TOP ACTIVATIONS
MAX = 9.564

I can find something gl immer ing ?"
drink it . It looks y ummy ." Mia says ,
and lettuce . It looks y ummy . "
sprink les . It looks y ummy . "
cher ries . It looks y ummy . "
says . " It looks y ummy !" "
cher ries . It looked y ummy . Lily and Ben
says . " It looks y ummy . Let 's have
the ground . It looks y ummy . "
cake . It looks so y ummy . Can we have
jam . " It looks y ummy !" They
red and shiny and looks y ummy . Ben says ,
Max thought it was too sm elly and didn 't want
" Wow , it looks y ummy ," Amy said .
is this ? It looks y ucky . I don 't
red strawberry . It looks y ummy . "
on it . It looked y ummy and sweet .
the apple . It looks y ummy ." Ben
table . It looks very y ummy and fancy . It
eat it ? It looks y ummy ." His

MIN ACTIVATIONS
MIN = -9.464

Every night , she would read a book that
her av oc ados . She thanked her mom and ate
difficult . Each time , she was happy when she saw
it . At the end she was happy and proud of
was happy and tired . He curled up in his cozy
she did good things . She decided to sell some of
cream . Every night , she would stand in front of
played and had fun . She saw the same bird from
my was so happy . He held the arrow up and
the wedding . Instead , she went to the park and
dream . Every night , she would close her eyes and
L ily and Ben also wake up and look out
got to the school , she saw many kids . They
my was so happy . He watched as the bright fireworks
became healthier and happier . He was no longer a thin
he was so happy . He wanted to try something else
my count the flowers . She felt wide awake and ready
found it . As the woman kept walking , she heard
the leash worked again . She was happy to go on
birds eat the jelly . She laughed and cl apped her

INTERVAL 4.807 - 9.564
CONTAINS 1.759%

't find it . B enny asked his
them to be careful . " Only feed the
it would go . As he was building
wanted to wear it . M ia put on
not like to wear warm clothes . They liked to wear

INTERVAL 0.050 - 4.807
CONTAINS 52.133%

it was time to go home . They waved goodbye to
named Lily who loved to read . She had a unique
He hugged his mom and said thank you . From that
to set . He slowly opened his eyes
different hairc uts . She asks Lily to choose one .

INTERVAL -4.707 - 0.050
CONTAINS 37.096%

His friends cheered him on and they all went back inside
sees the k ite . The k ite is red and
They walked back to their mom , holding hands
with her hands . It was hard to move , but
and Lily were playing in the living room . They liked

INTERVAL -9.464 - -4.707
CONTAINS 9.012%

best friend , Tommy . Tommy said , " You can
One day , Kitty found a pretty gem under
and Max became good friends and danced together every time they
He did so . He took off his head and
was confused , he had never seen a turkey before .
Token I
Token ID314
Feature activation-2.55e+00

Pos logit contributions
he+0.006
ing+0.006
ore+0.006
ay+0.006
ings+0.006
Neg logit contributions
 forgot-0.010
 remembered-0.008
 knew-0.008
 wished-0.008
 struggled-0.007
Token can
Token ID460
Feature activation-3.84e-01

Pos logit contributions
 forgot+0.052
 remembered+0.039
 wished+0.038
 knew+0.036
 regretted+0.035
Neg logit contributions
he-0.064
ore-0.061
ay-0.060
ings-0.059
room-0.058
Token find
Token ID1064
Feature activation+2.09e+00

Pos logit contributions
 forgot+0.009
 knew+0.007
 remembered+0.007
 wished+0.007
 knocked+0.006
Neg logit contributions
he-0.009
ing-0.008
ore-0.008
ay-0.008
ays-0.008
Token something
Token ID1223
Feature activation+7.50e-01

Pos logit contributions
he+0.042
room+0.037
ing+0.037
ings+0.037
ore+0.036
Neg logit contributions
 forgot-0.051
 knew-0.043
 remembered-0.043
 wished-0.043
 knocked-0.041
Token gl
Token ID1278
Feature activation+5.83e+00

Pos logit contributions
he+0.012
ing+0.011
room+0.011
ay+0.010
ays+0.010
Neg logit contributions
 forgot-0.022
 knew-0.019
 remembered-0.019
 wished-0.018
 bit-0.018
Tokenimmer
Token ID10957
Feature activation+9.56e+00

Pos logit contributions
he+0.096
room+0.087
ays+0.087
ings+0.083
ummy+0.080
Neg logit contributions
 forgot-0.143
'd-0.124
 remembered-0.120
 packed-0.118
 knew-0.114
Tokening
Token ID278
Feature activation+3.90e+00

Pos logit contributions
overed+0.177
uggle+0.166
aps+0.161
ooming+0.161
room+0.158
Neg logit contributions
 forgot-0.190
 bit-0.178
'd-0.172
 packed-0.164
 remembered-0.161
Token?"
Token ID1701
Feature activation+3.89e+00

Pos logit contributions
he+0.057
room+0.053
ay+0.053
ing+0.053
ays+0.053
Neg logit contributions
 forgot-0.113
 remembered-0.096
 knew-0.095
 bit-0.095
 wished-0.093
Token
Token ID220
Feature activation+4.81e+00

Pos logit contributions
he+0.072
ays+0.069
ay+0.069
ings+0.069
ing+0.069
Neg logit contributions
 forgot-0.102
 knew-0.087
 wished-0.085
 remembered-0.084
 learned-0.080
Token
Token ID198
Feature activation+5.10e+00

Pos logit contributions
room+0.093
ay+0.092
eddy+0.092
ays+0.091
ing+0.089
Neg logit contributions
 forgot-0.101
 knew-0.089
 told-0.082
 wished-0.079
 learned-0.079
Token
Token ID198
Feature activation+3.71e+00

Pos logit contributions
ays+0.111
ement+0.093
ings+0.091
ing+0.091
over+0.091
Neg logit contributions
 forgot-0.118
 knew-0.110
 learned-0.103
 remembered-0.098
 told-0.098
Token drink
Token ID4144
Feature activation+4.99e+00

Pos logit contributions
 forgot+0.069
 knew+0.055
 remembered+0.055
 known+0.053
 wished+0.053
Neg logit contributions
he-0.063
ore-0.057
ay-0.057
uggle-0.055
 with-0.054
Token it
Token ID340
Feature activation+3.08e+00

Pos logit contributions
he+0.108
over+0.099
room+0.099
ane+0.096
apping+0.096
Neg logit contributions
 forgot-0.111
 bit-0.094
 knew-0.092
 wished-0.087
 remembered-0.087
Token.
Token ID13
Feature activation+3.22e+00

Pos logit contributions
he+0.049
ore+0.047
ays+0.045
ings+0.045
ay+0.045
Neg logit contributions
 forgot-0.088
 bit-0.074
 knew-0.072
 remembered-0.071
 wished-0.069
Token It
Token ID632
Feature activation-1.89e+00

Pos logit contributions
he+0.068
ing+0.061
ings+0.061
ogging+0.057
ore+0.057
Neg logit contributions
 forgot-0.079
 remembered-0.065
 bit-0.064
 wished-0.064
 knew-0.063
Token looks
Token ID3073
Feature activation+2.31e-03

Pos logit contributions
 forgot+0.046
 known+0.040
 remembered+0.039
 wished+0.039
 robbers+0.038
Neg logit contributions
ing-0.033
ays-0.033
ings-0.032
ore-0.032
he-0.032
Token y
Token ID331
Feature activation+8.52e+00

Pos logit contributions
he+0.000
ay+0.000
ing+0.000
ings+0.000
ore+0.000
Neg logit contributions
 forgot-0.000
 remembered-0.000
 knew-0.000
 wished-0.000
 struggled-0.000
Tokenummy
Token ID13513
Feature activation+4.48e+00

Pos logit contributions
overed+0.106
he+0.102
ute+0.093
ane+0.092
eway+0.086
Neg logit contributions
 forgot-0.205
 bit-0.204
 mixed-0.176
 f-0.176
 sp-0.174
Token."
Token ID526
Feature activation+4.32e+00

Pos logit contributions
he+0.074
room+0.068
ings+0.066
ore+0.066
ay+0.064
Neg logit contributions
 forgot-0.128
 bit-0.109
 remembered-0.106
 knew-0.106
 wished-0.105
Token Mia
Token ID32189
Feature activation-5.84e+00

Pos logit contributions
he+0.073
ing+0.070
ay+0.067
ays+0.066
ings+0.063
Neg logit contributions
 forgot-0.124
 knew-0.103
 remembered-0.100
 wished-0.099
 learned-0.098
Token says
Token ID1139
Feature activation+1.20e+00

Pos logit contributions
 forgot+0.133
 remembered+0.102
 wished+0.101
 known+0.101
 forgotten+0.100
Neg logit contributions
he-0.129
ing-0.120
ore-0.118
ays-0.118
ay-0.116
Token,
Token ID11
Feature activation+6.02e-01

Pos logit contributions
he+0.021
ays+0.019
eddy+0.018
ay+0.018
ore+0.017
Neg logit contributions
 forgot-0.035
 knew-0.031
 remembered-0.030
 bit-0.029
 wished-0.028
Token and
Token ID290
Feature activation-3.88e+00

Pos logit contributions
he+0.067
ore+0.057
uggle+0.057
ay+0.056
room+0.054
Neg logit contributions
 forgot-0.124
 remembered-0.106
 knew-0.104
 wished-0.098
 bit-0.096
Token lettuce
Token ID39406
Feature activation+3.44e+00

Pos logit contributions
 forgot+0.097
 known+0.078
 knew+0.076
 punished+0.076
 struggled+0.074
Neg logit contributions
ing-0.062
ummy-0.061
ay-0.061
ays-0.061
he-0.061
Token.
Token ID13
Feature activation+3.75e+00

Pos logit contributions
he+0.035
ay+0.030
ore+0.029
ing+0.029
room+0.028
Neg logit contributions
 forgot-0.122
 remembered-0.105
 knew-0.104
 wished-0.104
 bit-0.099
Token It
Token ID632
Feature activation-3.50e+00

Pos logit contributions
he+0.083
ay+0.077
ing+0.076
room+0.075
uggle+0.071
Neg logit contributions
 forgot-0.087
 wished-0.074
 knew-0.073
 remembered-0.068
 knocked-0.067
Token looks
Token ID3073
Feature activation-5.13e-01

Pos logit contributions
 forgot+0.093
 known+0.079
 punished+0.077
 remembered+0.075
 forgotten+0.075
Neg logit contributions
ays-0.057
he-0.055
ore-0.052
 with-0.051
iffs-0.051
Token y
Token ID331
Feature activation+8.48e+00

Pos logit contributions
 forgot+0.011
 remembered+0.008
 knew+0.008
 wished+0.008
 struggled+0.008
Neg logit contributions
he-0.012
ing-0.011
ay-0.011
ore-0.011
ings-0.011
Tokenummy
Token ID13513
Feature activation+4.54e+00

Pos logit contributions
he+0.100
ute+0.091
overed+0.091
uggle+0.086
ane+0.086
Neg logit contributions
 forgot-0.226
 bit-0.201
 knew-0.192
 forgotten-0.186
 wasted-0.184
Token.
Token ID13
Feature activation+3.63e+00

Pos logit contributions
he+0.062
room+0.058
ay+0.055
ore+0.054
uggle+0.054
Neg logit contributions
 forgot-0.142
 knew-0.121
 bit-0.121
 remembered-0.121
 wished-0.120
Token
Token ID198
Feature activation+4.78e+00

Pos logit contributions
he+0.076
ay+0.070
room+0.070
ing+0.069
ings+0.065
Neg logit contributions
 forgot-0.089
 wished-0.075
 knew-0.075
 remembered-0.071
 knocked-0.069
Token
Token ID198
Feature activation+3.79e+00

Pos logit contributions
ays+0.096
ing+0.095
ement+0.094
ipping+0.094
ay+0.089
Neg logit contributions
 forgot-0.111
 knew-0.097
 wished-0.090
 also-0.088
 learned-0.088
Token"
Token ID1
Feature activation+4.83e-01

Pos logit contributions
he+0.055
ay+0.053
ore+0.051
ays+0.051
ing+0.050
Neg logit contributions
 forgot-0.118
 remembered-0.098
 knew-0.098
 wished-0.098
 bit-0.093
Token sprink
Token ID27562
Feature activation+2.32e+00

Pos logit contributions
he+0.048
ore+0.042
uggle+0.041
ay+0.040
room+0.040
Neg logit contributions
 forgot-0.074
 remembered-0.062
 knew-0.061
 wished-0.060
 bit-0.059
Tokenles
Token ID829
Feature activation+3.64e+00

Pos logit contributions
he+0.038
room+0.037
ore+0.036
uggle+0.036
ings+0.035
Neg logit contributions
 forgot-0.063
 knew-0.052
 remembered-0.051
 bit-0.050
 wished-0.049
Token.
Token ID13
Feature activation+3.66e+00

Pos logit contributions
he+0.046
uggle+0.042
ore+0.041
ay+0.039
ings+0.039
Neg logit contributions
 forgot-0.114
 knew-0.099
 remembered-0.098
 wished-0.096
 bit-0.095
Token It
Token ID632
Feature activation-4.57e+00

Pos logit contributions
he+0.095
uggle+0.088
ing+0.088
ings+0.087
ore+0.087
Neg logit contributions
 forgot-0.075
 knew-0.056
 wished-0.055
 remembered-0.055
 learned-0.051
Token looks
Token ID3073
Feature activation-9.45e-01

Pos logit contributions
 forgot+0.115
 known+0.099
 punished+0.096
 forgotten+0.095
 remembered+0.094
Neg logit contributions
ays-0.084
he-0.076
 with-0.075
ore-0.074
ay-0.072
Token y
Token ID331
Feature activation+8.41e+00

Pos logit contributions
 forgot+0.023
 remembered+0.018
 knew+0.018
 wished+0.018
 struggled+0.017
Neg logit contributions
he-0.019
ing-0.018
ay-0.018
ings-0.018
ore-0.018
Tokenummy
Token ID13513
Feature activation+3.90e+00

Pos logit contributions
he+0.101
overed+0.088
ore+0.086
ane+0.083
uggle+0.082
Neg logit contributions
 forgot-0.220
 bit-0.200
 knew-0.188
 f-0.183
 forgotten-0.178
Token.
Token ID13
Feature activation+3.60e+00

Pos logit contributions
he+0.053
room+0.048
ore+0.047
ay+0.046
uggle+0.046
Neg logit contributions
 forgot-0.122
 bit-0.105
 knew-0.104
 wished-0.104
 remembered-0.104
Token
Token ID198
Feature activation+4.82e+00

Pos logit contributions
he+0.087
ings+0.080
uggle+0.080
ing+0.080
ore+0.078
Neg logit contributions
 forgot-0.079
 knew-0.062
 wished-0.061
 remembered-0.060
 knocked-0.056
Token
Token ID198
Feature activation+3.84e+00

Pos logit contributions
ays+0.104
ing+0.100
ement+0.100
ipping+0.099
over+0.098
Neg logit contributions
 forgot-0.107
 knew-0.095
 wished-0.085
 learned-0.085
 told-0.084
Token"
Token ID1
Feature activation+2.73e-01

Pos logit contributions
he+0.055
ings+0.050
ore+0.050
ay+0.050
ays+0.050
Neg logit contributions
 forgot-0.122
 remembered-0.102
 knew-0.101
 wished-0.101
 bit-0.096
Token cher
Token ID21382
Feature activation+3.62e+00

Pos logit contributions
 forgot+0.108
 knew+0.089
 remembered+0.087
 struggled+0.084
 missed+0.084
Neg logit contributions
he-0.051
ing-0.049
ays-0.049
ay-0.047
room-0.047
Tokenries
Token ID1678
Feature activation+2.91e+00

Pos logit contributions
he+0.052
room+0.049
ore+0.047
uggle+0.046
time+0.045
Neg logit contributions
 forgot-0.109
 remembered-0.095
 knew-0.089
 knocked-0.088
 wasted-0.088
Token.
Token ID13
Feature activation+3.48e+00

Pos logit contributions
he+0.035
room+0.035
ay+0.034
ore+0.033
uggle+0.033
Neg logit contributions
 forgot-0.093
 remembered-0.079
 knew-0.079
 wished-0.078
 bit-0.077
Token It
Token ID632
Feature activation-4.04e+00

Pos logit contributions
he+0.078
ing+0.075
uggle+0.071
ingle+0.070
ipping+0.069
Neg logit contributions
 forgot-0.078
 wished-0.065
 knew-0.063
 learned-0.062
 remembered-0.060
Token looks
Token ID3073
Feature activation-7.19e-01

Pos logit contributions
 forgot+0.102
 known+0.085
 remembered+0.082
 struggled+0.082
 forgotten+0.081
Neg logit contributions
ays-0.072
he-0.069
 with-0.065
ore-0.062
ay-0.062
Token y
Token ID331
Feature activation+8.36e+00

Pos logit contributions
 forgot+0.017
 remembered+0.013
 knew+0.013
 wished+0.013
 struggled+0.012
Neg logit contributions
he-0.016
ing-0.015
ay-0.015
ings-0.014
ore-0.014
Tokenummy
Token ID13513
Feature activation+4.23e+00

Pos logit contributions
he+0.095
ute+0.090
overed+0.085
uggle+0.085
ore+0.085
Neg logit contributions
 forgot-0.222
 bit-0.200
 knew-0.191
 forgotten-0.185
 wasted-0.182
Token.
Token ID13
Feature activation+3.43e+00

Pos logit contributions
he+0.057
room+0.053
ore+0.052
ay+0.052
ings+0.051
Neg logit contributions
 forgot-0.133
 wished-0.113
 knew-0.112
 remembered-0.112
 bit-0.112
Token
Token ID198
Feature activation+3.60e+00

Pos logit contributions
he+0.074
ing+0.070
uggle+0.067
ings+0.066
ipping+0.065
Neg logit contributions
 forgot-0.080
 wished-0.067
 knew-0.067
 remembered-0.063
 learned-0.063
Token
Token ID198
Feature activation+3.25e+00

Pos logit contributions
ing+0.064
ays+0.064
over+0.061
he+0.061
ipping+0.061
Neg logit contributions
 forgot-0.093
 knew-0.082
 wished-0.079
 learned-0.076
 remembered-0.074
Token"
Token ID1
Feature activation+4.63e-01

Pos logit contributions
he+0.046
ore+0.042
ings+0.042
ay+0.042
ays+0.042
Neg logit contributions
 forgot-0.103
 remembered-0.087
 knew-0.086
 wished-0.086
 learned-0.081
Token says
Token ID1139
Feature activation+2.33e+00

Pos logit contributions
 forgot+0.096
 remembered+0.074
 knew+0.072
 wished+0.071
 forgotten+0.067
Neg logit contributions
he-0.106
ing-0.100
ore-0.099
ings-0.098
ays-0.097
Token.
Token ID13
Feature activation+5.57e+00

Pos logit contributions
he+0.042
ay+0.039
ays+0.037
eddy+0.036
uggle+0.036
Neg logit contributions
 forgot-0.064
 knew-0.055
 bit-0.054
 remembered-0.054
 knocked-0.054
Token "
Token ID366
Feature activation+2.98e+00

Pos logit contributions
he+0.135
ach+0.130
uggle+0.127
ays+0.124
ingle+0.123
Neg logit contributions
 forgot-0.100
 knew-0.096
 wished-0.088
 knocked-0.087
 remembered-0.085
TokenIt
Token ID1026
Feature activation-2.26e+00

Pos logit contributions
ings+0.068
ing+0.066
ummy+0.066
ays+0.065
ogging+0.065
Neg logit contributions
 forgot-0.062
 wished-0.054
 knew-0.052
 remembered-0.049
 learned-0.049
Token looks
Token ID3073
Feature activation+1.32e-01

Pos logit contributions
 forgot+0.062
 wished+0.051
 remembered+0.051
 struggled+0.049
 knew+0.049
Neg logit contributions
he-0.039
ing-0.038
ay-0.036
ays-0.036
ore-0.036
Token y
Token ID331
Feature activation+8.34e+00

Pos logit contributions
he+0.003
ay+0.003
ing+0.003
ings+0.003
ore+0.003
Neg logit contributions
 forgot-0.003
 remembered-0.002
 knew-0.002
 wished-0.002
 struggled-0.002
Tokenummy
Token ID13513
Feature activation+4.97e+00

Pos logit contributions
overed+0.104
ute+0.098
he+0.096
uggle+0.092
ane+0.091
Neg logit contributions
 forgot-0.199
 bit-0.191
 knew-0.172
 sp-0.168
 f-0.166
Token!"
Token ID2474
Feature activation+3.90e+00

Pos logit contributions
he+0.087
room+0.082
ore+0.082
uggle+0.081
ings+0.078
Neg logit contributions
 forgot-0.134
 bit-0.114
 remembered-0.111
 knew-0.110
 wished-0.107
Token
Token ID198
Feature activation+4.71e+00

Pos logit contributions
eddy+0.086
he+0.085
ay+0.084
uggle+0.083
ings+0.083
Neg logit contributions
 forgot-0.080
 knew-0.074
 learned-0.070
 wished-0.069
 thought-0.065
Token
Token ID198
Feature activation+4.11e+00

Pos logit contributions
ays+0.105
avored+0.104
over+0.103
ement+0.102
ipping+0.100
Neg logit contributions
 forgot-0.086
 knew-0.083
 learned-0.073
 wished-0.071
 told-0.070
Token"
Token ID1
Feature activation+7.58e-01

Pos logit contributions
he+0.073
ays+0.072
ore+0.071
ay+0.070
ings+0.068
Neg logit contributions
 forgot-0.111
 remembered-0.093
 wished-0.092
 knew-0.091
 learned-0.088
Token cher
Token ID21382
Feature activation+3.18e+00

Pos logit contributions
 forgot+0.083
 knew+0.067
 known+0.066
 remembered+0.066
 punished+0.066
Neg logit contributions
he-0.040
ing-0.040
ays-0.040
 with-0.040
time-0.039
Tokenries
Token ID1678
Feature activation+2.85e+00

Pos logit contributions
he+0.046
room+0.042
ore+0.042
ay+0.040
uggle+0.040
Neg logit contributions
 forgot-0.097
 remembered-0.084
 knew-0.078
 knocked-0.077
 bit-0.077
Token.
Token ID13
Feature activation+3.14e+00

Pos logit contributions
he+0.034
ay+0.034
room+0.032
ore+0.032
ings+0.031
Neg logit contributions
 forgot-0.094
 knew-0.079
 remembered-0.079
 bit-0.078
 wished-0.077
Token It
Token ID632
Feature activation-3.62e+00

Pos logit contributions
he+0.070
ing+0.070
ings+0.066
uggle+0.064
ane+0.064
Neg logit contributions
 forgot-0.069
 wished-0.057
 knew-0.056
 remembered-0.053
 learned-0.053
Token looked
Token ID3114
Feature activation-2.71e-01

Pos logit contributions
 forgot+0.078
 unknow+0.067
 robbers+0.067
 known+0.066
 punished+0.065
Neg logit contributions
he-0.069
 with-0.066
time-0.062
room-0.059
eddy-0.058
Token y
Token ID331
Feature activation+8.33e+00

Pos logit contributions
 forgot+0.006
 remembered+0.004
 knew+0.004
 wished+0.004
 knocked+0.004
Neg logit contributions
he-0.007
ing-0.006
ay-0.006
ore-0.006
ings-0.006
Tokenummy
Token ID13513
Feature activation+4.67e+00

Pos logit contributions
he+0.097
ute+0.094
overed+0.091
ane+0.085
uggle+0.084
Neg logit contributions
 forgot-0.222
 bit-0.198
 knew-0.190
 forgotten-0.183
 wasted-0.182
Token.
Token ID13
Feature activation+3.04e+00

Pos logit contributions
he+0.065
room+0.063
ay+0.060
ore+0.060
uggle+0.058
Neg logit contributions
 forgot-0.144
 bit-0.123
 knew-0.121
 remembered-0.121
 wished-0.121
Token Lily
Token ID20037
Feature activation-6.30e+00

Pos logit contributions
he+0.067
ing+0.066
ings+0.063
uggle+0.061
room+0.059
Neg logit contributions
 forgot-0.068
 wished-0.057
 knew-0.056
 remembered-0.053
 learned-0.052
Token and
Token ID290
Feature activation-1.05e+00

Pos logit contributions
 forgot+0.161
 known+0.152
 stolen+0.142
 forgotten+0.141
 unknow+0.137
Neg logit contributions
he-0.093
 with-0.085
ing-0.080
room-0.078
ore-0.073
Token Ben
Token ID3932
Feature activation-6.73e+00

Pos logit contributions
 forgot+0.030
 wished+0.025
 knew+0.024
 remembered+0.024
 bit+0.023
Neg logit contributions
he-0.017
ing-0.016
ays-0.016
ings-0.016
ay-0.015
Token says
Token ID1139
Feature activation+2.28e+00

Pos logit contributions
 forgot+0.101
 remembered+0.078
 knew+0.076
 wished+0.075
 forgotten+0.072
Neg logit contributions
he-0.104
ing-0.098
ore-0.098
ays-0.097
ings-0.096
Token.
Token ID13
Feature activation+5.38e+00

Pos logit contributions
he+0.039
ay+0.036
ays+0.034
eddy+0.033
uggle+0.033
Neg logit contributions
 forgot-0.066
 knew-0.057
 knocked-0.056
 remembered-0.055
 bit-0.055
Token "
Token ID366
Feature activation+2.88e+00

Pos logit contributions
he+0.142
ach+0.135
ingle+0.132
uggle+0.130
ating+0.129
Neg logit contributions
 forgot-0.085
 knew-0.082
 knocked-0.074
 wished-0.073
 remembered-0.071
TokenIt
Token ID1026
Feature activation-2.25e+00

Pos logit contributions
ings+0.065
ing+0.063
ummy+0.062
ement+0.062
ogging+0.062
Neg logit contributions
 forgot-0.061
 wished-0.052
 knew-0.051
 remembered-0.049
 learned-0.048
Token looks
Token ID3073
Feature activation+2.92e-01

Pos logit contributions
 forgot+0.063
 wished+0.053
 remembered+0.052
 known+0.050
 struggled+0.050
Neg logit contributions
ing-0.036
he-0.036
ays-0.035
ay-0.035
ore-0.034
Token y
Token ID331
Feature activation+8.32e+00

Pos logit contributions
he+0.006
ay+0.006
ing+0.006
ings+0.006
ore+0.006
Neg logit contributions
 forgot-0.007
 remembered-0.005
 knew-0.005
 wished-0.005
 struggled-0.005
Tokenummy
Token ID13513
Feature activation+4.84e+00

Pos logit contributions
overed+0.107
ute+0.102
uggle+0.097
he+0.096
ane+0.094
Neg logit contributions
 forgot-0.193
 bit-0.186
 sp-0.167
 knew-0.165
 f-0.164
Token.
Token ID13
Feature activation+3.42e+00

Pos logit contributions
he+0.080
ore+0.075
room+0.074
uggle+0.074
ings+0.073
Neg logit contributions
 forgot-0.137
 bit-0.114
 remembered-0.113
 knew-0.112
 wished-0.110
Token Let
Token ID3914
Feature activation+9.77e-01

Pos logit contributions
he+0.079
ing+0.073
ings+0.073
ore+0.067
ays+0.067
Neg logit contributions
 forgot-0.075
 knew-0.062
 wished-0.062
 remembered-0.061
 bit-0.057
Token's
Token ID338
Feature activation-2.65e+00

Pos logit contributions
he+0.023
ing+0.022
ay+0.022
ings+0.021
room+0.021
Neg logit contributions
 forgot-0.021
 remembered-0.017
 wished-0.016
 knew-0.016
 struggled-0.015
Token have
Token ID423
Feature activation+8.22e-01

Pos logit contributions
 forgot+0.059
 wished+0.046
 remembered+0.046
 knew+0.045
 struggled+0.042
Neg logit contributions
he-0.062
room-0.057
ay-0.057
ings-0.054
ore-0.054
Token the
Token ID262
Feature activation-5.21e-01

Pos logit contributions
ing+0.043
he+0.043
ore+0.042
ay+0.040
ays+0.039
Neg logit contributions
 forgot-0.040
 wished-0.033
 knew-0.032
 bit-0.031
 remembered-0.031
Token ground
Token ID2323
Feature activation+4.16e+00

Pos logit contributions
 forgot+0.011
 remembered+0.009
 wished+0.008
 knew+0.008
 struggled+0.008
Neg logit contributions
he-0.013
ay-0.012
ing-0.012
ings-0.012
room-0.012
Token.
Token ID13
Feature activation+3.55e+00

Pos logit contributions
ay+0.054
he+0.052
eddy+0.051
uggle+0.048
ing+0.048
Neg logit contributions
 forgot-0.129
 knew-0.111
 wished-0.110
 remembered-0.108
 bit-0.108
Token It
Token ID632
Feature activation-3.20e+00

Pos logit contributions
he+0.073
ing+0.069
over+0.065
eddy+0.064
uggle+0.062
Neg logit contributions
 forgot-0.086
 knew-0.074
 knocked-0.072
 wished-0.071
 remembered-0.067
Token looks
Token ID3073
Feature activation-7.69e-01

Pos logit contributions
 forgot+0.094
 known+0.078
 remembered+0.078
 punished+0.077
 forgotten+0.076
Neg logit contributions
ays-0.046
he-0.045
ore-0.042
ay-0.041
ing-0.041
Token y
Token ID331
Feature activation+8.27e+00

Pos logit contributions
 forgot+0.019
 remembered+0.015
 knew+0.014
 wished+0.014
 struggled+0.014
Neg logit contributions
he-0.015
ay-0.015
ing-0.015
ings-0.014
ore-0.014
Tokenummy
Token ID13513
Feature activation+4.06e+00

Pos logit contributions
he+0.103
overed+0.091
ane+0.088
ore+0.088
ute+0.086
Neg logit contributions
 forgot-0.211
 bit-0.193
 knew-0.177
 forgotten-0.175
 needed-0.173
Token.
Token ID13
Feature activation+3.57e+00

Pos logit contributions
he+0.056
room+0.052
ore+0.050
ay+0.049
ays+0.048
Neg logit contributions
 forgot-0.127
 bit-0.109
 knew-0.107
 wished-0.107
 remembered-0.106
Token
Token ID198
Feature activation+4.70e+00

Pos logit contributions
he+0.068
ing+0.062
over+0.059
uggle+0.059
eddy+0.059
Neg logit contributions
 forgot-0.094
 knew-0.082
 wished-0.079
 knocked-0.077
 remembered-0.075
Token
Token ID198
Feature activation+3.68e+00

Pos logit contributions
ays+0.092
ing+0.090
ipping+0.087
ement+0.086
over+0.085
Neg logit contributions
 forgot-0.113
 knew-0.103
 wished-0.093
 knocked-0.092
 had-0.092
Token"
Token ID1
Feature activation+3.98e-01

Pos logit contributions
he+0.058
ays+0.055
ore+0.054
ing+0.054
ay+0.053
Neg logit contributions
 forgot-0.109
 knew-0.092
 remembered-0.091
 wished-0.090
 bit-0.085
Token cake
Token ID12187
Feature activation+5.34e+00

Pos logit contributions
he+0.014
ing+0.013
ings+0.013
ore+0.013
ays+0.013
Neg logit contributions
 forgot-0.013
 knew-0.010
 wished-0.010
 remembered-0.010
 bit-0.010
Token.
Token ID13
Feature activation+2.87e+00

Pos logit contributions
ore+0.076
he+0.076
overed+0.073
ays+0.072
ay+0.072
Neg logit contributions
 forgot-0.155
 bit-0.129
 knew-0.128
 remembered-0.127
 forget-0.121
Token It
Token ID632
Feature activation-2.04e+00

Pos logit contributions
he+0.063
ing+0.062
ings+0.060
ays+0.055
ore+0.055
Neg logit contributions
 forgot-0.066
 wished-0.056
 knew-0.054
 remembered-0.054
 bit-0.053
Token looks
Token ID3073
Feature activation-6.43e-02

Pos logit contributions
 forgot+0.063
 remembered+0.054
 wished+0.053
 known+0.053
 struggled+0.052
Neg logit contributions
he-0.025
ays-0.024
ore-0.023
ing-0.023
ay-0.022
Token so
Token ID523
Feature activation-5.48e-01

Pos logit contributions
 forgot+0.001
 remembered+0.001
 knew+0.001
 wished+0.001
 struggled+0.001
Neg logit contributions
he-0.001
ay-0.001
ing-0.001
ings-0.001
ore-0.001
Token y
Token ID331
Feature activation+8.25e+00

Pos logit contributions
 forgot+0.011
 remembered+0.009
 wished+0.009
 knew+0.008
 struggled+0.008
Neg logit contributions
he-0.013
ay-0.012
ore-0.012
ing-0.012
ummy-0.012
Tokenummy
Token ID13513
Feature activation+5.73e+00

Pos logit contributions
overed+0.113
he+0.104
ute+0.096
eway+0.094
but+0.094
Neg logit contributions
 forgot-0.186
 bit-0.177
 f-0.162
 knew-0.157
 spoiled-0.157
Token.
Token ID13
Feature activation+2.74e+00

Pos logit contributions
he+0.095
ings+0.090
room+0.089
ay+0.087
uggle+0.084
Neg logit contributions
 forgot-0.161
 knew-0.136
 bit-0.136
 wished-0.135
 remembered-0.134
Token Can
Token ID1680
Feature activation+2.53e+00

Pos logit contributions
he+0.063
ing+0.059
ings+0.059
ays+0.055
ore+0.055
Neg logit contributions
 forgot-0.063
 wished-0.052
 remembered-0.051
 knew-0.051
 bit-0.049
Token we
Token ID356
Feature activation-2.99e+00

Pos logit contributions
he+0.060
ings+0.057
ing+0.057
ay+0.056
ore+0.056
Neg logit contributions
 forgot-0.057
 remembered-0.043
 knew-0.043
 wished-0.043
 bit-0.041
Token have
Token ID423
Feature activation+3.27e-01

Pos logit contributions
 forgot+0.062
 wished+0.050
 remembered+0.048
 knew+0.047
 struggled+0.046
Neg logit contributions
he-0.074
ay-0.068
room-0.066
ore-0.065
uggle-0.065
Token jam
Token ID16853
Feature activation+4.71e+00

Pos logit contributions
he+0.036
ore+0.034
uggle+0.034
ing+0.032
eddy+0.032
Neg logit contributions
 forgot-0.035
 knew-0.028
 remembered-0.028
 bit-0.027
 wished-0.026
Token.
Token ID13
Feature activation+4.27e+00

Pos logit contributions
he+0.057
ay+0.057
uggle+0.057
ore+0.055
room+0.054
Neg logit contributions
 forgot-0.146
 knew-0.124
 remembered-0.123
 wished-0.122
 bit-0.122
Token "
Token ID366
Feature activation+2.07e+00

Pos logit contributions
he+0.090
ing+0.085
uggle+0.082
ings+0.081
ays+0.081
Neg logit contributions
 forgot-0.105
 knew-0.084
 knocked-0.081
 remembered-0.081
 wished-0.078
TokenIt
Token ID1026
Feature activation-2.20e+00

Pos logit contributions
he+0.041
ings+0.040
ing+0.040
ay+0.039
ore+0.039
Neg logit contributions
 forgot-0.051
 wished-0.042
 knew-0.042
 remembered-0.041
 bit-0.040
Token looks
Token ID3073
Feature activation-5.38e-01

Pos logit contributions
 forgot+0.054
 wished+0.045
 remembered+0.043
 knew+0.042
 known+0.042
Neg logit contributions
ing-0.043
he-0.042
ays-0.041
ore-0.040
ay-0.040
Token y
Token ID331
Feature activation+8.25e+00

Pos logit contributions
 forgot+0.012
 remembered+0.009
 knew+0.009
 wished+0.009
 struggled+0.008
Neg logit contributions
he-0.012
ing-0.012
ay-0.011
ings-0.011
ore-0.011
Tokenummy
Token ID13513
Feature activation+5.29e+00

Pos logit contributions
overed+0.107
he+0.105
ute+0.100
uggle+0.096
ane+0.090
Neg logit contributions
 forgot-0.179
 bit-0.172
 packed-0.156
 needed-0.154
 knew-0.153
Token!"
Token ID2474
Feature activation+3.53e+00

Pos logit contributions
he+0.091
room+0.088
uggle+0.085
ore+0.085
ings+0.085
Neg logit contributions
 forgot-0.144
 bit-0.120
 remembered-0.120
 wished-0.118
 knew-0.118
Token
Token ID198
Feature activation+4.52e+00

Pos logit contributions
ays+0.057
ing+0.057
uggle+0.056
he+0.056
ings+0.055
Neg logit contributions
 forgot-0.102
 knew-0.084
 remembered-0.081
 wished-0.080
 learned-0.080
Token
Token ID198
Feature activation+3.54e+00

Pos logit contributions
owers+0.101
ays+0.100
ipping+0.098
ane+0.096
ement+0.095
Neg logit contributions
 forgot-0.092
 knew-0.082
 told-0.072
 learned-0.072
 had-0.071
TokenThey
Token ID2990
Feature activation-7.71e+00

Pos logit contributions
he+0.067
ore+0.066
ays+0.065
ings+0.063
uggle+0.062
Neg logit contributions
 forgot-0.096
 remembered-0.077
 knew-0.076
 wished-0.073
 learned-0.072
Token red
Token ID2266
Feature activation+4.55e+00

Pos logit contributions
 forgot+0.028
 remembered+0.022
 wished+0.022
 knew+0.022
 struggled+0.020
Neg logit contributions
he-0.023
ings-0.022
ay-0.022
ing-0.022
room-0.021
Token and
Token ID290
Feature activation-3.32e+00

Pos logit contributions
he+0.084
ore+0.077
uggle+0.074
ings+0.073
over+0.072
Neg logit contributions
 forgot-0.114
 bit-0.100
 knew-0.095
 remembered-0.093
 wished-0.092
Token shiny
Token ID22441
Feature activation+3.63e+00

Pos logit contributions
 forgot+0.077
 known+0.063
 remembered+0.061
 knew+0.060
 regretted+0.060
Neg logit contributions
ay-0.061
he-0.061
 with-0.060
ings-0.059
ing-0.058
Token and
Token ID290
Feature activation-3.49e+00

Pos logit contributions
he+0.058
uggle+0.054
ore+0.053
room+0.051
ay+0.050
Neg logit contributions
 forgot-0.106
 bit-0.092
 wished-0.091
 remembered-0.090
 knew-0.090
Token looks
Token ID3073
Feature activation-5.63e-01

Pos logit contributions
 forgot+0.073
 known+0.058
 remembered+0.057
 regretted+0.056
 knew+0.055
Neg logit contributions
ay-0.073
he-0.072
 with-0.070
ing-0.070
ings-0.068
Token y
Token ID331
Feature activation+8.24e+00

Pos logit contributions
 forgot+0.013
 remembered+0.011
 knew+0.010
 struggled+0.010
 wished+0.010
Neg logit contributions
he-0.011
ing-0.010
ings-0.010
ay-0.010
 anymore-0.010
Tokenummy
Token ID13513
Feature activation+4.15e+00

Pos logit contributions
he+0.101
overed+0.085
uggle+0.085
ore+0.083
ane+0.082
Neg logit contributions
 forgot-0.225
 bit-0.203
 forgotten-0.192
 knew-0.190
 needed-0.185
Token.
Token ID13
Feature activation+3.29e+00

Pos logit contributions
he+0.054
room+0.049
ore+0.048
ay+0.047
uggle+0.046
Neg logit contributions
 forgot-0.133
 bit-0.114
 wished-0.113
 knew-0.112
 remembered-0.112
Token Ben
Token ID3932
Feature activation-5.29e+00

Pos logit contributions
he+0.079
uggle+0.072
ay+0.072
ing+0.071
ore+0.071
Neg logit contributions
 forgot-0.074
 knew-0.058
 wished-0.057
 remembered-0.057
 knocked-0.052
Token says
Token ID1139
Feature activation+4.06e-01

Pos logit contributions
 forgot+0.172
 known+0.149
 forgotten+0.143
 remembered+0.143
 wished+0.141
Neg logit contributions
ing-0.052
he-0.052
ays-0.050
ore-0.049
ings-0.047
Token,
Token ID11
Feature activation+1.32e+00

Pos logit contributions
he+0.006
ays+0.005
ay+0.005
ings+0.005
ore+0.005
Neg logit contributions
 forgot-0.013
 remembered-0.011
 knew-0.011
 wished-0.011
 bit-0.010
Token Max
Token ID5436
Feature activation-4.90e+00

Pos logit contributions
 forgot+0.069
 wished+0.057
 knew+0.055
 remembered+0.055
 wasted+0.055
Neg logit contributions
he-0.037
ing-0.036
time-0.032
ay-0.032
room-0.032
Token thought
Token ID1807
Feature activation+1.31e+00

Pos logit contributions
 forgot+0.074
 known+0.059
 forget+0.055
 elves+0.055
 wished+0.054
Neg logit contributions
he-0.119
ing-0.119
ings-0.115
time-0.113
 with-0.112
Token it
Token ID340
Feature activation-9.04e-01

Pos logit contributions
he+0.022
ings+0.020
ore+0.020
ing+0.020
ay+0.020
Neg logit contributions
 forgot-0.037
 knew-0.031
 remembered-0.031
 wished-0.030
 knocked-0.029
Token was
Token ID373
Feature activation-1.79e+00

Pos logit contributions
 forgot+0.019
 remembered+0.015
 wished+0.015
 knew+0.014
 forget+0.013
Neg logit contributions
he-0.022
ing-0.021
ings-0.020
ore-0.020
room-0.020
Token too
Token ID1165
Feature activation+4.90e-01

Pos logit contributions
 forgot+0.042
 wished+0.034
 remembered+0.033
 knew+0.032
 struggled+0.031
Neg logit contributions
ings-0.033
time-0.033
ummy-0.033
ing-0.033
 anymore-0.033
Token sm
Token ID895
Feature activation+8.23e+00

Pos logit contributions
he+0.013
ing+0.013
ore+0.013
ay+0.013
ings+0.012
Neg logit contributions
 forgot-0.009
 remembered-0.007
 wished-0.007
 knew-0.006
 knocked-0.006
Tokenelly
Token ID6148
Feature activation+4.96e+00

Pos logit contributions
overed+0.113
ings+0.111
ement+0.107
ute+0.106
ently+0.099
Neg logit contributions
 forgot-0.206
 cared-0.180
 knew-0.172
 bit-0.171
 forgotten-0.166
Token and
Token ID290
Feature activation-4.09e+00

Pos logit contributions
he+0.076
ay+0.072
room+0.070
uggle+0.068
ays+0.066
Neg logit contributions
 forgot-0.144
 knew-0.126
 wished-0.122
 bit-0.121
 remembered-0.120
Token didn
Token ID1422
Feature activation+3.91e+00

Pos logit contributions
 forgot+0.089
 known+0.076
 remembered+0.074
 shown+0.072
 wished+0.072
Neg logit contributions
time-0.071
ing-0.071
he-0.068
 with-0.065
ore-0.063
Token't
Token ID470
Feature activation-1.47e+00

Pos logit contributions
he+0.086
ing+0.086
room+0.084
over+0.083
ays+0.083
Neg logit contributions
 forgot-0.079
 knew-0.064
 knocked-0.062
 cost-0.060
 never-0.059
Token want
Token ID765
Feature activation+2.17e+00

Pos logit contributions
 forgot+0.031
 wished+0.027
 remembered+0.026
 regretted+0.024
 pap+0.024
Neg logit contributions
he-0.032
ore-0.029
ing-0.029
 anymore-0.029
ings-0.028
Token"
Token ID1
Feature activation-1.32e-01

Pos logit contributions
ore+0.044
ings+0.044
he+0.044
ays+0.044
ing+0.042
Neg logit contributions
 forgot-0.081
 remembered-0.067
 knew-0.066
 wished-0.065
 learned-0.063
TokenWow
Token ID22017
Feature activation+3.02e+00

Pos logit contributions
 forgot+0.004
 remembered+0.003
 knew+0.003
 wished+0.003
 knocked+0.003
Neg logit contributions
he-0.002
room-0.002
ay-0.002
ing-0.002
ore-0.002
Token,
Token ID11
Feature activation-4.50e-01

Pos logit contributions
he+0.051
ore+0.049
eddy+0.048
uggle+0.046
ays+0.046
Neg logit contributions
 forgot-0.088
 remembered-0.074
 knew-0.071
 bit-0.068
 knocked-0.067
Token it
Token ID340
Feature activation-1.82e+00

Pos logit contributions
 forgot+0.014
 remembered+0.012
 struggled+0.012
 known+0.012
 knew+0.012
Neg logit contributions
he-0.006
ay-0.005
room-0.005
ing-0.005
ore-0.005
Token looks
Token ID3073
Feature activation+6.87e-02

Pos logit contributions
 forgot+0.045
 wished+0.037
 remembered+0.036
 knew+0.035
 forgotten+0.035
Neg logit contributions
he-0.037
ing-0.035
ay-0.034
room-0.033
ings-0.033
Token y
Token ID331
Feature activation+8.23e+00

Pos logit contributions
he+0.002
ing+0.002
ore+0.002
ay+0.002
ings+0.002
Neg logit contributions
 forgot-0.001
 remembered-0.001
 knew-0.001
 wished-0.001
 bit-0.001
Tokenummy
Token ID13513
Feature activation+5.30e+00

Pos logit contributions
ute+0.094
he+0.094
overed+0.092
ane+0.089
uggle+0.088
Neg logit contributions
 forgot-0.215
 bit-0.199
 knew-0.175
 sp-0.171
 ate-0.170
Token,"
Token ID553
Feature activation+1.19e+00

Pos logit contributions
he+0.090
ore+0.086
room+0.086
uggle+0.083
ays+0.082
Neg logit contributions
 forgot-0.145
 remembered-0.121
 bit-0.120
 knew-0.118
 wished-0.115
Token Amy
Token ID14235
Feature activation-4.55e+00

Pos logit contributions
he+0.017
ay+0.014
ing+0.014
room+0.014
ore+0.014
Neg logit contributions
 forgot-0.037
 remembered-0.031
 knew-0.031
 wished-0.031
 forgotten-0.029
Token said
Token ID531
Feature activation+2.53e+00

Pos logit contributions
 forgot+0.082
 known+0.060
 remembered+0.060
 forgotten+0.057
 wished+0.057
Neg logit contributions
he-0.117
ing-0.112
ore-0.107
room-0.107
 with-0.105
Token.
Token ID13
Feature activation+5.00e+00

Pos logit contributions
he+0.043
ay+0.042
ays+0.040
uggle+0.038
ore+0.037
Neg logit contributions
 forgot-0.071
 knew-0.060
 remembered-0.060
 bit-0.059
 knocked-0.058
Token is
Token ID318
Feature activation+4.63e-01

Pos logit contributions
he+0.010
ore+0.009
ings+0.009
ing+0.009
ays+0.009
Neg logit contributions
 forgot-0.009
 remembered-0.007
 knew-0.007
 wished-0.007
 learned-0.006
Token this
Token ID428
Feature activation-2.17e-01

Pos logit contributions
he+0.009
ay+0.008
ing+0.008
ings+0.008
ore+0.008
Neg logit contributions
 forgot-0.012
 remembered-0.010
 knew-0.010
 wished-0.010
 knocked-0.009
Token?
Token ID30
Feature activation+2.39e+00

Pos logit contributions
 forgot+0.007
 remembered+0.006
 knew+0.006
 wished+0.005
 bit+0.005
Neg logit contributions
he-0.003
ay-0.003
ing-0.003
room-0.003
ore-0.003
Token It
Token ID632
Feature activation-2.14e+00

Pos logit contributions
he+0.046
ing+0.041
ings+0.041
ay+0.040
room+0.038
Neg logit contributions
 forgot-0.063
 remembered-0.052
 wished-0.052
 knew-0.052
 bit-0.050
Token looks
Token ID3073
Feature activation+4.19e-01

Pos logit contributions
 forgot+0.060
 wished+0.050
 remembered+0.049
 known+0.048
 forgotten+0.047
Neg logit contributions
ing-0.035
ore-0.034
ays-0.033
he-0.032
ings-0.031
Token y
Token ID331
Feature activation+8.22e+00

Pos logit contributions
he+0.010
ing+0.010
ay+0.010
ore+0.010
ings+0.009
Neg logit contributions
 forgot-0.009
 remembered-0.007
 knew-0.007
 wished-0.007
 knocked-0.006
Tokenucky
Token ID5309
Feature activation+4.23e+00

Pos logit contributions
he+0.111
overed+0.107
ute+0.095
ane+0.092
uggle+0.090
Neg logit contributions
 forgot-0.195
 bit-0.187
 sp-0.169
 f-0.169
 knew-0.166
Token.
Token ID13
Feature activation+2.52e+00

Pos logit contributions
he+0.071
ay+0.065
room+0.065
uggle+0.062
ings+0.062
Neg logit contributions
 forgot-0.119
 bit-0.099
 remembered-0.098
 knew-0.098
 wished-0.096
Token I
Token ID314
Feature activation-4.11e+00

Pos logit contributions
he+0.051
ing+0.045
ings+0.045
ay+0.043
ore+0.042
Neg logit contributions
 forgot-0.064
 remembered-0.053
 wished-0.053
 knew-0.053
 bit-0.051
Token don
Token ID836
Feature activation+4.43e+00

Pos logit contributions
 forgot+0.086
 pap+0.069
 wished+0.069
 robbers+0.068
 Indians+0.067
Neg logit contributions
he-0.090
ore-0.089
ings-0.077
 with-0.076
uggle-0.074
Token't
Token ID470
Feature activation-2.48e+00

Pos logit contributions
ment+0.110
over+0.107
ings+0.102
oos+0.102
ement+0.102
Neg logit contributions
 forgot-0.069
 knew-0.058
 knocked-0.057
 cost-0.051
 never-0.051
Token red
Token ID2266
Feature activation+6.26e-01

Pos logit contributions
 forgot+0.025
 remembered+0.019
 wished+0.019
 knew+0.019
 learned+0.018
Neg logit contributions
he-0.021
ay-0.020
ings-0.020
ore-0.020
ing-0.019
Token strawberry
Token ID41236
Feature activation+3.16e+00

Pos logit contributions
he+0.015
ays+0.014
ing+0.014
ore+0.014
ings+0.014
Neg logit contributions
 forgot-0.012
 bit-0.010
 knew-0.010
 remembered-0.010
 wished-0.009
Token.
Token ID13
Feature activation+3.66e+00

Pos logit contributions
he+0.042
ay+0.039
room+0.038
ore+0.037
uggle+0.037
Neg logit contributions
 forgot-0.099
 knew-0.085
 bit-0.084
 remembered-0.083
 wished-0.082
Token It
Token ID632
Feature activation-2.96e+00

Pos logit contributions
he+0.075
ing+0.071
ings+0.068
ore+0.067
ently+0.066
Neg logit contributions
 forgot-0.091
 knew-0.072
 remembered-0.072
 wished-0.069
 knocked-0.069
Token looks
Token ID3073
Feature activation-6.09e-01

Pos logit contributions
 forgot+0.083
 wished+0.067
 punished+0.067
 remembered+0.067
 known+0.065
Neg logit contributions
he-0.048
ays-0.046
ay-0.045
ing-0.045
ore-0.045
Token y
Token ID331
Feature activation+8.21e+00

Pos logit contributions
 forgot+0.015
 remembered+0.012
 knew+0.012
 wished+0.012
 struggled+0.011
Neg logit contributions
he-0.012
ay-0.011
ing-0.011
ings-0.011
ore-0.011
Tokenummy
Token ID13513
Feature activation+4.68e+00

Pos logit contributions
he+0.103
overed+0.092
ane+0.091
ute+0.090
ore+0.088
Neg logit contributions
 forgot-0.203
 bit-0.187
 knew-0.171
 forgotten-0.169
 needed-0.167
Token.
Token ID13
Feature activation+3.69e+00

Pos logit contributions
he+0.072
room+0.069
ay+0.066
uggle+0.065
ore+0.064
Neg logit contributions
 forgot-0.137
 bit-0.120
 knew-0.117
 remembered-0.116
 wished-0.115
Token
Token ID198
Feature activation+4.85e+00

Pos logit contributions
he+0.074
ing+0.067
ings+0.067
ore+0.066
ay+0.066
Neg logit contributions
 forgot-0.097
 knew-0.076
 remembered-0.076
 wished-0.075
 learned-0.072
Token
Token ID198
Feature activation+3.76e+00

Pos logit contributions
ays+0.095
ement+0.091
ipping+0.090
ing+0.090
ment+0.089
Neg logit contributions
 forgot-0.120
 knew-0.101
 had-0.098
 learned-0.095
 remembered-0.092
Token"
Token ID1
Feature activation+1.36e-01

Pos logit contributions
he+0.052
ore+0.048
ings+0.047
ing+0.047
ays+0.047
Neg logit contributions
 forgot-0.120
 remembered-0.101
 knew-0.100
 wished-0.099
 bit-0.095
Token on
Token ID319
Feature activation+3.13e+00

Pos logit contributions
ore+0.066
he+0.065
uggle+0.063
room+0.061
ings+0.059
Neg logit contributions
 forgot-0.162
 remembered-0.136
 knew-0.135
 wished-0.129
 bit-0.129
Token it
Token ID340
Feature activation+2.40e+00

Pos logit contributions
ore+0.064
ing+0.063
ogging+0.062
he+0.062
uggle+0.062
Neg logit contributions
 forgot-0.071
 knew-0.059
 remembered-0.057
 wished-0.057
 bit-0.055
Token.
Token ID13
Feature activation+3.46e+00

Pos logit contributions
he+0.027
ays+0.026
ing+0.025
ore+0.025
eddy+0.025
Neg logit contributions
 forgot-0.079
 knew-0.067
 remembered-0.066
 wished-0.066
 bit-0.064
Token It
Token ID632
Feature activation-4.02e+00

Pos logit contributions
he+0.059
ing+0.056
ings+0.055
ore+0.052
ay+0.051
Neg logit contributions
 forgot-0.096
 wished-0.082
 knew-0.081
 remembered-0.080
 learned-0.076
Token looked
Token ID3114
Feature activation-1.06e-01

Pos logit contributions
 forgot+0.091
 robbers+0.082
 punished+0.076
 known+0.076
 stolen+0.075
Neg logit contributions
he-0.073
 with-0.070
time-0.062
room-0.062
ays-0.060
Token y
Token ID331
Feature activation+8.20e+00

Pos logit contributions
 forgot+0.003
 remembered+0.002
 knew+0.002
 wished+0.002
 knocked+0.002
Neg logit contributions
he-0.002
ing-0.002
ay-0.002
ore-0.002
ings-0.002
Tokenummy
Token ID13513
Feature activation+3.21e+00

Pos logit contributions
he+0.102
ute+0.095
ore+0.093
overed+0.091
uggle+0.089
Neg logit contributions
 forgot-0.205
 bit-0.194
 knew-0.175
 f-0.171
 mixed-0.171
Token and
Token ID290
Feature activation-3.33e+00

Pos logit contributions
he+0.046
room+0.041
ore+0.041
ays+0.040
ings+0.040
Neg logit contributions
 forgot-0.099
 knew-0.083
 remembered-0.083
 wished-0.082
 bit-0.082
Token sweet
Token ID6029
Feature activation+2.07e+00

Pos logit contributions
 forgot+0.065
 remembered+0.052
 wished+0.051
 knew+0.050
 struggled+0.050
Neg logit contributions
he-0.073
ing-0.071
 with-0.068
ay-0.068
time-0.067
Token.
Token ID13
Feature activation+3.31e+00

Pos logit contributions
he+0.031
ore+0.028
room+0.027
ays+0.027
ay+0.027
Neg logit contributions
 forgot-0.064
 knew-0.053
 remembered-0.053
 wished-0.053
 bit-0.052
Token
Token ID198
Feature activation+4.29e+00

Pos logit contributions
he+0.057
ings+0.052
ing+0.052
ay+0.049
room+0.049
Neg logit contributions
 forgot-0.093
 wished-0.080
 knew-0.079
 remembered-0.078
 learned-0.073
Token the
Token ID262
Feature activation-1.70e+00

Pos logit contributions
he+0.047
ing+0.042
ings+0.042
room+0.042
ore+0.042
Neg logit contributions
 forgot-0.058
 remembered-0.047
 knew-0.047
 wished-0.046
 bit-0.044
Token apple
Token ID17180
Feature activation+5.50e+00

Pos logit contributions
 forgot+0.043
 remembered+0.034
 knew+0.034
 wished+0.034
 bit+0.032
Neg logit contributions
he-0.035
ore-0.033
ing-0.032
ings-0.032
room-0.032
Token.
Token ID13
Feature activation+2.06e+00

Pos logit contributions
ement+0.094
overed+0.085
he+0.084
over+0.082
ment+0.082
Neg logit contributions
 forgot-0.144
 bit-0.128
 knew-0.118
 knocked-0.113
 remembered-0.113
Token It
Token ID632
Feature activation-2.02e+00

Pos logit contributions
he+0.037
ing+0.035
ings+0.034
ore+0.032
ogging+0.030
Neg logit contributions
 forgot-0.056
 remembered-0.047
 bit-0.047
 wished-0.046
 knew-0.046
Token looks
Token ID3073
Feature activation+2.38e-02

Pos logit contributions
 forgot+0.053
 wished+0.044
 remembered+0.044
 known+0.043
 forgotten+0.042
Neg logit contributions
he-0.035
ing-0.034
ays-0.033
ings-0.033
ore-0.032
Token y
Token ID331
Feature activation+8.20e+00

Pos logit contributions
he+0.001
ay+0.001
ing+0.001
ings+0.001
ore+0.001
Neg logit contributions
 forgot-0.000
 remembered-0.000
 knew-0.000
 wished-0.000
 struggled-0.000
Tokenummy
Token ID13513
Feature activation+5.55e+00

Pos logit contributions
overed+0.106
he+0.101
ute+0.089
eway+0.088
ane+0.086
Neg logit contributions
 forgot-0.196
 bit-0.187
 packed-0.166
 knew-0.164
 needed-0.162
Token."
Token ID526
Feature activation+4.28e+00

Pos logit contributions
he+0.105
room+0.103
ings+0.101
ay+0.097
ore+0.097
Neg logit contributions
 forgot-0.143
 bit-0.120
 remembered-0.115
 knew-0.115
 wished-0.115
Token
Token ID198
Feature activation+5.18e+00

Pos logit contributions
he+0.122
ay+0.117
ane+0.114
ing+0.114
eddy+0.114
Neg logit contributions
 forgot-0.076
 knew-0.052
 remembered-0.051
 learned-0.049
 tried-0.045
Token
Token ID198
Feature activation+3.94e+00

Pos logit contributions
ement+0.106
ays+0.105
overed+0.104
ment+0.102
ane+0.101
Neg logit contributions
 forgot-0.121
 knew-0.101
 learned-0.092
 remembered-0.091
 told-0.091
TokenBen
Token ID11696
Feature activation-5.22e+00

Pos logit contributions
he+0.085
ore+0.084
ays+0.082
ings+0.081
ay+0.080
Neg logit contributions
 forgot-0.096
 remembered-0.075
 knew-0.073
 wished-0.070
 bit-0.069
Token table
Token ID3084
Feature activation+4.04e+00

Pos logit contributions
 forgot+0.034
 remembered+0.026
 wished+0.025
 knew+0.025
 struggled+0.023
Neg logit contributions
he-0.037
ay-0.034
ing-0.034
room-0.034
ore-0.034
Token.
Token ID13
Feature activation+2.98e+00

Pos logit contributions
he+0.043
ay+0.041
ing+0.039
eddy+0.038
uggle+0.038
Neg logit contributions
 forgot-0.136
 knew-0.117
 remembered-0.116
 wished-0.116
 bit-0.113
Token It
Token ID632
Feature activation-3.82e+00

Pos logit contributions
he+0.054
ing+0.052
ings+0.050
ore+0.049
eddy+0.049
Neg logit contributions
 forgot-0.080
 knew-0.066
 wished-0.064
 remembered-0.064
 knocked-0.063
Token looks
Token ID3073
Feature activation-1.15e+00

Pos logit contributions
 forgot+0.105
 known+0.086
 remembered+0.085
 wished+0.083
 punished+0.082
Neg logit contributions
ays-0.061
he-0.059
ore-0.058
ay-0.056
ing-0.055
Token very
Token ID845
Feature activation-6.75e-01

Pos logit contributions
 forgot+0.027
 remembered+0.020
 knew+0.020
 wished+0.020
 struggled+0.019
Neg logit contributions
he-0.023
ing-0.022
ay-0.022
ings-0.022
ore-0.022
Token y
Token ID331
Feature activation+8.19e+00

Pos logit contributions
 forgot+0.013
 remembered+0.010
 wished+0.009
 knew+0.009
 knocked+0.008
Neg logit contributions
he-0.018
ing-0.017
ay-0.017
ore-0.017
ings-0.017
Tokenummy
Token ID13513
Feature activation+4.34e+00

Pos logit contributions
he+0.102
ute+0.088
overed+0.088
uggle+0.086
ane+0.085
Neg logit contributions
 forgot-0.214
 bit-0.196
 knew-0.183
 forgotten-0.180
 needed-0.175
Token and
Token ID290
Feature activation-4.73e+00

Pos logit contributions
he+0.062
room+0.057
ay+0.055
uggle+0.054
ore+0.054
Neg logit contributions
 forgot-0.134
 bit-0.115
 knew-0.114
 wished-0.114
 remembered-0.112
Token fancy
Token ID14996
Feature activation+2.20e+00

Pos logit contributions
 forgot+0.099
 known+0.082
 remembered+0.079
 knew+0.077
 regretted+0.076
Neg logit contributions
 with-0.092
he-0.089
ummy-0.088
ings-0.088
ing-0.087
Token.
Token ID13
Feature activation+3.23e+00

Pos logit contributions
he+0.027
ay+0.023
ore+0.022
uggle+0.022
room+0.021
Neg logit contributions
 forgot-0.073
 knew-0.062
 remembered-0.062
 wished-0.062
 bit-0.062
Token It
Token ID632
Feature activation-4.55e+00

Pos logit contributions
he+0.063
ing+0.060
ore+0.058
ings+0.058
ay+0.057
Neg logit contributions
 forgot-0.085
 knew-0.068
 wished-0.067
 remembered-0.066
 knocked-0.065
Token eat
Token ID4483
Feature activation+4.58e+00

Pos logit contributions
 forgot+0.049
 wished+0.037
 remembered+0.037
 knew+0.036
 struggled+0.033
Neg logit contributions
he-0.062
ay-0.058
ore-0.057
room-0.056
ings-0.055
Token it
Token ID340
Feature activation+3.34e+00

Pos logit contributions
he+0.106
ay+0.102
ings+0.101
room+0.101
ays+0.099
Neg logit contributions
 forgot-0.090
 bit-0.073
 knew-0.072
 remembered-0.071
 wished-0.069
Token?
Token ID30
Feature activation+2.53e+00

Pos logit contributions
ays+0.061
ings+0.060
ore+0.058
aps+0.058
over+0.058
Neg logit contributions
 forgot-0.083
 remembered-0.069
 knew-0.068
 reminded-0.066
 bit-0.066
Token It
Token ID632
Feature activation-1.66e+00

Pos logit contributions
he+0.050
ings+0.050
ing+0.049
ays+0.047
room+0.044
Neg logit contributions
 forgot-0.062
 knew-0.051
 remembered-0.051
 wished-0.051
 bit-0.049
Token looks
Token ID3073
Feature activation+2.00e-01

Pos logit contributions
 forgot+0.043
 known+0.037
 wished+0.036
 forgotten+0.036
 robbers+0.036
Neg logit contributions
ing-0.025
he-0.024
 with-0.024
ore-0.022
time-0.022
Token y
Token ID331
Feature activation+8.17e+00

Pos logit contributions
he+0.004
ay+0.004
ing+0.004
ings+0.004
ore+0.004
Neg logit contributions
 forgot-0.005
 remembered-0.004
 knew-0.003
 wished-0.003
 struggled-0.003
Tokenummy
Token ID13513
Feature activation+4.66e+00

Pos logit contributions
overed+0.106
ute+0.098
he+0.095
ane+0.089
ment+0.088
Neg logit contributions
 forgot-0.181
 bit-0.180
 packed-0.161
 knew-0.156
 needed-0.155
Token."
Token ID526
Feature activation+3.88e+00

Pos logit contributions
he+0.078
room+0.076
ings+0.075
ore+0.072
ay+0.072
Neg logit contributions
 forgot-0.129
 bit-0.108
 remembered-0.108
 knew-0.106
 wished-0.105
Token
Token ID198
Feature activation+4.62e+00

Pos logit contributions
he+0.075
ing+0.070
ays+0.068
ay+0.067
ings+0.066
Neg logit contributions
 forgot-0.102
 knew-0.087
 remembered-0.083
 wished-0.083
 knocked-0.079
Token
Token ID198
Feature activation+3.62e+00

Pos logit contributions
ays+0.091
ing+0.084
owers+0.079
ement+0.078
over+0.078
Neg logit contributions
 forgot-0.117
 knew-0.107
 remembered-0.099
 learned-0.094
 wished-0.093
TokenHis
Token ID6653
Feature activation-1.39e+00

Pos logit contributions
ays+0.068
he+0.068
ing+0.066
ore+0.064
ings+0.064
Neg logit contributions
 forgot-0.096
 knew-0.081
 remembered-0.080
 wished-0.077
 bit-0.073
Token
Token ID198
Feature activation+4.15e+00

Pos logit contributions
ing+0.068
ings+0.066
he+0.066
ment+0.065
ore+0.064
Neg logit contributions
 forgot-0.078
 knew-0.067
 learned-0.064
 wished-0.064
 remembered-0.062
Token
Token ID198
Feature activation+3.07e+00

Pos logit contributions
ays+0.080
ment+0.078
ing+0.078
ings+0.075
ipping+0.073
Neg logit contributions
 forgot-0.097
 knew-0.086
 had-0.082
 wished-0.081
 learned-0.081
TokenEvery
Token ID6109
Feature activation-1.62e+00

Pos logit contributions
he+0.061
ings+0.060
ore+0.060
ays+0.059
ing+0.059
Neg logit contributions
 forgot-0.076
 remembered-0.062
 wished-0.061
 knew-0.061
 bit-0.057
Token night
Token ID1755
Feature activation+3.46e+00

Pos logit contributions
 forgot+0.045
 remembered+0.035
 wished+0.035
 knew+0.034
 wasted+0.034
Neg logit contributions
he-0.030
ore-0.028
time-0.028
ay-0.028
ing-0.028
Token,
Token ID11
Feature activation+9.93e-01

Pos logit contributions
he+0.042
ore+0.036
ay+0.036
ing+0.036
ays+0.036
Neg logit contributions
 forgot-0.117
 remembered-0.099
 knew-0.099
 wished-0.097
 knocked-0.094
Token she
Token ID673
Feature activation-9.46e+00

Pos logit contributions
he+0.018
ing+0.017
ay+0.017
ings+0.017
room+0.016
Neg logit contributions
 forgot-0.027
 remembered-0.022
 knew-0.022
 wished-0.022
 struggled-0.020
Token would
Token ID561
Feature activation-3.45e+00

Pos logit contributions
 forgot+0.156
 stolen+0.130
 unknow+0.120
 known+0.120
 robbers+0.118
Neg logit contributions
he-0.211
room-0.194
ore-0.188
 with-0.188
ing-0.188
Token read
Token ID1100
Feature activation+3.40e+00

Pos logit contributions
 forgot+0.076
 remembered+0.058
 wished+0.057
 knew+0.055
 regretted+0.055
Neg logit contributions
he-0.079
room-0.072
ings-0.070
uggle-0.069
ore-0.069
Token a
Token ID257
Feature activation-1.14e+00

Pos logit contributions
he+0.071
ays+0.069
ore+0.069
ing+0.066
ings+0.066
Neg logit contributions
 forgot-0.077
 knew-0.065
 remembered-0.063
 wished-0.063
 learned-0.060
Token book
Token ID1492
Feature activation+4.65e+00

Pos logit contributions
 forgot+0.029
 remembered+0.024
 wished+0.023
 knew+0.023
 knocked+0.022
Neg logit contributions
he-0.023
ay-0.021
ore-0.021
ing-0.021
ings-0.021
Token that
Token ID326
Feature activation-3.42e+00

Pos logit contributions
ays+0.077
ings+0.072
eddy+0.071
ore+0.070
ach+0.069
Neg logit contributions
 forgot-0.124
 knew-0.108
 bit-0.102
 learned-0.102
 remembered-0.102
Token her
Token ID607
Feature activation-2.70e+00

Pos logit contributions
 forgot+0.020
 knew+0.016
 wished+0.016
 remembered+0.015
 bit+0.015
Neg logit contributions
he-0.016
ing-0.015
room-0.015
time-0.015
ay-0.015
Token av
Token ID1196
Feature activation+3.66e+00

Pos logit contributions
 forgot+0.066
 wished+0.053
 remembered+0.052
 knew+0.051
 knocked+0.048
Neg logit contributions
he-0.055
ay-0.052
ings-0.051
ing-0.051
ummy-0.050
Tokenoc
Token ID420
Feature activation+3.97e+00

Pos logit contributions
he+0.058
ay+0.054
ing+0.053
ings+0.052
ore+0.052
Neg logit contributions
 forgot-0.108
 remembered-0.090
 wished-0.089
 knew-0.089
 knocked-0.085
Tokenados
Token ID22484
Feature activation+2.86e+00

Pos logit contributions
he+0.049
ore+0.042
ay+0.039
ing+0.039
room+0.038
Neg logit contributions
 forgot-0.130
 remembered-0.113
 knew-0.112
 wished-0.110
 learned-0.105
Token.
Token ID13
Feature activation+2.26e+00

Pos logit contributions
he+0.034
ay+0.032
ore+0.032
ings+0.032
ays+0.031
Neg logit contributions
 forgot-0.095
 remembered-0.081
 knew-0.080
 wished-0.079
 bit-0.076
Token She
Token ID1375
Feature activation-9.33e+00

Pos logit contributions
he+0.032
ing+0.028
ore+0.028
ings+0.028
ay+0.027
Neg logit contributions
 forgot-0.073
 knew-0.061
 wished-0.060
 remembered-0.060
 missed-0.058
Token thanked
Token ID26280
Feature activation+1.23e+00

Pos logit contributions
chie+0.157
 mate+0.146
 robbers+0.139
 supervision+0.134
 brav+0.134
Neg logit contributions
 with-0.188
 and-0.179
he-0.168
 at-0.165
,-0.165
Token her
Token ID607
Feature activation-2.52e-01

Pos logit contributions
he+0.021
ing+0.019
ay+0.019
ore+0.019
ings+0.019
Neg logit contributions
 forgot-0.036
 remembered-0.029
 wished-0.029
 knew-0.029
 bit-0.027
Token mom
Token ID1995
Feature activation+2.04e+00

Pos logit contributions
 forgot+0.006
 remembered+0.004
 wished+0.004
 knew+0.004
 struggled+0.004
Neg logit contributions
he-0.006
ing-0.005
ay-0.005
ings-0.005
ore-0.005
Token and
Token ID290
Feature activation-5.85e+00

Pos logit contributions
he+0.038
ings+0.037
ays+0.037
ore+0.036
room+0.036
Neg logit contributions
 forgot-0.053
 knew-0.045
 remembered-0.044
 learned-0.041
 wished-0.041
Token ate
Token ID15063
Feature activation+2.60e+00

Pos logit contributions
 forgot+0.103
chie+0.090
'd+0.088
 fact+0.087
 smallest+0.087
Neg logit contributions
he-0.117
 with-0.117
ing-0.110
 at-0.104
 together-0.103
Token difficult
Token ID2408
Feature activation+3.46e+00

Pos logit contributions
 forgot+0.053
 wished+0.044
 remembered+0.042
 knew+0.041
 cared+0.037
Neg logit contributions
he-0.058
ing-0.057
ings-0.055
time-0.055
ay-0.053
Token.
Token ID13
Feature activation+2.62e+00

Pos logit contributions
he+0.041
ay+0.034
ore+0.034
ings+0.033
room+0.033
Neg logit contributions
 forgot-0.118
 knew-0.100
 remembered-0.100
 wished-0.100
 knocked-0.095
Token Each
Token ID5501
Feature activation-2.28e+00

Pos logit contributions
he+0.047
ing+0.043
ings+0.043
ore+0.041
ays+0.040
Neg logit contributions
 forgot-0.069
 knew-0.060
 wished-0.058
 remembered-0.057
 learned-0.056
Token time
Token ID640
Feature activation+1.08e+00

Pos logit contributions
 forgot+0.060
 remembered+0.048
 wished+0.046
 knew+0.046
 knocked+0.043
Neg logit contributions
he-0.045
ore-0.043
ing-0.043
ay-0.042
ings-0.042
Token,
Token ID11
Feature activation-7.36e-01

Pos logit contributions
he+0.017
ing+0.016
ings+0.015
ay+0.015
ore+0.015
Neg logit contributions
 forgot-0.032
 remembered-0.026
 knew-0.026
 wished-0.026
 bit-0.024
Token she
Token ID673
Feature activation-9.30e+00

Pos logit contributions
 forgot+0.022
 remembered+0.018
 wished+0.018
 knew+0.018
 wasted+0.017
Neg logit contributions
he-0.011
ing-0.010
ings-0.010
room-0.010
ay-0.010
Token was
Token ID373
Feature activation-3.26e+00

Pos logit contributions
 forgot+0.121
 pap+0.111
 stolen+0.111
 robbers+0.106
 unknow+0.105
Neg logit contributions
he-0.229
room-0.208
ing-0.207
 with-0.204
 at-0.198
Token happy
Token ID3772
Feature activation+7.13e-01

Pos logit contributions
 forgot+0.061
 wished+0.051
 knew+0.045
 regretted+0.044
 cost+0.044
Neg logit contributions
ing-0.073
he-0.072
 with-0.070
time-0.069
apping-0.067
Token when
Token ID618
Feature activation-6.01e-01

Pos logit contributions
he+0.010
ay+0.009
ore+0.009
ings+0.009
ays+0.009
Neg logit contributions
 forgot-0.023
 knew-0.019
 remembered-0.019
 wished-0.019
 bit-0.018
Token she
Token ID673
Feature activation-7.26e+00

Pos logit contributions
 forgot+0.017
 remembered+0.014
 wished+0.014
 knew+0.014
 bit+0.013
Neg logit contributions
he-0.010
ing-0.009
ings-0.009
ore-0.009
ay-0.009
Token saw
Token ID2497
Feature activation-1.55e+00

Pos logit contributions
 forgot+0.100
 pap+0.078
 wished+0.074
 known+0.068
 robbers+0.068
Neg logit contributions
he-0.199
ing-0.183
room-0.181
ore-0.178
time-0.175
Token it
Token ID340
Feature activation+2.10e+00

Pos logit contributions
he+0.004
ing+0.004
ings+0.004
ore+0.004
ay+0.004
Neg logit contributions
 forgot-0.007
 remembered-0.006
 knew-0.006
 wished-0.006
 bit-0.006
Token.
Token ID13
Feature activation+3.16e+00

Pos logit contributions
he+0.032
ays+0.031
ay+0.031
ore+0.030
room+0.030
Neg logit contributions
 forgot-0.063
 knew-0.053
 remembered-0.052
 bit-0.051
 wished-0.051
Token At
Token ID1629
Feature activation-9.90e-01

Pos logit contributions
he+0.056
ings+0.054
ing+0.053
room+0.052
ays+0.052
Neg logit contributions
 forgot-0.086
 knew-0.075
 wished-0.072
 remembered-0.071
 learned-0.070
Token the
Token ID262
Feature activation-3.11e+00

Pos logit contributions
 forgot+0.022
 remembered+0.017
 knew+0.017
 wished+0.017
 knocked+0.016
Neg logit contributions
he-0.022
ay-0.022
ings-0.021
ing-0.021
room-0.021
Token end
Token ID886
Feature activation+2.42e+00

Pos logit contributions
 forgot+0.085
 wished+0.070
 remembered+0.068
 struggled+0.067
 wasted+0.066
Neg logit contributions
he-0.056
time-0.052
ay-0.050
ummy-0.050
ing-0.049
Token she
Token ID673
Feature activation-9.30e+00

Pos logit contributions
he+0.038
ays+0.036
room+0.032
ikes+0.032
eddy+0.031
Neg logit contributions
 forgot-0.071
 knew-0.061
 remembered-0.059
 bit-0.056
 learned-0.056
Token was
Token ID373
Feature activation-2.60e+00

Pos logit contributions
 robbers+0.185
 forgot+0.166
 mate+0.163
 Indians+0.163
 sizes+0.154
Neg logit contributions
he-0.195
 with-0.172
uck-0.166
ore-0.162
time-0.162
Token happy
Token ID3772
Feature activation+1.00e+00

Pos logit contributions
 forgot+0.055
 wished+0.044
 remembered+0.042
 robbers+0.041
 knew+0.041
Neg logit contributions
he-0.059
ing-0.051
ings-0.050
time-0.050
over-0.050
Token and
Token ID290
Feature activation-4.60e+00

Pos logit contributions
he+0.015
ay+0.014
room+0.013
ays+0.013
ings+0.013
Neg logit contributions
 forgot-0.031
 knew-0.026
 remembered-0.026
 wished-0.026
 bit-0.025
Token proud
Token ID6613
Feature activation+1.73e+00

Pos logit contributions
 forgot+0.077
 robbers+0.065
 remembered+0.059
 mate+0.059
 knocked+0.058
Neg logit contributions
he-0.108
ing-0.104
 with-0.103
uck-0.100
uggle-0.099
Token of
Token ID286
Feature activation+8.63e-01

Pos logit contributions
he+0.025
room+0.021
ings+0.021
ays+0.021
ay+0.021
Neg logit contributions
 forgot-0.054
 knew-0.047
 remembered-0.046
 wished-0.046
 bit-0.044
Token was
Token ID373
Feature activation-2.50e+00

Pos logit contributions
 forgot+0.130
 setup+0.118
 dishonest+0.117
 error+0.116
chie+0.116
Neg logit contributions
 with-0.165
he-0.158
.-0.141
room-0.141
,-0.138
Token happy
Token ID3772
Feature activation+9.57e-01

Pos logit contributions
 forgot+0.058
 wished+0.047
 knew+0.045
 regretted+0.044
 remembered+0.044
Neg logit contributions
he-0.051
 with-0.047
ing-0.044
time-0.044
aps-0.044
Token and
Token ID290
Feature activation-4.71e+00

Pos logit contributions
he+0.016
ay+0.014
ings+0.014
room+0.014
ing+0.014
Neg logit contributions
 forgot-0.029
 knew-0.024
 remembered-0.024
 wished-0.024
 bit-0.022
Token tired
Token ID10032
Feature activation+3.44e+00

Pos logit contributions
 forgot+0.074
 wasted+0.062
 nurses+0.060
 punished+0.060
 setup+0.060
Neg logit contributions
 with-0.110
he-0.105
ing-0.100
uggle-0.097
time-0.097
Token.
Token ID13
Feature activation+2.26e+00

Pos logit contributions
ays+0.049
ay+0.047
he+0.045
ings+0.045
eddy+0.043
Neg logit contributions
 forgot-0.105
 knew-0.091
 bit-0.089
 wished-0.088
 remembered-0.085
Token He
Token ID679
Feature activation-9.28e+00

Pos logit contributions
he+0.038
ing+0.035
ings+0.035
ay+0.034
ore+0.033
Neg logit contributions
 forgot-0.066
 knew-0.055
 remembered-0.054
 wished-0.054
 learned-0.051
Token curled
Token ID41108
Feature activation+2.45e+00

Pos logit contributions
 forgot+0.135
chie+0.134
 known+0.127
 fact+0.126
 error+0.124
Neg logit contributions
 with-0.189
he-0.182
.-0.173
 again-0.173
 and-0.172
Token up
Token ID510
Feature activation+1.56e+00

Pos logit contributions
ay+0.036
he+0.035
room+0.032
ore+0.031
ummy+0.030
Neg logit contributions
 forgot-0.076
 remembered-0.067
 wished-0.065
 knew-0.064
 bit-0.063
Token in
Token ID287
Feature activation+4.98e-01

Pos logit contributions
he+0.023
ay+0.023
room+0.022
ays+0.021
ings+0.021
Neg logit contributions
 forgot-0.047
 remembered-0.040
 knew-0.040
 wished-0.039
 bit-0.039
Token his
Token ID465
Feature activation-2.09e+00

Pos logit contributions
he+0.009
ay+0.009
ing+0.009
ore+0.009
ings+0.009
Neg logit contributions
 forgot-0.013
 remembered-0.011
 knew-0.011
 wished-0.011
 knocked-0.010
Token cozy
Token ID37438
Feature activation-1.16e+00

Pos logit contributions
 forgot+0.048
 remembered+0.038
 wished+0.037
 knew+0.036
 known+0.034
Neg logit contributions
he-0.046
ay-0.044
ings-0.043
ore-0.042
ing-0.042
Token she
Token ID673
Feature activation-6.73e+00

Pos logit contributions
 forgot+0.025
 remembered+0.021
 wished+0.021
 knew+0.021
 bit+0.021
Neg logit contributions
he-0.012
ing-0.011
time-0.011
ings-0.011
room-0.010
Token did
Token ID750
Feature activation-9.77e-02

Pos logit contributions
 forgot+0.105
 pap+0.076
 wished+0.071
 forget+0.071
 remembered+0.069
Neg logit contributions
he-0.190
room-0.169
ay-0.164
ore-0.162
time-0.161
Token good
Token ID922
Feature activation+1.42e+00

Pos logit contributions
 forgot+0.002
 remembered+0.002
 knew+0.002
 wished+0.002
 knocked+0.002
Neg logit contributions
he-0.002
ay-0.002
ing-0.002
ore-0.002
ings-0.002
Token things
Token ID1243
Feature activation+5.82e-01

Pos logit contributions
he+0.036
room+0.036
ore+0.035
ays+0.035
ay+0.034
Neg logit contributions
 forgot-0.026
 wished-0.020
 knew-0.020
 remembered-0.020
 learned-0.019
Token.
Token ID13
Feature activation+2.26e+00

Pos logit contributions
he+0.006
ay+0.006
ore+0.006
room+0.006
ays+0.006
Neg logit contributions
 forgot-0.020
 knew-0.017
 remembered-0.017
 wished-0.017
 knocked-0.016
Token She
Token ID1375
Feature activation-9.28e+00

Pos logit contributions
he+0.035
ing+0.033
ore+0.032
ings+0.032
ay+0.030
Neg logit contributions
 forgot-0.068
 knew-0.057
 wished-0.057
 remembered-0.055
 learned-0.054
Token decided
Token ID3066
Feature activation-3.80e-01

Pos logit contributions
chie+0.144
 Indians+0.131
 pap+0.130
 robbers+0.127
 mate+0.123
Neg logit contributions
he-0.202
 with-0.198
 at-0.177
room-0.172
 and-0.170
Token to
Token ID284
Feature activation-4.57e+00

Pos logit contributions
 forgot+0.012
 remembered+0.010
 knew+0.010
 wished+0.010
 bit+0.009
Neg logit contributions
he-0.005
ing-0.005
ings-0.005
ore-0.005
ay-0.004
Token sell
Token ID3677
Feature activation+2.80e+00

Pos logit contributions
 forgot+0.102
 remembered+0.082
 wished+0.079
 knew+0.077
 struggled+0.073
Neg logit contributions
he-0.110
room-0.101
ay-0.098
uggle-0.097
ings-0.097
Token some
Token ID617
Feature activation-5.46e-01

Pos logit contributions
he+0.053
ore+0.050
ing+0.049
ay+0.049
ings+0.048
Neg logit contributions
 forgot-0.073
 remembered-0.059
 knew-0.059
 wished-0.059
 bit-0.055
Token of
Token ID286
Feature activation-1.26e+00

Pos logit contributions
 forgot+0.012
 knew+0.009
 wished+0.009
 remembered+0.009
 bit+0.008
Neg logit contributions
he-0.013
ore-0.012
ing-0.012
ays-0.012
room-0.012
Token cream
Token ID8566
Feature activation+3.52e+00

Pos logit contributions
 forgot+0.014
 knew+0.011
 remembered+0.011
 wished+0.011
 bit+0.010
Neg logit contributions
he-0.011
ing-0.011
ay-0.011
ore-0.010
ays-0.010
Token.
Token ID13
Feature activation+3.05e+00

Pos logit contributions
he+0.072
ay+0.071
ing+0.070
ore+0.070
ays+0.068
Neg logit contributions
 forgot-0.080
 knew-0.064
 remembered-0.063
 bit-0.063
 wished-0.062
Token Every
Token ID3887
Feature activation-1.24e+00

Pos logit contributions
ing+0.065
ore+0.062
he+0.062
ings+0.059
room+0.058
Neg logit contributions
 forgot-0.071
 knew-0.058
 learned-0.056
 remembered-0.055
 wished-0.053
Token night
Token ID1755
Feature activation+2.76e+00

Pos logit contributions
 forgot+0.032
 remembered+0.026
 wished+0.025
 knew+0.025
 knocked+0.023
Neg logit contributions
he-0.025
ore-0.023
ing-0.023
ay-0.023
ings-0.023
Token,
Token ID11
Feature activation+4.43e-01

Pos logit contributions
he+0.037
ay+0.032
ays+0.032
room+0.032
ore+0.031
Neg logit contributions
 forgot-0.092
 remembered-0.077
 knew-0.077
 wished-0.075
 knocked-0.072
Token she
Token ID673
Feature activation-9.25e+00

Pos logit contributions
he+0.009
ore+0.009
ay+0.009
ings+0.008
room+0.008
Neg logit contributions
 forgot-0.012
 remembered-0.009
 knew-0.009
 wished-0.009
 learned-0.009
Token would
Token ID561
Feature activation-3.96e+00

Pos logit contributions
 forgot+0.157
 robbers+0.134
 stolen+0.131
chie+0.130
 unknow+0.130
Neg logit contributions
he-0.201
time-0.178
ore-0.177
 with-0.176
ing-0.172
Token stand
Token ID1302
Feature activation+2.81e+00

Pos logit contributions
 forgot+0.088
 remembered+0.071
 wished+0.068
 knew+0.065
 regretted+0.064
Neg logit contributions
he-0.091
uggle-0.078
room-0.078
ore-0.077
time-0.076
Token in
Token ID287
Feature activation+2.92e-01

Pos logit contributions
ay+0.038
he+0.037
room+0.036
ings+0.034
ore+0.034
Neg logit contributions
 forgot-0.088
 remembered-0.078
 knew-0.075
 wished-0.073
 knocked-0.073
Token front
Token ID2166
Feature activation+6.37e-01

Pos logit contributions
he+0.005
ay+0.005
ings+0.005
ing+0.005
ore+0.005
Neg logit contributions
 forgot-0.008
 remembered-0.007
 knew-0.007
 wished-0.007
 knocked-0.006
Token of
Token ID286
Feature activation-6.46e-01

Pos logit contributions
he+0.008
ings+0.008
ing+0.007
ore+0.007
room+0.007
Neg logit contributions
 forgot-0.019
 knew-0.017
 wished-0.017
 remembered-0.017
 bit-0.016
Token played
Token ID2826
Feature activation+4.24e+00

Pos logit contributions
 forgot+0.089
 mistake+0.082
 importance+0.081
 regret+0.080
 enemy+0.080
Neg logit contributions
 with-0.108
 together-0.103
 down-0.102
 on-0.101
 at-0.098
Token and
Token ID290
Feature activation-4.39e+00

Pos logit contributions
ays+0.075
eddy+0.075
ay+0.070
ummy+0.069
ane+0.068
Neg logit contributions
 forgot-0.116
 learned-0.100
 knew-0.099
 remembered-0.096
 wished-0.096
Token had
Token ID550
Feature activation-1.74e-01

Pos logit contributions
 forgot+0.062
 punished+0.058
 pap+0.055
 wasted+0.051
 regretted+0.051
Neg logit contributions
 with-0.110
he-0.104
ing-0.100
 on-0.099
room-0.096
Token fun
Token ID1257
Feature activation+3.62e+00

Pos logit contributions
 forgot+0.004
 remembered+0.003
 knew+0.003
 wished+0.003
 knocked+0.003
Neg logit contributions
he-0.004
room-0.004
ay-0.004
ing-0.004
time-0.004
Token.
Token ID13
Feature activation+2.57e+00

Pos logit contributions
ay+0.062
eddy+0.061
ays+0.059
ummy+0.056
uggle+0.056
Neg logit contributions
 forgot-0.094
 wished-0.082
 bit-0.080
 knew-0.079
 knocked-0.078
Token She
Token ID1375
Feature activation-9.24e+00

Pos logit contributions
he+0.045
ing+0.043
ay+0.043
room+0.042
eddy+0.042
Neg logit contributions
 forgot-0.072
 knew-0.060
 wished-0.059
 remembered-0.057
 missed-0.057
Token saw
Token ID2497
Feature activation-1.69e+00

Pos logit contributions
chie+0.130
 robbers+0.121
 pap+0.111
 supervision+0.107
 setup+0.105
Neg logit contributions
 with-0.209
he-0.205
 at-0.184
 again-0.184
 and-0.182
Token the
Token ID262
Feature activation-2.52e+00

Pos logit contributions
 forgot+0.047
 remembered+0.037
 knew+0.035
 struggled+0.035
 bit+0.035
Neg logit contributions
ing-0.029
he-0.029
room-0.027
ings-0.026
ay-0.026
Token same
Token ID976
Feature activation-1.57e+00

Pos logit contributions
 forgot+0.071
 remembered+0.059
 struggled+0.056
 punished+0.056
 wished+0.055
Neg logit contributions
he-0.043
ay-0.039
ing-0.037
ings-0.036
time-0.036
Token bird
Token ID6512
Feature activation+1.69e-01

Pos logit contributions
 forgot+0.037
 remembered+0.029
 wished+0.029
 knew+0.028
 struggled+0.027
Neg logit contributions
he-0.034
ings-0.033
ing-0.033
ay-0.032
room-0.032
Token from
Token ID422
Feature activation-9.84e-01

Pos logit contributions
he+0.002
ing+0.002
ay+0.002
ore+0.002
ings+0.002
Neg logit contributions
 forgot-0.005
 remembered-0.004
 knew-0.004
 wished-0.004
 bit-0.004
Tokenmy
Token ID1820
Feature activation-7.70e+00

Pos logit contributions
 forgot+0.113
 wished+0.096
 knocked+0.090
 known+0.090
 forgotten+0.089
Neg logit contributions
he-0.082
ing-0.078
ay-0.077
eddy-0.076
room-0.072
Token was
Token ID373
Feature activation-2.88e+00

Pos logit contributions
 forgot+0.174
 known+0.156
 robbers+0.155
 stolen+0.148
 forgotten+0.147
Neg logit contributions
he-0.134
ing-0.129
 with-0.126
uck-0.120
eddy-0.118
Token so
Token ID523
Feature activation-2.45e+00

Pos logit contributions
 forgot+0.062
 wished+0.048
 remembered+0.046
 knew+0.046
 bit+0.046
Neg logit contributions
he-0.067
ing-0.057
room-0.057
ay-0.056
time-0.056
Token happy
Token ID3772
Feature activation+2.62e+00

Pos logit contributions
 forgot+0.046
 remembered+0.034
 wished+0.033
 knew+0.032
 bit+0.031
Neg logit contributions
he-0.066
ing-0.061
ay-0.061
ore-0.060
ings-0.060
Token.
Token ID13
Feature activation+2.46e+00

Pos logit contributions
he+0.046
ay+0.043
room+0.040
ays+0.040
ings+0.040
Neg logit contributions
 forgot-0.076
 knew-0.064
 wished-0.062
 remembered-0.062
 knocked-0.059
Token He
Token ID679
Feature activation-9.21e+00

Pos logit contributions
he+0.045
ings+0.041
ing+0.040
room+0.040
ore+0.040
Neg logit contributions
 forgot-0.068
 knew-0.058
 remembered-0.056
 wished-0.055
 knocked-0.052
Token held
Token ID2714
Feature activation+9.44e-01

Pos logit contributions
 forgot+0.138
 known+0.127
 robbers+0.122
 forgotten+0.120
 mate+0.119
Neg logit contributions
he-0.209
 with-0.201
ay-0.199
eddy-0.193
 and-0.180
Token the
Token ID262
Feature activation-3.10e+00

Pos logit contributions
he+0.020
ings+0.019
room+0.019
ing+0.018
ore+0.018
Neg logit contributions
 forgot-0.022
 remembered-0.019
 knew-0.019
 wished-0.018
 learned-0.017
Token arrow
Token ID15452
Feature activation+2.74e+00

Pos logit contributions
 forgot+0.071
 remembered+0.053
 knocked+0.052
 wished+0.052
 punished+0.051
Neg logit contributions
ay-0.069
he-0.067
ing-0.064
ings-0.061
eddy-0.059
Token up
Token ID510
Feature activation+1.55e+00

Pos logit contributions
he+0.039
ore+0.035
ing+0.035
ay+0.035
ings+0.035
Neg logit contributions
 forgot-0.086
 remembered-0.072
 knew-0.071
 wished-0.071
 bit-0.068
Token and
Token ID290
Feature activation-6.30e+00

Pos logit contributions
he+0.023
ore+0.021
ing+0.021
ays+0.021
ay+0.021
Neg logit contributions
 forgot-0.046
 remembered-0.039
 knew-0.039
 wished-0.038
 bit-0.038
Token the
Token ID262
Feature activation-3.22e+00

Pos logit contributions
 forgot+0.049
 remembered+0.040
 knew+0.039
 wished+0.039
 struggled+0.037
Neg logit contributions
he-0.030
ings-0.027
room-0.026
ore-0.026
ay-0.025
Token wedding
Token ID10614
Feature activation+1.75e+00

Pos logit contributions
 forgot+0.076
 remembered+0.058
 struggled+0.058
 knew+0.056
 wished+0.053
Neg logit contributions
he-0.070
time-0.068
ing-0.063
ore-0.062
ings-0.062
Token.
Token ID13
Feature activation+3.16e+00

Pos logit contributions
he+0.028
ing+0.027
ore+0.027
ay+0.027
ays+0.027
Neg logit contributions
 forgot-0.050
 remembered-0.041
 knew-0.041
 wished-0.041
 bit-0.039
Token Instead
Token ID5455
Feature activation+4.90e-01

Pos logit contributions
ing+0.055
he+0.054
ings+0.052
ays+0.051
ach+0.051
Neg logit contributions
 forgot-0.082
 knew-0.072
 learned-0.069
 wished-0.067
 remembered-0.066
Token,
Token ID11
Feature activation-1.43e+00

Pos logit contributions
he+0.005
ing+0.004
ay+0.004
ings+0.004
ore+0.004
Neg logit contributions
 forgot-0.017
 remembered-0.015
 knew-0.015
 wished-0.015
 bit-0.014
Token she
Token ID673
Feature activation-9.21e+00

Pos logit contributions
 forgot+0.042
 remembered+0.035
 knew+0.034
 wished+0.034
 struggled+0.032
Neg logit contributions
he-0.023
ing-0.021
ore-0.020
ay-0.020
ings-0.020
Token went
Token ID1816
Feature activation+1.61e+00

Pos logit contributions
 forgot+0.117
 robbers+0.111
 pap+0.107
 persever+0.105
 unknow+0.103
Neg logit contributions
he-0.213
 with-0.204
 at-0.192
.-0.189
room-0.187
Token to
Token ID284
Feature activation-1.05e+00

Pos logit contributions
he+0.025
ay+0.023
ing+0.023
ings+0.022
ore+0.022
Neg logit contributions
 forgot-0.048
 remembered-0.040
 knew-0.040
 wished-0.040
 knocked-0.038
Token the
Token ID262
Feature activation-3.29e+00

Pos logit contributions
 forgot+0.028
 remembered+0.023
 knew+0.023
 wished+0.022
 struggled+0.021
Neg logit contributions
he-0.020
ore-0.018
ings-0.018
room-0.018
ay-0.018
Token park
Token ID3952
Feature activation+3.28e+00

Pos logit contributions
 forgot+0.078
 struggled+0.060
 remembered+0.059
 knew+0.057
 wished+0.055
Neg logit contributions
he-0.072
time-0.069
ing-0.065
ore-0.065
eddy-0.064
Token and
Token ID290
Feature activation-4.86e+00

Pos logit contributions
he+0.049
ay+0.046
ing+0.045
ings+0.045
ore+0.044
Neg logit contributions
 forgot-0.100
 knew-0.083
 remembered-0.083
 wished-0.082
 bit-0.079
Token dream
Token ID4320
Feature activation+5.05e+00

Pos logit contributions
 forgot+0.037
 remembered+0.028
 knew+0.027
 wished+0.027
 bit+0.025
Neg logit contributions
he-0.043
ay-0.040
ing-0.040
ore-0.040
ings-0.040
Token.
Token ID13
Feature activation+2.63e+00

Pos logit contributions
ays+0.078
eddy+0.068
iffs+0.066
he+0.064
uggle+0.064
Neg logit contributions
 forgot-0.136
 knew-0.120
 always-0.113
 cost-0.112
 needed-0.109
Token Every
Token ID3887
Feature activation-1.18e+00

Pos logit contributions
ing+0.058
ings+0.053
he+0.053
ore+0.052
ays+0.051
Neg logit contributions
 forgot-0.058
 knew-0.049
 wished-0.048
 wanted-0.045
 learned-0.045
Token night
Token ID1755
Feature activation+3.31e+00

Pos logit contributions
 forgot+0.028
 remembered+0.022
 knew+0.022
 wished+0.022
 knocked+0.020
Neg logit contributions
he-0.026
ing-0.024
ore-0.024
ay-0.024
ings-0.024
Token,
Token ID11
Feature activation+7.75e-01

Pos logit contributions
he+0.060
room+0.056
ays+0.055
ay+0.053
ore+0.051
Neg logit contributions
 forgot-0.094
 knew-0.077
 remembered-0.075
 knocked-0.073
 wished-0.072
Token she
Token ID673
Feature activation-9.20e+00

Pos logit contributions
he+0.017
ay+0.017
ore+0.017
room+0.017
ings+0.016
Neg logit contributions
 forgot-0.018
 knew-0.014
 wished-0.014
 remembered-0.014
 knocked-0.013
Token would
Token ID561
Feature activation-3.75e+00

Pos logit contributions
 forgot+0.150
 stolen+0.123
 punished+0.115
 unknow+0.115
 forgotten+0.112
Neg logit contributions
he-0.209
ore-0.200
 with-0.194
ing-0.191
eddy-0.185
Token close
Token ID1969
Feature activation+1.98e+00

Pos logit contributions
 forgot+0.082
 remembered+0.061
 wished+0.060
 knew+0.058
 struggled+0.058
Neg logit contributions
he-0.088
ore-0.079
uggle-0.078
room-0.078
ay-0.077
Token her
Token ID607
Feature activation-1.26e+00

Pos logit contributions
he+0.046
room+0.043
ore+0.043
ay+0.042
over+0.041
Neg logit contributions
 forgot-0.041
 knew-0.034
 remembered-0.034
 wished-0.033
 bit-0.030
Token eyes
Token ID2951
Feature activation+3.46e+00

Pos logit contributions
 forgot+0.028
 remembered+0.022
 knew+0.022
 wished+0.021
 bit+0.020
Neg logit contributions
he-0.029
ing-0.027
ay-0.027
ore-0.027
ings-0.027
Token and
Token ID290
Feature activation-7.65e+00

Pos logit contributions
ays+0.063
ore+0.060
ings+0.060
room+0.058
aps+0.058
Neg logit contributions
 forgot-0.085
 bit-0.075
 knew-0.074
 remembered-0.070
 wished-0.069
Token
Token ID198
Feature activation+2.87e+00

Pos logit contributions
ays+0.054
ing+0.054
he+0.050
ings+0.050
ay+0.050
Neg logit contributions
 forgot-0.095
 knew-0.085
 wished-0.082
 remembered-0.078
 learned-0.078
TokenL
Token ID43
Feature activation+1.81e-02

Pos logit contributions
he+0.049
ing+0.045
ings+0.045
ay+0.045
ore+0.045
Neg logit contributions
 forgot-0.082
 remembered-0.068
 knew-0.068
 wished-0.067
 learned-0.063
Tokenily
Token ID813
Feature activation-5.83e+00

Pos logit contributions
he+0.000
ings+0.000
ay+0.000
ing+0.000
ore+0.000
Neg logit contributions
 forgot-0.000
 knew-0.000
 remembered-0.000
 wished-0.000
 learned-0.000
Token and
Token ID290
Feature activation-1.36e+00

Pos logit contributions
 forgot+0.134
 known+0.115
 stolen+0.110
 forgotten+0.110
 cost+0.106
Neg logit contributions
ing-0.101
he-0.098
ays-0.096
uck-0.096
ore-0.093
Token Ben
Token ID3932
Feature activation-7.27e+00

Pos logit contributions
 forgot+0.033
 remembered+0.027
 knew+0.026
 wished+0.026
 knocked+0.024
Neg logit contributions
he-0.029
ay-0.028
ing-0.027
ore-0.027
ings-0.027
Token also
Token ID635
Feature activation-9.20e+00

Pos logit contributions
 forgot+0.147
 pap+0.118
 stolen+0.118
 known+0.118
 shown+0.117
Neg logit contributions
he-0.162
room-0.125
uggle-0.124
 with-0.124
ore-0.123
Token wake
Token ID7765
Feature activation+2.37e+00

Pos logit contributions
 Indians+0.146
 forgot+0.141
 robbers+0.131
 crew+0.129
 pap+0.125
Neg logit contributions
he-0.185
.-0.171
 at-0.168
 with-0.167
.'-0.157
Token up
Token ID510
Feature activation+2.35e+00

Pos logit contributions
he+0.040
ay+0.040
unch+0.038
ane+0.037
room+0.035
Neg logit contributions
 forgot-0.064
 knew-0.054
 remembered-0.054
 missed-0.052
 wished-0.051
Token and
Token ID290
Feature activation-6.14e+00

Pos logit contributions
he+0.041
ays+0.040
room+0.040
ay+0.040
ings+0.037
Neg logit contributions
 forgot-0.063
 knew-0.055
 remembered-0.053
 bit-0.051
 learned-0.051
Token look
Token ID804
Feature activation-1.49e-01

Pos logit contributions
 forgot+0.122
 knew+0.091
 remembered+0.090
 known+0.089
 shown+0.087
Neg logit contributions
he-0.144
ore-0.135
uggle-0.132
ing-0.131
ays-0.128
Token out
Token ID503
Feature activation+9.53e-01

Pos logit contributions
 forgot+0.005
 remembered+0.004
 knew+0.004
 wished+0.004
 bit+0.004
Neg logit contributions
he-0.002
room-0.002
ay-0.002
ings-0.002
ore-0.002
Token got
Token ID1392
Feature activation+4.81e-01

Pos logit contributions
 forgot+0.116
 pap+0.084
 known+0.084
 wished+0.083
 remembered+0.082
Neg logit contributions
he-0.189
ore-0.168
ay-0.162
time-0.160
room-0.160
Token to
Token ID284
Feature activation-1.53e+00

Pos logit contributions
he+0.009
ay+0.008
ings+0.008
room+0.008
ing+0.008
Neg logit contributions
 forgot-0.013
 remembered-0.011
 knew-0.011
 wished-0.011
 knocked-0.010
Token the
Token ID262
Feature activation-2.21e+00

Pos logit contributions
 forgot+0.042
 remembered+0.035
 knew+0.034
 wished+0.034
 bit+0.032
Neg logit contributions
he-0.028
room-0.025
ay-0.025
ings-0.025
ore-0.024
Token school
Token ID1524
Feature activation+2.28e+00

Pos logit contributions
 forgot+0.053
 remembered+0.042
 wished+0.041
 knew+0.041
 struggled+0.039
Neg logit contributions
he-0.049
ay-0.044
ing-0.044
ore-0.044
room-0.044
Token,
Token ID11
Feature activation-3.84e-01

Pos logit contributions
he+0.029
ings+0.027
ore+0.027
ays+0.027
eddy+0.027
Neg logit contributions
 forgot-0.074
 knew-0.062
 remembered-0.062
 wished-0.061
 knocked-0.058
Token she
Token ID673
Feature activation-9.20e+00

Pos logit contributions
 forgot+0.010
 knew+0.008
 remembered+0.008
 wished+0.008
 knocked+0.007
Neg logit contributions
he-0.008
ay-0.007
ore-0.007
ing-0.007
ings-0.007
Token saw
Token ID2497
Feature activation-1.15e+00

Pos logit contributions
 sizes+0.135
 forgot+0.132
 vitamins+0.129
 fiber+0.126
 pap+0.125
Neg logit contributions
he-0.194
 with-0.186
 at-0.180
room-0.169
ing-0.167
Token many
Token ID867
Feature activation-1.86e+00

Pos logit contributions
 forgot+0.033
 remembered+0.027
 knew+0.027
 wished+0.027
 bit+0.025
Neg logit contributions
he-0.020
ing-0.018
ay-0.018
ore-0.018
ings-0.018
Token kids
Token ID3988
Feature activation-4.65e-01

Pos logit contributions
 forgot+0.043
 remembered+0.034
 wished+0.033
 knew+0.033
 knocked+0.031
Neg logit contributions
he-0.042
ing-0.039
ay-0.039
room-0.038
ings-0.037
Token.
Token ID13
Feature activation+2.30e+00

Pos logit contributions
 forgot+0.012
 remembered+0.010
 knew+0.010
 wished+0.010
 knocked+0.009
Neg logit contributions
he-0.009
ing-0.008
ay-0.008
ore-0.008
ings-0.008
Token They
Token ID1119
Feature activation-6.92e+00

Pos logit contributions
he+0.038
ing+0.034
ings+0.033
ore+0.033
ay+0.032
Neg logit contributions
 forgot-0.067
 knew-0.057
 wished-0.056
 remembered-0.055
 knocked-0.053
Tokenmy
Token ID1820
Feature activation-7.83e+00

Pos logit contributions
 forgot+0.118
 wished+0.100
 forgotten+0.093
 stolen+0.093
 known+0.093
Neg logit contributions
he-0.079
ing-0.075
room-0.072
eddy-0.071
ay-0.071
Token was
Token ID373
Feature activation-3.00e+00

Pos logit contributions
 forgot+0.151
 robbers+0.150
 dishonest+0.136
 stolen+0.133
 known+0.132
Neg logit contributions
he-0.159
ing-0.150
 with-0.145
room-0.145
ore-0.135
Token so
Token ID523
Feature activation-2.21e+00

Pos logit contributions
 forgot+0.073
 remembered+0.057
 knew+0.057
'd+0.056
 wished+0.056
Neg logit contributions
he-0.058
 with-0.053
ing-0.052
room-0.051
time-0.050
Token happy
Token ID3772
Feature activation+2.44e+00

Pos logit contributions
 forgot+0.046
 remembered+0.036
 wished+0.034
 knew+0.033
 bit+0.032
Neg logit contributions
he-0.056
ore-0.050
ing-0.050
ay-0.050
room-0.048
Token.
Token ID13
Feature activation+2.89e+00

Pos logit contributions
he+0.044
ay+0.042
ays+0.039
room+0.039
ings+0.038
Neg logit contributions
 forgot-0.069
 knew-0.059
 remembered-0.057
 wished-0.057
 knocked-0.054
Token He
Token ID679
Feature activation-9.20e+00

Pos logit contributions
he+0.047
ing+0.042
ings+0.042
ore+0.042
ay+0.042
Neg logit contributions
 forgot-0.085
 knew-0.071
 remembered-0.070
 wished-0.070
 knocked-0.066
Token watched
Token ID7342
Feature activation+2.36e+00

Pos logit contributions
 robbers+0.144
 supervision+0.140
 dishonest+0.134
 setup+0.130
 known+0.129
Neg logit contributions
 with-0.186
he-0.182
 and-0.174
 at-0.172
,-0.164
Token as
Token ID355
Feature activation+5.83e-01

Pos logit contributions
he+0.050
ays+0.048
ings+0.048
ay+0.047
ore+0.047
Neg logit contributions
 forgot-0.053
 knew-0.045
 remembered-0.043
 wished-0.042
 knocked-0.042
Token the
Token ID262
Feature activation-2.43e+00

Pos logit contributions
he+0.012
ings+0.012
ore+0.012
ay+0.012
ing+0.011
Neg logit contributions
 forgot-0.014
 knew-0.012
 remembered-0.011
 wished-0.011
 knocked-0.010
Token bright
Token ID6016
Feature activation-1.68e+00

Pos logit contributions
 forgot+0.074
 remembered+0.061
 wished+0.058
 knew+0.057
 regretted+0.057
Neg logit contributions
he-0.039
ay-0.037
time-0.036
ing-0.035
ore-0.033
Token fireworks
Token ID26056
Feature activation-2.75e+00

Pos logit contributions
 forgot+0.041
 remembered+0.032
 wished+0.031
 knew+0.031
 forgotten+0.029
Neg logit contributions
he-0.036
ing-0.035
ay-0.034
ore-0.033
ings-0.033
Token became
Token ID2627
Feature activation-2.94e+00

Pos logit contributions
 forgot+0.113
 known+0.090
 wished+0.087
 forget+0.086
 stolen+0.085
Neg logit contributions
he-0.137
ing-0.133
 with-0.125
ore-0.124
room-0.120
Token healthier
Token ID22841
Feature activation+2.42e+00

Pos logit contributions
 forgot+0.068
 wished+0.052
 knew+0.052
'd+0.051
 remembered+0.049
Neg logit contributions
he-0.059
ing-0.058
ings-0.058
time-0.056
over-0.054
Token and
Token ID290
Feature activation-5.48e+00

Pos logit contributions
he+0.032
ay+0.029
ore+0.028
ing+0.028
ays+0.028
Neg logit contributions
 forgot-0.078
 remembered-0.067
 knew-0.066
 wished-0.066
 bit-0.062
Token happier
Token ID23030
Feature activation+2.15e+00

Pos logit contributions
 forgot+0.114
 wished+0.095
 remembered+0.092
 struggled+0.091
 wasted+0.091
Neg logit contributions
ing-0.100
he-0.098
 with-0.097
time-0.095
 again-0.087
Token.
Token ID13
Feature activation+1.89e+00

Pos logit contributions
he+0.027
ay+0.024
room+0.023
ore+0.022
ays+0.022
Neg logit contributions
 forgot-0.074
 remembered-0.063
 knew-0.063
 wished-0.062
 bit-0.059
Token He
Token ID679
Feature activation-9.19e+00

Pos logit contributions
he+0.031
ing+0.028
ings+0.028
ay+0.027
ore+0.027
Neg logit contributions
 forgot-0.055
 knew-0.046
 wished-0.046
 remembered-0.045
 knocked-0.043
Token was
Token ID373
Feature activation-2.75e+00

Pos logit contributions
 forgot+0.125
chie+0.123
 robbers+0.116
 mate+0.111
 women+0.109
Neg logit contributions
he-0.208
 with-0.205
 again-0.188
 and-0.186
,-0.180
Token no
Token ID645
Feature activation-1.18e+00

Pos logit contributions
 forgot+0.062
 wished+0.048
 bit+0.046
 robbers+0.046
 struggled+0.046
Neg logit contributions
he-0.055
 with-0.053
time-0.049
ing-0.048
over-0.047
Token longer
Token ID2392
Feature activation-2.59e+00

Pos logit contributions
 forgot+0.033
 knew+0.028
 remembered+0.027
 wished+0.027
'd+0.027
Neg logit contributions
he-0.020
over-0.017
ing-0.017
ay-0.017
time-0.017
Token a
Token ID257
Feature activation-3.48e+00

Pos logit contributions
 forgot+0.047
'd+0.035
 wished+0.033
 knew+0.033
 remembered+0.033
Neg logit contributions
ing-0.062
he-0.061
over-0.059
ore-0.058
ings-0.057
Token thin
Token ID7888
Feature activation-1.65e+00

Pos logit contributions
 forgot+0.078
 struggled+0.066
 argued+0.062
 waited+0.061
 remembered+0.061
Neg logit contributions
time-0.062
 with-0.058
ay-0.058
ore-0.058
eddy-0.058
Token he
Token ID339
Feature activation-8.69e+00

Pos logit contributions
 forgot+0.010
 remembered+0.008
 knew+0.008
 wished+0.008
 knocked+0.008
Neg logit contributions
he-0.008
ore-0.007
ings-0.007
ay-0.007
ing-0.007
Token was
Token ID373
Feature activation-2.86e+00

Pos logit contributions
 forgot+0.134
 robbers+0.115
 stolen+0.107
 pap+0.106
 elves+0.104
Neg logit contributions
he-0.225
ing-0.198
ore-0.189
 with-0.188
room-0.187
Token so
Token ID523
Feature activation-2.39e+00

Pos logit contributions
 forgot+0.058
 wished+0.046
 knew+0.044
 remembered+0.043
 cared+0.041
Neg logit contributions
he-0.070
ing-0.064
ings-0.061
ore-0.061
room-0.060
Token happy
Token ID3772
Feature activation+3.23e+00

Pos logit contributions
 forgot+0.046
 remembered+0.034
 wished+0.034
 knew+0.034
 bit+0.030
Neg logit contributions
he-0.063
ing-0.060
ore-0.059
ay-0.059
ings-0.058
Token.
Token ID13
Feature activation+2.67e+00

Pos logit contributions
he+0.054
ay+0.052
ays+0.049
uggle+0.048
room+0.047
Neg logit contributions
 forgot-0.093
 knew-0.080
 wished-0.077
 remembered-0.076
 knocked-0.074
Token He
Token ID679
Feature activation-9.19e+00

Pos logit contributions
he+0.045
ing+0.042
ings+0.041
ore+0.041
ay+0.040
Neg logit contributions
 forgot-0.076
 knew-0.064
 remembered-0.063
 wished-0.063
 knocked-0.060
Token wanted
Token ID2227
Feature activation+8.32e-01

Pos logit contributions
 forgot+0.135
 robbers+0.126
 known+0.125
 pap+0.123
 setup+0.120
Neg logit contributions
he-0.214
 with-0.196
ing-0.184
ay-0.182
 at-0.181
Token to
Token ID284
Feature activation-3.65e+00

Pos logit contributions
he+0.010
ing+0.010
ore+0.010
ings+0.009
ay+0.009
Neg logit contributions
 forgot-0.027
 remembered-0.023
 wished-0.023
 knew-0.022
 bit-0.021
Token try
Token ID1949
Feature activation+1.86e+00

Pos logit contributions
 forgot+0.082
 remembered+0.064
 wished+0.064
 knew+0.062
 wasted+0.058
Neg logit contributions
he-0.085
ay-0.077
room-0.077
ings-0.076
ore-0.076
Token something
Token ID1223
Feature activation+1.78e-01

Pos logit contributions
he+0.037
ay+0.035
room+0.034
eddy+0.033
ore+0.033
Neg logit contributions
 forgot-0.048
 remembered-0.040
 knew-0.040
 wished-0.037
 learned-0.037
Token else
Token ID2073
Feature activation+1.77e+00

Pos logit contributions
he+0.004
ay+0.004
ays+0.004
room+0.004
ore+0.004
Neg logit contributions
 forgot-0.004
 knew-0.003
 remembered-0.003
 wished-0.003
 bit-0.003
Tokenmy
Token ID1820
Feature activation-1.61e+00

Pos logit contributions
 forgot+0.019
 remembered+0.014
 knew+0.014
 forget+0.012
 bit+0.012
Neg logit contributions
ore-0.031
he-0.031
ays-0.030
ings-0.030
overed-0.030
Token count
Token ID954
Feature activation+2.25e+00

Pos logit contributions
 forgot+0.050
 remembered+0.042
 knew+0.042
 wished+0.040
 bit+0.039
Neg logit contributions
he-0.022
ay-0.021
ing-0.021
ore-0.021
ays-0.021
Token the
Token ID262
Feature activation-2.36e+00

Pos logit contributions
he+0.049
ore+0.046
uggle+0.046
over+0.045
eddy+0.045
Neg logit contributions
 forgot-0.046
 knew-0.040
 remembered-0.038
 knocked-0.037
 bit-0.037
Token flowers
Token ID12734
Feature activation+1.28e+00

Pos logit contributions
 forgot+0.059
 wished+0.049
 remembered+0.049
 punished+0.048
 knew+0.047
Neg logit contributions
ay-0.043
he-0.042
ings-0.041
ing-0.041
ummy-0.039
Token.
Token ID13
Feature activation+2.50e+00

Pos logit contributions
he+0.016
ay+0.015
ore+0.014
ings+0.014
ing+0.014
Neg logit contributions
 forgot-0.042
 knew-0.036
 remembered-0.035
 wished-0.035
 bit-0.034
Token She
Token ID1375
Feature activation-9.18e+00

Pos logit contributions
he+0.044
ings+0.038
ore+0.038
ing+0.037
over+0.036
Neg logit contributions
 forgot-0.071
 knew-0.061
 wished-0.059
 remembered-0.058
 knocked-0.056
Token felt
Token ID2936
Feature activation-2.04e+00

Pos logit contributions
 stolen+0.145
 robbers+0.143
 dishonest+0.139
chie+0.139
 cruel+0.138
Neg logit contributions
 with-0.187
he-0.182
 at-0.172
 again-0.165
room-0.163
Token wide
Token ID3094
Feature activation+1.61e+00

Pos logit contributions
 forgot+0.043
 remembered+0.033
 knew+0.032
 wished+0.032
 bit+0.029
Neg logit contributions
he-0.050
ing-0.048
ay-0.047
ore-0.047
ings-0.047
Token awake
Token ID21693
Feature activation+2.59e+00

Pos logit contributions
he+0.027
ore+0.024
ay+0.024
ing+0.024
ings+0.024
Neg logit contributions
 forgot-0.047
 remembered-0.039
 knew-0.038
 wished-0.038
 bit-0.036
Token and
Token ID290
Feature activation-4.38e+00

Pos logit contributions
he+0.037
ay+0.035
ore+0.035
ays+0.034
ings+0.033
Neg logit contributions
 forgot-0.080
 remembered-0.067
 knew-0.067
 wished-0.066
 bit-0.063
Token ready
Token ID3492
Feature activation+1.45e+00

Pos logit contributions
 forgot+0.088
 wasted+0.069
 remembered+0.068
 wished+0.068
 stolen+0.066
Neg logit contributions
he-0.095
ing-0.089
 with-0.086
ore-0.084
ay-0.083
Token found
Token ID1043
Feature activation+4.65e-02

Pos logit contributions
 forgot+0.098
 robbers+0.073
 known+0.073
 pap+0.071
 remembered+0.068
Neg logit contributions
he-0.168
ing-0.159
ore-0.153
room-0.153
ays-0.149
Token it
Token ID340
Feature activation-6.01e-01

Pos logit contributions
he+0.001
ing+0.001
ay+0.001
ore+0.001
ings+0.001
Neg logit contributions
 forgot-0.001
 remembered-0.001
 knew-0.001
 wished-0.001
 bit-0.001
Token.
Token ID13
Feature activation+2.81e+00

Pos logit contributions
 forgot+0.020
 remembered+0.017
 knew+0.017
 wished+0.017
 bit+0.016
Neg logit contributions
he-0.007
ing-0.006
ay-0.006
ore-0.006
ings-0.006
Token As
Token ID1081
Feature activation-1.24e+00

Pos logit contributions
he+0.043
ing+0.039
ings+0.039
ay+0.038
ore+0.037
Neg logit contributions
 forgot-0.083
 knew-0.071
 wished-0.070
 remembered-0.069
 learned-0.067
Token the
Token ID262
Feature activation-2.65e+00

Pos logit contributions
 forgot+0.036
 remembered+0.030
 wished+0.029
 knew+0.029
 struggled+0.028
Neg logit contributions
he-0.019
ing-0.018
ore-0.017
ings-0.017
time-0.016
Token woman
Token ID2415
Feature activation-9.18e+00

Pos logit contributions
 forgot+0.061
 remembered+0.048
 known+0.046
 wasted+0.046
 struggled+0.046
Neg logit contributions
he-0.058
ay-0.055
ing-0.053
time-0.051
ummy-0.050
Token kept
Token ID4030
Feature activation-1.35e+00

Pos logit contributions
 forgot+0.152
 robbers+0.120
 vitamins+0.116
 known+0.114
 stolen+0.112
Neg logit contributions
he-0.240
ing-0.215
 with-0.205
ore-0.203
room-0.197
Token walking
Token ID6155
Feature activation+3.53e+00

Pos logit contributions
 forgot+0.036
 remembered+0.028
 knew+0.028
 wished+0.028
'd+0.026
Neg logit contributions
he-0.025
ore-0.024
ing-0.024
ings-0.024
ay-0.023
Token,
Token ID11
Feature activation-9.03e-01

Pos logit contributions
he+0.049
ays+0.046
ay+0.045
room+0.044
ore+0.044
Neg logit contributions
 forgot-0.110
 knew-0.094
 remembered-0.093
 wished-0.092
 bit-0.089
Token she
Token ID673
Feature activation-7.87e+00

Pos logit contributions
 forgot+0.026
 remembered+0.021
 wished+0.021
 knew+0.021
 wasted+0.020
Neg logit contributions
he-0.015
ing-0.014
ore-0.013
ings-0.013
ay-0.013
Token heard
Token ID2982
Feature activation+5.36e-01

Pos logit contributions
 forgot+0.119
 robbers+0.103
 vitamins+0.099
 wasted+0.095
 stolen+0.094
Neg logit contributions
ing-0.187
he-0.183
 with-0.177
 again-0.167
 at-0.165
Token the
Token ID262
Feature activation-3.08e+00

Pos logit contributions
he+0.008
ing+0.007
ore+0.007
ay+0.007
ings+0.007
Neg logit contributions
 forgot-0.013
 remembered-0.011
 wished-0.011
 knew-0.011
 struggled-0.010
Token leash
Token ID40529
Feature activation-2.99e+00

Pos logit contributions
 forgot+0.077
 remembered+0.060
 wished+0.060
 knew+0.059
 struggled+0.057
Neg logit contributions
he-0.065
ay-0.060
time-0.058
ings-0.057
ing-0.057
Token worked
Token ID3111
Feature activation+2.94e+00

Pos logit contributions
 forgot+0.061
 remembered+0.048
 pap+0.047
 known+0.045
 wished+0.045
Neg logit contributions
he-0.075
ing-0.065
ay-0.065
room-0.064
ore-0.063
Token again
Token ID757
Feature activation+1.53e+00

Pos logit contributions
he+0.044
ays+0.041
ore+0.041
ay+0.040
room+0.039
Neg logit contributions
 forgot-0.089
 knew-0.075
 remembered-0.075
 wished-0.073
 bit-0.070
Token.
Token ID13
Feature activation+1.83e+00

Pos logit contributions
he+0.019
ays+0.018
ay+0.017
ore+0.017
ing+0.017
Neg logit contributions
 forgot-0.050
 knew-0.043
 remembered-0.043
 wished-0.042
 bit-0.041
Token She
Token ID1375
Feature activation-9.17e+00

Pos logit contributions
he+0.026
ing+0.023
ings+0.023
ore+0.023
ay+0.023
Neg logit contributions
 forgot-0.058
 knew-0.049
 remembered-0.048
 wished-0.048
 learned-0.046
Token was
Token ID373
Feature activation-2.63e+00

Pos logit contributions
 forgot+0.149
 forgotten+0.133
 robbers+0.132
 stolen+0.129
 mate+0.126
Neg logit contributions
 with-0.202
he-0.195
 and-0.179
ing-0.176
eddy-0.173
Token happy
Token ID3772
Feature activation+7.12e-01

Pos logit contributions
 forgot+0.061
 wished+0.047
 bit+0.045
 struggled+0.045
 knew+0.045
Neg logit contributions
he-0.056
 with-0.050
ing-0.047
time-0.047
room-0.046
Token to
Token ID284
Feature activation-1.44e+00

Pos logit contributions
he+0.010
ing+0.009
ore+0.009
ay+0.009
ings+0.009
Neg logit contributions
 forgot-0.023
 remembered-0.019
 wished-0.019
 knew-0.019
 knocked-0.018
Token go
Token ID467
Feature activation+9.99e-01

Pos logit contributions
 forgot+0.026
 remembered+0.018
 wished+0.018
 knew+0.017
 struggled+0.016
Neg logit contributions
he-0.040
ay-0.038
room-0.037
ore-0.037
ings-0.037
Token on
Token ID319
Feature activation+8.50e-01

Pos logit contributions
he+0.016
ays+0.015
ings+0.015
room+0.015
ay+0.015
Neg logit contributions
 forgot-0.029
 remembered-0.024
 knew-0.024
 wished-0.023
 learned-0.023
Token birds
Token ID10087
Feature activation+2.14e-01

Pos logit contributions
 forgot+0.051
 remembered+0.041
 wished+0.038
 knocked+0.038
 struggled+0.038
Neg logit contributions
he-0.050
ing-0.047
ings-0.046
ay-0.045
ore-0.045
Token eat
Token ID4483
Feature activation+2.31e+00

Pos logit contributions
he+0.004
ay+0.003
ings+0.003
ore+0.003
ing+0.003
Neg logit contributions
 forgot-0.006
 remembered-0.005
 knew-0.005
 wished-0.005
 bit-0.005
Token the
Token ID262
Feature activation-1.51e+00

Pos logit contributions
he+0.039
ay+0.036
ore+0.035
ays+0.035
eddy+0.034
Neg logit contributions
 forgot-0.064
 knew-0.055
 remembered-0.054
 wished-0.053
 bit-0.053
Token jelly
Token ID27644
Feature activation+3.40e+00

Pos logit contributions
 forgot+0.032
 remembered+0.025
 wished+0.024
 knew+0.023
 knocked+0.022
Neg logit contributions
he-0.038
ing-0.036
ore-0.035
ings-0.035
ay-0.035
Token.
Token ID13
Feature activation+3.17e+00

Pos logit contributions
he+0.044
ay+0.043
room+0.041
ays+0.040
ings+0.040
Neg logit contributions
 forgot-0.107
 knew-0.090
 remembered-0.088
 bit-0.087
 wished-0.087
Token She
Token ID1375
Feature activation-9.17e+00

Pos logit contributions
he+0.051
ing+0.048
ings+0.047
room+0.046
ore+0.045
Neg logit contributions
 forgot-0.094
 knew-0.079
 wished-0.077
 remembered-0.075
 learned-0.074
Token laughed
Token ID13818
Feature activation+2.96e+00

Pos logit contributions
 robbers+0.137
chie+0.136
 supervision+0.131
 nurses+0.127
 prisoners+0.126
Neg logit contributions
 with-0.201
 and-0.180
he-0.177
 at-0.172
,-0.171
Token and
Token ID290
Feature activation-5.88e+00

Pos logit contributions
ore+0.067
ings+0.066
he+0.066
ays+0.065
ay+0.062
Neg logit contributions
 forgot-0.064
 knew-0.059
 remembered-0.057
 wished-0.054
 learned-0.053
Token cl
Token ID537
Feature activation+2.34e+00

Pos logit contributions
 forgot+0.083
 stolen+0.071
'd+0.071
 unknow+0.070
 punished+0.070
Neg logit contributions
 with-0.137
he-0.134
ing-0.128
 together-0.123
 on-0.122
Tokenapped
Token ID6320
Feature activation+3.94e+00

Pos logit contributions
he+0.038
ore+0.032
ings+0.031
ing+0.031
 anymore+0.030
Neg logit contributions
 forgot-0.066
 remembered-0.059
 knew-0.056
 learned-0.053
 wished-0.053
Token her
Token ID607
Feature activation-1.12e+00

Pos logit contributions
he+0.087
ays+0.083
ore+0.082
ane+0.078
ay+0.077
Neg logit contributions
 forgot-0.095
 knew-0.080
 remembered-0.078
 wished-0.074
 bit-0.072
Token't
Token ID470
Feature activation-1.34e+00

Pos logit contributions
ings+0.094
ing+0.094
over+0.092
ment+0.091
ore+0.089
Neg logit contributions
 forgot-0.089
 knew-0.071
 knocked-0.068
 remembered-0.067
 wished-0.066
Token find
Token ID1064
Feature activation+7.88e-01

Pos logit contributions
 forgot+0.019
 knew+0.013
 remembered+0.013
 wished+0.012
 knocked+0.010
Neg logit contributions
ing-0.042
he-0.041
ays-0.041
ay-0.041
ings-0.041
Token it
Token ID340
Feature activation+1.08e+00

Pos logit contributions
he+0.012
ing+0.011
ay+0.011
ore+0.011
ings+0.011
Neg logit contributions
 forgot-0.024
 remembered-0.020
 knew-0.020
 wished-0.019
 bit-0.018
Token.
Token ID13
Feature activation+3.87e+00

Pos logit contributions
ay+0.019
aps+0.019
ays+0.019
ore+0.019
he+0.018
Neg logit contributions
 forgot-0.030
 knew-0.025
 remembered-0.025
 wished-0.024
 forgotten-0.024
Token
Token ID220
Feature activation+5.37e+00

Pos logit contributions
he+0.078
ays+0.072
ings+0.071
room+0.071
ing+0.070
Neg logit contributions
 forgot-0.099
 knew-0.081
 remembered-0.080
 wished-0.078
 missed-0.076
Token
Token ID198
Feature activation+5.35e+00

Pos logit contributions
over+0.103
ipping+0.102
ach+0.100
ays+0.100
uggle+0.098
Neg logit contributions
 forgot-0.120
 knew-0.104
 had-0.095
 learned-0.092
 missed-0.091
Token
Token ID198
Feature activation+4.00e+00

Pos logit contributions
ays+0.104
ing+0.097
overed+0.096
ipping+0.095
over+0.094
Neg logit contributions
 forgot-0.127
 knew-0.111
 had-0.101
 learned-0.101
 told-0.101
TokenB
Token ID33
Feature activation+1.95e+00

Pos logit contributions
he+0.076
ays+0.070
ings+0.070
ore+0.070
ay+0.069
Neg logit contributions
 forgot-0.109
 remembered-0.089
 knew-0.088
 wished-0.087
 bit-0.082
Tokenenny
Token ID11870
Feature activation-5.79e+00

Pos logit contributions
he+0.037
ays+0.035
ings+0.034
ement+0.033
ore+0.032
Neg logit contributions
 forgot-0.051
 remembered-0.040
 knew-0.040
'd-0.037
 struggled-0.037
Token asked
Token ID1965
Feature activation+6.61e-01

Pos logit contributions
 forgot+0.114
 known+0.102
 forget+0.099
 groom+0.096
 humor+0.095
Neg logit contributions
ing-0.108
 with-0.104
uck-0.102
he-0.098
time-0.095
Token his
Token ID465
Feature activation-2.16e+00

Pos logit contributions
he+0.017
ays+0.016
ay+0.016
ings+0.015
ore+0.015
Neg logit contributions
 forgot-0.014
 knew-0.012
 remembered-0.012
 wished-0.011
 knocked-0.010
Token them
Token ID606
Feature activation+6.15e-01

Pos logit contributions
he+0.032
ing+0.029
ay+0.028
ings+0.028
room+0.028
Neg logit contributions
 forgot-0.067
 remembered-0.056
 wished-0.055
 knew-0.055
 bit-0.052
Token to
Token ID284
Feature activation-3.41e+00

Pos logit contributions
he+0.010
ing+0.009
ings+0.009
ore+0.009
ay+0.009
Neg logit contributions
 forgot-0.018
 remembered-0.015
 wished-0.015
 knew-0.015
 bit-0.014
Token be
Token ID307
Feature activation-1.28e+00

Pos logit contributions
 forgot+0.065
 remembered+0.050
 wished+0.050
 knew+0.048
 regretted+0.045
Neg logit contributions
he-0.088
ings-0.082
room-0.081
time-0.080
uggle-0.078
Token careful
Token ID8161
Feature activation+4.32e+00

Pos logit contributions
 forgot+0.023
 wished+0.016
 knew+0.016
 remembered+0.015
 knocked+0.015
Neg logit contributions
he-0.032
ings-0.031
over-0.031
ing-0.031
time-0.031
Token.
Token ID13
Feature activation+3.42e+00

Pos logit contributions
he+0.066
ay+0.064
ays+0.063
ore+0.059
room+0.056
Neg logit contributions
 forgot-0.127
 knew-0.107
 remembered-0.104
 wished-0.103
 never-0.101
Token
Token ID198
Feature activation+4.95e+00

Pos logit contributions
he+0.066
ings+0.063
ing+0.062
ore+0.062
ane+0.062
Neg logit contributions
 forgot-0.086
 remembered-0.066
 knew-0.066
 learned-0.066
 wished-0.063
Token
Token ID198
Feature activation+3.67e+00

Pos logit contributions
ays+0.089
ment+0.085
overed+0.082
ane+0.082
ement+0.081
Neg logit contributions
 forgot-0.127
 knew-0.105
 learned-0.101
 had-0.100
 remembered-0.096
Token"
Token ID1
Feature activation+1.08e-01

Pos logit contributions
he+0.062
ore+0.060
ings+0.059
ing+0.058
ays+0.058
Neg logit contributions
 forgot-0.105
 remembered-0.086
 knew-0.085
 wished-0.084
 bit-0.080
TokenOnly
Token ID10049
Feature activation-2.01e+00

Pos logit contributions
he+0.002
ing+0.002
ore+0.002
ay+0.002
room+0.002
Neg logit contributions
 forgot-0.003
 remembered-0.002
 knew-0.002
 wished-0.002
 knocked-0.002
Token feed
Token ID3745
Feature activation+3.16e+00

Pos logit contributions
 forgot+0.058
 wished+0.048
 knew+0.047
 remembered+0.047
 missed+0.045
Neg logit contributions
ing-0.029
ore-0.028
he-0.027
ings-0.026
ummy-0.026
Token the
Token ID262
Feature activation-2.66e+00

Pos logit contributions
he+0.065
ay+0.061
ays+0.059
room+0.059
ore+0.059
Neg logit contributions
 forgot-0.076
 knew-0.062
 remembered-0.061
 wished-0.059
 bit-0.058
Token it
Token ID340
Feature activation-2.73e+00

Pos logit contributions
he+0.059
ay+0.056
ore+0.056
room+0.055
uggle+0.055
Neg logit contributions
 forgot-0.068
 knew-0.054
 remembered-0.054
 wished-0.053
 bit-0.051
Token would
Token ID561
Feature activation-8.50e-01

Pos logit contributions
 forgot+0.058
 remembered+0.046
 wished+0.045
 knew+0.042
 pap+0.042
Neg logit contributions
he-0.065
ing-0.060
room-0.059
ay-0.058
ore-0.058
Token go
Token ID467
Feature activation+3.08e+00

Pos logit contributions
 forgot+0.016
 remembered+0.012
 wished+0.012
 knew+0.011
 struggled+0.011
Neg logit contributions
he-0.023
room-0.021
ay-0.021
ings-0.021
ing-0.021
Token.
Token ID13
Feature activation+3.87e+00

Pos logit contributions
ays+0.043
aps+0.043
he+0.043
ore+0.041
ings+0.040
Neg logit contributions
 forgot-0.091
 knew-0.078
 bit-0.078
 knocked-0.074
 remembered-0.074
Token
Token ID220
Feature activation+5.23e+00

Pos logit contributions
he+0.068
ing+0.067
ays+0.067
ings+0.065
over+0.064
Neg logit contributions
 forgot-0.106
 knew-0.089
 remembered-0.084
 wished-0.084
 knocked-0.082
Token
Token ID198
Feature activation+5.51e+00

Pos logit contributions
ipping+0.102
over+0.102
ays+0.101
ing+0.100
ingly+0.098
Neg logit contributions
 forgot-0.113
 knew-0.098
 learned-0.088
 had-0.087
 told-0.085
Token
Token ID198
Feature activation+3.82e+00

Pos logit contributions
ays+0.112
ing+0.100
overed+0.100
over+0.100
ipping+0.099
Neg logit contributions
 forgot-0.131
 knew-0.114
 told-0.103
 learned-0.103
 had-0.102
TokenAs
Token ID1722
Feature activation+7.19e-01

Pos logit contributions
he+0.066
ing+0.061
ore+0.061
ay+0.061
ings+0.061
Neg logit contributions
 forgot-0.109
 remembered-0.090
 knew-0.089
 wished-0.089
 bit-0.083
Token he
Token ID339
Feature activation-6.76e+00

Pos logit contributions
he+0.015
ing+0.014
ore+0.014
ay+0.014
ings+0.014
Neg logit contributions
 forgot-0.018
 remembered-0.015
 knew-0.014
 wished-0.014
 knocked-0.013
Token was
Token ID373
Feature activation-1.63e+00

Pos logit contributions
 forgot+0.134
 wished+0.099
 pap+0.099
 forget+0.098
 remembered+0.097
Neg logit contributions
he-0.168
room-0.161
ay-0.155
ing-0.154
ore-0.150
Token building
Token ID2615
Feature activation+1.79e+00

Pos logit contributions
 forgot+0.038
 wished+0.031
 remembered+0.030
 knew+0.029
 forget+0.028
Neg logit contributions
he-0.035
room-0.033
ing-0.032
time-0.032
ore-0.032
Token wanted
Token ID2227
Feature activation+1.22e+00

Pos logit contributions
 forgot+0.107
 known+0.103
 forgotten+0.094
 robbers+0.092
 setup+0.091
Neg logit contributions
 with-0.130
he-0.123
room-0.115
ing-0.113
 at-0.111
Token to
Token ID284
Feature activation-1.82e+00

Pos logit contributions
he+0.012
ing+0.012
ore+0.012
ay+0.011
ings+0.011
Neg logit contributions
 forgot-0.042
 remembered-0.035
 wished-0.034
 knew-0.034
 struggled-0.034
Token wear
Token ID5806
Feature activation+2.64e+00

Pos logit contributions
 forgot+0.042
 remembered+0.033
 wished+0.033
 knew+0.031
 struggled+0.031
Neg logit contributions
he-0.041
ay-0.039
room-0.038
ings-0.037
ore-0.037
Token it
Token ID340
Feature activation+3.71e+00

Pos logit contributions
he+0.047
ing+0.044
ore+0.044
ings+0.043
room+0.043
Neg logit contributions
 forgot-0.074
 remembered-0.061
 knew-0.060
 wished-0.060
 learned-0.056
Token.
Token ID13
Feature activation+3.80e+00

Pos logit contributions
ays+0.075
ore+0.070
he+0.069
eddy+0.068
ings+0.068
Neg logit contributions
 forgot-0.094
 knew-0.078
 remembered-0.077
 bit-0.075
 reminded-0.073
Token
Token ID198
Feature activation+5.06e+00

Pos logit contributions
he+0.068
over+0.062
eddy+0.062
ing+0.060
ays+0.060
Neg logit contributions
 forgot-0.104
 knew-0.093
 remembered-0.085
 wished-0.083
 learned-0.082
Token
Token ID198
Feature activation+4.14e+00

Pos logit contributions
ays+0.112
ipping+0.109
ing+0.104
overed+0.104
over+0.102
Neg logit contributions
 forgot-0.104
 knew-0.100
 told-0.088
 learned-0.086
 had-0.084
TokenM
Token ID44
Feature activation+2.22e+00

Pos logit contributions
ays+0.076
eddy+0.076
he+0.074
ing+0.073
ings+0.072
Neg logit contributions
 forgot-0.110
 knew-0.094
 remembered-0.090
 wished-0.086
 learned-0.085
Tokenia
Token ID544
Feature activation-5.35e+00

Pos logit contributions
he+0.048
ore+0.044
ing+0.043
ays+0.042
uggle+0.042
Neg logit contributions
 forgot-0.053
 knew-0.043
 remembered-0.042
 wished-0.039
 knocked-0.038
Token put
Token ID1234
Feature activation+9.88e-01

Pos logit contributions
 forgot+0.089
 known+0.069
 forget+0.066
 forgotten+0.065
 punished+0.060
Neg logit contributions
ing-0.131
he-0.131
 with-0.124
ore-0.123
room-0.121
Token on
Token ID319
Feature activation+1.72e+00

Pos logit contributions
he+0.015
ore+0.014
ings+0.014
ing+0.014
room+0.013
Neg logit contributions
 forgot-0.029
 remembered-0.025
 knew-0.025
 wished-0.024
 knocked-0.023
Token not
Token ID407
Feature activation-2.51e+00

Pos logit contributions
he+0.004
ay+0.004
ing+0.004
ore+0.004
ings+0.004
Neg logit contributions
 forgot-0.002
 remembered-0.001
 knew-0.001
 wished-0.001
 knocked-0.001
Token like
Token ID588
Feature activation-3.48e-01

Pos logit contributions
 forgot+0.056
 remembered+0.045
 wished+0.045
 knew+0.043
 struggled+0.042
Neg logit contributions
he-0.057
ore-0.050
ay-0.050
room-0.049
ing-0.049
Token to
Token ID284
Feature activation-3.27e+00

Pos logit contributions
 forgot+0.010
 remembered+0.008
 knew+0.008
 wished+0.008
 bit+0.008
Neg logit contributions
he-0.006
ing-0.005
ore-0.005
ings-0.005
ay-0.005
Token wear
Token ID5806
Feature activation+1.40e+00

Pos logit contributions
 forgot+0.069
 remembered+0.054
 knew+0.053
 wished+0.052
 struggled+0.047
Neg logit contributions
he-0.080
ings-0.074
room-0.073
ay-0.073
ore-0.073
Token warm
Token ID5814
Feature activation+2.87e+00

Pos logit contributions
he+0.028
ing+0.026
ay+0.026
ore+0.026
ings+0.025
Neg logit contributions
 forgot-0.036
 remembered-0.029
 knew-0.029
 wished-0.029
 knocked-0.027
Token clothes
Token ID8242
Feature activation+5.31e+00

Pos logit contributions
he+0.085
ore+0.077
room+0.076
ay+0.076
ing+0.075
Neg logit contributions
 forgot-0.047
 remembered-0.035
 wished-0.034
 knew-0.033
 knocked-0.031
Token.
Token ID13
Feature activation+3.43e+00

Pos logit contributions
ore+0.063
ay+0.062
he+0.061
room+0.060
ays+0.059
Neg logit contributions
 forgot-0.175
 remembered-0.145
 knew-0.144
 wished-0.143
 bit-0.141
Token They
Token ID1119
Feature activation-7.86e+00

Pos logit contributions
he+0.076
ing+0.068
ay+0.065
ore+0.065
uggle+0.063
Neg logit contributions
 forgot-0.083
 knew-0.067
 wished-0.064
 remembered-0.063
 knocked-0.061
Token liked
Token ID8288
Feature activation-1.51e-01

Pos logit contributions
 forgot+0.117
 known+0.116
 shown+0.116
 payoff+0.116
 setup+0.108
Neg logit contributions
he-0.196
.-0.150
 at-0.150
 with-0.146
room-0.144
Token to
Token ID284
Feature activation-3.12e+00

Pos logit contributions
 forgot+0.005
 remembered+0.004
 knew+0.004
 struggled+0.004
 wished+0.004
Neg logit contributions
ing-0.002
ore-0.002
he-0.002
time-0.002
eddy-0.002
Token wear
Token ID5806
Feature activation+1.09e+00

Pos logit contributions
 forgot+0.071
 remembered+0.057
 knew+0.055
 wished+0.055
 regretted+0.052
Neg logit contributions
he-0.070
room-0.063
ings-0.063
uggle-0.062
ay-0.062
Token it
Token ID340
Feature activation-2.13e+00

Pos logit contributions
he+0.005
ing+0.004
ay+0.004
ore+0.004
ings+0.004
Neg logit contributions
 forgot-0.009
 remembered-0.008
 knew-0.008
 wished-0.008
 knocked-0.007
Token was
Token ID373
Feature activation-1.33e+00

Pos logit contributions
 forgot+0.045
 wished+0.035
 remembered+0.035
 wasted+0.033
 knew+0.033
Neg logit contributions
ing-0.046
he-0.046
ings-0.044
room-0.043
time-0.042
Token time
Token ID640
Feature activation+2.54e+00

Pos logit contributions
 forgot+0.035
 wished+0.028
 remembered+0.028
 knew+0.028
 bit+0.026
Neg logit contributions
he-0.025
ings-0.023
ay-0.023
ing-0.023
ore-0.022
Token to
Token ID284
Feature activation-3.08e+00

Pos logit contributions
he+0.040
ing+0.037
ay+0.036
ore+0.036
ings+0.036
Neg logit contributions
 forgot-0.075
 remembered-0.063
 knew-0.062
 wished-0.062
 knocked-0.058
Token go
Token ID467
Feature activation+1.80e+00

Pos logit contributions
 forgot+0.066
 remembered+0.050
 knew+0.049
 wished+0.049
 bit+0.045
Neg logit contributions
he-0.075
ing-0.071
ore-0.070
ay-0.070
ings-0.070
Token home
Token ID1363
Feature activation+2.66e+00

Pos logit contributions
he+0.036
room+0.035
ings+0.035
ay+0.035
ing+0.034
Neg logit contributions
 forgot-0.043
 remembered-0.034
 knew-0.034
 wished-0.033
 forget-0.032
Token.
Token ID13
Feature activation+2.95e+00

Pos logit contributions
he+0.030
ay+0.029
ays+0.027
ing+0.026
ore+0.026
Neg logit contributions
 forgot-0.091
 knew-0.077
 remembered-0.077
 wished-0.076
 bit-0.073
Token They
Token ID1119
Feature activation-7.34e+00

Pos logit contributions
he+0.050
ing+0.045
ay+0.044
ings+0.044
ore+0.044
Neg logit contributions
 forgot-0.085
 remembered-0.070
 knew-0.070
 wished-0.069
 knocked-0.065
Token waved
Token ID26834
Feature activation+1.93e+00

Pos logit contributions
 forgot+0.123
 pap+0.102
 cruel+0.098
 known+0.098
 destructive+0.096
Neg logit contributions
he-0.186
ay-0.145
 with-0.143
room-0.141
 at-0.140
Token goodbye
Token ID24829
Feature activation+1.88e+00

Pos logit contributions
he+0.038
ore+0.034
room+0.034
ings+0.034
ay+0.034
Neg logit contributions
 forgot-0.051
 remembered-0.042
 knew-0.042
 wished-0.042
 bit-0.039
Token to
Token ID284
Feature activation-4.35e-01

Pos logit contributions
he+0.028
ay+0.025
ore+0.025
ings+0.025
ing+0.025
Neg logit contributions
 forgot-0.058
 remembered-0.048
 knew-0.048
 wished-0.048
 bit-0.045
Token named
Token ID3706
Feature activation+2.10e+00

Pos logit contributions
he+0.045
ings+0.043
ing+0.042
ay+0.042
ays+0.042
Neg logit contributions
 forgot-0.026
 knew-0.019
 remembered-0.019
 wished-0.019
 bit-0.017
Token Lily
Token ID20037
Feature activation+6.51e-01

Pos logit contributions
he+0.048
ings+0.046
ing+0.046
ore+0.045
ays+0.045
Neg logit contributions
 forgot-0.049
 knew-0.038
 remembered-0.037
 wished-0.036
 knocked-0.035
Token who
Token ID508
Feature activation-2.49e+00

Pos logit contributions
he+0.008
ings+0.008
ays+0.008
ay+0.007
ing+0.007
Neg logit contributions
 forgot-0.021
 knew-0.019
 wished-0.019
 remembered-0.019
 learned-0.018
Token loved
Token ID6151
Feature activation+1.04e+00

Pos logit contributions
 forgot+0.046
 remembered+0.032
 wished+0.031
 struggled+0.030
 forget+0.029
Neg logit contributions
he-0.067
ay-0.061
ore-0.061
ing-0.060
uggle-0.060
Token to
Token ID284
Feature activation-1.16e+00

Pos logit contributions
he+0.012
ing+0.012
ore+0.012
eddy+0.011
ay+0.011
Neg logit contributions
 forgot-0.035
 remembered-0.029
 knew-0.028
 wished-0.028
 struggled-0.027
Token read
Token ID1100
Feature activation+4.77e+00

Pos logit contributions
 forgot+0.023
 remembered+0.018
 knew+0.018
 wished+0.017
 knocked+0.016
Neg logit contributions
he-0.029
ing-0.028
ings-0.028
ore-0.028
ay-0.027
Token.
Token ID13
Feature activation+3.54e+00

Pos logit contributions
ays+0.134
he+0.132
ore+0.132
uck+0.131
ing+0.127
Neg logit contributions
 forgot-0.063
 knew-0.054
 wished-0.053
 sc-0.051
 learned-0.049
Token She
Token ID1375
Feature activation-6.70e+00

Pos logit contributions
ing+0.073
ore+0.072
he+0.070
ings+0.070
ach+0.069
Neg logit contributions
 forgot-0.081
 knew-0.068
 had-0.063
 learned-0.063
 remembered-0.062
Token had
Token ID550
Feature activation-1.03e+00

Pos logit contributions
 forgot+0.108
 punished+0.084
 forget+0.084
 unknow+0.082
 claw+0.080
Neg logit contributions
he-0.162
 with-0.145
time-0.142
room-0.140
ay-0.140
Token a
Token ID257
Feature activation-2.08e+00

Pos logit contributions
 forgot+0.032
 remembered+0.026
 knew+0.026
 struggled+0.025
 bit+0.025
Neg logit contributions
he-0.014
ay-0.013
room-0.012
ing-0.012
time-0.011
Token unique
Token ID3748
Feature activation-2.27e+00

Pos logit contributions
 forgot+0.055
 wished+0.042
 struggled+0.042
 remembered+0.042
 punished+0.040
Neg logit contributions
ay-0.040
he-0.039
ings-0.036
time-0.036
ummy-0.036
Token He
Token ID679
Feature activation-8.63e+00

Pos logit contributions
he+0.038
ings+0.034
ing+0.034
ore+0.034
ay+0.033
Neg logit contributions
 forgot-0.070
 knew-0.060
 wished-0.058
 remembered-0.058
 knocked-0.055
Token hugged
Token ID41130
Feature activation+2.61e+00

Pos logit contributions
 forgot+0.122
 known+0.118
chie+0.112
 sizes+0.111
 supervision+0.111
Neg logit contributions
he-0.190
 with-0.186
eddy-0.175
ay-0.166
 and-0.165
Token his
Token ID465
Feature activation-1.25e+00

Pos logit contributions
he+0.059
ore+0.057
ays+0.056
ings+0.056
ing+0.056
Neg logit contributions
 forgot-0.060
 knew-0.049
 remembered-0.048
 wished-0.047
 knocked-0.043
Token mom
Token ID1995
Feature activation+1.87e+00

Pos logit contributions
 forgot+0.030
 wished+0.023
 remembered+0.023
 knew+0.023
 struggled+0.022
Neg logit contributions
he-0.026
ay-0.025
ings-0.024
eddy-0.024
ing-0.024
Token and
Token ID290
Feature activation-6.01e+00

Pos logit contributions
he+0.028
ings+0.028
ays+0.028
room+0.027
ore+0.027
Neg logit contributions
 forgot-0.055
 knew-0.047
 remembered-0.046
 wished-0.045
 bit-0.045
Token said
Token ID531
Feature activation+1.12e+00

Pos logit contributions
 forgot+0.103
chie+0.095
'd+0.094
 tricked+0.087
 known+0.086
Neg logit contributions
 with-0.125
he-0.120
eddy-0.117
ing-0.113
 from-0.109
Token thank
Token ID5875
Feature activation+3.43e+00

Pos logit contributions
he+0.019
ays+0.018
room+0.017
ay+0.017
ings+0.017
Neg logit contributions
 forgot-0.033
 knew-0.029
 remembered-0.028
 wished-0.028
 learned-0.026
Token you
Token ID345
Feature activation+2.39e+00

Pos logit contributions
ays+0.083
ings+0.079
ay+0.077
he+0.076
overed+0.075
Neg logit contributions
 forgot-0.070
 knew-0.056
 remembered-0.055
 wished-0.052
 bit-0.051
Token.
Token ID13
Feature activation+2.47e+00

Pos logit contributions
he+0.046
ays+0.044
ings+0.042
ay+0.042
room+0.042
Neg logit contributions
 forgot-0.064
 knew-0.053
 remembered-0.051
 wished-0.049
 bit-0.048
Token From
Token ID3574
Feature activation-1.31e-01

Pos logit contributions
he+0.048
ings+0.044
ing+0.043
ore+0.043
ay+0.042
Neg logit contributions
 forgot-0.066
 knew-0.055
 remembered-0.054
 wished-0.054
 knocked-0.050
Token that
Token ID326
Feature activation-1.12e+00

Pos logit contributions
 forgot+0.004
 wished+0.003
 knew+0.003
 remembered+0.003
 learned+0.003
Neg logit contributions
ings-0.002
room-0.002
he-0.002
ay-0.002
ore-0.002
Token to
Token ID284
Feature activation-1.28e+00

Pos logit contributions
he+0.015
ing+0.014
ay+0.014
ings+0.014
ore+0.014
Neg logit contributions
 forgot-0.032
 remembered-0.026
 knew-0.026
 wished-0.026
 bit-0.024
Token set
Token ID900
Feature activation+4.36e+00

Pos logit contributions
 forgot+0.026
 remembered+0.020
 wished+0.019
 knew+0.019
 struggled+0.017
Neg logit contributions
he-0.032
ing-0.030
ay-0.030
ings-0.030
ore-0.030
Token.
Token ID13
Feature activation+3.37e+00

Pos logit contributions
ays+0.069
he+0.063
iffs+0.060
ore+0.060
aps+0.059
Neg logit contributions
 forgot-0.118
 knew-0.104
 bit-0.102
 remembered-0.099
 wished-0.098
Token
Token ID220
Feature activation+4.90e+00

Pos logit contributions
he+0.055
ing+0.052
ings+0.051
ay+0.049
ays+0.049
Neg logit contributions
 forgot-0.097
 knew-0.081
 wished-0.080
 remembered-0.080
 knocked-0.075
Token
Token ID198
Feature activation+4.30e+00

Pos logit contributions
he+0.077
ing+0.075
ays+0.070
room+0.070
ement+0.069
Neg logit contributions
 forgot-0.129
 knew-0.109
 remembered-0.103
 learned-0.102
 wished-0.101
Token
Token ID198
Feature activation+3.21e+00

Pos logit contributions
ays+0.054
ing+0.054
he+0.054
ings+0.052
ay+0.050
Neg logit contributions
 forgot-0.134
 knew-0.117
 remembered-0.114
 wished-0.114
 learned-0.110
TokenHe
Token ID1544
Feature activation-7.02e+00

Pos logit contributions
he+0.051
ing+0.046
ay+0.045
ore+0.045
ings+0.045
Neg logit contributions
 forgot-0.096
 remembered-0.080
 knew-0.079
 wished-0.079
 bit-0.075
Token slowly
Token ID6364
Feature activation-4.68e+00

Pos logit contributions
 forgot+0.131
 robbers+0.109
 known+0.100
 stolen+0.100
 forget+0.098
Neg logit contributions
he-0.153
ing-0.145
 with-0.143
ore-0.140
time-0.138
Token opened
Token ID4721
Feature activation+1.74e+00

Pos logit contributions
 forgot+0.082
 remembered+0.056
 wished+0.055
 knew+0.054
 known+0.054
Neg logit contributions
he-0.120
ing-0.119
ore-0.114
ings-0.109
ays-0.107
Token his
Token ID465
Feature activation-1.86e+00

Pos logit contributions
he+0.041
room+0.039
ay+0.038
ore+0.038
uggle+0.038
Neg logit contributions
 forgot-0.035
 remembered-0.030
 knew-0.029
 wished-0.028
 knocked-0.027
Token eyes
Token ID2951
Feature activation+4.04e+00

Pos logit contributions
 forgot+0.049
 wished+0.039
 remembered+0.038
 knew+0.038
 struggled+0.036
Neg logit contributions
ings-0.035
ay-0.035
ore-0.034
ing-0.033
he-0.033
Token different
Token ID1180
Feature activation-2.38e-01

Pos logit contributions
 forgot+0.038
 remembered+0.030
 knew+0.029
 wished+0.029
 struggled+0.028
Neg logit contributions
he-0.026
ay-0.024
ings-0.024
room-0.024
ing-0.024
Token hairc
Token ID37363
Feature activation+5.37e+00

Pos logit contributions
 forgot+0.006
 remembered+0.004
 knew+0.004
 wished+0.004
 bit+0.004
Neg logit contributions
he-0.005
ing-0.005
ay-0.005
ore-0.005
ings-0.005
Tokenuts
Token ID5500
Feature activation+3.94e+00

Pos logit contributions
he+0.065
ore+0.064
uggle+0.060
time+0.060
ing+0.057
Neg logit contributions
 forgot-0.159
 knew-0.127
 learned-0.122
 wished-0.122
 wasted-0.121
Token.
Token ID13
Feature activation+2.86e+00

Pos logit contributions
room+0.052
ings+0.052
ay+0.052
he+0.049
uggle+0.049
Neg logit contributions
 forgot-0.120
 knew-0.104
 wished-0.100
 remembered-0.099
 had-0.098
Token She
Token ID1375
Feature activation-7.58e+00

Pos logit contributions
he+0.061
ing+0.056
ore+0.054
ings+0.053
uggle+0.051
Neg logit contributions
 forgot-0.069
 knew-0.058
 wished-0.056
 knocked-0.055
 remembered-0.054
Token asks
Token ID7893
Feature activation+1.73e+00

Pos logit contributions
 forgot+0.167
 known+0.135
 stolen+0.132
 forgotten+0.131
 remembered+0.130
Neg logit contributions
ays-0.148
he-0.144
aps-0.133
ay-0.131
 with-0.129
Token Lily
Token ID20037
Feature activation+5.93e-01

Pos logit contributions
he+0.039
ays+0.037
ore+0.037
ay+0.036
uggle+0.035
Neg logit contributions
 forgot-0.041
 knew-0.036
 remembered-0.035
 wished-0.032
 learned-0.031
Token to
Token ID284
Feature activation-2.77e+00

Pos logit contributions
he+0.010
ay+0.009
room+0.009
ore+0.009
ings+0.009
Neg logit contributions
 forgot-0.017
 remembered-0.015
 knew-0.015
 wished-0.014
 learned-0.013
Token choose
Token ID3853
Feature activation+2.44e+00

Pos logit contributions
 forgot+0.057
 wished+0.043
 remembered+0.043
 knew+0.042
 struggled+0.039
Neg logit contributions
he-0.070
room-0.064
ay-0.064
ings-0.063
ore-0.063
Token one
Token ID530
Feature activation+9.59e-01

Pos logit contributions
he+0.054
ore+0.050
ay+0.049
room+0.049
ing+0.048
Neg logit contributions
 forgot-0.057
 remembered-0.046
 knew-0.046
 wished-0.046
 knocked-0.043
Token.
Token ID13
Feature activation+3.21e+00

Pos logit contributions
he+0.016
room+0.014
ays+0.014
ing+0.014
ings+0.014
Neg logit contributions
 forgot-0.027
 knew-0.024
 bit-0.023
 remembered-0.023
 wished-0.023
Token His
Token ID2399
Feature activation-3.16e+00

Pos logit contributions
he+0.041
ing+0.037
ings+0.037
ay+0.036
ore+0.036
Neg logit contributions
 forgot-0.076
 knew-0.063
 remembered-0.063
 wished-0.063
 learned-0.059
Token friends
Token ID2460
Feature activation-4.77e+00

Pos logit contributions
 forgot+0.077
 wished+0.058
 remembered+0.058
 knew+0.057
 wasted+0.055
Neg logit contributions
he-0.065
ay-0.064
ings-0.062
ing-0.061
room-0.059
Token cheered
Token ID34767
Feature activation+2.74e+00

Pos logit contributions
 forgot+0.085
 pap+0.071
 forgotten+0.060
 robbers+0.060
 elves+0.059
Neg logit contributions
he-0.124
ing-0.109
room-0.105
 with-0.098
ore-0.097
Token him
Token ID683
Feature activation+6.93e-02

Pos logit contributions
he+0.049
ays+0.047
ore+0.046
ings+0.044
room+0.044
Neg logit contributions
 forgot-0.074
 knew-0.064
 remembered-0.062
 wished-0.060
 bit-0.057
Token on
Token ID319
Feature activation+1.88e+00

Pos logit contributions
he+0.001
ays+0.001
ore+0.001
ummy+0.001
room+0.001
Neg logit contributions
 forgot-0.002
 remembered-0.002
 knew-0.002
 bit-0.002
 wished-0.002
Token and
Token ID290
Feature activation-4.26e+00

Pos logit contributions
he+0.033
ays+0.032
ing+0.031
ore+0.031
ay+0.030
Neg logit contributions
 forgot-0.051
 knew-0.043
 wished-0.042
 remembered-0.042
 bit-0.040
Token they
Token ID484
Feature activation-6.83e+00

Pos logit contributions
 forgot+0.088
 stolen+0.071
 remembered+0.071
'd+0.070
 wasted+0.070
Neg logit contributions
 with-0.080
he-0.078
ing-0.076
 together-0.070
ore-0.068
Token all
Token ID477
Feature activation-4.26e+00

Pos logit contributions
 forgot+0.155
 pap+0.127
 forgotten+0.121
 stolen+0.121
 remembered+0.120
Neg logit contributions
he-0.157
ay-0.125
room-0.119
ing-0.119
uggle-0.113
Token went
Token ID1816
Feature activation+1.66e+00

Pos logit contributions
 forgot+0.068
 pap+0.059
 robbers+0.052
 setup+0.050
 stolen+0.049
Neg logit contributions
he-0.113
ing-0.099
ay-0.095
ore-0.093
room-0.093
Token back
Token ID736
Feature activation+6.64e-01

Pos logit contributions
he+0.028
ay+0.025
ings+0.025
room+0.025
ore+0.025
Neg logit contributions
 forgot-0.048
 remembered-0.040
 knew-0.039
 wished-0.039
 knocked-0.037
Token inside
Token ID2641
Feature activation+2.19e+00

Pos logit contributions
he+0.013
room+0.012
ay+0.012
ings+0.011
ays+0.011
Neg logit contributions
 forgot-0.017
 remembered-0.014
 knew-0.014
 wished-0.014
 knocked-0.014
Token sees
Token ID7224
Feature activation+4.14e-01

Pos logit contributions
 forgot+0.077
 remembered+0.058
 wished+0.058
 knew+0.058
 struggled+0.053
Neg logit contributions
he-0.088
ing-0.085
ore-0.084
ings-0.084
ays-0.083
Token the
Token ID262
Feature activation-2.10e+00

Pos logit contributions
he+0.008
ay+0.007
ing+0.007
ings+0.007
ore+0.007
Neg logit contributions
 forgot-0.011
 remembered-0.009
 knew-0.009
 wished-0.009
 knocked-0.008
Token k
Token ID479
Feature activation+2.36e+00

Pos logit contributions
 forgot+0.058
 remembered+0.048
 wished+0.047
 knew+0.047
 struggled+0.044
Neg logit contributions
he-0.039
ay-0.037
ing-0.035
ings-0.035
room-0.034
Tokenite
Token ID578
Feature activation+1.54e+00

Pos logit contributions
ently+0.029
ings+0.029
he+0.029
ment+0.028
room+0.027
Neg logit contributions
 forgot-0.065
 bit-0.059
 remembered-0.055
 knew-0.052
 wasted-0.052
Token.
Token ID13
Feature activation+2.43e+00

Pos logit contributions
ay+0.019
he+0.019
ore+0.018
ings+0.018
ing+0.017
Neg logit contributions
 forgot-0.048
 knew-0.041
 remembered-0.041
 wished-0.040
 bit-0.040
Token The
Token ID383
Feature activation-2.98e+00

Pos logit contributions
he+0.051
ing+0.049
ings+0.045
ays+0.044
ement+0.044
Neg logit contributions
 forgot-0.056
 knew-0.049
 wished-0.047
 bit-0.045
 remembered-0.045
Token k
Token ID479
Feature activation+1.00e+00

Pos logit contributions
 forgot+0.084
 remembered+0.066
 knew+0.064
 wished+0.062
 struggled+0.062
Neg logit contributions
he-0.055
ay-0.054
ings-0.050
room-0.049
eddy-0.049
Tokenite
Token ID578
Feature activation-2.54e+00

Pos logit contributions
he+0.012
ings+0.011
time+0.011
room+0.011
ing+0.011
Neg logit contributions
 forgot-0.030
 bit-0.026
 remembered-0.026
 wished-0.026
 knew-0.025
Token is
Token ID318
Feature activation-2.11e+00

Pos logit contributions
 forgot+0.068
 remembered+0.055
 wished+0.054
 knew+0.054
 known+0.053
Neg logit contributions
he-0.048
ing-0.043
ays-0.043
ings-0.043
ore-0.043
Token red
Token ID2266
Feature activation+4.69e+00

Pos logit contributions
 forgot+0.045
 knew+0.034
 wished+0.033
 remembered+0.033
 cared+0.032
Neg logit contributions
he-0.049
ay-0.047
ings-0.047
ing-0.045
ays-0.045
Token and
Token ID290
Feature activation-3.94e+00

Pos logit contributions
ore+0.094
he+0.093
uggle+0.092
ings+0.088
ay+0.086
Neg logit contributions
 forgot-0.105
 bit-0.093
 knew-0.089
 wished-0.087
 remembered-0.087
Token
Token ID198
Feature activation+3.55e+00

Pos logit contributions
he+0.083
ings+0.079
ay+0.078
room+0.077
ing+0.077
Neg logit contributions
 forgot-0.094
 knew-0.081
 learned-0.078
 wished-0.078
 remembered-0.076
Token
Token ID198
Feature activation+3.09e+00

Pos logit contributions
ays+0.069
ement+0.067
ing+0.066
over+0.065
he+0.065
Neg logit contributions
 forgot-0.080
 knew-0.075
 learned-0.071
 wished-0.069
 remembered-0.065
TokenThey
Token ID2990
Feature activation-7.09e+00

Pos logit contributions
ays+0.064
he+0.063
ore+0.063
ings+0.062
ay+0.062
Neg logit contributions
 forgot-0.073
 remembered-0.059
 knew-0.059
 wished-0.057
 learned-0.057
Token walked
Token ID6807
Feature activation+3.06e+00

Pos logit contributions
 forgot+0.118
 pap+0.099
 wished+0.088
 forgotten+0.087
 stolen+0.087
Neg logit contributions
he-0.191
room-0.167
ay-0.154
ing-0.150
 with-0.144
Token back
Token ID736
Feature activation+6.01e-01

Pos logit contributions
he+0.060
ummy+0.058
ays+0.058
ay+0.057
ore+0.056
Neg logit contributions
 forgot-0.076
 knew-0.064
 remembered-0.063
 learned-0.062
 bit-0.060
Token to
Token ID284
Feature activation+1.11e-02

Pos logit contributions
he+0.013
ement+0.012
over+0.012
room+0.012
ay+0.012
Neg logit contributions
 forgot-0.013
 knew-0.011
 remembered-0.011
 knocked-0.011
 bit-0.010
Token their
Token ID511
Feature activation-2.56e+00

Pos logit contributions
he+0.000
ing+0.000
ay+0.000
ings+0.000
ore+0.000
Neg logit contributions
 forgot-0.000
 knew-0.000
 remembered-0.000
 wished-0.000
 knocked-0.000
Token mom
Token ID1995
Feature activation+9.48e-01

Pos logit contributions
 forgot+0.066
 remembered+0.053
 knew+0.052
 wished+0.051
 known+0.051
Neg logit contributions
he-0.048
eddy-0.046
ings-0.046
ay-0.045
ummy-0.045
Token,
Token ID11
Feature activation-1.55e+00

Pos logit contributions
he+0.016
ings+0.015
ay+0.015
ore+0.015
overed+0.015
Neg logit contributions
 forgot-0.026
 remembered-0.021
 knew-0.021
 learned-0.021
 bit-0.021
Token holding
Token ID4769
Feature activation+1.51e+00

Pos logit contributions
 forgot+0.043
 wished+0.035
 knew+0.035
 struggled+0.035
 remembered+0.035
Neg logit contributions
he-0.024
 with-0.022
ing-0.021
room-0.021
aps-0.020
Token hands
Token ID2832
Feature activation+4.09e+00

Pos logit contributions
he+0.036
ore+0.033
ing+0.032
ummy+0.032
ay+0.032
Neg logit contributions
 forgot-0.029
 remembered-0.024
 bit-0.023
 knew-0.023
 learned-0.023
Token with
Token ID351
Feature activation-1.77e+00

Pos logit contributions
he+0.044
ore+0.042
ays+0.041
ay+0.041
ings+0.040
Neg logit contributions
 forgot-0.080
 knew-0.069
 bit-0.066
 remembered-0.066
 wished-0.065
Token her
Token ID607
Feature activation-1.56e+00

Pos logit contributions
 forgot+0.054
 remembered+0.044
 knew+0.043
 wished+0.043
 knocked+0.040
Neg logit contributions
he-0.028
ay-0.026
ings-0.025
ing-0.025
room-0.025
Token hands
Token ID2832
Feature activation+2.70e+00

Pos logit contributions
 forgot+0.034
 remembered+0.026
 knew+0.025
 wished+0.025
 knocked+0.023
Neg logit contributions
he-0.038
ing-0.035
ay-0.035
ings-0.035
ore-0.035
Token.
Token ID13
Feature activation+4.00e+00

Pos logit contributions
he+0.039
ore+0.037
ay+0.037
ays+0.035
uggle+0.035
Neg logit contributions
 forgot-0.081
 knew-0.069
 remembered-0.069
 wished-0.068
 bit-0.067
Token It
Token ID632
Feature activation-4.68e+00

Pos logit contributions
he+0.077
ing+0.070
ach+0.070
eddy+0.069
over+0.069
Neg logit contributions
 forgot-0.095
 knew-0.083
 knocked-0.077
 wished-0.077
 learned-0.076
Token was
Token ID373
Feature activation-1.74e+00

Pos logit contributions
 forgot+0.098
 pap+0.082
 forgotten+0.079
 remembered+0.078
 robbers+0.078
Neg logit contributions
he-0.096
 with-0.090
room-0.089
ing-0.085
ays-0.085
Token hard
Token ID1327
Feature activation+1.63e+00

Pos logit contributions
 forgot+0.044
 knew+0.036
 wished+0.035
 remembered+0.035
 struggled+0.033
Neg logit contributions
he-0.031
ay-0.030
ings-0.030
ing-0.029
time-0.028
Token to
Token ID284
Feature activation-1.61e+00

Pos logit contributions
he+0.022
ing+0.019
ay+0.019
ore+0.019
ings+0.019
Neg logit contributions
 forgot-0.053
 remembered-0.045
 knew-0.044
 wished-0.044
 bit-0.042
Token move
Token ID1445
Feature activation+4.38e+00

Pos logit contributions
 forgot+0.030
 remembered+0.022
 knew+0.022
 wished+0.021
 knocked+0.020
Neg logit contributions
he-0.042
ing-0.042
ore-0.041
ay-0.041
ays-0.041
Token,
Token ID11
Feature activation-9.97e-01

Pos logit contributions
he+0.059
ore+0.054
ays+0.054
uggle+0.054
ay+0.053
Neg logit contributions
 forgot-0.136
 knew-0.118
 wished-0.115
 remembered-0.115
 bit-0.112
Token but
Token ID475
Feature activation-3.26e+00

Pos logit contributions
 forgot+0.030
 remembered+0.025
 knew+0.025
 wished+0.025
 struggled+0.024
Neg logit contributions
he-0.014
ing-0.013
ings-0.013
room-0.013
ay-0.012
Token and
Token ID290
Feature activation-1.43e+00

Pos logit contributions
 forgot+0.040
 wished+0.031
 remembered+0.031
 knew+0.030
 bit+0.028
Neg logit contributions
he-0.041
ing-0.038
ay-0.038
ore-0.037
ings-0.037
Token Lily
Token ID20037
Feature activation-4.25e+00

Pos logit contributions
 forgot+0.038
 remembered+0.030
 wished+0.030
 knew+0.030
 bit+0.028
Neg logit contributions
he-0.027
ay-0.025
ore-0.025
ing-0.025
ings-0.024
Token were
Token ID547
Feature activation+1.78e+00

Pos logit contributions
 forgot+0.078
 remembered+0.057
 wished+0.056
 knew+0.055
 bit+0.052
Neg logit contributions
he-0.113
ing-0.103
room-0.103
ay-0.102
ings-0.100
Token playing
Token ID2712
Feature activation+3.40e+00

Pos logit contributions
he+0.033
ing+0.031
ay+0.031
ore+0.030
ays+0.030
Neg logit contributions
 forgot-0.049
 wished-0.039
 knew-0.039
 remembered-0.039
 bit-0.037
Token in
Token ID287
Feature activation+1.76e+00

Pos logit contributions
ay+0.049
he+0.046
ays+0.046
eddy+0.045
ummy+0.045
Neg logit contributions
 forgot-0.098
 remembered-0.086
 knew-0.085
 bit-0.084
 wished-0.083
Token the
Token ID262
Feature activation-1.10e+00

Pos logit contributions
he+0.034
ing+0.032
ore+0.031
ay+0.031
ings+0.031
Neg logit contributions
 forgot-0.046
 remembered-0.037
 knew-0.037
 wished-0.037
 knocked-0.035
Token living
Token ID2877
Feature activation+3.10e-01

Pos logit contributions
 forgot+0.026
 remembered+0.020
 knew+0.020
 wished+0.020
 struggled+0.018
Neg logit contributions
he-0.025
ing-0.023
room-0.023
ay-0.023
ore-0.023
Token room
Token ID2119
Feature activation+4.91e+00

Pos logit contributions
he+0.008
ing+0.008
ore+0.008
ings+0.007
ay+0.007
Neg logit contributions
 forgot-0.006
 remembered-0.005
 knew-0.005
 wished-0.004
 bit-0.004
Token.
Token ID13
Feature activation+2.91e+00

Pos logit contributions
he+0.053
ay+0.050
ore+0.047
ings+0.046
ays+0.046
Neg logit contributions
 forgot-0.167
 knew-0.143
 remembered-0.143
 wished-0.142
 bit-0.138
Token They
Token ID1119
Feature activation-6.04e+00

Pos logit contributions
he+0.056
ing+0.055
ore+0.054
ings+0.051
uggle+0.050
Neg logit contributions
 forgot-0.078
 remembered-0.062
 knew-0.062
 wished-0.059
 knocked-0.059
Token liked
Token ID8288
Feature activation+4.47e-01

Pos logit contributions
 forgot+0.110
 pap+0.098
 setup+0.094
 regretted+0.093
 unknow+0.093
Neg logit contributions
he-0.144
room-0.128
 with-0.119
ay-0.112
 at-0.110
Token best
Token ID1266
Feature activation-6.91e-01

Pos logit contributions
 forgot+0.022
 remembered+0.017
 knew+0.017
 wished+0.017
 bit+0.015
Neg logit contributions
he-0.026
ing-0.024
ay-0.024
ore-0.024
ings-0.024
Token friend
Token ID1545
Feature activation+6.14e-01

Pos logit contributions
 forgot+0.014
 remembered+0.011
 wished+0.010
 knew+0.010
 knocked+0.009
Neg logit contributions
he-0.017
ay-0.016
ing-0.016
ore-0.016
ings-0.016
Token,
Token ID11
Feature activation-2.64e+00

Pos logit contributions
he+0.008
ore+0.007
ays+0.007
ings+0.007
ay+0.007
Neg logit contributions
 forgot-0.020
 remembered-0.017
 knew-0.017
 wished-0.017
 knocked-0.016
Token Tommy
Token ID19919
Feature activation+4.82e-01

Pos logit contributions
 forgot+0.070
 remembered+0.056
 knew+0.055
 wished+0.055
 bit+0.052
Neg logit contributions
he-0.050
ing-0.046
ore-0.046
ay-0.046
ings-0.045
Token.
Token ID13
Feature activation+2.80e+00

Pos logit contributions
he+0.004
ing+0.003
ore+0.003
ings+0.003
ay+0.003
Neg logit contributions
 forgot-0.018
 remembered-0.016
 wished-0.016
 knew-0.016
 bit-0.015
Token Tommy
Token ID19919
Feature activation-5.80e+00

Pos logit contributions
he+0.051
ore+0.048
ing+0.046
ays+0.046
ings+0.046
Neg logit contributions
 forgot-0.077
 knew-0.064
 remembered-0.063
 wished-0.062
 learned-0.059
Token said
Token ID531
Feature activation+2.21e-01

Pos logit contributions
 forgot+0.107
 known+0.085
 forgotten+0.085
 stolen+0.084
 vitamins+0.081
Neg logit contributions
he-0.128
ing-0.124
 with-0.120
ore-0.119
,-0.111
Token,
Token ID11
Feature activation+9.69e-01

Pos logit contributions
he+0.003
ays+0.003
ay+0.003
ore+0.003
ings+0.003
Neg logit contributions
 forgot-0.007
 knew-0.006
 remembered-0.006
 wished-0.006
 knocked-0.006
Token "
Token ID366
Feature activation+1.46e-01

Pos logit contributions
he+0.021
ays+0.020
ings+0.020
ay+0.020
room+0.019
Neg logit contributions
 forgot-0.023
 knew-0.019
 remembered-0.019
 wished-0.018
 knocked-0.017
TokenYou
Token ID1639
Feature activation-2.99e+00

Pos logit contributions
he+0.003
ings+0.003
ore+0.003
ing+0.003
ay+0.003
Neg logit contributions
 forgot-0.004
 remembered-0.003
 knew-0.003
 wished-0.003
 bit-0.003
Token can
Token ID460
Feature activation-1.20e+00

Pos logit contributions
 forgot+0.059
 pap+0.047
 wished+0.046
 forgotten+0.045
 knocked+0.044
Neg logit contributions
he-0.068
ay-0.064
room-0.058
ing-0.056
ore-0.055
Token
Token ID198
Feature activation+3.83e+00

Pos logit contributions
ing+0.053
he+0.050
ings+0.050
ays+0.048
ore+0.048
Neg logit contributions
 forgot-0.070
 knew-0.060
 had-0.058
 remembered-0.056
 learned-0.056
Token
Token ID198
Feature activation+3.17e+00

Pos logit contributions
ays+0.081
ipping+0.078
ing+0.076
ement+0.073
overed+0.073
Neg logit contributions
 forgot-0.084
 knew-0.077
 had-0.073
 learned-0.070
 told-0.069
TokenOne
Token ID3198
Feature activation-1.86e+00

Pos logit contributions
ays+0.074
he+0.073
ore+0.073
ings+0.072
ing+0.072
Neg logit contributions
 forgot-0.067
 knew-0.054
 remembered-0.053
 wished-0.051
 knocked-0.050
Token day
Token ID1110
Feature activation+3.57e+00

Pos logit contributions
 forgot+0.039
 knew+0.031
 remembered+0.030
 wished+0.028
 bit+0.027
Neg logit contributions
ays-0.045
he-0.045
ings-0.045
ing-0.043
ore-0.043
Token,
Token ID11
Feature activation+2.20e+00

Pos logit contributions
he+0.072
eddy+0.065
ays+0.064
ore+0.060
room+0.060
Neg logit contributions
 forgot-0.095
 remembered-0.082
 knew-0.082
 knocked-0.080
 bit-0.079
Token Kitty
Token ID21168
Feature activation-4.88e+00

Pos logit contributions
he+0.047
ore+0.045
ays+0.044
ings+0.043
eddy+0.043
Neg logit contributions
 forgot-0.055
 knew-0.044
 remembered-0.044
 bit-0.042
 wished-0.042
Token found
Token ID1043
Feature activation+5.33e-01

Pos logit contributions
 forgot+0.129
 remembered+0.105
 wished+0.105
 knew+0.104
 known+0.100
Neg logit contributions
he-0.087
ing-0.081
ore-0.078
room-0.076
ay-0.076
Token a
Token ID257
Feature activation-3.17e+00

Pos logit contributions
he+0.011
ay+0.011
ore+0.010
ings+0.010
ing+0.010
Neg logit contributions
 forgot-0.013
 remembered-0.010
 knew-0.010
 wished-0.010
 knocked-0.010
Token pretty
Token ID2495
Feature activation-1.34e+00

Pos logit contributions
 forgot+0.080
 wished+0.064
 remembered+0.063
 struggled+0.061
 learned+0.060
Neg logit contributions
he-0.060
ay-0.060
ings-0.058
ing-0.058
time-0.057
Token gem
Token ID16840
Feature activation+3.50e+00

Pos logit contributions
 forgot+0.030
 remembered+0.023
 knew+0.023
 wished+0.023
 bit+0.021
Neg logit contributions
he-0.031
ing-0.029
ay-0.029
ore-0.029
ings-0.029
Token under
Token ID739
Feature activation+1.30e+00

Pos logit contributions
he+0.045
ings+0.043
ay+0.042
room+0.042
ore+0.040
Neg logit contributions
 forgot-0.110
 knew-0.095
 remembered-0.094
 wished-0.094
 bit-0.093
Token and
Token ID290
Feature activation-3.89e-01

Pos logit contributions
 forgot+0.175
 stolen+0.151
 wished+0.150
 forgotten+0.147
 wasted+0.146
Neg logit contributions
ing-0.076
he-0.073
ay-0.065
room-0.065
ore-0.062
Token Max
Token ID5436
Feature activation-6.31e+00

Pos logit contributions
 forgot+0.010
 remembered+0.008
 knew+0.008
 wished+0.008
 knocked+0.007
Neg logit contributions
he-0.008
ore-0.007
ings-0.007
ing-0.007
ay-0.007
Token became
Token ID2627
Feature activation-1.58e+00

Pos logit contributions
 forgot+0.104
 stolen+0.082
 punished+0.079
 wished+0.077
 forgotten+0.077
Neg logit contributions
he-0.160
ing-0.157
ay-0.153
room-0.151
uggle-0.147
Token good
Token ID922
Feature activation+7.58e-02

Pos logit contributions
 forgot+0.034
 wished+0.027
 remembered+0.026
 knew+0.026
 knocked+0.025
Neg logit contributions
he-0.036
ing-0.035
ay-0.034
ings-0.034
room-0.034
Token friends
Token ID2460
Feature activation+2.57e+00

Pos logit contributions
he+0.002
ore+0.002
ings+0.002
ay+0.002
ing+0.002
Neg logit contributions
 forgot-0.001
 remembered-0.001
 knew-0.001
 wished-0.001
 learned-0.001
Token and
Token ID290
Feature activation-5.20e+00

Pos logit contributions
ays+0.041
he+0.040
ings+0.040
ore+0.039
room+0.037
Neg logit contributions
 forgot-0.074
 knew-0.064
 learned-0.062
 remembered-0.062
 wished-0.061
Token danced
Token ID39480
Feature activation+3.58e+00

Pos logit contributions
 forgot+0.091
 stolen+0.081
 punished+0.080
 unknow+0.078
 wished+0.076
Neg logit contributions
ing-0.113
he-0.108
uggle-0.107
room-0.107
 with-0.105
Token together
Token ID1978
Feature activation+8.33e-01

Pos logit contributions
ays+0.052
he+0.049
ore+0.047
ummy+0.047
ings+0.047
Neg logit contributions
 forgot-0.107
 knew-0.093
 learned-0.090
 remembered-0.089
 wished-0.086
Token every
Token ID790
Feature activation-1.99e+00

Pos logit contributions
he+0.012
ay+0.011
ays+0.011
ore+0.011
ings+0.011
Neg logit contributions
 forgot-0.026
 knew-0.022
 remembered-0.022
 wished-0.021
 bit-0.020
Token time
Token ID640
Feature activation+2.15e+00

Pos logit contributions
 forgot+0.050
 remembered+0.039
 wished+0.039
 knew+0.038
 knocked+0.035
Neg logit contributions
he-0.042
ing-0.039
ay-0.039
ore-0.039
room-0.038
Token they
Token ID484
Feature activation-5.98e+00

Pos logit contributions
he+0.050
ore+0.047
ay+0.046
ings+0.045
room+0.044
Neg logit contributions
 forgot-0.050
 remembered-0.038
 knew-0.038
 wished-0.038
 knocked-0.035
Token
Token ID198
Feature activation+3.68e+00

Pos logit contributions
ays+0.104
ipping+0.093
ing+0.092
overed+0.092
ement+0.092
Neg logit contributions
 forgot-0.109
 knew-0.102
 learned-0.091
 told-0.090
 had-0.090
TokenHe
Token ID1544
Feature activation-7.44e+00

Pos logit contributions
he+0.061
ing+0.056
ay+0.056
ore+0.056
ings+0.055
Neg logit contributions
 forgot-0.108
 remembered-0.089
 knew-0.089
 wished-0.088
 bit-0.083
Token did
Token ID750
Feature activation+9.55e-01

Pos logit contributions
 forgot+0.120
 forget+0.111
 known+0.102
 forgotten+0.095
 forgetting+0.093
Neg logit contributions
ing-0.176
he-0.167
room-0.164
 with-0.164
uck-0.160
Token so
Token ID523
Feature activation-7.95e-01

Pos logit contributions
he+0.028
ay+0.027
ing+0.026
ore+0.026
ays+0.026
Neg logit contributions
 forgot-0.016
 remembered-0.011
 knew-0.011
 wished-0.011
 knocked-0.010
Token.
Token ID13
Feature activation+2.71e+00

Pos logit contributions
 forgot+0.017
 knew+0.014
 remembered+0.012
 wished+0.012
 learned+0.012
Neg logit contributions
eddy-0.020
ays-0.019
ay-0.019
uggle-0.019
ings-0.019
Token He
Token ID679
Feature activation-8.08e+00

Pos logit contributions
he+0.047
ings+0.043
ing+0.043
ore+0.041
ay+0.041
Neg logit contributions
 forgot-0.076
 knew-0.064
 remembered-0.063
 wished-0.063
 knocked-0.060
Token took
Token ID1718
Feature activation+1.93e-02

Pos logit contributions
 forgot+0.126
 forget+0.116
 known+0.114
 forgotten+0.105
 stolen+0.104
Neg logit contributions
he-0.185
room-0.183
ing-0.182
 with-0.181
ore-0.167
Token off
Token ID572
Feature activation+1.89e+00

Pos logit contributions
he+0.000
ing+0.000
ore+0.000
ings+0.000
ay+0.000
Neg logit contributions
 forgot-0.001
 remembered-0.000
 knew-0.000
 wished-0.000
 knocked-0.000
Token his
Token ID465
Feature activation-1.96e+00

Pos logit contributions
he+0.036
ing+0.033
ay+0.033
ore+0.033
ings+0.033
Neg logit contributions
 forgot-0.051
 remembered-0.041
 wished-0.041
 knew-0.041
 bit-0.038
Token head
Token ID1182
Feature activation+3.24e+00

Pos logit contributions
 forgot+0.042
 wished+0.032
 remembered+0.031
 knew+0.031
 known+0.029
Neg logit contributions
he-0.046
ay-0.044
ing-0.044
ore-0.044
ings-0.044
Token and
Token ID290
Feature activation-5.93e+00

Pos logit contributions
he+0.055
ay+0.051
ays+0.050
ings+0.050
ore+0.049
Neg logit contributions
 forgot-0.089
 knew-0.077
 remembered-0.075
 wished-0.073
 bit-0.072
Token was
Token ID373
Feature activation-2.69e+00

Pos logit contributions
 forgot+0.101
 known+0.084
 forget+0.081
 forgotten+0.077
 stolen+0.076
Neg logit contributions
he-0.128
ing-0.125
 with-0.118
room-0.116
uck-0.116
Token confused
Token ID10416
Feature activation+3.24e+00

Pos logit contributions
 forgot+0.065
 knew+0.048
 remembered+0.047
 wished+0.047
 forget+0.047
Neg logit contributions
he-0.055
 with-0.050
ing-0.050
over-0.050
room-0.050
Token,
Token ID11
Feature activation-1.02e+00

Pos logit contributions
room+0.060
uggle+0.060
he+0.060
ays+0.059
ay+0.056
Neg logit contributions
 forgot-0.085
 knew-0.077
 remembered-0.073
 wished-0.071
 bit-0.070
Token he
Token ID339
Feature activation-6.29e+00

Pos logit contributions
 forgot+0.032
 remembered+0.027
 wasted+0.027
 known+0.026
 tricked+0.026
Neg logit contributions
he-0.010
ing-0.010
 with-0.010
time-0.009
ummy-0.008
Token had
Token ID550
Feature activation-2.02e+00

Pos logit contributions
 forgot+0.093
 known+0.078
 pap+0.075
 forget+0.073
 payoff+0.073
Neg logit contributions
he-0.144
ing-0.144
time-0.141
 at-0.134
 with-0.134
Token never
Token ID1239
Feature activation-5.55e+00

Pos logit contributions
 forgot+0.038
 forget+0.030
 pap+0.029
 elves+0.028
 sizes+0.026
Neg logit contributions
time-0.039
he-0.039
 with-0.038
 at-0.038
ore-0.037
Token seen
Token ID1775
Feature activation+9.00e-02

Pos logit contributions
 forgot+0.094
 pap+0.080
 wished+0.075
 majesty+0.073
 elves+0.069
Neg logit contributions
he-0.120
ing-0.116
ore-0.109
 with-0.108
.-0.108
Token a
Token ID257
Feature activation-3.21e+00

Pos logit contributions
he+0.002
ing+0.002
ay+0.002
ore+0.002
ings+0.002
Neg logit contributions
 forgot-0.002
 remembered-0.002
 knew-0.002
 wished-0.002
 bit-0.002
Token turkey
Token ID26876
Feature activation+1.75e+00

Pos logit contributions
 forgot+0.088
 wished+0.072
 remembered+0.070
 knew+0.068
 struggled+0.067
Neg logit contributions
he-0.057
ing-0.054
ore-0.054
ay-0.053
ings-0.052
Token before
Token ID878
Feature activation+1.73e+00

Pos logit contributions
he+0.029
ay+0.027
ings+0.027
ing+0.026
ays+0.026
Neg logit contributions
 forgot-0.051
 remembered-0.042
 knew-0.042
 wished-0.041
 bit-0.040
Token.
Token ID13
Feature activation+2.62e+00

Pos logit contributions
he+0.025
ay+0.023
uggle+0.023
room+0.022
ays+0.022
Neg logit contributions
 forgot-0.051
 knew-0.045
 wished-0.044
 remembered-0.044
 knocked-0.043