit contributions
 angle+0.135
away+0.131
 Joey+0.125
 side+0.123
 Mill+0.122
Neg logit contributions
istered-0.270
hes-0.259
as-0.248
uts-0.245
oll-0.244
Token his
Token ID465
Feature activation-1.29e-02

Pos logit contributions
 Joey+0.015
 angle+0.014
 Sus+0.014
away+0.014
 inspect+0.014
Neg logit contributions
istered-0.047
uts-0.047
as-0.046
hes-0.046
zel-0.045
Token fishing
Token ID12478
Feature activation-4.33e-03

Pos logit contributions
 home+0.018
 Joey+0.018
 side+0.017
 angle+0.017
away+0.016
Neg logit contributions
hes-0.026
istered-0.025
uts-0.024
ows-0.024
oll-0.024
Token rod
Token ID15299
Feature activation+2.80e-02

Pos logit contributions
 angle+0.006
away+0.006
 home+0.006
 side+0.006
 Joey+0.006
Neg logit contributions
istered-0.008
hes-0.008
uts-0.008
as-0.007
oll-0.007
Token.
Token ID13
Feature activation-2.95e-02

Pos logit contributions
istered+0.068
hes+0.065
as+0.062
aj+0.062
zel+0.062
Neg logit contributions
 angle-0.022
 Joey-0.022
 home-0.022
 side-0.021
 itself-0.021
Token girl
Token ID2576
Feature activation-2.50e-02

Pos logit contributions
 angle+0.035
riage+0.034
away+0.033
 Sus+0.033
 Mill+0.032
Neg logit contributions
hes-0.103
istered-0.102
as-0.099
uts-0.098
oll-0.095
Token called
Token ID1444
Feature activation-6.36e-03

Pos logit contributions
riage+0.042
 angle+0.040
 itself+0.039
away+0.038
keeping+0.038
Neg logit contributions
hes-0.039
as-0.036
itt-0.036
istered-0.035
oll-0.033
Token Daisy
Token ID40355
Feature activation-4.10e-02

Pos logit contributions
 angle+0.009
away+0.009
riage+0.008
eled+0.008
 state+0.008
Neg logit contributions
uts-0.011
hes-0.010
istered-0.010
oll-0.010
as-0.010
Token went
Token ID1816
Feature activation-2.65e-02

Pos logit contributions
riage+0.063
 angle+0.062
away+0.055
 itself+0.053
 basis+0.053
Neg logit contributions
hes-0.072
istered-0.068
as-0.066
uts-0.063
ends-0.061
Token outside
Token ID2354
Feature activation-6.26e-02

Pos logit contributions
 angle+0.035
riage+0.034
 Sus+0.032
 behold+0.032
away+0.031
Neg logit contributions
hes-0.052
istered-0.049
uts-0.048
as-0.048
agging-0.047
Token to
Token ID284
Feature activation-1.20e-01

Pos logit contributions
riage+0.108
 angle+0.097
away+0.097
 behold+0.096
keeping+0.095
Neg logit contributions
hes-0.092
uts-0.083
istered-0.083
as-0.081
agging-0.079
Token wander
Token ID27776
Feature activation-5.40e-02

Pos logit contributions
keeping+0.162
away+0.159
 angle+0.158
riage+0.148
 itself+0.140
Neg logit contributions
hes-0.216
istered-0.184
aj-0.183
 allergic-0.182
ends-0.177
Token.
Token ID13
Feature activation-4.72e-02

Pos logit contributions
riage+0.092
 angle+0.091
 inspect+0.083
ACK+0.082
 behold+0.082
Neg logit contributions
hes-0.081
agging-0.076
istered-0.075
uts-0.074
as-0.074
Token She
Token ID1375
Feature activation-3.59e-02

Pos logit contributions
riage+0.067
away+0.066
 angle+0.063
keeping+0.060
 Sus+0.057
Neg logit contributions
hes-0.081
uts-0.080
istered-0.079
ows-0.075
umbs-0.075
Token picked
Token ID6497
Feature activation-5.06e-02

Pos logit contributions
 angle+0.044
away+0.042
 Sus+0.042
riage+0.042
 Mill+0.042
Neg logit contributions
hes-0.071
istered-0.070
as-0.068
uts-0.067
aj-0.065
Token up
Token ID510
Feature activation-4.07e-02

Pos logit contributions
 angle+0.074
riage+0.073
away+0.070
 Sus+0.068
keeping+0.067
Neg logit contributions
hes-0.092
istered-0.085
uts-0.084
as-0.084
oll-0.082
Token big
Token ID1263
Feature activation-5.73e-02

Pos logit contributions
 angle+0.012
 side+0.012
 home+0.012
away+0.012
 Sus+0.012
Neg logit contributions
istered-0.018
hes-0.018
uts-0.017
as-0.017
aj-0.017
Token slide
Token ID10649
Feature activation+9.15e-03

Pos logit contributions
riage+0.082
keeping+0.082
 Mill+0.082
 Sus+0.081
 angle+0.080
Neg logit contributions
hes-0.090
as-0.089
aj-0.085
istered-0.085
agging-0.085
Token and
Token ID290
Feature activation-4.22e-02

Pos logit contributions
istered+0.019
hes+0.018
as+0.017
uts+0.017
aj+0.017
Neg logit contributions
 angle-0.011
 Sus-0.011
away-0.011
 Mill-0.011
 Joey-0.011
Token climbed
Token ID19952
Feature activation-4.39e-02

Pos logit contributions
 angle+0.054
away+0.052
 Sus+0.051
 Mill+0.051
riage+0.051
Neg logit contributions
hes-0.082
istered-0.080
as-0.078
uts-0.077
aj-0.074
Token up
Token ID510
Feature activation-5.21e-02

Pos logit contributions
 angle+0.050
riage+0.049
 Mill+0.048
away+0.048
 Sus+0.047
Neg logit contributions
hes-0.092
uts-0.085
as-0.084
istered-0.084
ends-0.083
Token to
Token ID284
Feature activation-1.20e-01

Pos logit contributions
riage+0.067
 angle+0.063
 Mill+0.059
away+0.058
 generation+0.057
Neg logit contributions
hes-0.100
uts-0.097
as-0.091
ows-0.090
istered-0.090
Token the
Token ID262
Feature activation-9.22e-03

Pos logit contributions
keeping+0.115
 Mill+0.112
 angle+0.112
away+0.107
riage+0.107
Neg logit contributions
hes-0.263
as-0.235
istered-0.233
ends-0.230
agging-0.227
Token top
Token ID1353
Feature activation-3.21e-03

Pos logit contributions
 angle+0.012
 side+0.012
 home+0.011
 investigate+0.011
 Sus+0.011
Neg logit contributions
istered-0.019
hes-0.018
uts-0.017
as-0.017
aj-0.017
Token.
Token ID13
Feature activation-5.19e-02

Pos logit contributions
 angle+0.004
away+0.004
 Sus+0.004
 side+0.004
 Mill+0.004
Neg logit contributions
istered-0.006
hes-0.006
as-0.006
aj-0.006
uts-0.006
Token He
Token ID679
Feature activation-4.07e-02

Pos logit contributions
away+0.060
 angle+0.056
riage+0.053
keeping+0.052
 Mill+0.049
Neg logit contributions
hes-0.110
uts-0.105
istered-0.101
cases-0.101
ends-0.100
Token slid
Token ID27803
Feature activation-2.13e-02

Pos logit contributions
 Sus+0.043
 angle+0.043
 Joey+0.041
 inspect+0.041
 Mill+0.041
Neg logit contributions
istered-0.090
hes-0.086
uts-0.085
agging-0.082
as-0.082
Token girl
Token ID2576
Feature activation-2.45e-02

Pos logit contributions
keeping+0.038
riage+0.038
 angle+0.036
 itself+0.036
 inspect+0.035
Neg logit contributions
hes-0.063
istered-0.063
as-0.062
oll-0.061
uts-0.060
Token called
Token ID1444
Feature activation-5.32e-03

Pos logit contributions
riage+0.043
 angle+0.041
 itself+0.040
away+0.040
keeping+0.039
Neg logit contributions
hes-0.037
as-0.034
itt-0.033
istered-0.031
oll-0.030
Token Lucy
Token ID22162
Feature activation-2.93e-02

Pos logit contributions
 angle+0.007
away+0.007
riage+0.007
eled+0.006
 state+0.006
Neg logit contributions
hes-0.009
uts-0.009
as-0.009
oll-0.009
istered-0.009
Token went
Token ID1816
Feature activation-2.74e-02

Pos logit contributions
riage+0.045
 angle+0.044
away+0.040
 Joey+0.039
 basis+0.039
Neg logit contributions
hes-0.053
as-0.048
istered-0.047
uts-0.045
ends-0.044
Token outside
Token ID2354
Feature activation-6.29e-02

Pos logit contributions
 angle+0.038
riage+0.038
 behold+0.036
 Sus+0.035
keeping+0.035
Neg logit contributions
hes-0.051
uts-0.047
istered-0.047
agging-0.046
as-0.046
Token to
Token ID284
Feature activation-1.20e-01

Pos logit contributions
riage+0.111
away+0.100
 angle+0.099
keeping+0.099
 behold+0.098
Neg logit contributions
hes-0.090
uts-0.079
as-0.078
istered-0.077
agging-0.076
Token play
Token ID711
Feature activation-4.86e-02

Pos logit contributions
keeping+0.176
away+0.167
 angle+0.166
riage+0.157
 itself+0.149
Neg logit contributions
hes-0.200
 allergic-0.167
 cute-0.164
aj-0.161
istered-0.160
Token in
Token ID287
Feature activation-7.75e-02

Pos logit contributions
riage+0.091
 angle+0.088
 Sus+0.084
 Mill+0.084
keeping+0.083
Neg logit contributions
hes-0.074
agging-0.066
istered-0.064
uts-0.063
as-0.061
Token the
Token ID262
Feature activation-3.24e-02

Pos logit contributions
riage+0.108
away+0.098
 angle+0.095
 zone+0.091
ened+0.090
Neg logit contributions
hes-0.140
uts-0.121
as-0.117
aj-0.117
istered-0.117
Token park
Token ID3952
Feature activation-3.78e-02

Pos logit contributions
riage+0.045
 angle+0.045
 Sus+0.044
away+0.043
 Mill+0.043
Neg logit contributions
hes-0.058
istered-0.055
as-0.053
aj-0.052
uts-0.052
Token.
Token ID13
Feature activation-4.77e-02

Pos logit contributions
 angle+0.055
riage+0.054
away+0.054
 Mill+0.054
 inspect+0.053
Neg logit contributions
hes-0.065
uts-0.062
as-0.062
istered-0.062
ows-0.058
Token the
Token ID262
Feature activation+3.16e-03

Pos logit contributions
riage+0.114
 angle+0.113
away+0.111
keeping+0.110
 zone+0.109
Neg logit contributions
hes-0.171
ends-0.152
istered-0.149
ows-0.143
 allergic-0.143
Token stairs
Token ID16046
Feature activation+1.08e-02

Pos logit contributions
istered+0.006
hes+0.005
as+0.005
ows+0.005
aj+0.005
Neg logit contributions
 Joey-0.005
 Sus-0.004
 Mill-0.004
riage-0.004
keeping-0.004
Token and
Token ID290
Feature activation-2.55e-02

Pos logit contributions
hes+0.018
istered+0.017
as+0.017
uts+0.017
ows+0.016
Neg logit contributions
riage-0.016
 Mill-0.016
 Joey-0.016
 Sus-0.016
 angle-0.016
Token climbed
Token ID19952
Feature activation-4.37e-02

Pos logit contributions
 angle+0.039
away+0.038
riage+0.037
 Joey+0.036
 Mill+0.036
Neg logit contributions
hes-0.045
istered-0.041
as-0.041
ends-0.040
ows-0.039
Token up
Token ID510
Feature activation-5.02e-02

Pos logit contributions
 angle+0.071
 Joey+0.069
riage+0.068
 basis+0.068
 Mill+0.067
Neg logit contributions
hes-0.075
ows-0.066
ends-0.064
as-0.061
uts-0.060
Token to
Token ID284
Feature activation-1.20e-01

Pos logit contributions
riage+0.068
 angle+0.064
 Joey+0.062
 inspect+0.060
 zone+0.060
Neg logit contributions
hes-0.097
ows-0.088
uts-0.088
as-0.083
cases-0.082
Token the
Token ID262
Feature activation-4.86e-03

Pos logit contributions
keeping+0.166
 angle+0.153
crow+0.150
 zone+0.148
away+0.146
Neg logit contributions
hes-0.210
ends-0.181
ows-0.172
istered-0.170
as-0.168
Token attic
Token ID46113
Feature activation-3.03e-02

Pos logit contributions
 Joey+0.007
 Sus+0.007
riage+0.007
 angle+0.007
 Mill+0.007
Neg logit contributions
hes-0.009
istered-0.009
as-0.008
uts-0.008
ows-0.008
Token.
Token ID13
Feature activation-4.26e-02

Pos logit contributions
 angle+0.040
 Mill+0.040
 Sus+0.040
 Joey+0.039
riage+0.039
Neg logit contributions
hes-0.056
istered-0.055
as-0.054
uts-0.054
ows-0.051
Token They
Token ID1119
Feature activation+5.12e-03

Pos logit contributions
 angle+0.057
away+0.056
riage+0.052
keeping+0.051
 Mill+0.051
Neg logit contributions
hes-0.086
istered-0.079
uts-0.079
ends-0.077
aj-0.076
Token opened
Token ID4721
Feature activation-4.96e-02

Pos logit contributions
istered+0.010
hes+0.010
uts+0.010
as+0.009
aj+0.009
Neg logit contributions
 angle-0.006
 Sus-0.006
 Mill-0.006
away-0.006
riage-0.006
Token a
Token ID257
Feature activation+3.30e-02

Pos logit contributions
hes+0.077
istered+0.062
ends+0.061
oll+0.059
as+0.058
Neg logit contributions
riage-0.075
 angle-0.074
away-0.073
 basis-0.072
 generation-0.068
Token chair
Token ID5118
Feature activation+4.26e-02

Pos logit contributions
istered+0.051
hes+0.051
as+0.047
uts+0.046
aj+0.045
Neg logit contributions
riage-0.055
 angle-0.054
 Sus-0.053
 Mill-0.053
away-0.053
Token and
Token ID290
Feature activation-1.78e-02

Pos logit contributions
hes+0.082
istered+0.081
as+0.078
uts+0.077
aj+0.075
Neg logit contributions
 angle-0.054
away-0.054
 Sus-0.053
 Mill-0.053
 Joey-0.051
Token climbed
Token ID19952
Feature activation-4.39e-02

Pos logit contributions
 angle+0.026
away+0.025
riage+0.025
 Sus+0.024
keeping+0.024
Neg logit contributions
hes-0.032
istered-0.031
as-0.030
uts-0.029
aj-0.029
Token up
Token ID510
Feature activation-5.48e-02

Pos logit contributions
riage+0.075
 angle+0.073
 zone+0.069
 generation+0.069
 basis+0.068
Neg logit contributions
hes-0.076
ends-0.064
uts-0.061
ows-0.060
br-0.060
Token to
Token ID284
Feature activation-1.19e-01

Pos logit contributions
riage+0.091
 angle+0.076
 zone+0.076
 exp+0.074
roc+0.073
Neg logit contributions
hes-0.098
uts-0.092
br-0.082
 long-0.081
ows-0.081
Token get
Token ID651
Feature activation+5.10e-04

Pos logit contributions
riage+0.141
 angle+0.137
keeping+0.135
 zone+0.125
away+0.125
Neg logit contributions
hes-0.237
istered-0.215
ends-0.205
zel-0.204
 allergic-0.203
Token the
Token ID262
Feature activation+2.82e-03

Pos logit contributions
hes+0.001
istered+0.001
ends+0.001
oll+0.001
uts+0.001
Neg logit contributions
 angle-0.001
riage-0.001
away-0.001
 basis-0.001
 exp-0.001
Token ball
Token ID2613
Feature activation-1.48e-03

Pos logit contributions
as+0.005
istered+0.005
hes+0.005
uts+0.005
keys+0.005
Neg logit contributions
 angle-0.004
 side-0.004
 Sus-0.004
away-0.004
 home-0.004
Token down
Token ID866
Feature activation-5.43e-02

Pos logit contributions
riage+0.002
 angle+0.002
 Mill+0.002
 Sus+0.002
 inspect+0.002
Neg logit contributions
istered-0.003
uts-0.003
hes-0.003
as-0.003
ows-0.003
Token.
Token ID13
Feature activation-5.16e-02

Pos logit contributions
riage+0.099
 inspect+0.092
 angle+0.091
 exp+0.090
 basis+0.090
Neg logit contributions
hes-0.080
uts-0.076
as-0.069
istered-0.068
ows-0.066
Token slide
Token ID10649
Feature activation+1.52e-02

Pos logit contributions
 home+0.005
 angle+0.005
 side+0.005
 investigate+0.005
away+0.005
Neg logit contributions
istered-0.011
uts-0.010
as-0.010
hes-0.010
aj-0.010
Token and
Token ID290
Feature activation-3.72e-02

Pos logit contributions
istered+0.031
hes+0.031
uts+0.029
as+0.029
oll+0.028
Neg logit contributions
 angle-0.018
away-0.018
 Sus-0.018
 Mill-0.017
 Joey-0.017
Token Max
Token ID5436
Feature activation-5.22e-02

Pos logit contributions
 angle+0.047
away+0.046
riage+0.045
 Mill+0.044
keeping+0.043
Neg logit contributions
hes-0.073
istered-0.069
as-0.069
uts-0.067
ends-0.066
Token climbed
Token ID19952
Feature activation-4.63e-02

Pos logit contributions
 Sus+0.062
 angle+0.061
 Mill+0.060
 Joey+0.060
 inspect+0.059
Neg logit contributions
istered-0.105
hes-0.103
uts-0.099
as-0.097
oll-0.095
Token up
Token ID510
Feature activation-5.35e-02

Pos logit contributions
 angle+0.052
riage+0.050
 Mill+0.050
away+0.049
 Sus+0.049
Neg logit contributions
hes-0.098
istered-0.092
as-0.091
uts-0.091
ends-0.089
Token to
Token ID284
Feature activation-1.19e-01

Pos logit contributions
riage+0.074
 angle+0.070
 Mill+0.068
 inspect+0.066
away+0.065
Neg logit contributions
hes-0.099
uts-0.094
as-0.089
istered-0.089
ends-0.088
Token the
Token ID262
Feature activation-1.12e-02

Pos logit contributions
 angle+0.111
 Mill+0.110
keeping+0.110
away+0.107
 Sus+0.103
Neg logit contributions
hes-0.266
istered-0.244
as-0.240
ends-0.237
agging-0.230
Token top
Token ID1353
Feature activation-6.56e-03

Pos logit contributions
 angle+0.013
 side+0.013
 home+0.012
 investigate+0.012
away+0.012
Neg logit contributions
istered-0.024
aj-0.023
as-0.023
hes-0.023
uts-0.023
Token.
Token ID13
Feature activation-4.93e-02

Pos logit contributions
 angle+0.009
away+0.009
 side+0.008
 Sus+0.008
riage+0.008
Neg logit contributions
istered-0.013
hes-0.012
aj-0.012
as-0.012
uts-0.012
Token Tim
Token ID5045
Feature activation-3.59e-02

Pos logit contributions
 angle+0.051
away+0.050
 Sus+0.049
riage+0.048
 Mill+0.048
Neg logit contributions
hes-0.108
istered-0.106
as-0.103
uts-0.103
aj-0.099
Token watched
Token ID7342
Feature activation-4.57e-02

Pos logit contributions
 angle+0.040
 Sus+0.039
 Mill+0.038
 Joey+0.038
away+0.037
Neg logit contributions
istered-0.076
hes-0.074
uts-0.071
as-0.070
oll-0.068
Token and
Token ID290
Feature activation-4.27e-02

Pos logit contributions
 angle+0.068
riage+0.068
 Sus+0.067
 Mill+0.066
away+0.065
Neg logit contributions
hes-0.088
istered-0.086
as-0.084
uts-0.084
aj-0.081
Token managed
Token ID5257
Feature activation-4.52e-02

Pos logit contributions
 angle+0.056
 Sus+0.056
away+0.055
 Mill+0.055
 Joey+0.054
Neg logit contributions
istered-0.080
hes-0.080
as-0.077
uts-0.077
aj-0.074
Token to
Token ID284
Feature activation-9.33e-02

Pos logit contributions
 angle+0.047
riage+0.044
away+0.044
 Sus+0.044
 Mill+0.044
Neg logit contributions
hes-0.098
istered-0.097
as-0.095
uts-0.095
aj-0.091
Token slide
Token ID10649
Feature activation+1.31e-03

Pos logit contributions
 angle+0.106
 Sus+0.103
away+0.103
 Mill+0.102
riage+0.101
Neg logit contributions
hes-0.193
istered-0.191
as-0.185
uts-0.184
aj-0.177
Token down
Token ID866
Feature activation-5.62e-02

Pos logit contributions
hes+0.003
uts+0.003
istered+0.003
as+0.003
ows+0.003
Neg logit contributions
 angle-0.001
riage-0.001
 Mill-0.001
 inspect-0.001
 Sus-0.001
Token to
Token ID284
Feature activation-1.19e-01

Pos logit contributions
riage+0.059
 angle+0.057
 Mill+0.054
away+0.053
 inspect+0.052
Neg logit contributions
hes-0.120
uts-0.114
as-0.114
istered-0.111
ows-0.107
Token the
Token ID262
Feature activation-1.64e-02

Pos logit contributions
 angle+0.134
riage+0.134
keeping+0.128
 Mill+0.127
away+0.124
Neg logit contributions
hes-0.237
istered-0.217
as-0.213
ows-0.207
zel-0.207
Token bottom
Token ID4220
Feature activation-9.73e-03

Pos logit contributions
 angle+0.023
away+0.023
 Sus+0.022
 Mill+0.022
riage+0.022
Neg logit contributions
hes-0.030
istered-0.029
as-0.028
uts-0.028
aj-0.027
Token.
Token ID13
Feature activation-5.02e-02

Pos logit contributions
 angle+0.016
 Sus+0.016
away+0.016
 Joey+0.016
 side+0.015
Neg logit contributions
istered-0.016
hes-0.015
aj-0.014
uts-0.014
as-0.014
Token
Token ID220
Feature activation+8.75e-02

Pos logit contributions
 angle+0.062
away+0.061
riage+0.053
keeping+0.052
 behold+0.049
Neg logit contributions
hes-0.107
uts-0.102
oll-0.096
umbs-0.096
istered-0.095
Token
Token ID198
Feature activation+6.15e-02

Pos logit contributions
istered+0.199
hes+0.191
as+0.187
agging+0.181
uts+0.180
Neg logit contributions
 Sus-0.083
 Joey-0.083
 Mill-0.081
 investigate-0.075
 inspect-0.075
Token he
Token ID339
Feature activation-2.72e-02

Pos logit contributions
 angle+0.009
riage+0.009
away+0.009
 Mill+0.008
 Sus+0.008
Neg logit contributions
hes-0.012
istered-0.012
uts-0.012
as-0.012
oll-0.011
Token decided
Token ID3066
Feature activation-2.55e-02

Pos logit contributions
 angle+0.031
 Sus+0.030
away+0.030
riage+0.030
 Mill+0.030
Neg logit contributions
hes-0.056
istered-0.056
as-0.054
uts-0.053
aj-0.052
Token to
Token ID284
Feature activation-7.33e-02

Pos logit contributions
 angle+0.036
riage+0.035
 Mill+0.033
 Sus+0.033
away+0.032
Neg logit contributions
uts-0.045
hes-0.045
istered-0.042
as-0.042
ends-0.042
Token slide
Token ID10649
Feature activation+1.15e-02

Pos logit contributions
 angle+0.081
away+0.079
riage+0.079
 Sus+0.078
 Mill+0.077
Neg logit contributions
hes-0.151
istered-0.148
as-0.145
uts-0.144
oll-0.138
Token down
Token ID866
Feature activation-5.05e-02

Pos logit contributions
hes+0.025
as+0.024
uts+0.024
istered+0.024
ends+0.023
Neg logit contributions
 angle-0.011
riage-0.011
 Mill-0.011
keeping-0.011
 generation-0.010
Token to
Token ID284
Feature activation-1.19e-01

Pos logit contributions
riage+0.064
 Mill+0.060
 angle+0.059
 Sus+0.057
 Joey+0.057
Neg logit contributions
hes-0.098
as-0.093
uts-0.092
ends-0.088
istered-0.086
Token get
Token ID651
Feature activation+5.58e-03

Pos logit contributions
keeping+0.140
 angle+0.134
 Mill+0.134
riage+0.132
away+0.129
Neg logit contributions
hes-0.235
istered-0.217
as-0.215
ends-0.211
ows-0.210
Token a
Token ID257
Feature activation+4.85e-03

Pos logit contributions
hes+0.011
istered+0.011
as+0.011
uts+0.011
oll+0.010
Neg logit contributions
 angle-0.007
away-0.006
riage-0.006
 Mill-0.006
 Sus-0.006
Token better
Token ID1365
Feature activation-6.96e-02

Pos logit contributions
hes+0.010
istered+0.010
uts+0.010
as+0.010
aj+0.009
Neg logit contributions
 angle-0.006
away-0.006
 Sus-0.006
 Mill-0.005
 Joey-0.005
Token look
Token ID804
Feature activation+1.67e-02

Pos logit contributions
 angle+0.080
away+0.078
 Sus+0.076
 Mill+0.075
 Joey+0.073
Neg logit contributions
hes-0.147
istered-0.145
as-0.141
uts-0.140
aj-0.135
Token.
Token ID13
Feature activation-4.15e-02

Pos logit contributions
istered+0.036
hes+0.035
as+0.033
uts+0.033
aj+0.033
Neg logit contributions
away-0.019
 angle-0.019
 Sus-0.018
riage-0.018
 Mill-0.017
Token to
Token ID284
Feature activation-7.49e-02

Pos logit contributions
istered+0.010
hes+0.010
as+0.010
uts+0.010
aj+0.009
Neg logit contributions
 angle-0.009
 Sus-0.009
 Mill-0.009
away-0.009
 Joey-0.009
Token climb
Token ID12080
Feature activation-5.92e-02

Pos logit contributions
 angle+0.086
 Sus+0.084
away+0.084
 Mill+0.083
riage+0.082
Neg logit contributions
hes-0.154
istered-0.153
as-0.148
uts-0.147
aj-0.141
Token over
Token ID625
Feature activation-6.44e-02

Pos logit contributions
 angle+0.072
riage+0.070
 inspect+0.066
 Mill+0.065
 basis+0.064
Neg logit contributions
hes-0.113
ends-0.103
uts-0.103
istered-0.101
as-0.100
Token a
Token ID257
Feature activation-1.30e-02

Pos logit contributions
riage+0.072
 angle+0.069
 Mill+0.064
 inspect+0.064
away+0.063
Neg logit contributions
hes-0.130
as-0.128
uts-0.126
istered-0.124
ows-0.117
Token fence
Token ID13990
Feature activation-1.93e-02

Pos logit contributions
 angle+0.019
away+0.018
 Sus+0.017
 home+0.017
 Joey+0.017
Neg logit contributions
hes-0.025
istered-0.024
uts-0.023
as-0.023
oll-0.023
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
 angle+0.024
 Sus+0.023
 Mill+0.023
away+0.023
riage+0.023
Neg logit contributions
hes-0.037
istered-0.037
as-0.036
uts-0.036
aj-0.034
Token get
Token ID651
Feature activation+7.86e-03

Pos logit contributions
 angle+0.137
 Sus+0.133
 Mill+0.132
away+0.131
riage+0.130
Neg logit contributions
hes-0.243
istered-0.233
as-0.223
uts-0.219
ows-0.215
Token it
Token ID340
Feature activation-7.21e-03

Pos logit contributions
hes+0.018
istered+0.017
as+0.016
uts+0.016
ends+0.015
Neg logit contributions
 angle-0.008
riage-0.008
away-0.008
 Mill-0.007
 inspect-0.007
Token.
Token ID13
Feature activation-4.43e-02

Pos logit contributions
 angle+0.008
away+0.008
 Sus+0.008
 Mill+0.008
 Joey+0.008
Neg logit contributions
hes-0.015
istered-0.015
as-0.015
uts-0.014
aj-0.014
Token Tim
Token ID5045
Feature activation-3.59e-02

Pos logit contributions
away+0.045
 angle+0.044
riage+0.038
keeping+0.036
 inspect+0.034
Neg logit contributions
hes-0.105
uts-0.102
istered-0.100
ends-0.096
aj-0.096
Tokenmy
Token ID1820
Feature activation-1.76e-02

Pos logit contributions
 angle+0.046
away+0.045
riage+0.045
 Sus+0.044
 Mill+0.044
Neg logit contributions
hes-0.069
istered-0.068
as-0.068
uts-0.065
aj-0.064
Token<|endoftext|>
Token ID50256
Feature activation+3.36e-02

Pos logit contributions
 Joey+0.014
 Sus+0.014
 Mill+0.014
 angle+0.013
 inspect+0.012
Neg logit contributions
istered-0.036
hes-0.035
as-0.035
uts-0.034
aj-0.033
TokenJimmy
Token ID40335
Feature activation-3.75e-02

Pos logit contributions
hes+0.058
as+0.053
uts+0.053
umbs+0.051
istered+0.051
Neg logit contributions
away-0.052
 angle-0.051
riage-0.050
 inspect-0.048
 Mill-0.047
Token wanted
Token ID2227
Feature activation-3.49e-03

Pos logit contributions
 angle+0.061
riage+0.061
away+0.059
keeping+0.056
 Mill+0.055
Neg logit contributions
hes-0.059
as-0.056
istered-0.056
aj-0.054
uts-0.054
Token to
Token ID284
Feature activation-6.82e-02

Pos logit contributions
riage+0.005
 angle+0.004
away+0.004
 Mill+0.004
keeping+0.004
Neg logit contributions
hes-0.007
uts-0.006
istered-0.006
as-0.006
aj-0.006
Token climb
Token ID12080
Feature activation-6.50e-02

Pos logit contributions
 angle+0.079
away+0.079
riage+0.077
 Sus+0.075
 Mill+0.075
Neg logit contributions
hes-0.137
istered-0.132
uts-0.131
as-0.129
aj-0.124
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
 angle+0.079
riage+0.076
away+0.072
keeping+0.071
 Sus+0.069
Neg logit contributions
hes-0.129
uts-0.122
istered-0.121
as-0.118
aj-0.117
Token the
Token ID262
Feature activation-2.68e-02

Pos logit contributions
keeping+0.150
away+0.139
riage+0.136
 angle+0.135
ened+0.130
Neg logit contributions
hes-0.208
 allergic-0.183
ends-0.180
as-0.179
istered-0.177
Token top
Token ID1353
Feature activation-7.81e-03

Pos logit contributions
 angle+0.034
away+0.033
 Sus+0.033
 Mill+0.033
keeping+0.033
Neg logit contributions
hes-0.051
istered-0.050
as-0.048
uts-0.047
aj-0.046
Token of
Token ID286
Feature activation-9.44e-02

Pos logit contributions
 angle+0.014
 Sus+0.013
away+0.013
 Mill+0.013
riage+0.013
Neg logit contributions
hes-0.011
istered-0.011
as-0.011
uts-0.010
aj-0.010
Token the
Token ID262
Feature activation-1.60e-02

Pos logit contributions
away+0.092
riage+0.089
 angle+0.086
keeping+0.085
 Mill+0.079
Neg logit contributions
hes-0.208
istered-0.189
uts-0.185
as-0.183
 allergic-0.180
Token tree
Token ID5509
Feature activation-2.52e-02

Pos logit contributions
 angle+0.021
away+0.021
 Sus+0.021
 Mill+0.021
riage+0.020
Neg logit contributions
hes-0.030
istered-0.029
as-0.028
uts-0.028
aj-0.027
Token and
Token ID290
Feature activation-3.30e-02

Pos logit contributions
riage+0.080
 angle+0.077
 Sus+0.073
away+0.072
 Mill+0.071
Neg logit contributions
hes-0.104
istered-0.098
as-0.095
uts-0.094
oll-0.092
Token his
Token ID465
Feature activation-8.72e-03

Pos logit contributions
 angle+0.035
riage+0.034
away+0.029
ACK+0.027
keeping+0.027
Neg logit contributions
hes-0.073
uts-0.068
istered-0.066
oll-0.065
as-0.064
Token Mom
Token ID11254
Feature activation+7.12e-03

Pos logit contributions
 home+0.011
 side+0.010
 Mill+0.010
 Joey+0.010
 Sus+0.010
Neg logit contributions
hes-0.018
istered-0.017
uts-0.017
aj-0.017
ucker-0.016
Token went
Token ID1816
Feature activation-2.27e-02

Pos logit contributions
hes+0.013
as+0.012
oll+0.011
istered+0.011
umbs+0.011
Neg logit contributions
away-0.011
 angle-0.011
 itself-0.010
riage-0.010
 behold-0.010
Token outside
Token ID2354
Feature activation-6.10e-02

Pos logit contributions
 angle+0.026
 Sus+0.025
riage+0.025
 Mill+0.025
away+0.025
Neg logit contributions
hes-0.047
istered-0.046
as-0.045
uts-0.045
aj-0.043
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
riage+0.093
 angle+0.087
away+0.086
 Mill+0.085
keeping+0.085
Neg logit contributions
hes-0.101
as-0.092
uts-0.092
istered-0.091
agging-0.087
Token play
Token ID711
Feature activation-4.85e-02

Pos logit contributions
 angle+0.135
keeping+0.135
away+0.134
riage+0.129
 itself+0.123
Neg logit contributions
hes-0.233
istered-0.207
aj-0.200
ends-0.197
 allergic-0.196
Token in
Token ID287
Feature activation-8.05e-02

Pos logit contributions
 angle+0.078
riage+0.077
 Mill+0.075
 Sus+0.075
keeping+0.073
Neg logit contributions
hes-0.083
istered-0.077
agging-0.073
as-0.073
uts-0.072
Token the
Token ID262
Feature activation-3.76e-02

Pos logit contributions
riage+0.092
away+0.090
 angle+0.090
keeping+0.083
 generation+0.082
Neg logit contributions
hes-0.163
istered-0.142
as-0.141
aj-0.140
uts-0.138
Token snow
Token ID6729
Feature activation-2.40e-02

Pos logit contributions
 angle+0.051
riage+0.050
 Sus+0.050
away+0.049
 Mill+0.049
Neg logit contributions
hes-0.070
istered-0.068
as-0.066
uts-0.064
aj-0.063
Token.
Token ID13
Feature activation-5.06e-02

Pos logit contributions
 side+0.021
 angle+0.021
keeping+0.020
away+0.020
 home+0.020
Neg logit contributions
istered-0.057
hes-0.054
as-0.054
aj-0.052
oll-0.052
Token cat
Token ID3797
Feature activation-4.28e-03

Pos logit contributions
 angle+0.052
riage+0.051
keeping+0.051
 Sus+0.050
away+0.050
Neg logit contributions
hes-0.069
istered-0.067
as-0.065
uts-0.065
oll-0.063
Token named
Token ID3706
Feature activation-1.03e-02

Pos logit contributions
 angle+0.007
riage+0.007
keeping+0.007
 Sus+0.006
away+0.006
Neg logit contributions
hes-0.007
as-0.007
uts-0.006
istered-0.006
oll-0.006
Token Tim
Token ID5045
Feature activation-4.55e-02

Pos logit contributions
 Mill+0.009
 Sus+0.009
 Joey+0.009
 angle+0.009
away+0.009
Neg logit contributions
istered-0.024
hes-0.024
as-0.023
uts-0.023
aj-0.022
Token went
Token ID1816
Feature activation-2.42e-02

Pos logit contributions
riage+0.067
 angle+0.066
 Sus+0.062
 Mill+0.062
 basis+0.060
Neg logit contributions
hes-0.086
as-0.079
istered-0.073
uts-0.073
uch-0.071
Token outside
Token ID2354
Feature activation-6.26e-02

Pos logit contributions
away+0.025
 angle+0.025
 Mill+0.024
 Sus+0.024
 Joey+0.024
Neg logit contributions
istered-0.053
hes-0.052
as-0.051
uts-0.050
aj-0.049
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
riage+0.100
 Mill+0.093
 angle+0.092
keeping+0.092
 Sus+0.091
Neg logit contributions
hes-0.102
uts-0.091
as-0.090
istered-0.089
agging-0.088
Token play
Token ID711
Feature activation-4.49e-02

Pos logit contributions
keeping+0.152
 angle+0.148
away+0.146
riage+0.135
ened+0.134
Neg logit contributions
hes-0.229
 allergic-0.193
istered-0.191
agging-0.186
uch-0.185
Token on
Token ID319
Feature activation-8.92e-02

Pos logit contributions
 angle+0.069
riage+0.069
 Sus+0.069
 Mill+0.068
keeping+0.066
Neg logit contributions
hes-0.080
istered-0.073
agging-0.071
uts-0.070
as-0.070
Token the
Token ID262
Feature activation-3.38e-02

Pos logit contributions
riage+0.094
 angle+0.093
away+0.089
 Sus+0.088
 Mill+0.086
Neg logit contributions
hes-0.195
uts-0.182
istered-0.182
aj-0.175
as-0.173
Token lawn
Token ID21133
Feature activation-4.03e-02

Pos logit contributions
 angle+0.039
 home+0.038
 side+0.038
away+0.038
 Mill+0.037
Neg logit contributions
istered-0.073
hes-0.070
as-0.069
uts-0.066
oll-0.065
Token.
Token ID13
Feature activation-5.23e-02

Pos logit contributions
 Mill+0.059
away+0.059
 Sus+0.058
 angle+0.058
riage+0.057
Neg logit contributions
hes-0.071
istered-0.066
uts-0.066
as-0.066
ows-0.063
Token would
Token ID561
Feature activation+1.33e-03

Pos logit contributions
 angle+0.046
 Sus+0.045
away+0.045
riage+0.045
 Mill+0.044
Neg logit contributions
hes-0.056
as-0.052
istered-0.052
uts-0.052
oll-0.050
Token often
Token ID1690
Feature activation-3.69e-02

Pos logit contributions
hes+0.002
istered+0.002
as+0.002
uts+0.002
aj+0.002
Neg logit contributions
 angle-0.002
away-0.002
 Sus-0.002
 Mill-0.002
riage-0.002
Token climb
Token ID12080
Feature activation-5.32e-02

Pos logit contributions
 angle+0.038
away+0.036
 Mill+0.036
 Sus+0.036
 inspect+0.035
Neg logit contributions
hes-0.081
istered-0.081
as-0.078
uts-0.077
aj-0.076
Token the
Token ID262
Feature activation-1.48e-02

Pos logit contributions
riage+0.073
 angle+0.073
keeping+0.066
 basis+0.065
 Sus+0.065
Neg logit contributions
hes-0.100
as-0.089
ows-0.088
uts-0.087
istered-0.086
Token ladder
Token ID18002
Feature activation-1.53e-02

Pos logit contributions
 Sus+0.021
riage+0.021
 Joey+0.021
 angle+0.020
 Mill+0.020
Neg logit contributions
hes-0.026
istered-0.025
as-0.025
uts-0.024
ows-0.023
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
 Sus+0.021
 angle+0.021
riage+0.021
 Mill+0.020
 Joey+0.020
Neg logit contributions
hes-0.028
as-0.027
istered-0.026
uts-0.026
ows-0.025
Token peek
Token ID27185
Feature activation-9.02e-03

Pos logit contributions
keeping+0.131
 angle+0.130
riage+0.126
 Sus+0.124
 Mill+0.123
Neg logit contributions
hes-0.242
istered-0.216
as-0.211
 allergic-0.207
oll-0.206
Token over
Token ID625
Feature activation-3.68e-02

Pos logit contributions
 angle+0.012
riage+0.012
 Sus+0.012
 Mill+0.012
 Joey+0.011
Neg logit contributions
hes-0.017
istered-0.017
as-0.016
uts-0.016
oll-0.015
Token the
Token ID262
Feature activation-3.08e-03

Pos logit contributions
 angle+0.044
riage+0.044
keeping+0.044
 Mill+0.041
gent+0.041
Neg logit contributions
hes-0.067
istered-0.065
as-0.065
agging-0.061
ends-0.061
Token fence
Token ID13990
Feature activation-1.95e-02

Pos logit contributions
 Sus+0.004
 angle+0.004
away+0.004
 Mill+0.004
 Joey+0.004
Neg logit contributions
hes-0.006
istered-0.006
as-0.006
uts-0.006
aj-0.005
Token and
Token ID290
Feature activation-3.71e-02

Pos logit contributions
 Sus+0.028
 angle+0.027
riage+0.027
 Mill+0.027
 Joey+0.027
Neg logit contributions
hes-0.034
istered-0.033
as-0.033
uts-0.032
ows-0.031
Token to
Token ID284
Feature activation-1.01e-01

Pos logit contributions
riage+0.074
 generation+0.062
 Acc+0.060
 angle+0.059
 state+0.059
Neg logit contributions
as-0.048
hes-0.048
itt-0.043
 long-0.041
uts-0.041
Token it
Token ID340
Feature activation-1.86e-02

Pos logit contributions
keeping+0.136
riage+0.133
 angle+0.132
 zone+0.127
 Mill+0.126
Neg logit contributions
hes-0.172
 allergic-0.145
ends-0.140
istered-0.140
as-0.138
Token and
Token ID290
Feature activation-3.78e-02

Pos logit contributions
 angle+0.029
riage+0.028
away+0.028
 Sus+0.028
 Mill+0.027
Neg logit contributions
hes-0.031
istered-0.030
uts-0.030
as-0.029
ows-0.028
Token climbed
Token ID19952
Feature activation-4.12e-02

Pos logit contributions
 angle+0.062
riage+0.059
away+0.058
 basis+0.056
keeping+0.056
Neg logit contributions
hes-0.062
istered-0.058
uts-0.055
ends-0.055
as-0.054
Token up
Token ID510
Feature activation-5.00e-02

Pos logit contributions
riage+0.059
 angle+0.058
 generation+0.055
 basis+0.054
away+0.054
Neg logit contributions
hes-0.074
ends-0.069
uts-0.066
ows-0.064
istered-0.064
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
riage+0.086
 angle+0.077
 zone+0.076
 generation+0.071
 basis+0.071
Neg logit contributions
hes-0.077
uts-0.075
ows-0.070
istered-0.068
ends-0.067
Token the
Token ID262
Feature activation-1.41e-02

Pos logit contributions
keeping+0.141
 angle+0.139
riage+0.138
 zone+0.135
 Mill+0.127
Neg logit contributions
hes-0.222
istered-0.196
 allergic-0.191
ends-0.188
ows-0.184
Token top
Token ID1353
Feature activation-5.39e-03

Pos logit contributions
 angle+0.013
 Sus+0.013
away+0.013
 Mill+0.013
riage+0.013
Neg logit contributions
hes-0.032
istered-0.031
as-0.030
uts-0.030
aj-0.029
Token.
Token ID13
Feature activation-4.77e-02

Pos logit contributions
 angle+0.008
away+0.008
 Sus+0.008
 Mill+0.008
 Joey+0.008
Neg logit contributions
istered-0.009
hes-0.009
as-0.009
uts-0.009
aj-0.009
Token From
Token ID3574
Feature activation-2.36e-02

Pos logit contributions
away+0.070
 angle+0.066
riage+0.065
ie+0.060
keeping+0.060
Neg logit contributions
hes-0.090
uts-0.089
ows-0.082
aj-0.080
ends-0.079
Token the
Token ID262
Feature activation+2.91e-03

Pos logit contributions
 angle+0.032
 Sus+0.031
away+0.031
keeping+0.030
riage+0.030
Neg logit contributions
hes-0.044
istered-0.042
as-0.041
uts-0.040
aj-0.040
Token and
Token ID290
Feature activation-2.78e-02

Pos logit contributions
 angle+0.003
 Sus+0.003
away+0.003
 Mill+0.003
riage+0.003
Neg logit contributions
hes-0.004
istered-0.004
as-0.004
uts-0.004
aj-0.004
Token boots
Token ID14412
Feature activation+2.10e-02

Pos logit contributions
 Sus+0.034
 Mill+0.034
 angle+0.032
away+0.032
 Joey+0.032
Neg logit contributions
istered-0.056
hes-0.054
uts-0.054
as-0.053
oll-0.051
Token and
Token ID290
Feature activation-2.53e-02

Pos logit contributions
istered+0.047
hes+0.045
keys+0.043
as+0.043
aj+0.043
Neg logit contributions
 Sus-0.022
 Mill-0.021
away-0.021
 angle-0.021
 home-0.020
Token went
Token ID1816
Feature activation-3.21e-02

Pos logit contributions
 Sus+0.027
 Mill+0.027
 angle+0.027
away+0.026
 Joey+0.026
Neg logit contributions
istered-0.054
hes-0.053
uts-0.052
as-0.051
aj-0.049
Token outside
Token ID2354
Feature activation-6.67e-02

Pos logit contributions
 angle+0.033
 Sus+0.032
away+0.032
 Mill+0.032
riage+0.031
Neg logit contributions
hes-0.070
istered-0.069
as-0.067
uts-0.067
aj-0.064
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
riage+0.099
 angle+0.094
away+0.093
 Joey+0.090
 behold+0.090
Neg logit contributions
hes-0.114
uts-0.108
istered-0.107
as-0.106
ends-0.101
Token jump
Token ID4391
Feature activation-2.23e-02

Pos logit contributions
 angle+0.121
keeping+0.117
away+0.117
riage+0.111
 Mill+0.106
Neg logit contributions
hes-0.261
istered-0.241
as-0.233
aj-0.230
agging-0.226
Token in
Token ID287
Feature activation-6.77e-02

Pos logit contributions
riage+0.033
 angle+0.033
 Mill+0.031
 Sus+0.031
keeping+0.030
Neg logit contributions
hes-0.040
istered-0.038
as-0.037
uts-0.037
ows-0.036
Token p
Token ID279
Feature activation-1.48e-02

Pos logit contributions
riage+0.095
away+0.083
 angle+0.083
 generation+0.083
keeping+0.082
Neg logit contributions
hes-0.115
istered-0.107
as-0.105
 long-0.099
aj-0.098
Tokenudd
Token ID4185
Feature activation-5.86e-02

Pos logit contributions
away+0.034
 Sus+0.033
 angle+0.033
 home+0.032
riage+0.032
Neg logit contributions
hes-0.013
istered-0.013
keys-0.013
uts-0.012
oll-0.011
Tokenles
Token ID829
Feature activation-2.09e-02

Pos logit contributions
 angle+0.105
away+0.104
 Sus+0.103
 Mill+0.102
 Joey+0.101
Neg logit contributions
hes-0.083
istered-0.082
as-0.078
uts-0.077
aj-0.073
Token to
Token ID284
Feature activation-6.40e-02

Pos logit contributions
 angle+0.039
riage+0.038
 Mill+0.036
 behold+0.036
 Sus+0.035
Neg logit contributions
uts-0.041
hes-0.039
 not-0.037
istered-0.037
ends-0.037
Token jump
Token ID4391
Feature activation-8.15e-03

Pos logit contributions
 angle+0.072
away+0.070
riage+0.069
 Sus+0.069
 Mill+0.068
Neg logit contributions
hes-0.133
istered-0.131
uts-0.127
as-0.127
aj-0.122
Token over
Token ID625
Feature activation-5.11e-02

Pos logit contributions
riage+0.011
 angle+0.011
 Mill+0.011
keeping+0.011
 Sus+0.010
Neg logit contributions
hes-0.015
istered-0.014
uts-0.014
as-0.014
zel-0.013
Token a
Token ID257
Feature activation-1.65e-02

Pos logit contributions
riage+0.056
 angle+0.052
 Mill+0.050
 Sus+0.048
away+0.047
Neg logit contributions
hes-0.107
as-0.103
uts-0.103
istered-0.102
agging-0.097
Token fence
Token ID13990
Feature activation-1.77e-02

Pos logit contributions
 angle+0.021
away+0.020
 Mill+0.019
 Sus+0.019
 Joey+0.019
Neg logit contributions
hes-0.033
istered-0.033
as-0.032
uts-0.032
oll-0.030
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
 angle+0.023
riage+0.023
 Mill+0.023
 Sus+0.023
away+0.022
Neg logit contributions
hes-0.033
istered-0.033
as-0.032
uts-0.032
oll-0.030
Token get
Token ID651
Feature activation+6.35e-03

Pos logit contributions
riage+0.153
 angle+0.152
keeping+0.151
 Mill+0.151
 Sus+0.149
Neg logit contributions
hes-0.223
istered-0.204
as-0.195
uts-0.192
zel-0.191
Token it
Token ID340
Feature activation-6.80e-03

Pos logit contributions
hes+0.013
istered+0.013
as+0.013
uts+0.013
aj+0.012
Neg logit contributions
 angle-0.007
away-0.007
 Sus-0.007
 Mill-0.007
riage-0.007
Token.
Token ID13
Feature activation-4.09e-02

Pos logit contributions
 angle+0.008
away+0.008
 Sus+0.008
 Mill+0.008
 Joey+0.008
Neg logit contributions
hes-0.014
istered-0.014
as-0.013
uts-0.013
aj-0.013
Token But
Token ID887
Feature activation-5.03e-02

Pos logit contributions
riage+0.070
 angle+0.069
away+0.069
keeping+0.065
 generation+0.063
Neg logit contributions
hes-0.064
uts-0.063
aj-0.060
istered-0.058
ucker-0.057
Token as
Token ID355
Feature activation+3.03e-02

Pos logit contributions
keeping+0.095
 angle+0.086
riage+0.083
away+0.081
 inspect+0.079
Neg logit contributions
 not-0.070
hes-0.070
ittens-0.068
ends-0.068
zel-0.067
Token to
Token ID284
Feature activation-6.95e-02

Pos logit contributions
 angle+0.036
riage+0.034
away+0.032
 Joey+0.032
 behold+0.032
Neg logit contributions
uts-0.044
hes-0.043
istered-0.041
aj-0.039
ends-0.039
Token climb
Token ID12080
Feature activation-6.66e-02

Pos logit contributions
 angle+0.080
away+0.078
riage+0.077
 Sus+0.076
 Mill+0.075
Neg logit contributions
hes-0.141
istered-0.138
as-0.134
uts-0.133
oll-0.128
Token onto
Token ID4291
Feature activation-8.24e-02

Pos logit contributions
 angle+0.092
riage+0.085
 exp+0.081
 inspect+0.080
away+0.079
Neg logit contributions
hes-0.127
uts-0.114
istered-0.114
as-0.109
ends-0.109
Token the
Token ID262
Feature activation-2.50e-02

Pos logit contributions
 angle+0.081
riage+0.078
away+0.076
 Joey+0.071
 inspect+0.070
Neg logit contributions
hes-0.185
istered-0.176
uts-0.172
as-0.166
zel-0.164
Token counter
Token ID3753
Feature activation-1.96e-02

Pos logit contributions
 angle+0.034
riage+0.034
 Sus+0.034
away+0.033
 Mill+0.033
Neg logit contributions
istered-0.045
hes-0.045
uts-0.042
as-0.042
oll-0.041
Token to
Token ID284
Feature activation-1.18e-01

Pos logit contributions
 angle+0.024
 Sus+0.024
away+0.024
 Mill+0.024
riage+0.023
Neg logit contributions
hes-0.038
istered-0.038
as-0.037
uts-0.036
aj-0.035
Token reach
Token ID3151
Feature activation-6.11e-02

Pos logit contributions
 angle+0.143
riage+0.134
away+0.134
 Mill+0.133
keeping+0.133
Neg logit contributions
hes-0.233
istered-0.217
 allergic-0.202
oll-0.202
ends-0.199
Token the
Token ID262
Feature activation-8.85e-03

Pos logit contributions
 angle+0.065
riage+0.059
away+0.058
keeping+0.056
 inspect+0.056
Neg logit contributions
hes-0.134
istered-0.129
uts-0.125
as-0.124
oll-0.122
Token top
Token ID1353
Feature activation-1.03e-02

Pos logit contributions
 angle+0.011
 Sus+0.011
away+0.011
 Mill+0.011
riage+0.010
Neg logit contributions
hes-0.017
istered-0.017
as-0.017
uts-0.017
aj-0.016
Token shelf
Token ID18316
Feature activation-7.44e-03

Pos logit contributions
 angle+0.015
 Sus+0.014
away+0.014
 Mill+0.014
riage+0.014
Neg logit contributions
istered-0.018
hes-0.018
as-0.017
uts-0.017
aj-0.017
Token of
Token ID286
Feature activation-8.11e-02

Pos logit contributions
 angle+0.009
 Sus+0.009
away+0.009
 Mill+0.008
riage+0.008
Neg logit contributions
hes-0.015
istered-0.015
as-0.014
uts-0.014
aj-0.014
Token sick
Token ID6639
Feature activation-2.30e-02

Pos logit contributions
hes+0.073
istered+0.072
as+0.069
uts+0.069
aj+0.066
Neg logit contributions
 angle-0.053
 Sus-0.052
away-0.052
 Mill-0.052
riage-0.051
Token person
Token ID1048
Feature activation+1.68e-02

Pos logit contributions
 angle+0.029
 Sus+0.029
away+0.028
 Mill+0.028
riage+0.028
Neg logit contributions
istered-0.044
hes-0.044
as-0.042
uts-0.042
aj-0.041
Token.
Token ID13
Feature activation-5.56e-03

Pos logit contributions
hes+0.034
istered+0.034
as+0.032
uts+0.032
aj+0.031
Neg logit contributions
 angle-0.020
away-0.020
 Sus-0.020
 Mill-0.019
riage-0.019
Tokenâ
Token ID22940
Feature activation+2.01e-02

Pos logit contributions
away+0.007
 angle+0.007
riage+0.006
 inspect+0.006
keeping+0.006
Neg logit contributions
hes-0.011
uts-0.011
as-0.010
istered-0.010
aj-0.010
Token€
Token ID26391
Feature activation+3.36e-02

Pos logit contributions
hes+0.037
uch+0.034
agging+0.034
zel+0.034
ows+0.033
Neg logit contributions
 angle-0.027
 inspect-0.026
 Sus-0.026
 Mill-0.026
 Where-0.026
Token
Token ID220
Feature activation+1.37e-01

Pos logit contributions
istered+0.078
hes+0.078
oll+0.077
agging+0.076
as+0.073
Neg logit contributions
 Sus-0.029
 Mill-0.028
keeping-0.027
 Joey-0.026
riage-0.025
Token
Token ID198
Feature activation+9.24e-02

Pos logit contributions
istered+0.319
hes+0.303
agging+0.292
as+0.292
oll+0.289
Neg logit contributions
 Sus-0.127
 Mill-0.123
 Joey-0.122
 itself-0.109
riage-0.109
Token
Token ID198
Feature activation+7.38e-02

Pos logit contributions
hes+0.184
istered+0.181
as+0.176
uts+0.175
aj+0.168
Neg logit contributions
 angle-0.112
away-0.109
 Sus-0.108
riage-0.108
 Mill-0.107
TokenKevin
Token ID23865
Feature activation-5.54e-03

Pos logit contributions
hes+0.129
istered+0.127
as+0.122
uts+0.121
aj+0.116
Neg logit contributions
 angle-0.108
away-0.106
 Sus-0.106
 Mill-0.105
riage-0.104
Token was
Token ID373
Feature activation+7.98e-02

Pos logit contributions
 angle+0.007
away+0.006
riage+0.006
 Sus+0.006
 Mill+0.006
Neg logit contributions
hes-0.011
istered-0.011
as-0.011
uts-0.010
aj-0.010
Token so
Token ID523
Feature activation+6.13e-02

Pos logit contributions
istered+0.137
uts+0.133
as+0.132
hes+0.126
aj+0.126
Neg logit contributions
 Sus-0.121
 Mill-0.119
 angle-0.117
 Joey-0.117
 home-0.116
Token vegetables
Token ID13701
Feature activation+4.81e-02

Pos logit contributions
istered+0.036
hes+0.035
as+0.034
uts+0.034
aj+0.033
Neg logit contributions
 angle-0.023
 Sus-0.023
 Mill-0.023
 Joey-0.023
away-0.022
Token in
Token ID287
Feature activation-2.70e-02

Pos logit contributions
istered+0.107
hes+0.104
as+0.101
aj+0.099
uts+0.098
Neg logit contributions
 Joey-0.048
 Mill-0.048
away-0.048
 Sus-0.048
 angle-0.047
Token the
Token ID262
Feature activation+1.23e-02

Pos logit contributions
riage+0.020
 angle+0.020
away+0.020
 Sus+0.019
 Mill+0.019
Neg logit contributions
hes-0.065
istered-0.064
as-0.062
uts-0.061
aj-0.059
Token back
Token ID736
Feature activation-3.86e-02

Pos logit contributions
hes+0.023
istered+0.023
as+0.022
aj+0.021
uts+0.021
Neg logit contributions
 angle-0.016
 Sus-0.016
away-0.016
 Mill-0.016
riage-0.015
Token.
Token ID13
Feature activation-3.48e-03

Pos logit contributions
 angle+0.052
away+0.051
 Sus+0.051
 Mill+0.051
 Joey+0.050
Neg logit contributions
istered-0.073
hes-0.072
uts-0.069
as-0.068
zel-0.066
Token
Token ID220
Feature activation+1.35e-01

Pos logit contributions
away+0.004
 angle+0.004
riage+0.004
 inspect+0.004
keeping+0.004
Neg logit contributions
hes-0.007
uts-0.007
istered-0.007
as-0.007
aj-0.007
Token
Token ID198
Feature activation+8.76e-02

Pos logit contributions
istered+0.316
hes+0.298
as+0.288
agging+0.287
oll+0.286
Neg logit contributions
 Joey-0.123
 Sus-0.123
 Mill-0.122
 angle-0.111
 itself-0.111
Token
Token ID198
Feature activation+7.22e-02

Pos logit contributions
hes+0.162
as+0.156
istered+0.152
uts+0.152
aj+0.146
Neg logit contributions
 angle-0.121
away-0.120
riage-0.120
 Sus-0.113
 Mill-0.113
TokenOne
Token ID3198
Feature activation+5.44e-02

Pos logit contributions
hes+0.116
istered+0.112
as+0.112
uts+0.109
zel+0.103
Neg logit contributions
 angle-0.117
riage-0.114
away-0.114
 Sus-0.113
 Mill-0.112
Token day
Token ID1110
Feature activation+2.83e-02

Pos logit contributions
hes+0.138
istered+0.134
as+0.134
uts+0.130
oll+0.129
Neg logit contributions
 angle-0.034
keeping-0.033
riage-0.033
 Sus-0.033
away-0.033
Token,
Token ID11
Feature activation+6.42e-03

Pos logit contributions
hes+0.058
istered+0.057
as+0.056
uts+0.056
oll+0.053
Neg logit contributions
 angle-0.033
riage-0.031
away-0.031
keeping-0.030
 inspect-0.029
Token picking
Token ID10868
Feature activation-1.76e-02

Pos logit contributions
istered+0.108
hes+0.107
as+0.104
uts+0.102
oll+0.100
Neg logit contributions
 Sus-0.064
 Mill-0.063
 angle-0.063
 Joey-0.062
 inspect-0.060
Token up
Token ID510
Feature activation-2.67e-03

Pos logit contributions
 angle+0.019
riage+0.019
 Mill+0.018
 Sus+0.018
away+0.017
Neg logit contributions
hes-0.038
istered-0.036
as-0.036
uts-0.035
oll-0.034
Token the
Token ID262
Feature activation+2.87e-02

Pos logit contributions
 angle+0.002
 Sus+0.002
away+0.002
 Mill+0.002
riage+0.002
Neg logit contributions
hes-0.007
istered-0.006
as-0.006
uts-0.006
aj-0.006
Token phone
Token ID3072
Feature activation+1.31e-02

Pos logit contributions
hes+0.053
istered+0.053
uts+0.051
as+0.050
aj+0.048
Neg logit contributions
 angle-0.040
away-0.038
 Sus-0.038
 Mill-0.038
 Joey-0.037
Token.
Token ID13
Feature activation-1.77e-02

Pos logit contributions
hes+0.026
istered+0.026
as+0.025
uts+0.025
aj+0.024
Neg logit contributions
 angle-0.016
 Sus-0.016
away-0.016
 Mill-0.015
riage-0.015
Token
Token ID220
Feature activation+1.29e-01

Pos logit contributions
away+0.024
 angle+0.024
riage+0.023
keeping+0.022
 inspect+0.021
Neg logit contributions
hes-0.034
uts-0.033
istered-0.032
aj-0.031
as-0.031
Token
Token ID198
Feature activation+8.98e-02

Pos logit contributions
istered+0.305
hes+0.288
oll+0.280
agging+0.279
as+0.279
Neg logit contributions
 Sus-0.110
 Mill-0.108
 Joey-0.108
 itself-0.103
 angle-0.096
Token
Token ID198
Feature activation+7.44e-02

Pos logit contributions
hes+0.169
as+0.161
istered+0.160
uts+0.160
aj+0.152
Neg logit contributions
 angle-0.123
riage-0.119
away-0.118
 Sus-0.115
 Mill-0.114
Token"
Token ID1
Feature activation+3.03e-02

Pos logit contributions
hes+0.123
istered+0.122
as+0.117
uts+0.115
aj+0.111
Neg logit contributions
 angle-0.115
away-0.113
 Sus-0.113
 Mill-0.112
riage-0.111
TokenHi
Token ID17250
Feature activation-1.55e-02

Pos logit contributions
istered+0.072
as+0.069
hes+0.069
aj+0.069
ows+0.068
Neg logit contributions
 Sus-0.025
 angle-0.025
 Mill-0.024
away-0.024
riage-0.024
Token,
Token ID11
Feature activation+1.00e-02

Pos logit contributions
 angle+0.020
away+0.019
keeping+0.019
 inspect+0.018
 generation+0.018
Neg logit contributions
istered-0.029
hes-0.029
as-0.029
aj-0.028
uts-0.028
Token and
Token ID290
Feature activation+1.70e-02

Pos logit contributions
hes+0.041
istered+0.041
uts+0.039
as+0.039
aj+0.038
Neg logit contributions
 angle-0.025
away-0.025
 Sus-0.025
 Mill-0.024
 Joey-0.024
Token healthy
Token ID5448
Feature activation-8.01e-03

Pos logit contributions
istered+0.030
hes+0.030
uts+0.029
as+0.029
oll+0.028
Neg logit contributions
 Sus-0.024
 Mill-0.024
 angle-0.024
 Joey-0.023
riage-0.023
Token once
Token ID1752
Feature activation+4.15e-02

Pos logit contributions
 angle+0.009
away+0.009
 Sus+0.009
 Mill+0.009
 Joey+0.009
Neg logit contributions
hes-0.016
istered-0.016
as-0.016
uts-0.015
aj-0.015
Token more
Token ID517
Feature activation+4.75e-02

Pos logit contributions
hes+0.086
istered+0.086
as+0.082
uts+0.082
aj+0.079
Neg logit contributions
 angle-0.047
 Sus-0.046
away-0.046
 Mill-0.046
 Joey-0.045
Token.
Token ID13
Feature activation-6.21e-03

Pos logit contributions
hes+0.096
istered+0.094
as+0.089
uts+0.088
aj+0.087
Neg logit contributions
 Sus-0.059
 Mill-0.058
 angle-0.058
 Joey-0.057
away-0.057
Token
Token ID220
Feature activation+1.33e-01

Pos logit contributions
away+0.008
 angle+0.008
riage+0.007
 inspect+0.007
keeping+0.007
Neg logit contributions
hes-0.012
uts-0.012
as-0.012
istered-0.012
aj-0.011
Token
Token ID198
Feature activation+9.00e-02

Pos logit contributions
istered+0.300
hes+0.288
as+0.278
uts+0.274
oll+0.273
Neg logit contributions
 Sus-0.129
 Mill-0.126
 Joey-0.125
 angle-0.118
 itself-0.116
Token
Token ID198
Feature activation+7.36e-02

Pos logit contributions
hes+0.164
as+0.157
uts+0.152
istered+0.151
aj+0.146
Neg logit contributions
away-0.128
riage-0.128
 angle-0.128
 inspect-0.119
 Sus-0.119
TokenMax
Token ID11518
Feature activation-1.86e-02

Pos logit contributions
hes+0.120
as+0.116
istered+0.114
uts+0.112
zel+0.107
Neg logit contributions
 angle-0.117
riage-0.115
away-0.115
 Sus-0.113
 Mill-0.112
Token's
Token ID338
Feature activation+4.72e-02

Pos logit contributions
 angle+0.026
away+0.025
riage+0.025
 Sus+0.024
 Mill+0.024
Neg logit contributions
hes-0.034
istered-0.033
as-0.032
uts-0.032
aj-0.031
Token owner
Token ID4870
Feature activation+3.40e-02

Pos logit contributions
hes+0.087
istered+0.086
as+0.083
uts+0.082
aj+0.079
Neg logit contributions
 angle-0.064
 Sus-0.062
away-0.062
 Mill-0.062
riage-0.061
Token was
Token ID373
Feature activation+9.24e-02

Pos logit contributions
istered+0.133
hes+0.130
as+0.128
uts+0.128
agging+0.124
Neg logit contributions
 angle-0.065
 Sus-0.065
 Mill-0.064
away-0.063
riage-0.062
Token bright
Token ID6016
Feature activation+5.33e-02

Pos logit contributions
istered+0.164
uts+0.159
as+0.157
hes+0.152
umbs+0.148
Neg logit contributions
 Sus-0.140
 Mill-0.138
 Joey-0.137
 home-0.131
 itself-0.131
Token and
Token ID290
Feature activation+3.48e-02

Pos logit contributions
hes+0.107
istered+0.107
as+0.103
uts+0.102
zel+0.100
Neg logit contributions
 Sus-0.067
 angle-0.067
 Mill-0.066
away-0.065
 Joey-0.064
Token clear
Token ID1598
Feature activation+1.54e-02

Pos logit contributions
istered+0.052
as+0.050
uts+0.050
hes+0.050
agging+0.048
Neg logit contributions
 Sus-0.062
 Mill-0.062
 Joey-0.060
 angle-0.059
riage-0.058
Token.
Token ID13
Feature activation-5.34e-03

Pos logit contributions
hes+0.030
istered+0.030
as+0.029
uts+0.029
aj+0.028
Neg logit contributions
 angle-0.019
 Sus-0.019
away-0.019
 Mill-0.019
riage-0.018
Token
Token ID220
Feature activation+1.30e-01

Pos logit contributions
 angle+0.006
away+0.006
riage+0.005
 inspect+0.005
keeping+0.005
Neg logit contributions
hes-0.011
istered-0.011
as-0.011
uts-0.011
aj-0.011
Token
Token ID198
Feature activation+9.16e-02

Pos logit contributions
istered+0.294
hes+0.280
as+0.269
agging+0.269
oll+0.268
Neg logit contributions
 Sus-0.124
 Mill-0.122
 Joey-0.120
 angle-0.116
 itself-0.114
Token
Token ID198
Feature activation+7.50e-02

Pos logit contributions
hes+0.167
as+0.162
uts+0.157
istered+0.157
aj+0.151
Neg logit contributions
 angle-0.129
riage-0.127
away-0.126
 inspect-0.120
 Mill-0.119
TokenJimmy
Token ID40335
Feature activation-4.25e-03

Pos logit contributions
hes+0.123
istered+0.120
as+0.120
uts+0.117
zel+0.111
Neg logit contributions
 angle-0.117
riage-0.114
away-0.113
 Sus-0.113
 Mill-0.113
Token shouted
Token ID17293
Feature activation+4.32e-02

Pos logit contributions
 angle+0.005
 Sus+0.005
away+0.005
 Mill+0.005
riage+0.005
Neg logit contributions
hes-0.008
istered-0.008
as-0.008
uts-0.008
aj-0.008
Token in
Token ID287
Feature activation-1.54e-02

Pos logit contributions
hes+0.085
istered+0.084
uts+0.082
as+0.081
aj+0.078
Neg logit contributions
 angle-0.055
riage-0.052
away-0.051
keeping-0.051
 Mill-0.051
Token,
Token ID11
Feature activation+3.18e-02

Pos logit contributions
hes+0.031
uts+0.031
istered+0.030
as+0.030
oll+0.029
Neg logit contributions
 angle-0.023
away-0.021
keeping-0.021
 generation-0.021
riage-0.021
Token "
Token ID366
Feature activation+1.34e-04

Pos logit contributions
hes+0.064
istered+0.060
uts+0.060
as+0.059
aj+0.057
Neg logit contributions
 angle-0.041
away-0.041
keeping-0.038
 Sus-0.038
 Mill-0.038
TokenYou
Token ID1639
Feature activation+5.61e-03

Pos logit contributions
istered+0.000
hes+0.000
as+0.000
uts+0.000
aj+0.000
Neg logit contributions
 angle-0.000
 Sus-0.000
riage-0.000
away-0.000
 Mill-0.000
Token are
Token ID389
Feature activation+7.73e-02

Pos logit contributions
istered+0.009
aj+0.009
uts+0.009
agging+0.009
hes+0.009
Neg logit contributions
 Sus-0.008
riage-0.008
 Joey-0.008
 investigate-0.008
 Mill-0.008
Token good
Token ID922
Feature activation-1.96e-02

Pos logit contributions
istered+0.150
uts+0.146
aj+0.143
as+0.142
hes+0.137
Neg logit contributions
riage-0.096
 Sus-0.096
 Joey-0.095
 Mill-0.095
 home-0.093
Token twins
Token ID20345
Feature activation+5.12e-02

Pos logit contributions
 Sus+0.030
away+0.029
 angle+0.029
 home+0.028
 Joey+0.028
Neg logit contributions
hes-0.033
istered-0.033
as-0.031
uts-0.031
zel-0.030
Token.
Token ID13
Feature activation-3.57e-02

Pos logit contributions
hes+0.111
istered+0.111
as+0.106
uts+0.106
aj+0.106
Neg logit contributions
away-0.049
riage-0.049
 Sus-0.049
 Mill-0.046
 angle-0.046
Token I
Token ID314
Feature activation+2.87e-04

Pos logit contributions
 angle+0.054
away+0.053
 inspect+0.049
 Mill+0.049
 Sus+0.049
Neg logit contributions
hes-0.063
istered-0.061
uts-0.059
as-0.059
oll-0.057
Token am
Token ID716
Feature activation+6.32e-02

Pos logit contributions
hes+0.001
istered+0.001
uts+0.000
as+0.000
aj+0.000
Neg logit contributions
 angle-0.000
 Sus-0.000
away-0.000
riage-0.000
 Mill-0.000
Token glad
Token ID9675
Feature activation-2.54e-02

Pos logit contributions
istered+0.102
as+0.098
uts+0.098
cases+0.097
aj+0.096
Neg logit contributions
 home-0.106
 Sus-0.103
 Joey-0.100
 Mill-0.100
 Sally-0.097
Token you
Token ID345
Feature activation+1.03e-02

Pos logit contributions
 angle+0.046
away+0.045
 Sus+0.045
 Mill+0.044
riage+0.044
Neg logit contributions
hes-0.036
istered-0.035
as-0.034
uts-0.033
aj-0.031
Token I
Token ID314
Feature activation-9.27e-03

Pos logit contributions
istered+0.011
hes+0.011
as+0.010
aj+0.010
uts+0.010
Neg logit contributions
 Sus-0.009
 Mill-0.009
 Joey-0.009
 angle-0.009
away-0.009
Token'm
Token ID1101
Feature activation+4.77e-02

Pos logit contributions
 Sus+0.010
 investigate+0.010
 Joey+0.009
 inspect+0.009
 Mill+0.009
Neg logit contributions
hes-0.020
agging-0.019
istered-0.019
keys-0.019
umbs-0.019
Token OK
Token ID7477
Feature activation-4.83e-02

Pos logit contributions
as+0.087
istered+0.086
uts+0.084
umbs+0.082
hes+0.080
Neg logit contributions
 Sus-0.079
 Joey-0.077
 Mill-0.077
 home-0.074
 Bobby-0.072
Token.
Token ID13
Feature activation-4.33e-02

Pos logit contributions
 angle+0.068
away+0.067
riage+0.065
 Mill+0.064
 Sus+0.064
Neg logit contributions
hes-0.085
istered-0.084
as-0.083
uts-0.082
aj-0.078
Token It
Token ID632
Feature activation+1.45e-02

Pos logit contributions
 angle+0.053
away+0.052
 Sus+0.050
 Mill+0.050
riage+0.049
Neg logit contributions
hes-0.086
istered-0.085
as-0.082
uts-0.082
aj-0.079
Token was
Token ID373
Feature activation+5.06e-02

Pos logit contributions
istered+0.032
agging+0.032
ittens+0.032
uts+0.030
aj+0.030
Neg logit contributions
 Mill-0.015
 investigate-0.015
 Joey-0.015
 Sus-0.015
 Bobby-0.014
Token a
Token ID257
Feature activation+4.46e-02

Pos logit contributions
istered+0.091
uts+0.090
as+0.089
hes+0.088
ows+0.083
Neg logit contributions
 Sus-0.076
 Mill-0.075
 Joey-0.075
 angle-0.071
 home-0.070
Token bit
Token ID1643
Feature activation+2.85e-02

Pos logit contributions
hes+0.063
istered+0.062
as+0.060
uts+0.059
aj+0.056
Neg logit contributions
 angle-0.080
away-0.078
 Sus-0.078
 Mill-0.078
riage-0.077
Token of
Token ID286
Feature activation-4.17e-02

Pos logit contributions
hes+0.042
istered+0.040
uts+0.040
as+0.038
aj+0.036
Neg logit contributions
 angle-0.051
away-0.050
 Sus-0.050
 Joey-0.050
 Mill-0.049
Token a
Token ID257
Feature activation+1.90e-02

Pos logit contributions
 angle+0.060
 Sus+0.059
away+0.059
 Mill+0.059
 Joey+0.058
Neg logit contributions
hes-0.074
istered-0.073
as-0.071
uts-0.070
aj-0.067
Token silly
Token ID14397
Feature activation-5.49e-02

Pos logit contributions
hes+0.032
istered+0.031
as+0.030
uts+0.030
aj+0.029
Neg logit contributions
 angle-0.029
 Sus-0.028
away-0.028
 Mill-0.028
riage-0.028
Token happily
Token ID18177
Feature activation+1.73e-02

Pos logit contributions
hes+0.104
istered+0.102
as+0.100
uts+0.098
aj+0.094
Neg logit contributions
 angle-0.046
riage-0.043
 Sus-0.043
 Mill-0.043
away-0.043
Token ever
Token ID1683
Feature activation+1.42e-02

Pos logit contributions
hes+0.034
istered+0.033
aj+0.031
uts+0.031
as+0.031
Neg logit contributions
away-0.022
 Sus-0.022
 home-0.022
keeping-0.022
 Mill-0.022
Token after
Token ID706
Feature activation-3.95e-02

Pos logit contributions
hes+0.037
istered+0.036
as+0.035
uts+0.035
aj+0.034
Neg logit contributions
 angle-0.009
away-0.009
 Sus-0.009
 Mill-0.009
riage-0.009
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
 angle+0.050
away+0.048
 Sus+0.048
 Mill+0.048
riage+0.047
Neg logit contributions
hes-0.077
istered-0.076
as-0.074
uts-0.073
aj-0.070
Token<|endoftext|>
Token ID50256
Feature activation+5.88e-02

Pos logit contributions
 angle+0.038
away+0.037
 inspect+0.035
riage+0.034
keeping+0.033
Neg logit contributions
hes-0.038
uts-0.037
istered-0.035
as-0.035
oll-0.034
TokenOnce
Token ID7454
Feature activation+4.91e-02

Pos logit contributions
hes+0.109
as+0.101
istered+0.101
uts+0.100
ows+0.094
Neg logit contributions
 angle-0.082
away-0.081
riage-0.080
 Mill-0.079
 Sus-0.079
Token upon
Token ID2402
Feature activation+6.27e-02

Pos logit contributions
hes+0.079
istered+0.079
uts+0.074
as+0.072
oll+0.069
Neg logit contributions
 angle-0.080
 Sus-0.079
 Mill-0.078
 Joey-0.077
away-0.077
Token a
Token ID257
Feature activation+1.52e-02

Pos logit contributions
as+0.146
istered+0.146
aj+0.146
ucker+0.144
uts+0.142
Neg logit contributions
 itself-0.058
 investigate-0.055
 him-0.053
 his-0.051
 Sus-0.050
Token time
Token ID640
Feature activation+6.63e-04

Pos logit contributions
uts+0.040
plets+0.038
hes+0.038
acas+0.038
cases+0.037
Neg logit contributions
 angle-0.011
away-0.010
 day-0.010
 moment-0.010
 home-0.010
Token,
Token ID11
Feature activation-4.25e-03

Pos logit contributions
hes+0.002
istered+0.002
as+0.002
uts+0.002
oll+0.002
Neg logit contributions
 Sus-0.000
 Mill-0.000
 Joey-0.000
 angle-0.000
 Bobby-0.000
Token there
Token ID612
Feature activation+5.18e-03

Pos logit contributions
 angle+0.006
 Sus+0.006
away+0.006
 Mill+0.006
riage+0.006
Neg logit contributions
hes-0.008
istered-0.008
as-0.007
uts-0.007
aj-0.007
Token the
Token ID262
Feature activation+4.57e-02

Pos logit contributions
 Sus+0.008
 Mill+0.008
 Joey+0.008
 itself+0.008
 angle+0.008
Neg logit contributions
istered-0.018
as-0.018
aj-0.017
uts-0.017
hes-0.017
Token bee
Token ID20697
Feature activation+3.57e-02

Pos logit contributions
istered+0.096
uts+0.095
hes+0.094
cases+0.091
oll+0.090
Neg logit contributions
 angle-0.056
 home-0.054
 side-0.051
 Mill-0.049
away-0.049
Token on
Token ID319
Feature activation-4.99e-02

Pos logit contributions
istered+0.069
aj+0.067
hes+0.066
as+0.066
uts+0.065
Neg logit contributions
 side-0.043
 itself-0.043
 angle-0.043
away-0.042
 Mill-0.042
Token their
Token ID511
Feature activation-2.55e-02

Pos logit contributions
 angle+0.045
riage+0.044
 Sus+0.043
away+0.043
 Mill+0.042
Neg logit contributions
hes-0.115
istered-0.112
uts-0.109
as-0.109
aj-0.104
Token flower
Token ID15061
Feature activation+2.30e-02

Pos logit contributions
 angle+0.038
away+0.037
 Mill+0.036
 Sus+0.036
 Joey+0.035
Neg logit contributions
hes-0.045
istered-0.044
as-0.042
uts-0.042
oll-0.040
Token necklace
Token ID38987
Feature activation+5.02e-02

Pos logit contributions
hes+0.046
istered+0.046
as+0.044
uts+0.043
aj+0.043
Neg logit contributions
 angle-0.027
away-0.026
 Mill-0.026
 Sus-0.026
keeping-0.025
Token.
Token ID13
Feature activation-2.15e-02

Pos logit contributions
istered+0.118
hes+0.113
aj+0.110
as+0.108
ends+0.108
Neg logit contributions
 itself-0.045
away-0.044
 side-0.043
 home-0.043
 angle-0.042
Token They
Token ID1119
Feature activation+1.97e-02

Pos logit contributions
 angle+0.026
away+0.024
riage+0.023
keeping+0.023
 Sus+0.023
Neg logit contributions
hes-0.047
istered-0.044
uts-0.044
as-0.042
ends-0.041
Token felt
Token ID2936
Feature activation+7.67e-02

Pos logit contributions
istered+0.046
aj+0.045
agging+0.045
oll+0.044
uts+0.044
Neg logit contributions
 Sus-0.018
 Mill-0.017
 Joey-0.017
 investigate-0.017
 inspect-0.017
Token sorry
Token ID7926
Feature activation+1.10e-02

Pos logit contributions
as+0.128
aj+0.124
istered+0.123
uts+0.122
hes+0.119
Neg logit contributions
 Joey-0.117
 itself-0.117
 Mill-0.116
 Sus-0.115
 inspect-0.113
Token for
Token ID329
Feature activation-2.82e-02

Pos logit contributions
hes+0.022
istered+0.022
as+0.021
uts+0.021
aj+0.020
Neg logit contributions
 angle-0.014
away-0.013
 Sus-0.013
 Mill-0.013
riage-0.013
Token happy
Token ID3772
Feature activation-4.78e-03

Pos logit contributions
plets+0.156
arest+0.154
br+0.153
uts+0.153
istered+0.152
Neg logit contributions
 home-0.113
 Sus-0.110
 Mill-0.108
 Joey-0.108
 Tim-0.106
Token and
Token ID290
Feature activation+2.81e-03

Pos logit contributions
 angle+0.006
away+0.006
 Mill+0.006
riage+0.006
 Sus+0.006
Neg logit contributions
hes-0.009
istered-0.009
as-0.009
uts-0.009
aj-0.008
Token hungry
Token ID14720
Feature activation-1.66e-02

Pos logit contributions
as+0.005
agging+0.005
aj+0.005
istered+0.005
oll+0.005
Neg logit contributions
 Joey-0.004
 Mill-0.004
 Sus-0.004
 Sally-0.004
 home-0.004
Token.
Token ID13
Feature activation-2.88e-02

Pos logit contributions
 angle+0.020
away+0.019
 Sus+0.019
 Mill+0.019
riage+0.019
Neg logit contributions
hes-0.033
istered-0.033
as-0.032
uts-0.032
aj-0.031
Token
Token ID198
Feature activation+8.46e-02

Pos logit contributions
 angle+0.034
away+0.034
 Mill+0.033
 Sus+0.032
riage+0.032
Neg logit contributions
hes-0.059
istered-0.058
uts-0.056
as-0.056
aj-0.054
Token
Token ID198
Feature activation+5.92e-02

Pos logit contributions
hes+0.163
istered+0.156
as+0.154
uts+0.153
aj+0.146
Neg logit contributions
 angle-0.111
away-0.108
riage-0.108
 Sus-0.107
 Mill-0.107
TokenBut
Token ID1537
Feature activation-4.67e-02

Pos logit contributions
istered+0.123
hes+0.118
oll+0.117
uts+0.116
ows+0.115
Neg logit contributions
away-0.067
 angle-0.065
riage-0.063
 Sus-0.063
 behold-0.061
Token then
Token ID788
Feature activation-3.34e-03

Pos logit contributions
 angle+0.071
away+0.069
keeping+0.067
 inspect+0.066
 Mill+0.063
Neg logit contributions
hes-0.082
ends-0.076
istered-0.076
aj-0.074
as-0.073
Token,
Token ID11
Feature activation-8.93e-03

Pos logit contributions
 angle+0.005
away+0.004
keeping+0.004
 Mill+0.004
 inspect+0.004
Neg logit contributions
hes-0.006
istered-0.006
as-0.006
uts-0.006
aj-0.006
Token a
Token ID257
Feature activation+2.37e-02

Pos logit contributions
 angle+0.013
away+0.013
 Mill+0.012
 inspect+0.012
 Sus+0.012
Neg logit contributions
hes-0.016
istered-0.016
as-0.015
uts-0.015
aj-0.014
Token loud
Token ID7812
Feature activation-4.61e-02

Pos logit contributions
hes+0.034
istered+0.034
uts+0.033
as+0.033
ows+0.031
Neg logit contributions
 angle-0.043
 Sus-0.041
away-0.041
 Mill-0.041
 Joey-0.040
Token He
Token ID679
Feature activation+1.66e-02

Pos logit contributions
hes+0.027
uts+0.027
as+0.026
istered+0.026
aj+0.026
Neg logit contributions
away-0.017
 angle-0.017
riage-0.015
 inspect-0.015
keeping-0.015
Token felt
Token ID2936
Feature activation+9.73e-02

Pos logit contributions
istered+0.035
hes+0.034
uts+0.034
as+0.033
oll+0.032
Neg logit contributions
 angle-0.018
 Sus-0.018
 Mill-0.018
away-0.018
 Joey-0.018
Token proud
Token ID6613
Feature activation+4.80e-02

Pos logit contributions
istered+0.165
as+0.164
uts+0.163
hes+0.161
aj+0.158
Neg logit contributions
 Joey-0.147
 Sus-0.146
 Mill-0.144
away-0.144
 angle-0.142
Token to
Token ID284
Feature activation-9.66e-03

Pos logit contributions
istered+0.093
hes+0.091
as+0.085
uts+0.084
zel+0.083
Neg logit contributions
 Sus-0.065
 angle-0.064
away-0.063
 Mill-0.063
 Joey-0.063
Token be
Token ID307
Feature activation+7.06e-02

Pos logit contributions
 angle+0.012
away+0.011
 Sus+0.011
 Mill+0.011
riage+0.011
Neg logit contributions
hes-0.019
istered-0.019
uts-0.018
as-0.018
agging-0.017
Token visiting
Token ID10013
Feature activation-2.06e-02

Pos logit contributions
istered+0.130
hes+0.128
uts+0.125
as+0.125
aj+0.120
Neg logit contributions
away-0.095
 angle-0.094
 Sus-0.094
 Mill-0.094
 Joey-0.092
Token his
Token ID465
Feature activation-5.34e-03

Pos logit contributions
 angle+0.021
riage+0.019
 Sus+0.018
 Mill+0.018
keeping+0.018
Neg logit contributions
hes-0.044
as-0.043
istered-0.043
uts-0.042
ows-0.041
Token grandparents
Token ID28571
Feature activation+9.47e-03

Pos logit contributions
 angle+0.007
 Sus+0.007
away+0.007
 Mill+0.007
riage+0.007
Neg logit contributions
hes-0.010
istered-0.010
as-0.010
uts-0.010
aj-0.009
Token for
Token ID329
Feature activation-3.65e-02

Pos logit contributions
hes+0.018
istered+0.018
as+0.018
uts+0.018
aj+0.017
Neg logit contributions
 angle-0.012
 Sus-0.011
away-0.011
 Mill-0.011
riage-0.011
Token the
Token ID262
Feature activation+3.63e-02

Pos logit contributions
 angle+0.049
 Sus+0.048
away+0.048
 Mill+0.047
riage+0.047
Neg logit contributions
hes-0.068
istered-0.066
as-0.064
uts-0.063
aj-0.061
Token very
Token ID845
Feature activation+3.42e-02

Pos logit contributions
hes+0.074
istered+0.073
uts+0.070
as+0.070
aj+0.067
Neg logit contributions
 angle-0.044
away-0.043
 Mill-0.043
 Sus-0.043
riage-0.042
Token you
Token ID345
Feature activation-2.73e-02

Pos logit contributions
 Sus+0.045
riage+0.045
 angle+0.044
 Joey+0.044
 Mill+0.044
Neg logit contributions
istered-0.055
hes-0.054
as-0.052
uts-0.052
aj-0.049
Token trust
Token ID3774
Feature activation-4.25e-02

Pos logit contributions
 investigate+0.035
 Joey+0.033
 inspect+0.032
 angle+0.031
away+0.031
Neg logit contributions
istered-0.055
uts-0.053
hes-0.053
aj-0.051
as-0.051
Token your
Token ID534
Feature activation-6.19e-03

Pos logit contributions
 angle+0.042
away+0.041
 Sus+0.041
 Mill+0.040
 Joey+0.040
Neg logit contributions
istered-0.094
hes-0.094
as-0.090
uts-0.090
aj-0.087
Token mom
Token ID1995
Feature activation-1.55e-02

Pos logit contributions
 angle+0.008
away+0.008
 Joey+0.008
 home+0.008
 side+0.008
Neg logit contributions
hes-0.012
istered-0.011
uts-0.011
as-0.011
oll-0.011
Token's
Token ID338
Feature activation+2.13e-02

Pos logit contributions
riage+0.018
 Sus+0.018
 Mill+0.018
 angle+0.018
 Joey+0.018
Neg logit contributions
istered-0.031
hes-0.030
uts-0.029
agging-0.028
as-0.028
Token suggestions
Token ID11776
Feature activation-2.23e-04

Pos logit contributions
hes+0.044
istered+0.043
as+0.042
uts+0.041
oll+0.040
Neg logit contributions
 angle-0.025
away-0.024
 Sus-0.024
 Joey-0.024
 Mill-0.024
Token,
Token ID11
Feature activation-4.62e-03

Pos logit contributions
 angle+0.000
away+0.000
 Sus+0.000
riage+0.000
 Joey+0.000
Neg logit contributions
istered-0.000
hes-0.000
as-0.000
oll-0.000
uts-0.000
Token you
Token ID345
Feature activation+5.50e-03

Pos logit contributions
 Sus+0.005
 Mill+0.005
 Joey+0.005
riage+0.005
 angle+0.005
Neg logit contributions
istered-0.010
hes-0.010
uts-0.009
as-0.009
agging-0.009
Token can
Token ID460
Feature activation+1.10e-02

Pos logit contributions
istered+0.011
hes+0.011
aj+0.011
ittens+0.011
uts+0.011
Neg logit contributions
 investigate-0.007
 Joey-0.006
 Sus-0.006
 Mill-0.006
 angle-0.006
Token be
Token ID307
Feature activation+2.68e-02

Pos logit contributions
istered+0.024
hes+0.023
as+0.023
aj+0.023
uts+0.021
Neg logit contributions
 investigate-0.014
 Joey-0.013
 inspect-0.013
 Mill-0.012
 Sus-0.012
Token very
Token ID845
Feature activation+4.93e-02

Pos logit contributions
istered+0.042
uts+0.042
oll+0.041
as+0.040
umbs+0.039
Neg logit contributions
 home-0.043
away-0.042
 Sus-0.042
 Joey-0.042
 Mill-0.041
Token.
Token ID13
Feature activation-4.49e-02

Pos logit contributions
hes+0.045
istered+0.045
as+0.043
uts+0.042
aj+0.041
Neg logit contributions
 angle-0.023
 Sus-0.022
away-0.022
riage-0.022
 Mill-0.022
Token They
Token ID1119
Feature activation+2.86e-03

Pos logit contributions
 angle+0.051
away+0.051
keeping+0.049
 Mill+0.049
 inspect+0.047
Neg logit contributions
hes-0.095
istered-0.092
as-0.090
uts-0.089
aj-0.088
Token share
Token ID2648
Feature activation-2.47e-02

Pos logit contributions
istered+0.006
hes+0.006
uts+0.005
as+0.005
aj+0.005
Neg logit contributions
 angle-0.004
 Sus-0.004
away-0.004
 Mill-0.004
 Joey-0.003
Token their
Token ID511
Feature activation-2.17e-02

Pos logit contributions
 angle+0.017
away+0.017
 Sus+0.017
 Mill+0.016
riage+0.016
Neg logit contributions
istered-0.063
hes-0.061
as-0.060
uts-0.059
oll-0.059
Token toys
Token ID14958
Feature activation+1.77e-02

Pos logit contributions
 angle+0.038
 Sus+0.038
 Mill+0.038
away+0.038
riage+0.037
Neg logit contributions
hes-0.031
istered-0.030
as-0.029
uts-0.029
aj-0.028
Token and
Token ID290
Feature activation-2.09e-02

Pos logit contributions
istered+0.042
hes+0.041
aj+0.038
plets+0.038
uch+0.038
Neg logit contributions
 home-0.017
away-0.016
 side-0.015
 Sus-0.014
 itself-0.014
Token have
Token ID423
Feature activation+1.11e-02

Pos logit contributions
 Sus+0.024
 Mill+0.023
 angle+0.023
riage+0.023
 Joey+0.023
Neg logit contributions
istered-0.044
hes-0.043
as-0.042
uts-0.042
oll-0.041
Token fun
Token ID1257
Feature activation-1.91e-02

Pos logit contributions
istered+0.013
hes+0.012
as+0.012
uts+0.012
aj+0.011
Neg logit contributions
 angle-0.023
 Sus-0.023
 Mill-0.022
away-0.022
 Joey-0.022
Token.
Token ID13
Feature activation-3.63e-02

Pos logit contributions
 angle+0.026
 Mill+0.025
 Sus+0.025
away+0.025
riage+0.025
Neg logit contributions
hes-0.036
istered-0.035
as-0.034
uts-0.034
aj-0.032
Token
Token ID198
Feature activation+7.88e-02

Pos logit contributions
away+0.043
 angle+0.043
keeping+0.042
 Mill+0.041
 Sus+0.040
Neg logit contributions
hes-0.075
istered-0.073
as-0.071
uts-0.070
aj-0.070
Token
Token ID198
Feature activation+5.41e-02

Pos logit contributions
istered+0.179
hes+0.171
uts+0.168
as+0.166
oll+0.165
Neg logit contributions
 angle-0.077
 Sus-0.074
away-0.073
 Mill-0.072
 Joey-0.071
Token should
Token ID815
Feature activation-1.20e-02

Pos logit contributions
istered+0.111
hes+0.110
uts+0.106
as+0.105
aj+0.103
Neg logit contributions
 angle-0.068
 Sus-0.068
away-0.067
 Mill-0.067
 Joey-0.066
Token play
Token ID711
Feature activation+9.82e-03

Pos logit contributions
 Sus+0.019
 Mill+0.019
 Joey+0.019
 angle+0.019
away+0.019
Neg logit contributions
istered-0.020
hes-0.020
as-0.019
uts-0.018
oll-0.018
Token the
Token ID262
Feature activation+4.66e-02

Pos logit contributions
hes+0.018
istered+0.017
as+0.017
uts+0.017
ends+0.016
Neg logit contributions
 angle-0.014
 Mill-0.013
 Sus-0.013
riage-0.013
 Joey-0.013
Token drum
Token ID13026
Feature activation+3.33e-02

Pos logit contributions
istered+0.092
hes+0.091
uts+0.090
as+0.085
oll+0.084
Neg logit contributions
 angle-0.062
away-0.060
 home-0.058
 side-0.058
 Sus-0.057
Token together
Token ID1978
Feature activation+2.63e-03

Pos logit contributions
istered+0.074
hes+0.073
as+0.069
uts+0.069
aj+0.068
Neg logit contributions
away-0.034
 angle-0.033
 Sus-0.032
 home-0.031
 Mill-0.031
Token,
Token ID11
Feature activation+2.27e-02

Pos logit contributions
hes+0.005
istered+0.005
as+0.005
uts+0.005
aj+0.005
Neg logit contributions
 angle-0.003
away-0.003
 Sus-0.003
 Mill-0.003
riage-0.003
Token and
Token ID290
Feature activation+1.51e-02

Pos logit contributions
istered+0.045
hes+0.044
as+0.042
uts+0.042
oll+0.041
Neg logit contributions
 Sus-0.029
 Mill-0.028
 Joey-0.028
 angle-0.028
away-0.027
Token make
Token ID787
Feature activation+2.35e-02

Pos logit contributions
istered+0.030
hes+0.030
as+0.028
uts+0.028
aj+0.027
Neg logit contributions
 angle-0.018
 Sus-0.018
 Mill-0.018
away-0.018
riage-0.018
Token a
Token ID257
Feature activation+6.69e-02

Pos logit contributions
istered+0.049
hes+0.048
uts+0.048
as+0.047
oll+0.046
Neg logit contributions
 Sus-0.026
away-0.025
 angle-0.025
 Mill-0.025
 itself-0.025
Token happy
Token ID3772
Feature activation-1.14e-02

Pos logit contributions
hes+0.110
uts+0.108
as+0.104
oll+0.103
istered+0.102
Neg logit contributions
 angle-0.109
away-0.109
 home-0.106
 Sus-0.105
 side-0.103
Token sound
Token ID2128
Feature activation+6.00e-02

Pos logit contributions
riage+0.017
 angle+0.016
 Mill+0.016
 inspect+0.016
away+0.016
Neg logit contributions
istered-0.018
hes-0.018
uts-0.018
as-0.017
aj-0.017
Token gave
Token ID2921
Feature activation-7.80e-03

Pos logit contributions
hes+0.029
istered+0.029
as+0.028
uts+0.028
aj+0.027
Neg logit contributions
 angle-0.017
 Sus-0.017
 Mill-0.017
away-0.017
 Joey-0.016
Token him
Token ID683
Feature activation-1.13e-02

Pos logit contributions
 angle+0.006
riage+0.005
away+0.005
 Mill+0.004
 exp+0.004
Neg logit contributions
hes-0.020
as-0.019
istered-0.019
uts-0.018
oll-0.018
Token a
Token ID257
Feature activation+2.74e-02

Pos logit contributions
 angle+0.017
away+0.016
 Mill+0.016
 Sus+0.016
riage+0.016
Neg logit contributions
hes-0.020
istered-0.019
as-0.019
uts-0.019
aj-0.018
Token pe
Token ID613
Feature activation+6.65e-02

Pos logit contributions
hes+0.046
istered+0.046
uts+0.045
as+0.044
ows+0.042
Neg logit contributions
 angle-0.044
 Sus-0.044
 Joey-0.043
 Mill-0.043
 inspect-0.043
Tokenb
Token ID65
Feature activation+5.77e-02

Pos logit contributions
istered+0.121
uts+0.117
hes+0.115
umbs+0.114
plets+0.106
Neg logit contributions
away-0.096
 angle-0.091
 inspect-0.091
 home-0.089
 Sus-0.084
Tokenble
Token ID903
Feature activation+1.82e-02

Pos logit contributions
istered+0.087
hes+0.085
uts+0.077
as+0.077
aj+0.076
Neg logit contributions
 Sus-0.098
 angle-0.098
away-0.096
 Mill-0.096
 Joey-0.095
Token.
Token ID13
Feature activation-3.62e-02

Pos logit contributions
hes+0.037
istered+0.037
as+0.036
uts+0.035
aj+0.034
Neg logit contributions
 angle-0.021
away-0.021
 Sus-0.021
 Mill-0.020
 Joey-0.020
Token He
Token ID679
Feature activation-2.20e-02

Pos logit contributions
 angle+0.034
away+0.033
 Sus+0.032
 Mill+0.032
riage+0.032
Neg logit contributions
hes-0.082
istered-0.081
as-0.079
uts-0.079
aj-0.076
Token had
Token ID550
Feature activation+2.82e-02

Pos logit contributions
 Sus+0.023
 Joey+0.023
 Mill+0.022
 angle+0.022
 inspect+0.022
Neg logit contributions
istered-0.048
hes-0.047
uts-0.046
as-0.045
agging-0.045
Token never
Token ID1239
Feature activation+6.84e-04

Pos logit contributions
istered+0.059
ows+0.053
uts+0.053
as+0.053
aj+0.053
Neg logit contributions
 Sus-0.037
 Mill-0.035
 Joey-0.035
 home-0.034
 inspect-0.033
Token seen
Token ID1775
Feature activation-1.75e-02

Pos logit contributions
istered+0.001
hes+0.001
uts+0.001
as+0.001
aj+0.001
Neg logit contributions
 angle-0.001
 Sus-0.001
away-0.001
 Mill-0.001
riage-0.001
Token a
Token ID257
Feature activation+1.07e-02

Pos logit contributions
hes+0.017
istered+0.016
as+0.015
ends+0.015
uts+0.015
Neg logit contributions
 angle-0.013
riage-0.012
away-0.012
keeping-0.012
 Mill-0.012
Token better
Token ID1365
Feature activation-6.53e-02

Pos logit contributions
hes+0.018
istered+0.018
as+0.018
uts+0.018
oll+0.016
Neg logit contributions
 angle-0.016
away-0.016
 Sus-0.016
 Mill-0.016
 inspect-0.016
Token look
Token ID804
Feature activation+1.07e-02

Pos logit contributions
 angle+0.076
away+0.075
 Sus+0.072
 Mill+0.070
 Joey+0.069
Neg logit contributions
hes-0.140
istered-0.136
as-0.134
uts-0.132
keys-0.128
Token.
Token ID13
Feature activation-4.76e-02

Pos logit contributions
istered+0.023
hes+0.023
as+0.022
uts+0.022
aj+0.021
Neg logit contributions
away-0.011
 angle-0.011
 Sus-0.011
 Mill-0.011
riage-0.011
Token He
Token ID679
Feature activation-4.19e-02

Pos logit contributions
away+0.060
 angle+0.058
riage+0.054
keeping+0.052
 Mill+0.049
Neg logit contributions
hes-0.098
uts-0.092
ows-0.091
istered-0.090
ends-0.090
Token noticed
Token ID6810
Feature activation-4.79e-02

Pos logit contributions
 angle+0.050
 Sus+0.049
away+0.049
 Mill+0.049
riage+0.048
Neg logit contributions
hes-0.083
istered-0.083
uts-0.080
as-0.080
aj-0.076
Token the
Token ID262
Feature activation+2.30e-02

Pos logit contributions
 angle+0.068
riage+0.067
away+0.066
 Mill+0.064
 Sus+0.064
Neg logit contributions
hes-0.086
istered-0.081
as-0.081
uts-0.080
oll-0.076
Token body
Token ID1767
Feature activation+2.46e-02

Pos logit contributions
hes+0.042
uts+0.041
as+0.041
istered+0.041
keys+0.040
Neg logit contributions
 angle-0.035
 side-0.033
 home-0.032
away-0.031
 inspect-0.031
Token had
Token ID550
Feature activation-8.67e-04

Pos logit contributions
istered+0.051
hes+0.049
aj+0.048
as+0.047
oll+0.046
Neg logit contributions
 angle-0.029
 Sus-0.028
 Mill-0.027
away-0.027
 Joey-0.027
Token all
Token ID477
Feature activation+1.45e-03

Pos logit contributions
 Sus+0.001
 Mill+0.001
 itself+0.001
 angle+0.001
 Joey+0.001
Neg logit contributions
istered-0.002
as-0.002
hes-0.002
uts-0.001
aj-0.001
Token different
Token ID1180
Feature activation-2.39e-02

Pos logit contributions
hes+0.003
istered+0.003
as+0.003
uts+0.003
aj+0.003
Neg logit contributions
 angle-0.001
 Sus-0.001
away-0.001
 Mill-0.001
riage-0.001
TokenLook
Token ID8567
Feature activation+1.05e-02

Pos logit contributions
 angle+0.008
riage+0.008
 Sus+0.008
 Mill+0.008
away+0.008
Neg logit contributions
hes-0.014
istered-0.014
uts-0.013
as-0.013
zel-0.013
Token,
Token ID11
Feature activation+6.12e-04

Pos logit contributions
istered+0.018
hes+0.018
as+0.017
uts+0.017
oll+0.017
Neg logit contributions
 angle-0.016
keeping-0.015
riage-0.014
away-0.014
 generation-0.014
Token it
Token ID340
Feature activation+1.94e-02

Pos logit contributions
hes+0.001
istered+0.001
uts+0.001
as+0.001
keys+0.001
Neg logit contributions
 angle-0.001
away-0.001
keeping-0.001
riage-0.001
 Sus-0.001
Token's
Token ID338
Feature activation+1.94e-02

Pos logit contributions
istered+0.043
aj+0.043
as+0.043
ittens+0.042
uts+0.042
Neg logit contributions
 investigate-0.019
 inspect-0.016
 Where-0.016
 away-0.016
 Bobby-0.016
Token coming
Token ID2406
Feature activation-1.56e-02

Pos logit contributions
as+0.039
istered+0.037
uts+0.037
itt+0.037
aj+0.036
Neg logit contributions
 home-0.027
 Tim-0.026
 Joey-0.025
 Sally-0.025
 side-0.024
Token towards
Token ID3371
Feature activation-7.87e-02

Pos logit contributions
 angle+0.021
riage+0.020
 Mill+0.020
 Sus+0.019
 Joey+0.019
Neg logit contributions
istered-0.028
as-0.027
hes-0.027
uts-0.026
ends-0.026
Token us
Token ID514
Feature activation-3.79e-02

Pos logit contributions
 angle+0.106
riage+0.102
 Sus+0.101
away+0.101
 Mill+0.101
Neg logit contributions
hes-0.146
istered-0.137
uts-0.136
as-0.136
aj-0.127
Token!"
Token ID2474
Feature activation-2.18e-02

Pos logit contributions
 angle+0.054
away+0.053
 Sus+0.053
 Mill+0.052
 Joey+0.052
Neg logit contributions
hes-0.068
istered-0.067
as-0.062
uts-0.062
oll-0.061
Token
Token ID220
Feature activation+8.87e-02

Pos logit contributions
away+0.035
 angle+0.034
riage+0.032
keeping+0.030
 generation+0.029
Neg logit contributions
hes-0.039
uts-0.037
ittens-0.035
aj-0.034
ows-0.034
Token
Token ID198
Feature activation+6.78e-02

Pos logit contributions
istered+0.208
hes+0.199
as+0.194
oll+0.193
agging+0.191
Neg logit contributions
 Sus-0.077
 Joey-0.076
 Mill-0.074
 angle-0.071
 itself-0.069
Token
Token ID198
Feature activation+4.90e-02

Pos logit contributions
hes+0.130
istered+0.126
as+0.124
uts+0.123
aj+0.118
Neg logit contributions
 angle-0.089
riage-0.086
away-0.086
 Sus-0.084
 Mill-0.084
Token years
Token ID812
Feature activation+9.09e-02

Pos logit contributions
istered+0.089
hes+0.089
uts+0.084
plets+0.082
ends+0.081
Neg logit contributions
away-0.070
 angle-0.068
 Sus-0.067
 Mill-0.065
 side-0.065
Token old
Token ID1468
Feature activation+1.15e-02

Pos logit contributions
hes+0.163
istered+0.163
as+0.154
uts+0.153
aj+0.146
Neg logit contributions
 angle-0.129
away-0.127
 Sus-0.127
 Mill-0.126
 Joey-0.124
Token and
Token ID290
Feature activation+1.57e-02

Pos logit contributions
istered+0.018
hes+0.018
uts+0.018
as+0.018
oll+0.017
Neg logit contributions
 angle-0.019
riage-0.018
 Mill-0.018
 inspect-0.018
away-0.017
Token liked
Token ID8288
Feature activation+1.00e-03

Pos logit contributions
istered+0.032
as+0.031
uts+0.031
aj+0.031
umbs+0.030
Neg logit contributions
 Sus-0.019
 Joey-0.018
 Mill-0.018
riage-0.017
 inspect-0.017
Token to
Token ID284
Feature activation-5.67e-02

Pos logit contributions
hes+0.002
uts+0.002
as+0.002
istered+0.002
aj+0.002
Neg logit contributions
riage-0.001
 angle-0.001
away-0.001
 behold-0.001
 Mill-0.001
Token explore
Token ID7301
Feature activation-5.33e-02

Pos logit contributions
 angle+0.053
away+0.051
riage+0.051
 Sus+0.050
 Mill+0.049
Neg logit contributions
hes-0.128
istered-0.124
uts-0.122
as-0.120
oll-0.117
Token outside
Token ID2354
Feature activation-4.62e-02

Pos logit contributions
 angle+0.064
riage+0.063
keeping+0.062
 Sus+0.060
 Mill+0.059
Neg logit contributions
hes-0.105
istered-0.101
as-0.100
uts-0.098
agging-0.096
Token with
Token ID351
Feature activation-1.87e-02

Pos logit contributions
 angle+0.065
riage+0.064
 Mill+0.064
 Sus+0.063
keeping+0.063
Neg logit contributions
hes-0.082
istered-0.078
as-0.076
uts-0.076
agging-0.073
Token his
Token ID465
Feature activation-5.01e-03

Pos logit contributions
 angle+0.008
 Sus+0.008
 Mill+0.007
away+0.007
 Joey+0.007
Neg logit contributions
istered-0.052
hes-0.052
as-0.051
uts-0.051
zel-0.049
Token parents
Token ID3397
Feature activation+1.49e-02

Pos logit contributions
 Mill+0.008
 Sus+0.008
 Joey+0.008
 home+0.008
 angle+0.008
Neg logit contributions
hes-0.008
istered-0.008
uts-0.008
as-0.008
zel-0.007
Token.
Token ID13
Feature activation-2.39e-02

Pos logit contributions
hes+0.029
istered+0.029
as+0.028
uts+0.028
aj+0.026
Neg logit contributions
 angle-0.019
away-0.018
 Sus-0.018
 Mill-0.018
riage-0.018
Token,
Token ID11
Feature activation+2.30e-03

Pos logit contributions
 angle+0.053
away+0.051
 Sus+0.050
 Mill+0.050
riage+0.050
Neg logit contributions
hes-0.109
istered-0.108
as-0.105
uts-0.104
aj-0.101
Token Anna
Token ID11735
Feature activation-2.07e-02

Pos logit contributions
hes+0.005
istered+0.004
as+0.004
uts+0.004
aj+0.004
Neg logit contributions
 angle-0.003
 Sus-0.003
away-0.003
 Mill-0.003
riage-0.003
Token and
Token ID290
Feature activation-9.58e-03

Pos logit contributions
 angle+0.027
 Sus+0.027
away+0.027
 Mill+0.027
riage+0.026
Neg logit contributions
hes-0.039
istered-0.039
as-0.037
uts-0.037
aj-0.035
Token Emma
Token ID18966
Feature activation-3.20e-02

Pos logit contributions
 angle+0.014
away+0.012
keeping+0.012
 inspect+0.011
 Sus+0.011
Neg logit contributions
hes-0.018
oll-0.017
istered-0.017
as-0.016
aj-0.016
Token,
Token ID11
Feature activation+1.68e-02

Pos logit contributions
 angle+0.041
 Sus+0.040
away+0.040
 Mill+0.039
riage+0.039
Neg logit contributions
hes-0.062
istered-0.061
as-0.059
uts-0.058
aj-0.056
Token let
Token ID1309
Feature activation-4.68e-02

Pos logit contributions
istered+0.029
hes+0.029
uts+0.027
as+0.027
aj+0.026
Neg logit contributions
 angle-0.025
 Sus-0.024
away-0.024
 Mill-0.024
 Joey-0.024
Token's
Token ID338
Feature activation-1.40e-02

Pos logit contributions
 angle+0.074
 Sus+0.074
 Mill+0.074
away+0.073
 Joey+0.072
Neg logit contributions
istered-0.075
hes-0.075
as-0.072
uts-0.071
aj-0.068
Token go
Token ID467
Feature activation-1.04e-02

Pos logit contributions
 angle+0.015
 Sus+0.015
 Mill+0.015
away+0.015
 Joey+0.014
Neg logit contributions
istered-0.030
hes-0.030
as-0.029
uts-0.028
aj-0.027
Token have
Token ID423
Feature activation-7.31e-03

Pos logit contributions
 angle+0.013
away+0.012
 Sus+0.012
 Mill+0.012
riage+0.012
Neg logit contributions
hes-0.021
istered-0.021
as-0.020
uts-0.020
aj-0.019
Token dinner
Token ID8073
Feature activation-1.23e-02

Pos logit contributions
 Sus+0.009
 Mill+0.009
 Joey+0.009
away+0.009
 angle+0.009
Neg logit contributions
istered-0.014
hes-0.014
as-0.014
uts-0.014
ows-0.013
Token.
Token ID13
Feature activation-3.74e-02

Pos logit contributions
 home+0.012
away+0.011
 again+0.011
keeping+0.010
 Sus+0.010
Neg logit contributions
istered-0.030
hes-0.028
plets-0.027
agging-0.026
as-0.026
Token with
Token ID351
Feature activation-1.12e-02

Pos logit contributions
hes+0.005
istered+0.005
as+0.005
uts+0.005
aj+0.005
Neg logit contributions
 angle-0.003
away-0.003
 Sus-0.003
 Mill-0.003
 Joey-0.003
Token his
Token ID465
Feature activation-5.36e-03

Pos logit contributions
 Sus+0.013
 angle+0.013
 Mill+0.013
 Joey+0.013
away+0.012
Neg logit contributions
istered-0.023
hes-0.023
uts-0.022
as-0.022
aj-0.022
Token pencil
Token ID21613
Feature activation+4.16e-02

Pos logit contributions
 angle+0.008
 home+0.007
 side+0.007
 Sus+0.007
 inspect+0.007
Neg logit contributions
hes-0.010
istered-0.010
as-0.009
uts-0.009
umbs-0.009
Token and
Token ID290
Feature activation-1.51e-02

Pos logit contributions
hes+0.092
istered+0.091
agging+0.082
aj+0.081
as+0.081
Neg logit contributions
away-0.042
 home-0.040
 angle-0.040
 Sus-0.040
 itself-0.039
Token at
Token ID379
Feature activation-6.08e-02

Pos logit contributions
 Sus+0.018
 angle+0.017
 Mill+0.017
 Joey+0.017
 inspect+0.017
Neg logit contributions
istered-0.031
hes-0.030
as-0.029
uts-0.029
agging-0.029
Token last
Token ID938
Feature activation-6.05e-02

Pos logit contributions
 angle+0.062
riage+0.061
away+0.061
 Mill+0.061
 Sus+0.059
Neg logit contributions
hes-0.130
istered-0.126
as-0.125
uts-0.122
aj-0.119
Token,
Token ID11
Feature activation-8.40e-03

Pos logit contributions
riage+0.075
 angle+0.075
 Mill+0.074
 Sus+0.074
away+0.073
Neg logit contributions
hes-0.116
istered-0.113
uts-0.111
as-0.110
oll-0.104
Token he
Token ID339
Feature activation-2.73e-02

Pos logit contributions
 Sus+0.010
 angle+0.010
 Mill+0.010
 Joey+0.010
away+0.010
Neg logit contributions
istered-0.017
hes-0.017
as-0.016
uts-0.016
aj-0.015
Token was
Token ID373
Feature activation+5.40e-02

Pos logit contributions
 angle+0.031
 Sus+0.031
away+0.029
 Mill+0.029
 inspect+0.029
Neg logit contributions
istered-0.057
hes-0.055
uts-0.053
as-0.052
agging-0.052
Token done
Token ID1760
Feature activation+5.90e-03

Pos logit contributions
istered+0.093
uts+0.089
as+0.088
aj+0.085
ittens+0.084
Neg logit contributions
 Sus-0.082
 Mill-0.079
 home-0.078
 Joey-0.078
 angle-0.077
Token!
Token ID0
Feature activation-4.12e-02

Pos logit contributions
hes+0.011
istered+0.011
uts+0.011
as+0.011
aj+0.010
Neg logit contributions
 angle-0.008
away-0.007
 Mill-0.007
 Sus-0.007
riage-0.007