TOP ACTIVATIONS
MAX = 0.110

says , sad . " Maybe the snail did not like
Maybe it is sad . Maybe it is lonely ."
. She thought , " Maybe he is hungry or lonely
was scared too . " Maybe we are lost ."
is scared too . " Maybe we should go out ."
cloud . " Maybe we can ask it to
sorry . " Maybe we are both bad fisher
too . " Maybe we should say sorry to
. He thinks , " Maybe I was wrong . Maybe
happy . They ask mom if they can talk to the
want to move . " Maybe I can stay a little
it . " Maybe we can find another one
hungry . They ask Mom if they can have a snack
don 't have permission . Maybe we should go back ."
was sad and didn 't know what to do .
say sorry and hug . Maybe later , they would .
them . " Maybe we can fly with the
afraid and careful . " Maybe there are monsters or ghosts
" Me too . Maybe we should sell the book
. Maybe they forgot . Maybe they are late . Maybe

MIN ACTIVATIONS
MIN = -0.148

different styles like stars or spir als . Everyone was so
on the lake in a kay ak . As they were
the air , leaving a spir aling trail of black ink
liked to paddle in the kay ak and splash in the
butterfly , tw irling and spir aling in the air .
saw a man in a kay ak . The man said
Fro ggy , in a kay ak . Fro ggy said
for a ride in the kay ak and her mom said
forth in the sky , spir alling higher and higher .
can we go in the kay ak ?" Her dad said
river . They saw a kay ak on the shore .
hopped on , and started ped alling . She felt like
and Anne jumped in the kay ak and padd led away
day , they saw a kay ak on the shore .
on their bikes and started ped aling . When
saw a big , red kay ak . Lily said ,
food . He saw a kay ak by the river .
to join them in the kay ak . "
Anne began to build a kay ak . They found some
hose . They have a kay ak toy . It is

INTERVAL 0.046 - 0.110
CONTAINS 4.902%

cooking . " Mom my , what are you
ily says , " Ben , do you want to marry
faster . The end . <|endoftext|> Once upon a
he was playing with his friend , a little bird named
job . I don 't know what to do ."

INTERVAL -0.019 - 0.046
CONTAINS 69.248%

But Tim my , it 's cold outside . You need
to his mom and ate his vegetables . <|endoftext|> Once upon
to work together . They learned that even though they were
on a ch ute . He had to go down the
special - a spring ! Inside were lots of colourful coils

INTERVAL -0.083 - -0.019
CONTAINS 25.612%

to him . Tim was so happy and thanked the lady
was still by her side . Mom my said ,
a box . The man looked up at the girl ,
do ggy !" and he ran away .
had to come with her . She said he would be

INTERVAL -0.148 - -0.083
CONTAINS 0.238%

<|endoftext|> One day , a sw an named Sam was swimming
play with her toys and c uddle with her t eddy
messy . Like when I sne e ze . Ach oo
" R ib bit , rib bit !" Lily says .
parent smiled and said you will
Token says
Token ID1139
Feature activation+1.52e-02

Pos logit contributions
aling+0.023
alling+0.021
irl+0.019
astics+0.019
ophone+0.018
Neg logit contributions
guard-0.022
 poisoning-0.022
 injuries-0.021
 payoff-0.020
 last-0.020
Token,
Token ID11
Feature activation+2.10e-02

Pos logit contributions
guard+0.015
 poisoning+0.014
 injuries+0.014
 payoff+0.012
 unfortunately+0.012
Neg logit contributions
aling-0.032
alling-0.029
astics-0.027
irl-0.027
 sings-0.027
Token sad
Token ID6507
Feature activation+4.76e-02

Pos logit contributions
guard+0.031
 poisoning+0.030
 injuries+0.030
 payoff+0.027
 unfortunately+0.027
Neg logit contributions
aling-0.034
alling-0.031
astics-0.028
irl-0.028
 sings-0.027
Token.
Token ID13
Feature activation-2.94e-02

Pos logit contributions
 poisoning+0.069
 injuries+0.067
guard+0.063
 unfortunately+0.063
 payoff+0.060
Neg logit contributions
aling-0.072
 dances-0.064
 sings-0.064
alling-0.062
astics-0.062
Token "
Token ID366
Feature activation-1.47e-02

Pos logit contributions
aling+0.034
alling+0.028
 sings+0.027
astics+0.026
 Hide+0.026
Neg logit contributions
 injuries-0.053
 poisoning-0.053
guard-0.052
 hesitation-0.049
 payoff-0.049
TokenMaybe
Token ID13300
Feature activation+1.02e-01

Pos logit contributions
aling+0.029
alling+0.026
astics+0.025
irl+0.025
 sings+0.024
Neg logit contributions
 poisoning-0.017
 injuries-0.017
guard-0.016
 payoff-0.015
 unfortunately-0.014
Token the
Token ID262
Feature activation+2.79e-02

Pos logit contributions
 poisoning+0.120
guard+0.116
 injuries+0.115
 unfortunately+0.103
 payoff+0.103
Neg logit contributions
aling-0.191
alling-0.172
astics-0.166
irl-0.165
 sings-0.161
Token snail
Token ID47374
Feature activation-1.02e-02

Pos logit contributions
 poisoning+0.043
 injuries+0.042
guard+0.041
 department+0.039
 unfortunately+0.039
Neg logit contributions
aling-0.042
alling-0.036
irl-0.036
astics-0.034
 sings-0.034
Token did
Token ID750
Feature activation-1.26e-03

Pos logit contributions
aling+0.016
alling+0.014
 sings+0.013
irl+0.013
astics+0.013
Neg logit contributions
 injuries-0.016
 poisoning-0.015
guard-0.015
 unfortunately-0.015
 payoff-0.014
Token not
Token ID407
Feature activation+3.73e-02

Pos logit contributions
aling+0.003
alling+0.002
astics+0.002
irl+0.002
 sings+0.002
Neg logit contributions
guard-0.001
 poisoning-0.001
 injuries-0.001
 unfortunately-0.001
ho-0.001
Token like
Token ID588
Feature activation+9.32e-03

Pos logit contributions
guard+0.051
 poisoning+0.051
 injuries+0.051
 unfortunately+0.047
 payoff+0.046
Neg logit contributions
aling-0.065
alling-0.057
astics-0.054
irl-0.053
 sings-0.051
TokenMaybe
Token ID13300
Feature activation+9.62e-02

Pos logit contributions
aling+0.009
alling+0.008
astics+0.008
irl+0.008
 sings+0.008
Neg logit contributions
 poisoning-0.006
 injuries-0.006
guard-0.006
 payoff-0.005
 unfortunately-0.005
Token it
Token ID340
Feature activation+3.14e-02

Pos logit contributions
 poisoning+0.125
 injuries+0.118
guard+0.115
t+0.108
ho+0.108
Neg logit contributions
aling-0.159
astics-0.144
 sings-0.143
irl-0.141
alling-0.139
Token is
Token ID318
Feature activation-3.37e-02

Pos logit contributions
guard+0.044
 poisoning+0.042
 department+0.040
 last+0.040
 injuries+0.039
Neg logit contributions
aling-0.054
alling-0.050
irl-0.046
astics-0.046
 essay-0.043
Token sad
Token ID6507
Feature activation+5.15e-02

Pos logit contributions
aling+0.059
alling+0.052
astics+0.051
irl+0.050
 sings+0.050
Neg logit contributions
guard-0.045
 poisoning-0.042
 injuries-0.042
t-0.039
ho-0.039
Token.
Token ID13
Feature activation-1.56e-02

Pos logit contributions
 poisoning+0.089
 injuries+0.088
+0.087
 hesitation+0.087
guard+0.085
Neg logit contributions
 plays-0.054
 sings-0.054
aling-0.053
 dances-0.053
 shows-0.048
Token Maybe
Token ID6674
Feature activation+1.02e-01

Pos logit contributions
aling+0.027
alling+0.025
irl+0.023
astics+0.023
 sings+0.021
Neg logit contributions
guard-0.021
 poisoning-0.021
 injuries-0.020
 unfortunately-0.019
 payoff-0.019
Token it
Token ID340
Feature activation+2.54e-02

Pos logit contributions
 poisoning+0.132
guard+0.127
 injuries+0.125
t+0.119
ho+0.114
Neg logit contributions
aling-0.167
 sings-0.153
irl-0.148
astics-0.147
alling-0.146
Token is
Token ID318
Feature activation-3.42e-02

Pos logit contributions
guard+0.037
 poisoning+0.036
 injuries+0.034
 last+0.033
 department+0.032
Neg logit contributions
aling-0.042
alling-0.039
astics-0.036
irl-0.036
 essay-0.033
Token lonely
Token ID21757
Feature activation+3.50e-02

Pos logit contributions
aling+0.059
alling+0.053
astics+0.051
irl+0.051
 sings+0.050
Neg logit contributions
guard-0.046
 poisoning-0.044
 injuries-0.043
t-0.040
 unfortunately-0.040
Token."
Token ID526
Feature activation-7.52e-03

Pos logit contributions
ho+0.059
 injuries+0.059
 poisoning+0.058
guard+0.056
 unfortunately+0.056
Neg logit contributions
aling-0.043
 sings-0.038
 plays-0.036
 dances-0.035
astics-0.035
Token
Token ID198
Feature activation+2.82e-02

Pos logit contributions
aling+0.014
alling+0.012
astics+0.012
 sings+0.012
irl+0.011
Neg logit contributions
guard-0.010
 poisoning-0.010
 injuries-0.009
 payoff-0.008
 unfortunately-0.008
Token.
Token ID13
Feature activation-2.19e-02

Pos logit contributions
guard+0.034
 injuries+0.033
 poisoning+0.032
 unfortunately+0.032
 payoff+0.031
Neg logit contributions
aling-0.037
alling-0.032
astics-0.030
 sings-0.030
ophone-0.030
Token She
Token ID1375
Feature activation+5.71e-03

Pos logit contributions
aling+0.029
alling+0.025
astics+0.024
irl+0.023
 sings+0.023
Neg logit contributions
guard-0.039
 poisoning-0.038
 injuries-0.038
 payoff-0.036
 unfortunately-0.034
Token thought
Token ID1807
Feature activation+5.66e-02

Pos logit contributions
guard+0.007
 poisoning+0.007
 injuries+0.007
 unfortunately+0.006
 payoff+0.006
Neg logit contributions
aling-0.011
alling-0.010
astics-0.009
irl-0.009
ophone-0.009
Token,
Token ID11
Feature activation+3.35e-02

Pos logit contributions
 poisoning+0.066
 extingu+0.066
 injuries+0.066
guard+0.065
 payoff+0.065
Neg logit contributions
aling-0.102
irl-0.091
alling-0.089
 dances-0.088
 sings-0.088
Token "
Token ID366
Feature activation-6.06e-03

Pos logit contributions
guard+0.050
 poisoning+0.048
 injuries+0.048
 payoff+0.045
 unfortunately+0.043
Neg logit contributions
aling-0.053
alling-0.047
 sings-0.044
astics-0.044
irl-0.044
TokenMaybe
Token ID13300
Feature activation+1.01e-01

Pos logit contributions
aling+0.011
astics+0.009
alling+0.009
irl+0.009
 sings+0.009
Neg logit contributions
 poisoning-0.009
 injuries-0.008
 hesitation-0.008
guard-0.008
 department-0.008
Token he
Token ID339
Feature activation+3.11e-02

Pos logit contributions
 poisoning+0.136
 injuries+0.127
ho+0.126
 extingu+0.121
guard+0.119
Neg logit contributions
aling-0.167
astics-0.148
 sings-0.144
irl-0.144
 plays-0.142
Token is
Token ID318
Feature activation-3.20e-02

Pos logit contributions
 poisoning+0.042
guard+0.041
 injuries+0.038
 last+0.038
t+0.036
Neg logit contributions
aling-0.054
alling-0.050
astics-0.046
irl-0.045
arg-0.043
Token hungry
Token ID14720
Feature activation+3.09e-02

Pos logit contributions
aling+0.065
alling+0.059
astics+0.055
irl+0.055
 sings+0.052
Neg logit contributions
 poisoning-0.035
 injuries-0.035
guard-0.034
 unfortunately-0.031
 payoff-0.030
Token or
Token ID393
Feature activation-3.39e-02

Pos logit contributions
 injuries+0.048
 unfortunately+0.046
ho+0.046
 poisoning+0.046
 hesitation+0.045
Neg logit contributions
aling-0.044
alling-0.035
astics-0.033
v-0.033
 plays-0.033
Token lonely
Token ID21757
Feature activation+3.40e-02

Pos logit contributions
aling+0.055
alling+0.048
astics+0.045
irl+0.044
 sings+0.044
Neg logit contributions
 poisoning-0.050
guard-0.049
 injuries-0.049
 unfortunately-0.045
 payoff-0.045
Token was
Token ID373
Feature activation-5.26e-02

Pos logit contributions
guard+0.003
 poisoning+0.003
 injuries+0.003
 unfortunately+0.002
 last+0.002
Neg logit contributions
aling-0.006
alling-0.005
astics-0.005
irl-0.005
opl-0.005
Token scared
Token ID12008
Feature activation+3.17e-02

Pos logit contributions
aling+0.101
alling+0.091
irl+0.088
astics+0.088
ophone+0.085
Neg logit contributions
 injuries-0.063
 poisoning-0.062
guard-0.057
 last-0.055
 though-0.053
Token too
Token ID1165
Feature activation+3.16e-02

Pos logit contributions
 poisoning+0.055
comb+0.053
 payoff+0.053
ho+0.052
guard+0.051
Neg logit contributions
aling-0.041
alling-0.034
v-0.030
 plays-0.030
 zoom-0.029
Token.
Token ID13
Feature activation-2.56e-02

Pos logit contributions
 poisoning+0.040
guard+0.039
 injuries+0.038
 payoff+0.037
 unfortunately+0.036
Neg logit contributions
aling-0.057
alling-0.050
irl-0.048
astics-0.048
 sings-0.045
Token "
Token ID366
Feature activation-1.24e-02

Pos logit contributions
aling+0.026
alling+0.021
astics+0.020
atters+0.019
learn+0.018
Neg logit contributions
 poisoning-0.051
guard-0.051
 injuries-0.051
 payoff-0.049
ho-0.046
TokenMaybe
Token ID13300
Feature activation+1.01e-01

Pos logit contributions
aling+0.026
alling+0.023
astics+0.022
irl+0.022
 sings+0.021
Neg logit contributions
 poisoning-0.013
 injuries-0.013
guard-0.013
 payoff-0.011
 unfortunately-0.011
Token we
Token ID356
Feature activation+2.59e-02

Pos logit contributions
 poisoning+0.115
 injuries+0.109
guard+0.108
ho+0.101
 payoff+0.098
Neg logit contributions
aling-0.193
alling-0.174
astics-0.168
irl-0.166
 sings-0.163
Token are
Token ID389
Feature activation-1.64e-02

Pos logit contributions
guard+0.030
 poisoning+0.030
 injuries+0.029
 unfortunately+0.027
 payoff+0.027
Neg logit contributions
aling-0.050
alling-0.045
irl-0.042
astics-0.042
 sings-0.041
Token lost
Token ID2626
Feature activation+8.37e-03

Pos logit contributions
aling+0.030
alling+0.027
astics+0.026
irl+0.025
irled+0.024
Neg logit contributions
 injuries-0.021
 poisoning-0.021
guard-0.020
 unfortunately-0.018
 last-0.018
Token."
Token ID526
Feature activation-1.02e-02

Pos logit contributions
 poisoning+0.013
 payoff+0.013
 unfortunately+0.013
 injuries+0.013
ho+0.013
Neg logit contributions
aling-0.011
alling-0.009
astics-0.009
v-0.008
ab-0.008
Token
Token ID198
Feature activation+2.56e-02

Pos logit contributions
aling+0.011
astics+0.009
learn+0.009
irl+0.009
alling+0.009
Neg logit contributions
 poisoning-0.020
 injuries-0.019
guard-0.019
ho-0.018
 payoff-0.018
Token is
Token ID318
Feature activation-4.47e-02

Pos logit contributions
 injuries+0.003
 last+0.003
 poisoning+0.003
 or+0.003
 department+0.003
Neg logit contributions
aling-0.004
alling-0.004
astics-0.003
ophone-0.003
beh-0.003
Token scared
Token ID12008
Feature activation+3.31e-02

Pos logit contributions
aling+0.080
alling+0.071
astics+0.068
irl+0.067
 sings+0.064
Neg logit contributions
 injuries-0.060
 poisoning-0.059
guard-0.057
 unfortunately-0.053
 last-0.052
Token too
Token ID1165
Feature activation+3.24e-02

Pos logit contributions
 poisoning+0.056
 payoff+0.053
guard+0.052
comb+0.051
 injuries+0.050
Neg logit contributions
aling-0.046
alling-0.039
v-0.034
 plays-0.033
 dances-0.033
Token.
Token ID13
Feature activation-2.91e-02

Pos logit contributions
 poisoning+0.042
guard+0.041
 injuries+0.039
 payoff+0.039
 unfortunately+0.038
Neg logit contributions
aling-0.056
alling-0.050
irl-0.048
astics-0.047
 sings-0.045
Token "
Token ID366
Feature activation-1.28e-02

Pos logit contributions
aling+0.032
alling+0.026
astics+0.026
atters+0.024
 sings+0.024
Neg logit contributions
 injuries-0.056
guard-0.056
 poisoning-0.056
 payoff-0.052
 unfortunately-0.050
TokenMaybe
Token ID13300
Feature activation+1.01e-01

Pos logit contributions
aling+0.026
alling+0.024
astics+0.022
irl+0.022
 sings+0.021
Neg logit contributions
 poisoning-0.014
 injuries-0.014
guard-0.013
 payoff-0.012
 unfortunately-0.012
Token we
Token ID356
Feature activation+2.46e-02

Pos logit contributions
 poisoning+0.122
 injuries+0.117
guard+0.116
ho+0.107
 payoff+0.104
Neg logit contributions
aling-0.182
alling-0.162
astics-0.159
irl-0.158
 sings-0.155
Token should
Token ID815
Feature activation+3.09e-02

Pos logit contributions
guard+0.029
 poisoning+0.029
 injuries+0.029
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.047
alling-0.042
irl-0.040
astics-0.039
 sings-0.038
Token go
Token ID467
Feature activation-2.45e-02

Pos logit contributions
guard+0.042
 poisoning+0.041
 injuries+0.041
 unfortunately+0.038
 payoff+0.037
Neg logit contributions
aling-0.054
alling-0.049
astics-0.046
irl-0.045
 sings-0.043
Token out
Token ID503
Feature activation+2.66e-02

Pos logit contributions
aling+0.043
alling+0.038
astics+0.035
irl+0.035
atters+0.033
Neg logit contributions
 poisoning-0.033
 injuries-0.033
guard-0.032
 unfortunately-0.032
 payoff-0.031
Token."
Token ID526
Feature activation-1.09e-02

Pos logit contributions
 poisoning+0.036
ho+0.035
guard+0.034
 injuries+0.033
 payoff+0.032
Neg logit contributions
aling-0.047
alling-0.041
astics-0.037
irl-0.037
 sings-0.035
Token cloud
Token ID6279
Feature activation-9.46e-03

Pos logit contributions
aling+0.054
alling+0.049
astics+0.046
irl+0.046
 sings+0.044
Neg logit contributions
guard-0.032
 injuries-0.032
 poisoning-0.032
 payoff-0.029
 unfortunately-0.028
Token.
Token ID13
Feature activation-2.42e-02

Pos logit contributions
aling+0.015
alling+0.013
c+0.013
irl+0.013
ophone+0.012
Neg logit contributions
 injuries-0.013
 poisoning-0.013
 payoff-0.013
guard-0.012
 hesitation-0.012
Token
Token ID198
Feature activation+2.78e-02

Pos logit contributions
aling+0.023
astics+0.021
alling+0.019
 sings+0.019
irl+0.019
Neg logit contributions
guard-0.049
 injuries-0.049
 poisoning-0.048
 hesitation-0.045
 payoff-0.045
Token
Token ID198
Feature activation+2.29e-02

Pos logit contributions
 poisoning+0.031
guard+0.030
 injuries+0.030
 unfortunately+0.027
 payoff+0.027
Neg logit contributions
aling-0.055
alling-0.050
irl-0.048
astics-0.048
 sings-0.045
Token"
Token ID1
Feature activation+3.31e-02

Pos logit contributions
guard+0.022
 poisoning+0.021
 injuries+0.021
 unfortunately+0.019
 payoff+0.018
Neg logit contributions
aling-0.049
alling-0.045
astics-0.043
irl-0.042
 sings-0.041
TokenMaybe
Token ID13300
Feature activation+1.01e-01

Pos logit contributions
 poisoning+0.048
 hesitation+0.047
 injuries+0.046
guard+0.045
 unfortunately+0.044
Neg logit contributions
aling-0.045
 sings-0.043
irl-0.043
astics-0.042
 presses-0.040
Token we
Token ID356
Feature activation+3.20e-02

Pos logit contributions
 poisoning+0.146
 injuries+0.140
guard+0.133
t+0.128
 payoff+0.125
Neg logit contributions
aling-0.153
irl-0.139
 sings-0.137
astics-0.135
alling-0.133
Token can
Token ID460
Feature activation+4.64e-02

Pos logit contributions
guard+0.038
 poisoning+0.037
 injuries+0.036
 unfortunately+0.033
 payoff+0.033
Neg logit contributions
aling-0.062
alling-0.056
astics-0.052
irl-0.052
 sings-0.051
Token ask
Token ID1265
Feature activation+5.39e-02

Pos logit contributions
guard+0.077
 poisoning+0.075
 injuries+0.074
 unfortunately+0.070
 payoff+0.070
Neg logit contributions
aling-0.065
irl-0.059
alling-0.058
astics-0.054
atters-0.052
Token it
Token ID340
Feature activation+4.43e-02

Pos logit contributions
guard+0.071
 poisoning+0.067
t+0.064
 unfortunately+0.063
 injuries+0.063
Neg logit contributions
aling-0.092
alling-0.083
astics-0.081
irl-0.080
 sings-0.077
Token to
Token ID284
Feature activation+4.29e-02

Pos logit contributions
 poisoning+0.051
guard+0.051
 injuries+0.049
 unfortunately+0.045
 payoff+0.044
Neg logit contributions
aling-0.084
alling-0.076
astics-0.075
irl-0.073
 sings-0.073
Token sorry
Token ID7926
Feature activation+6.23e-02

Pos logit contributions
aling+0.069
alling+0.062
astics+0.059
irl+0.058
 sings+0.056
Neg logit contributions
 poisoning-0.045
 injuries-0.044
guard-0.044
 unfortunately-0.040
 payoff-0.039
Token.
Token ID13
Feature activation-2.89e-02

Pos logit contributions
 poisoning+0.096
guard+0.096
 injuries+0.092
 payoff+0.090
 hesitation+0.088
Neg logit contributions
aling-0.092
 sings-0.078
alling-0.077
 plays-0.075
 dances-0.074
Token
Token ID198
Feature activation+2.50e-02

Pos logit contributions
aling+0.043
alling+0.037
astics+0.036
irl+0.034
 sings+0.033
Neg logit contributions
 poisoning-0.047
 injuries-0.047
guard-0.046
 payoff-0.042
ho-0.042
Token
Token ID198
Feature activation+2.22e-02

Pos logit contributions
 poisoning+0.030
guard+0.029
 injuries+0.028
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.049
alling-0.044
astics-0.042
irl-0.042
 sings-0.039
Token"
Token ID1
Feature activation+2.79e-02

Pos logit contributions
 poisoning+0.026
 injuries+0.025
guard+0.025
 unfortunately+0.023
 payoff+0.022
Neg logit contributions
aling-0.044
alling-0.039
astics-0.038
irl-0.037
 sings-0.035
TokenMaybe
Token ID13300
Feature activation+1.00e-01

Pos logit contributions
 injuries+0.041
 poisoning+0.040
 hesitation+0.039
 department+0.036
guard+0.035
Neg logit contributions
aling-0.043
astics-0.038
irl-0.037
Check-0.036
alling-0.036
Token we
Token ID356
Feature activation+3.08e-02

Pos logit contributions
 poisoning+0.140
 injuries+0.131
guard+0.130
ho+0.126
t+0.121
Neg logit contributions
aling-0.161
irl-0.141
astics-0.141
alling-0.137
 sings-0.135
Token are
Token ID389
Feature activation-9.98e-03

Pos logit contributions
guard+0.034
 poisoning+0.034
 injuries+0.034
 unfortunately+0.030
 payoff+0.030
Neg logit contributions
aling-0.061
alling-0.055
astics-0.052
irl-0.052
 sings-0.050
Token both
Token ID1111
Feature activation+1.76e-02

Pos logit contributions
aling+0.018
alling+0.017
astics+0.016
irl+0.015
 sings+0.015
Neg logit contributions
 poisoning-0.013
 injuries-0.012
guard-0.012
 unfortunately-0.011
 payoff-0.011
Token bad
Token ID2089
Feature activation+5.58e-02

Pos logit contributions
 poisoning+0.028
guard+0.028
 injuries+0.027
ho+0.027
t+0.027
Neg logit contributions
aling-0.024
alling-0.021
astics-0.020
irl-0.019
 sings-0.019
Token fisher
Token ID17685
Feature activation-3.17e-02

Pos logit contributions
 poisoning+0.078
guard+0.075
 unfortunately+0.074
t+0.073
ho+0.072
Neg logit contributions
aling-0.082
 sings-0.080
astics-0.079
 dances-0.078
irl-0.076
Token too
Token ID1165
Feature activation+3.51e-02

Pos logit contributions
 poisoning+0.081
guard+0.080
 injuries+0.079
 payoff+0.072
 unfortunately+0.072
Neg logit contributions
aling-0.105
alling-0.092
 sings-0.091
astics-0.088
irl-0.086
Token.
Token ID13
Feature activation-2.05e-02

Pos logit contributions
guard+0.032
 injuries+0.032
 poisoning+0.031
 unfortunately+0.027
 payoff+0.026
Neg logit contributions
aling-0.078
alling-0.072
astics-0.069
irl-0.068
 sings-0.065
Token
Token ID198
Feature activation+3.37e-02

Pos logit contributions
aling+0.029
alling+0.025
astics+0.024
irl+0.022
 sings+0.022
Neg logit contributions
 poisoning-0.035
 injuries-0.035
guard-0.034
 payoff-0.032
 unfortunately-0.031
Token
Token ID198
Feature activation+3.23e-02

Pos logit contributions
 poisoning+0.040
guard+0.039
 injuries+0.038
 unfortunately+0.035
 payoff+0.035
Neg logit contributions
aling-0.065
alling-0.059
irl-0.056
astics-0.056
 sings-0.053
Token"
Token ID1
Feature activation+3.06e-02

Pos logit contributions
 poisoning+0.039
 injuries+0.038
guard+0.038
 payoff+0.034
 unfortunately+0.034
Neg logit contributions
aling-0.063
alling-0.056
astics-0.054
irl-0.052
 sings-0.050
TokenMaybe
Token ID13300
Feature activation+1.00e-01

Pos logit contributions
 injuries+0.046
 poisoning+0.045
 hesitation+0.044
 department+0.040
 extingu+0.040
Neg logit contributions
aling-0.046
astics-0.040
irl-0.039
alling-0.039
 sings-0.038
Token we
Token ID356
Feature activation+3.14e-02

Pos logit contributions
 poisoning+0.127
 injuries+0.117
guard+0.115
ho+0.110
t+0.110
Neg logit contributions
aling-0.170
astics-0.153
irl-0.152
 sings-0.152
alling-0.148
Token should
Token ID815
Feature activation+3.69e-02

Pos logit contributions
guard+0.035
 poisoning+0.035
 injuries+0.034
 unfortunately+0.031
 payoff+0.030
Neg logit contributions
aling-0.062
alling-0.057
astics-0.053
irl-0.053
 sings-0.051
Token say
Token ID910
Feature activation+1.32e-02

Pos logit contributions
guard+0.050
 poisoning+0.050
 injuries+0.049
 unfortunately+0.045
 payoff+0.044
Neg logit contributions
aling-0.063
alling-0.057
irl-0.054
astics-0.054
 sings-0.052
Token sorry
Token ID7926
Feature activation+5.02e-02

Pos logit contributions
guard+0.021
 poisoning+0.021
 injuries+0.021
 unfortunately+0.019
 payoff+0.019
Neg logit contributions
aling-0.019
alling-0.017
astics-0.016
irl-0.016
 sings-0.016
Token to
Token ID284
Feature activation+2.73e-02

Pos logit contributions
 poisoning+0.060
 injuries+0.059
+0.057
 unfortunately+0.057
guard+0.057
Neg logit contributions
aling-0.087
 sings-0.082
 dances-0.081
 plays-0.080
astics-0.079
Token.
Token ID13
Feature activation-2.99e-02

Pos logit contributions
 injuries+0.028
 poisoning+0.028
guard+0.028
 payoff+0.026
 unfortunately+0.025
Neg logit contributions
aling-0.039
alling-0.035
astics-0.033
irl-0.033
 sings-0.032
Token He
Token ID679
Feature activation-2.86e-03

Pos logit contributions
aling+0.037
astics+0.031
alling+0.029
 sings+0.027
 plays+0.027
Neg logit contributions
 injuries-0.057
 poisoning-0.056
guard-0.054
ho-0.052
 hesitation-0.051
Token thinks
Token ID6834
Feature activation+6.47e-02

Pos logit contributions
aling+0.005
alling+0.004
vel+0.004
astics+0.004
etition+0.004
Neg logit contributions
guard-0.004
 last-0.004
 poisoning-0.004
 injuries-0.003
 that-0.003
Token,
Token ID11
Feature activation+3.08e-02

Pos logit contributions
 poisoning+0.069
 injuries+0.066
guard+0.064
 payoff+0.062
 extingu+0.062
Neg logit contributions
aling-0.124
irl-0.114
alling-0.111
 sings-0.110
 dances-0.109
Token "
Token ID366
Feature activation-9.59e-03

Pos logit contributions
 poisoning+0.033
guard+0.033
 injuries+0.033
 unfortunately+0.031
 payoff+0.029
Neg logit contributions
aling-0.062
alling-0.056
astics-0.054
irl-0.053
irled-0.050
TokenMaybe
Token ID13300
Feature activation+1.00e-01

Pos logit contributions
aling+0.020
alling+0.018
astics+0.017
irl+0.017
 sings+0.017
Neg logit contributions
 injuries-0.010
 poisoning-0.010
guard-0.009
 unfortunately-0.009
 hesitation-0.009
Token I
Token ID314
Feature activation+3.12e-02

Pos logit contributions
 poisoning+0.135
 injuries+0.131
ho+0.118
guard+0.118
 extingu+0.117
Neg logit contributions
aling-0.168
astics-0.150
alling-0.145
 sings-0.144
irl-0.144
Token was
Token ID373
Feature activation-3.46e-02

Pos logit contributions
guard+0.034
 poisoning+0.033
 injuries+0.032
 unfortunately+0.029
 payoff+0.029
Neg logit contributions
aling-0.063
alling-0.058
astics-0.055
irl-0.054
 sings-0.052
Token wrong
Token ID2642
Feature activation+3.89e-02

Pos logit contributions
aling+0.066
alling+0.061
irl+0.055
 cycl+0.055
astics+0.055
Neg logit contributions
 poisoning-0.041
guard-0.039
 injuries-0.038
 though-0.036
 that-0.036
Token.
Token ID13
Feature activation-3.12e-02

Pos logit contributions
 poisoning+0.058
 injuries+0.056
guard+0.055
 unfortunately+0.054
ho+0.052
Neg logit contributions
aling-0.060
astics-0.054
alling-0.051
irl-0.051
 sings-0.048
Token Maybe
Token ID6674
Feature activation+9.14e-02

Pos logit contributions
aling+0.054
alling+0.049
irl+0.046
astics+0.046
 sings+0.043
Neg logit contributions
guard-0.042
 poisoning-0.042
 injuries-0.041
 unfortunately-0.038
 payoff-0.038
Token happy
Token ID3772
Feature activation+4.19e-02

Pos logit contributions
aling+0.085
alling+0.078
irl+0.073
astics+0.073
vel+0.072
Neg logit contributions
guard-0.069
 poisoning-0.062
 they-0.062
 last-0.061
 payoff-0.060
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
 poisoning+0.064
 injuries+0.061
guard+0.061
 unfortunately+0.059
ho+0.056
Neg logit contributions
aling-0.058
 sings-0.053
 dances-0.052
astics-0.051
 plays-0.051
Token They
Token ID1119
Feature activation+6.27e-04

Pos logit contributions
aling+0.037
alling+0.033
irl+0.031
astics+0.031
 sings+0.029
Neg logit contributions
guard-0.029
 poisoning-0.028
 injuries-0.028
 unfortunately-0.026
 payoff-0.026
Token ask
Token ID1265
Feature activation+5.15e-02

Pos logit contributions
guard+0.001
 poisoning+0.001
 injuries+0.001
 last+0.001
 payoff+0.001
Neg logit contributions
aling-0.001
alling-0.001
astics-0.001
irl-0.001
irled-0.001
Token mom
Token ID1995
Feature activation+7.10e-02

Pos logit contributions
guard+0.073
t+0.072
 poisoning+0.071
comb+0.070
storm+0.068
Neg logit contributions
aling-0.071
alling-0.062
irl-0.062
 plays-0.061
 zoom-0.060
Token if
Token ID611
Feature activation+9.99e-02

Pos logit contributions
 injuries+0.097
 poisoning+0.092
guard+0.090
 unfortunately+0.089
storm+0.088
Neg logit contributions
aling-0.116
alling-0.101
 plays-0.097
 sings-0.095
ophone-0.093
Token they
Token ID484
Feature activation+4.73e-02

Pos logit contributions
 injuries+0.090
 poisoning+0.089
 extingu+0.085
guard+0.080
t+0.077
Neg logit contributions
aling-0.179
 sings-0.170
irl-0.168
astics-0.168
 rode-0.165
Token can
Token ID460
Feature activation+4.95e-02

Pos logit contributions
 poisoning+0.059
 injuries+0.059
guard+0.059
 unfortunately+0.055
 payoff+0.052
Neg logit contributions
aling-0.087
alling-0.077
irl-0.072
astics-0.072
 sings-0.071
Token talk
Token ID1561
Feature activation-5.97e-03

Pos logit contributions
 poisoning+0.080
 injuries+0.078
guard+0.077
 unfortunately+0.075
 payoff+0.074
Neg logit contributions
aling-0.073
alling-0.064
irl-0.062
astics-0.059
 sings-0.054
Token to
Token ID284
Feature activation+2.36e-02

Pos logit contributions
aling+0.011
 sings+0.010
alling+0.010
ophone+0.010
astics+0.009
Neg logit contributions
 injuries-0.007
 poisoning-0.007
guard-0.006
ho-0.006
 unfortunately-0.006
Token the
Token ID262
Feature activation+1.29e-02

Pos logit contributions
 poisoning+0.030
guard+0.030
t+0.028
 injuries+0.028
 payoff+0.027
Neg logit contributions
aling-0.039
alling-0.035
irl-0.034
 sings-0.034
astics-0.033
Token want
Token ID765
Feature activation+3.61e-02

Pos logit contributions
guard+0.055
 poisoning+0.055
 injuries+0.054
 payoff+0.049
 unfortunately+0.049
Neg logit contributions
aling-0.070
alling-0.063
astics-0.059
irl-0.058
 sings-0.056
Token to
Token ID284
Feature activation+2.01e-02

Pos logit contributions
 poisoning+0.041
 injuries+0.040
 extingu+0.039
 department+0.038
 unfortunately+0.037
Neg logit contributions
aling-0.065
astics-0.060
 sings-0.059
alling-0.058
 plays-0.057
Token move
Token ID1445
Feature activation-2.99e-02

Pos logit contributions
 poisoning+0.033
guard+0.032
 injuries+0.032
 unfortunately+0.030
ho+0.029
Neg logit contributions
aling-0.030
alling-0.026
astics-0.025
irl-0.025
 sings-0.023
Token.
Token ID13
Feature activation-1.53e-02

Pos logit contributions
aling+0.052
alling+0.046
irl+0.044
astics+0.044
 sings+0.041
Neg logit contributions
 poisoning-0.040
 injuries-0.039
guard-0.038
 payoff-0.037
 unfortunately-0.036
Token "
Token ID366
Feature activation+2.84e-03

Pos logit contributions
aling+0.028
alling+0.025
irl+0.024
astics+0.023
 sings+0.022
Neg logit contributions
guard-0.019
 poisoning-0.018
 injuries-0.018
 unfortunately-0.018
 payoff-0.017
TokenMaybe
Token ID13300
Feature activation+9.99e-02

Pos logit contributions
guard+0.002
 poisoning+0.002
 injuries+0.002
Not+0.002
 payoff+0.002
Neg logit contributions
aling-0.007
alling-0.006
irl-0.006
astics-0.006
 sings-0.006
Token I
Token ID314
Feature activation+2.99e-02

Pos logit contributions
 poisoning+0.127
 injuries+0.122
guard+0.118
ho+0.111
t+0.110
Neg logit contributions
aling-0.174
alling-0.156
astics-0.155
irl-0.153
 sings-0.149
Token can
Token ID460
Feature activation+4.41e-02

Pos logit contributions
 poisoning+0.036
guard+0.036
 injuries+0.036
 unfortunately+0.033
 payoff+0.032
Neg logit contributions
aling-0.056
alling-0.050
astics-0.047
irl-0.047
 sings-0.045
Token stay
Token ID2652
Feature activation-3.63e-02

Pos logit contributions
 poisoning+0.062
guard+0.062
 injuries+0.061
 unfortunately+0.056
 payoff+0.056
Neg logit contributions
aling-0.075
alling-0.067
irl-0.063
astics-0.062
 sings-0.058
Token a
Token ID257
Feature activation-4.62e-04

Pos logit contributions
aling+0.070
alling+0.064
astics+0.060
irl+0.059
 sings+0.058
Neg logit contributions
 poisoning-0.042
guard-0.042
 injuries-0.041
 unfortunately-0.037
 payoff-0.036
Token little
Token ID1310
Feature activation+5.77e-03

Pos logit contributions
aling+0.001
alling+0.001
irl+0.001
astics+0.001
 sings+0.001
Neg logit contributions
 poisoning-0.001
guard-0.001
 injuries-0.000
 unfortunately-0.000
t-0.000
Token it
Token ID340
Feature activation+2.23e-02

Pos logit contributions
aling+0.038
alling+0.035
astics+0.033
ophone+0.032
irl+0.032
Neg logit contributions
 poisoning-0.012
 injuries-0.012
 last-0.011
 that-0.011
 help-0.010
Token.
Token ID13
Feature activation-2.55e-02

Pos logit contributions
guard+0.029
 poisoning+0.027
 injuries+0.027
 payoff+0.026
 unfortunately+0.025
Neg logit contributions
aling-0.040
alling-0.036
irl-0.034
astics-0.034
 sings-0.032
Token
Token ID198
Feature activation+3.15e-02

Pos logit contributions
aling+0.039
alling+0.034
astics+0.033
irl+0.032
 sings+0.031
Neg logit contributions
guard-0.040
 poisoning-0.040
 injuries-0.039
 payoff-0.036
 unfortunately-0.036
Token
Token ID198
Feature activation+2.99e-02

Pos logit contributions
 poisoning+0.039
guard+0.039
 injuries+0.037
 payoff+0.035
 unfortunately+0.035
Neg logit contributions
aling-0.059
alling-0.053
irl-0.051
astics-0.051
 sings-0.048
Token"
Token ID1
Feature activation+3.30e-02

Pos logit contributions
guard+0.032
 poisoning+0.032
 injuries+0.032
 unfortunately+0.028
 payoff+0.028
Neg logit contributions
aling-0.061
alling-0.055
astics-0.053
irl-0.052
 sings-0.050
TokenMaybe
Token ID13300
Feature activation+9.97e-02

Pos logit contributions
 injuries+0.047
 poisoning+0.046
 hesitation+0.045
 department+0.044
guard+0.043
Neg logit contributions
aling-0.048
astics-0.046
irl-0.045
 sings-0.043
learn-0.041
Token we
Token ID356
Feature activation+2.83e-02

Pos logit contributions
 poisoning+0.141
 injuries+0.137
guard+0.137
t+0.126
 payoff+0.123
Neg logit contributions
aling-0.157
irl-0.139
astics-0.138
 sings-0.137
alling-0.137
Token can
Token ID460
Feature activation+4.33e-02

Pos logit contributions
guard+0.030
 poisoning+0.030
 injuries+0.029
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.058
alling-0.053
astics-0.050
irl-0.050
 sings-0.048
Token find
Token ID1064
Feature activation+7.96e-03

Pos logit contributions
guard+0.068
 poisoning+0.068
 injuries+0.067
 payoff+0.064
 unfortunately+0.063
Neg logit contributions
aling-0.064
alling-0.057
irl-0.056
astics-0.053
atters-0.050
Token another
Token ID1194
Feature activation+5.15e-02

Pos logit contributions
guard+0.011
 poisoning+0.010
 injuries+0.010
 unfortunately+0.010
t+0.009
Neg logit contributions
aling-0.014
alling-0.012
irl-0.012
astics-0.012
 sings-0.012
Token one
Token ID530
Feature activation+6.49e-02

Pos logit contributions
 injuries+0.075
 poisoning+0.073
guard+0.073
 unfortunately+0.067
 department+0.064
Neg logit contributions
aling-0.080
 sings-0.071
c-0.071
irl-0.070
alling-0.070
Token hungry
Token ID14720
Feature activation+3.22e-02

Pos logit contributions
aling+0.010
alling+0.008
irl+0.008
astics+0.008
 sings+0.007
Neg logit contributions
guard-0.010
 poisoning-0.010
 injuries-0.010
 unfortunately-0.010
 payoff-0.010
Token.
Token ID13
Feature activation-2.95e-02

Pos logit contributions
 unfortunately+0.050
guard+0.049
 injuries+0.047
 poisoning+0.047
etsy+0.047
Neg logit contributions
aling-0.042
 goals-0.034
alling-0.034
 rides-0.033
 dances-0.033
Token They
Token ID1119
Feature activation-4.43e-03

Pos logit contributions
aling+0.029
astics+0.026
 sings+0.024
alling+0.024
irl+0.023
Neg logit contributions
guard-0.060
 poisoning-0.060
 injuries-0.059
 hesitation-0.054
 mercy-0.054
Token ask
Token ID1265
Feature activation+5.18e-02

Pos logit contributions
aling+0.008
alling+0.007
astics+0.006
irl+0.006
 sings+0.006
Neg logit contributions
guard-0.006
 poisoning-0.006
 injuries-0.006
 unfortunately-0.005
 payoff-0.005
Token Mom
Token ID11254
Feature activation+5.97e-02

Pos logit contributions
guard+0.074
t+0.068
 poisoning+0.067
 payoff+0.065
 unfortunately+0.063
Neg logit contributions
aling-0.075
irl-0.068
alling-0.067
astics-0.066
 dances-0.064
Token if
Token ID611
Feature activation+9.97e-02

Pos logit contributions
 injuries+0.081
guard+0.079
 poisoning+0.077
 unfortunately+0.075
 payoff+0.073
Neg logit contributions
aling-0.098
alling-0.085
 sings-0.083
 plays-0.081
astics-0.080
Token they
Token ID484
Feature activation+4.76e-02

Pos logit contributions
 injuries+0.093
 poisoning+0.088
 extingu+0.088
guard+0.086
 hesitation+0.076
Neg logit contributions
aling-0.176
irl-0.172
 sings-0.171
astics-0.169
 plays-0.163
Token can
Token ID460
Feature activation+4.96e-02

Pos logit contributions
guard+0.077
 injuries+0.075
 poisoning+0.074
 unfortunately+0.074
 extingu+0.072
Neg logit contributions
aling-0.069
alling-0.056
irl-0.053
 sings-0.053
astics-0.053
Token have
Token ID423
Feature activation+1.97e-03

Pos logit contributions
 poisoning+0.089
guard+0.089
 injuries+0.088
 unfortunately+0.086
ately+0.085
Neg logit contributions
aling-0.057
irl-0.052
alling-0.046
astics-0.043
 plays-0.042
Token a
Token ID257
Feature activation+1.24e-02

Pos logit contributions
guard+0.003
ho+0.003
 poisoning+0.003
 department+0.003
ately+0.003
Neg logit contributions
aling-0.003
irl-0.002
astics-0.002
 sings-0.002
alling-0.002
Token snack
Token ID26906
Feature activation+2.95e-02

Pos logit contributions
guard+0.016
ately+0.016
 unfortunately+0.016
 injuries+0.015
 poisoning+0.015
Neg logit contributions
aling-0.019
irl-0.018
 sings-0.016
astics-0.016
alling-0.016
Token don
Token ID836
Feature activation+2.36e-02

Pos logit contributions
guard+0.029
 poisoning+0.028
 injuries+0.028
 unfortunately+0.025
 payoff+0.025
Neg logit contributions
aling-0.046
alling-0.041
astics-0.039
irl-0.039
 sings-0.037
Token't
Token ID470
Feature activation+3.62e-02

Pos logit contributions
guard+0.012
 poisoning+0.011
 injuries+0.011
 unfortunately+0.009
t+0.009
Neg logit contributions
aling-0.061
alling-0.057
astics-0.054
irl-0.053
 sings-0.053
Token have
Token ID423
Feature activation+3.13e-04

Pos logit contributions
 poisoning+0.050
guard+0.050
 injuries+0.049
 payoff+0.046
 last+0.045
Neg logit contributions
aling-0.061
alling-0.055
astics-0.052
irl-0.052
irled-0.049
Token permission
Token ID7170
Feature activation+5.38e-02

Pos logit contributions
guard+0.000
 poisoning+0.000
 injuries+0.000
 unfortunately+0.000
ho+0.000
Neg logit contributions
aling-0.001
alling-0.001
astics-0.001
irl-0.001
 sings-0.001
Token.
Token ID13
Feature activation-2.51e-02

Pos logit contributions
 poisoning+0.068
 injuries+0.068
guard+0.067
 unfortunately+0.062
 payoff+0.061
Neg logit contributions
aling-0.098
alling-0.088
astics-0.083
ophone-0.081
irl-0.077
Token Maybe
Token ID6674
Feature activation+9.96e-02

Pos logit contributions
aling+0.046
alling+0.041
irl+0.040
astics+0.037
irled+0.036
Neg logit contributions
 poisoning-0.031
 injuries-0.030
guard-0.030
 unfortunately-0.030
 sir-0.029
Token we
Token ID356
Feature activation+2.42e-02

Pos logit contributions
 poisoning+0.118
guard+0.115
 injuries+0.111
t+0.103
ho+0.101
Neg logit contributions
aling-0.179
astics-0.160
alling-0.160
irl-0.158
 sings-0.156
Token should
Token ID815
Feature activation+3.32e-02

Pos logit contributions
guard+0.032
 injuries+0.032
 poisoning+0.032
 payoff+0.030
 unfortunately+0.029
Neg logit contributions
aling-0.043
alling-0.037
irl-0.035
astics-0.034
ophone-0.034
Token go
Token ID467
Feature activation-2.24e-02

Pos logit contributions
guard+0.045
 poisoning+0.045
 injuries+0.044
 unfortunately+0.041
 payoff+0.040
Neg logit contributions
aling-0.058
alling-0.052
astics-0.049
irl-0.049
 sings-0.046
Token back
Token ID736
Feature activation+1.07e-02

Pos logit contributions
aling+0.043
alling+0.039
astics+0.036
irl+0.036
 sings+0.034
Neg logit contributions
 poisoning-0.027
guard-0.027
 injuries-0.027
 unfortunately-0.025
 payoff-0.024
Token."
Token ID526
Feature activation-9.27e-03

Pos logit contributions
 poisoning+0.010
 injuries+0.010
guard+0.009
 unfortunately+0.009
 payoff+0.008
Neg logit contributions
aling-0.023
alling-0.021
astics-0.020
irl-0.020
 sings-0.020
Token was
Token ID373
Feature activation-4.04e-02

Pos logit contributions
guard+0.026
 poisoning+0.026
 injuries+0.026
 unfortunately+0.023
 payoff+0.023
Neg logit contributions
aling-0.037
alling-0.034
astics-0.032
irl-0.031
 sings-0.030
Token sad
Token ID6507
Feature activation+5.59e-02

Pos logit contributions
aling+0.068
alling+0.061
astics+0.057
irl+0.057
 sings+0.055
Neg logit contributions
 poisoning-0.056
 injuries-0.055
guard-0.055
 unfortunately-0.052
 payoff-0.050
Token and
Token ID290
Feature activation-3.08e-02

Pos logit contributions
guard+0.099
 payoff+0.096
tha+0.096
 poisoning+0.095
 injuries+0.095
Neg logit contributions
 plays-0.055
aling-0.053
 dances-0.051
 sings-0.048
 zoom-0.045
Token didn
Token ID1422
Feature activation+1.56e-02

Pos logit contributions
aling+0.062
alling+0.056
astics+0.053
irl+0.053
 sings+0.050
Neg logit contributions
guard-0.034
 poisoning-0.033
 injuries-0.033
 unfortunately-0.030
 payoff-0.029
Token't
Token ID470
Feature activation+5.42e-02

Pos logit contributions
 poisoning+0.013
guard+0.012
 injuries+0.012
 payoff+0.011
 unfortunately+0.010
Neg logit contributions
aling-0.036
alling-0.033
irl-0.032
astics-0.031
 sings-0.030
Token know
Token ID760
Feature activation+9.96e-02

Pos logit contributions
guard+0.069
 injuries+0.066
 poisoning+0.066
 help+0.063
 last+0.063
Neg logit contributions
aling-0.095
alling-0.089
astics-0.084
irl-0.083
irled-0.081
Token what
Token ID644
Feature activation+7.19e-02

Pos logit contributions
guard+0.136
 poisoning+0.132
 payoff+0.127
 injuries+0.126
ho+0.122
Neg logit contributions
aling-0.162
alling-0.144
astics-0.137
irl-0.136
 sings-0.132
Token to
Token ID284
Feature activation+4.75e-02

Pos logit contributions
 injuries+0.074
 poisoning+0.066
guard+0.064
 unfortunately+0.062
 extingu+0.060
Neg logit contributions
aling-0.140
astics-0.131
 sings-0.130
 plays-0.128
ophone-0.127
Token do
Token ID466
Feature activation+3.43e-02

Pos logit contributions
 department+0.094
 injuries+0.091
guard+0.091
 poisoning+0.089
 extingu+0.088
Neg logit contributions
aling-0.042
irl-0.041
astics-0.037
 zoom-0.036
 sings-0.036
Token.
Token ID13
Feature activation-2.73e-03

Pos logit contributions
 injuries+0.056
ho+0.056
guard+0.055
 poisoning+0.054
 extingu+0.053
Neg logit contributions
aling-0.047
astics-0.042
irl-0.040
alling-0.039
vel-0.037
Token
Token ID220
Feature activation+4.76e-03

Pos logit contributions
aling+0.005
alling+0.004
astics+0.004
irl+0.004
 sings+0.004
Neg logit contributions
guard-0.004
 poisoning-0.004
 injuries-0.004
 unfortunately-0.003
 payoff-0.003
Token say
Token ID910
Feature activation+2.23e-02

Pos logit contributions
 poisoning+0.079
guard+0.078
 injuries+0.077
 unfortunately+0.073
 payoff+0.072
Neg logit contributions
aling-0.070
alling-0.061
irl-0.059
astics-0.057
 sings-0.052
Token sorry
Token ID7926
Feature activation+5.46e-02

Pos logit contributions
guard+0.029
 poisoning+0.029
 injuries+0.028
 unfortunately+0.026
 payoff+0.025
Neg logit contributions
aling-0.040
alling-0.036
astics-0.034
irl-0.034
 sings-0.033
Token and
Token ID290
Feature activation-2.17e-02

Pos logit contributions
guard+0.068
 poisoning+0.067
 injuries+0.067
 unfortunately+0.062
 payoff+0.061
Neg logit contributions
aling-0.093
 dances-0.085
 sings-0.085
astics-0.084
 plays-0.083
Token hug
Token ID16225
Feature activation+3.19e-02

Pos logit contributions
aling+0.032
 sings+0.027
alling+0.027
astics+0.026
 plays+0.026
Neg logit contributions
 poisoning-0.033
 injuries-0.032
guard-0.032
t-0.030
ho-0.030
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
guard+0.046
 poisoning+0.045
 injuries+0.042
 payoff+0.042
t+0.042
Neg logit contributions
aling-0.046
 sings-0.041
irl-0.041
astics-0.040
alling-0.040
Token Maybe
Token ID6674
Feature activation+9.95e-02

Pos logit contributions
aling+0.031
alling+0.027
astics+0.027
 sings+0.025
irl+0.024
Neg logit contributions
guard-0.041
 poisoning-0.041
 injuries-0.040
 payoff-0.036
 unfortunately-0.036
Token later
Token ID1568
Feature activation+4.29e-02

Pos logit contributions
 poisoning+0.100
 injuries+0.094
ho+0.093
guard+0.093
t+0.088
Neg logit contributions
aling-0.180
 sings-0.168
astics-0.165
irl-0.163
alling-0.161
Token,
Token ID11
Feature activation+2.39e-02

Pos logit contributions
 extingu+0.047
 poisoning+0.047
 injuries+0.046
ousy+0.042
guard+0.042
Neg logit contributions
aling-0.075
astics-0.068
 sings-0.067
alling-0.066
 rides-0.065
Token they
Token ID484
Feature activation+7.47e-03

Pos logit contributions
 injuries+0.025
guard+0.025
 poisoning+0.024
ho+0.023
t+0.022
Neg logit contributions
aling-0.046
alling-0.042
 sings-0.041
astics-0.040
irl-0.039
Token would
Token ID561
Feature activation+9.73e-03

Pos logit contributions
guard+0.008
 poisoning+0.008
 injuries+0.008
 unfortunately+0.007
 payoff+0.007
Neg logit contributions
aling-0.015
alling-0.014
astics-0.013
irl-0.013
 sings-0.013
Token.
Token ID13
Feature activation-3.06e-02

Pos logit contributions
guard+0.015
 poisoning+0.015
 injuries+0.014
 unfortunately+0.013
 payoff+0.013
Neg logit contributions
aling-0.015
alling-0.013
irl-0.013
astics-0.013
 sings-0.012
Token them
Token ID606
Feature activation+3.59e-02

Pos logit contributions
guard+0.001
 poisoning+0.001
Not+0.001
 department+0.001
t+0.001
Neg logit contributions
aling-0.002
irl-0.001
astics-0.001
alling-0.001
 sings-0.001
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
guard+0.045
 injuries+0.044
 poisoning+0.044
 payoff+0.043
 unfortunately+0.042
Neg logit contributions
aling-0.062
alling-0.055
astics-0.053
irl-0.052
ophone-0.050
Token
Token ID198
Feature activation+2.96e-02

Pos logit contributions
aling+0.037
alling+0.033
astics+0.031
irl+0.030
 sings+0.029
Neg logit contributions
guard-0.029
 poisoning-0.029
 injuries-0.029
 payoff-0.026
 unfortunately-0.026
Token
Token ID198
Feature activation+2.20e-02

Pos logit contributions
 poisoning+0.036
guard+0.035
 injuries+0.034
 unfortunately+0.032
 payoff+0.031
Neg logit contributions
aling-0.057
alling-0.051
astics-0.049
irl-0.048
 sings-0.046
Token"
Token ID1
Feature activation+3.22e-02

Pos logit contributions
guard+0.022
 poisoning+0.022
 injuries+0.022
 unfortunately+0.019
 payoff+0.019
Neg logit contributions
aling-0.046
alling-0.042
astics-0.040
irl-0.040
 sings-0.038
TokenMaybe
Token ID13300
Feature activation+9.94e-02

Pos logit contributions
 poisoning+0.043
 injuries+0.042
 hesitation+0.040
guard+0.038
ful+0.038
Neg logit contributions
aling-0.053
astics-0.048
alling-0.046
irl-0.045
 sings-0.045
Token we
Token ID356
Feature activation+3.21e-02

Pos logit contributions
 poisoning+0.127
 injuries+0.121
guard+0.118
t+0.112
ho+0.107
Neg logit contributions
aling-0.164
astics-0.149
 sings-0.147
irl-0.146
alling-0.143
Token can
Token ID460
Feature activation+4.98e-02

Pos logit contributions
guard+0.040
 poisoning+0.039
 injuries+0.039
 unfortunately+0.036
 payoff+0.035
Neg logit contributions
aling-0.060
alling-0.053
astics-0.050
irl-0.050
 sings-0.048
Token fly
Token ID6129
Feature activation-4.27e-02

Pos logit contributions
guard+0.084
 poisoning+0.082
 injuries+0.081
 payoff+0.076
 unfortunately+0.075
Neg logit contributions
aling-0.067
alling-0.058
irl-0.058
astics-0.055
atters-0.050
Token with
Token ID351
Feature activation-8.54e-03

Pos logit contributions
aling+0.067
irl+0.057
alling+0.056
c+0.056
astics+0.054
Neg logit contributions
 injuries-0.064
 poisoning-0.063
 unfortunately-0.059
 hesitation-0.059
guard-0.058
Token the
Token ID262
Feature activation+7.99e-03

Pos logit contributions
aling+0.013
irl+0.012
alling+0.011
 sings+0.011
astics+0.011
Neg logit contributions
guard-0.013
 poisoning-0.012
t-0.012
 department-0.012
 unfortunately-0.011
Token afraid
Token ID7787
Feature activation+5.34e-02

Pos logit contributions
aling+0.090
alling+0.081
astics+0.078
irl+0.077
ophone+0.074
Neg logit contributions
 poisoning-0.064
 injuries-0.063
guard-0.062
 unfortunately-0.058
 last-0.057
Token and
Token ID290
Feature activation-4.26e-02

Pos logit contributions
 poisoning+0.073
guard+0.067
 payoff+0.066
 injuries+0.065
 extingu+0.064
Neg logit contributions
aling-0.087
 sings-0.078
 dances-0.076
alling-0.076
 plays-0.074
Token careful
Token ID8161
Feature activation+3.47e-02

Pos logit contributions
aling+0.077
alling+0.070
astics+0.066
irl+0.066
ophone+0.062
Neg logit contributions
guard-0.054
 poisoning-0.054
 injuries-0.053
 unfortunately-0.050
 payoff-0.048
Token.
Token ID13
Feature activation-2.37e-02

Pos logit contributions
 poisoning+0.048
 injuries+0.047
guard+0.046
 payoff+0.044
 extingu+0.041
Neg logit contributions
aling-0.059
alling-0.053
astics-0.049
irl-0.048
ophone-0.048
Token "
Token ID366
Feature activation-6.89e-03

Pos logit contributions
aling+0.037
alling+0.033
astics+0.032
irl+0.031
 sings+0.030
Neg logit contributions
guard-0.038
 poisoning-0.036
 injuries-0.035
 payoff-0.033
 unfortunately-0.031
TokenMaybe
Token ID13300
Feature activation+9.94e-02

Pos logit contributions
aling+0.014
alling+0.013
astics+0.012
irl+0.012
 sings+0.012
Neg logit contributions
guard-0.007
 poisoning-0.007
 injuries-0.007
 unfortunately-0.006
 payoff-0.006
Token there
Token ID612
Feature activation+4.92e-02

Pos logit contributions
 poisoning+0.135
 injuries+0.129
guard+0.128
t+0.121
ho+0.121
Neg logit contributions
aling-0.158
irl-0.141
astics-0.140
 sings-0.140
alling-0.140
Token are
Token ID389
Feature activation-1.50e-02

Pos logit contributions
 injuries+0.079
 poisoning+0.079
guard+0.077
 unfortunately+0.073
 payoff+0.072
Neg logit contributions
aling-0.072
alling-0.062
astics-0.058
irl-0.058
 sings-0.058
Token monsters
Token ID10963
Feature activation+2.18e-02

Pos logit contributions
aling+0.022
astics+0.019
 sings+0.019
alling+0.019
irl+0.019
Neg logit contributions
guard-0.022
 poisoning-0.022
 injuries-0.020
 payoff-0.019
 unfortunately-0.019
Token or
Token ID393
Feature activation-1.59e-02

Pos logit contributions
 injuries+0.026
 poisoning+0.026
guard+0.025
 unfortunately+0.024
 payoff+0.024
Neg logit contributions
aling-0.041
alling-0.036
 sings-0.034
astics-0.033
ophone-0.033
Token ghosts
Token ID25899
Feature activation+6.42e-03

Pos logit contributions
aling+0.022
 sings+0.020
alling+0.019
 dances+0.019
astics+0.019
Neg logit contributions
 poisoning-0.025
guard-0.025
 unfortunately-0.023
t-0.023
 injuries-0.023
Token
Token ID198
Feature activation+1.83e-02

Pos logit contributions
guard+0.018
 injuries+0.016
 poisoning+0.015
 unfortunately+0.013
t+0.013
Neg logit contributions
aling-0.057
alling-0.053
astics-0.051
irl-0.050
 sings-0.050
Token"
Token ID1
Feature activation+3.33e-02

Pos logit contributions
guard+0.014
Not+0.012
 injuries+0.011
 poisoning+0.011
 unfortunately+0.011
Neg logit contributions
aling-0.041
alling-0.039
 sings-0.038
irl-0.037
ophone-0.037
TokenMe
Token ID5308
Feature activation-1.29e-02

Pos logit contributions
 injuries+0.046
 poisoning+0.044
 hesitation+0.041
ately+0.040
guard+0.040
Neg logit contributions
aling-0.056
alling-0.048
astics-0.046
 sings-0.046
irl-0.045
Token too
Token ID1165
Feature activation+4.90e-02

Pos logit contributions
aling+0.023
alling+0.021
irl+0.020
astics+0.020
 sings+0.019
Neg logit contributions
 poisoning-0.016
 injuries-0.016
guard-0.016
 unfortunately-0.014
 payoff-0.014
Token.
Token ID13
Feature activation-1.91e-02

Pos logit contributions
 poisoning+0.055
 injuries+0.053
guard+0.052
 payoff+0.046
 unfortunately+0.046
Neg logit contributions
aling-0.092
alling-0.082
irl-0.081
astics-0.079
 sings-0.077
Token Maybe
Token ID6674
Feature activation+9.93e-02

Pos logit contributions
aling+0.032
alling+0.028
irl+0.027
astics+0.027
 sings+0.025
Neg logit contributions
guard-0.027
 poisoning-0.027
 injuries-0.027
 unfortunately-0.025
 payoff-0.025
Token we
Token ID356
Feature activation+2.69e-02

Pos logit contributions
 poisoning+0.128
 injuries+0.123
guard+0.120
ho+0.116
t+0.115
Neg logit contributions
aling-0.162
alling-0.142
 sings-0.140
astics-0.139
irl-0.139
Token should
Token ID815
Feature activation+3.72e-02

Pos logit contributions
guard+0.031
 poisoning+0.031
 injuries+0.031
 unfortunately+0.028
 payoff+0.027
Neg logit contributions
aling-0.052
alling-0.047
astics-0.044
irl-0.044
 sings-0.043
Token sell
Token ID3677
Feature activation-1.01e-05

Pos logit contributions
guard+0.052
 poisoning+0.051
 injuries+0.049
 unfortunately+0.046
 payoff+0.045
Neg logit contributions
aling-0.062
alling-0.056
irl-0.054
astics-0.054
 sings-0.051
Token the
Token ID262
Feature activation+2.44e-02

Pos logit contributions
aling+0.000
astics+0.000
 rides+0.000
irl+0.000
alling+0.000
Neg logit contributions
 injuries-0.000
 unfortunately-0.000
 poisoning-0.000
 department-0.000
guard-0.000
Token book
Token ID1492
Feature activation+1.13e-02

Pos logit contributions
 poisoning+0.032
 injuries+0.032
guard+0.032
 unfortunately+0.031
t+0.029
Neg logit contributions
aling-0.040
alling-0.036
irl-0.034
astics-0.034
 sings-0.034
Token.
Token ID13
Feature activation-1.89e-02

Pos logit contributions
 poisoning+0.083
guard+0.082
 injuries+0.082
 payoff+0.074
 unfortunately+0.073
Neg logit contributions
aling-0.157
alling-0.142
astics-0.136
irl-0.135
 sings-0.133
Token Maybe
Token ID6674
Feature activation+9.60e-02

Pos logit contributions
aling+0.033
alling+0.030
irl+0.028
astics+0.028
 sings+0.026
Neg logit contributions
guard-0.025
 poisoning-0.025
 injuries-0.025
 unfortunately-0.023
 payoff-0.022
Token they
Token ID484
Feature activation+2.05e-02

Pos logit contributions
 poisoning+0.120
 injuries+0.114
guard+0.110
t+0.108
ho+0.106
Neg logit contributions
aling-0.166
 sings-0.147
astics-0.146
alling-0.145
irl-0.141
Token forgot
Token ID16453
Feature activation+4.96e-02

Pos logit contributions
guard+0.023
 poisoning+0.022
 injuries+0.022
 unfortunately+0.020
 payoff+0.020
Neg logit contributions
aling-0.041
alling-0.037
astics-0.035
irl-0.035
 sings-0.034
Token.
Token ID13
Feature activation-2.04e-02

Pos logit contributions
 injuries+0.058
 poisoning+0.057
guard+0.056
ho+0.056
 unfortunately+0.053
Neg logit contributions
aling-0.086
 sings-0.078
alling-0.076
ophone-0.076
astics-0.075
Token Maybe
Token ID6674
Feature activation+9.93e-02

Pos logit contributions
aling+0.032
alling+0.028
astics+0.027
irl+0.026
 sings+0.025
Neg logit contributions
guard-0.031
 poisoning-0.031
 injuries-0.031
 unfortunately-0.028
 payoff-0.028
Token they
Token ID484
Feature activation+1.85e-02

Pos logit contributions
 poisoning+0.104
 injuries+0.100
guard+0.093
t+0.090
ho+0.089
Neg logit contributions
aling-0.184
 sings-0.174
astics-0.166
alling-0.164
 plays-0.162
Token are
Token ID389
Feature activation-1.84e-02

Pos logit contributions
guard+0.021
 poisoning+0.021
 injuries+0.021
 unfortunately+0.019
 payoff+0.019
Neg logit contributions
aling-0.036
alling-0.032
astics-0.030
irl-0.030
 sings-0.030
Token late
Token ID2739
Feature activation+5.21e-02

Pos logit contributions
aling+0.032
alling+0.028
astics+0.027
irl+0.027
 sings+0.025
Neg logit contributions
guard-0.025
 poisoning-0.025
 injuries-0.024
 unfortunately-0.023
 payoff-0.022
Token.
Token ID13
Feature activation-2.91e-02

Pos logit contributions
 poisoning+0.069
guard+0.069
 injuries+0.066
 unfortunately+0.063
ho+0.059
Neg logit contributions
aling-0.089
alling-0.079
astics-0.078
 sings-0.076
irl-0.073
Token Maybe
Token ID6674
Feature activation+9.52e-02

Pos logit contributions
aling+0.042
alling+0.036
astics+0.034
 sings+0.033
irl+0.031
Neg logit contributions
guard-0.048
 poisoning-0.048
 injuries-0.048
 payoff-0.043
 unfortunately-0.043
Token different
Token ID1180
Feature activation-1.44e-02

Pos logit contributions
aling+0.031
alling+0.027
irl+0.026
astics+0.026
 sings+0.025
Neg logit contributions
guard-0.030
 poisoning-0.029
 injuries-0.027
t-0.026
ho-0.025
Token styles
Token ID12186
Feature activation-1.49e-02

Pos logit contributions
aling+0.021
alling+0.018
astics+0.017
irl+0.017
 sings+0.016
Neg logit contributions
guard-0.024
 poisoning-0.024
 injuries-0.023
 unfortunately-0.022
 payoff-0.021
Token like
Token ID588
Feature activation-1.39e-04

Pos logit contributions
aling+0.028
alling+0.026
astics+0.024
irl+0.024
 sings+0.023
Neg logit contributions
 injuries-0.018
 poisoning-0.018
guard-0.017
 unfortunately-0.016
 payoff-0.016
Token stars
Token ID5788
Feature activation-9.11e-03

Pos logit contributions
aling+0.000
alling+0.000
irl+0.000
astics+0.000
 sings+0.000
Neg logit contributions
guard-0.000
 poisoning-0.000
 injuries-0.000
 unfortunately-0.000
t-0.000
Token or
Token ID393
Feature activation-2.73e-02

Pos logit contributions
aling+0.017
alling+0.016
irl+0.015
astics+0.014
 sings+0.014
Neg logit contributions
 injuries-0.011
 poisoning-0.011
guard-0.011
 unfortunately-0.010
 payoff-0.010
Token spir
Token ID9158
Feature activation-1.48e-01

Pos logit contributions
aling+0.038
alling+0.033
 sings+0.032
astics+0.031
irl+0.031
Neg logit contributions
guard-0.046
 poisoning-0.045
 injuries-0.042
t-0.042
 unfortunately-0.041
Tokenals
Token ID874
Feature activation-2.15e-02

Pos logit contributions
astics+0.270
irl+0.265
aling+0.259
 essay+0.246
 sings+0.243
Neg logit contributions
 poisoning-0.141
ho-0.140
 fourteen-0.139
 last-0.138
t-0.134
Token.
Token ID13
Feature activation-2.29e-02

Pos logit contributions
aling+0.039
alling+0.035
irl+0.032
 sings+0.031
astics+0.031
Neg logit contributions
guard-0.027
 poisoning-0.027
 injuries-0.027
 unfortunately-0.025
 payoff-0.025
Token Everyone
Token ID11075
Feature activation+1.28e-03

Pos logit contributions
aling+0.042
alling+0.038
irl+0.037
astics+0.035
ophone+0.033
Neg logit contributions
 unfortunately-0.027
 poisoning-0.027
guard-0.026
t-0.025
 injuries-0.025
Token was
Token ID373
Feature activation-4.51e-02

Pos logit contributions
guard+0.002
 poisoning+0.001
 injuries+0.001
 unfortunately+0.001
 payoff+0.001
Neg logit contributions
aling-0.002
alling-0.002
astics-0.002
irl-0.002
 sings-0.002
Token so
Token ID523
Feature activation-1.71e-02

Pos logit contributions
aling+0.071
alling+0.063
astics+0.059
irl+0.058
 sings+0.055
Neg logit contributions
guard-0.069
 poisoning-0.068
 injuries-0.068
 unfortunately-0.063
 payoff-0.062
Token on
Token ID319
Feature activation-7.54e-03

Pos logit contributions
guard+0.005
 poisoning+0.005
 injuries+0.005
 unfortunately+0.004
 last+0.004
Neg logit contributions
aling-0.007
alling-0.006
astics-0.006
irl-0.006
 sings-0.005
Token the
Token ID262
Feature activation-1.35e-02

Pos logit contributions
aling+0.012
alling+0.011
astics+0.011
irl+0.011
 sings+0.010
Neg logit contributions
guard-0.011
 poisoning-0.011
 injuries-0.010
 unfortunately-0.010
t-0.010
Token lake
Token ID13546
Feature activation-5.03e-02

Pos logit contributions
aling+0.022
alling+0.020
astics+0.019
irl+0.018
 sings+0.017
Neg logit contributions
guard-0.020
 poisoning-0.020
 injuries-0.019
 unfortunately-0.018
 payoff-0.017
Token in
Token ID287
Feature activation-2.19e-02

Pos logit contributions
aling+0.110
alling+0.102
astics+0.096
irl+0.095
 sings+0.092
Neg logit contributions
guard-0.047
 poisoning-0.046
 injuries-0.044
 unfortunately-0.040
ho-0.039
Token a
Token ID257
Feature activation-3.52e-02

Pos logit contributions
aling+0.045
alling+0.041
astics+0.039
irl+0.039
 sings+0.037
Neg logit contributions
guard-0.023
 poisoning-0.023
 injuries-0.023
 unfortunately-0.020
 payoff-0.020
Token kay
Token ID34681
Feature activation-1.48e-01

Pos logit contributions
aling+0.070
alling+0.065
aled+0.060
astics+0.060
ophone+0.059
Neg logit contributions
guard-0.039
 poisoning-0.036
 injuries-0.036
 last-0.034
 payoff-0.033
Tokenak
Token ID461
Feature activation-3.60e-02

Pos logit contributions
 sings+0.246
 dances+0.229
 presses+0.221
aling+0.220
alling+0.216
Neg logit contributions
guard-0.200
t-0.189
 poisoning-0.176
ll-0.163
side-0.158
Token.
Token ID13
Feature activation-9.20e-03

Pos logit contributions
aling+0.080
alling+0.074
astics+0.071
irl+0.071
 sings+0.068
Neg logit contributions
guard-0.032
 poisoning-0.030
 injuries-0.028
 unfortunately-0.027
 last-0.025
Token As
Token ID1081
Feature activation+2.61e-02

Pos logit contributions
aling+0.018
alling+0.017
irl+0.016
astics+0.015
irled+0.015
Neg logit contributions
 poisoning-0.011
guard-0.010
 injuries-0.010
 unfortunately-0.009
 Papa-0.009
Token they
Token ID484
Feature activation+5.93e-04

Pos logit contributions
 poisoning+0.018
 injuries+0.018
guard+0.017
 unfortunately+0.014
 payoff+0.014
Neg logit contributions
aling-0.062
alling-0.057
irl-0.056
astics-0.055
 sings-0.054
Token were
Token ID547
Feature activation-2.33e-02

Pos logit contributions
 poisoning+0.001
guard+0.001
 injuries+0.001
 unfortunately+0.001
 payoff+0.001
Neg logit contributions
aling-0.001
alling-0.001
irl-0.001
astics-0.001
 sings-0.001
Token the
Token ID262
Feature activation-1.92e-02

Pos logit contributions
aling+0.032
 spir+0.029
 sorts+0.028
alling+0.027
irl+0.027
Neg logit contributions
guard-0.045
t-0.043
 poisoning-0.042
ll-0.039
 injuries-0.039
Token air
Token ID1633
Feature activation-1.17e-02

Pos logit contributions
aling+0.030
alling+0.027
astics+0.025
irl+0.025
 sings+0.023
Neg logit contributions
guard-0.029
 poisoning-0.029
 injuries-0.029
 payoff-0.027
 unfortunately-0.027
Token,
Token ID11
Feature activation+8.76e-03

Pos logit contributions
aling+0.021
alling+0.019
irl+0.018
astics+0.018
 sings+0.018
Neg logit contributions
guard-0.014
 poisoning-0.014
 injuries-0.014
 unfortunately-0.013
 payoff-0.012
Token leaving
Token ID4305
Feature activation-6.80e-03

Pos logit contributions
guard+0.009
 poisoning+0.009
 injuries+0.009
 unfortunately+0.008
 payoff+0.008
Neg logit contributions
aling-0.018
alling-0.016
astics-0.015
irl-0.015
 sings-0.015
Token a
Token ID257
Feature activation-1.72e-02

Pos logit contributions
aling+0.010
alling+0.009
irl+0.009
 sings+0.009
c+0.008
Neg logit contributions
guard-0.010
 injuries-0.009
 poisoning-0.009
 unfortunately-0.008
 extingu-0.008
Token spir
Token ID9158
Feature activation-1.47e-01

Pos logit contributions
aling+0.027
alling+0.024
irl+0.023
astics+0.022
 sings+0.021
Neg logit contributions
guard-0.025
 poisoning-0.025
 injuries-0.024
 unfortunately-0.023
 payoff-0.022
Tokenaling
Token ID4272
Feature activation-4.56e-02

Pos logit contributions
irl+0.237
astics+0.228
irling+0.202
 essay+0.196
 sings+0.196
Neg logit contributions
ho-0.194
 poisoning-0.182
ful-0.181
 last-0.181
 fourteen-0.180
Token trail
Token ID8025
Feature activation-4.09e-02

Pos logit contributions
aling+0.078
alling+0.070
astics+0.065
irl+0.064
 sings+0.062
Neg logit contributions
 poisoning-0.063
guard-0.063
 injuries-0.062
 unfortunately-0.057
 payoff-0.056
Token of
Token ID286
Feature activation-2.01e-02

Pos logit contributions
aling+0.075
alling+0.068
astics+0.066
irl+0.065
 sings+0.063
Neg logit contributions
guard-0.052
 poisoning-0.050
 injuries-0.049
 unfortunately-0.045
t-0.045
Token black
Token ID2042
Feature activation-3.51e-02

Pos logit contributions
aling+0.028
alling+0.024
 sings+0.024
irl+0.023
astics+0.022
Neg logit contributions
guard-0.033
 poisoning-0.032
 injuries-0.031
t-0.031
 unfortunately-0.030
Token ink
Token ID16882
Feature activation-2.92e-02

Pos logit contributions
aling+0.052
alling+0.046
astics+0.042
irl+0.041
 sings+0.041
Neg logit contributions
guard-0.056
 poisoning-0.056
 injuries-0.055
 unfortunately-0.053
 payoff-0.051
Token liked
Token ID8288
Feature activation-8.00e-03

Pos logit contributions
aling+0.001
alling+0.001
astics+0.001
irl+0.001
 sings+0.001
Neg logit contributions
guard-0.001
 poisoning-0.001
 injuries-0.001
 unfortunately-0.001
 last-0.001
Token to
Token ID284
Feature activation-1.56e-02

Pos logit contributions
aling+0.018
alling+0.017
astics+0.016
irl+0.016
 sings+0.016
Neg logit contributions
 injuries-0.006
 poisoning-0.006
guard-0.006
 unfortunately-0.005
t-0.005
Token paddle
Token ID39517
Feature activation-6.32e-02

Pos logit contributions
aling+0.028
alling+0.025
astics+0.023
ophone+0.022
 sings+0.022
Neg logit contributions
guard-0.020
 poisoning-0.020
 injuries-0.019
 last-0.019
 help-0.018
Token in
Token ID287
Feature activation-2.24e-02

Pos logit contributions
aling+0.114
alling+0.102
irl+0.098
astics+0.097
 sings+0.094
Neg logit contributions
 injuries-0.080
 poisoning-0.078
guard-0.076
 unfortunately-0.073
 payoff-0.072
Token the
Token ID262
Feature activation-2.48e-02

Pos logit contributions
aling+0.046
alling+0.042
irl+0.040
astics+0.040
 sings+0.038
Neg logit contributions
guard-0.023
 poisoning-0.023
 injuries-0.022
 unfortunately-0.020
 payoff-0.020
Token kay
Token ID34681
Feature activation-1.45e-01

Pos logit contributions
aling+0.046
alling+0.042
astics+0.040
irl+0.039
 sings+0.037
Neg logit contributions
guard-0.031
 poisoning-0.031
 injuries-0.031
 unfortunately-0.028
 payoff-0.028
Tokenak
Token ID461
Feature activation-3.26e-02

Pos logit contributions
 sings+0.221
aling+0.220
alling+0.214
 dances+0.204
 presses+0.200
Neg logit contributions
guard-0.205
t-0.189
 poisoning-0.183
ll-0.167
 last-0.166
Token and
Token ID290
Feature activation-4.27e-02

Pos logit contributions
aling+0.067
alling+0.061
astics+0.058
irl+0.057
ophone+0.055
Neg logit contributions
 injuries-0.034
 poisoning-0.033
guard-0.033
 payoff-0.030
 unfortunately-0.029
Token splash
Token ID22870
Feature activation-6.06e-02

Pos logit contributions
aling+0.074
alling+0.066
astics+0.062
irl+0.061
 sings+0.058
Neg logit contributions
guard-0.059
 poisoning-0.058
 injuries-0.057
 unfortunately-0.053
 payoff-0.052
Token in
Token ID287
Feature activation-2.89e-02

Pos logit contributions
aling+0.097
alling+0.086
irl+0.083
astics+0.081
 sings+0.077
Neg logit contributions
 injuries-0.090
 poisoning-0.089
guard-0.087
 unfortunately-0.083
 payoff-0.083
Token the
Token ID262
Feature activation-2.38e-02

Pos logit contributions
aling+0.059
alling+0.054
irl+0.051
astics+0.051
 sings+0.049
Neg logit contributions
guard-0.030
 poisoning-0.030
 injuries-0.029
 unfortunately-0.026
 payoff-0.026
Token butterfly
Token ID35113
Feature activation-4.23e-02

Pos logit contributions
aling+0.060
alling+0.055
irl+0.052
astics+0.051
 sings+0.050
Neg logit contributions
 injuries-0.040
guard-0.040
 poisoning-0.040
 unfortunately-0.036
 payoff-0.035
Token,
Token ID11
Feature activation+2.19e-03

Pos logit contributions
aling+0.089
alling+0.082
astics+0.079
irl+0.077
aled+0.074
Neg logit contributions
guard-0.042
 poisoning-0.041
 injuries-0.039
 unfortunately-0.037
ho-0.035
Token tw
Token ID665
Feature activation-9.11e-02

Pos logit contributions
 poisoning+0.003
guard+0.003
 injuries+0.003
 unfortunately+0.003
 payoff+0.002
Neg logit contributions
aling-0.004
alling-0.004
astics-0.003
irl-0.003
ophone-0.003
Tokenirling
Token ID24297
Feature activation-6.52e-02

Pos logit contributions
aling+0.117
alling+0.099
ophone+0.097
 sings+0.090
 cycl+0.089
Neg logit contributions
ho-0.151
 injuries-0.145
ful-0.143
oles-0.140
 that-0.139
Token and
Token ID290
Feature activation-4.60e-02

Pos logit contributions
aling+0.105
alling+0.093
astics+0.087
irl+0.086
 sings+0.081
Neg logit contributions
guard-0.097
 poisoning-0.096
 injuries-0.094
 unfortunately-0.088
 payoff-0.087
Token spir
Token ID9158
Feature activation-1.44e-01

Pos logit contributions
aling+0.078
alling+0.070
astics+0.066
irl+0.063
ophone+0.062
Neg logit contributions
ho-0.061
 poisoning-0.060
guard-0.060
 payoff-0.059
 unfortunately-0.059
Tokenaling
Token ID4272
Feature activation-4.45e-02

Pos logit contributions
astics+0.279
irl+0.275
aling+0.257
 essay+0.249
 sings+0.246
Neg logit contributions
ho-0.128
 fourteen-0.118
ful-0.118
t-0.117
 last-0.117
Token in
Token ID287
Feature activation-2.84e-02

Pos logit contributions
aling+0.086
alling+0.077
astics+0.074
irl+0.073
 sings+0.069
Neg logit contributions
guard-0.053
 poisoning-0.051
 injuries-0.049
 payoff-0.046
ho-0.046
Token the
Token ID262
Feature activation-2.36e-02

Pos logit contributions
aling+0.047
alling+0.042
irl+0.039
astics+0.038
 sings+0.037
Neg logit contributions
guard-0.042
 poisoning-0.041
 injuries-0.040
 unfortunately-0.037
 payoff-0.036
Token air
Token ID1633
Feature activation-1.74e-02

Pos logit contributions
aling+0.043
alling+0.038
astics+0.037
irl+0.037
ophone+0.034
Neg logit contributions
guard-0.030
 poisoning-0.030
 injuries-0.029
 payoff-0.029
 last-0.029
Token.
Token ID13
Feature activation-2.07e-02

Pos logit contributions
aling+0.032
alling+0.029
irl+0.027
astics+0.027
 sings+0.026
Neg logit contributions
guard-0.021
 poisoning-0.021
 injuries-0.021
 unfortunately-0.019
 payoff-0.018
Token saw
Token ID2497
Feature activation-8.27e-03

Pos logit contributions
aling+0.011
alling+0.010
astics+0.009
irl+0.009
 sings+0.008
Neg logit contributions
guard-0.007
 poisoning-0.007
 injuries-0.007
 unfortunately-0.007
 payoff-0.007
Token a
Token ID257
Feature activation-1.78e-02

Pos logit contributions
aling+0.012
alling+0.011
 sings+0.011
ophone+0.010
irl+0.010
Neg logit contributions
 poisoning-0.012
 injuries-0.012
guard-0.012
t-0.011
 payoff-0.011
Token man
Token ID582
Feature activation-1.83e-02

Pos logit contributions
aling+0.030
alling+0.027
irl+0.026
astics+0.025
 sings+0.025
Neg logit contributions
 poisoning-0.024
guard-0.024
 injuries-0.023
 unfortunately-0.022
 payoff-0.021
Token in
Token ID287
Feature activation-2.46e-02

Pos logit contributions
aling+0.033
irl+0.030
alling+0.030
astics+0.030
irled+0.028
Neg logit contributions
guard-0.022
 poisoning-0.022
 injuries-0.021
 unfortunately-0.020
 last-0.019
Token a
Token ID257
Feature activation-3.88e-02

Pos logit contributions
aling+0.042
alling+0.038
astics+0.035
irl+0.035
 sings+0.033
Neg logit contributions
guard-0.033
 poisoning-0.033
 injuries-0.031
t-0.031
ho-0.030
Token kay
Token ID34681
Feature activation-1.44e-01

Pos logit contributions
aling+0.068
alling+0.061
astics+0.057
irl+0.056
 sings+0.054
Neg logit contributions
guard-0.053
 poisoning-0.053
 injuries-0.052
 unfortunately-0.047
 payoff-0.047
Tokenak
Token ID461
Feature activation-2.68e-02

Pos logit contributions
 sings+0.218
 dances+0.207
aling+0.207
alling+0.198
astics+0.193
Neg logit contributions
guard-0.211
 poisoning-0.194
t-0.192
 injuries-0.176
 last-0.173
Token.
Token ID13
Feature activation-2.82e-02

Pos logit contributions
aling+0.058
alling+0.053
irl+0.052
astics+0.052
 sings+0.050
Neg logit contributions
guard-0.025
 poisoning-0.024
 injuries-0.022
 unfortunately-0.021
 last-0.020
Token The
Token ID383
Feature activation+3.25e-03

Pos logit contributions
aling+0.048
alling+0.043
irl+0.041
astics+0.041
 sings+0.038
Neg logit contributions
guard-0.039
 poisoning-0.039
 injuries-0.038
 unfortunately-0.036
 payoff-0.035
Token man
Token ID582
Feature activation-2.17e-02

Pos logit contributions
guard+0.003
 poisoning+0.003
 injuries+0.003
 unfortunately+0.002
 payoff+0.002
Neg logit contributions
aling-0.007
alling-0.007
astics-0.006
irl-0.006
 sings-0.006
Token said
Token ID531
Feature activation+3.06e-02

Pos logit contributions
aling+0.046
alling+0.043
astics+0.042
irl+0.041
etition+0.038
Neg logit contributions
guard-0.023
 poisoning-0.020
 injuries-0.018
 last-0.018
 unfortunately-0.018
Token Fro
Token ID9734
Feature activation-4.53e-02

Pos logit contributions
 injuries+0.060
guard+0.059
 poisoning+0.057
 payoff+0.057
 unfortunately+0.053
Neg logit contributions
aling-0.061
alling-0.055
astics-0.051
irl-0.050
ophone-0.050
Tokenggy
Token ID19970
Feature activation-3.42e-02

Pos logit contributions
aling+0.067
irl+0.058
alling+0.058
astics+0.057
 sings+0.052
Neg logit contributions
guard-0.070
 poisoning-0.070
 injuries-0.069
 payoff-0.066
 unfortunately-0.065
Token,
Token ID11
Feature activation+3.70e-02

Pos logit contributions
aling+0.082
alling+0.076
astics+0.073
irl+0.073
 sings+0.070
Neg logit contributions
guard-0.023
 poisoning-0.023
 injuries-0.023
 unfortunately-0.019
 payoff-0.019
Token in
Token ID287
Feature activation-1.55e-02

Pos logit contributions
 injuries+0.055
guard+0.055
 poisoning+0.052
 payoff+0.051
 unfortunately+0.049
Neg logit contributions
aling-0.060
alling-0.053
 sings-0.049
astics-0.047
irl-0.047
Token a
Token ID257
Feature activation-3.67e-02

Pos logit contributions
aling+0.031
alling+0.028
astics+0.026
irl+0.026
 sings+0.025
Neg logit contributions
guard-0.017
 poisoning-0.017
 injuries-0.017
 unfortunately-0.015
 payoff-0.015
Token kay
Token ID34681
Feature activation-1.43e-01

Pos logit contributions
aling+0.067
alling+0.061
astics+0.057
irl+0.056
 sings+0.054
Neg logit contributions
guard-0.047
 poisoning-0.047
 injuries-0.046
 unfortunately-0.042
 payoff-0.041
Tokenak
Token ID461
Feature activation-3.23e-02

Pos logit contributions
aling+0.212
 sings+0.210
alling+0.203
 dances+0.195
astics+0.189
Neg logit contributions
guard-0.205
 poisoning-0.201
t-0.193
 injuries-0.185
 last-0.175
Token.
Token ID13
Feature activation-1.31e-02

Pos logit contributions
aling+0.071
alling+0.065
astics+0.063
irl+0.062
 sings+0.061
Neg logit contributions
guard-0.029
 poisoning-0.029
 injuries-0.027
 unfortunately-0.025
ho-0.024
Token Fro
Token ID9734
Feature activation-5.62e-02

Pos logit contributions
aling+0.027
alling+0.024
astics+0.023
irl+0.023
ophone+0.022
Neg logit contributions
 poisoning-0.014
guard-0.014
 injuries-0.013
 unfortunately-0.012
ho-0.012
Tokenggy
Token ID19970
Feature activation-5.24e-02

Pos logit contributions
aling+0.092
alling+0.081
astics+0.077
irl+0.077
 sings+0.072
Neg logit contributions
guard-0.082
 poisoning-0.081
 injuries-0.080
 unfortunately-0.074
 payoff-0.074
Token said
Token ID531
Feature activation+3.26e-02

Pos logit contributions
aling+0.108
alling+0.100
astics+0.093
irl+0.092
 sings+0.087
Neg logit contributions
guard-0.055
 poisoning-0.054
 injuries-0.052
 unfortunately-0.048
ho-0.047
Token for
Token ID329
Feature activation+2.08e-02

Pos logit contributions
aling+0.048
alling+0.043
astics+0.042
 sings+0.042
irl+0.041
Neg logit contributions
guard-0.029
 poisoning-0.028
 injuries-0.026
 last-0.026
t-0.024
Token a
Token ID257
Feature activation+1.44e-02

Pos logit contributions
guard+0.026
 poisoning+0.026
 injuries+0.025
 unfortunately+0.023
 payoff+0.023
Neg logit contributions
aling-0.038
alling-0.035
astics-0.033
irl-0.033
 sings-0.031
Token ride
Token ID6594
Feature activation-3.19e-02

Pos logit contributions
guard+0.021
 poisoning+0.021
 injuries+0.020
 unfortunately+0.019
 payoff+0.019
Neg logit contributions
aling-0.024
alling-0.021
astics-0.020
irl-0.020
 sings-0.019
Token in
Token ID287
Feature activation-1.62e-02

Pos logit contributions
aling+0.059
alling+0.053
astics+0.051
irl+0.050
 sings+0.047
Neg logit contributions
 poisoning-0.039
 injuries-0.039
guard-0.039
 payoff-0.035
 unfortunately-0.035
Token the
Token ID262
Feature activation-2.17e-02

Pos logit contributions
aling+0.027
alling+0.025
irl+0.025
astics+0.023
c+0.022
Neg logit contributions
 poisoning-0.022
 department-0.021
guard-0.021
 injuries-0.021
ho-0.020
Token kay
Token ID34681
Feature activation-1.41e-01

Pos logit contributions
aling+0.043
alling+0.039
astics+0.039
irl+0.037
ophone+0.037
Neg logit contributions
guard-0.026
 poisoning-0.023
 last-0.022
 injuries-0.022
 payoff-0.022
Tokenak
Token ID461
Feature activation-2.25e-02

Pos logit contributions
 sings+0.234
 dances+0.213
aling+0.210
alling+0.205
 presses+0.201
Neg logit contributions
guard-0.196
t-0.179
 poisoning-0.174
ll-0.157
 last-0.154
Token and
Token ID290
Feature activation-4.52e-02

Pos logit contributions
aling+0.043
alling+0.039
astics+0.036
irl+0.035
ophone+0.035
Neg logit contributions
 injuries-0.027
 poisoning-0.026
guard-0.026
 payoff-0.024
 unfortunately-0.023
Token her
Token ID607
Feature activation-8.92e-03

Pos logit contributions
aling+0.096
alling+0.088
astics+0.083
irl+0.083
 sings+0.079
Neg logit contributions
guard-0.044
 poisoning-0.043
 injuries-0.042
 unfortunately-0.039
 payoff-0.038
Token mom
Token ID1995
Feature activation+2.99e-02

Pos logit contributions
aling+0.017
irl+0.017
alling+0.016
 nic+0.016
astics+0.016
Neg logit contributions
guard-0.008
 poisoning-0.008
 injuries-0.007
 unfortunately-0.007
 payoff-0.007
Token said
Token ID531
Feature activation+2.55e-02

Pos logit contributions
 injuries+0.033
 poisoning+0.032
guard+0.032
 unfortunately+0.029
 payoff+0.029
Neg logit contributions
aling-0.059
alling-0.054
astics-0.051
irl-0.051
 sings-0.049
Token forth
Token ID6071
Feature activation-1.46e-03

Pos logit contributions
aling+0.082
alling+0.077
astics+0.073
 essay+0.071
vel+0.070
Neg logit contributions
 poisoning-0.059
guard-0.059
 though-0.057
 last-0.056
 while-0.055
Token in
Token ID287
Feature activation-2.39e-02

Pos logit contributions
aling+0.003
alling+0.003
irl+0.003
astics+0.003
 sings+0.003
Neg logit contributions
guard-0.001
 injuries-0.001
 poisoning-0.001
 payoff-0.001
 unfortunately-0.001
Token the
Token ID262
Feature activation-2.95e-02

Pos logit contributions
aling+0.040
alling+0.036
irl+0.033
astics+0.032
 sings+0.031
Neg logit contributions
guard-0.035
 poisoning-0.034
 injuries-0.032
t-0.031
ho-0.031
Token sky
Token ID6766
Feature activation-1.91e-02

Pos logit contributions
aling+0.060
alling+0.054
astics+0.054
irl+0.051
ophone+0.050
Neg logit contributions
guard-0.032
 injuries-0.032
 last-0.031
 poisoning-0.030
 payoff-0.029
Token,
Token ID11
Feature activation+2.89e-03

Pos logit contributions
aling+0.041
alling+0.037
astics+0.035
irl+0.035
 sings+0.034
Neg logit contributions
guard-0.019
 poisoning-0.018
 injuries-0.018
 unfortunately-0.016
 payoff-0.016
Token spir
Token ID9158
Feature activation-1.41e-01

Pos logit contributions
 poisoning+0.004
guard+0.003
 injuries+0.003
 unfortunately+0.003
 last+0.003
Neg logit contributions
aling-0.005
alling-0.005
astics-0.005
irl-0.005
ophone-0.004
Tokenalling
Token ID9221
Feature activation-5.34e-02

Pos logit contributions
astics+0.227
irl+0.222
 essay+0.202
aling+0.198
 sings+0.196
Neg logit contributions
guard-0.169
 last-0.167
 poisoning-0.165
ho-0.165
t-0.164
Token higher
Token ID2440
Feature activation-2.02e-02

Pos logit contributions
aling+0.090
alling+0.080
astics+0.074
irl+0.073
 sings+0.070
Neg logit contributions
 injuries-0.075
 poisoning-0.075
guard-0.075
 unfortunately-0.068
 payoff-0.068
Token and
Token ID290
Feature activation-5.60e-02

Pos logit contributions
aling+0.031
alling+0.027
irl+0.026
 spir+0.026
 sings+0.026
Neg logit contributions
 poisoning-0.030
 injuries-0.030
guard-0.030
 mercy-0.028
 unfortunately-0.027
Token higher
Token ID2440
Feature activation-1.97e-02

Pos logit contributions
aling+0.091
astics+0.086
alling+0.084
irl+0.082
 essay+0.081
Neg logit contributions
 last-0.076
guard-0.076
 payoff-0.072
 though-0.071
 poisoning-0.070
Token.
Token ID13
Feature activation-2.21e-02

Pos logit contributions
aling+0.032
alling+0.028
irl+0.027
 sings+0.026
astics+0.026
Neg logit contributions
 poisoning-0.028
guard-0.028
 injuries-0.027
 unfortunately-0.025
 department-0.025
Token can
Token ID460
Feature activation+4.05e-02

Pos logit contributions
 injuries+0.065
guard+0.065
 poisoning+0.064
 unfortunately+0.058
 payoff+0.058
Neg logit contributions
aling-0.091
alling-0.081
astics-0.077
irl-0.077
 sings-0.074
Token we
Token ID356
Feature activation+3.11e-02

Pos logit contributions
guard+0.037
 poisoning+0.037
 injuries+0.036
 unfortunately+0.032
 payoff+0.031
Neg logit contributions
aling-0.088
alling-0.081
astics-0.077
irl-0.077
 sings-0.074
Token go
Token ID467
Feature activation-3.31e-02

Pos logit contributions
guard+0.051
 poisoning+0.051
 injuries+0.050
 unfortunately+0.047
 payoff+0.047
Neg logit contributions
aling-0.045
alling-0.039
astics-0.036
irl-0.036
 sings-0.034
Token in
Token ID287
Feature activation-1.10e-02

Pos logit contributions
aling+0.064
alling+0.058
astics+0.056
irl+0.055
 sings+0.054
Neg logit contributions
guard-0.038
 poisoning-0.037
 injuries-0.036
 last-0.033
 unfortunately-0.032
Token the
Token ID262
Feature activation-3.14e-02

Pos logit contributions
aling+0.019
alling+0.017
irl+0.017
astics+0.016
c+0.015
Neg logit contributions
 injuries-0.015
 poisoning-0.015
ho-0.014
guard-0.014
t-0.013
Token kay
Token ID34681
Feature activation-1.41e-01

Pos logit contributions
aling+0.061
alling+0.055
astics+0.052
irl+0.051
ophone+0.050
Neg logit contributions
guard-0.039
 poisoning-0.037
 injuries-0.036
 last-0.034
 payoff-0.034
Tokenak
Token ID461
Feature activation-2.73e-02

Pos logit contributions
 sings+0.223
aling+0.211
 dances+0.207
alling+0.204
 spir+0.192
Neg logit contributions
guard-0.201
t-0.181
 poisoning-0.180
 last-0.161
 injuries-0.159
Token?"
Token ID1701
Feature activation+2.30e-03

Pos logit contributions
aling+0.052
alling+0.047
astics+0.044
irl+0.044
ophone+0.042
Neg logit contributions
 injuries-0.034
 poisoning-0.032
guard-0.032
 payoff-0.030
 unfortunately-0.029
Token Her
Token ID2332
Feature activation-9.20e-03

Pos logit contributions
guard+0.003
 poisoning+0.003
 injuries+0.003
 unfortunately+0.002
 payoff+0.002
Neg logit contributions
aling-0.004
alling-0.004
astics-0.004
irl-0.004
 sings-0.004
Token dad
Token ID9955
Feature activation+6.75e-03

Pos logit contributions
aling+0.021
alling+0.020
irl+0.020
astics+0.020
 sings+0.019
Neg logit contributions
 poisoning-0.006
guard-0.006
 injuries-0.006
 unfortunately-0.005
 payoff-0.005
Token said
Token ID531
Feature activation+1.93e-02

Pos logit contributions
guard+0.007
 poisoning+0.006
 unfortunately+0.006
 injuries+0.005
ho+0.005
Neg logit contributions
aling-0.015
alling-0.014
astics-0.013
irl-0.013
atters-0.012
Token river
Token ID7850
Feature activation-3.94e-02

Pos logit contributions
aling+0.021
alling+0.019
astics+0.018
irl+0.017
 sings+0.017
Neg logit contributions
 injuries-0.018
 poisoning-0.018
guard-0.018
 unfortunately-0.017
 payoff-0.016
Token.
Token ID13
Feature activation-4.37e-03

Pos logit contributions
aling+0.075
alling+0.067
astics+0.064
irl+0.063
 sings+0.060
Neg logit contributions
 injuries-0.047
 poisoning-0.046
guard-0.046
 payoff-0.042
 unfortunately-0.042
Token They
Token ID1119
Feature activation-9.87e-04

Pos logit contributions
aling+0.007
alling+0.006
astics+0.006
irl+0.006
 sings+0.005
Neg logit contributions
 poisoning-0.006
guard-0.006
 injuries-0.006
 unfortunately-0.006
 payoff-0.006
Token saw
Token ID2497
Feature activation-8.15e-03

Pos logit contributions
aling+0.002
alling+0.002
astics+0.001
irl+0.001
 sings+0.001
Neg logit contributions
guard-0.001
 poisoning-0.001
 injuries-0.001
 unfortunately-0.001
 payoff-0.001
Token a
Token ID257
Feature activation-1.37e-02

Pos logit contributions
aling+0.010
alling+0.009
 sings+0.009
irl+0.009
astics+0.008
Neg logit contributions
 injuries-0.014
 poisoning-0.014
guard-0.014
 payoff-0.013
t-0.013
Token kay
Token ID34681
Feature activation-1.40e-01

Pos logit contributions
aling+0.024
alling+0.021
irl+0.020
astics+0.020
 sings+0.019
Neg logit contributions
 poisoning-0.018
guard-0.018
 injuries-0.018
 unfortunately-0.016
 payoff-0.016
Tokenak
Token ID461
Feature activation-2.84e-02

Pos logit contributions
aling+0.202
 sings+0.199
alling+0.191
astics+0.181
 dances+0.179
Neg logit contributions
guard-0.224
 poisoning-0.209
t-0.209
 last-0.185
 injuries-0.184
Token on
Token ID319
Feature activation-2.58e-02

Pos logit contributions
aling+0.061
alling+0.056
astics+0.054
irl+0.053
 sings+0.051
Neg logit contributions
guard-0.027
 poisoning-0.026
 injuries-0.026
 unfortunately-0.023
 payoff-0.022
Token the
Token ID262
Feature activation-1.49e-02

Pos logit contributions
aling+0.045
alling+0.040
irl+0.039
astics+0.038
 sings+0.036
Neg logit contributions
guard-0.034
 injuries-0.034
 poisoning-0.033
 department-0.030
t-0.030
Token shore
Token ID15191
Feature activation-3.59e-02

Pos logit contributions
aling+0.018
alling+0.016
 spir+0.015
 zoom+0.014
 nic+0.014
Neg logit contributions
 injuries-0.025
guard-0.024
 poisoning-0.024
 unfortunately-0.024
 payoff-0.024
Token.
Token ID13
Feature activation-1.65e-02

Pos logit contributions
aling+0.064
alling+0.057
astics+0.053
irl+0.052
ophone+0.051
Neg logit contributions
 injuries-0.046
 poisoning-0.044
guard-0.043
 payoff-0.040
 mercy-0.040
Token hopped
Token ID45230
Feature activation-7.79e-02

Pos logit contributions
aling+0.001
alling+0.001
astics+0.001
etition+0.001
atters+0.001
Neg logit contributions
guard-0.000
 last-0.000
 poisoning-0.000
 now-0.000
 to-0.000
Token on
Token ID319
Feature activation-1.18e-02

Pos logit contributions
aling+0.136
alling+0.121
astics+0.116
irl+0.113
 sings+0.109
Neg logit contributions
guard-0.107
 poisoning-0.105
 injuries-0.099
 unfortunately-0.095
 payoff-0.094
Token,
Token ID11
Feature activation+3.43e-03

Pos logit contributions
aling+0.026
alling+0.024
astics+0.023
ophone+0.022
irl+0.022
Neg logit contributions
 poisoning-0.011
 injuries-0.010
guard-0.010
 last-0.009
 unfortunately-0.009
Token and
Token ID290
Feature activation-4.48e-02

Pos logit contributions
 but+0.004
 while+0.004
 to+0.004
 though+0.004
 poisoning+0.004
Neg logit contributions
aling-0.006
astics-0.006
irl-0.006
alling-0.006
atters-0.005
Token started
Token ID2067
Feature activation-2.10e-02

Pos logit contributions
aling+0.088
alling+0.081
astics+0.079
irl+0.078
atters+0.075
Neg logit contributions
guard-0.049
 poisoning-0.046
 unfortunately-0.044
 though-0.043
 they-0.043
Token ped
Token ID7190
Feature activation-1.39e-01

Pos logit contributions
aling+0.048
alling+0.043
irl+0.043
ophone+0.043
astics+0.043
Neg logit contributions
 to-0.020
 that-0.016
 while-0.015
 they-0.014
guard-0.013
Tokenalling
Token ID9221
Feature activation-5.63e-02

Pos logit contributions
 sings+0.091
 nic+0.062
 sights+0.061
 dances+0.060
 revers+0.058
Neg logit contributions
 since-0.293
t-0.291
guard-0.282
 then-0.278
 last-0.275
Token.
Token ID13
Feature activation-1.11e-02

Pos logit contributions
aling+0.115
alling+0.105
astics+0.100
irl+0.099
 sings+0.096
Neg logit contributions
guard-0.059
 poisoning-0.057
 injuries-0.057
 unfortunately-0.051
 payoff-0.050
Token She
Token ID1375
Feature activation-4.14e-03

Pos logit contributions
aling+0.021
alling+0.019
astics+0.018
irl+0.017
 sings+0.017
Neg logit contributions
 poisoning-0.013
guard-0.012
 injuries-0.012
t-0.012
 unfortunately-0.012
Token felt
Token ID2936
Feature activation-4.76e-02

Pos logit contributions
aling+0.008
astics+0.007
alling+0.007
etition+0.007
atters+0.007
Neg logit contributions
guard-0.005
 last-0.004
t-0.004
 to-0.004
 while-0.004
Token like
Token ID588
Feature activation-9.21e-03

Pos logit contributions
aling+0.103
astics+0.095
alling+0.094
irl+0.092
ophone+0.091
Neg logit contributions
 poisoning-0.039
guard-0.039
 unfortunately-0.036
 though-0.036
 injuries-0.034
Token and
Token ID290
Feature activation-3.77e-03

Pos logit contributions
aling+0.055
alling+0.051
astics+0.050
irl+0.049
atters+0.046
Neg logit contributions
guard-0.020
 poisoning-0.017
 last-0.016
 payoff-0.015
ho-0.015
Token Anne
Token ID15397
Feature activation+3.77e-03

Pos logit contributions
aling+0.005
alling+0.005
 sings+0.004
astics+0.004
irl+0.004
Neg logit contributions
 poisoning-0.006
 injuries-0.006
guard-0.006
 unfortunately-0.006
ho-0.006
Token jumped
Token ID11687
Feature activation-6.56e-02

Pos logit contributions
guard+0.005
 last+0.005
 poisoning+0.005
 help+0.005
ho+0.004
Neg logit contributions
aling-0.006
astics-0.006
alling-0.006
irl-0.005
 sings-0.005
Token in
Token ID287
Feature activation-1.77e-02

Pos logit contributions
aling+0.115
astics+0.101
alling+0.100
 sings+0.097
irl+0.096
Neg logit contributions
guard-0.090
 poisoning-0.085
 injuries-0.081
 last-0.080
 unfortunately-0.079
Token the
Token ID262
Feature activation-2.67e-02

Pos logit contributions
aling+0.028
alling+0.026
irl+0.024
 ped+0.023
astics+0.023
Neg logit contributions
guard-0.026
 poisoning-0.025
t-0.023
 injuries-0.023
ho-0.022
Token kay
Token ID34681
Feature activation-1.39e-01

Pos logit contributions
aling+0.052
astics+0.048
vel+0.047
alling+0.046
irl+0.046
Neg logit contributions
guard-0.032
 last-0.030
 payoff-0.028
 injuries-0.027
 poisoning-0.026
Tokenak
Token ID461
Feature activation-2.73e-02

Pos logit contributions
 sings+0.237
 dances+0.222
aling+0.210
alling+0.202
astics+0.200
Neg logit contributions
guard-0.186
t-0.170
 poisoning-0.159
less-0.149
ll-0.144
Token and
Token ID290
Feature activation-5.60e-02

Pos logit contributions
aling+0.060
alling+0.055
astics+0.053
irl+0.052
 sings+0.050
Neg logit contributions
guard-0.024
 poisoning-0.024
 injuries-0.023
 unfortunately-0.021
 payoff-0.020
Token padd
Token ID14098
Feature activation-9.06e-02

Pos logit contributions
aling+0.105
astics+0.098
vel+0.095
alling+0.095
irl+0.092
Neg logit contributions
 they-0.069
guard-0.065
 while-0.062
 last-0.060
 to-0.059
Tokenled
Token ID992
Feature activation-4.10e-02

Pos logit contributions
aling+0.157
alling+0.142
astics+0.140
irl+0.133
 sings+0.126
Neg logit contributions
guard-0.127
 poisoning-0.117
 injuries-0.113
 last-0.108
 payoff-0.106
Token away
Token ID1497
Feature activation+1.94e-02

Pos logit contributions
aling+0.074
alling+0.066
astics+0.065
irl+0.062
ophone+0.059
Neg logit contributions
guard-0.052
 poisoning-0.051
 injuries-0.050
 last-0.048
 payoff-0.048
Token day
Token ID1110
Feature activation+6.54e-02

Pos logit contributions
 injuries+0.039
 poisoning+0.038
guard+0.037
 unfortunately+0.034
 payoff+0.034
Neg logit contributions
aling-0.059
alling-0.053
irl-0.051
astics-0.050
 sings-0.048
Token,
Token ID11
Feature activation+2.40e-02

Pos logit contributions
 injuries+0.085
guard+0.077
 poisoning+0.076
 mercy+0.074
 payoff+0.073
Neg logit contributions
irl-0.108
aling-0.107
astics-0.101
irled-0.101
 dances-0.101
Token they
Token ID484
Feature activation-7.71e-03

Pos logit contributions
 injuries+0.036
 mercy+0.036
guard+0.033
ache+0.032
 payoff+0.031
Neg logit contributions
irl-0.034
 plays-0.031
irled-0.030
 dances-0.030
astics-0.029
Token saw
Token ID2497
Feature activation-8.82e-03

Pos logit contributions
aling+0.012
alling+0.011
irl+0.010
astics+0.010
 sings+0.009
Neg logit contributions
 injuries-0.012
 poisoning-0.012
guard-0.012
 unfortunately-0.011
 mercy-0.011
Token a
Token ID257
Feature activation-1.51e-02

Pos logit contributions
aling+0.013
alling+0.011
irl+0.011
 sings+0.010
astics+0.010
Neg logit contributions
 injuries-0.014
guard-0.014
 poisoning-0.014
 payoff-0.013
t-0.013
Token kay
Token ID34681
Feature activation-1.39e-01

Pos logit contributions
aling+0.029
alling+0.026
astics+0.025
irl+0.024
 sings+0.023
Neg logit contributions
guard-0.019
 poisoning-0.018
 injuries-0.018
 unfortunately-0.017
 payoff-0.016
Tokenak
Token ID461
Feature activation-2.86e-02

Pos logit contributions
 sings+0.209
aling+0.203
alling+0.194
 dances+0.191
astics+0.188
Neg logit contributions
guard-0.212
t-0.195
 poisoning-0.191
 last-0.171
 injuries-0.171
Token on
Token ID319
Feature activation-2.30e-02

Pos logit contributions
aling+0.060
alling+0.055
astics+0.054
irl+0.053
 sings+0.052
Neg logit contributions
guard-0.028
 poisoning-0.027
 injuries-0.025
 unfortunately-0.024
ho-0.023
Token the
Token ID262
Feature activation-1.44e-02

Pos logit contributions
aling+0.041
alling+0.037
irl+0.035
astics+0.035
 sings+0.033
Neg logit contributions
guard-0.029
 injuries-0.029
 poisoning-0.029
t-0.027
 department-0.026
Token shore
Token ID15191
Feature activation-3.17e-02

Pos logit contributions
aling+0.020
alling+0.018
astics+0.016
irl+0.016
 spir+0.016
Neg logit contributions
 injuries-0.022
guard-0.022
 poisoning-0.022
 unfortunately-0.021
 payoff-0.020
Token.
Token ID13
Feature activation-2.61e-02

Pos logit contributions
aling+0.058
alling+0.052
astics+0.048
irl+0.047
ophone+0.046
Neg logit contributions
 injuries-0.040
 poisoning-0.039
guard-0.038
 payoff-0.035
 unfortunately-0.035
Token on
Token ID319
Feature activation-4.22e-03

Pos logit contributions
aling+0.122
alling+0.108
astics+0.103
irl+0.101
 sings+0.097
Neg logit contributions
guard-0.111
 poisoning-0.108
 injuries-0.105
 unfortunately-0.099
 payoff-0.097
Token their
Token ID511
Feature activation-6.21e-03

Pos logit contributions
aling+0.008
alling+0.007
astics+0.007
irl+0.007
 sings+0.007
Neg logit contributions
guard-0.005
 poisoning-0.005
 injuries-0.005
 unfortunately-0.004
 payoff-0.004
Token bikes
Token ID16715
Feature activation-3.38e-02

Pos logit contributions
aling+0.010
alling+0.009
astics+0.008
irl+0.008
 sings+0.008
Neg logit contributions
guard-0.010
 poisoning-0.009
 injuries-0.009
 unfortunately-0.009
 payoff-0.009
Token and
Token ID290
Feature activation-4.79e-02

Pos logit contributions
aling+0.068
alling+0.062
astics+0.059
irl+0.058
 sings+0.056
Neg logit contributions
guard-0.036
 poisoning-0.036
 injuries-0.035
 unfortunately-0.032
 payoff-0.031
Token started
Token ID2067
Feature activation-2.36e-02

Pos logit contributions
aling+0.083
alling+0.074
astics+0.070
irl+0.070
 sings+0.065
Neg logit contributions
guard-0.065
 poisoning-0.064
 injuries-0.063
 unfortunately-0.059
 payoff-0.058
Token ped
Token ID7190
Feature activation-1.38e-01

Pos logit contributions
aling+0.056
alling+0.051
irl+0.049
astics+0.049
ophone+0.049
Neg logit contributions
guard-0.016
 injuries-0.016
 poisoning-0.015
 payoff-0.013
 to-0.013
Tokenaling
Token ID4272
Feature activation-4.38e-02

Pos logit contributions
 sings+0.088
 dances+0.057
 sights+0.049
azes+0.045
 nic+0.042
Neg logit contributions
guard-0.324
t-0.317
 last-0.294
Not-0.289
 since-0.288
Token.
Token ID13
Feature activation-2.71e-02

Pos logit contributions
aling+0.083
alling+0.076
astics+0.071
irl+0.071
 sings+0.067
Neg logit contributions
 poisoning-0.052
guard-0.052
 injuries-0.051
 unfortunately-0.047
 payoff-0.046
Token
Token ID198
Feature activation+2.22e-02

Pos logit contributions
aling+0.049
alling+0.045
astics+0.042
irl+0.042
 sings+0.040
Neg logit contributions
guard-0.033
 poisoning-0.033
 injuries-0.033
 unfortunately-0.031
 payoff-0.030
Token
Token ID198
Feature activation+1.71e-02

Pos logit contributions
 poisoning+0.027
guard+0.026
 injuries+0.026
 payoff+0.023
ho+0.023
Neg logit contributions
aling-0.043
alling-0.039
irl-0.037
astics-0.037
 sings-0.034
TokenWhen
Token ID2215
Feature activation+5.24e-02

Pos logit contributions
 poisoning+0.018
guard+0.018
 injuries+0.018
 payoff+0.016
 unfortunately+0.016
Neg logit contributions
aling-0.035
alling-0.032
astics-0.030
irl-0.030
 sings-0.029
Token saw
Token ID2497
Feature activation-6.82e-03

Pos logit contributions
aling+0.084
alling+0.077
astics+0.073
irl+0.072
 sings+0.068
Neg logit contributions
guard-0.047
 poisoning-0.047
 injuries-0.044
 unfortunately-0.041
 payoff-0.040
Token a
Token ID257
Feature activation-1.92e-02

Pos logit contributions
aling+0.009
alling+0.007
 sings+0.007
irl+0.007
astics+0.007
Neg logit contributions
 injuries-0.012
guard-0.011
 poisoning-0.011
t-0.011
 payoff-0.011
Token big
Token ID1263
Feature activation-3.77e-02

Pos logit contributions
aling+0.037
alling+0.033
astics+0.031
irl+0.030
 sings+0.029
Neg logit contributions
guard-0.024
 poisoning-0.023
 injuries-0.022
 unfortunately-0.021
 payoff-0.020
Token,
Token ID11
Feature activation-1.14e-02

Pos logit contributions
aling+0.075
alling+0.068
astics+0.065
irl+0.063
ophone+0.060
Neg logit contributions
guard-0.043
 poisoning-0.043
 injuries-0.041
 unfortunately-0.038
 payoff-0.037
Token red
Token ID2266
Feature activation-4.52e-02

Pos logit contributions
aling+0.023
alling+0.021
astics+0.020
aled+0.019
irled+0.019
Neg logit contributions
guard-0.014
 poisoning-0.013
 last-0.012
 injuries-0.012
 but-0.012
Token kay
Token ID34681
Feature activation-1.38e-01

Pos logit contributions
aling+0.081
alling+0.073
irl+0.070
astics+0.069
 sings+0.066
Neg logit contributions
 injuries-0.056
 poisoning-0.056
guard-0.056
 unfortunately-0.051
 payoff-0.050
Tokenak
Token ID461
Feature activation-3.69e-02

Pos logit contributions
 sings+0.213
aling+0.207
alling+0.201
 dances+0.196
astics+0.189
Neg logit contributions
guard-0.202
t-0.194
 poisoning-0.190
ho-0.170
 last-0.162
Token.
Token ID13
Feature activation-2.08e-02

Pos logit contributions
aling+0.086
alling+0.080
astics+0.078
irl+0.077
 sings+0.075
Neg logit contributions
guard-0.029
 poisoning-0.027
 that-0.026
t-0.025
ho-0.024
Token Lily
Token ID20037
Feature activation-2.36e-02

Pos logit contributions
aling+0.038
alling+0.034
astics+0.032
irl+0.032
 sings+0.030
Neg logit contributions
guard-0.027
 poisoning-0.027
 injuries-0.026
 unfortunately-0.024
 payoff-0.024
Token said
Token ID531
Feature activation+3.06e-02

Pos logit contributions
aling+0.046
alling+0.043
astics+0.039
irl+0.038
 sings+0.037
Neg logit contributions
guard-0.027
 poisoning-0.027
 unfortunately-0.025
 injuries-0.024
ho-0.024
Token,
Token ID11
Feature activation+3.86e-02

Pos logit contributions
 injuries+0.032
 poisoning+0.029
guard+0.029
 payoff+0.025
 extingu+0.025
Neg logit contributions
aling-0.061
alling-0.056
irl-0.056
astics-0.056
 sings-0.055
Token food
Token ID2057
Feature activation+2.34e-02

Pos logit contributions
guard+0.011
 injuries+0.010
 poisoning+0.010
ho+0.010
 unfortunately+0.010
Neg logit contributions
aling-0.013
alling-0.012
astics-0.011
irl-0.011
 sings-0.010
Token.
Token ID13
Feature activation+5.51e-03

Pos logit contributions
 payoff+0.040
 injuries+0.038
Girls+0.037
 fourteen+0.037
 extingu+0.037
Neg logit contributions
aling-0.025
astics-0.023
irl-0.021
 rides-0.021
 nic-0.020
Token He
Token ID679
Feature activation+2.60e-04

Pos logit contributions
 poisoning+0.007
guard+0.007
 injuries+0.007
 unfortunately+0.007
t+0.006
Neg logit contributions
aling-0.010
alling-0.009
irl-0.008
astics-0.008
 sings-0.008
Token saw
Token ID2497
Feature activation-5.06e-03

Pos logit contributions
 poisoning+0.000
guard+0.000
 injuries+0.000
 unfortunately+0.000
 payoff+0.000
Neg logit contributions
aling-0.000
alling-0.000
astics-0.000
irl-0.000
 sings-0.000
Token a
Token ID257
Feature activation-1.82e-02

Pos logit contributions
aling+0.005
 nic+0.004
alling+0.004
 rides+0.004
 zoom+0.004
Neg logit contributions
guard-0.009
 injuries-0.009
t-0.009
ho-0.009
 poisoning-0.009
Token kay
Token ID34681
Feature activation-1.38e-01

Pos logit contributions
aling+0.025
irl+0.023
alling+0.022
astics+0.022
 sings+0.021
Neg logit contributions
 injuries-0.027
guard-0.027
 poisoning-0.026
 unfortunately-0.025
 payoff-0.025
Tokenak
Token ID461
Feature activation-2.69e-02

Pos logit contributions
 sings+0.216
aling+0.204
 dances+0.198
alling+0.197
 presses+0.187
Neg logit contributions
guard-0.208
t-0.200
 poisoning-0.199
 injuries-0.171
 last-0.169
Token by
Token ID416
Feature activation-1.40e-02

Pos logit contributions
aling+0.055
alling+0.051
astics+0.048
irl+0.048
 sings+0.046
Neg logit contributions
guard-0.028
 poisoning-0.028
 injuries-0.027
 unfortunately-0.025
 payoff-0.024
Token the
Token ID262
Feature activation-1.44e-02

Pos logit contributions
aling+0.027
alling+0.024
astics+0.023
irl+0.023
 sings+0.022
Neg logit contributions
guard-0.016
 poisoning-0.016
 injuries-0.016
 unfortunately-0.014
 payoff-0.014
Token river
Token ID7850
Feature activation-3.50e-02

Pos logit contributions
aling+0.023
alling+0.020
astics+0.019
irl+0.018
 sings+0.018
Neg logit contributions
guard-0.020
 poisoning-0.020
 injuries-0.020
 unfortunately-0.019
t-0.018
Token.
Token ID13
Feature activation-8.73e-03

Pos logit contributions
aling+0.072
alling+0.066
astics+0.063
irl+0.062
 sings+0.060
Neg logit contributions
guard-0.036
 poisoning-0.036
 injuries-0.035
 unfortunately-0.031
 payoff-0.031
Token to
Token ID284
Feature activation+2.60e-02

Pos logit contributions
 injuries+0.013
 poisoning+0.012
guard+0.012
 unfortunately+0.012
 payoff+0.010
Neg logit contributions
aling-0.047
alling-0.043
astics-0.042
irl-0.041
 sings-0.041
Token join
Token ID4654
Feature activation-3.25e-02

Pos logit contributions
guard+0.040
 poisoning+0.040
 injuries+0.039
 unfortunately+0.038
t+0.037
Neg logit contributions
aling-0.040
alling-0.035
astics-0.033
irl-0.033
 sings-0.031
Token them
Token ID606
Feature activation+2.38e-02

Pos logit contributions
aling+0.071
alling+0.066
astics+0.062
irl+0.062
 sings+0.060
Neg logit contributions
guard-0.029
 poisoning-0.029
 injuries-0.028
 unfortunately-0.025
 payoff-0.024
Token in
Token ID287
Feature activation-6.98e-03

Pos logit contributions
 injuries+0.028
 poisoning+0.028
guard+0.028
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.045
alling-0.040
astics-0.039
irl-0.037
 sings-0.037
Token the
Token ID262
Feature activation-2.41e-02

Pos logit contributions
aling+0.012
alling+0.011
irl+0.011
astics+0.011
 sings+0.010
Neg logit contributions
 poisoning-0.009
guard-0.009
 injuries-0.008
 unfortunately-0.008
t-0.008
Token kay
Token ID34681
Feature activation-1.38e-01

Pos logit contributions
aling+0.046
alling+0.043
astics+0.039
irl+0.038
ophone+0.038
Neg logit contributions
guard-0.030
 poisoning-0.028
 injuries-0.027
 last-0.027
 payoff-0.027
Tokenak
Token ID461
Feature activation-2.35e-02

Pos logit contributions
 sings+0.204
aling+0.203
alling+0.196
 dances+0.187
etition+0.181
Neg logit contributions
guard-0.209
t-0.179
 poisoning-0.179
 last-0.165
 injuries-0.163
Token.
Token ID13
Feature activation-3.96e-02

Pos logit contributions
aling+0.046
alling+0.041
astics+0.039
irl+0.038
 sings+0.037
Neg logit contributions
 injuries-0.027
 poisoning-0.027
guard-0.026
 unfortunately-0.024
 payoff-0.024
Token
Token ID198
Feature activation+1.69e-02

Pos logit contributions
aling+0.057
alling+0.049
astics+0.047
irl+0.045
 sings+0.045
Neg logit contributions
 injuries-0.066
 poisoning-0.066
guard-0.065
 unfortunately-0.060
 payoff-0.060
Token
Token ID198
Feature activation+1.35e-02

Pos logit contributions
 poisoning+0.018
guard+0.018
 injuries+0.018
 unfortunately+0.016
 payoff+0.016
Neg logit contributions
aling-0.034
alling-0.031
irl-0.030
astics-0.030
 sings-0.028
Token"
Token ID1
Feature activation+2.51e-02

Pos logit contributions
 poisoning+0.015
 injuries+0.015
guard+0.014
 unfortunately+0.013
 payoff+0.013
Neg logit contributions
aling-0.027
alling-0.025
astics-0.024
irl-0.023
 sings-0.022
Token Anne
Token ID15397
Feature activation+1.05e-02

Pos logit contributions
 poisoning+0.016
 injuries+0.016
guard+0.016
 unfortunately+0.014
 payoff+0.014
Neg logit contributions
aling-0.015
alling-0.013
astics-0.012
 sings-0.012
irl-0.012
Token began
Token ID2540
Feature activation-1.37e-02

Pos logit contributions
guard+0.015
 poisoning+0.015
 injuries+0.015
 payoff+0.014
 unfortunately+0.014
Neg logit contributions
aling-0.017
alling-0.015
astics-0.014
irl-0.014
 sings-0.013
Token to
Token ID284
Feature activation-6.74e-03

Pos logit contributions
aling+0.038
alling+0.035
astics+0.034
irl+0.034
 sings+0.033
Neg logit contributions
guard-0.004
 poisoning-0.004
 injuries-0.004
 unfortunately-0.003
 payoff-0.003
Token build
Token ID1382
Feature activation-2.47e-02

Pos logit contributions
aling+0.012
alling+0.011
astics+0.010
 sings+0.010
irl+0.010
Neg logit contributions
 poisoning-0.008
guard-0.008
 injuries-0.008
 last-0.008
 payoff-0.008
Token a
Token ID257
Feature activation-1.84e-02

Pos logit contributions
aling+0.057
alling+0.052
astics+0.049
ophone+0.048
 sings+0.047
Neg logit contributions
 injuries-0.020
 poisoning-0.019
 last-0.019
guard-0.019
 unfortunately-0.018
Token kay
Token ID34681
Feature activation-1.38e-01

Pos logit contributions
aling+0.035
alling+0.032
astics+0.030
 sings+0.030
irl+0.029
Neg logit contributions
guard-0.023
 injuries-0.022
 poisoning-0.022
 last-0.020
 unfortunately-0.020
Tokenak
Token ID461
Feature activation-2.53e-02

Pos logit contributions
 sings+0.230
 dances+0.209
aling+0.202
alling+0.195
 presses+0.193
Neg logit contributions
guard-0.198
t-0.180
 poisoning-0.167
 last-0.162
 injuries-0.154
Token.
Token ID13
Feature activation-2.25e-02

Pos logit contributions
aling+0.053
alling+0.049
astics+0.047
irl+0.046
 sings+0.045
Neg logit contributions
guard-0.025
 poisoning-0.024
 injuries-0.024
 unfortunately-0.022
 payoff-0.021
Token They
Token ID1119
Feature activation-9.58e-03

Pos logit contributions
aling+0.038
alling+0.034
irl+0.033
astics+0.032
 sings+0.030
Neg logit contributions
guard-0.031
 poisoning-0.031
 injuries-0.029
 unfortunately-0.029
 payoff-0.028
Token found
Token ID1043
Feature activation-2.04e-02

Pos logit contributions
aling+0.016
alling+0.014
astics+0.014
irl+0.013
 sings+0.013
Neg logit contributions
guard-0.014
 poisoning-0.014
 injuries-0.013
 unfortunately-0.013
 last-0.012
Token some
Token ID617
Feature activation+1.81e-02

Pos logit contributions
aling+0.029
alling+0.025
irl+0.023
astics+0.022
 sings+0.022
Neg logit contributions
 injuries-0.034
guard-0.033
 poisoning-0.033
 payoff-0.030
 unfortunately-0.030
Token hose
Token ID31489
Feature activation-3.68e-02

Pos logit contributions
aling+0.010
 nic+0.009
alling+0.009
irl+0.009
astics+0.008
Neg logit contributions
 poisoning-0.011
 injuries-0.011
guard-0.011
 unfortunately-0.011
t-0.011
Token.
Token ID13
Feature activation-1.49e-02

Pos logit contributions
aling+0.060
alling+0.054
astics+0.050
ophone+0.049
c+0.049
Neg logit contributions
 injuries-0.052
guard-0.049
 poisoning-0.048
 payoff-0.046
 unfortunately-0.045
Token They
Token ID1119
Feature activation-5.91e-03

Pos logit contributions
aling+0.026
alling+0.024
astics+0.022
irl+0.022
 sings+0.021
Neg logit contributions
guard-0.020
 poisoning-0.020
 injuries-0.019
 unfortunately-0.018
 payoff-0.018
Token have
Token ID423
Feature activation-1.61e-02

Pos logit contributions
aling+0.010
alling+0.009
astics+0.009
irl+0.008
 sings+0.008
Neg logit contributions
 poisoning-0.008
guard-0.008
 injuries-0.008
 last-0.007
 unfortunately-0.007
Token a
Token ID257
Feature activation-8.69e-03

Pos logit contributions
aling+0.024
astics+0.022
irl+0.021
alling+0.021
 sings+0.019
Neg logit contributions
guard-0.026
 injuries-0.024
 poisoning-0.024
 unfortunately-0.023
ho-0.023
Token kay
Token ID34681
Feature activation-1.38e-01

Pos logit contributions
aling+0.015
irl+0.013
alling+0.013
astics+0.013
 sings+0.012
Neg logit contributions
guard-0.012
 poisoning-0.011
 injuries-0.011
 unfortunately-0.011
 payoff-0.010
Tokenak
Token ID461
Feature activation-2.98e-02

Pos logit contributions
aling+0.207
alling+0.199
 sings+0.196
astics+0.181
 presses+0.175
Neg logit contributions
guard-0.205
 poisoning-0.202
t-0.198
 injuries-0.181
 last-0.176
Token toy
Token ID13373
Feature activation-2.58e-03

Pos logit contributions
aling+0.060
alling+0.055
astics+0.052
irl+0.051
 sings+0.050
Neg logit contributions
 injuries-0.033
 poisoning-0.031
guard-0.031
 unfortunately-0.028
 payoff-0.028
Token.
Token ID13
Feature activation-1.75e-02

Pos logit contributions
aling+0.004
alling+0.004
astics+0.003
ophone+0.003
am+0.003
Neg logit contributions
 injuries-0.004
 poisoning-0.003
guard-0.003
 payoff-0.003
 unfortunately-0.003
Token It
Token ID632
Feature activation+4.39e-03

Pos logit contributions
aling+0.031
alling+0.028
astics+0.027
irl+0.026
 sings+0.025
Neg logit contributions
guard-0.023
 poisoning-0.023
 injuries-0.023
 unfortunately-0.021
 payoff-0.020
Token is
Token ID318
Feature activation-4.07e-02

Pos logit contributions
guard+0.006
 last+0.005
 poisoning+0.005
 they+0.005
 that+0.005
Neg logit contributions
aling-0.008
alling-0.008
astics-0.007
vel-0.007
 essay-0.007
Token cooking
Token ID10801
Feature activation-1.34e-02

Pos logit contributions
aling+0.027
alling+0.024
irl+0.023
astics+0.023
 sings+0.021
Neg logit contributions
guard-0.021
 poisoning-0.020
 injuries-0.020
 payoff-0.018
 unfortunately-0.018
Token.
Token ID13
Feature activation-1.43e-02

Pos logit contributions
aling+0.022
irl+0.020
alling+0.019
astics+0.019
 sings+0.018
Neg logit contributions
guard-0.019
 injuries-0.018
 poisoning-0.017
 department-0.017
ho-0.017
Token
Token ID198
Feature activation+2.83e-02

Pos logit contributions
aling+0.029
alling+0.026
irl+0.025
astics+0.025
 sings+0.023
Neg logit contributions
 poisoning-0.015
guard-0.015
t-0.015
 unfortunately-0.014
Not-0.014
Token
Token ID198
Feature activation+2.24e-02

Pos logit contributions
guard+0.023
 poisoning+0.022
 injuries+0.020
t+0.019
 unfortunately+0.019
Neg logit contributions
aling-0.065
alling-0.060
astics-0.057
irl-0.056
 sings-0.055
Token"
Token ID1
Feature activation+2.85e-02

Pos logit contributions
Not+0.014
t+0.014
guard+0.014
 unfortunately+0.012
 poisoning+0.012
Neg logit contributions
aling-0.054
alling-0.051
 sings-0.048
astics-0.047
irl-0.046
TokenMom
Token ID29252
Feature activation+5.53e-02

Pos logit contributions
 injuries+0.031
guard+0.031
 poisoning+0.030
 mercy+0.027
 unfortunately+0.026
Neg logit contributions
aling-0.056
alling-0.051
astics-0.048
irl-0.047
 sings-0.046
Tokenmy
Token ID1820
Feature activation+4.37e-02

Pos logit contributions
guard+0.065
 poisoning+0.063
 injuries+0.063
 unfortunately+0.056
 payoff+0.055
Neg logit contributions
aling-0.107
alling-0.097
astics-0.091
irl-0.091
 sings-0.088
Token,
Token ID11
Feature activation+5.25e-02

Pos logit contributions
guard+0.031
 injuries+0.030
 poisoning+0.030
 unfortunately+0.025
 payoff+0.024
Neg logit contributions
aling-0.104
alling-0.096
astics-0.092
irl-0.091
 sings-0.089
Token what
Token ID644
Feature activation+5.55e-02

Pos logit contributions
 injuries+0.068
guard+0.068
 poisoning+0.064
 payoff+0.059
 unfortunately+0.058
Neg logit contributions
aling-0.092
alling-0.082
irl-0.078
astics-0.078
 sings-0.077
Token are
Token ID389
Feature activation+1.39e-02

Pos logit contributions
 poisoning+0.078
guard+0.077
 injuries+0.068
 last+0.068
t+0.068
Neg logit contributions
aling-0.099
alling-0.091
astics-0.079
irl-0.078
aled-0.077
Token you
Token ID345
Feature activation+2.30e-02

Pos logit contributions
 injuries+0.017
guard+0.017
 poisoning+0.016
 payoff+0.015
 department+0.015
Neg logit contributions
aling-0.024
irl-0.022
astics-0.022
alling-0.022
 sings-0.021
Tokenily
Token ID813
Feature activation-6.37e-03

Pos logit contributions
aling+0.056
alling+0.051
astics+0.049
irl+0.048
 sings+0.046
Neg logit contributions
guard-0.030
 poisoning-0.030
 injuries-0.029
 unfortunately-0.026
 payoff-0.026
Token says
Token ID1139
Feature activation+4.03e-02

Pos logit contributions
aling+0.015
alling+0.015
astics+0.014
irl+0.014
beh+0.014
Neg logit contributions
guard-0.004
 last-0.003
 poisoning-0.003
 payoff-0.003
 that-0.003
Token,
Token ID11
Feature activation+4.60e-02

Pos logit contributions
 poisoning+0.036
 injuries+0.034
guard+0.033
 extingu+0.028
 unfortunately+0.028
Neg logit contributions
aling-0.085
alling-0.078
 sings-0.078
irl-0.077
astics-0.077
Token "
Token ID366
Feature activation-8.10e-03

Pos logit contributions
guard+0.064
 poisoning+0.064
 injuries+0.063
 unfortunately+0.057
 payoff+0.057
Neg logit contributions
aling-0.078
alling-0.069
astics-0.065
irl-0.064
 sings-0.063
TokenBen
Token ID11696
Feature activation+9.16e-03

Pos logit contributions
aling+0.016
alling+0.014
astics+0.013
irl+0.013
 sings+0.013
Neg logit contributions
 injuries-0.010
 poisoning-0.010
guard-0.009
 unfortunately-0.009
 department-0.008
Token,
Token ID11
Feature activation+5.07e-02

Pos logit contributions
 injuries+0.008
guard+0.008
 poisoning+0.007
 unfortunately+0.007
 payoff+0.006
Neg logit contributions
aling-0.020
alling-0.018
astics-0.018
 sings-0.018
irl-0.018
Token do
Token ID466
Feature activation+1.90e-02

Pos logit contributions
 injuries+0.070
guard+0.067
 poisoning+0.066
 extingu+0.060
 department+0.060
Neg logit contributions
aling-0.082
alling-0.072
 sings-0.071
irl-0.070
astics-0.069
Token you
Token ID345
Feature activation+1.30e-02

Pos logit contributions
 that+0.009
 they+0.008
t+0.007
guard+0.007
 injuries+0.006
Neg logit contributions
aling-0.052
alling-0.050
aled-0.046
acked-0.045
wn-0.045
Token want
Token ID765
Feature activation+2.83e-02

Pos logit contributions
guard+0.016
 poisoning+0.016
 last+0.015
t+0.015
 injuries+0.015
Neg logit contributions
aling-0.023
alling-0.021
astics-0.020
irl-0.019
 sings-0.018
Token to
Token ID284
Feature activation+1.40e-02

Pos logit contributions
 unfortunately+0.030
ho+0.030
 injuries+0.029
 poisoning+0.029
comb+0.029
Neg logit contributions
aling-0.051
astics-0.048
 plays-0.047
 sings-0.046
alling-0.046
Token marry
Token ID12479
Feature activation-1.82e-02

Pos logit contributions
guard+0.025
 poisoning+0.025
 injuries+0.025
 unfortunately+0.024
 payoff+0.023
Neg logit contributions
aling-0.018
alling-0.016
astics-0.014
irl-0.014
 sings-0.013
Token faster
Token ID5443
Feature activation-2.59e-03

Pos logit contributions
aling+0.024
alling+0.021
astics+0.020
irl+0.019
 sings+0.019
Neg logit contributions
 poisoning-0.020
guard-0.020
 injuries-0.019
 unfortunately-0.018
 payoff-0.018
Token.
Token ID13
Feature activation-1.81e-02

Pos logit contributions
aling+0.004
alling+0.004
irl+0.004
astics+0.004
 sings+0.004
Neg logit contributions
 poisoning-0.004
guard-0.003
 injuries-0.003
 unfortunately-0.003
 department-0.003
Token
Token ID198
Feature activation+2.53e-02

Pos logit contributions
aling+0.034
alling+0.031
irl+0.030
astics+0.030
 sings+0.028
Neg logit contributions
guard-0.021
 poisoning-0.021
 injuries-0.020
 unfortunately-0.019
 payoff-0.019
Token
Token ID198
Feature activation+1.90e-02

Pos logit contributions
 poisoning+0.031
guard+0.031
 injuries+0.030
 unfortunately+0.028
 payoff+0.027
Neg logit contributions
aling-0.048
alling-0.043
astics-0.041
irl-0.040
 sings-0.038
TokenThe
Token ID464
Feature activation+1.08e-02

Pos logit contributions
guard+0.020
 poisoning+0.020
 injuries+0.020
 unfortunately+0.018
 payoff+0.017
Neg logit contributions
aling-0.039
alling-0.035
astics-0.033
irl-0.033
 sings-0.032
Token end
Token ID886
Feature activation+6.99e-02

Pos logit contributions
guard+0.012
 injuries+0.011
 payoff+0.010
 last+0.010
 poisoning+0.010
Neg logit contributions
aling-0.023
alling-0.021
astics-0.020
irl-0.019
atters-0.019
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
 poisoning+0.076
guard+0.076
 injuries+0.075
 unfortunately+0.068
 payoff+0.066
Neg logit contributions
aling-0.140
alling-0.127
astics-0.121
irl-0.120
 sings-0.116
Token<|endoftext|>
Token ID50256
Feature activation-3.07e-03

Pos logit contributions
aling+0.041
alling+0.037
astics+0.035
irl+0.034
 sings+0.033
Neg logit contributions
 poisoning-0.025
 injuries-0.025
guard-0.025
 unfortunately-0.021
 payoff-0.021
TokenOnce
Token ID7454
Feature activation+3.39e-02

Pos logit contributions
aling+0.007
alling+0.006
astics+0.006
irl+0.006
 sings+0.006
Neg logit contributions
guard-0.003
 poisoning-0.002
 injuries-0.002
 payoff-0.002
t-0.002
Token upon
Token ID2402
Feature activation-1.58e-02

Pos logit contributions
 injuries+0.051
 poisoning+0.051
 extingu+0.048
 payoff+0.047
guard+0.047
Neg logit contributions
aling-0.052
irl-0.047
alling-0.045
ophone-0.045
 sings-0.043
Token a
Token ID257
Feature activation-1.22e-03

Pos logit contributions
aling+0.030
alling+0.027
astics+0.025
irl+0.025
 sings+0.024
Neg logit contributions
guard-0.019
 poisoning-0.019
 injuries-0.019
 unfortunately-0.017
 payoff-0.017
Token he
Token ID339
Feature activation-3.47e-03

Pos logit contributions
guard+0.040
 injuries+0.039
 payoff+0.038
 department+0.037
 extingu+0.036
Neg logit contributions
aling-0.040
irl-0.039
 zoom-0.035
 plays-0.035
astics-0.034
Token was
Token ID373
Feature activation-3.69e-02

Pos logit contributions
aling+0.006
alling+0.006
astics+0.005
irl+0.005
 sings+0.005
Neg logit contributions
guard-0.005
 poisoning-0.005
 injuries-0.004
 unfortunately-0.004
 payoff-0.004
Token playing
Token ID2712
Feature activation-2.74e-02

Pos logit contributions
aling+0.066
alling+0.060
 essay+0.058
ophone+0.057
etition+0.056
Neg logit contributions
 poisoning-0.048
 to-0.047
 that-0.044
 injuries-0.043
 while-0.043
Token with
Token ID351
Feature activation-2.45e-02

Pos logit contributions
aling+0.048
alling+0.043
astics+0.040
irl+0.040
 sings+0.038
Neg logit contributions
guard-0.037
 poisoning-0.037
 injuries-0.036
 unfortunately-0.034
 payoff-0.033
Token his
Token ID465
Feature activation+1.63e-02

Pos logit contributions
aling+0.033
irl+0.029
alling+0.028
astics+0.027
 sings+0.025
Neg logit contributions
guard-0.043
 injuries-0.041
 poisoning-0.040
ho-0.040
 payoff-0.040
Token friend
Token ID1545
Feature activation+5.10e-02

Pos logit contributions
 payoff+0.033
guard+0.033
 injuries+0.033
 poisoning+0.033
 unfortunately+0.032
Neg logit contributions
irl-0.014
 nic-0.012
 zoom-0.011
irled-0.011
astics-0.011
Token,
Token ID11
Feature activation+4.24e-02

Pos logit contributions
guard+0.066
 injuries+0.064
 payoff+0.061
 poisoning+0.057
ho+0.056
Neg logit contributions
aling-0.089
alling-0.080
c-0.080
astics-0.078
ophone-0.076
Token a
Token ID257
Feature activation-1.66e-02

Pos logit contributions
 injuries+0.064
guard+0.062
 poisoning+0.059
 payoff+0.059
ho+0.057
Neg logit contributions
aling-0.064
alling-0.058
astics-0.054
ophone-0.053
irl-0.052
Token little
Token ID1310
Feature activation-5.85e-03

Pos logit contributions
aling+0.025
irl+0.024
astics+0.022
alling+0.022
 sings+0.022
Neg logit contributions
 injuries-0.022
 poisoning-0.022
guard-0.022
 payoff-0.020
ately-0.020
Token bird
Token ID6512
Feature activation-3.20e-02

Pos logit contributions
aling+0.012
alling+0.011
astics+0.010
irl+0.010
 sings+0.010
Neg logit contributions
 injuries-0.006
 poisoning-0.006
guard-0.006
 unfortunately-0.005
 payoff-0.005
Token named
Token ID3706
Feature activation-3.40e-02

Pos logit contributions
aling+0.045
alling+0.038
irl+0.036
astics+0.035
c+0.034
Neg logit contributions
guard-0.055
 poisoning-0.054
 injuries-0.054
 payoff-0.052
 unfortunately-0.050
Token job
Token ID1693
Feature activation+5.80e-02

Pos logit contributions
guard+0.012
 poisoning+0.012
 injuries+0.011
 unfortunately+0.011
t+0.011
Neg logit contributions
aling-0.010
alling-0.009
astics-0.009
irl-0.009
 sings-0.008
Token.
Token ID13
Feature activation-3.83e-02

Pos logit contributions
 poisoning+0.074
guard+0.074
 injuries+0.073
 unfortunately+0.068
ho+0.068
Neg logit contributions
aling-0.103
alling-0.092
astics-0.088
irl-0.086
 sings-0.082
Token I
Token ID314
Feature activation+2.37e-02

Pos logit contributions
aling+0.066
alling+0.059
irl+0.055
astics+0.055
 sings+0.051
Neg logit contributions
guard-0.053
 poisoning-0.052
 injuries-0.051
 unfortunately-0.048
 payoff-0.047
Token don
Token ID836
Feature activation+2.80e-02

Pos logit contributions
 injuries+0.030
 poisoning+0.030
guard+0.030
 unfortunately+0.027
 payoff+0.027
Neg logit contributions
aling-0.043
alling-0.038
irl-0.036
astics-0.036
 sings-0.035
Token't
Token ID470
Feature activation+3.41e-02

Pos logit contributions
guard+0.011
t+0.009
 poisoning+0.008
 injuries+0.007
 unfortunately+0.007
Neg logit contributions
aling-0.076
alling-0.071
astics-0.069
 sings-0.067
irl-0.067
Token know
Token ID760
Feature activation+7.04e-02

Pos logit contributions
 poisoning+0.044
 injuries+0.043
guard+0.043
 last+0.040
 payoff+0.039
Neg logit contributions
aling-0.060
alling-0.056
astics-0.053
irl-0.053
irled-0.051
Token what
Token ID644
Feature activation+6.70e-02

Pos logit contributions
guard+0.079
 poisoning+0.078
 injuries+0.076
 payoff+0.071
 unfortunately+0.068
Neg logit contributions
aling-0.137
alling-0.124
astics-0.118
irl-0.117
 sings-0.114
Token to
Token ID284
Feature activation+3.78e-02

Pos logit contributions
 injuries+0.056
 poisoning+0.055
guard+0.053
 unfortunately+0.049
ho+0.046
Neg logit contributions
aling-0.148
astics-0.134
alling-0.134
 sings-0.132
irl-0.131
Token do
Token ID466
Feature activation+2.11e-02

Pos logit contributions
guard+0.057
 poisoning+0.057
 injuries+0.057
 unfortunately+0.055
 payoff+0.055
Neg logit contributions
aling-0.054
irl-0.049
astics-0.047
alling-0.047
 sings-0.044
Token."
Token ID526
Feature activation-1.90e-02

Pos logit contributions
 injuries+0.032
 poisoning+0.031
guard+0.031
ho+0.031
 extingu+0.029
Neg logit contributions
aling-0.032
alling-0.028
astics-0.028
irl-0.027
 sings-0.025
Token
Token ID198
Feature activation+1.93e-02

Pos logit contributions
aling+0.024
 sings+0.021
irl+0.021
alling+0.020
astics+0.020
Neg logit contributions
 poisoning-0.032
 injuries-0.031
 department-0.030
guard-0.030
 extingu-0.029
TokenBut
Token ID1537
Feature activation+5.33e-02

Pos logit contributions
 injuries+0.042
 hesitation+0.038
 poisoning+0.038
 extingu+0.037
guard+0.037
Neg logit contributions
aling-0.051
astics-0.045
irl-0.045
alling-0.044
learn-0.042
Token Tim
Token ID5045
Feature activation+6.77e-03

Pos logit contributions
guard+0.064
 injuries+0.063
 poisoning+0.062
 payoff+0.057
t+0.055
Neg logit contributions
aling-0.097
alling-0.088
irl-0.086
astics-0.084
 sings-0.081
Tokenmy
Token ID1820
Feature activation+3.81e-02

Pos logit contributions
guard+0.013
 injuries+0.012
 poisoning+0.012
 unfortunately+0.011
 payoff+0.011
Neg logit contributions
aling-0.009
alling-0.007
astics-0.007
irl-0.007
c-0.006
Token,
Token ID11
Feature activation+5.10e-02

Pos logit contributions
 poisoning+0.016
guard+0.014
 injuries+0.012
 unfortunately+0.009
 last+0.009
Neg logit contributions
aling-0.105
alling-0.100
astics-0.092
aled-0.091
irl-0.090
Token it
Token ID340
Feature activation+2.54e-02

Pos logit contributions
guard+0.056
 injuries+0.055
 poisoning+0.055
 payoff+0.049
 unfortunately+0.048
Neg logit contributions
aling-0.100
alling-0.091
irl-0.087
astics-0.087
 sings-0.084
Token's
Token ID338
Feature activation-1.78e-02

Pos logit contributions
guard+0.039
 poisoning+0.038
 last+0.037
t+0.036
 unfortunately+0.036
Neg logit contributions
aling-0.041
alling-0.038
astics-0.033
aled-0.031
 essay-0.031
Token cold
Token ID4692
Feature activation+1.95e-02

Pos logit contributions
aling+0.031
alling+0.028
irl+0.026
astics+0.026
 sings+0.025
Neg logit contributions
guard-0.025
 poisoning-0.024
 injuries-0.024
 payoff-0.022
 unfortunately-0.022
Token outside
Token ID2354
Feature activation-5.03e-03

Pos logit contributions
 injuries+0.032
guard+0.031
 poisoning+0.031
 payoff+0.030
 unfortunately+0.030
Neg logit contributions
aling-0.028
alling-0.024
irl-0.023
astics-0.022
 sings-0.022
Token.
Token ID13
Feature activation-1.96e-02

Pos logit contributions
aling+0.010
alling+0.009
astics+0.008
irl+0.008
c+0.008
Neg logit contributions
 poisoning-0.006
 injuries-0.006
guard-0.005
 payoff-0.005
ho-0.005
Token You
Token ID921
Feature activation+3.27e-04

Pos logit contributions
aling+0.036
alling+0.033
irl+0.033
astics+0.031
irled+0.030
Neg logit contributions
 unfortunately-0.023
 poisoning-0.023
guard-0.022
 injuries-0.022
 Papa-0.022
Token need
Token ID761
Feature activation+3.92e-02

Pos logit contributions
guard+0.000
 poisoning+0.000
 injuries+0.000
 unfortunately+0.000
 payoff+0.000
Neg logit contributions
aling-0.001
alling-0.001
astics-0.001
irl-0.001
 sings-0.000
Token to
Token ID284
Feature activation+1.35e-02

Pos logit contributions
aling+0.000
alling+0.000
astics+0.000
irl+0.000
 sings+0.000
Neg logit contributions
guard-0.000
 poisoning-0.000
 injuries-0.000
 unfortunately-0.000
 payoff-0.000
Token his
Token ID465
Feature activation+1.53e-02

Pos logit contributions
 injuries+0.013
 poisoning+0.012
 last+0.012
 unfortunately+0.011
guard+0.011
Neg logit contributions
aling-0.029
alling-0.027
astics-0.025
irl-0.024
ab-0.024
Token mom
Token ID1995
Feature activation+5.47e-02

Pos logit contributions
guard+0.011
 poisoning+0.011
 injuries+0.011
 unfortunately+0.009
 payoff+0.009
Neg logit contributions
aling-0.037
alling-0.034
astics-0.032
irl-0.032
 sings-0.031
Token and
Token ID290
Feature activation-1.26e-03

Pos logit contributions
guard+0.063
 poisoning+0.063
 injuries+0.062
 unfortunately+0.056
 payoff+0.055
Neg logit contributions
aling-0.106
alling-0.096
astics-0.091
irl-0.091
 sings-0.087
Token ate
Token ID15063
Feature activation-2.27e-02

Pos logit contributions
aling+0.002
alling+0.002
astics+0.002
irl+0.002
 sings+0.002
Neg logit contributions
 injuries-0.002
guard-0.002
 poisoning-0.002
t-0.001
 unfortunately-0.001
Token his
Token ID465
Feature activation+2.12e-02

Pos logit contributions
aling+0.035
alling+0.031
irl+0.030
astics+0.029
 sings+0.027
Neg logit contributions
guard-0.034
 poisoning-0.032
 injuries-0.032
t-0.031
 payoff-0.031
Token vegetables
Token ID13701
Feature activation+1.52e-02

Pos logit contributions
 poisoning+0.028
guard+0.028
 injuries+0.028
 unfortunately+0.026
 payoff+0.025
Neg logit contributions
aling-0.037
alling-0.033
astics-0.031
irl-0.031
 sings-0.029
Token.
Token ID13
Feature activation-1.88e-02

Pos logit contributions
 injuries+0.021
guard+0.021
 payoff+0.021
 department+0.020
 poisoning+0.020
Neg logit contributions
aling-0.024
alling-0.021
irl-0.019
astics-0.019
vel-0.018
Token<|endoftext|>
Token ID50256
Feature activation+6.20e-03

Pos logit contributions
aling+0.042
alling+0.039
irl+0.037
astics+0.037
 sings+0.035
Neg logit contributions
 poisoning-0.016
guard-0.015
 injuries-0.015
 unfortunately-0.014
t-0.014
TokenOnce
Token ID7454
Feature activation+4.68e-02

Pos logit contributions
 injuries+0.008
 poisoning+0.008
guard+0.008
 unfortunately+0.007
 payoff+0.007
Neg logit contributions
aling-0.011
alling-0.010
irl-0.010
astics-0.010
 sings-0.009
Token upon
Token ID2402
Feature activation-1.49e-02

Pos logit contributions
 poisoning+0.074
 injuries+0.074
 extingu+0.072
 payoff+0.072
guard+0.071
Neg logit contributions
aling-0.062
irl-0.059
ophone-0.054
c-0.052
alling-0.052
Token to
Token ID284
Feature activation+3.80e-02

Pos logit contributions
guard+0.035
 poisoning+0.034
 injuries+0.032
 unfortunately+0.029
 last+0.029
Neg logit contributions
aling-0.068
alling-0.062
irl-0.058
astics-0.058
 sings-0.055
Token work
Token ID670
Feature activation+2.15e-02

Pos logit contributions
guard+0.049
 poisoning+0.049
 injuries+0.048
 unfortunately+0.044
t+0.043
Neg logit contributions
aling-0.068
alling-0.061
irl-0.059
astics-0.058
 sings-0.054
Token together
Token ID1978
Feature activation+6.56e-02

Pos logit contributions
guard+0.027
 poisoning+0.027
 injuries+0.026
 payoff+0.024
 unfortunately+0.024
Neg logit contributions
aling-0.040
alling-0.036
astics-0.034
irl-0.033
 sings-0.032
Token.
Token ID13
Feature activation-2.43e-02

Pos logit contributions
guard+0.083
 injuries+0.083
 poisoning+0.083
 unfortunately+0.075
 payoff+0.072
Neg logit contributions
aling-0.117
alling-0.105
astics-0.101
irl-0.098
 sings-0.094
Token They
Token ID1119
Feature activation-2.60e-03

Pos logit contributions
aling+0.036
alling+0.031
irl+0.029
astics+0.029
 sings+0.027
Neg logit contributions
guard-0.039
 poisoning-0.039
 injuries-0.038
 unfortunately-0.037
 payoff-0.036
Token learned
Token ID4499
Feature activation+2.50e-02

Pos logit contributions
aling+0.005
alling+0.004
astics+0.004
irl+0.004
 sings+0.004
Neg logit contributions
guard-0.004
 poisoning-0.004
 injuries-0.003
 unfortunately-0.003
 payoff-0.003
Token that
Token ID326
Feature activation+4.85e-02

Pos logit contributions
 injuries+0.024
 poisoning+0.022
guard+0.021
ho+0.020
 payoff+0.018
Neg logit contributions
aling-0.052
alling-0.048
astics-0.048
irl-0.046
 sings-0.045
Token even
Token ID772
Feature activation+1.18e-02

Pos logit contributions
guard+0.057
 poisoning+0.057
 injuries+0.055
 unfortunately+0.052
 payoff+0.051
Neg logit contributions
aling-0.094
alling-0.086
astics-0.080
irl-0.079
ophone-0.076
Token though
Token ID996
Feature activation+6.97e-02

Pos logit contributions
 injuries+0.011
guard+0.010
 poisoning+0.010
ho+0.009
 unfortunately+0.008
Neg logit contributions
aling-0.026
alling-0.024
astics-0.023
irl-0.022
 sings-0.022
Token they
Token ID484
Feature activation+1.14e-02

Pos logit contributions
 poisoning+0.068
guard+0.066
 injuries+0.066
 payoff+0.057
 extingu+0.057
Neg logit contributions
aling-0.140
irl-0.127
alling-0.127
astics-0.125
 sings-0.123
Token were
Token ID547
Feature activation-2.85e-02

Pos logit contributions
 injuries+0.015
guard+0.015
 poisoning+0.015
 unfortunately+0.014
 extingu+0.013
Neg logit contributions
aling-0.020
alling-0.018
irl-0.017
astics-0.016
 sings-0.016
Token on
Token ID319
Feature activation-1.16e-02

Pos logit contributions
aling+0.081
 sings+0.074
alling+0.074
astics+0.073
irl+0.072
Neg logit contributions
guard-0.051
 poisoning-0.051
 injuries-0.050
 last-0.046
 unfortunately-0.045
Token a
Token ID257
Feature activation-1.55e-02

Pos logit contributions
aling+0.018
alling+0.016
irl+0.015
astics+0.015
aled+0.014
Neg logit contributions
guard-0.018
 poisoning-0.017
 injuries-0.016
ho-0.016
 hesitation-0.015
Token ch
Token ID442
Feature activation-8.16e-02

Pos logit contributions
aling+0.027
alling+0.024
astics+0.023
irl+0.023
 sings+0.022
Neg logit contributions
guard-0.021
 poisoning-0.021
 injuries-0.020
 unfortunately-0.019
 payoff-0.019
Tokenute
Token ID1133
Feature activation-4.03e-02

Pos logit contributions
aling+0.135
 sings+0.128
astics+0.127
alling+0.125
beh+0.119
Neg logit contributions
guard-0.104
 injuries-0.101
 last-0.096
 poisoning-0.095
t-0.094
Token.
Token ID13
Feature activation-2.45e-02

Pos logit contributions
aling+0.086
alling+0.079
astics+0.076
irl+0.075
 sings+0.073
Neg logit contributions
guard-0.038
 poisoning-0.038
 injuries-0.037
 unfortunately-0.033
 last-0.032
Token He
Token ID679
Feature activation-9.83e-03

Pos logit contributions
aling+0.057
alling+0.053
irl+0.051
astics+0.051
 sings+0.049
Neg logit contributions
 poisoning-0.017
t-0.017
guard-0.017
 unfortunately-0.016
Not-0.016
Token had
Token ID550
Feature activation-2.10e-02

Pos logit contributions
aling+0.020
alling+0.019
astics+0.018
vel+0.017
beh+0.017
Neg logit contributions
guard-0.010
 last-0.009
 they-0.009
 that-0.009
 now-0.009
Token to
Token ID284
Feature activation-9.23e-03

Pos logit contributions
aling+0.044
alling+0.040
irl+0.039
astics+0.038
 sings+0.037
Neg logit contributions
 poisoning-0.020
 injuries-0.019
guard-0.019
 payoff-0.018
 last-0.018
Token go
Token ID467
Feature activation-2.94e-02

Pos logit contributions
aling+0.016
alling+0.014
astics+0.013
ophone+0.013
 sings+0.013
Neg logit contributions
 injuries-0.012
guard-0.012
 poisoning-0.012
 last-0.012
 payoff-0.011
Token down
Token ID866
Feature activation-1.44e-02

Pos logit contributions
aling+0.059
alling+0.054
astics+0.053
irl+0.052
 sings+0.051
Neg logit contributions
guard-0.032
 poisoning-0.030
 injuries-0.029
 last-0.027
 unfortunately-0.026
Token the
Token ID262
Feature activation-1.14e-02

Pos logit contributions
aling+0.027
alling+0.024
astics+0.022
irl+0.022
 ped+0.021
Neg logit contributions
 injuries-0.018
 poisoning-0.017
guard-0.017
ho-0.016
 unfortunately-0.016
Token special
Token ID2041
Feature activation+2.17e-02

Pos logit contributions
 injuries+0.036
guard+0.036
 poisoning+0.036
 payoff+0.034
 unfortunately+0.033
Neg logit contributions
aling-0.038
alling-0.033
irl-0.031
astics-0.031
 sings-0.029
Token -
Token ID532
Feature activation+1.13e-02

Pos logit contributions
 injuries+0.032
guard+0.030
 poisoning+0.029
 payoff+0.029
 unfortunately+0.028
Neg logit contributions
aling-0.034
alling-0.030
astics-0.029
irl-0.028
 sings-0.028
Token a
Token ID257
Feature activation-2.02e-02

Pos logit contributions
 poisoning+0.011
guard+0.011
 injuries+0.010
 unfortunately+0.010
 last+0.010
Neg logit contributions
aling-0.024
alling-0.022
astics-0.021
irl-0.021
aled-0.020
Token spring
Token ID6076
Feature activation-2.88e-02

Pos logit contributions
aling+0.036
alling+0.032
astics+0.030
irl+0.030
 sings+0.028
Neg logit contributions
guard-0.027
 poisoning-0.027
 injuries-0.026
 unfortunately-0.024
 payoff-0.024
Token!
Token ID0
Feature activation-2.15e-02

Pos logit contributions
aling+0.051
alling+0.046
irl+0.044
astics+0.043
 sings+0.042
Neg logit contributions
 injuries-0.037
 poisoning-0.036
guard-0.036
 unfortunately-0.033
 payoff-0.032
Token Inside
Token ID14384
Feature activation+8.51e-03

Pos logit contributions
aling+0.035
alling+0.031
irl+0.030
astics+0.029
 sings+0.027
Neg logit contributions
guard-0.032
 poisoning-0.031
 injuries-0.031
 unfortunately-0.029
 payoff-0.029
Token were
Token ID547
Feature activation-3.01e-02

Pos logit contributions
 injuries+0.013
 poisoning+0.012
 mercy+0.011
 extingu+0.011
 payoff+0.011
Neg logit contributions
aling-0.014
irl-0.013
alling-0.012
c-0.012
 sings-0.011
Token lots
Token ID6041
Feature activation+2.57e-03

Pos logit contributions
aling+0.045
alling+0.039
astics+0.037
irl+0.036
 sings+0.035
Neg logit contributions
 poisoning-0.048
guard-0.048
 injuries-0.048
 unfortunately-0.043
 payoff-0.043
Token of
Token ID286
Feature activation-1.62e-03

Pos logit contributions
 injuries+0.002
 poisoning+0.002
guard+0.002
 unfortunately+0.002
 payoff+0.002
Neg logit contributions
aling-0.006
alling-0.005
irl-0.005
astics-0.005
 sings-0.005
Token colourful
Token ID44799
Feature activation-2.34e-02

Pos logit contributions
aling+0.003
alling+0.003
astics+0.002
ophone+0.002
irl+0.002
Neg logit contributions
guard-0.002
 poisoning-0.002
 injuries-0.002
 last-0.002
 unfortunately-0.002
Token coils
Token ID41331
Feature activation-1.42e-02

Pos logit contributions
aling+0.038
alling+0.033
astics+0.031
irl+0.031
 sings+0.029
Neg logit contributions
guard-0.035
 poisoning-0.035
 injuries-0.034
 unfortunately-0.032
 payoff-0.031
Token to
Token ID284
Feature activation+9.22e-03

Pos logit contributions
guard+0.002
 injuries+0.002
 poisoning+0.002
 unfortunately+0.001
 payoff+0.001
Neg logit contributions
aling-0.006
alling-0.006
astics-0.006
irl-0.006
ophone-0.006
Token him
Token ID683
Feature activation+3.94e-03

Pos logit contributions
guard+0.011
 poisoning+0.010
 injuries+0.010
 payoff+0.009
 unfortunately+0.009
Neg logit contributions
aling-0.018
alling-0.016
irl-0.015
astics-0.015
 sings-0.014
Token.
Token ID13
Feature activation-2.99e-02

Pos logit contributions
 poisoning+0.003
guard+0.003
 injuries+0.003
 unfortunately+0.003
 last+0.003
Neg logit contributions
aling-0.009
alling-0.008
irl-0.008
astics-0.008
 sings-0.008
Token Tim
Token ID5045
Feature activation-2.24e-02

Pos logit contributions
aling+0.057
alling+0.052
irl+0.050
astics+0.049
 sings+0.047
Neg logit contributions
 poisoning-0.035
guard-0.035
 injuries-0.034
 unfortunately-0.032
 payoff-0.031
Token was
Token ID373
Feature activation-4.45e-02

Pos logit contributions
aling+0.043
alling+0.039
astics+0.037
irl+0.037
 sings+0.035
Neg logit contributions
guard-0.026
 poisoning-0.026
 injuries-0.025
 unfortunately-0.024
 payoff-0.023
Token so
Token ID523
Feature activation-2.02e-02

Pos logit contributions
aling+0.082
alling+0.074
astics+0.070
irl+0.070
 sings+0.067
Neg logit contributions
 poisoning-0.055
guard-0.055
 injuries-0.054
 unfortunately-0.051
 payoff-0.049
Token happy
Token ID3772
Feature activation+2.96e-02

Pos logit contributions
aling+0.028
alling+0.024
irl+0.022
astics+0.022
 sings+0.021
Neg logit contributions
 injuries-0.035
 poisoning-0.035
guard-0.035
 payoff-0.032
 unfortunately-0.032
Token and
Token ID290
Feature activation-3.22e-02

Pos logit contributions
 injuries+0.036
guard+0.034
 poisoning+0.033
 payoff+0.031
ho+0.031
Neg logit contributions
aling-0.053
alling-0.046
 ped-0.045
astics-0.045
 sings-0.043
Token thanked
Token ID26280
Feature activation+5.37e-02

Pos logit contributions
aling+0.068
alling+0.063
irl+0.060
astics+0.060
ophone+0.057
Neg logit contributions
guard-0.032
 poisoning-0.032
 unfortunately-0.029
 injuries-0.029
 payoff-0.027
Token the
Token ID262
Feature activation+1.80e-02

Pos logit contributions
 poisoning+0.079
guard+0.078
t+0.075
 injuries+0.075
less+0.073
Neg logit contributions
aling-0.082
alling-0.072
irl-0.068
 sings-0.065
astics-0.064
Token lady
Token ID10846
Feature activation+1.10e-02

Pos logit contributions
guard+0.009
 poisoning+0.009
 injuries+0.008
 payoff+0.007
 last+0.006
Neg logit contributions
aling-0.048
alling-0.045
astics-0.043
irl-0.043
ophone-0.041
Token was
Token ID373
Feature activation-3.30e-02

Pos logit contributions
guard+0.015
 injuries+0.015
 poisoning+0.015
 unfortunately+0.013
 payoff+0.013
Neg logit contributions
aling-0.020
alling-0.018
astics-0.017
irl-0.017
 sings-0.016
Token still
Token ID991
Feature activation-6.19e-03

Pos logit contributions
aling+0.057
alling+0.051
astics+0.048
irl+0.048
 sings+0.046
Neg logit contributions
guard-0.045
 poisoning-0.045
 injuries-0.044
 unfortunately-0.041
 payoff-0.040
Token by
Token ID416
Feature activation-3.70e-03

Pos logit contributions
aling+0.010
alling+0.009
astics+0.008
irl+0.008
 sings+0.008
Neg logit contributions
guard-0.009
 poisoning-0.009
 injuries-0.009
 unfortunately-0.008
 payoff-0.008
Token her
Token ID607
Feature activation+4.03e-03

Pos logit contributions
aling+0.006
alling+0.006
astics+0.005
irl+0.005
 sings+0.005
Neg logit contributions
guard-0.005
 poisoning-0.005
 injuries-0.005
t-0.004
 department-0.004
Token side
Token ID1735
Feature activation+2.41e-02

Pos logit contributions
guard+0.006
 poisoning+0.006
 injuries+0.005
 unfortunately+0.005
t+0.005
Neg logit contributions
aling-0.007
alling-0.006
astics-0.006
irl-0.005
 sings-0.005
Token.
Token ID13
Feature activation-3.03e-02

Pos logit contributions
guard+0.027
 poisoning+0.026
 injuries+0.025
 payoff+0.024
 unfortunately+0.023
Neg logit contributions
aling-0.047
alling-0.043
astics-0.039
c-0.039
irl-0.039
Token Mom
Token ID11254
Feature activation+1.83e-02

Pos logit contributions
aling+0.056
alling+0.051
irl+0.049
astics+0.048
 sings+0.045
Neg logit contributions
 poisoning-0.037
guard-0.037
 injuries-0.036
 unfortunately-0.035
 payoff-0.033
Tokenmy
Token ID1820
Feature activation+7.30e-03

Pos logit contributions
guard+0.025
 poisoning+0.024
 injuries+0.023
 unfortunately+0.022
 last+0.022
Neg logit contributions
aling-0.032
alling-0.028
astics-0.027
irl-0.027
 sings-0.025
Token said
Token ID531
Feature activation+2.65e-02

Pos logit contributions
guard+0.009
 poisoning+0.009
 injuries+0.009
 unfortunately+0.008
 payoff+0.008
Neg logit contributions
aling-0.013
alling-0.012
astics-0.011
irl-0.011
 sings-0.011
Token,
Token ID11
Feature activation+4.09e-02

Pos logit contributions
 injuries+0.030
guard+0.028
 poisoning+0.028
 payoff+0.024
 extingu+0.024
Neg logit contributions
aling-0.048
alling-0.044
irl-0.043
vel-0.043
 sings-0.043
Token
Token ID6184
Feature activation-5.94e-02

Pos logit contributions
guard+0.051
 poisoning+0.051
 injuries+0.050
 unfortunately+0.046
 payoff+0.045
Neg logit contributions
aling-0.075
alling-0.068
astics-0.064
irl-0.063
 sings-0.061
Token a
Token ID257
Feature activation-4.50e-03

Pos logit contributions
aling+0.018
alling+0.016
astics+0.015
irl+0.015
 sings+0.014
Neg logit contributions
guard-0.011
 poisoning-0.011
 injuries-0.011
 unfortunately-0.010
 payoff-0.010
Token box
Token ID3091
Feature activation-2.18e-02

Pos logit contributions
aling+0.008
alling+0.007
astics+0.007
irl+0.007
 sings+0.007
Neg logit contributions
 poisoning-0.006
guard-0.006
 injuries-0.006
 unfortunately-0.005
 payoff-0.005
Token.
Token ID13
Feature activation-1.12e-02

Pos logit contributions
aling+0.045
alling+0.041
astics+0.039
irl+0.038
 sings+0.037
Neg logit contributions
guard-0.023
 poisoning-0.022
 injuries-0.022
 unfortunately-0.020
 payoff-0.019
Token The
Token ID383
Feature activation+2.69e-03

Pos logit contributions
aling+0.021
alling+0.019
irl+0.019
astics+0.019
irled+0.017
Neg logit contributions
guard-0.013
 poisoning-0.013
 unfortunately-0.012
 injuries-0.012
 payoff-0.011
Token man
Token ID582
Feature activation-2.57e-02

Pos logit contributions
guard+0.003
 poisoning+0.003
 injuries+0.003
 payoff+0.002
 unfortunately+0.002
Neg logit contributions
aling-0.006
alling-0.005
astics-0.005
irl-0.005
 sings-0.005
Token looked
Token ID3114
Feature activation-4.44e-02

Pos logit contributions
aling+0.053
alling+0.048
astics+0.047
irl+0.046
etition+0.043
Neg logit contributions
guard-0.027
 poisoning-0.026
 injuries-0.024
 unfortunately-0.024
 last-0.023
Token up
Token ID510
Feature activation-1.63e-02

Pos logit contributions
aling+0.072
alling+0.064
astics+0.063
irl+0.059
 sings+0.055
Neg logit contributions
 poisoning-0.066
guard-0.064
 injuries-0.063
 unfortunately-0.061
 though-0.060
Token at
Token ID379
Feature activation-1.47e-02

Pos logit contributions
aling+0.020
alling+0.017
irl+0.015
 ped+0.014
irled+0.014
Neg logit contributions
guard-0.030
 injuries-0.029
 poisoning-0.029
 payoff-0.027
 mercy-0.027
Token the
Token ID262
Feature activation+4.31e-03

Pos logit contributions
aling+0.016
 sings+0.015
c+0.014
alling+0.014
aled+0.014
Neg logit contributions
guard-0.025
 poisoning-0.025
t-0.025
 injuries-0.024
 unfortunately-0.024
Token girl
Token ID2576
Feature activation-6.67e-03

Pos logit contributions
guard+0.005
 poisoning+0.005
 injuries+0.005
 unfortunately+0.004
 payoff+0.004
Neg logit contributions
aling-0.009
alling-0.008
astics-0.008
irl-0.008
 sings-0.007
Token,
Token ID11
Feature activation+1.79e-02

Pos logit contributions
aling+0.016
irl+0.015
alling+0.015
astics+0.015
irled+0.014
Neg logit contributions
 poisoning-0.004
guard-0.004
 injuries-0.004
 unfortunately-0.004
 last-0.003
Token do
Token ID466
Feature activation+2.31e-02

Pos logit contributions
guard+0.026
 poisoning+0.025
 injuries+0.024
 unfortunately+0.022
 last+0.021
Neg logit contributions
aling-0.055
alling-0.051
astics-0.049
irl-0.048
irled-0.046
Tokenggy
Token ID19970
Feature activation-3.18e-02

Pos logit contributions
guard+0.024
 poisoning+0.022
 that+0.022
 injuries+0.022
 unfortunately+0.022
Neg logit contributions
aling-0.049
alling-0.046
aled-0.041
atters-0.040
ophone-0.039
Token!"
Token ID2474
Feature activation-2.73e-03

Pos logit contributions
aling+0.062
alling+0.057
astics+0.054
irl+0.053
 sings+0.051
Neg logit contributions
guard-0.036
 poisoning-0.036
 injuries-0.035
 unfortunately-0.032
 payoff-0.031
Token and
Token ID290
Feature activation-3.14e-02

Pos logit contributions
aling+0.005
alling+0.004
astics+0.004
irl+0.004
 sings+0.004
Neg logit contributions
guard-0.004
 poisoning-0.004
 injuries-0.004
 unfortunately-0.004
 payoff-0.004
Token he
Token ID339
Feature activation-9.85e-03

Pos logit contributions
aling+0.065
alling+0.060
astics+0.057
irl+0.056
 sings+0.054
Neg logit contributions
guard-0.032
 poisoning-0.032
 injuries-0.031
 unfortunately-0.028
 payoff-0.027
Token ran
Token ID4966
Feature activation-4.63e-02

Pos logit contributions
aling+0.017
alling+0.016
astics+0.015
irl+0.014
atters+0.014
Neg logit contributions
guard-0.014
 poisoning-0.013
 injuries-0.012
 unfortunately-0.012
 last-0.012
Token away
Token ID1497
Feature activation+2.08e-02

Pos logit contributions
aling+0.091
alling+0.082
astics+0.078
irl+0.078
 sings+0.075
Neg logit contributions
guard-0.053
 poisoning-0.053
 injuries-0.051
 unfortunately-0.046
 payoff-0.046
Token.
Token ID13
Feature activation-2.14e-02

Pos logit contributions
 poisoning+0.026
 injuries+0.026
guard+0.025
 unfortunately+0.023
 payoff+0.023
Neg logit contributions
aling-0.036
alling-0.032
astics-0.030
irl-0.030
 ped-0.029
Token
Token ID220
Feature activation-6.71e-03

Pos logit contributions
aling+0.042
alling+0.038
astics+0.036
irl+0.036
 sings+0.034
Neg logit contributions
guard-0.024
 poisoning-0.024
 injuries-0.024
 unfortunately-0.022
 payoff-0.021
Token
Token ID198
Feature activation+2.63e-02

Pos logit contributions
aling+0.015
alling+0.014
astics+0.013
irl+0.013
 sings+0.013
Neg logit contributions
guard-0.005
 poisoning-0.005
 injuries-0.005
 unfortunately-0.005
t-0.004
Token
Token ID198
Feature activation+2.19e-02

Pos logit contributions
 injuries+0.034
 poisoning+0.034
guard+0.033
 payoff+0.030
 unfortunately+0.029
Neg logit contributions
aling-0.048
alling-0.043
irl-0.041
astics-0.041
 sings-0.039
Token had
Token ID550
Feature activation-3.74e-03

Pos logit contributions
 injuries+0.014
 poisoning+0.014
guard+0.014
 unfortunately+0.014
 payoff+0.014
Neg logit contributions
aling-0.018
alling-0.016
astics-0.015
 sings-0.015
irl-0.015
Token to
Token ID284
Feature activation+4.30e-03

Pos logit contributions
aling+0.007
alling+0.007
astics+0.006
irl+0.006
 sings+0.006
Neg logit contributions
guard-0.004
 poisoning-0.004
 injuries-0.004
 unfortunately-0.004
 payoff-0.004
Token come
Token ID1282
Feature activation-1.77e-02

Pos logit contributions
guard+0.006
 poisoning+0.006
 injuries+0.006
 unfortunately+0.006
 payoff+0.005
Neg logit contributions
aling-0.007
alling-0.006
astics-0.006
irl-0.006
 sings-0.006
Token with
Token ID351
Feature activation-8.16e-03

Pos logit contributions
aling+0.031
alling+0.029
astics+0.026
irl+0.026
ophone+0.025
Neg logit contributions
guard-0.023
 poisoning-0.023
 injuries-0.022
 payoff-0.022
 unfortunately-0.022
Token her
Token ID607
Feature activation+3.95e-02

Pos logit contributions
aling+0.014
alling+0.013
irl+0.012
astics+0.012
 sings+0.011
Neg logit contributions
guard-0.011
 poisoning-0.010
 unfortunately-0.010
 department-0.010
ho-0.010
Token.
Token ID13
Feature activation-2.65e-02

Pos logit contributions
guard+0.060
 payoff+0.058
 unfortunately+0.056
 poisoning+0.056
Not+0.056
Neg logit contributions
aling-0.053
ophone-0.048
alling-0.047
 sings-0.046
c-0.046
Token She
Token ID1375
Feature activation-6.68e-04

Pos logit contributions
aling+0.042
alling+0.038
astics+0.036
irl+0.035
 sings+0.034
Neg logit contributions
guard-0.041
 poisoning-0.039
 injuries-0.039
 payoff-0.036
 unfortunately-0.035
Token said
Token ID531
Feature activation+3.47e-02

Pos logit contributions
aling+0.001
alling+0.001
astics+0.001
irl+0.001
 sings+0.001
Neg logit contributions
guard-0.001
 poisoning-0.001
 injuries-0.001
 unfortunately-0.001
 payoff-0.001
Token he
Token ID339
Feature activation+3.33e-02

Pos logit contributions
 poisoning+0.037
guard+0.037
 injuries+0.036
 extingu+0.034
 payoff+0.033
Neg logit contributions
aling-0.064
 sings-0.059
irl-0.058
alling-0.057
astics-0.057
Token would
Token ID561
Feature activation+1.65e-02

Pos logit contributions
guard+0.035
 poisoning+0.035
 injuries+0.034
 unfortunately+0.030
t+0.030
Neg logit contributions
aling-0.069
alling-0.063
astics-0.059
irl-0.059
 sings-0.056
Token be
Token ID307
Feature activation-4.68e-02

Pos logit contributions
guard+0.023
 poisoning+0.022
 injuries+0.021
 unfortunately+0.020
 department+0.020
Neg logit contributions
aling-0.028
alling-0.025
irl-0.024
astics-0.023
 sings-0.022
Token<|endoftext|>
Token ID50256
Feature activation-4.17e-03

Pos logit contributions
aling+0.056
alling+0.052
astics+0.050
irl+0.049
 sings+0.047
Neg logit contributions
 poisoning-0.020
guard-0.020
 injuries-0.020
 unfortunately-0.018
 payoff-0.017
TokenOne
Token ID3198
Feature activation+1.79e-02

Pos logit contributions
aling+0.008
alling+0.008
irl+0.007
astics+0.007
 sings+0.007
Neg logit contributions
 poisoning-0.004
 injuries-0.004
guard-0.004
 unfortunately-0.004
 payoff-0.004
Token day
Token ID1110
Feature activation+4.63e-02

Pos logit contributions
guard+0.020
 poisoning+0.019
 injuries+0.019
 unfortunately+0.017
 payoff+0.017
Neg logit contributions
aling-0.036
alling-0.033
astics-0.031
irl-0.031
 sings-0.030
Token,
Token ID11
Feature activation+2.27e-02

Pos logit contributions
 injuries+0.054
guard+0.051
 poisoning+0.050
 payoff+0.047
 department+0.045
Neg logit contributions
aling-0.086
irl-0.085
irled-0.080
astics-0.078
alling-0.076
Token a
Token ID257
Feature activation-2.15e-02

Pos logit contributions
 injuries+0.036
guard+0.035
 poisoning+0.034
 payoff+0.032
 department+0.031
Neg logit contributions
aling-0.034
alling-0.030
irl-0.030
astics-0.029
irled-0.028
Token sw
Token ID1509
Feature activation-1.28e-01

Pos logit contributions
aling+0.051
alling+0.048
astics+0.045
irl+0.043
aled+0.043
Neg logit contributions
guard-0.017
 poisoning-0.017
 injuries-0.016
 last-0.014
 unfortunately-0.013
Tokenan
Token ID272
Feature activation-6.47e-02

Pos logit contributions
aling+0.229
 sings+0.210
alling+0.204
astics+0.200
vel+0.188
Neg logit contributions
guard-0.172
ho-0.165
 poisoning-0.153
 injuries-0.145
t-0.142
Token named
Token ID3706
Feature activation-4.04e-02

Pos logit contributions
aling+0.120
alling+0.109
astics+0.103
irl+0.101
 sings+0.095
Neg logit contributions
guard-0.081
 poisoning-0.079
 injuries-0.077
 unfortunately-0.071
ho-0.070
Token Sam
Token ID3409
Feature activation-1.33e-02

Pos logit contributions
aling+0.098
alling+0.091
astics+0.086
irl+0.083
 essay+0.080
Neg logit contributions
 poisoning-0.028
 that-0.027
ho-0.027
 last-0.026
guard-0.025
Token was
Token ID373
Feature activation-3.74e-02

Pos logit contributions
aling+0.024
alling+0.022
astics+0.020
irl+0.020
 sings+0.019
Neg logit contributions
guard-0.017
 poisoning-0.017
 injuries-0.017
 unfortunately-0.015
 payoff-0.015
Token swimming
Token ID14899
Feature activation-7.07e-02

Pos logit contributions
aling+0.079
alling+0.072
astics+0.069
etition+0.069
 essay+0.068
Neg logit contributions
 that-0.037
 to-0.037
 poisoning-0.034
 help-0.033
 last-0.033
Token play
Token ID711
Feature activation-3.17e-02

Pos logit contributions
aling+0.024
alling+0.022
astics+0.019
ophone+0.019
irled+0.018
Neg logit contributions
 help-0.016
 poisoning-0.016
guard-0.016
 injuries-0.016
 last-0.015
Token with
Token ID351
Feature activation-1.09e-02

Pos logit contributions
aling+0.057
alling+0.051
astics+0.046
ophone+0.044
 sings+0.044
Neg logit contributions
 poisoning-0.044
guard-0.042
 injuries-0.041
 unfortunately-0.039
 while-0.037
Token her
Token ID607
Feature activation+1.79e-02

Pos logit contributions
aling+0.022
alling+0.020
astics+0.019
irl+0.019
 sings+0.018
Neg logit contributions
 poisoning-0.012
guard-0.012
 injuries-0.011
 unfortunately-0.010
 payoff-0.010
Token toys
Token ID14958
Feature activation+1.13e-02

Pos logit contributions
guard+0.035
 payoff+0.034
 injuries+0.034
 poisoning+0.034
 unfortunately+0.033
Neg logit contributions
irl-0.016
 sings-0.016
 nic-0.015
astics-0.014
irled-0.014
Token and
Token ID290
Feature activation-1.98e-02

Pos logit contributions
 injuries+0.019
 payoff+0.018
guard+0.018
 mercy+0.018
 hesitation+0.017
Neg logit contributions
aling-0.015
alling-0.013
irl-0.012
astics-0.012
 rides-0.012
Token c
Token ID269
Feature activation-8.42e-02

Pos logit contributions
aling+0.028
alling+0.024
irl+0.023
astics+0.023
 sings+0.023
Neg logit contributions
guard-0.033
 poisoning-0.032
 injuries-0.032
 unfortunately-0.030
 payoff-0.030
Tokenuddle
Token ID24500
Feature activation-1.56e-02

Pos logit contributions
aling+0.168
alling+0.151
astics+0.133
 sings+0.132
ophone+0.129
Neg logit contributions
 poisoning-0.102
 injuries-0.099
guard-0.094
ho-0.087
t-0.086
Token with
Token ID351
Feature activation-1.54e-02

Pos logit contributions
aling+0.026
alling+0.023
irl+0.022
astics+0.022
 sings+0.020
Neg logit contributions
guard-0.022
 injuries-0.022
 poisoning-0.022
 payoff-0.021
 unfortunately-0.020
Token her
Token ID607
Feature activation+1.61e-02

Pos logit contributions
aling+0.027
alling+0.025
irl+0.024
astics+0.024
 sings+0.023
Neg logit contributions
guard-0.020
 poisoning-0.020
 injuries-0.019
 payoff-0.018
 unfortunately-0.018
Token t
Token ID256
Feature activation-2.22e-02

Pos logit contributions
guard+0.032
 payoff+0.032
 unfortunately+0.030
 poisoning+0.030
 injuries+0.030
Neg logit contributions
irl-0.014
 nic-0.014
 sings-0.013
c-0.013
 zoom-0.012
Tokeneddy
Token ID21874
Feature activation-1.02e-02

Pos logit contributions
aling+0.034
alling+0.030
irl+0.029
astics+0.028
ophone+0.027
Neg logit contributions
 poisoning-0.033
 injuries-0.033
 payoff-0.032
guard-0.032
 unfortunately-0.031
Token messy
Token ID24097
Feature activation+1.43e-02

Pos logit contributions
aling+0.089
alling+0.083
astics+0.076
irl+0.073
irled+0.070
Neg logit contributions
guard-0.062
 poisoning-0.062
 injuries-0.062
 unfortunately-0.057
 last-0.057
Token.
Token ID13
Feature activation-3.61e-02

Pos logit contributions
 injuries+0.018
 poisoning+0.018
guard+0.018
 unfortunately+0.017
 payoff+0.017
Neg logit contributions
aling-0.025
alling-0.022
astics-0.022
irl-0.022
 sings-0.021
Token Like
Token ID4525
Feature activation+6.77e-03

Pos logit contributions
aling+0.058
alling+0.052
irl+0.049
astics+0.047
irled+0.044
Neg logit contributions
 poisoning-0.051
guard-0.050
 injuries-0.050
 unfortunately-0.050
ful-0.048
Token when
Token ID618
Feature activation+2.48e-02

Pos logit contributions
guard+0.007
 poisoning+0.007
 injuries+0.007
 unfortunately+0.006
 payoff+0.006
Neg logit contributions
aling-0.014
alling-0.012
astics-0.012
irl-0.012
 sings-0.011
Token I
Token ID314
Feature activation+1.68e-02

Pos logit contributions
 poisoning+0.029
guard+0.029
 injuries+0.028
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.047
alling-0.042
irl-0.041
astics-0.041
 sings-0.039
Token sne
Token ID10505
Feature activation-1.04e-01

Pos logit contributions
guard+0.021
 poisoning+0.021
 injuries+0.021
 unfortunately+0.019
 payoff+0.019
Neg logit contributions
aling-0.031
alling-0.028
irl-0.026
astics-0.026
 sings-0.025
Tokene
Token ID68
Feature activation-5.90e-02

Pos logit contributions
aling+0.122
alling+0.112
etition+0.097
astics+0.094
 nic+0.092
Neg logit contributions
 poisoning-0.179
 injuries-0.179
ful-0.176
guard-0.169
 help-0.166
Tokenze
Token ID2736
Feature activation-1.43e-02

Pos logit contributions
aling+0.090
alling+0.080
astics+0.073
irl+0.070
 sings+0.065
Neg logit contributions
 poisoning-0.092
guard-0.091
 injuries-0.090
 unfortunately-0.084
ho-0.083
Token.
Token ID13
Feature activation-3.96e-02

Pos logit contributions
aling+0.027
alling+0.024
astics+0.023
irl+0.023
 sings+0.022
Neg logit contributions
 injuries-0.017
guard-0.017
 poisoning-0.017
 payoff-0.015
 unfortunately-0.015
Token Ach
Token ID26219
Feature activation-1.64e-02

Pos logit contributions
aling+0.064
alling+0.057
irl+0.054
astics+0.053
irled+0.050
Neg logit contributions
 poisoning-0.056
 injuries-0.055
guard-0.054
 unfortunately-0.053
Not-0.051
Tokenoo
Token ID2238
Feature activation-3.65e-02

Pos logit contributions
aling+0.027
alling+0.025
astics+0.023
irl+0.022
 sings+0.021
Neg logit contributions
 injuries-0.023
 poisoning-0.023
guard-0.022
 unfortunately-0.021
t-0.020
Token"
Token ID1
Feature activation+2.48e-02

Pos logit contributions
guard+0.020
 poisoning+0.020
 injuries+0.020
 unfortunately+0.017
 payoff+0.017
Neg logit contributions
aling-0.041
alling-0.037
astics-0.035
irl-0.035
 sings-0.033
TokenR
Token ID49
Feature activation-2.36e-02

Pos logit contributions
guard+0.031
 poisoning+0.030
 injuries+0.029
 payoff+0.027
 unfortunately+0.027
Neg logit contributions
aling-0.047
alling-0.042
astics-0.040
irl-0.040
 sings-0.038
Tokenib
Token ID571
Feature activation-3.53e-02

Pos logit contributions
aling+0.040
alling+0.036
irl+0.031
astics+0.031
ophone+0.031
Neg logit contributions
 injuries-0.034
 poisoning-0.034
guard-0.034
 payoff-0.030
ho-0.030
Tokenbit
Token ID2545
Feature activation+2.40e-02

Pos logit contributions
aling+0.056
alling+0.050
astics+0.046
irl+0.044
ophone+0.041
Neg logit contributions
 injuries-0.053
 poisoning-0.052
guard-0.051
ho-0.049
 last-0.048
Token,
Token ID11
Feature activation+2.11e-02

Pos logit contributions
 poisoning+0.024
 injuries+0.024
guard+0.024
 unfortunately+0.021
 payoff+0.020
Neg logit contributions
aling-0.050
alling-0.045
irl-0.044
astics-0.043
 sings-0.043
Token rib
Token ID12183
Feature activation-8.63e-02

Pos logit contributions
guard+0.034
 poisoning+0.033
 injuries+0.032
 unfortunately+0.031
 payoff+0.030
Neg logit contributions
aling-0.032
alling-0.028
irl-0.026
astics-0.026
 sings-0.025
Tokenbit
Token ID2545
Feature activation+2.35e-02

Pos logit contributions
aling+0.139
alling+0.127
ophone+0.115
astics+0.114
irl+0.114
Neg logit contributions
 injuries-0.122
 poisoning-0.121
 payoff-0.113
guard-0.112
 last-0.111
Token!"
Token ID2474
Feature activation+1.79e-02

Pos logit contributions
 poisoning+0.028
 injuries+0.027
guard+0.027
 unfortunately+0.025
 payoff+0.024
Neg logit contributions
aling-0.045
alling-0.040
irl-0.039
astics-0.038
 sings-0.038
Token Lily
Token ID20037
Feature activation-7.66e-03

Pos logit contributions
 poisoning+0.024
 injuries+0.024
guard+0.023
 department+0.021
ho+0.021
Neg logit contributions
aling-0.032
alling-0.028
 sings-0.027
irl-0.027
astics-0.026
Token says
Token ID1139
Feature activation+1.27e-02

Pos logit contributions
aling+0.013
alling+0.012
irl+0.011
astics+0.011
vel+0.011
Neg logit contributions
guard-0.010
 last-0.009
 poisoning-0.009
 payoff-0.009
 but-0.009
Token.
Token ID13
Feature activation-3.42e-02

Pos logit contributions
guard+0.011
 injuries+0.011
 poisoning+0.011
 unfortunately+0.010
 payoff+0.009
Neg logit contributions
aling-0.028
alling-0.026
astics-0.025
irl-0.025
 sings-0.023
Token parent
Token ID2560
Feature activation+2.75e-02

Pos logit contributions
 injuries+0.000
 poisoning+0.000
guard+0.000
 unfortunately+0.000
 last+0.000
Neg logit contributions
aling-0.001
alling-0.001
astics-0.001
irl-0.001
 sings-0.001
Token smiled
Token ID13541
Feature activation-1.42e-02

Pos logit contributions
guard+0.030
 poisoning+0.029
 injuries+0.029
 unfortunately+0.026
 payoff+0.026
Neg logit contributions
aling-0.056
alling-0.051
astics-0.048
irl-0.048
 sings-0.046
Token and
Token ID290
Feature activation-3.56e-02

Pos logit contributions
aling+0.028
alling+0.025
astics+0.023
irl+0.023
 sings+0.023
Neg logit contributions
guard-0.016
 injuries-0.016
 poisoning-0.016
 payoff-0.014
 unfortunately-0.014
Token said
Token ID531
Feature activation+3.59e-02

Pos logit contributions
aling+0.087
alling+0.080
astics+0.076
irl+0.074
atters+0.073
Neg logit contributions
guard-0.025
 poisoning-0.023
 unfortunately-0.022
 payoff-0.020
 they-0.020
Token
Token ID6184
Feature activation-4.62e-02

Pos logit contributions
 injuries+0.026
 poisoning+0.025
guard+0.025
 payoff+0.020
 unfortunately+0.020
Neg logit contributions
aling-0.085
alling-0.078
astics-0.075
irl-0.075
 sings-0.073
Token
Token ID95
Feature activation-9.37e-02

Pos logit contributions
aling+0.082
 sings+0.076
alling+0.075
ophone+0.072
c+0.071
Neg logit contributions
guard-0.060
 poisoning-0.055
t-0.051
-0.048
ho-0.048
Token
Token ID26391
Feature activation-5.27e-02

Pos logit contributions
aling+0.134
 sings+0.134
 goals+0.130
irl+0.130
am+0.127
Neg logit contributions
 poisoning-0.124
t-0.121
guard-0.118
 payoff-0.114
-0.107
Token
Token ID129
Feature activation-9.27e-02

Pos logit contributions
aling+0.115
alling+0.108
aled+0.100
alled+0.094
irl+0.090
Neg logit contributions
t-0.049
 poisoning-0.048
 injuries-0.048
guard-0.046
 they-0.046
Token
Token ID241
Feature activation-5.09e-03

Pos logit contributions
aling+0.135
 sings+0.131
 cycl+0.131
ophone+0.128
c+0.125
Neg logit contributions
 that-0.109
 poisoning-0.105
t-0.099
 first-0.094
 yet-0.094
Tokenyou
Token ID5832
Feature activation-4.24e-03

Pos logit contributions
aling+0.009
alling+0.008
astics+0.008
irl+0.008
 sings+0.007
Neg logit contributions
 injuries-0.007
 poisoning-0.007
guard-0.006
 unfortunately-0.006
 payoff-0.006
Token will
Token ID481
Feature activation+1.56e-02

Pos logit contributions
aling+0.008
alling+0.007
astics+0.007
irl+0.007
 sings+0.006
Neg logit contributions
guard-0.005
 poisoning-0.005
 injuries-0.005
 unfortunately-0.005
t-0.005