TOP ACTIVATIONS
MAX = 10.080

by and saw how sad Z ippy was . The girl
asked what was wrong and Z ippy told her about her
behind a big rock . Z ippy was so happy and
From that day on , Z ippy made sure to keep
The tree was gone ! Z eb was happy because his
her brown fur . <|endoftext|> Ark ie wanted his own equipment
that he wanted . <|endoftext|> Ak io was a 3 year
, a gentle girl named Ci ara stumbled upon a lovely
! â Car la felt sad and disappointed
, something bad happened . Z ee split in two .
R uf us now ." El le was so happy that
, " No problem , Z ippy . That 's what
stories said . Mama helped Ci ara and together they read
nowhere to be found . Z ippy was very sad and
missed you so much , Z ipper ! I promise I
using the ladder , then R uf us can come over
than the park . K ait lyn ran over to the
again . â <|endoftext|> K itty was outside in the
many different things . <|endoftext|> K athy was a helpful girl
, " Thank you , Z og . You are very

MIN ACTIVATIONS
MIN = -7.131

very much . It was a cute , intelligent t eddy
very much . It was a t eddy bear with soft
very much . It was a fragile doll that she always
very much . It was a small , pretty ball er
to wear . It was a flexible skirt that moved when
very much . It was a big , pink t eddy
loved using . It was a lamp with a pink top
a lot . It was a red car that went fast
color in and it was a great way to pass the
the snow . They made a big snow man with a
shovel s . They made a big castle with towers and
a roof . It was a pretty roof . It was
the snow . They made a big snow man with a
had a small shovel and a bucket . He looked for
. It made the sky a deep orange color and the
the dark . It was a hand that looked like a
his garden . It was a very special plant , and
pink shape . It was a nice shape and it kept
she went . It was a big , red hat that
the snow . They made a big ball and a small

INTERVAL 5.777 - 10.080
CONTAINS 0.661%

So , Qu ackers and Rib bit left
a brand new house ! <|endoftext|> Tom my was out playing
us escape the mystery ? <|endoftext|> Once upon a time there
end of the march . <|endoftext|> Tom was a silly boy
, funny fire man !" <|endoftext|> Once upon a time ,

INTERVAL 1.475 - 5.777
CONTAINS 20.286%

over and began playing . The olive was
. The car sees Ben . It hon ks
make new discoveries . <|endoftext|> Once upon a time , there
doing here ?". The fle a replied , "
silver stars . Her mom saw her crying and came to

INTERVAL -2.828 - 1.475
CONTAINS 73.342%

doesn 't mean it 's good for you .
, ˜ Can I have one of those
stuck in the mud . He was scared and alone .
three years old , to help her . Together , they
park with a big slide . Lily went up and down

INTERVAL -7.131 - -2.828
CONTAINS 5.711%

, bright orange carrot he could find . He wanted it
still explore , but he would manage to stay close to
much noise !" Sara says , frown ing . She tries
so excited , she could hardly wait to go inside .
him it was a qu ilt . Jack asked his mom
Token by
Token ID416
Feature activation+2.47e+00

Pos logit contributions
ople+0.047
ipped+0.044
ebra+0.043
pa+0.042
father+0.042
Neg logit contributions
 exc-0.026
 bend-0.025
 sturdy-0.025
 flat-0.024
 heal-0.023
Token and
Token ID290
Feature activation-1.84e-01

Pos logit contributions
ople+0.081
ipped+0.076
ebra+0.075
father+0.074
pa+0.073
Neg logit contributions
 exc-0.039
 sturdy-0.038
 bend-0.037
 heal-0.036
 flat-0.036
Token saw
Token ID2497
Feature activation+3.31e+00

Pos logit contributions
 sturdy+0.004
 exc+0.004
 bend+0.004
 flat+0.004
 heal+0.004
Neg logit contributions
ople-0.005
ipped-0.005
pa-0.005
father-0.004
ebra-0.004
Token how
Token ID703
Feature activation+5.22e-01

Pos logit contributions
ople+0.108
ipped+0.101
 End+0.097
ated+0.097
father+0.097
Neg logit contributions
 exc-0.053
 sew-0.049
 sturdy-0.048
 bend-0.048
 unt-0.046
Token sad
Token ID6507
Feature activation+3.32e+00

Pos logit contributions
ople+0.019
ebra+0.018
father+0.018
ipped+0.018
pa+0.018
Neg logit contributions
 sturdy-0.008
 exc-0.007
 flat-0.007
 boun-0.006
 bend-0.006
Token Z
Token ID1168
Feature activation+9.94e+00

Pos logit contributions
ople+0.115
ipped+0.113
father+0.111
eled+0.111
ated+0.108
Neg logit contributions
 exc-0.043
 sturdy-0.040
 bend-0.040
 heal-0.039
 sew-0.038
Tokenippy
Token ID41214
Feature activation+2.61e+00

Pos logit contributions
ople+0.201
pa+0.197
father+0.190
iece+0.180
 End+0.169
Neg logit contributions
onna-0.233
 exc-0.216
 bend-0.212
 gently-0.208
 opener-0.206
Token was
Token ID373
Feature activation-9.78e-01

Pos logit contributions
ople+0.066
father+0.064
pa+0.063
eled+0.062
ipped+0.062
Neg logit contributions
 sturdy-0.056
 bend-0.054
 exc-0.054
 flat-0.053
 heal-0.051
Token.
Token ID13
Feature activation-7.03e-02

Pos logit contributions
 sturdy+0.023
 exc+0.022
 bleed+0.021
 bend+0.020
onna+0.020
Neg logit contributions
ople-0.024
pa-0.023
ipped-0.023
ie-0.022
father-0.022
Token The
Token ID383
Feature activation+3.23e+00

Pos logit contributions
 sturdy+0.001
 exc+0.001
 flat+0.001
 bend+0.001
onna+0.001
Neg logit contributions
ople-0.002
ipped-0.002
father-0.002
pa-0.002
ie-0.002
Token girl
Token ID2576
Feature activation+1.26e+00

Pos logit contributions
ople+0.099
ipped+0.095
pa+0.093
ated+0.091
father+0.090
Neg logit contributions
 exc-0.063
 sturdy-0.061
 boun-0.060
 bend-0.059
 cozy-0.059
Token asked
Token ID1965
Feature activation+3.61e+00

Pos logit contributions
ople+0.030
ipped+0.029
pa+0.028
father+0.027
ebra+0.026
Neg logit contributions
 sturdy-0.030
 exc-0.030
 flat-0.028
 bend-0.028
 heal-0.028
Token what
Token ID644
Feature activation+2.86e+00

Pos logit contributions
ople+0.118
ipped+0.112
ated+0.109
eled+0.109
aked+0.107
Neg logit contributions
 exc-0.054
 heal-0.053
 gently-0.053
 bend-0.052
 safe-0.050
Token was
Token ID373
Feature activation-2.90e-01

Pos logit contributions
ople+0.061
ipped+0.057
pa+0.056
father+0.054
ie+0.053
Neg logit contributions
 sturdy-0.077
 exc-0.075
 flat-0.072
 heal-0.072
 bend-0.072
Token wrong
Token ID2642
Feature activation+6.82e-01

Pos logit contributions
 sturdy+0.006
 exc+0.006
 flat+0.006
 heal+0.006
 bend+0.006
Neg logit contributions
ople-0.008
ipped-0.007
pa-0.007
eled-0.007
father-0.007
Token and
Token ID290
Feature activation+1.92e-01

Pos logit contributions
ople+0.019
ipped+0.018
pa+0.017
father+0.017
ebra+0.017
Neg logit contributions
 sturdy-0.014
 exc-0.014
 flat-0.013
 heal-0.013
 bend-0.013
Token Z
Token ID1168
Feature activation+9.72e+00

Pos logit contributions
ople+0.006
ipped+0.006
ebra+0.006
father+0.006
pa+0.006
Neg logit contributions
 exc-0.003
 sturdy-0.003
 bend-0.002
 heal-0.002
 sew-0.002
Tokenippy
Token ID41214
Feature activation+2.93e+00

Pos logit contributions
ople+0.190
iece+0.184
father+0.169
pa+0.165
 End+0.164
Neg logit contributions
onna-0.233
 exc-0.222
 bend-0.213
 gently-0.212
 unt-0.210
Token told
Token ID1297
Feature activation+3.01e+00

Pos logit contributions
ople+0.081
ipped+0.078
father+0.076
ebra+0.075
pa+0.074
Neg logit contributions
 exc-0.057
 sturdy-0.055
 bend-0.053
 sew-0.050
 heal-0.050
Token her
Token ID607
Feature activation+1.54e+00

Pos logit contributions
ople+0.095
ipped+0.090
eled+0.088
aked+0.087
ebra+0.087
Neg logit contributions
 heal-0.053
 bend-0.050
 sew-0.050
 sturdy-0.050
 exc-0.049
Token about
Token ID546
Feature activation+1.96e+00

Pos logit contributions
ople+0.042
ipped+0.040
pa+0.040
father+0.039
ie+0.039
Neg logit contributions
 sturdy-0.031
 exc-0.030
 flat-0.029
 bend-0.029
 bleed-0.029
Token her
Token ID607
Feature activation-3.61e-01

Pos logit contributions
ople+0.045
pa+0.043
ipped+0.043
father+0.042
ie+0.041
Neg logit contributions
 sturdy-0.048
 exc-0.047
 bleed-0.045
 flat-0.044
 bend-0.044
Token behind
Token ID2157
Feature activation+3.03e+00

Pos logit contributions
ople+0.036
ipped+0.034
pa+0.033
father+0.033
atted+0.032
Neg logit contributions
 sturdy-0.023
 flat-0.021
 exc-0.020
 bend-0.020
 boun-0.020
Token a
Token ID257
Feature activation-1.00e+00

Pos logit contributions
ople+0.099
ipped+0.095
ebra+0.095
atted+0.094
father+0.090
Neg logit contributions
 sturdy-0.046
 flat-0.044
 heal-0.044
 bend-0.044
 sew-0.043
Token big
Token ID1263
Feature activation-8.09e-01

Pos logit contributions
 exc+0.021
 bleed+0.020
 sturdy+0.019
onna+0.018
 heal+0.018
Neg logit contributions
ie-0.023
les-0.023
ebra-0.023
ople-0.023
mother-0.023
Token rock
Token ID3881
Feature activation-1.85e+00

Pos logit contributions
 exc+0.017
 sturdy+0.016
 bleed+0.015
 heal+0.015
onna+0.015
Neg logit contributions
ople-0.022
pa-0.021
ipped-0.021
ie-0.020
el-0.020
Token.
Token ID13
Feature activation+3.04e-01

Pos logit contributions
 sturdy+0.037
 exc+0.036
 flat+0.033
 bend+0.033
 heal+0.033
Neg logit contributions
ople-0.052
ipped-0.050
pa-0.048
ie-0.048
father-0.047
Token Z
Token ID1168
Feature activation+9.61e+00

Pos logit contributions
ople+0.009
pa+0.009
ipped+0.009
ie+0.009
father+0.009
Neg logit contributions
 sturdy-0.005
 exc-0.005
 flat-0.005
 bend-0.005
onna-0.005
Tokenippy
Token ID41214
Feature activation+2.07e+00

Pos logit contributions
ople+0.203
iece+0.191
pa+0.185
father+0.180
 End+0.170
Neg logit contributions
onna-0.219
 exc-0.210
 unt-0.201
 gently-0.199
 bend-0.197
Token was
Token ID373
Feature activation-2.33e+00

Pos logit contributions
ople+0.050
ipped+0.048
pa+0.047
father+0.047
ebra+0.045
Neg logit contributions
 sturdy-0.049
 exc-0.048
 bend-0.047
 sew-0.045
 flat-0.045
Token so
Token ID523
Feature activation-2.35e+00

Pos logit contributions
 sturdy+0.044
 exc+0.043
onna+0.042
 bleed+0.042
 heal+0.041
Neg logit contributions
ople-0.064
ipped-0.061
atted-0.060
pa-0.060
eled-0.060
Token happy
Token ID3772
Feature activation-7.66e-02

Pos logit contributions
 sturdy+0.035
onna+0.032
 sew+0.032
 heal+0.032
 bleed+0.032
Neg logit contributions
ople-0.073
ipped-0.071
ed-0.070
atted-0.070
pa-0.069
Token and
Token ID290
Feature activation-2.31e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
 flat+0.002
 bleed+0.002
 heal+0.002
Neg logit contributions
ople-0.002
ipped-0.002
pa-0.002
ie-0.002
father-0.002
Token From
Token ID3574
Feature activation+1.75e+00

Pos logit contributions
 exc+0.003
 sturdy+0.003
onna+0.003
 flat+0.003
 bend+0.003
Neg logit contributions
ople-0.005
father-0.004
pa-0.004
ipped-0.004
ie-0.004
Token that
Token ID326
Feature activation+1.74e+00

Pos logit contributions
ople+0.057
ipped+0.055
ado+0.053
ated+0.053
aked+0.053
Neg logit contributions
 sturdy-0.028
 exc-0.025
 boun-0.025
 flat-0.024
 bend-0.024
Token day
Token ID1110
Feature activation+8.38e-01

Pos logit contributions
ople+0.038
ipped+0.034
atted+0.032
io+0.031
eled+0.031
Neg logit contributions
 exc-0.050
 sturdy-0.049
 heal-0.048
 flat-0.047
 bend-0.047
Token on
Token ID319
Feature activation+3.08e+00

Pos logit contributions
ople+0.024
pa+0.022
ipped+0.021
ie+0.021
les+0.021
Neg logit contributions
 exc-0.017
 sturdy-0.016
 flat-0.016
 heal-0.015
 bend-0.015
Token,
Token ID11
Feature activation+2.67e-01

Pos logit contributions
ople+0.094
eled+0.088
aked+0.087
ipped+0.086
father+0.085
Neg logit contributions
 sturdy-0.054
 exc-0.053
 bend-0.053
 flat-0.052
onna-0.052
Token Z
Token ID1168
Feature activation+9.33e+00

Pos logit contributions
ople+0.008
ipped+0.008
pa+0.008
eled+0.008
father+0.008
Neg logit contributions
 exc-0.005
 sturdy-0.004
 boun-0.004
 bend-0.004
 flat-0.004
Tokenippy
Token ID41214
Feature activation+2.52e+00

Pos logit contributions
ople+0.205
iece+0.188
pa+0.175
father+0.167
oos+0.158
Neg logit contributions
onna-0.218
 exc-0.218
 gently-0.203
 unt-0.202
 bend-0.193
Token made
Token ID925
Feature activation+1.27e+00

Pos logit contributions
ople+0.069
ipped+0.066
eled+0.064
father+0.063
pa+0.063
Neg logit contributions
 exc-0.050
 sturdy-0.049
 bend-0.047
 heal-0.046
 sew-0.046
Token sure
Token ID1654
Feature activation+1.34e+00

Pos logit contributions
ople+0.046
ipped+0.044
ebra+0.043
pa+0.042
eled+0.041
Neg logit contributions
 sturdy-0.017
 exc-0.016
 heal-0.016
 crunch-0.015
 bend-0.015
Token to
Token ID284
Feature activation-2.48e+00

Pos logit contributions
ople+0.048
ipped+0.047
ebra+0.047
pa+0.045
les+0.045
Neg logit contributions
 sturdy-0.016
 exc-0.015
 heal-0.015
 flat-0.014
onna-0.014
Token keep
Token ID1394
Feature activation+1.85e+00

Pos logit contributions
 sturdy+0.039
 exc+0.036
 flat+0.035
 bleed+0.035
onna+0.035
Neg logit contributions
ople-0.077
pa-0.075
father-0.074
ipped-0.074
ie-0.074
Token The
Token ID383
Feature activation+1.23e+00

Pos logit contributions
ople+0.023
ie+0.022
pa+0.022
father+0.022
ipped+0.022
Neg logit contributions
 exc-0.015
 sturdy-0.014
 flat-0.014
 bend-0.014
onna-0.013
Token tree
Token ID5509
Feature activation-8.82e-02

Pos logit contributions
father+0.032
ople+0.032
 End+0.031
pa+0.031
ipped+0.031
Neg logit contributions
onna-0.027
 sturdy-0.024
 exc-0.023
 bleed-0.023
 moist-0.022
Token was
Token ID373
Feature activation-3.23e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
onna+0.002
 bleed+0.002
 heal+0.002
Neg logit contributions
ople-0.002
ipped-0.002
pa-0.002
ie-0.002
atted-0.002
Token gone
Token ID3750
Feature activation-1.24e+00

Pos logit contributions
 sew+0.063
onna+0.063
 knit+0.060
 exc+0.060
 heal+0.059
Neg logit contributions
atted-0.076
eled-0.075
ople-0.073
 End-0.073
ed-0.072
Token!
Token ID0
Feature activation+8.91e-01

Pos logit contributions
 exc+0.028
 sturdy+0.027
 bleed+0.026
 sew+0.026
onna+0.026
Neg logit contributions
ople-0.030
ipped-0.029
ie-0.029
pa-0.029
el-0.028
Token Z
Token ID1168
Feature activation+9.26e+00

Pos logit contributions
ople+0.028
ie+0.028
father+0.027
ipped+0.027
pa+0.027
Neg logit contributions
 exc-0.015
 flat-0.014
 sturdy-0.014
 bend-0.014
onna-0.014
Tokeneb
Token ID1765
Feature activation+4.57e+00

Pos logit contributions
ople+0.155
mother+0.125
zie+0.125
father+0.122
mo+0.122
Neg logit contributions
onna-0.268
 exc-0.263
 unt-0.243
 flat-0.235
 bend-0.232
Token was
Token ID373
Feature activation-2.55e+00

Pos logit contributions
ople+0.115
ipped+0.112
ted+0.106
eled+0.102
ays+0.102
Neg logit contributions
 exc-0.102
 bend-0.097
 sturdy-0.097
 heal-0.096
 sew-0.092
Token happy
Token ID3772
Feature activation-1.13e+00

Pos logit contributions
 sturdy+0.046
 exc+0.044
onna+0.042
 bend+0.042
 heal+0.042
Neg logit contributions
ople-0.076
ipped-0.073
pa-0.071
father-0.070
atted-0.070
Token because
Token ID780
Feature activation-5.39e-01

Pos logit contributions
 sturdy+0.028
 exc+0.027
 bleed+0.026
 sew+0.025
 flat+0.025
Neg logit contributions
ople-0.026
ipped-0.025
ie-0.025
pa-0.025
father-0.024
Token his
Token ID465
Feature activation+6.69e-01

Pos logit contributions
 exc+0.009
 sturdy+0.009
 flat+0.008
 bend+0.008
 heal+0.008
Neg logit contributions
ople-0.018
ipped-0.017
pa-0.016
father-0.016
atted-0.016
Token her
Token ID607
Feature activation+6.69e-01

Pos logit contributions
ople+0.056
ipped+0.055
ebra+0.052
eled+0.051
atted+0.050
Neg logit contributions
 sturdy-0.043
 exc-0.040
 heal-0.040
 flat-0.040
 crunch-0.039
Token brown
Token ID7586
Feature activation+2.34e+00

Pos logit contributions
ople+0.019
ipped+0.018
pa+0.018
father+0.018
ie+0.017
Neg logit contributions
 sturdy-0.013
 exc-0.013
 flat-0.012
 bend-0.012
 heal-0.012
Token fur
Token ID9230
Feature activation-9.51e-01

Pos logit contributions
ople+0.066
ipped+0.062
eled+0.059
father+0.059
ie+0.059
Neg logit contributions
 sturdy-0.046
 exc-0.044
 flat-0.044
 bend-0.043
 heal-0.042
Token.
Token ID13
Feature activation-1.31e-01

Pos logit contributions
 sturdy+0.019
 exc+0.019
 bend+0.018
 flat+0.017
 bleed+0.017
Neg logit contributions
ople-0.026
ipped-0.025
pa-0.025
father-0.024
ie-0.024
Token<|endoftext|>
Token ID50256
Feature activation+7.05e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
onna+0.002
 flat+0.002
 bend+0.002
Neg logit contributions
ople-0.004
ipped-0.004
ie-0.004
pa-0.004
eled-0.004
TokenArk
Token ID42007
Feature activation+9.01e+00

Pos logit contributions
ople+0.192
atted+0.180
father+0.178
pa+0.176
ie+0.172
Neg logit contributions
 opener-0.142
onna-0.133
 boun-0.133
 exc-0.132
 bend-0.131
Tokenie
Token ID494
Feature activation+2.07e+00

Pos logit contributions
ipped+0.191
ted+0.187
rified+0.180
ement+0.179
ebra+0.170
Neg logit contributions
 gently-0.184
 fold-0.178
 clean-0.175
 exc-0.172
 put-0.168
Token wanted
Token ID2227
Feature activation+1.93e+00

Pos logit contributions
ople+0.057
ipped+0.055
father+0.055
ebra+0.054
eled+0.053
Neg logit contributions
 exc-0.038
 sturdy-0.037
 heal-0.036
 bend-0.036
 sew-0.036
Token his
Token ID465
Feature activation+8.00e-01

Pos logit contributions
ople+0.059
ipped+0.056
pa+0.056
father+0.054
ie+0.054
Neg logit contributions
 sturdy-0.034
 exc-0.033
 bleed-0.031
 flat-0.031
 bend-0.030
Token own
Token ID898
Feature activation+8.14e-01

Pos logit contributions
ople+0.025
ipped+0.024
les+0.022
aked+0.022
atted+0.022
Neg logit contributions
 exc-0.015
 sturdy-0.015
 flat-0.014
 boun-0.014
 bend-0.014
Token equipment
Token ID5112
Feature activation-3.17e+00

Pos logit contributions
ople+0.024
ipped+0.023
pa+0.022
father+0.022
eled+0.022
Neg logit contributions
 sturdy-0.015
 exc-0.015
 flat-0.014
 heal-0.014
 bend-0.014
Token that
Token ID326
Feature activation+1.29e+00

Pos logit contributions
 exc+0.060
 sturdy+0.060
 bleed+0.056
 sew+0.055
 bend+0.054
Neg logit contributions
ople-0.046
ipped-0.044
ie-0.043
ed-0.042
les-0.041
Token he
Token ID339
Feature activation+1.69e+00

Pos logit contributions
ople+0.031
ipped+0.029
pa+0.028
father+0.028
ebra+0.027
Neg logit contributions
 sturdy-0.031
 exc-0.031
 flat-0.029
 heal-0.029
 bend-0.029
Token wanted
Token ID2227
Feature activation+1.60e+00

Pos logit contributions
ople+0.047
ipped+0.045
father+0.043
pa+0.043
ebra+0.043
Neg logit contributions
 sturdy-0.035
 exc-0.034
 bend-0.033
 heal-0.032
 flat-0.032
Token.
Token ID13
Feature activation-5.91e-01

Pos logit contributions
ople+0.049
ipped+0.047
pa+0.045
father+0.045
ebra+0.045
Neg logit contributions
 sturdy-0.028
 exc-0.027
 flat-0.026
 heal-0.026
 bend-0.026
Token<|endoftext|>
Token ID50256
Feature activation+6.78e+00

Pos logit contributions
 sturdy+0.010
 exc+0.010
onna+0.010
 flat+0.009
 heal+0.009
Neg logit contributions
ople-0.018
ipped-0.018
ie-0.017
pa-0.017
ebra-0.017
TokenAk
Token ID33901
Feature activation+8.99e+00

Pos logit contributions
ople+0.228
atted+0.216
father+0.207
pa+0.206
eled+0.205
Neg logit contributions
 opener-0.102
onna-0.096
 boun-0.090
 bend-0.090
 exc-0.090
Tokenio
Token ID952
Feature activation+4.51e+00

Pos logit contributions
ople+0.122
 nights+0.117
 mornings+0.117
 End+0.113
ting+0.112
Neg logit contributions
 unt-0.244
onna-0.239
 unw-0.229
 gently-0.225
 tie-0.220
Token was
Token ID373
Feature activation-2.66e+00

Pos logit contributions
ople+0.119
ipped+0.116
 End+0.109
ebra+0.109
eled+0.107
Neg logit contributions
 exc-0.089
onna-0.087
 sew-0.087
 sturdy-0.087
 bend-0.086
Token a
Token ID257
Feature activation-4.88e+00

Pos logit contributions
 sturdy+0.050
 exc+0.049
 flat+0.047
 bend+0.046
 heal+0.045
Neg logit contributions
ople-0.079
ipped-0.076
pa-0.073
father-0.072
ebra-0.072
Token 3
Token ID513
Feature activation+2.01e+00

Pos logit contributions
 sturdy+0.106
 exc+0.103
 bleed+0.102
 heal+0.100
onna+0.099
Neg logit contributions
ople-0.126
ipped-0.118
pa-0.117
father-0.115
ebra-0.114
Token year
Token ID614
Feature activation+8.15e-01

Pos logit contributions
ipped+0.043
ople+0.041
ance+0.040
ays+0.040
father+0.039
Neg logit contributions
 exc-0.055
 sturdy-0.050
 strong-0.049
 crunch-0.048
 boun-0.048
Token,
Token ID11
Feature activation+2.73e-01

Pos logit contributions
ople+0.043
les+0.040
eled+0.040
father+0.039
pa+0.039
Neg logit contributions
 exc-0.024
 bend-0.022
 boun-0.021
 flat-0.021
 sturdy-0.020
Token a
Token ID257
Feature activation-1.62e+00

Pos logit contributions
ople+0.009
pa+0.009
ipped+0.008
father+0.008
mother+0.008
Neg logit contributions
 exc-0.005
 sturdy-0.004
 flat-0.004
 bend-0.004
 boun-0.004
Token gentle
Token ID10296
Feature activation-9.68e-02

Pos logit contributions
 exc+0.031
 sturdy+0.031
 bend+0.029
 flat+0.028
 boun+0.028
Neg logit contributions
ople-0.049
ipped-0.047
pa-0.045
father-0.044
atted-0.044
Token girl
Token ID2576
Feature activation+7.86e-01

Pos logit contributions
 sturdy+0.002
 exc+0.002
 heal+0.002
 bend+0.002
 crunch+0.002
Neg logit contributions
ople-0.003
ipped-0.003
father-0.003
pa-0.003
atted-0.003
Token named
Token ID3706
Feature activation+1.93e+00

Pos logit contributions
ople+0.027
ipped+0.026
les+0.025
father+0.025
or+0.025
Neg logit contributions
 exc-0.012
 sew-0.010
 bend-0.009
 wrap-0.009
 clean-0.009
Token Ci
Token ID37685
Feature activation+8.88e+00

Pos logit contributions
ople+0.061
ipped+0.060
father+0.058
pa+0.058
ebra+0.057
Neg logit contributions
 sturdy-0.033
 exc-0.032
 bend-0.031
 sew-0.030
 flat-0.030
Tokenara
Token ID3301
Feature activation+3.75e+00

Pos logit contributions
ople+0.169
years+0.168
atted+0.168
mo+0.161
eled+0.156
Neg logit contributions
 sew-0.217
onna-0.211
 tie-0.208
 fold-0.208
 wrap-0.207
Token stumbled
Token ID24241
Feature activation+2.63e+00

Pos logit contributions
ople+0.122
ipped+0.120
atted+0.112
ebra+0.112
or+0.110
Neg logit contributions
 exc-0.064
 clean-0.052
 sew-0.052
 cozy-0.052
 bend-0.051
Token upon
Token ID2402
Feature activation+2.67e+00

Pos logit contributions
ople+0.081
pa+0.076
father+0.074
ipped+0.073
ebra+0.071
Neg logit contributions
 flat-0.047
 exc-0.047
 sturdy-0.047
onna-0.046
 bend-0.046
Token a
Token ID257
Feature activation-2.99e+00

Pos logit contributions
ople+0.081
ipped+0.077
pa+0.075
father+0.074
ebra+0.073
Neg logit contributions
 sturdy-0.049
 exc-0.047
 flat-0.045
 bend-0.045
 heal-0.044
Token lovely
Token ID14081
Feature activation-1.60e+00

Pos logit contributions
 exc+0.056
 sturdy+0.054
 bleed+0.054
onna+0.051
 opener+0.051
Neg logit contributions
ie-0.080
by-0.079
ople-0.078
ebra-0.078
pa-0.077
Token!
Token ID0
Feature activation+1.31e+00

Pos logit contributions
ople+0.021
ipped+0.020
pa+0.020
ie+0.020
el+0.019
Neg logit contributions
 sturdy-0.019
 exc-0.019
 bend-0.017
 bleed-0.017
 sew-0.017
Tokenâ
Token ID22940
Feature activation+6.10e+00

Pos logit contributions
ople+0.036
ipped+0.034
pa+0.033
father+0.033
ie+0.032
Neg logit contributions
 sturdy-0.027
 exc-0.027
 flat-0.025
 heal-0.025
 bend-0.025
Token
Token ID26391
Feature activation+2.91e+00

Pos logit contributions
eled+0.144
ople+0.142
ipped+0.142
father+0.141
ie+0.140
Neg logit contributions
 sturdy-0.135
 heal-0.133
 sew-0.132
 exc-0.132
 flat-0.126
Token
Token ID198
Feature activation+2.30e+00

Pos logit contributions
ople+0.079
ipped+0.076
pa+0.074
father+0.074
ie+0.074
Neg logit contributions
 sturdy-0.060
 exc-0.059
 flat-0.056
 bend-0.055
 heal-0.055
Token
Token ID198
Feature activation+1.18e+00

Pos logit contributions
ople+0.068
ipped+0.065
pa+0.064
father+0.063
ie+0.062
Neg logit contributions
 sturdy-0.043
 exc-0.043
onna-0.040
 bend-0.040
 flat-0.040
TokenCar
Token ID9914
Feature activation+8.81e+00

Pos logit contributions
ople+0.034
ipped+0.033
pa+0.032
father+0.031
ie+0.031
Neg logit contributions
 sturdy-0.023
 exc-0.022
 flat-0.021
 heal-0.021
 bend-0.021
Tokenla
Token ID5031
Feature activation+4.32e+00

Pos logit contributions
father+0.157
pa+0.133
uses+0.132
igh+0.129
icer+0.128
Neg logit contributions
 exc-0.230
 wrap-0.229
onna-0.227
 sil-0.226
 gently-0.225
Token felt
Token ID2936
Feature activation-1.93e+00

Pos logit contributions
ople+0.110
father+0.108
ipped+0.107
pa+0.105
ebra+0.103
Neg logit contributions
 exc-0.091
 sturdy-0.089
 bend-0.087
 sew-0.084
onna-0.082
Token sad
Token ID6507
Feature activation-1.49e+00

Pos logit contributions
 sturdy+0.035
 exc+0.034
onna+0.034
 sew+0.033
 bend+0.032
Neg logit contributions
ople-0.054
ipped-0.052
atted-0.050
ed-0.050
pa-0.050
Token and
Token ID290
Feature activation-3.55e-01

Pos logit contributions
 sturdy+0.031
 exc+0.030
 flat+0.028
 bend+0.028
 bleed+0.028
Neg logit contributions
ople-0.041
ipped-0.039
pa-0.038
father-0.038
ie-0.037
Token disappointed
Token ID11679
Feature activation+5.80e-02

Pos logit contributions
 sturdy+0.006
 exc+0.006
 flat+0.006
 bleed+0.006
 bend+0.006
Neg logit contributions
ople-0.011
ipped-0.011
ebra-0.010
father-0.010
pa-0.010
Token,
Token ID11
Feature activation+2.21e-01

Pos logit contributions
ople+0.059
father+0.056
pa+0.055
ipped+0.055
eled+0.055
Neg logit contributions
 bend-0.033
 exc-0.032
 sturdy-0.032
 flat-0.031
 boun-0.030
Token something
Token ID1223
Feature activation+2.61e-01

Pos logit contributions
ople+0.008
ipped+0.007
pa+0.007
father+0.007
atted+0.007
Neg logit contributions
 exc-0.003
 sturdy-0.003
 bend-0.003
 flat-0.003
 boun-0.003
Token bad
Token ID2089
Feature activation+1.81e+00

Pos logit contributions
ople+0.006
ed+0.006
les+0.006
ipped+0.006
ie+0.005
Neg logit contributions
 sturdy-0.006
 exc-0.006
onna-0.006
 sew-0.006
 bleed-0.006
Token happened
Token ID3022
Feature activation+6.81e-01

Pos logit contributions
ople+0.043
ipped+0.040
pa+0.039
father+0.039
ebra+0.038
Neg logit contributions
 sturdy-0.044
 exc-0.043
 bend-0.041
 flat-0.041
 heal-0.041
Token.
Token ID13
Feature activation-5.53e-01

Pos logit contributions
ople+0.020
ipped+0.019
pa+0.019
father+0.018
atted+0.018
Neg logit contributions
 sturdy-0.013
 exc-0.013
 flat-0.012
 heal-0.012
 bend-0.012
Token Z
Token ID1168
Feature activation+8.76e+00

Pos logit contributions
 sturdy+0.012
 exc+0.011
 flat+0.011
 bend+0.011
 bleed+0.011
Neg logit contributions
ople-0.015
ipped-0.014
pa-0.014
ie-0.014
father-0.014
Tokenee
Token ID1453
Feature activation+3.78e+00

Pos logit contributions
ople+0.178
father+0.159
mother+0.155
mo+0.154
years+0.150
Neg logit contributions
 fold-0.221
 bend-0.213
onna-0.213
 exc-0.207
 tie-0.200
Token split
Token ID6626
Feature activation+3.86e+00

Pos logit contributions
ople+0.100
father+0.099
pa+0.099
ipped+0.096
mother+0.093
Neg logit contributions
 exc-0.075
 bend-0.075
 bleed-0.073
 sturdy-0.073
 flat-0.070
Token in
Token ID287
Feature activation+1.85e+00

Pos logit contributions
ople+0.124
father+0.122
pa+0.118
mother+0.116
parents+0.115
Neg logit contributions
 flat-0.067
 bend-0.064
 sturdy-0.062
 boun-0.059
 exc-0.058
Token two
Token ID734
Feature activation+1.13e+00

Pos logit contributions
ople+0.051
ipped+0.048
pa+0.047
ebra+0.046
ie+0.046
Neg logit contributions
 sturdy-0.040
 exc-0.039
 flat-0.039
 bend-0.038
 bleed-0.037
Token.
Token ID13
Feature activation-6.12e-01

Pos logit contributions
ople+0.027
ipped+0.025
pa+0.024
father+0.024
el+0.024
Neg logit contributions
 sturdy-0.028
 exc-0.027
 heal-0.026
onna-0.026
 bleed-0.026
Token R
Token ID371
Feature activation+7.52e+00

Pos logit contributions
ople+0.081
ipped+0.077
eled+0.075
father+0.074
pa+0.074
Neg logit contributions
 sturdy-0.059
 exc-0.057
 bend-0.056
 heal-0.055
 flat-0.054
Tokenuf
Token ID3046
Feature activation+6.43e+00

Pos logit contributions
ople+0.150
ipped+0.142
eled+0.140
ed+0.138
aked+0.133
Neg logit contributions
onna-0.201
 exc-0.192
 heal-0.187
 bend-0.182
 sturdy-0.181
Tokenus
Token ID385
Feature activation+1.52e+00

Pos logit contributions
ipped+0.107
ople+0.107
eled+0.100
atted+0.093
pa+0.090
Neg logit contributions
onna-0.199
 exc-0.187
 sturdy-0.178
 opener-0.178
 heal-0.178
Token now
Token ID783
Feature activation+6.59e-01

Pos logit contributions
ople+0.050
ipped+0.049
father+0.048
pa+0.048
eled+0.048
Neg logit contributions
 sturdy-0.022
 exc-0.022
 flat-0.020
 heal-0.020
 bend-0.019
Token."
Token ID526
Feature activation+2.51e-01

Pos logit contributions
ople+0.020
ipped+0.020
pa+0.019
father+0.019
ie+0.019
Neg logit contributions
 sturdy-0.011
 exc-0.011
 flat-0.010
 bleed-0.010
 bend-0.010
Token El
Token ID2574
Feature activation+8.73e+00

Pos logit contributions
ople+0.007
ipped+0.007
pa+0.007
father+0.007
ie+0.006
Neg logit contributions
 exc-0.005
 sturdy-0.005
 bend-0.005
onna-0.005
 heal-0.005
Tokenle
Token ID293
Feature activation+2.07e+00

Pos logit contributions
ipped+0.164
ople+0.163
oos+0.162
father+0.160
irens+0.160
Neg logit contributions
 exc-0.202
 gently-0.189
 fold-0.184
onna-0.175
 nib-0.175
Token was
Token ID373
Feature activation-2.10e+00

Pos logit contributions
ople+0.052
ipped+0.049
father+0.049
pa+0.048
ebra+0.047
Neg logit contributions
 exc-0.047
 sturdy-0.045
 bend-0.045
 heal-0.043
 sew-0.043
Token so
Token ID523
Feature activation-2.03e+00

Pos logit contributions
 sturdy+0.040
 exc+0.038
 heal+0.037
 bleed+0.037
 flat+0.036
Neg logit contributions
ople-0.060
ipped-0.058
pa-0.056
atted-0.056
eled-0.055
Token happy
Token ID3772
Feature activation+8.57e-02

Pos logit contributions
 sturdy+0.030
 exc+0.028
 heal+0.027
 bleed+0.027
 flat+0.027
Neg logit contributions
ople-0.066
ipped-0.065
pa-0.063
atted-0.062
eled-0.061
Token that
Token ID326
Feature activation+1.34e+00

Pos logit contributions
ople+0.002
ipped+0.002
pa+0.002
ie+0.002
father+0.002
Neg logit contributions
 sturdy-0.002
 exc-0.002
 flat-0.002
 bleed-0.002
 heal-0.002
Token,
Token ID11
Feature activation-2.55e+00

Pos logit contributions
 sturdy+0.012
 exc+0.012
 flat+0.011
 heal+0.011
 bend+0.011
Neg logit contributions
ople-0.012
ipped-0.012
pa-0.011
father-0.011
ebra-0.011
Token "
Token ID366
Feature activation-2.09e-01

Pos logit contributions
 sturdy+0.049
 bleed+0.047
 exc+0.042
 sew+0.042
 flat+0.042
Neg logit contributions
ople-0.071
ipped-0.071
atted-0.069
ed-0.067
eled-0.065
TokenNo
Token ID2949
Feature activation+2.51e+00

Pos logit contributions
 sturdy+0.004
 exc+0.004
 bend+0.004
 heal+0.004
 boun+0.004
Neg logit contributions
ople-0.006
ipped-0.006
pa-0.006
father-0.006
atted-0.006
Token problem
Token ID1917
Feature activation+9.11e-01

Pos logit contributions
ople+0.067
ipped+0.064
eled+0.061
father+0.061
pa+0.060
Neg logit contributions
 sturdy-0.056
 exc-0.054
 bend-0.053
 flat-0.052
 heal-0.052
Token,
Token ID11
Feature activation-1.44e+00

Pos logit contributions
ople+0.022
ipped+0.021
pa+0.020
ie+0.019
atted+0.019
Neg logit contributions
 sturdy-0.022
 exc-0.021
 flat-0.021
 bleed-0.020
 bend-0.020
Token Z
Token ID1168
Feature activation+8.73e+00

Pos logit contributions
 sturdy+0.022
 exc+0.021
 flat+0.020
 heal+0.020
 bend+0.019
Neg logit contributions
ople-0.048
ipped-0.046
pa-0.045
father-0.044
atted-0.044
Tokenippy
Token ID41214
Feature activation+2.86e+00

Pos logit contributions
ople+0.158
pa+0.140
father+0.139
iece+0.129
eled+0.123
Neg logit contributions
 exc-0.245
onna-0.244
 bend-0.228
 boun-0.222
 please-0.221
Token.
Token ID13
Feature activation-1.97e-01

Pos logit contributions
ople+0.086
ipped+0.082
father+0.081
pa+0.080
eled+0.080
Neg logit contributions
 sturdy-0.051
 exc-0.050
 heal-0.046
 flat-0.046
 bend-0.046
Token That
Token ID1320
Feature activation+2.96e+00

Pos logit contributions
 exc+0.003
 sturdy+0.003
 heal+0.003
 sew+0.003
 bend+0.003
Neg logit contributions
ople-0.006
ipped-0.006
pa-0.006
ie-0.006
les-0.006
Token's
Token ID338
Feature activation-4.53e-01

Pos logit contributions
ople+0.066
ipped+0.063
pa+0.059
father+0.059
atted+0.058
Neg logit contributions
 sturdy-0.077
 exc-0.076
 heal-0.073
 bend-0.073
 flat-0.073
Token what
Token ID644
Feature activation+2.98e+00

Pos logit contributions
 sturdy+0.008
 exc+0.008
 flat+0.007
 heal+0.007
 bend+0.007
Neg logit contributions
ople-0.014
ipped-0.013
pa-0.013
father-0.013
ie-0.013
Token stories
Token ID3923
Feature activation-6.93e-01

Pos logit contributions
ople+0.024
pa+0.023
father+0.023
mother+0.022
ipped+0.022
Neg logit contributions
onna-0.020
 sturdy-0.020
 exc-0.020
 bleed-0.019
 flat-0.019
Token said
Token ID531
Feature activation-2.92e-01

Pos logit contributions
 sturdy+0.018
 exc+0.018
 bleed+0.017
 flat+0.017
onna+0.017
Neg logit contributions
ople-0.015
ipped-0.014
pa-0.014
ie-0.014
ed-0.013
Token.
Token ID13
Feature activation-2.75e-01

Pos logit contributions
 sturdy+0.006
 exc+0.006
 flat+0.005
 bleed+0.005
 bend+0.005
Neg logit contributions
ople-0.008
ipped-0.008
pa-0.008
father-0.008
ie-0.008
Token Mama
Token ID40084
Feature activation+3.67e+00

Pos logit contributions
 sturdy+0.006
 exc+0.006
 flat+0.006
 bleed+0.006
 heal+0.006
Neg logit contributions
ople-0.007
ipped-0.007
pa-0.007
father-0.007
atted-0.007
Token helped
Token ID4193
Feature activation+2.52e+00

Pos logit contributions
ople+0.103
ipped+0.101
ebra+0.095
father+0.095
or+0.095
Neg logit contributions
 exc-0.068
 sturdy-0.067
 bend-0.065
 crunch-0.064
 sew-0.064
Token Ci
Token ID37685
Feature activation+8.72e+00

Pos logit contributions
ople+0.073
ebra+0.069
ipped+0.068
father+0.068
atted+0.066
Neg logit contributions
 heal-0.048
 sew-0.048
 exc-0.048
 bend-0.046
 sturdy-0.046
Tokenara
Token ID3301
Feature activation+1.72e-01

Pos logit contributions
father+0.180
ople+0.165
years+0.162
pa+0.155
mother+0.155
Neg logit contributions
 sew-0.199
onna-0.195
 wrap-0.191
 fold-0.191
 knit-0.188
Token and
Token ID290
Feature activation+5.54e-01

Pos logit contributions
ople+0.006
ipped+0.005
pa+0.005
father+0.005
ebra+0.005
Neg logit contributions
 sturdy-0.003
 exc-0.003
 heal-0.003
 bend-0.003
 flat-0.002
Token together
Token ID1978
Feature activation+8.51e-02

Pos logit contributions
ople+0.016
ipped+0.016
ebra+0.015
father+0.015
pa+0.015
Neg logit contributions
 exc-0.010
 sturdy-0.010
 bend-0.009
 sew-0.009
 heal-0.009
Token they
Token ID484
Feature activation-4.56e-01

Pos logit contributions
ople+0.003
ipped+0.003
pa+0.002
ie+0.002
father+0.002
Neg logit contributions
 sturdy-0.002
 exc-0.001
 bleed-0.001
 flat-0.001
 heal-0.001
Token read
Token ID1100
Feature activation+1.23e+00

Pos logit contributions
 sturdy+0.010
 exc+0.010
 bend+0.010
 flat+0.009
 sew+0.009
Neg logit contributions
ople-0.012
ipped-0.012
ebra-0.011
atted-0.011
pa-0.011
Token nowhere
Token ID12062
Feature activation-1.05e+00

Pos logit contributions
 exc+0.028
 sew+0.028
 heal+0.027
 sturdy+0.027
 bleed+0.027
Neg logit contributions
ople-0.031
ie-0.029
el-0.029
 End-0.029
pa-0.029
Token to
Token ID284
Feature activation+7.96e-01

Pos logit contributions
 sturdy+0.022
 exc+0.022
 bleed+0.022
 heal+0.021
 click+0.020
Neg logit contributions
ie-0.027
ople-0.026
ing-0.026
pa-0.026
ipped-0.026
Token be
Token ID307
Feature activation+4.18e-01

Pos logit contributions
ople+0.027
ipped+0.027
pa+0.026
atted+0.026
father+0.026
Neg logit contributions
 sturdy-0.011
 exc-0.011
 heal-0.010
 bend-0.010
 flat-0.010
Token found
Token ID1043
Feature activation+1.60e+00

Pos logit contributions
ople+0.011
ipped+0.011
ebra+0.011
pa+0.011
father+0.010
Neg logit contributions
 sturdy-0.009
 exc-0.009
 flat-0.008
 bend-0.008
 heal-0.008
Token.
Token ID13
Feature activation+2.86e-01

Pos logit contributions
ople+0.053
ipped+0.050
pa+0.050
father+0.050
ebra+0.049
Neg logit contributions
 sturdy-0.024
 exc-0.023
 heal-0.022
 flat-0.022
 bend-0.021
Token Z
Token ID1168
Feature activation+8.72e+00

Pos logit contributions
ople+0.009
ipped+0.009
pa+0.009
father+0.009
ie+0.009
Neg logit contributions
 sturdy-0.005
 exc-0.004
onna-0.004
 bend-0.004
 flat-0.004
Tokenippy
Token ID41214
Feature activation+2.75e+00

Pos logit contributions
ople+0.170
father+0.163
 End+0.162
pa+0.160
years+0.157
Neg logit contributions
onna-0.214
 exc-0.198
 tie-0.196
 unt-0.196
 bend-0.196
Token was
Token ID373
Feature activation-1.40e+00

Pos logit contributions
ople+0.072
father+0.070
ipped+0.069
pa+0.069
ebra+0.066
Neg logit contributions
 sturdy-0.058
 exc-0.057
 bend-0.057
 sew-0.053
 flat-0.052
Token very
Token ID845
Feature activation-2.10e+00

Pos logit contributions
 sturdy+0.021
 exc+0.020
 flat+0.019
 bend+0.019
 crunch+0.018
Neg logit contributions
ople-0.048
ipped-0.047
pa-0.046
ebra-0.045
father-0.045
Token sad
Token ID6507
Feature activation-1.47e+00

Pos logit contributions
 sturdy+0.035
 exc+0.034
 flat+0.031
 bend+0.031
 heal+0.031
Neg logit contributions
ople-0.068
ipped-0.066
pa-0.064
father-0.063
ebra-0.062
Token and
Token ID290
Feature activation-3.68e-01

Pos logit contributions
 sturdy+0.033
 exc+0.032
 bleed+0.030
 heal+0.030
 flat+0.030
Neg logit contributions
ople-0.037
ipped-0.036
pa-0.035
ie-0.035
father-0.034
Token missed
Token ID6825
Feature activation+2.37e+00

Pos logit contributions
ople+0.015
ipped+0.015
pa+0.015
ie+0.014
father+0.014
Neg logit contributions
 sturdy-0.012
 exc-0.011
 flat-0.011
 bend-0.011
 bleed-0.011
Token you
Token ID345
Feature activation+2.07e-01

Pos logit contributions
ople+0.064
ipped+0.062
father+0.059
pa+0.059
eled+0.059
Neg logit contributions
 sturdy-0.050
 exc-0.049
 bend-0.047
 flat-0.047
 heal-0.047
Token so
Token ID523
Feature activation-7.38e-01

Pos logit contributions
ople+0.005
ipped+0.005
pa+0.005
ie+0.005
father+0.005
Neg logit contributions
 sturdy-0.005
 exc-0.004
 bleed-0.004
 bend-0.004
 flat-0.004
Token much
Token ID881
Feature activation+3.38e-02

Pos logit contributions
 sturdy+0.009
 exc+0.009
 flat+0.008
 bend+0.008
 heal+0.008
Neg logit contributions
ople-0.027
ipped-0.026
pa-0.026
father-0.025
ebra-0.025
Token,
Token ID11
Feature activation-1.51e+00

Pos logit contributions
ople+0.001
pa+0.001
ipped+0.001
ie+0.001
ebra+0.001
Neg logit contributions
 sturdy-0.001
 exc-0.001
 bleed-0.001
 flat-0.001
 sew-0.001
Token Z
Token ID1168
Feature activation+8.70e+00

Pos logit contributions
 sturdy+0.029
 exc+0.029
 bend+0.027
 flat+0.027
 heal+0.026
Neg logit contributions
ople-0.044
ipped-0.043
pa-0.040
father-0.040
eled-0.040
Tokenipper
Token ID14710
Feature activation+1.76e+00

Pos logit contributions
ople+0.167
father+0.164
mother+0.152
years+0.151
 End+0.148
Neg logit contributions
 exc-0.203
 boun-0.202
 tie-0.196
 fold-0.196
onna-0.195
Token!
Token ID0
Feature activation+8.60e-02

Pos logit contributions
ople+0.054
ipped+0.052
pa+0.051
father+0.051
eled+0.050
Neg logit contributions
 sturdy-0.030
 exc-0.029
 flat-0.028
 heal-0.027
 bend-0.027
Token I
Token ID314
Feature activation+1.69e-01

Pos logit contributions
ople+0.002
ipped+0.002
pa+0.002
father+0.002
ie+0.002
Neg logit contributions
 sturdy-0.002
 exc-0.002
 heal-0.002
 bend-0.002
 flat-0.002
Token promise
Token ID6991
Feature activation+3.01e-01

Pos logit contributions
ople+0.005
ipped+0.004
pa+0.004
ie+0.004
atted+0.004
Neg logit contributions
 sturdy-0.003
 exc-0.003
 flat-0.003
 bend-0.003
onna-0.003
Token I
Token ID314
Feature activation+3.35e-02

Pos logit contributions
ople+0.009
pa+0.008
ipped+0.008
ie+0.008
ed+0.008
Neg logit contributions
 sturdy-0.006
 exc-0.006
 flat-0.005
 bleed-0.005
 bend-0.005
Token using
Token ID1262
Feature activation+2.46e+00

Pos logit contributions
ople+0.032
ipped+0.031
pa+0.030
father+0.029
ie+0.029
Neg logit contributions
 sturdy-0.026
 exc-0.025
 flat-0.024
 heal-0.024
 bend-0.024
Token the
Token ID262
Feature activation+1.52e+00

Pos logit contributions
ople+0.063
ipped+0.061
pa+0.060
el+0.058
father+0.058
Neg logit contributions
 sturdy-0.052
 exc-0.052
 bleed-0.050
 heal-0.049
onna-0.048
Token ladder
Token ID18002
Feature activation+1.84e-02

Pos logit contributions
ople+0.043
ipped+0.041
pa+0.040
father+0.039
ie+0.039
Neg logit contributions
 sturdy-0.030
 exc-0.029
 bleed-0.028
 heal-0.027
 flat-0.027
Token,
Token ID11
Feature activation-1.96e+00

Pos logit contributions
ople+0.001
ipped+0.001
pa+0.001
father+0.001
ebra+0.000
Neg logit contributions
 sturdy-0.000
 exc-0.000
 flat-0.000
 heal-0.000
 bend-0.000
Token then
Token ID788
Feature activation+4.17e-01

Pos logit contributions
 sturdy+0.034
 exc+0.033
 flat+0.031
 bleed+0.031
 heal+0.031
Neg logit contributions
ople-0.059
ipped-0.057
pa-0.055
father-0.054
ie-0.054
Token R
Token ID371
Feature activation+8.68e+00

Pos logit contributions
ople+0.012
ipped+0.012
pa+0.012
father+0.011
ebra+0.011
Neg logit contributions
 sturdy-0.008
 exc-0.007
 bend-0.007
 flat-0.007
 heal-0.007
Tokenuf
Token ID3046
Feature activation+6.70e+00

Pos logit contributions
ople+0.213
ipped+0.207
aked+0.196
ed+0.196
atted+0.194
Neg logit contributions
onna-0.192
 exc-0.182
 heal-0.172
 bend-0.158
 sew-0.157
Tokenus
Token ID385
Feature activation+2.42e+00

Pos logit contributions
ople+0.113
ipped+0.109
eled+0.099
atted+0.097
pa+0.094
Neg logit contributions
onna-0.205
 exc-0.191
 heal-0.187
 opener-0.182
 sturdy-0.180
Token can
Token ID460
Feature activation-2.96e+00

Pos logit contributions
ople+0.064
ipped+0.061
father+0.060
pa+0.059
ebra+0.059
Neg logit contributions
 exc-0.050
 sturdy-0.050
 heal-0.046
 flat-0.046
 bend-0.046
Token come
Token ID1282
Feature activation+5.70e-01

Pos logit contributions
 sturdy+0.056
 boun+0.049
 flat+0.049
 sil+0.049
 cozy+0.049
Neg logit contributions
el-0.084
ople-0.082
pa-0.081
ie-0.080
or-0.078
Token over
Token ID625
Feature activation+2.27e+00

Pos logit contributions
ople+0.013
ipped+0.013
ie+0.013
pa+0.013
el+0.013
Neg logit contributions
 sturdy-0.013
 exc-0.013
 bleed-0.013
onna-0.012
 crunch-0.012
Token than
Token ID621
Feature activation+8.91e-02

Pos logit contributions
ople+0.009
ipped+0.009
pa+0.009
father+0.008
ie+0.008
Neg logit contributions
 sturdy-0.006
 exc-0.006
 flat-0.005
 heal-0.005
 bend-0.005
Token the
Token ID262
Feature activation+2.23e+00

Pos logit contributions
ople+0.003
ipped+0.003
ebra+0.003
father+0.003
pa+0.003
Neg logit contributions
 sturdy-0.002
 exc-0.002
 flat-0.002
 bend-0.001
onna-0.001
Token park
Token ID3952
Feature activation+1.08e+00

Pos logit contributions
ople+0.058
ipped+0.056
pa+0.054
father+0.053
atted+0.053
Neg logit contributions
 sturdy-0.050
 exc-0.049
 flat-0.047
 bend-0.046
 heal-0.045
Token.
Token ID13
Feature activation-5.73e-01

Pos logit contributions
ople+0.033
ipped+0.032
pa+0.031
father+0.031
atted+0.031
Neg logit contributions
 sturdy-0.019
 exc-0.018
 flat-0.017
 bend-0.017
 heal-0.017
Token K
Token ID509
Feature activation+7.62e+00

Pos logit contributions
 sturdy+0.011
 exc+0.011
 flat+0.010
 bend+0.010
 sil+0.010
Neg logit contributions
ople-0.017
ipped-0.016
pa-0.016
father-0.016
ebra-0.016
Tokenait
Token ID4548
Feature activation+8.68e+00

Pos logit contributions
ople+0.199
ipped+0.185
father+0.183
io+0.177
ed+0.176
Neg logit contributions
 exc-0.158
 sturdy-0.137
onna-0.132
 sil-0.126
 sew-0.125
Tokenlyn
Token ID6213
Feature activation+1.77e+00

Pos logit contributions
ipped+0.208
ople+0.199
joy+0.196
oken+0.195
atted+0.195
Neg logit contributions
 exc-0.155
 gently-0.150
 heal-0.140
 wrap-0.136
 sweep-0.134
Token ran
Token ID4966
Feature activation+7.80e-01

Pos logit contributions
ople+0.046
ipped+0.044
father+0.044
pa+0.043
ebra+0.042
Neg logit contributions
 exc-0.038
 sturdy-0.038
 bend-0.036
 flat-0.035
 boun-0.035
Token over
Token ID625
Feature activation+1.44e+00

Pos logit contributions
ople+0.022
ipped+0.021
pa+0.020
father+0.020
ebra+0.019
Neg logit contributions
 sturdy-0.016
 exc-0.015
 flat-0.015
 bend-0.015
 heal-0.015
Token to
Token ID284
Feature activation-3.75e-01

Pos logit contributions
ople+0.051
father+0.049
pa+0.048
ipped+0.048
ebra+0.047
Neg logit contributions
 sturdy-0.019
 bend-0.018
 flat-0.018
 boun-0.018
 exc-0.017
Token the
Token ID262
Feature activation+2.02e+00

Pos logit contributions
 sturdy+0.008
 exc+0.008
 bend+0.007
 flat+0.007
 heal+0.007
Neg logit contributions
ople-0.010
ipped-0.010
pa-0.009
father-0.009
ebra-0.009
Token again
Token ID757
Feature activation+2.04e+00

Pos logit contributions
ople+0.029
ipped+0.028
father+0.027
pa+0.027
eled+0.027
Neg logit contributions
 sturdy-0.016
 exc-0.015
 heal-0.015
 flat-0.014
 bend-0.014
Token.
Token ID13
Feature activation+1.18e-01

Pos logit contributions
ople+0.064
ipped+0.061
pa+0.059
father+0.059
atted+0.058
Neg logit contributions
 sturdy-0.035
 exc-0.034
 flat-0.032
 heal-0.031
 bend-0.031
Tokenâ
Token ID22940
Feature activation+5.32e+00

Pos logit contributions
ople+0.003
ipped+0.003
pa+0.003
father+0.003
atted+0.003
Neg logit contributions
 sturdy-0.003
 exc-0.002
 flat-0.002
 bleed-0.002
 bend-0.002
Token
Token ID26391
Feature activation+2.61e+00

Pos logit contributions
ople+0.123
ipped+0.118
father+0.118
eled+0.116
pa+0.115
Neg logit contributions
 sturdy-0.125
 heal-0.122
 exc-0.122
 sew-0.119
 flat-0.116
Token<|endoftext|>
Token ID50256
Feature activation+7.32e+00

Pos logit contributions
ople+0.072
ipped+0.068
pa+0.066
father+0.066
ie+0.064
Neg logit contributions
 sturdy-0.054
 exc-0.053
 flat-0.050
 bend-0.050
 heal-0.050
TokenK
Token ID42
Feature activation+8.65e+00

Pos logit contributions
ople+0.173
atted+0.160
father+0.156
pa+0.156
eled+0.151
Neg logit contributions
 opener-0.173
 boun-0.170
onna-0.166
 bend-0.165
 flat-0.163
Tokenitty
Token ID9760
Feature activation+3.90e+00

Pos logit contributions
ople+0.177
eled+0.151
iece+0.149
ed+0.145
father+0.145
Neg logit contributions
onna-0.230
 sew-0.204
 knit-0.194
 gently-0.193
 bend-0.193
Token was
Token ID373
Feature activation-2.68e+00

Pos logit contributions
ople+0.120
ipped+0.116
father+0.115
eled+0.111
or+0.110
Neg logit contributions
 sturdy-0.059
 exc-0.058
 bend-0.057
 sew-0.056
 boun-0.056
Token outside
Token ID2354
Feature activation-3.98e-01

Pos logit contributions
 sturdy+0.051
 exc+0.050
 flat+0.047
 bend+0.047
 heal+0.046
Neg logit contributions
ople-0.079
ipped-0.075
pa-0.073
father-0.072
ebra-0.071
Token in
Token ID287
Feature activation+9.80e-01

Pos logit contributions
 exc+0.008
 sturdy+0.008
 flat+0.008
 bend+0.008
 heal+0.008
Neg logit contributions
ople-0.011
ipped-0.010
father-0.010
pa-0.010
ebra-0.010
Token the
Token ID262
Feature activation+2.41e+00

Pos logit contributions
ople+0.029
ipped+0.027
io+0.026
or+0.026
ebra+0.026
Neg logit contributions
 sturdy-0.019
 bend-0.019
 exc-0.019
 flat-0.018
 cozy-0.018
Token many
Token ID867
Feature activation+1.06e+00

Pos logit contributions
ople+0.010
ipped+0.009
pa+0.009
father+0.009
ebra+0.009
Neg logit contributions
 sturdy-0.008
 exc-0.007
 flat-0.007
 heal-0.007
 bend-0.007
Token different
Token ID1180
Feature activation+8.14e-01

Pos logit contributions
ople+0.025
ipped+0.025
ebra+0.023
father+0.023
pa+0.023
Neg logit contributions
 sturdy-0.027
 exc-0.026
 bend-0.024
 flat-0.024
 boun-0.024
Token things
Token ID1243
Feature activation-2.51e+00

Pos logit contributions
ipped+0.018
ople+0.018
ebra+0.017
eled+0.016
father+0.016
Neg logit contributions
 sturdy-0.021
 exc-0.021
 bend-0.020
 flat-0.019
 heal-0.019
Token.
Token ID13
Feature activation-4.90e-01

Pos logit contributions
 sturdy+0.054
 exc+0.053
 bleed+0.052
 bend+0.049
 sew+0.049
Neg logit contributions
ople-0.065
ipped-0.061
pa-0.061
ie-0.060
ed-0.059
Token<|endoftext|>
Token ID50256
Feature activation+6.95e+00

Pos logit contributions
 exc+0.009
 sturdy+0.009
onna+0.008
 heal+0.008
 bend+0.008
Neg logit contributions
ople-0.014
ipped-0.014
pa-0.013
ie-0.013
les-0.013
TokenK
Token ID42
Feature activation+8.64e+00

Pos logit contributions
ople+0.231
atted+0.219
ipped+0.214
eled+0.212
father+0.211
Neg logit contributions
onna-0.100
 opener-0.100
 exc-0.097
 flat-0.095
 boun-0.094
Tokenathy
Token ID10036
Feature activation+4.27e+00

Pos logit contributions
ople+0.164
ed+0.149
iece+0.142
led+0.141
ipped+0.140
Neg logit contributions
onna-0.239
 exc-0.225
 sturdy-0.206
 flat-0.203
 branch-0.202
Token was
Token ID373
Feature activation-2.13e+00

Pos logit contributions
ople+0.125
ipped+0.124
ebra+0.120
father+0.120
atted+0.116
Neg logit contributions
 exc-0.069
 sew-0.068
 sturdy-0.066
 heal-0.065
 bend-0.064
Token a
Token ID257
Feature activation-4.36e+00

Pos logit contributions
 sturdy+0.041
 exc+0.040
 flat+0.038
 bend+0.038
 heal+0.037
Neg logit contributions
ople-0.062
ipped-0.059
pa-0.057
father-0.057
ebra-0.056
Token helpful
Token ID7613
Feature activation-1.11e+00

Pos logit contributions
 sturdy+0.090
 exc+0.086
 bleed+0.086
 heal+0.085
onna+0.085
Neg logit contributions
ople-0.116
ipped-0.110
pa-0.109
father-0.107
ebra-0.107
Token girl
Token ID2576
Feature activation-1.12e+00

Pos logit contributions
 sturdy+0.027
 exc+0.027
 bend+0.025
 flat+0.025
 boun+0.025
Neg logit contributions
ople-0.027
ipped-0.026
father-0.024
pa-0.024
atted-0.024
Token,
Token ID11
Feature activation-1.76e+00

Pos logit contributions
ople+0.040
ipped+0.038
pa+0.037
father+0.037
ebra+0.036
Neg logit contributions
 sturdy-0.028
 exc-0.027
 flat-0.026
 heal-0.026
 bend-0.026
Token "
Token ID366
Feature activation+5.39e-01

Pos logit contributions
 sturdy+0.034
 bleed+0.033
 exc+0.030
 flat+0.029
 heal+0.029
Neg logit contributions
ople-0.047
ipped-0.047
pa-0.046
atted-0.045
ed-0.045
TokenThank
Token ID10449
Feature activation+1.70e+00

Pos logit contributions
ople+0.016
ipped+0.016
pa+0.015
atted+0.015
father+0.015
Neg logit contributions
 sturdy-0.009
 exc-0.009
 bend-0.009
 sew-0.009
 heal-0.009
Token you
Token ID345
Feature activation+1.45e+00

Pos logit contributions
eled+0.053
ople+0.053
atted+0.051
les+0.050
ipped+0.050
Neg logit contributions
 exc-0.025
 heal-0.024
 bend-0.023
 boun-0.023
 sturdy-0.023
Token,
Token ID11
Feature activation-1.43e+00

Pos logit contributions
ople+0.042
ipped+0.041
eled+0.040
ebra+0.038
atted+0.038
Neg logit contributions
 sturdy-0.028
 exc-0.027
 heal-0.026
 crunch-0.026
 flat-0.026
Token Z
Token ID1168
Feature activation+8.63e+00

Pos logit contributions
 sturdy+0.031
 bleed+0.030
 exc+0.029
 flat+0.029
 heal+0.029
Neg logit contributions
ople-0.037
pa-0.035
ipped-0.034
ie-0.034
father-0.034
Tokenog
Token ID519
Feature activation+4.67e+00

Pos logit contributions
eled+0.154
ople+0.146
father+0.138
years+0.136
 nights+0.131
Neg logit contributions
onna-0.226
 exc-0.207
 bend-0.204
 tie-0.204
 boun-0.202
Token.
Token ID13
Feature activation-4.68e-01

Pos logit contributions
ople+0.121
ipped+0.116
eled+0.114
father+0.109
atted+0.108
Neg logit contributions
 sturdy-0.100
 exc-0.097
 heal-0.096
 sew-0.093
 flat-0.093
Token You
Token ID921
Feature activation+1.51e+00

Pos logit contributions
 exc+0.009
 sturdy+0.009
 heal+0.009
 flat+0.009
 bend+0.008
Neg logit contributions
ople-0.013
ipped-0.013
pa-0.012
ie-0.012
father-0.012
Token are
Token ID389
Feature activation-1.70e+00

Pos logit contributions
ople+0.035
ipped+0.033
pa+0.032
father+0.031
ie+0.031
Neg logit contributions
 sturdy-0.038
 exc-0.037
 flat-0.036
onna-0.035
 bend-0.035
Token very
Token ID845
Feature activation-1.50e+00

Pos logit contributions
 sturdy+0.035
 exc+0.034
onna+0.034
 bleed+0.033
 heal+0.033
Neg logit contributions
ople-0.045
pa-0.042
father-0.042
ipped-0.042
mother-0.041
Token very
Token ID845
Feature activation-2.44e+00

Pos logit contributions
 sturdy+0.017
 exc+0.016
 bleed+0.016
onna+0.015
 bend+0.015
Neg logit contributions
ople-0.022
pa-0.022
ie-0.022
ipped-0.021
father-0.021
Token much
Token ID881
Feature activation-1.80e+00

Pos logit contributions
 sturdy+0.029
 exc+0.029
 flat+0.026
 bend+0.025
 heal+0.025
Neg logit contributions
ople-0.090
ipped-0.087
pa-0.084
father-0.083
eled-0.083
Token.
Token ID13
Feature activation-1.50e+00

Pos logit contributions
 sturdy+0.044
 bleed+0.043
 exc+0.043
 bend+0.041
 sew+0.041
Neg logit contributions
ople-0.040
ie-0.038
pa-0.038
les-0.037
ipped-0.037
Token It
Token ID632
Feature activation-1.12e+00

Pos logit contributions
 sturdy+0.033
 exc+0.032
 bleed+0.031
 flat+0.030
 heal+0.030
Neg logit contributions
ople-0.039
ipped-0.037
pa-0.036
ie-0.035
father-0.035
Token was
Token ID373
Feature activation-4.14e+00

Pos logit contributions
 sturdy+0.033
 exc+0.032
 flat+0.031
 bleed+0.031
onna+0.031
Neg logit contributions
ople-0.020
ipped-0.018
ie-0.018
pa-0.018
father-0.017
Token a
Token ID257
Feature activation-7.13e+00

Pos logit contributions
 sturdy+0.087
 bleed+0.086
 exc+0.084
 heal+0.083
onna+0.082
Neg logit contributions
ie-0.099
pa-0.097
ople-0.096
el-0.096
father-0.095
Token cute
Token ID13779
Feature activation-2.35e+00

Pos logit contributions
 bleed+0.167
uts+0.151
onna+0.149
 opener+0.147
 distract+0.145
Neg logit contributions
 End-0.149
by-0.143
ebra-0.137
ie-0.136
ma-0.133
Token,
Token ID11
Feature activation-4.04e+00

Pos logit contributions
 sturdy+0.051
 exc+0.050
 bleed+0.048
onna+0.047
 heal+0.046
Neg logit contributions
ople-0.060
ipped-0.058
ie-0.057
pa-0.057
ebra-0.056
Token intelligent
Token ID12661
Feature activation-1.25e+00

Pos logit contributions
onna+0.069
 bleed+0.067
 opener+0.065
 exc+0.064
 sturdy+0.061
Neg logit contributions
ie-0.101
pa-0.097
el-0.096
elle-0.096
ople-0.093
Token t
Token ID256
Feature activation-1.90e-02

Pos logit contributions
 sturdy+0.028
 exc+0.027
 bleed+0.027
onna+0.026
 heal+0.026
Neg logit contributions
ople-0.031
pa-0.030
ebra-0.030
ipped-0.030
ie-0.029
Tokeneddy
Token ID21874
Feature activation-1.94e-01

Pos logit contributions
 exc+0.000
 sturdy+0.000
onna+0.000
 flat+0.000
 heal+0.000
Neg logit contributions
ople-0.000
ipped-0.000
pa-0.000
atted-0.000
 End-0.000
Token very
Token ID845
Feature activation-2.31e+00

Pos logit contributions
 sturdy+0.016
 exc+0.015
 bleed+0.014
 bend+0.014
 heal+0.014
Neg logit contributions
ople-0.022
pa-0.022
ie-0.022
ipped-0.021
father-0.021
Token much
Token ID881
Feature activation-1.65e+00

Pos logit contributions
 exc+0.027
 sturdy+0.027
 flat+0.024
 bend+0.023
 heal+0.023
Neg logit contributions
ople-0.086
ipped-0.083
pa-0.080
father-0.079
eled-0.079
Token.
Token ID13
Feature activation-1.67e+00

Pos logit contributions
 sturdy+0.039
 exc+0.038
 bleed+0.037
 bend+0.036
 flat+0.036
Neg logit contributions
ople-0.039
pa-0.037
ie-0.037
les-0.036
ipped-0.036
Token It
Token ID632
Feature activation-1.09e+00

Pos logit contributions
 sturdy+0.037
 exc+0.036
 bleed+0.035
 flat+0.033
 heal+0.033
Neg logit contributions
ople-0.043
ipped-0.041
pa-0.041
ie-0.040
father-0.040
Token was
Token ID373
Feature activation-4.04e+00

Pos logit contributions
 sturdy+0.032
 exc+0.031
 flat+0.030
 bleed+0.030
onna+0.030
Neg logit contributions
ople-0.020
ipped-0.019
ie-0.018
pa-0.018
father-0.018
Token a
Token ID257
Feature activation-7.12e+00

Pos logit contributions
 sturdy+0.085
 bleed+0.084
 exc+0.083
 heal+0.081
onna+0.080
Neg logit contributions
ie-0.096
pa-0.095
ople-0.094
el-0.093
father-0.093
Token t
Token ID256
Feature activation-8.08e-01

Pos logit contributions
 bleed+0.167
uts+0.149
onna+0.147
 opener+0.145
 distract+0.144
Neg logit contributions
 End-0.149
by-0.143
ebra-0.138
ie-0.136
ma-0.134
Tokeneddy
Token ID21874
Feature activation-3.57e-01

Pos logit contributions
 exc+0.019
 sturdy+0.019
 flat+0.018
onna+0.018
 bend+0.018
Neg logit contributions
ople-0.020
ipped-0.018
pa-0.018
atted-0.017
father-0.017
Token bear
Token ID6842
Feature activation-2.97e+00

Pos logit contributions
 sturdy+0.009
 exc+0.009
 flat+0.009
 bend+0.009
 heal+0.009
Neg logit contributions
ople-0.008
ipped-0.008
pa-0.007
father-0.007
ie-0.007
Token with
Token ID351
Feature activation-6.55e-01

Pos logit contributions
 sturdy+0.070
 exc+0.068
 bleed+0.065
 flat+0.063
onna+0.063
Neg logit contributions
ople-0.071
ie-0.068
pa-0.068
ipped-0.067
father-0.067
Token soft
Token ID2705
Feature activation-5.86e-01

Pos logit contributions
 sturdy+0.010
 exc+0.010
 flat+0.010
 crunch+0.010
 bend+0.010
Neg logit contributions
ople-0.022
ipped-0.021
pa-0.020
ebra-0.020
or-0.020
Token very
Token ID845
Feature activation-2.47e+00

Pos logit contributions
 sturdy+0.017
 exc+0.016
 bleed+0.015
 bend+0.015
onna+0.015
Neg logit contributions
ople-0.022
pa-0.022
ie-0.022
ipped-0.021
father-0.021
Token much
Token ID881
Feature activation-1.79e+00

Pos logit contributions
 sturdy+0.029
 exc+0.029
 flat+0.026
 bend+0.025
 heal+0.025
Neg logit contributions
ople-0.091
ipped-0.088
pa-0.085
father-0.084
eled-0.084
Token.
Token ID13
Feature activation-1.61e+00

Pos logit contributions
 sturdy+0.044
 bleed+0.043
 exc+0.043
 sew+0.041
 bend+0.041
Neg logit contributions
ople-0.040
pa-0.038
ie-0.038
les-0.037
ipped-0.037
Token It
Token ID632
Feature activation-1.17e+00

Pos logit contributions
 sturdy+0.037
 exc+0.035
 bleed+0.035
 flat+0.033
 heal+0.033
Neg logit contributions
ople-0.040
ipped-0.038
pa-0.038
el-0.037
father-0.037
Token was
Token ID373
Feature activation-4.10e+00

Pos logit contributions
 sturdy+0.034
 exc+0.033
 flat+0.032
 bleed+0.032
onna+0.032
Neg logit contributions
ople-0.022
ipped-0.020
pa-0.020
ie-0.020
father-0.019
Token a
Token ID257
Feature activation-7.08e+00

Pos logit contributions
 sturdy+0.087
 bleed+0.085
 exc+0.084
 heal+0.082
onna+0.081
Neg logit contributions
ie-0.097
pa-0.096
ople-0.095
el-0.095
father-0.094
Token fragile
Token ID21049
Feature activation-2.85e+00

Pos logit contributions
 bleed+0.165
uts+0.147
onna+0.146
 opener+0.145
 distract+0.142
Neg logit contributions
 End-0.148
by-0.145
ebra-0.138
ie-0.136
ma-0.134
Token doll
Token ID3654
Feature activation-4.24e+00

Pos logit contributions
 sturdy+0.065
 exc+0.064
 bleed+0.062
onna+0.061
 heal+0.059
Neg logit contributions
ie-0.065
ebra-0.064
ople-0.064
ipped-0.063
el-0.063
Token that
Token ID326
Feature activation-1.06e+00

Pos logit contributions
 sturdy+0.099
 exc+0.094
 bleed+0.094
onna+0.092
 flat+0.090
Neg logit contributions
ople-0.102
ie-0.099
pa-0.097
mother-0.096
father-0.096
Token she
Token ID673
Feature activation+2.88e-01

Pos logit contributions
 sturdy+0.021
 exc+0.020
 flat+0.019
 bleed+0.019
 heal+0.019
Neg logit contributions
ople-0.029
ipped-0.028
pa-0.028
father-0.027
ie-0.027
Token always
Token ID1464
Feature activation-1.19e+00

Pos logit contributions
ople+0.007
ipped+0.007
pa+0.007
father+0.007
ie+0.007
Neg logit contributions
 sturdy-0.007
 exc-0.006
 flat-0.006
 heal-0.006
 bend-0.006
Token very
Token ID845
Feature activation-2.41e+00

Pos logit contributions
 sturdy+0.013
 exc+0.012
 bleed+0.011
 bend+0.011
onna+0.011
Neg logit contributions
ople-0.018
pa-0.018
ie-0.017
ipped-0.017
father-0.017
Token much
Token ID881
Feature activation-1.63e+00

Pos logit contributions
 exc+0.028
 sturdy+0.028
 flat+0.025
 bend+0.024
 heal+0.024
Neg logit contributions
ople-0.090
ipped-0.087
pa-0.084
father-0.083
eled-0.083
Token.
Token ID13
Feature activation-1.64e+00

Pos logit contributions
 sturdy+0.039
 exc+0.038
 bleed+0.037
 bend+0.037
 sew+0.036
Neg logit contributions
ople-0.038
pa-0.036
ie-0.036
les-0.035
ipped-0.035
Token It
Token ID632
Feature activation-1.15e+00

Pos logit contributions
 sturdy+0.035
 exc+0.034
 bleed+0.033
 flat+0.032
 heal+0.032
Neg logit contributions
ople-0.044
ipped-0.042
pa-0.041
father-0.040
ie-0.040
Token was
Token ID373
Feature activation-3.90e+00

Pos logit contributions
 sturdy+0.031
 exc+0.029
 flat+0.028
onna+0.028
 bleed+0.028
Neg logit contributions
ople-0.024
ipped-0.023
pa-0.022
ie-0.022
father-0.022
Token a
Token ID257
Feature activation-7.05e+00

Pos logit contributions
 sturdy+0.078
 exc+0.077
 bleed+0.075
 heal+0.074
onna+0.073
Neg logit contributions
ie-0.097
ople-0.096
pa-0.095
el-0.094
father-0.094
Token small
Token ID1402
Feature activation-4.11e+00

Pos logit contributions
 bleed+0.158
uts+0.144
 opener+0.143
onna+0.143
 distract+0.141
Neg logit contributions
 End-0.148
by-0.143
ie-0.138
ebra-0.136
mother-0.133
Token,
Token ID11
Feature activation-4.24e+00

Pos logit contributions
 sturdy+0.091
 exc+0.089
 bleed+0.086
onna+0.085
 heal+0.084
Neg logit contributions
ople-0.099
ie-0.096
pa-0.096
ipped-0.095
el-0.095
Token pretty
Token ID2495
Feature activation-2.34e+00

Pos logit contributions
onna+0.075
 opener+0.071
 bleed+0.071
 exc+0.070
comb+0.069
Neg logit contributions
ie-0.104
el-0.102
pa-0.098
elle-0.097
ople-0.095
Token ball
Token ID2613
Feature activation-4.56e+00

Pos logit contributions
 sturdy+0.048
 exc+0.048
 bleed+0.046
onna+0.044
 heal+0.044
Neg logit contributions
ople-0.061
ie-0.059
ipped-0.059
ebra-0.059
pa-0.058
Tokener
Token ID263
Feature activation-2.07e+00

Pos logit contributions
 sturdy+0.120
 exc+0.116
 bleed+0.116
onna+0.111
 heal+0.110
Neg logit contributions
ie-0.095
ople-0.094
ipped-0.089
ing-0.088
el-0.087
Token to
Token ID284
Feature activation-2.58e+00

Pos logit contributions
 sturdy+0.005
 exc+0.005
 bleed+0.005
 heal+0.005
 flat+0.005
Neg logit contributions
ople-0.009
ipped-0.008
pa-0.008
ie-0.008
father-0.008
Token wear
Token ID5806
Feature activation-8.66e-01

Pos logit contributions
 sturdy+0.053
 bleed+0.048
 exc+0.046
 flat+0.046
onna+0.045
Neg logit contributions
ie-0.070
ipped-0.068
les-0.066
ople-0.066
pa-0.065
Token.
Token ID13
Feature activation-1.71e+00

Pos logit contributions
 sturdy+0.020
 exc+0.020
 bleed+0.020
onna+0.019
 heal+0.018
Neg logit contributions
ople-0.021
ipped-0.020
les-0.020
ie-0.020
pa-0.020
Token It
Token ID632
Feature activation-7.66e-01

Pos logit contributions
 sturdy+0.039
 exc+0.038
 bleed+0.037
 flat+0.036
 heal+0.035
Neg logit contributions
ople-0.043
ipped-0.041
pa-0.040
ie-0.039
father-0.039
Token was
Token ID373
Feature activation-3.90e+00

Pos logit contributions
 sturdy+0.021
 exc+0.020
 flat+0.019
 bend+0.019
 heal+0.019
Neg logit contributions
ople-0.016
ipped-0.015
pa-0.015
father-0.015
atted-0.014
Token a
Token ID257
Feature activation-7.04e+00

Pos logit contributions
 sturdy+0.077
 bleed+0.076
 exc+0.075
 heal+0.074
onna+0.072
Neg logit contributions
ie-0.098
ople-0.097
pa-0.097
ipped-0.095
el-0.094
Token flexible
Token ID12846
Feature activation-2.45e+00

Pos logit contributions
 bleed+0.163
 opener+0.146
 distract+0.143
onna+0.142
 heal+0.142
Neg logit contributions
 End-0.150
by-0.147
ebra-0.144
ie-0.141
 morning-0.134
Token skirt
Token ID23967
Feature activation-5.86e+00

Pos logit contributions
 sturdy+0.049
 exc+0.048
 bleed+0.046
 heal+0.045
onna+0.045
Neg logit contributions
ople-0.067
ipped-0.064
ie-0.063
pa-0.062
ebra-0.062
Token that
Token ID326
Feature activation-1.14e+00

Pos logit contributions
 bleed+0.151
 sturdy+0.151
 exc+0.149
onna+0.143
 twitch+0.138
Neg logit contributions
ople-0.120
ie-0.118
ed-0.116
ipped-0.113
ing-0.113
Token moved
Token ID3888
Feature activation+2.91e-02

Pos logit contributions
 sturdy+0.021
 exc+0.020
 flat+0.019
 heal+0.019
 bleed+0.019
Neg logit contributions
ople-0.033
ipped-0.032
pa-0.031
father-0.030
ie-0.030
Token when
Token ID618
Feature activation-3.54e-02

Pos logit contributions
ople+0.001
ipped+0.001
pa+0.001
father+0.001
eled+0.001
Neg logit contributions
 sturdy-0.001
 flat-0.001
 exc-0.000
 bend-0.000
 boun-0.000
Token very
Token ID845
Feature activation-2.32e+00

Pos logit contributions
 sturdy+0.016
 exc+0.015
 bleed+0.015
 heal+0.014
 bend+0.014
Neg logit contributions
ople-0.023
pa-0.023
ie-0.022
ipped-0.022
father-0.022
Token much
Token ID881
Feature activation-1.72e+00

Pos logit contributions
 exc+0.027
 sturdy+0.027
 flat+0.024
 bend+0.023
 heal+0.023
Neg logit contributions
ople-0.086
ipped-0.083
pa-0.081
father-0.080
eled-0.080
Token.
Token ID13
Feature activation-1.66e+00

Pos logit contributions
 sturdy+0.041
 exc+0.040
 bleed+0.039
 bend+0.038
 sew+0.038
Neg logit contributions
ople-0.040
pa-0.038
ie-0.038
les-0.037
ipped-0.037
Token It
Token ID632
Feature activation-1.04e+00

Pos logit contributions
 sturdy+0.036
 exc+0.035
 bleed+0.034
 flat+0.033
 heal+0.032
Neg logit contributions
ople-0.043
ipped-0.041
pa-0.041
ie-0.040
father-0.040
Token was
Token ID373
Feature activation-4.09e+00

Pos logit contributions
 sturdy+0.030
 exc+0.029
 flat+0.028
 bleed+0.028
 heal+0.028
Neg logit contributions
ople-0.020
ipped-0.019
pa-0.018
ie-0.018
father-0.018
Token a
Token ID257
Feature activation-7.03e+00

Pos logit contributions
 sturdy+0.084
 bleed+0.083
 exc+0.082
 heal+0.081
onna+0.079
Neg logit contributions
ie-0.099
pa-0.098
ople-0.098
el-0.096
father-0.096
Token big
Token ID1263
Feature activation-4.06e+00

Pos logit contributions
 bleed+0.163
onna+0.144
uts+0.143
 opener+0.143
 distract+0.139
Neg logit contributions
 End-0.151
by-0.146
ebra-0.142
ie-0.138
ma-0.134
Token,
Token ID11
Feature activation-4.43e+00

Pos logit contributions
 sturdy+0.098
 exc+0.096
 bleed+0.095
onna+0.093
 heal+0.092
Neg logit contributions
ople-0.093
pa-0.091
ie-0.088
ipped-0.087
ebra-0.087
Token pink
Token ID11398
Feature activation-1.98e+00

Pos logit contributions
onna+0.083
 bleed+0.080
 opener+0.077
 exc+0.075
comb+0.075
Neg logit contributions
ie-0.104
el-0.102
pa-0.099
elle-0.097
ma-0.096
Token t
Token ID256
Feature activation-2.28e-01

Pos logit contributions
 sturdy+0.043
 exc+0.042
 bleed+0.040
 heal+0.039
onna+0.039
Neg logit contributions
ople-0.052
ipped-0.050
ebra-0.049
ie-0.049
pa-0.049
Tokeneddy
Token ID21874
Feature activation-5.68e-01

Pos logit contributions
 exc+0.005
 sturdy+0.005
onna+0.005
 flat+0.005
 bend+0.005
Neg logit contributions
ople-0.006
ipped-0.005
pa-0.005
atted-0.005
father-0.005
Token loved
Token ID6151
Feature activation-7.02e-01

Pos logit contributions
ople+0.006
ipped+0.006
pa+0.006
father+0.005
ie+0.005
Neg logit contributions
 sturdy-0.005
 exc-0.005
 flat-0.005
 bend-0.005
 heal-0.005
Token using
Token ID1262
Feature activation-1.24e-01

Pos logit contributions
 sturdy+0.014
 exc+0.013
 bleed+0.013
onna+0.012
 bend+0.012
Neg logit contributions
ople-0.020
pa-0.019
ie-0.019
ipped-0.019
father-0.019
Token.
Token ID13
Feature activation-8.00e-01

Pos logit contributions
 exc+0.003
 sturdy+0.003
onna+0.003
 bleed+0.003
 heal+0.003
Neg logit contributions
ople-0.003
ipped-0.003
pa-0.003
el-0.003
ie-0.003
Token It
Token ID632
Feature activation-9.39e-01

Pos logit contributions
 sturdy+0.018
 exc+0.017
 flat+0.017
 bleed+0.017
 heal+0.016
Neg logit contributions
ople-0.020
ipped-0.019
pa-0.019
father-0.019
el-0.018
Token was
Token ID373
Feature activation-3.87e+00

Pos logit contributions
 sturdy+0.026
 exc+0.025
 flat+0.024
 heal+0.024
 bend+0.024
Neg logit contributions
ople-0.019
ipped-0.018
pa-0.017
father-0.017
ie-0.017
Token a
Token ID257
Feature activation-7.03e+00

Pos logit contributions
 sturdy+0.079
 exc+0.078
 bleed+0.077
 heal+0.075
onna+0.073
Neg logit contributions
ople-0.098
pa-0.096
ie-0.095
ipped-0.095
father-0.094
Token lamp
Token ID20450
Feature activation-4.16e+00

Pos logit contributions
 bleed+0.162
uts+0.144
onna+0.144
 opener+0.141
 distract+0.136
Neg logit contributions
by-0.149
 End-0.149
ebra-0.142
ie-0.137
mother-0.136
Token with
Token ID351
Feature activation-4.46e-01

Pos logit contributions
 sturdy+0.089
 exc+0.087
 bleed+0.083
 flat+0.083
 bend+0.082
Neg logit contributions
ople-0.110
ipped-0.105
pa-0.102
father-0.101
ie-0.100
Token a
Token ID257
Feature activation-5.58e+00

Pos logit contributions
 sturdy+0.008
 exc+0.007
 flat+0.007
 crunch+0.007
 bend+0.007
Neg logit contributions
ople-0.015
ipped-0.014
ebra-0.014
pa-0.014
father-0.013
Token pink
Token ID11398
Feature activation-2.69e+00

Pos logit contributions
onna+0.116
 opener+0.114
 exc+0.113
 bleed+0.111
 distract+0.110
Neg logit contributions
el-0.120
 End-0.109
ebra-0.107
les-0.106
ing-0.103
Token top
Token ID1353
Feature activation-3.38e+00

Pos logit contributions
 sturdy+0.057
 exc+0.056
 bleed+0.051
onna+0.051
 nutritious+0.048
Neg logit contributions
el-0.071
ipped-0.069
ebra-0.068
les-0.067
ie-0.067
Token a
Token ID257
Feature activation-4.48e+00

Pos logit contributions
 sturdy+0.014
 exc+0.013
 bleed+0.013
onna+0.013
 heal+0.012
Neg logit contributions
ople-0.017
ipped-0.017
pa-0.017
father-0.016
ie-0.016
Token lot
Token ID1256
Feature activation-1.05e+00

Pos logit contributions
 bleed+0.120
 exc+0.115
 sturdy+0.111
 distract+0.109
onna+0.109
Neg logit contributions
by-0.088
ebra-0.074
ie-0.074
les-0.073
or-0.071
Token.
Token ID13
Feature activation-1.51e+00

Pos logit contributions
 sturdy+0.024
 exc+0.023
 flat+0.022
 heal+0.022
 bend+0.022
Neg logit contributions
ople-0.027
ipped-0.025
pa-0.025
father-0.024
ie-0.024
Token It
Token ID632
Feature activation-1.17e+00

Pos logit contributions
 sturdy+0.036
 bleed+0.035
 exc+0.034
 heal+0.033
 sew+0.032
Neg logit contributions
ople-0.035
ipped-0.034
pa-0.033
el-0.032
father-0.032
Token was
Token ID373
Feature activation-4.27e+00

Pos logit contributions
 sturdy+0.036
 bleed+0.034
 flat+0.034
 exc+0.034
onna+0.034
Neg logit contributions
ople-0.019
ie-0.018
ipped-0.017
pa-0.017
father-0.017
Token a
Token ID257
Feature activation-6.99e+00

Pos logit contributions
 bleed+0.094
 heal+0.089
 sturdy+0.089
onna+0.088
 exc+0.087
Neg logit contributions
pa-0.095
ie-0.094
el-0.093
father-0.092
ople-0.089
Token red
Token ID2266
Feature activation-2.82e+00

Pos logit contributions
 bleed+0.162
uts+0.143
onna+0.143
 heal+0.141
 opener+0.139
Neg logit contributions
by-0.143
ebra-0.139
 End-0.137
ma-0.125
 morning-0.124
Token car
Token ID1097
Feature activation-4.54e+00

Pos logit contributions
 sturdy+0.064
 exc+0.062
 bleed+0.060
onna+0.059
 heal+0.057
Neg logit contributions
ipped-0.070
ebra-0.068
el-0.068
ie-0.067
ople-0.067
Token that
Token ID326
Feature activation-9.83e-01

Pos logit contributions
 sturdy+0.102
 bleed+0.097
 exc+0.097
onna+0.096
 heal+0.091
Neg logit contributions
ople-0.115
ipped-0.107
pa-0.107
ie-0.106
father-0.105
Token went
Token ID1816
Feature activation-9.73e-02

Pos logit contributions
 sturdy+0.020
 exc+0.019
 flat+0.018
 bend+0.018
 heal+0.018
Neg logit contributions
ople-0.028
ipped-0.027
pa-0.026
father-0.025
ie-0.025
Token fast
Token ID3049
Feature activation-1.29e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
 flat+0.002
 bend+0.002
 heal+0.001
Neg logit contributions
ople-0.003
ipped-0.003
pa-0.003
father-0.003
eled-0.003
Token color
Token ID3124
Feature activation-1.48e+00

Pos logit contributions
 sturdy+0.026
 exc+0.025
 bleed+0.025
 boun+0.024
onna+0.024
Neg logit contributions
ople-0.055
pa-0.055
father-0.053
ie-0.053
el-0.052
Token in
Token ID287
Feature activation-5.48e-01

Pos logit contributions
 sturdy+0.030
 exc+0.029
 bend+0.029
 flat+0.029
 heal+0.028
Neg logit contributions
ople-0.042
ipped-0.040
pa-0.038
eled-0.038
father-0.037
Token and
Token ID290
Feature activation-3.21e+00

Pos logit contributions
 sturdy+0.010
 exc+0.010
 flat+0.009
 bend+0.009
 heal+0.009
Neg logit contributions
ople-0.017
ipped-0.017
atted-0.016
ebra-0.016
pa-0.015
Token it
Token ID340
Feature activation-1.64e+00

Pos logit contributions
 sturdy+0.054
 exc+0.052
 flat+0.049
 bleed+0.049
 heal+0.049
Neg logit contributions
ople-0.100
ipped-0.095
pa-0.094
father-0.092
ie-0.091
Token was
Token ID373
Feature activation-3.79e+00

Pos logit contributions
 sturdy+0.041
 exc+0.040
 flat+0.039
 bleed+0.039
 heal+0.038
Neg logit contributions
ople-0.037
ipped-0.035
pa-0.034
father-0.033
ie-0.033
Token a
Token ID257
Feature activation-6.95e+00

Pos logit contributions
 bleed+0.075
 heal+0.071
 exc+0.070
 sturdy+0.069
onna+0.068
Neg logit contributions
ie-0.096
ople-0.095
pa-0.094
father-0.093
el-0.091
Token great
Token ID1049
Feature activation-3.41e+00

Pos logit contributions
 bleed+0.158
 opener+0.138
onna+0.137
 heal+0.136
 distract+0.133
Neg logit contributions
by-0.159
 End-0.155
mother-0.149
father-0.144
ie-0.144
Token way
Token ID835
Feature activation-1.74e+00

Pos logit contributions
 bleed+0.074
 exc+0.071
 sturdy+0.071
 heal+0.067
onna+0.067
Neg logit contributions
ople-0.084
pa-0.083
by-0.081
ie-0.081
father-0.081
Token to
Token ID284
Feature activation-1.61e+00

Pos logit contributions
 sturdy+0.030
 exc+0.030
 bleed+0.028
 bend+0.027
 heal+0.027
Neg logit contributions
ople-0.053
ipped-0.051
pa-0.050
father-0.049
ie-0.049
Token pass
Token ID1208
Feature activation+1.43e+00

Pos logit contributions
 sturdy+0.023
 exc+0.022
 bleed+0.022
 boun+0.021
onna+0.021
Neg logit contributions
ople-0.052
pa-0.050
father-0.050
ipped-0.050
ie-0.049
Token the
Token ID262
Feature activation+1.54e+00

Pos logit contributions
ople+0.044
ipped+0.043
ebra+0.043
eled+0.041
atted+0.041
Neg logit contributions
 sturdy-0.025
 flat-0.024
 bend-0.024
 exc-0.023
 heal-0.022
Token the
Token ID262
Feature activation+1.40e+00

Pos logit contributions
ople+0.022
ipped+0.021
atted+0.020
father+0.020
pa+0.020
Neg logit contributions
 sturdy-0.020
 exc-0.020
 flat-0.019
 bend-0.019
 sew-0.018
Token snow
Token ID6729
Feature activation-6.51e-01

Pos logit contributions
ople+0.037
ipped+0.035
pa+0.033
atted+0.033
father+0.033
Neg logit contributions
 sturdy-0.032
 exc-0.031
 flat-0.031
 bend-0.030
 boun-0.030
Token.
Token ID13
Feature activation-2.42e+00

Pos logit contributions
 exc+0.009
 sturdy+0.009
 heal+0.009
 bend+0.009
 flat+0.009
Neg logit contributions
ople-0.021
ipped-0.021
atted-0.020
eled-0.020
father-0.020
Token They
Token ID1119
Feature activation-1.61e+00

Pos logit contributions
 sturdy+0.049
 exc+0.047
 bleed+0.045
 heal+0.045
 flat+0.045
Neg logit contributions
ople-0.067
ipped-0.065
pa-0.063
father-0.062
ie-0.061
Token made
Token ID925
Feature activation+1.12e-01

Pos logit contributions
 sturdy+0.039
 bleed+0.038
 heal+0.038
onna+0.036
 click+0.036
Neg logit contributions
ople-0.036
pa-0.036
ie-0.035
father-0.033
eled-0.033
Token a
Token ID257
Feature activation-6.94e+00

Pos logit contributions
ople+0.004
ipped+0.003
pa+0.003
father+0.003
ebra+0.003
Neg logit contributions
 sturdy-0.002
 exc-0.002
 flat-0.002
 bend-0.002
 crunch-0.002
Token big
Token ID1263
Feature activation-4.77e+00

Pos logit contributions
 bleed+0.156
 sew+0.151
 heal+0.149
 opener+0.148
uts+0.148
Neg logit contributions
by-0.155
ie-0.144
el-0.137
icer-0.134
 End-0.133
Token snow
Token ID6729
Feature activation-2.67e+00

Pos logit contributions
 exc+0.111
 sturdy+0.108
 bleed+0.107
onna+0.104
 heal+0.104
Neg logit contributions
pa-0.109
ie-0.108
el-0.106
ople-0.104
ebra-0.103
Tokenman
Token ID805
Feature activation-3.24e+00

Pos logit contributions
 sturdy+0.065
 exc+0.064
 flat+0.061
 bend+0.061
 heal+0.061
Neg logit contributions
ople-0.064
ipped-0.060
pa-0.058
father-0.057
atted-0.056
Token with
Token ID351
Feature activation-6.70e-01

Pos logit contributions
 sturdy+0.064
 exc+0.061
 bleed+0.059
onna+0.058
 flat+0.057
Neg logit contributions
ople-0.092
pa-0.087
ipped-0.087
ie-0.086
father-0.085
Token a
Token ID257
Feature activation-4.70e+00

Pos logit contributions
 sturdy+0.011
 flat+0.011
 exc+0.011
 crunch+0.011
 bend+0.010
Neg logit contributions
ople-0.022
ipped-0.021
or-0.020
ebra-0.020
ie-0.020
Token shovel
Token ID33677
Feature activation+2.98e-01

Pos logit contributions
 sturdy+0.071
 heal+0.070
 bleed+0.070
 exc+0.069
onna+0.067
Neg logit contributions
ople-0.101
pa-0.096
ipped-0.094
ie-0.091
atted-0.090
Tokens
Token ID82
Feature activation-1.41e+00

Pos logit contributions
ople+0.009
father+0.008
ipped+0.008
ebra+0.008
mother+0.008
Neg logit contributions
 flat-0.005
 sturdy-0.005
 exc-0.005
 bend-0.005
 boun-0.005
Token.
Token ID13
Feature activation-1.43e+00

Pos logit contributions
 sturdy+0.030
 exc+0.029
 heal+0.028
 bleed+0.027
onna+0.027
Neg logit contributions
ople-0.037
ipped-0.037
pa-0.035
ed-0.035
atted-0.034
Token They
Token ID1119
Feature activation-1.67e+00

Pos logit contributions
 sturdy+0.027
 exc+0.026
 heal+0.025
 bleed+0.025
 flat+0.025
Neg logit contributions
ople-0.042
ipped-0.040
pa-0.039
father-0.038
atted-0.038
Token made
Token ID925
Feature activation+1.64e-01

Pos logit contributions
 sturdy+0.040
 bleed+0.040
 heal+0.039
onna+0.038
 exc+0.038
Neg logit contributions
ople-0.039
pa-0.036
ipped-0.036
ie-0.035
eled-0.034
Token a
Token ID257
Feature activation-6.93e+00

Pos logit contributions
ople+0.005
ipped+0.005
pa+0.005
father+0.005
ebra+0.005
Neg logit contributions
 sturdy-0.003
 exc-0.003
 flat-0.003
 bend-0.003
 crunch-0.003
Token big
Token ID1263
Feature activation-4.84e+00

Pos logit contributions
 bleed+0.163
 heal+0.159
 sew+0.151
 exc+0.145
onna+0.143
Neg logit contributions
by-0.159
ie-0.140
andy-0.138
ebra-0.138
 End-0.138
Token castle
Token ID16669
Feature activation-4.38e+00

Pos logit contributions
 exc+0.108
 bleed+0.108
 heal+0.105
 sturdy+0.104
 sew+0.101
Neg logit contributions
ople-0.114
ie-0.113
pa-0.111
el-0.111
ipped-0.110
Token with
Token ID351
Feature activation-8.72e-01

Pos logit contributions
 sturdy+0.091
 exc+0.088
 bleed+0.088
 heal+0.084
onna+0.082
Neg logit contributions
ople-0.116
ipped-0.110
pa-0.107
ie-0.107
ed-0.106
Token towers
Token ID18028
Feature activation-3.50e+00

Pos logit contributions
 sturdy+0.018
 exc+0.018
 flat+0.017
 bend+0.017
 heal+0.016
Neg logit contributions
ople-0.024
ipped-0.023
pa-0.022
father-0.022
ebra-0.022
Token and
Token ID290
Feature activation-2.61e+00

Pos logit contributions
 sturdy+0.079
 exc+0.077
 heal+0.073
 bleed+0.073
 flat+0.072
Neg logit contributions
ople-0.090
ipped-0.086
pa-0.083
father-0.081
atted-0.080
Token a
Token ID257
Feature activation-5.70e+00

Pos logit contributions
ople+0.055
ipped+0.053
pa+0.051
father+0.051
ebra+0.051
Neg logit contributions
 sturdy-0.036
 exc-0.035
 flat-0.033
 bend-0.033
 heal-0.033
Token roof
Token ID9753
Feature activation-3.26e+00

Pos logit contributions
 bleed+0.114
 exc+0.108
onna+0.107
 heal+0.104
 sturdy+0.104
Neg logit contributions
by-0.143
ie-0.141
pa-0.140
father-0.135
mother-0.134
Token.
Token ID13
Feature activation-1.27e+00

Pos logit contributions
 sturdy+0.068
 exc+0.067
 bleed+0.064
 bend+0.063
 heal+0.063
Neg logit contributions
ople-0.087
ipped-0.082
pa-0.081
ie-0.081
father-0.078
Token It
Token ID632
Feature activation-6.31e-01

Pos logit contributions
 sturdy+0.030
 exc+0.030
 bleed+0.029
 heal+0.028
 flat+0.028
Neg logit contributions
ople-0.031
ipped-0.030
pa-0.028
eled-0.028
el-0.028
Token was
Token ID373
Feature activation-3.33e+00

Pos logit contributions
 sturdy+0.021
 bleed+0.021
 exc+0.020
onna+0.020
 heal+0.020
Neg logit contributions
ople-0.008
ie-0.008
el-0.007
pa-0.007
ipped-0.007
Token a
Token ID257
Feature activation-6.87e+00

Pos logit contributions
 bleed+0.089
 heal+0.085
onna+0.082
 sew+0.081
 exc+0.079
Neg logit contributions
el-0.063
pa-0.062
 End-0.059
ie-0.058
ma-0.056
Token pretty
Token ID2495
Feature activation-3.60e+00

Pos logit contributions
 bleed+0.179
uts+0.162
onna+0.155
 heal+0.155
 opener+0.152
Neg logit contributions
 End-0.132
ma-0.130
 morning-0.124
by-0.123
ie-0.123
Token roof
Token ID9753
Feature activation-2.77e+00

Pos logit contributions
 bleed+0.076
onna+0.070
 heal+0.069
 exc+0.069
 opener+0.067
Neg logit contributions
ie-0.095
el-0.090
ople-0.088
ebra-0.087
ipped-0.087
Token.
Token ID13
Feature activation-8.65e-01

Pos logit contributions
 sturdy+0.056
 exc+0.055
 flat+0.051
 heal+0.051
 bend+0.051
Neg logit contributions
ople-0.077
ipped-0.074
pa-0.072
father-0.071
ie-0.070
Token It
Token ID632
Feature activation-5.54e-01

Pos logit contributions
 sturdy+0.021
 exc+0.020
 bleed+0.020
 heal+0.019
 flat+0.019
Neg logit contributions
ople-0.021
ipped-0.020
pa-0.019
eled-0.019
el-0.019
Token was
Token ID373
Feature activation-3.46e+00

Pos logit contributions
 bleed+0.018
 sturdy+0.018
 exc+0.018
onna+0.018
 heal+0.017
Neg logit contributions
ople-0.008
ie-0.007
el-0.007
ipped-0.007
pa-0.007
Token the
Token ID262
Feature activation+1.38e+00

Pos logit contributions
ople+0.024
ipped+0.023
atted+0.021
father+0.021
ed+0.021
Neg logit contributions
 sturdy-0.020
 exc-0.020
 bend-0.019
 flat-0.019
 sew-0.019
Token snow
Token ID6729
Feature activation-5.84e-01

Pos logit contributions
ople+0.038
ipped+0.036
atted+0.034
pa+0.034
father+0.034
Neg logit contributions
 sturdy-0.031
 exc-0.029
 flat-0.029
 boun-0.028
 bend-0.028
Token.
Token ID13
Feature activation-2.20e+00

Pos logit contributions
 sturdy+0.009
 exc+0.009
 heal+0.009
 bend+0.008
 flat+0.008
Neg logit contributions
ople-0.019
ipped-0.018
atted-0.018
eled-0.018
father-0.018
Token They
Token ID1119
Feature activation-1.59e+00

Pos logit contributions
 sturdy+0.050
 exc+0.048
 bleed+0.048
 heal+0.046
 flat+0.046
Neg logit contributions
ople-0.056
ipped-0.052
pa-0.051
ie-0.050
el-0.050
Token made
Token ID925
Feature activation+2.52e-01

Pos logit contributions
 sturdy+0.039
 bleed+0.039
 heal+0.038
 flat+0.036
onna+0.036
Neg logit contributions
ople-0.035
ie-0.035
pa-0.034
father-0.032
ipped-0.032
Token a
Token ID257
Feature activation-6.85e+00

Pos logit contributions
ople+0.008
ipped+0.007
pa+0.007
father+0.007
ebra+0.007
Neg logit contributions
 sturdy-0.005
 exc-0.004
 bend-0.004
 flat-0.004
 crunch-0.004
Token big
Token ID1263
Feature activation-4.79e+00

Pos logit contributions
 bleed+0.151
 sew+0.146
 heal+0.144
 opener+0.144
uts+0.142
Neg logit contributions
by-0.158
ie-0.149
el-0.140
andy-0.138
 End-0.137
Token snow
Token ID6729
Feature activation-2.48e+00

Pos logit contributions
 exc+0.109
 bleed+0.106
 sturdy+0.106
 heal+0.102
onna+0.102
Neg logit contributions
pa-0.113
ie-0.112
el-0.109
ople-0.108
ebra-0.107
Tokenman
Token ID805
Feature activation-3.16e+00

Pos logit contributions
 sturdy+0.060
 exc+0.059
 flat+0.057
 bend+0.057
 heal+0.057
Neg logit contributions
ople-0.059
ipped-0.056
pa-0.054
father-0.053
atted-0.052
Token with
Token ID351
Feature activation-7.10e-01

Pos logit contributions
 sturdy+0.063
 exc+0.060
 bleed+0.058
 flat+0.057
onna+0.057
Neg logit contributions
ople-0.088
pa-0.084
ipped-0.084
ie-0.083
father-0.082
Token a
Token ID257
Feature activation-4.75e+00

Pos logit contributions
 sturdy+0.012
 flat+0.012
 exc+0.012
 crunch+0.011
 bend+0.011
Neg logit contributions
ople-0.023
ipped-0.022
or-0.021
ebra-0.021
pa-0.021
Token had
Token ID550
Feature activation-1.39e-01

Pos logit contributions
 sturdy+0.026
 exc+0.026
 heal+0.025
 bleed+0.025
 flat+0.024
Neg logit contributions
ople-0.030
ipped-0.029
pa-0.028
father-0.027
ie-0.027
Token a
Token ID257
Feature activation-6.17e+00

Pos logit contributions
 sturdy+0.003
 exc+0.003
 flat+0.003
 bend+0.002
 heal+0.002
Neg logit contributions
ople-0.004
ipped-0.004
pa-0.004
father-0.004
ie-0.004
Token small
Token ID1402
Feature activation-4.76e+00

Pos logit contributions
 bleed+0.129
 heal+0.127
 sew+0.117
 exc+0.116
onna+0.114
Neg logit contributions
by-0.154
ebra-0.142
 End-0.134
ie-0.132
ipped-0.131
Token shovel
Token ID33677
Feature activation-2.91e+00

Pos logit contributions
 exc+0.091
 sturdy+0.091
 heal+0.089
 bleed+0.087
onna+0.087
Neg logit contributions
ipped-0.124
ople-0.123
el-0.121
pa-0.121
ebra-0.121
Token and
Token ID290
Feature activation-2.87e+00

Pos logit contributions
 sturdy+0.054
 exc+0.052
 flat+0.052
 bend+0.051
onna+0.049
Neg logit contributions
ople-0.087
ipped-0.081
father-0.079
pa-0.079
elle-0.077
Token a
Token ID257
Feature activation-6.82e+00

Pos logit contributions
 sturdy+0.053
 exc+0.052
 bend+0.049
 flat+0.049
 boun+0.047
Neg logit contributions
ople-0.089
ipped-0.085
pa-0.082
father-0.081
ebra-0.080
Token bucket
Token ID19236
Feature activation-5.30e+00

Pos logit contributions
 bleed+0.143
 heal+0.141
uts+0.131
 exc+0.130
 sew+0.126
Neg logit contributions
by-0.156
 End-0.148
ebra-0.147
ipped-0.132
ael-0.129
Token.
Token ID13
Feature activation-6.87e-01

Pos logit contributions
 sturdy+0.123
 exc+0.118
 bleed+0.117
 heal+0.114
 sew+0.111
Neg logit contributions
ople-0.128
ipped-0.126
ie-0.123
pa-0.119
atted-0.119
Token He
Token ID679
Feature activation-1.11e+00

Pos logit contributions
 sturdy+0.013
 exc+0.012
 flat+0.012
 bend+0.012
onna+0.011
Neg logit contributions
ople-0.021
ipped-0.020
pa-0.019
father-0.019
ie-0.019
Token looked
Token ID3114
Feature activation-3.50e+00

Pos logit contributions
 sturdy+0.025
 bleed+0.025
 heal+0.024
 exc+0.024
 flat+0.023
Neg logit contributions
ople-0.028
ipped-0.026
pa-0.026
ie-0.025
eled-0.025
Token for
Token ID329
Feature activation+4.21e-01

Pos logit contributions
 sturdy+0.079
 exc+0.077
 flat+0.074
 bend+0.073
 heal+0.073
Neg logit contributions
ople-0.090
ipped-0.085
pa-0.083
father-0.081
ebra-0.080
Token.
Token ID13
Feature activation-3.08e-01

Pos logit contributions
 sturdy+0.008
 exc+0.007
 flat+0.007
 heal+0.007
 bend+0.007
Neg logit contributions
ople-0.013
ipped-0.013
father-0.012
ebra-0.012
pa-0.012
Token It
Token ID632
Feature activation+1.18e-01

Pos logit contributions
 sturdy+0.007
 exc+0.006
 bleed+0.006
 heal+0.006
 bend+0.006
Neg logit contributions
ople-0.008
ipped-0.008
pa-0.007
eled-0.007
atted-0.007
Token made
Token ID925
Feature activation+5.25e-01

Pos logit contributions
ople+0.002
ipped+0.002
pa+0.002
father+0.002
ie+0.002
Neg logit contributions
 sturdy-0.003
 exc-0.003
 flat-0.003
 heal-0.003
 bleed-0.003
Token the
Token ID262
Feature activation-3.23e-01

Pos logit contributions
ople+0.015
ipped+0.014
ebra+0.014
father+0.013
pa+0.013
Neg logit contributions
 sturdy-0.011
 crunch-0.011
 exc-0.011
 heal-0.011
 bend-0.010
Token sky
Token ID6766
Feature activation-4.26e+00

Pos logit contributions
 sturdy+0.007
 exc+0.007
 flat+0.007
 bend+0.007
 heal+0.007
Neg logit contributions
ople-0.008
ipped-0.008
pa-0.008
father-0.008
atted-0.008
Token a
Token ID257
Feature activation-6.82e+00

Pos logit contributions
 exc+0.064
 sturdy+0.063
onna+0.061
 bleed+0.059
 sew+0.056
Neg logit contributions
ie-0.128
ople-0.127
el-0.126
pa-0.123
ipped-0.122
Token deep
Token ID2769
Feature activation-2.61e+00

Pos logit contributions
 opener+0.143
 bleed+0.139
onna+0.137
uts+0.133
 sew+0.131
Neg logit contributions
 End-0.159
ie-0.156
by-0.154
el-0.137
ma-0.137
Token orange
Token ID10912
Feature activation-3.15e+00

Pos logit contributions
 exc+0.061
 sturdy+0.059
 bleed+0.057
 heal+0.056
onna+0.056
Neg logit contributions
ople-0.064
ie-0.061
ipped-0.060
el-0.060
pa-0.060
Token color
Token ID3124
Feature activation-3.38e+00

Pos logit contributions
 sturdy+0.072
 exc+0.071
 bleed+0.068
 heal+0.066
onna+0.066
Neg logit contributions
ople-0.077
ipped-0.075
ie-0.074
pa-0.073
el-0.072
Token and
Token ID290
Feature activation-2.31e+00

Pos logit contributions
 exc+0.076
 sturdy+0.075
 bleed+0.073
 heal+0.070
 sew+0.069
Neg logit contributions
ople-0.081
ie-0.079
ipped-0.079
el-0.078
ing-0.077
Token the
Token ID262
Feature activation+1.21e+00

Pos logit contributions
 exc+0.045
 sturdy+0.045
 bleed+0.044
 heal+0.043
onna+0.043
Neg logit contributions
ople-0.064
ipped-0.060
pa-0.059
ie-0.057
eled-0.057
Token the
Token ID262
Feature activation+2.05e+00

Pos logit contributions
 sturdy+0.003
 exc+0.003
 flat+0.003
 bend+0.003
 heal+0.003
Neg logit contributions
ople-0.004
ipped-0.004
atted-0.004
ebra-0.004
ed-0.004
Token dark
Token ID3223
Feature activation+1.48e+00

Pos logit contributions
ople+0.055
ipped+0.053
atted+0.051
pa+0.050
father+0.049
Neg logit contributions
 sturdy-0.045
 flat-0.044
 exc-0.043
 bend-0.042
 boun-0.042
Token.
Token ID13
Feature activation-3.68e-01

Pos logit contributions
ipped+0.053
atted+0.053
ople+0.051
father+0.048
ied+0.048
Neg logit contributions
 sturdy-0.017
 bend-0.016
 flat-0.016
 exc-0.016
 cozy-0.016
Token It
Token ID632
Feature activation-7.97e-01

Pos logit contributions
 sturdy+0.008
 exc+0.007
 bleed+0.007
 heal+0.007
 flat+0.007
Neg logit contributions
ople-0.010
ipped-0.010
pa-0.009
atted-0.009
eled-0.009
Token was
Token ID373
Feature activation-3.53e+00

Pos logit contributions
 sturdy+0.023
 bleed+0.022
 exc+0.022
 flat+0.022
onna+0.021
Neg logit contributions
ople-0.014
ie-0.014
ipped-0.013
pa-0.013
ed-0.013
Token a
Token ID257
Feature activation-6.81e+00

Pos logit contributions
 bleed+0.071
 sturdy+0.070
 exc+0.070
 heal+0.069
onna+0.068
Neg logit contributions
ople-0.085
pa-0.085
ie-0.084
father-0.083
el-0.081
Token hand
Token ID1021
Feature activation-3.17e+00

Pos logit contributions
 bleed+0.166
onna+0.146
 opener+0.141
 heal+0.141
uts+0.140
Neg logit contributions
 End-0.154
by-0.145
ebra-0.144
 audience-0.136
ie-0.136
Token that
Token ID326
Feature activation-4.63e-01

Pos logit contributions
 sturdy+0.068
 exc+0.066
 flat+0.063
 bend+0.062
 heal+0.062
Neg logit contributions
ople-0.085
ipped-0.081
pa-0.078
father-0.077
ie-0.076
Token looked
Token ID3114
Feature activation-2.95e+00

Pos logit contributions
 sturdy+0.011
 exc+0.011
 flat+0.010
 heal+0.010
 bleed+0.010
Neg logit contributions
ople-0.011
ipped-0.011
pa-0.010
father-0.010
ie-0.010
Token like
Token ID588
Feature activation-5.23e-01

Pos logit contributions
 exc+0.038
 sturdy+0.037
onna+0.037
 bleed+0.036
 sew+0.035
Neg logit contributions
ople-0.097
ipped-0.094
pa-0.094
ie-0.094
les-0.093
Token a
Token ID257
Feature activation-5.63e+00

Pos logit contributions
 sturdy+0.010
 exc+0.010
 flat+0.009
 bend+0.009
 heal+0.009
Neg logit contributions
ople-0.015
ipped-0.015
pa-0.014
father-0.014
ebra-0.014
Token his
Token ID465
Feature activation+1.10e+00

Pos logit contributions
 sturdy+0.003
 exc+0.003
 bleed+0.003
onna+0.003
 heal+0.003
Neg logit contributions
ople-0.003
ipped-0.003
pa-0.003
father-0.003
el-0.003
Token garden
Token ID11376
Feature activation-1.87e+00

Pos logit contributions
ople+0.032
ipped+0.030
pa+0.029
father+0.029
ebra+0.028
Neg logit contributions
 sturdy-0.022
 exc-0.022
 flat-0.021
 bend-0.020
 cozy-0.020
Token.
Token ID13
Feature activation-1.94e+00

Pos logit contributions
 sturdy+0.042
 bleed+0.041
 exc+0.040
 heal+0.038
 flat+0.038
Neg logit contributions
ople-0.047
ipped-0.046
ie-0.046
pa-0.044
el-0.044
Token It
Token ID632
Feature activation-1.11e+00

Pos logit contributions
 bleed+0.050
 sturdy+0.049
 exc+0.047
 heal+0.046
 flat+0.045
Neg logit contributions
ople-0.042
pa-0.040
ipped-0.039
el-0.038
father-0.038
Token was
Token ID373
Feature activation-4.14e+00

Pos logit contributions
 sturdy+0.032
 exc+0.031
 bleed+0.031
 flat+0.030
 heal+0.030
Neg logit contributions
ople-0.020
ie-0.020
ipped-0.019
ed-0.019
el-0.019
Token a
Token ID257
Feature activation-6.80e+00

Pos logit contributions
 bleed+0.089
 heal+0.085
 sew+0.083
 click+0.082
 exc+0.081
Neg logit contributions
ie-0.094
el-0.094
 End-0.089
aked-0.086
eled-0.086
Token very
Token ID845
Feature activation-4.92e+00

Pos logit contributions
 bleed+0.167
 loosen+0.144
 opener+0.144
 sew+0.144
 distract+0.143
Neg logit contributions
 End-0.138
by-0.134
ie-0.132
andy-0.122
 morning-0.121
Token special
Token ID2041
Feature activation-2.32e+00

Pos logit contributions
 bleed+0.088
 sew+0.082
 heal+0.082
 loosen+0.076
 opener+0.076
Neg logit contributions
ie-0.136
by-0.129
atted-0.128
 End-0.126
el-0.126
Token plant
Token ID4618
Feature activation-3.01e+00

Pos logit contributions
 bleed+0.043
 sturdy+0.040
 exc+0.040
 heal+0.037
 flat+0.037
Neg logit contributions
ie-0.066
by-0.063
el-0.062
ipped-0.061
ing-0.061
Token,
Token ID11
Feature activation-3.33e+00

Pos logit contributions
 sturdy+0.072
 bleed+0.072
 exc+0.070
 flat+0.067
 heal+0.066
Neg logit contributions
ie-0.070
ople-0.068
ing-0.065
pa-0.064
ipped-0.064
Token and
Token ID290
Feature activation-1.40e+00

Pos logit contributions
 sturdy+0.071
 bleed+0.068
onna+0.066
 heal+0.066
 exc+0.065
Neg logit contributions
ople-0.079
ie-0.077
ipped-0.076
el-0.076
pa-0.075
Token pink
Token ID11398
Feature activation-1.19e+00

Pos logit contributions
 bleed+0.109
 sturdy+0.108
 heal+0.104
onna+0.103
 sew+0.100
Neg logit contributions
ebra-0.080
ie-0.078
ople-0.078
by-0.077
pa-0.075
Token shape
Token ID5485
Feature activation-2.21e+00

Pos logit contributions
 sturdy+0.021
 exc+0.021
 bend+0.019
 flat+0.019
 heal+0.019
Neg logit contributions
ople-0.037
ipped-0.035
pa-0.034
father-0.034
atted-0.033
Token.
Token ID13
Feature activation-1.42e+00

Pos logit contributions
 exc+0.039
 sturdy+0.038
 bend+0.036
 flat+0.036
 heal+0.035
Neg logit contributions
ople-0.067
ipped-0.064
pa-0.063
father-0.062
atted-0.060
Token It
Token ID632
Feature activation-1.15e+00

Pos logit contributions
 sturdy+0.034
 bleed+0.032
 exc+0.032
 flat+0.031
 heal+0.031
Neg logit contributions
ople-0.033
el-0.032
ipped-0.031
pa-0.031
father-0.030
Token was
Token ID373
Feature activation-3.97e+00

Pos logit contributions
 sturdy+0.035
 exc+0.033
 bleed+0.033
 heal+0.032
 flat+0.032
Neg logit contributions
ople-0.019
ipped-0.019
ie-0.018
pa-0.018
el-0.018
Token a
Token ID257
Feature activation-6.79e+00

Pos logit contributions
 bleed+0.078
 heal+0.078
 sturdy+0.077
 exc+0.074
onna+0.074
Neg logit contributions
el-0.100
ie-0.097
pa-0.095
ople-0.095
ipped-0.094
Token nice
Token ID3621
Feature activation-2.90e+00

Pos logit contributions
 bleed+0.158
onna+0.143
 heal+0.143
 opener+0.141
 distract+0.138
Neg logit contributions
ebra-0.152
 End-0.148
ie-0.143
el-0.142
by-0.139
Token shape
Token ID5485
Feature activation-3.66e+00

Pos logit contributions
 sturdy+0.058
 bleed+0.057
 exc+0.057
 heal+0.054
 sew+0.052
Neg logit contributions
ople-0.075
ie-0.074
ipped-0.074
ebra-0.073
pa-0.073
Token and
Token ID290
Feature activation-2.50e+00

Pos logit contributions
 sturdy+0.085
 bleed+0.081
 exc+0.080
 heal+0.078
 flat+0.077
Neg logit contributions
ople-0.089
ipped-0.087
ie-0.086
el-0.084
ebra-0.084
Token it
Token ID340
Feature activation-1.10e+00

Pos logit contributions
 bleed+0.047
 sturdy+0.046
onna+0.046
 heal+0.045
 exc+0.044
Neg logit contributions
ople-0.067
ipped-0.066
pa-0.064
ie-0.064
el-0.064
Token kept
Token ID4030
Feature activation-1.26e+00

Pos logit contributions
 sturdy+0.028
 bleed+0.026
onna+0.026
 exc+0.026
 heal+0.026
Neg logit contributions
ople-0.023
ie-0.022
ipped-0.022
ed-0.021
les-0.021
Token she
Token ID673
Feature activation+2.10e+00

Pos logit contributions
 sturdy+0.022
 bleed+0.021
onna+0.020
 exc+0.020
 flat+0.019
Neg logit contributions
ie-0.026
ople-0.025
el-0.025
pa-0.025
ipped-0.025
Token went
Token ID1816
Feature activation+1.63e-01

Pos logit contributions
ople+0.060
ipped+0.057
atted+0.057
pa+0.055
father+0.055
Neg logit contributions
 exc-0.042
 sew-0.040
 bend-0.040
 sturdy-0.039
 heal-0.038
Token.
Token ID13
Feature activation-2.10e+00

Pos logit contributions
ople+0.004
ipped+0.004
ie+0.004
pa+0.004
el+0.004
Neg logit contributions
 sturdy-0.004
 exc-0.003
onna-0.003
 bleed-0.003
 bend-0.003
Token It
Token ID632
Feature activation-8.46e-01

Pos logit contributions
 sturdy+0.047
 exc+0.045
 bleed+0.044
 flat+0.042
 heal+0.042
Neg logit contributions
ople-0.054
ipped-0.052
pa-0.051
ie-0.050
father-0.050
Token was
Token ID373
Feature activation-3.61e+00

Pos logit contributions
 sturdy+0.024
 exc+0.024
 flat+0.023
onna+0.023
 bleed+0.023
Neg logit contributions
ople-0.016
ipped-0.015
ie-0.015
pa-0.015
father-0.014
Token a
Token ID257
Feature activation-6.78e+00

Pos logit contributions
 sturdy+0.075
 exc+0.074
 bleed+0.073
 heal+0.072
onna+0.070
Neg logit contributions
ie-0.088
ople-0.088
pa-0.087
father-0.085
ipped-0.084
Token big
Token ID1263
Feature activation-3.98e+00

Pos logit contributions
 bleed+0.147
 opener+0.132
uts+0.130
onna+0.129
 distract+0.128
Neg logit contributions
 End-0.151
by-0.151
ie-0.148
ma-0.146
ebra-0.145
Token,
Token ID11
Feature activation-4.20e+00

Pos logit contributions
 sturdy+0.100
 exc+0.099
 bleed+0.097
onna+0.095
 heal+0.094
Neg logit contributions
ople-0.088
pa-0.085
ipped-0.082
ie-0.082
ebra-0.080
Token red
Token ID2266
Feature activation-1.68e+00

Pos logit contributions
onna+0.074
 opener+0.069
 bleed+0.069
 exc+0.067
 heal+0.064
Neg logit contributions
ie-0.107
pa-0.103
elle-0.102
el-0.102
ople-0.101
Token hat
Token ID6877
Feature activation-4.44e+00

Pos logit contributions
 sturdy+0.030
 exc+0.029
 flat+0.027
 heal+0.027
 bend+0.027
Neg logit contributions
ople-0.051
ipped-0.049
pa-0.048
father-0.047
ie-0.046
Token that
Token ID326
Feature activation-9.25e-01

Pos logit contributions
 sturdy+0.103
 exc+0.100
 bleed+0.094
 flat+0.093
onna+0.093
Neg logit contributions
ople-0.111
ipped-0.106
pa-0.103
ie-0.103
father-0.101
Token the
Token ID262
Feature activation+1.76e+00

Pos logit contributions
ople+0.042
ipped+0.041
atted+0.040
ed+0.040
les+0.040
Neg logit contributions
 exc-0.028
 sturdy-0.027
 bend-0.027
 flat-0.027
 boun-0.026
Token snow
Token ID6729
Feature activation-1.74e-01

Pos logit contributions
ople+0.051
ipped+0.048
atted+0.047
ed+0.046
or+0.045
Neg logit contributions
 sturdy-0.037
 flat-0.036
 boun-0.035
 cozy-0.035
 exc-0.035
Token.
Token ID13
Feature activation-1.93e+00

Pos logit contributions
 flat+0.002
 exc+0.002
 sturdy+0.002
 bend+0.002
 heal+0.002
Neg logit contributions
ople-0.006
ipped-0.006
atted-0.006
eled-0.006
father-0.006
Token They
Token ID1119
Feature activation-1.70e+00

Pos logit contributions
 sturdy+0.037
 exc+0.036
 heal+0.034
 bleed+0.034
 flat+0.034
Neg logit contributions
ople-0.055
ipped-0.053
pa-0.051
ie-0.050
father-0.050
Token made
Token ID925
Feature activation+6.00e-01

Pos logit contributions
 sturdy+0.039
 heal+0.036
 bleed+0.036
onna+0.036
 exc+0.035
Neg logit contributions
ople-0.041
pa-0.039
ie-0.038
ipped-0.038
el-0.036
Token a
Token ID257
Feature activation-6.77e+00

Pos logit contributions
ople+0.019
ipped+0.018
father+0.018
pa+0.018
ebra+0.018
Neg logit contributions
 sturdy-0.010
 exc-0.010
 flat-0.010
 bend-0.009
 crunch-0.009
Token big
Token ID1263
Feature activation-4.66e+00

Pos logit contributions
 bleed+0.150
 heal+0.146
 sew+0.144
 opener+0.144
onna+0.143
Neg logit contributions
by-0.155
ie-0.144
el-0.134
andy-0.134
ebra-0.132
Token ball
Token ID2613
Feature activation-5.38e+00

Pos logit contributions
 exc+0.105
 sturdy+0.102
 bleed+0.101
onna+0.100
 opener+0.099
Neg logit contributions
ie-0.109
pa-0.109
el-0.107
ebra-0.104
ople-0.103
Token and
Token ID290
Feature activation-2.96e+00

Pos logit contributions
 sturdy+0.150
 bleed+0.143
 sil+0.141
 heal+0.137
 exc+0.137
Neg logit contributions
ie-0.105
ing-0.099
el-0.097
ipped-0.092
les-0.090
Token a
Token ID257
Feature activation-6.31e+00

Pos logit contributions
 heal+0.064
 sturdy+0.064
onna+0.064
 exc+0.062
 bleed+0.062
Neg logit contributions
ople-0.073
ipped-0.069
pa-0.067
ed-0.067
atted-0.066
Token small
Token ID1402
Feature activation-4.33e+00

Pos logit contributions
 heal+0.140
 bleed+0.140
 exc+0.138
 opener+0.137
uts+0.135
Neg logit contributions
by-0.131
ebra-0.124
 End-0.122
ie-0.109
ma-0.106
Token
Token ID220
Feature activation+1.66e+00

Pos logit contributions
 exc+0.013
 sturdy+0.013
 heal+0.012
 flat+0.012
 sew+0.012
Neg logit contributions
ople-0.019
ipped-0.018
pa-0.017
father-0.017
ebra-0.017
Token
Token ID198
Feature activation+4.70e-01

Pos logit contributions
ople+0.047
ipped+0.045
pa+0.044
father+0.043
ebra+0.043
Neg logit contributions
 sturdy-0.033
 exc-0.032
 flat-0.030
 heal-0.030
 bend-0.030
Token
Token ID198
Feature activation-5.69e-01

Pos logit contributions
ople+0.014
ipped+0.013
pa+0.013
father+0.012
atted+0.012
Neg logit contributions
 sturdy-0.009
 exc-0.009
 heal-0.008
 bend-0.008
onna-0.008
TokenSo
Token ID2396
Feature activation+9.26e-01

Pos logit contributions
 sturdy+0.012
 bleed+0.011
 flat+0.011
 exc+0.011
 heal+0.011
Neg logit contributions
ople-0.015
ipped-0.015
father-0.014
pa-0.014
ie-0.014
Token,
Token ID11
Feature activation-1.23e+00

Pos logit contributions
ople+0.029
ipped+0.029
eled+0.028
father+0.028
ebra+0.028
Neg logit contributions
 exc-0.013
 sturdy-0.013
 bend-0.012
 boun-0.012
 strong-0.012
Token Qu
Token ID2264
Feature activation+6.29e+00

Pos logit contributions
 sturdy+0.022
 exc+0.022
 bend+0.020
 flat+0.020
 sew+0.020
Neg logit contributions
ople-0.037
ipped-0.036
father-0.034
pa-0.034
eled-0.034
Tokenackers
Token ID28874
Feature activation+2.39e+00

Pos logit contributions
ople+0.141
pa+0.140
father+0.135
eled+0.129
ted+0.128
Neg logit contributions
onna-0.160
 exc-0.157
 bend-0.155
 sturdy-0.147
 opener-0.147
Token and
Token ID290
Feature activation+8.52e-01

Pos logit contributions
ople+0.068
ipped+0.064
father+0.063
pa+0.063
ebra+0.061
Neg logit contributions
 exc-0.046
 sturdy-0.046
 bend-0.043
 flat-0.042
 heal-0.042
Token Rib
Token ID23133
Feature activation+5.22e+00

Pos logit contributions
ipped+0.026
ople+0.025
atted+0.024
les+0.023
father+0.023
Neg logit contributions
 exc-0.016
 boun-0.014
 bend-0.014
onna-0.014
 sturdy-0.014
Tokenbit
Token ID2545
Feature activation+2.90e-02

Pos logit contributions
ople+0.133
ipped+0.129
pa+0.126
eled+0.124
father+0.123
Neg logit contributions
 exc-0.109
 sturdy-0.108
 crunch-0.108
 heal-0.107
 bleed-0.106
Token left
Token ID1364
Feature activation+1.93e+00

Pos logit contributions
ople+0.001
ipped+0.001
pa+0.001
father+0.001
ie+0.001
Neg logit contributions
 sturdy-0.001
 exc-0.001
 flat-0.001
 heal-0.001
 bend-0.001
Token a
Token ID257
Feature activation-4.52e+00

Pos logit contributions
ople+0.032
ipped+0.031
pa+0.030
father+0.030
ebra+0.029
Neg logit contributions
 sturdy-0.024
 exc-0.023
 flat-0.022
 bend-0.022
 heal-0.022
Token brand
Token ID4508
Feature activation-3.18e+00

Pos logit contributions
 bleed+0.101
 exc+0.100
onna+0.091
 opener+0.088
 sew+0.087
Neg logit contributions
by-0.106
ie-0.098
mother-0.095
father-0.093
ing-0.091
Token new
Token ID649
Feature activation-1.67e+00

Pos logit contributions
 exc+0.070
 bleed+0.070
 sturdy+0.066
 twitch+0.064
 bend+0.064
Neg logit contributions
ie-0.079
ipped-0.078
by-0.076
ople-0.076
or-0.076
Token house
Token ID2156
Feature activation-3.03e+00

Pos logit contributions
 bleed+0.045
 exc+0.045
onna+0.043
 sturdy+0.042
 heal+0.041
Neg logit contributions
ie-0.033
pa-0.033
by-0.032
mother-0.031
father-0.031
Token!
Token ID0
Feature activation-1.02e-01

Pos logit contributions
 exc+0.070
 bleed+0.067
 sturdy+0.066
 bend+0.064
onna+0.063
Neg logit contributions
ople-0.074
pa-0.070
ipped-0.069
ie-0.069
mother-0.067
Token<|endoftext|>
Token ID50256
Feature activation+6.53e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
 sew+0.002
 bend+0.002
 heal+0.002
Neg logit contributions
ople-0.003
pa-0.003
ipped-0.003
ie-0.003
father-0.003
TokenTom
Token ID13787
Feature activation+4.26e+00

Pos logit contributions
ople+0.177
atted+0.164
pa+0.162
father+0.160
eled+0.154
Neg logit contributions
 opener-0.134
 boun-0.129
 bend-0.125
 cozy-0.124
 flat-0.124
Tokenmy
Token ID1820
Feature activation+1.53e+00

Pos logit contributions
ople+0.081
ipped+0.071
ebra+0.070
or+0.068
elle+0.063
Neg logit contributions
 clean-0.118
 exc-0.118
 bend-0.116
 sew-0.116
 sturdy-0.116
Token was
Token ID373
Feature activation-2.78e+00

Pos logit contributions
ople+0.039
ipped+0.037
father+0.035
ebra+0.035
pa+0.035
Neg logit contributions
 sturdy-0.034
 exc-0.034
 bend-0.032
 flat-0.032
 sew-0.032
Token out
Token ID503
Feature activation-3.70e-01

Pos logit contributions
 sturdy+0.054
 exc+0.053
 flat+0.050
 bend+0.049
 heal+0.049
Neg logit contributions
ople-0.081
ipped-0.077
pa-0.075
father-0.074
ebra-0.073
Token playing
Token ID2712
Feature activation-1.31e-01

Pos logit contributions
 exc+0.008
 sturdy+0.008
 flat+0.007
 bend+0.007
 heal+0.007
Neg logit contributions
ople-0.010
ipped-0.010
pa-0.009
father-0.009
io-0.009
Token us
Token ID514
Feature activation-1.34e+00

Pos logit contributions
ople+0.007
ie+0.006
pa+0.006
ipped+0.006
father+0.006
Neg logit contributions
 sturdy-0.006
 exc-0.006
 bleed-0.006
 flat-0.006
 boun-0.006
Token escape
Token ID6654
Feature activation-4.40e-02

Pos logit contributions
 sturdy+0.028
 boun+0.026
 bleed+0.026
 exc+0.026
 flat+0.026
Neg logit contributions
ie-0.034
pa-0.033
father-0.032
ople-0.032
el-0.032
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
 exc+0.001
 sturdy+0.001
 bleed+0.001
 flat+0.001
 sew+0.001
Neg logit contributions
ople-0.001
ie-0.001
pa-0.001
ipped-0.001
el-0.001
Token mystery
Token ID10715
Feature activation+5.58e-01

Pos logit contributions
ople+0.032
ipped+0.030
pa+0.030
father+0.029
ie+0.029
Neg logit contributions
 sturdy-0.024
 exc-0.024
 bleed-0.022
 flat-0.022
 heal-0.022
Token?
Token ID30
Feature activation-5.98e-01

Pos logit contributions
ople+0.015
ipped+0.014
pa+0.014
father+0.014
ie+0.014
Neg logit contributions
 sturdy-0.012
 exc-0.011
 bleed-0.011
 flat-0.011
 bend-0.011
Token<|endoftext|>
Token ID50256
Feature activation+6.71e+00

Pos logit contributions
 sturdy+0.013
 bleed+0.012
 flat+0.012
 exc+0.012
 crunch+0.012
Neg logit contributions
ople-0.016
ipped-0.015
pa-0.015
father-0.015
el-0.014
TokenOnce
Token ID7454
Feature activation+3.52e+00

Pos logit contributions
ople+0.175
atted+0.162
pa+0.157
eled+0.156
father+0.155
Neg logit contributions
 opener-0.143
onna-0.141
 exc-0.139
 bend-0.137
 sew-0.136
Token upon
Token ID2402
Feature activation+2.79e+00

Pos logit contributions
ople+0.102
ipped+0.095
ated+0.091
aked+0.091
ted+0.088
Neg logit contributions
 exc-0.073
onna-0.072
 flat-0.068
 sturdy-0.067
 bend-0.067
Token a
Token ID257
Feature activation-1.78e-01

Pos logit contributions
ople+0.090
ipped+0.087
pa+0.083
atted+0.082
father+0.082
Neg logit contributions
 sturdy-0.047
 exc-0.046
 bend-0.043
 flat-0.042
 heal-0.042
Token time
Token ID640
Feature activation+1.00e+00

Pos logit contributions
 sturdy+0.005
 exc+0.005
 bend+0.005
 crunch+0.005
onna+0.005
Neg logit contributions
ople-0.004
ipped-0.004
pa-0.003
father-0.003
ed-0.003
Token there
Token ID612
Feature activation+3.62e+00

Pos logit contributions
ople+0.029
ipped+0.026
father+0.026
eled+0.026
pa+0.026
Neg logit contributions
 exc-0.021
 sturdy-0.020
 bend-0.020
 flat-0.019
onna-0.019
Token end
Token ID886
Feature activation+6.02e-01

Pos logit contributions
ipped+0.022
ople+0.022
atted+0.021
eled+0.021
ated+0.020
Neg logit contributions
 sturdy-0.020
 exc-0.020
 bend-0.019
 boun-0.019
 cozy-0.019
Token of
Token ID286
Feature activation+1.84e+00

Pos logit contributions
ople+0.016
ipped+0.015
eled+0.014
pa+0.014
ebra+0.014
Neg logit contributions
 sturdy-0.013
 exc-0.013
 heal-0.012
 bend-0.012
 flat-0.012
Token the
Token ID262
Feature activation+1.69e+00

Pos logit contributions
ople+0.053
ipped+0.050
les+0.048
atted+0.047
ebra+0.047
Neg logit contributions
 exc-0.038
 sturdy-0.037
 flat-0.035
 heal-0.035
 bend-0.035
Token march
Token ID9960
Feature activation+9.60e-02

Pos logit contributions
ople+0.039
ipped+0.039
atted+0.038
ed+0.036
io+0.036
Neg logit contributions
 exc-0.045
 sturdy-0.044
 cozy-0.042
 boun-0.042
 bend-0.042
Token.
Token ID13
Feature activation-2.63e-01

Pos logit contributions
ople+0.003
ipped+0.003
eled+0.003
father+0.003
pa+0.003
Neg logit contributions
 sturdy-0.002
 exc-0.002
 heal-0.002
 flat-0.002
 bend-0.002
Token<|endoftext|>
Token ID50256
Feature activation+6.21e+00

Pos logit contributions
 exc+0.004
 sturdy+0.004
 heal+0.004
onna+0.004
 cozy+0.003
Neg logit contributions
ople-0.009
ipped-0.009
ebra-0.008
pa-0.008
les-0.008
TokenTom
Token ID13787
Feature activation+4.65e+00

Pos logit contributions
ople+0.231
atted+0.220
ipped+0.216
father+0.214
pa+0.213
Neg logit contributions
 opener-0.067
onna-0.062
 exc-0.062
 cozy-0.060
 bend-0.060
Token was
Token ID373
Feature activation-2.83e+00

Pos logit contributions
ople+0.097
ipped+0.092
ebra+0.087
io+0.086
oos+0.085
Neg logit contributions
 clean-0.121
 exc-0.116
 fit-0.114
 cozy-0.114
 sew-0.114
Token a
Token ID257
Feature activation-5.00e+00

Pos logit contributions
 sturdy+0.054
 exc+0.053
 flat+0.050
 bend+0.050
 heal+0.049
Neg logit contributions
ople-0.082
ipped-0.079
pa-0.076
father-0.075
ebra-0.074
Token silly
Token ID14397
Feature activation-1.91e+00

Pos logit contributions
 bleed+0.109
 sturdy+0.108
onna+0.105
 heal+0.105
 exc+0.104
Neg logit contributions
ople-0.125
pa-0.118
ipped-0.116
ebra-0.115
ie-0.114
Token boy
Token ID2933
Feature activation-7.20e-01

Pos logit contributions
 exc+0.043
 sturdy+0.043
 boun+0.041
 flat+0.041
 bend+0.041
Neg logit contributions
ople-0.049
ipped-0.047
father-0.044
pa-0.044
atted-0.044
Token,
Token ID11
Feature activation-2.20e+00

Pos logit contributions
ople+0.033
ipped+0.032
pa+0.031
father+0.030
ebra+0.029
Neg logit contributions
 sturdy-0.033
 exc-0.032
 flat-0.031
 heal-0.030
 bend-0.030
Token funny
Token ID8258
Feature activation+1.70e+00

Pos logit contributions
 bleed+0.053
 sturdy+0.052
 exc+0.050
 flat+0.049
 twitch+0.049
Neg logit contributions
ople-0.048
pa-0.047
ie-0.045
father-0.044
mother-0.044
Token fire
Token ID2046
Feature activation+7.27e-01

Pos logit contributions
ople+0.049
pa+0.047
ipped+0.047
ie+0.047
father+0.045
Neg logit contributions
 sturdy-0.033
 exc-0.031
 flat-0.030
 bend-0.030
 bleed-0.030
Tokenman
Token ID805
Feature activation+1.21e+00

Pos logit contributions
ople+0.017
ipped+0.017
pa+0.016
father+0.016
atted+0.015
Neg logit contributions
 sturdy-0.018
 exc-0.017
 heal-0.016
 flat-0.016
 bend-0.016
Token!"
Token ID2474
Feature activation-9.38e-01

Pos logit contributions
ople+0.041
ipped+0.040
pa+0.038
eled+0.038
father+0.038
Neg logit contributions
 sturdy-0.017
 exc-0.017
 heal-0.016
 flat-0.015
 bend-0.015
Token<|endoftext|>
Token ID50256
Feature activation+5.81e+00

Pos logit contributions
 sturdy+0.018
 exc+0.018
 heal+0.017
 flat+0.017
 bend+0.017
Neg logit contributions
ople-0.027
ipped-0.025
pa-0.025
father-0.024
ebra-0.024
TokenOnce
Token ID7454
Feature activation+3.87e+00

Pos logit contributions
ople+0.208
ipped+0.195
atted+0.193
ebra+0.192
father+0.191
Neg logit contributions
 opener-0.071
onna-0.070
 exc-0.069
 bend-0.067
 flat-0.067
Token upon
Token ID2402
Feature activation+1.89e+00

Pos logit contributions
ople+0.107
ipped+0.096
ebra+0.093
aked+0.092
ated+0.091
Neg logit contributions
 exc-0.089
onna-0.085
 flat-0.082
 sturdy-0.081
 bend-0.081
Token a
Token ID257
Feature activation+2.46e-01

Pos logit contributions
ople+0.051
ipped+0.049
pa+0.048
father+0.047
ie+0.047
Neg logit contributions
 sturdy-0.039
 exc-0.037
 flat-0.036
onna-0.035
 heal-0.035
Token time
Token ID640
Feature activation+1.35e+00

Pos logit contributions
ople+0.005
ipped+0.005
father+0.004
pa+0.004
ebra+0.004
Neg logit contributions
 sturdy-0.007
 exc-0.007
 bend-0.007
 flat-0.007
 crunch-0.007
Token,
Token ID11
Feature activation+6.50e-01

Pos logit contributions
ople+0.043
eled+0.040
father+0.040
les+0.040
ipped+0.039
Neg logit contributions
 exc-0.025
 bend-0.022
 boun-0.022
 flat-0.021
 sturdy-0.021
Token over
Token ID625
Feature activation+2.32e+00

Pos logit contributions
ople+0.084
father+0.083
pa+0.080
mother+0.079
parents+0.077
Neg logit contributions
 sturdy-0.040
 flat-0.036
 bend-0.036
 gently-0.035
 boun-0.034
Token and
Token ID290
Feature activation-8.65e-01

Pos logit contributions
ople+0.083
father+0.079
pa+0.077
ie+0.076
ipped+0.076
Neg logit contributions
 sturdy-0.031
 bend-0.029
 flat-0.029
 boun-0.028
 exc-0.028
Token began
Token ID2540
Feature activation-5.62e-01

Pos logit contributions
 sturdy+0.016
 boun+0.016
 bend+0.016
 flat+0.015
 exc+0.015
Neg logit contributions
father-0.025
ople-0.024
pa-0.024
ebra-0.024
ipped-0.024
Token playing
Token ID2712
Feature activation+7.68e-01

Pos logit contributions
 sturdy+0.009
 exc+0.009
 heal+0.009
 flat+0.009
 bleed+0.008
Neg logit contributions
ople-0.018
ipped-0.017
pa-0.017
father-0.016
atted-0.016
Token.
Token ID13
Feature activation-2.09e-01

Pos logit contributions
ople+0.022
ipped+0.021
pa+0.021
father+0.020
ie+0.020
Neg logit contributions
 sturdy-0.015
 exc-0.014
 flat-0.014
 heal-0.014
 bend-0.014
Token
Token ID220
Feature activation+1.82e+00

Pos logit contributions
 sturdy+0.004
 exc+0.004
 flat+0.004
 bend+0.004
 heal+0.004
Neg logit contributions
ople-0.006
ipped-0.005
pa-0.005
father-0.005
ie-0.005
Token
Token ID198
Feature activation+5.79e-01

Pos logit contributions
ople+0.052
ipped+0.050
pa+0.049
father+0.048
ie+0.047
Neg logit contributions
 sturdy-0.036
 exc-0.035
 flat-0.033
 bend-0.033
 heal-0.033
Token
Token ID198
Feature activation-4.84e-01

Pos logit contributions
ople+0.015
ipped+0.014
pa+0.014
father+0.013
ebra+0.013
Neg logit contributions
 sturdy-0.013
 exc-0.013
 bleed-0.012
 flat-0.012
 heal-0.012
TokenThe
Token ID464
Feature activation+2.53e+00

Pos logit contributions
 sturdy+0.011
 bleed+0.010
 exc+0.010
 heal+0.010
 flat+0.010
Neg logit contributions
ople-0.012
ipped-0.012
or-0.011
aked-0.011
ebra-0.011
Token olive
Token ID19450
Feature activation+1.63e+00

Pos logit contributions
ople+0.072
ipped+0.069
pa+0.065
atted+0.065
eled+0.064
Neg logit contributions
 sturdy-0.053
 exc-0.051
 bend-0.049
 flat-0.048
 crunch-0.047
Token was
Token ID373
Feature activation-2.51e+00

Pos logit contributions
ople+0.039
ipped+0.037
pa+0.036
father+0.035
atted+0.034
Neg logit contributions
 sturdy-0.040
 exc-0.039
 bend-0.037
 flat-0.037
 heal-0.037
Token.
Token ID13
Feature activation-5.65e-01

Pos logit contributions
 sturdy+0.022
 exc+0.021
 flat+0.020
 heal+0.020
 bend+0.020
Neg logit contributions
ople-0.035
ipped-0.034
pa-0.033
father-0.032
atted-0.032
Token
Token ID198
Feature activation+1.76e+00

Pos logit contributions
 sturdy+0.012
 exc+0.011
 flat+0.011
 heal+0.011
 bend+0.011
Neg logit contributions
ople-0.015
ipped-0.015
pa-0.014
father-0.014
ebra-0.014
Token
Token ID198
Feature activation+1.45e-01

Pos logit contributions
ople+0.051
ipped+0.048
pa+0.046
father+0.046
ebra+0.045
Neg logit contributions
 sturdy-0.035
 exc-0.034
 heal-0.032
 bend-0.032
 flat-0.032
TokenThe
Token ID464
Feature activation+2.77e+00

Pos logit contributions
ople+0.005
ipped+0.004
pa+0.004
father+0.004
ebra+0.004
Neg logit contributions
 sturdy-0.002
 exc-0.002
 flat-0.002
 heal-0.002
 bleed-0.002
Token car
Token ID1097
Feature activation+5.43e-01

Pos logit contributions
ople+0.070
ipped+0.067
pa+0.065
father+0.064
ie+0.062
Neg logit contributions
 sturdy-0.064
 exc-0.063
 flat-0.060
 heal-0.060
 bend-0.060
Token sees
Token ID7224
Feature activation+1.93e+00

Pos logit contributions
ople+0.013
ipped+0.012
pa+0.012
father+0.012
ie+0.011
Neg logit contributions
 exc-0.013
 sturdy-0.013
 bend-0.013
 flat-0.013
 heal-0.012
Token Ben
Token ID3932
Feature activation+3.48e-01

Pos logit contributions
or+0.063
ople+0.063
ipped+0.061
father+0.061
ebra+0.061
Neg logit contributions
 crunch-0.026
 bend-0.026
 exc-0.025
 heal-0.024
 sturdy-0.024
Token.
Token ID13
Feature activation-1.02e+00

Pos logit contributions
ople+0.010
ipped+0.010
pa+0.009
father+0.009
ebra+0.009
Neg logit contributions
 sturdy-0.007
 exc-0.007
 flat-0.006
 heal-0.006
 bend-0.006
Token It
Token ID632
Feature activation+2.26e-01

Pos logit contributions
 sturdy+0.024
 exc+0.023
 flat+0.022
 heal+0.022
 bend+0.022
Neg logit contributions
ople-0.024
ipped-0.024
ed-0.022
ebra-0.022
el-0.022
Token hon
Token ID3032
Feature activation+1.08e+00

Pos logit contributions
ople+0.004
ipped+0.004
pa+0.004
father+0.004
ie+0.004
Neg logit contributions
 sturdy-0.007
 exc-0.006
 flat-0.006
 bend-0.006
 heal-0.006
Tokenks
Token ID591
Feature activation-2.85e-01

Pos logit contributions
ople+0.023
pa+0.022
father+0.022
ie+0.021
ipped+0.021
Neg logit contributions
 bleed-0.029
 flat-0.028
 bend-0.028
 exc-0.028
 sturdy-0.028
Token make
Token ID787
Feature activation+1.40e+00

Pos logit contributions
 sturdy+0.036
 exc+0.035
 flat+0.032
 heal+0.032
 bend+0.032
Neg logit contributions
ople-0.075
ipped-0.072
pa-0.070
father-0.070
ie-0.069
Token new
Token ID649
Feature activation+5.54e-01

Pos logit contributions
ople+0.043
ipped+0.042
ebra+0.041
father+0.040
eled+0.039
Neg logit contributions
 sturdy-0.026
 heal-0.024
 exc-0.024
 cozy-0.023
 crunch-0.023
Token discoveries
Token ID24362
Feature activation-2.26e+00

Pos logit contributions
ople+0.014
ipped+0.013
eled+0.012
ebra+0.012
father+0.012
Neg logit contributions
 sturdy-0.013
 exc-0.013
 cozy-0.013
 heal-0.013
 boun-0.013
Token.
Token ID13
Feature activation-6.76e-01

Pos logit contributions
 exc+0.049
 sturdy+0.049
 bleed+0.048
onna+0.046
 bend+0.046
Neg logit contributions
ople-0.056
ie-0.054
pa-0.054
ipped-0.053
les-0.052
Token<|endoftext|>
Token ID50256
Feature activation+6.34e+00

Pos logit contributions
 exc+0.011
 sturdy+0.010
onna+0.010
 heal+0.009
 sew+0.009
Neg logit contributions
ople-0.022
ipped-0.021
pa-0.020
ie-0.020
ebra-0.020
TokenOnce
Token ID7454
Feature activation+3.75e+00

Pos logit contributions
ople+0.189
atted+0.179
pa+0.173
father+0.172
ipped+0.169
Neg logit contributions
 opener-0.112
onna-0.109
 boun-0.107
 exc-0.106
 flat-0.106
Token upon
Token ID2402
Feature activation+1.86e+00

Pos logit contributions
ople+0.105
ipped+0.095
ated+0.092
aked+0.092
eled+0.090
Neg logit contributions
 exc-0.086
onna-0.081
 flat-0.079
 sturdy-0.078
 bend-0.077
Token a
Token ID257
Feature activation-1.52e-01

Pos logit contributions
ople+0.050
ipped+0.048
pa+0.047
father+0.046
ie+0.046
Neg logit contributions
 sturdy-0.038
 exc-0.036
 flat-0.035
onna-0.035
 heal-0.034
Token time
Token ID640
Feature activation+1.20e+00

Pos logit contributions
 sturdy+0.004
 exc+0.004
 bend+0.004
 flat+0.004
 crunch+0.004
Neg logit contributions
ople-0.003
ipped-0.003
father-0.003
pa-0.003
atted-0.003
Token,
Token ID11
Feature activation+8.98e-01

Pos logit contributions
ople+0.040
les+0.037
eled+0.037
father+0.037
ipped+0.036
Neg logit contributions
 exc-0.021
 bend-0.019
 boun-0.019
 flat-0.018
 sturdy-0.018
Token there
Token ID612
Feature activation+3.29e+00

Pos logit contributions
ople+0.030
pa+0.028
father+0.028
ipped+0.028
mother+0.027
Neg logit contributions
 exc-0.016
 sturdy-0.015
 bend-0.015
 flat-0.015
 boun-0.014
Token doing
Token ID1804
Feature activation+6.52e-01

Pos logit contributions
ople+0.020
ipped+0.019
pa+0.018
atted+0.018
father+0.018
Neg logit contributions
 exc-0.015
 sturdy-0.015
 flat-0.014
 bend-0.014
 heal-0.014
Token here
Token ID994
Feature activation-9.78e-01

Pos logit contributions
ople+0.015
pa+0.015
ipped+0.014
ie+0.014
father+0.014
Neg logit contributions
 sturdy-0.016
 bleed-0.016
 exc-0.015
onna-0.015
 flat-0.015
Token?".
Token ID43634
Feature activation-4.02e-01

Pos logit contributions
 sturdy+0.025
 bleed+0.025
 exc+0.024
 flat+0.023
onna+0.022
Neg logit contributions
ople-0.021
pa-0.020
ie-0.020
ipped-0.020
father-0.019
Token
Token ID198
Feature activation+1.21e+00

Pos logit contributions
 sturdy+0.009
 exc+0.009
 flat+0.009
 bend+0.009
 bleed+0.009
Neg logit contributions
ople-0.010
ipped-0.010
pa-0.009
father-0.009
ie-0.009
Token
Token ID198
Feature activation-4.22e-01

Pos logit contributions
ople+0.034
ipped+0.033
pa+0.032
father+0.031
ebra+0.031
Neg logit contributions
 sturdy-0.024
 exc-0.024
 heal-0.022
 bend-0.022
 flat-0.022
TokenThe
Token ID464
Feature activation+2.54e+00

Pos logit contributions
 bleed+0.010
 sturdy+0.010
 exc+0.010
onna+0.010
 flat+0.010
Neg logit contributions
ople-0.010
ipped-0.009
father-0.009
pa-0.009
mother-0.009
Token fle
Token ID5104
Feature activation+3.15e+00

Pos logit contributions
ople+0.079
ipped+0.075
pa+0.074
father+0.072
atted+0.071
Neg logit contributions
 sturdy-0.044
 exc-0.043
 flat-0.040
 bend-0.040
 heal-0.040
Tokena
Token ID64
Feature activation+2.47e+00

Pos logit contributions
ople+0.067
pa+0.063
ipped+0.062
father+0.060
eled+0.057
Neg logit contributions
 sturdy-0.085
 exc-0.084
 bend-0.082
onna-0.081
 heal-0.081
Token replied
Token ID8712
Feature activation-1.80e-02

Pos logit contributions
ople+0.063
ipped+0.060
pa+0.058
father+0.057
ebra+0.056
Neg logit contributions
 sturdy-0.056
 exc-0.055
 bend-0.053
 flat-0.052
 heal-0.052
Token,
Token ID11
Feature activation-1.93e+00

Pos logit contributions
 sturdy+0.000
 exc+0.000
 bleed+0.000
 flat+0.000
 bend+0.000
Neg logit contributions
ople-0.000
ipped-0.000
pa-0.000
ie-0.000
atted-0.000
Token "
Token ID366
Feature activation+3.80e-01

Pos logit contributions
 sturdy+0.039
 bleed+0.038
 exc+0.036
onna+0.034
 flat+0.033
Neg logit contributions
ople-0.051
ipped-0.051
ed-0.050
atted-0.049
ing-0.049
Token silver
Token ID8465
Feature activation+5.70e-01

Pos logit contributions
ople+0.011
ipped+0.010
father+0.010
pa+0.010
ebra+0.010
Neg logit contributions
 sturdy-0.008
 exc-0.008
 flat-0.008
 bend-0.007
 heal-0.007
Token stars
Token ID5788
Feature activation-1.91e+00

Pos logit contributions
ople+0.017
ipped+0.016
pa+0.016
father+0.016
eled+0.016
Neg logit contributions
 sturdy-0.010
 exc-0.010
 flat-0.010
 bend-0.009
 heal-0.009
Token.
Token ID13
Feature activation-9.74e-01

Pos logit contributions
 sturdy+0.043
 exc+0.042
 bleed+0.039
 heal+0.039
 bend+0.039
Neg logit contributions
ople-0.048
ipped-0.047
ie-0.045
pa-0.045
les-0.044
Token Her
Token ID2332
Feature activation+4.74e-01

Pos logit contributions
 sturdy+0.018
 exc+0.017
 flat+0.016
 heal+0.016
 bleed+0.016
Neg logit contributions
ople-0.029
ipped-0.028
pa-0.027
father-0.027
atted-0.026
Token mom
Token ID1995
Feature activation+1.89e+00

Pos logit contributions
ople+0.014
ipped+0.014
father+0.013
pa+0.013
eled+0.013
Neg logit contributions
 exc-0.009
 sturdy-0.009
 bend-0.008
 flat-0.008
 boun-0.008
Token saw
Token ID2497
Feature activation+3.02e+00

Pos logit contributions
ople+0.055
ipped+0.053
ebra+0.052
father+0.051
or+0.050
Neg logit contributions
 exc-0.034
 bend-0.033
 sew-0.033
 sturdy-0.033
 heal-0.032
Token her
Token ID607
Feature activation-2.55e-02

Pos logit contributions
ople+0.101
father+0.094
ated+0.093
ebra+0.092
or+0.090
Neg logit contributions
 exc-0.043
 crunch-0.040
 gently-0.040
 sew-0.039
 unt-0.038
Token crying
Token ID13774
Feature activation+1.82e-01

Pos logit contributions
 sturdy+0.000
 exc+0.000
 flat+0.000
 bend+0.000
 heal+0.000
Neg logit contributions
ople-0.001
ipped-0.001
pa-0.001
father-0.001
ebra-0.001
Token and
Token ID290
Feature activation-1.58e+00

Pos logit contributions
ople+0.006
ebra+0.006
ipped+0.006
father+0.006
pa+0.005
Neg logit contributions
 heal-0.003
 bleed-0.003
 exc-0.003
 flat-0.003
 sturdy-0.003
Token came
Token ID1625
Feature activation+1.07e+00

Pos logit contributions
 sturdy+0.034
 exc+0.034
 flat+0.032
 bend+0.032
 heal+0.032
Neg logit contributions
ople-0.042
ipped-0.040
pa-0.039
father-0.038
ebra-0.038
Token to
Token ID284
Feature activation-7.37e-01

Pos logit contributions
ople+0.036
ipped+0.034
ebra+0.034
father+0.033
pa+0.033
Neg logit contributions
 exc-0.016
 sturdy-0.015
 bend-0.015
 flat-0.015
 heal-0.014
Token doesn
Token ID1595
Feature activation-8.71e-01

Pos logit contributions
 sturdy+0.009
 exc+0.008
 flat+0.008
 bend+0.008
 heal+0.008
Neg logit contributions
ople-0.007
ipped-0.007
pa-0.007
father-0.007
ie-0.006
Token't
Token ID470
Feature activation-2.18e+00

Pos logit contributions
 sturdy+0.019
 exc+0.018
 flat+0.017
 boun+0.017
 crunch+0.017
Neg logit contributions
ople-0.022
ipped-0.021
el-0.021
ie-0.021
pa-0.020
Token mean
Token ID1612
Feature activation+8.40e-01

Pos logit contributions
 sturdy+0.044
onna+0.042
 sil+0.041
 opener+0.041
 sew+0.040
Neg logit contributions
ie-0.058
ople-0.052
by-0.049
ado-0.048
pa-0.048
Token it
Token ID340
Feature activation+1.87e-01

Pos logit contributions
ople+0.028
ipped+0.027
ebra+0.027
father+0.027
eled+0.026
Neg logit contributions
 exc-0.012
 sturdy-0.012
 heal-0.012
 flat-0.011
 bend-0.011
Token's
Token ID338
Feature activation-3.41e-01

Pos logit contributions
ople+0.005
ipped+0.004
pa+0.004
father+0.004
eled+0.004
Neg logit contributions
 sturdy-0.004
 exc-0.004
 heal-0.004
 bend-0.004
 flat-0.004
Token good
Token ID922
Feature activation+1.87e-01

Pos logit contributions
 sturdy+0.005
 exc+0.004
 bend+0.004
 flat+0.004
onna+0.004
Neg logit contributions
ople-0.012
ipped-0.011
pa-0.011
ie-0.011
father-0.011
Token for
Token ID329
Feature activation+9.79e-01

Pos logit contributions
ople+0.006
ipped+0.006
pa+0.006
father+0.005
ebra+0.005
Neg logit contributions
 sturdy-0.003
 exc-0.003
 flat-0.003
 heal-0.003
 bend-0.003
Token you
Token ID345
Feature activation+7.92e-01

Pos logit contributions
ople+0.025
ipped+0.024
pa+0.023
ebra+0.023
father+0.023
Neg logit contributions
 sturdy-0.022
 exc-0.022
 heal-0.021
 bend-0.021
 flat-0.021
Token.
Token ID13
Feature activation-5.96e-01

Pos logit contributions
ople+0.023
ipped+0.022
pa+0.021
father+0.021
ie+0.021
Neg logit contributions
 sturdy-0.015
 exc-0.015
 flat-0.014
 bend-0.014
 heal-0.014
Token
Token ID198
Feature activation+6.28e-01

Pos logit contributions
 exc+0.011
 sturdy+0.011
 heal+0.010
onna+0.010
 sew+0.010
Neg logit contributions
ople-0.018
ipped-0.017
ebra-0.016
pa-0.016
father-0.016
Token
Token ID198
Feature activation-9.63e-01

Pos logit contributions
ople+0.019
ipped+0.018
pa+0.017
ebra+0.017
father+0.017
Neg logit contributions
 exc-0.011
 sturdy-0.011
 heal-0.011
onna-0.011
 bend-0.011
Token,
Token ID11
Feature activation-1.80e+00

Pos logit contributions
ople+0.035
ipped+0.034
ebra+0.033
eled+0.032
aked+0.031
Neg logit contributions
 sturdy-0.019
 heal-0.019
 bend-0.019
 exc-0.019
 sew-0.019
Token
Token ID6184
Feature activation+1.80e+00

Pos logit contributions
 sturdy+0.033
 bleed+0.032
 exc+0.029
 flat+0.029
onna+0.028
Neg logit contributions
pa-0.051
ople-0.051
ipped-0.050
father-0.050
mother-0.050
Token
Token ID95
Feature activation+2.89e+00

Pos logit contributions
ople+0.043
ipped+0.043
ebra+0.042
or+0.038
io+0.038
Neg logit contributions
 exc-0.043
 sturdy-0.043
 bleed-0.042
 flat-0.039
 heal-0.039
Token
Token ID26391
Feature activation+2.18e+00

Pos logit contributions
ople+0.069
ipped+0.062
ebra+0.061
father+0.060
io+0.059
Neg logit contributions
 exc-0.074
 sturdy-0.072
 bleed-0.070
 flat-0.069
 bend-0.066
Token˜
Token ID41185
Feature activation+2.48e+00

Pos logit contributions
ople+0.056
ipped+0.053
pa+0.052
father+0.051
ie+0.050
Neg logit contributions
 sturdy-0.049
 exc-0.048
 flat-0.045
 bend-0.045
 heal-0.045
TokenCan
Token ID6090
Feature activation+5.43e-01

Pos logit contributions
ople+0.065
ipped+0.061
eled+0.058
pa+0.058
father+0.058
Neg logit contributions
 sturdy-0.054
 exc-0.053
 heal-0.052
 bend-0.052
onna-0.051
Token I
Token ID314
Feature activation-8.48e-01

Pos logit contributions
ople+0.017
ipped+0.016
pa+0.016
father+0.015
ie+0.015
Neg logit contributions
 sturdy-0.009
 exc-0.009
 flat-0.009
 bleed-0.008
 bend-0.008
Token have
Token ID423
Feature activation+2.71e+00

Pos logit contributions
 sturdy+0.015
 bleed+0.014
 flat+0.014
 exc+0.013
 sil+0.013
Neg logit contributions
ople-0.025
ipped-0.024
ie-0.024
or-0.024
pa-0.024
Token one
Token ID530
Feature activation-1.27e+00

Pos logit contributions
ople+0.067
ipped+0.064
pa+0.061
father+0.061
ebra+0.060
Neg logit contributions
 sturdy-0.064
 exc-0.062
 flat-0.060
 bend-0.059
 heal-0.059
Token of
Token ID286
Feature activation+2.33e+00

Pos logit contributions
 sturdy+0.028
 exc+0.027
 bleed+0.026
 heal+0.026
 flat+0.026
Neg logit contributions
ople-0.032
ipped-0.031
ie-0.030
pa-0.030
ebra-0.029
Token those
Token ID883
Feature activation+2.00e+00

Pos logit contributions
ople+0.063
ipped+0.060
pa+0.057
father+0.057
eled+0.057
Neg logit contributions
 sturdy-0.049
 exc-0.049
 bend-0.046
 flat-0.046
 heal-0.046
Token stuck
Token ID7819
Feature activation+1.26e+00

Pos logit contributions
ople+0.016
ipped+0.015
pa+0.015
father+0.015
ebra+0.015
Neg logit contributions
 sturdy-0.008
 exc-0.008
 flat-0.008
 bend-0.008
 heal-0.008
Token in
Token ID287
Feature activation+1.93e+00

Pos logit contributions
ople+0.034
ipped+0.032
pa+0.031
father+0.031
ebra+0.030
Neg logit contributions
 sturdy-0.027
 exc-0.026
 flat-0.025
 bend-0.025
 heal-0.025
Token the
Token ID262
Feature activation+2.01e+00

Pos logit contributions
ople+0.053
ipped+0.050
pa+0.048
ebra+0.048
father+0.048
Neg logit contributions
 sturdy-0.042
 exc-0.040
 flat-0.039
 bend-0.039
 heal-0.038
Token mud
Token ID17492
Feature activation-1.06e+00

Pos logit contributions
ople+0.053
ipped+0.051
pa+0.049
father+0.049
ie+0.048
Neg logit contributions
 exc-0.043
 sturdy-0.042
onna-0.040
 heal-0.039
 bend-0.039
Token.
Token ID13
Feature activation+1.19e-01

Pos logit contributions
 sturdy+0.022
 exc+0.022
 heal+0.021
 flat+0.020
 bend+0.020
Neg logit contributions
ople-0.028
ipped-0.027
pa-0.026
father-0.026
ie-0.026
Token He
Token ID679
Feature activation+6.62e-01

Pos logit contributions
ople+0.004
ipped+0.003
pa+0.003
father+0.003
ebra+0.003
Neg logit contributions
 sturdy-0.002
 exc-0.002
 flat-0.002
 bend-0.002
onna-0.002
Token was
Token ID373
Feature activation-1.48e+00

Pos logit contributions
ople+0.016
ipped+0.015
pa+0.015
father+0.015
ebra+0.014
Neg logit contributions
 sturdy-0.016
 exc-0.016
 bend-0.015
 flat-0.015
 heal-0.015
Token scared
Token ID12008
Feature activation+1.98e-01

Pos logit contributions
 sturdy+0.028
 exc+0.027
onna+0.027
 sew+0.027
 heal+0.026
Neg logit contributions
ople-0.040
ipped-0.039
atted-0.038
eled-0.037
ed-0.037
Token and
Token ID290
Feature activation-8.42e-02

Pos logit contributions
ople+0.006
ipped+0.006
father+0.006
ebra+0.006
pa+0.006
Neg logit contributions
 sturdy-0.003
 exc-0.003
 flat-0.003
 bend-0.003
 crunch-0.003
Token alone
Token ID3436
Feature activation+1.05e+00

Pos logit contributions
 sturdy+0.002
 exc+0.002
 flat+0.002
 bend+0.002
 bleed+0.002
Neg logit contributions
ople-0.002
ipped-0.002
father-0.002
pa-0.002
ebra-0.002
Token.
Token ID13
Feature activation+9.44e-02

Pos logit contributions
ople+0.030
ipped+0.029
pa+0.028
father+0.027
ie+0.027
Neg logit contributions
 sturdy-0.021
 exc-0.020
 flat-0.019
 bend-0.019
 heal-0.019
Token three
Token ID1115
Feature activation+1.41e+00

Pos logit contributions
 sturdy+0.040
 exc+0.038
 bleed+0.037
onna+0.036
 bend+0.036
Neg logit contributions
ople-0.055
ipped-0.051
pa-0.051
ie-0.050
father-0.050
Token years
Token ID812
Feature activation-4.02e-01

Pos logit contributions
ople+0.023
ipped+0.022
father+0.021
pa+0.020
ebra+0.020
Neg logit contributions
 exc-0.044
 sturdy-0.044
 crunch-0.042
 flat-0.041
 bend-0.041
Token old
Token ID1468
Feature activation+7.76e-01

Pos logit contributions
 exc+0.008
 sturdy+0.008
 bleed+0.008
 heal+0.008
 bend+0.008
Neg logit contributions
ople-0.011
ipped-0.010
pa-0.010
ie-0.010
father-0.010
Token,
Token ID11
Feature activation-1.96e+00

Pos logit contributions
ople+0.022
ipped+0.022
ebra+0.021
father+0.021
pa+0.021
Neg logit contributions
 sturdy-0.014
 exc-0.014
 heal-0.013
 bend-0.013
 sew-0.013
Token to
Token ID284
Feature activation-5.46e-01

Pos logit contributions
 sturdy+0.041
 exc+0.040
 bleed+0.039
onna+0.039
 flat+0.038
Neg logit contributions
ople-0.051
pa-0.049
ipped-0.049
atted-0.048
ie-0.047
Token help
Token ID1037
Feature activation+5.76e-01

Pos logit contributions
 sturdy+0.008
 flat+0.007
 exc+0.007
 boun+0.007
 bleed+0.007
Neg logit contributions
ople-0.018
pa-0.018
ipped-0.017
ie-0.017
father-0.017
Token her
Token ID607
Feature activation-2.45e-01

Pos logit contributions
ople+0.013
pa+0.012
ie+0.012
el+0.012
father+0.011
Neg logit contributions
 sturdy-0.015
 exc-0.014
 boun-0.014
 flat-0.014
onna-0.014
Token.
Token ID13
Feature activation-2.21e+00

Pos logit contributions
 sturdy+0.005
onna+0.005
 flat+0.005
 exc+0.005
 boun+0.005
Neg logit contributions
ople-0.007
pa-0.006
ie-0.006
ipped-0.006
father-0.006
Token Together
Token ID17083
Feature activation+2.58e-01

Pos logit contributions
 sturdy+0.055
 exc+0.053
 bleed+0.052
 flat+0.051
 boun+0.050
Neg logit contributions
ople-0.052
ipped-0.048
pa-0.048
atted-0.046
el-0.046
Token,
Token ID11
Feature activation-2.19e+00

Pos logit contributions
ople+0.006
ipped+0.006
pa+0.006
father+0.006
ebra+0.006
Neg logit contributions
 sturdy-0.006
 exc-0.006
 flat-0.006
 bend-0.006
 heal-0.006
Token they
Token ID484
Feature activation-1.03e+00

Pos logit contributions
 sturdy+0.035
 exc+0.034
 flat+0.032
 heal+0.031
 bend+0.031
Neg logit contributions
ople-0.071
ipped-0.068
pa-0.066
father-0.065
ie-0.065
Token park
Token ID3952
Feature activation-1.80e+00

Pos logit contributions
 exc+0.077
 bleed+0.075
 sturdy+0.071
onna+0.071
 heal+0.069
Neg logit contributions
by-0.091
ie-0.089
mother-0.087
ma-0.086
ople-0.086
Token with
Token ID351
Feature activation+6.92e-01

Pos logit contributions
 sturdy+0.035
 exc+0.035
 bleed+0.033
 heal+0.033
 bend+0.032
Neg logit contributions
ople-0.051
ipped-0.049
pa-0.048
ie-0.047
ed-0.046
Token a
Token ID257
Feature activation-3.70e+00

Pos logit contributions
ople+0.021
ipped+0.021
ebra+0.020
father+0.020
pa+0.020
Neg logit contributions
 sturdy-0.013
 exc-0.012
 flat-0.012
 boun-0.012
 crunch-0.011
Token big
Token ID1263
Feature activation-1.95e+00

Pos logit contributions
 exc+0.072
 heal+0.068
 bleed+0.067
 opener+0.067
onna+0.066
Neg logit contributions
ople-0.090
by-0.090
ebra-0.090
ipped-0.088
ma-0.087
Token slide
Token ID10649
Feature activation-9.77e-01

Pos logit contributions
 exc+0.042
 sturdy+0.040
 heal+0.039
 bleed+0.039
onna+0.038
Neg logit contributions
ople-0.051
ipped-0.049
pa-0.049
ebra-0.048
el-0.046
Token.
Token ID13
Feature activation-1.27e+00

Pos logit contributions
 sturdy+0.017
 flat+0.017
 exc+0.016
 bend+0.016
 boun+0.015
Neg logit contributions
ople-0.030
ipped-0.028
father-0.028
pa-0.028
mother-0.027
Token Lily
Token ID20037
Feature activation+1.06e+00

Pos logit contributions
 sturdy+0.028
 exc+0.028
 bleed+0.026
 heal+0.026
 bend+0.026
Neg logit contributions
ople-0.033
ipped-0.031
pa-0.030
atted-0.029
father-0.029
Token went
Token ID1816
Feature activation-3.15e-01

Pos logit contributions
ople+0.026
ipped+0.024
pa+0.024
father+0.023
ie+0.023
Neg logit contributions
 sturdy-0.025
 exc-0.025
 heal-0.024
 flat-0.024
 bleed-0.023
Token up
Token ID510
Feature activation+7.34e-01

Pos logit contributions
 sturdy+0.008
 exc+0.007
 heal+0.007
 bleed+0.007
 bend+0.007
Neg logit contributions
ople-0.008
ipped-0.007
pa-0.007
father-0.007
ie-0.007
Token and
Token ID290
Feature activation-1.61e+00

Pos logit contributions
ople+0.024
ipped+0.023
father+0.023
pa+0.022
ebra+0.022
Neg logit contributions
 sturdy-0.011
 exc-0.011
 flat-0.011
 bend-0.011
onna-0.010
Token down
Token ID866
Feature activation+7.94e-01

Pos logit contributions
 sturdy+0.038
 exc+0.037
 flat+0.035
 bend+0.035
 heal+0.035
Neg logit contributions
ople-0.040
ipped-0.038
pa-0.037
father-0.036
ie-0.035
Token,
Token ID11
Feature activation-1.56e+00

Pos logit contributions
 exc+0.002
 sturdy+0.002
 bleed+0.002
 heal+0.002
onna+0.002
Neg logit contributions
les-0.003
ople-0.003
ie-0.003
ipped-0.003
el-0.002
Token bright
Token ID6016
Feature activation+6.95e-01

Pos logit contributions
 exc+0.020
 sturdy+0.020
 bleed+0.019
onna+0.019
 heal+0.018
Neg logit contributions
ople-0.052
ipped-0.050
pa-0.049
ie-0.049
les-0.049
Token orange
Token ID10912
Feature activation+1.33e+00

Pos logit contributions
ople+0.021
ipped+0.020
pa+0.020
father+0.019
atted+0.019
Neg logit contributions
 sturdy-0.012
 exc-0.012
 flat-0.011
 bend-0.011
 heal-0.011
Token carrot
Token ID40562
Feature activation-2.25e+00

Pos logit contributions
ople+0.036
ipped+0.035
pa+0.034
ie+0.034
les+0.033
Neg logit contributions
 sturdy-0.028
 exc-0.027
 bleed-0.026
 heal-0.025
 flat-0.025
Token he
Token ID339
Feature activation+9.35e-01

Pos logit contributions
 sturdy+0.051
 exc+0.051
 bleed+0.049
 heal+0.047
 flat+0.047
Neg logit contributions
ople-0.055
ipped-0.053
ed-0.052
ie-0.052
pa-0.051
Token could
Token ID714
Feature activation-3.38e+00

Pos logit contributions
ople+0.026
ipped+0.025
pa+0.024
father+0.024
atted+0.023
Neg logit contributions
 sturdy-0.019
 exc-0.019
 flat-0.018
 bend-0.018
 heal-0.018
Token find
Token ID1064
Feature activation+1.98e+00

Pos logit contributions
 sturdy+0.065
 bleed+0.060
 click+0.056
 flat+0.056
 sil+0.055
Neg logit contributions
ie-0.090
ipped-0.089
ople-0.088
el-0.086
or-0.085
Token.
Token ID13
Feature activation-1.41e+00

Pos logit contributions
ople+0.067
ipped+0.063
father+0.062
pa+0.061
ebra+0.061
Neg logit contributions
 sturdy-0.030
 exc-0.027
 crunch-0.027
 heal-0.027
 flat-0.027
Token He
Token ID679
Feature activation+4.86e-01

Pos logit contributions
 sturdy+0.033
 exc+0.032
 bleed+0.031
 heal+0.030
 flat+0.030
Neg logit contributions
ople-0.035
ipped-0.033
pa-0.031
father-0.031
atted-0.031
Token wanted
Token ID2227
Feature activation+1.38e+00

Pos logit contributions
ople+0.012
ipped+0.011
pa+0.011
father+0.011
ie+0.011
Neg logit contributions
 sturdy-0.011
 exc-0.011
 flat-0.011
 heal-0.011
 bleed-0.011
Token it
Token ID340
Feature activation-1.21e+00

Pos logit contributions
ople+0.048
ipped+0.047
ebra+0.045
pa+0.045
father+0.044
Neg logit contributions
 sturdy-0.019
 exc-0.018
 flat-0.018
 bend-0.018
 crunch-0.017
Token still
Token ID991
Feature activation-2.34e+00

Pos logit contributions
 sturdy+0.042
onna+0.038
 flat+0.038
 bleed+0.036
 boun+0.036
Neg logit contributions
ople-0.078
pa-0.076
ipped-0.075
ie-0.074
el-0.072
Token explore
Token ID7301
Feature activation+1.41e+00

Pos logit contributions
 sturdy+0.036
onna+0.034
 flat+0.033
 exc+0.032
 boun+0.032
Neg logit contributions
ople-0.073
pa-0.071
ipped-0.070
ie-0.068
father-0.068
Token,
Token ID11
Feature activation-2.37e+00

Pos logit contributions
ople+0.040
ipped+0.038
atted+0.036
ebra+0.036
eled+0.036
Neg logit contributions
 sturdy-0.030
 flat-0.028
 exc-0.027
 heal-0.027
 bend-0.027
Token but
Token ID475
Feature activation+2.90e-01

Pos logit contributions
 sturdy+0.038
 exc+0.037
 flat+0.034
 bend+0.034
 heal+0.034
Neg logit contributions
ople-0.077
ipped-0.074
pa-0.072
father-0.071
ebra-0.070
Token he
Token ID339
Feature activation+2.65e+00

Pos logit contributions
ople+0.009
ipped+0.009
ebra+0.008
pa+0.008
father+0.008
Neg logit contributions
 sturdy-0.005
 exc-0.005
 bend-0.005
 flat-0.005
 heal-0.005
Token would
Token ID561
Feature activation-2.85e+00

Pos logit contributions
ople+0.074
ipped+0.072
pa+0.068
ebra+0.067
father+0.067
Neg logit contributions
 exc-0.054
 sturdy-0.054
 bend-0.052
 heal-0.049
 flat-0.048
Token manage
Token ID6687
Feature activation+8.70e-01

Pos logit contributions
 sturdy+0.056
 exc+0.052
 flat+0.051
 boun+0.051
 bleed+0.050
Neg logit contributions
ople-0.077
pa-0.073
ie-0.072
father-0.071
mother-0.070
Token to
Token ID284
Feature activation-1.25e+00

Pos logit contributions
ople+0.025
ipped+0.024
pa+0.023
father+0.023
ebra+0.022
Neg logit contributions
 sturdy-0.017
 exc-0.017
 flat-0.016
 bend-0.016
 heal-0.016
Token stay
Token ID2652
Feature activation-7.39e-01

Pos logit contributions
onna+0.019
 sturdy+0.019
 exc+0.018
 boun+0.018
 bleed+0.018
Neg logit contributions
ople-0.037
pa-0.036
ie-0.035
ipped-0.035
father-0.035
Token close
Token ID1969
Feature activation+4.88e-01

Pos logit contributions
onna+0.014
 exc+0.014
 sturdy+0.013
 bleed+0.013
 bend+0.013
Neg logit contributions
ople-0.020
pa-0.020
ipped-0.020
ie-0.019
el-0.019
Token to
Token ID284
Feature activation+7.43e-01

Pos logit contributions
ople+0.016
ipped+0.016
ebra+0.015
father+0.015
pa+0.015
Neg logit contributions
 sturdy-0.008
 flat-0.007
 exc-0.007
 bend-0.007
onna-0.007
Token much
Token ID881
Feature activation-1.65e+00

Pos logit contributions
ople+0.021
ipped+0.020
pa+0.020
father+0.019
ie+0.019
Neg logit contributions
 sturdy-0.008
 exc-0.008
 flat-0.007
 bend-0.007
 heal-0.007
Token noise
Token ID7838
Feature activation-1.14e+00

Pos logit contributions
 sturdy+0.033
 exc+0.031
onna+0.030
 flat+0.029
 boun+0.028
Neg logit contributions
les-0.044
pa-0.044
ople-0.044
ipped-0.043
ing-0.042
Token!"
Token ID2474
Feature activation-9.78e-01

Pos logit contributions
 sturdy+0.024
 exc+0.023
 sew+0.022
 heal+0.022
 flat+0.021
Neg logit contributions
ople-0.031
pa-0.029
ipped-0.029
ie-0.029
les-0.028
Token Sara
Token ID24799
Feature activation+6.77e-01

Pos logit contributions
 sturdy+0.022
 exc+0.021
 flat+0.020
 bend+0.020
 heal+0.020
Neg logit contributions
ople-0.026
ipped-0.024
pa-0.024
father-0.023
ie-0.023
Token says
Token ID1139
Feature activation-1.31e+00

Pos logit contributions
ople+0.014
ipped+0.014
pa+0.013
atted+0.013
les+0.013
Neg logit contributions
 sturdy-0.019
 exc-0.018
 flat-0.017
 heal-0.017
onna-0.017
Token,
Token ID11
Feature activation-3.49e+00

Pos logit contributions
 sturdy+0.038
 exc+0.036
 flat+0.034
onna+0.034
 boun+0.034
Neg logit contributions
ople-0.026
ipped-0.025
atted-0.024
ed-0.024
ie-0.024
Token frown
Token ID22002
Feature activation+1.19e+00

Pos logit contributions
 sturdy+0.083
onna+0.081
 opener+0.077
 valuable+0.076
 suitable+0.071
Neg logit contributions
ople-0.082
ipped-0.078
eled-0.077
atted-0.076
ie-0.075
Tokening
Token ID278
Feature activation+1.48e+00

Pos logit contributions
ople+0.027
ipped+0.024
pa+0.024
father+0.024
ebra+0.023
Neg logit contributions
 exc-0.030
 sturdy-0.030
 bleed-0.030
 bend-0.030
 flat-0.030
Token.
Token ID13
Feature activation-1.98e+00

Pos logit contributions
ople+0.050
ipped+0.047
father+0.047
pa+0.046
ebra+0.045
Neg logit contributions
 exc-0.022
 sturdy-0.022
 flat-0.021
 bleed-0.021
 bend-0.021
Token She
Token ID1375
Feature activation-6.67e-01

Pos logit contributions
 sturdy+0.047
 exc+0.043
 heal+0.042
 flat+0.041
 bleed+0.041
Neg logit contributions
ople-0.048
ipped-0.048
pa-0.045
ed-0.045
father-0.045
Token tries
Token ID8404
Feature activation+4.81e-01

Pos logit contributions
 exc+0.017
 sturdy+0.017
 bend+0.017
 flat+0.016
 bleed+0.016
Neg logit contributions
ople-0.015
ipped-0.014
pa-0.014
father-0.013
ie-0.013
Token so
Token ID523
Feature activation-1.97e+00

Pos logit contributions
 sturdy+0.024
 exc+0.023
 flat+0.022
 bend+0.022
 heal+0.022
Neg logit contributions
ople-0.040
ipped-0.038
pa-0.037
father-0.037
ie-0.036
Token excited
Token ID6568
Feature activation+5.82e-01

Pos logit contributions
 sturdy+0.025
 exc+0.024
 flat+0.022
 bend+0.022
 heal+0.021
Neg logit contributions
ople-0.071
ipped-0.069
pa-0.067
father-0.067
ebra-0.066
Token,
Token ID11
Feature activation-1.86e+00

Pos logit contributions
ople+0.018
ipped+0.017
eled+0.017
father+0.017
ebra+0.017
Neg logit contributions
 sturdy-0.010
 exc-0.009
 bend-0.009
 flat-0.009
onna-0.009
Token she
Token ID673
Feature activation+8.99e-01

Pos logit contributions
 sturdy+0.031
 exc+0.030
 flat+0.028
 bend+0.028
 heal+0.028
Neg logit contributions
ople-0.058
ipped-0.056
pa-0.055
father-0.054
ebra-0.053
Token could
Token ID714
Feature activation-3.76e+00

Pos logit contributions
ople+0.023
ipped+0.021
pa+0.021
ie+0.020
les+0.020
Neg logit contributions
 sturdy-0.021
 exc-0.020
 heal-0.019
 bleed-0.019
 flat-0.019
Token hardly
Token ID8941
Feature activation-3.09e+00

Pos logit contributions
 sturdy+0.062
 exc+0.055
 flat+0.054
 bleed+0.054
 sil+0.053
Neg logit contributions
ople-0.116
pa-0.109
ipped-0.109
ie-0.106
ebra-0.106
Token wait
Token ID4043
Feature activation-9.17e-01

Pos logit contributions
 sturdy+0.050
 flat+0.046
 exc+0.046
 boun+0.045
 sil+0.044
Neg logit contributions
ople-0.095
pa-0.088
ipped-0.088
ie-0.087
or-0.087
Token to
Token ID284
Feature activation-2.09e+00

Pos logit contributions
 sturdy+0.017
 exc+0.017
 bleed+0.017
 heal+0.016
 sew+0.016
Neg logit contributions
ople-0.025
ipped-0.025
pa-0.024
atted-0.024
ie-0.024
Token go
Token ID467
Feature activation+6.88e-01

Pos logit contributions
 sturdy+0.034
 exc+0.032
 bleed+0.030
 flat+0.030
 heal+0.030
Neg logit contributions
ople-0.067
ipped-0.063
pa-0.063
father-0.062
ie-0.061
Token inside
Token ID2641
Feature activation-1.88e-01

Pos logit contributions
ople+0.020
ipped+0.018
father+0.018
pa+0.018
atted+0.018
Neg logit contributions
 sturdy-0.014
 exc-0.013
 flat-0.013
 bend-0.013
 heal-0.013
Token.
Token ID13
Feature activation-6.78e-01

Pos logit contributions
 sturdy+0.004
 exc+0.004
 bleed+0.004
 flat+0.004
 bend+0.004
Neg logit contributions
ople-0.005
pa-0.005
ipped-0.005
ie-0.004
father-0.004
Token him
Token ID683
Feature activation+1.71e+00

Pos logit contributions
ople+0.036
ipped+0.035
ebra+0.034
aked+0.033
atted+0.032
Neg logit contributions
 heal-0.049
 sturdy-0.049
 bend-0.048
 exc-0.047
 sew-0.047
Token it
Token ID340
Feature activation+1.36e+00

Pos logit contributions
ople+0.050
ipped+0.048
pa+0.046
father+0.046
ebra+0.045
Neg logit contributions
 sturdy-0.032
 exc-0.031
 flat-0.030
 heal-0.030
 bend-0.029
Token was
Token ID373
Feature activation-1.78e+00

Pos logit contributions
ople+0.024
ipped+0.022
pa+0.022
ie+0.021
father+0.021
Neg logit contributions
 sturdy-0.041
 exc-0.040
 flat-0.039
 bleed-0.039
onna-0.038
Token a
Token ID257
Feature activation-4.58e+00

Pos logit contributions
 bleed+0.042
 click+0.042
 sturdy+0.041
 sew+0.040
 exc+0.040
Neg logit contributions
ie-0.040
pa-0.040
el-0.036
father-0.036
mother-0.035
Token qu
Token ID627
Feature activation+1.16e+00

Pos logit contributions
 bleed+0.099
uts+0.093
 exc+0.087
 opener+0.085
onna+0.085
Neg logit contributions
by-0.105
mother-0.090
ebra-0.089
ie-0.089
 End-0.088
Tokenilt
Token ID2326
Feature activation-3.36e+00

Pos logit contributions
ople+0.030
pa+0.029
father+0.028
ipped+0.028
eled+0.027
Neg logit contributions
 sturdy-0.025
 exc-0.024
 bend-0.024
 flat-0.024
 heal-0.024
Token.
Token ID13
Feature activation-1.78e+00

Pos logit contributions
 sturdy+0.081
 exc+0.080
 bleed+0.078
 bend+0.074
onna+0.073
Neg logit contributions
ie-0.075
ople-0.074
ipped-0.071
les-0.071
pa-0.071
Token Jack
Token ID3619
Feature activation+1.74e+00

Pos logit contributions
 sturdy+0.043
 exc+0.042
 flat+0.040
 bend+0.040
 heal+0.040
Neg logit contributions
ople-0.043
ipped-0.041
pa-0.039
father-0.039
ie-0.038
Token asked
Token ID1965
Feature activation+1.25e+00

Pos logit contributions
ople+0.044
ipped+0.042
pa+0.040
father+0.040
ebra+0.039
Neg logit contributions
 sturdy-0.040
 exc-0.040
 bend-0.038
 heal-0.037
 sew-0.037
Token his
Token ID465
Feature activation+6.76e-01

Pos logit contributions
ople+0.033
ipped+0.031
ebra+0.030
atted+0.029
father+0.029
Neg logit contributions
 sturdy-0.028
 exc-0.028
 heal-0.027
 bend-0.026
 flat-0.026
Token mom
Token ID1995
Feature activation+5.84e-01

Pos logit contributions
ople+0.017
pa+0.016
ipped+0.016
father+0.016
ie+0.015
Neg logit contributions
 sturdy-0.015
 exc-0.015
 bleed-0.015
 flat-0.014
 heal-0.014