TOP ACTIVATIONS
MAX = 4.925

chooses the word " ign or ant " and the number
okay if you can 't whistle yet . It 's interesting
you teach me how to whistle too ?"
" But I like to chew gum , it 's so
pieces to try and check mate your opponent ."
that said " Do not write on the rail ". She
bird said " t weet tweet " and flew away .
songs you know and songs you make up ." <|endoftext|> L
like " t weet , tweet , tweet ". Tim and
book . But you cannot write on things that are not
and says " t weet tweet ". Ben is a lion
should be compassionate . Comp assion ate means you care about
that said " Do not whistle here ". They wondered why
t weet , tweet , tweet ". Tim and Lily think
a new word - " ignore ". She learned that it
Anna . " Can you say cat ?" Anna
. " Maybe you can cut it in half , or
and had the word " w elcome " written in big
played a game called " shoot the w affle ." In
" S ss ss ," Tom says .

MIN ACTIVATIONS
MIN = -6.387

happy ! <|endoftext|> On a sunny day , a little girl
forever . <|endoftext|> On a sunny day , a mom my
happy . <|endoftext|> On a sunny day , a little girl
after ! <|endoftext|> On a sunny day , Billy was out
together . <|endoftext|> On a sunny day , Joe decided he
best friends . On a sunny afternoon , they decided to
<|endoftext|> On a bright and sunny day , Jimmy was so
<|endoftext|> On a bright and sunny day , a little boy
. On a sunny day , a little girl
. On a sunny day , Sally and Tom
beach . It was a sunny day and Kate couldn 't
to explore . On one sunny day , she went on
and Patrick . On a sunny day , they decided to
to the park on a sunny day . Mack enzie loved
in the park on a sunny day . It was a
. On a warm and sunny day , John and his
<|endoftext|> Mark was very excited because today he got to go
On one sunny day , Jack 's mom
Everybody was outside on a sunny day when suddenly Jack noticed
garden . It was a sunny day and the grass was

INTERVAL 2.097 - 4.925
CONTAINS 1.989%

Please just lend it to me ". But
ringing the bell and making loud noises . They had so
puzzle that he wanted to put together . He tried and
You have sharp teeth and claws . You want to hurt
You should come down . Mom said not to climb the

INTERVAL -0.731 - 2.097
CONTAINS 72.906%

, " Let your shadow go on a walk . I
L ily , I like you a lot , but I
. One day she found a zip at the beach .
and very attractive . One day she was in her bedroom
toys in the attic and she put them in a different

INTERVAL -3.559 - -0.731
CONTAINS 25.069%

and the house was very busy . Everyone knew they could
We need to go home and rest ."
man . Every day he would take a big bucket outside
monkey yelled back , " But why ? I 'm trying
Lily want to play the puzzle game . They

INTERVAL -6.387 - -3.559
CONTAINS 0.035%

to the poppy in the sunny field . She put the
, Mike saw a big container of cream !
to go to the beach and play with the sand .
She only saw a dusty yard and a broken fence .
Tim my was very excited because he was going on a
Token chooses
Token ID19769
Feature activation+2.38e+00

Pos logit contributions
ps+0.069
ss+0.062
rier+0.061
mus+0.060
iddles+0.059
Neg logit contributions
 she-0.069
 he-0.061
 they-0.061
 day-0.060
 hallway-0.059
Token the
Token ID262
Feature activation+2.25e+00

Pos logit contributions
ps+0.121
ches+0.110
rier+0.108
ss+0.108
ealing+0.107
Neg logit contributions
 she-0.082
 he-0.071
 day-0.071
 they-0.068
 weekend-0.063
Token word
Token ID1573
Feature activation+1.99e+00

Pos logit contributions
ps+0.102
ches+0.096
mus+0.090
ealing+0.085
ss+0.085
Neg logit contributions
 day-0.088
 stair-0.081
 she-0.081
 weekend-0.078
 particular-0.078
Token "
Token ID366
Feature activation+2.60e+00

Pos logit contributions
ps+0.094
ss+0.081
wy+0.077
iddles+0.077
ches+0.077
Neg logit contributions
 she-0.075
 day-0.067
 he-0.064
 they-0.064
 weekend-0.064
Tokenign
Token ID570
Feature activation+3.80e+00

Pos logit contributions
ps+0.126
 teeth+0.106
ches+0.100
rier+0.098
ealing+0.097
Neg logit contributions
 she-0.093
 day-0.091
 weekend-0.088
days-0.083
 particular-0.083
Tokenor
Token ID273
Feature activation+4.92e+00

Pos logit contributions
ps+0.209
 teeth+0.184
 wounds+0.174
iddles+0.173
ags+0.173
Neg logit contributions
 day-0.109
 she-0.102
 weekend-0.097
 they-0.089
 he-0.087
Tokenant
Token ID415
Feature activation+3.64e+00

Pos logit contributions
ps+0.177
ksh+0.175
 wounds+0.165
 nicer+0.163
 unnecessary+0.162
Neg logit contributions
 she-0.188
 day-0.167
 he-0.160
 they-0.157
where-0.153
Token"
Token ID1
Feature activation+3.41e+00

Pos logit contributions
ps+0.158
 teeth+0.136
iddles+0.134
 wounds+0.131
 infected+0.126
Neg logit contributions
 she-0.140
 day-0.129
 he-0.117
 they-0.116
 weekend-0.115
Token and
Token ID290
Feature activation+1.39e+00

Pos logit contributions
ps+0.174
ealing+0.140
 teeth+0.138
ches+0.137
 unnecessary+0.137
Neg logit contributions
 she-0.132
 day-0.119
 weekend-0.112
 he-0.109
 they-0.105
Token the
Token ID262
Feature activation+1.32e+00

Pos logit contributions
ps+0.052
ss+0.047
ches+0.045
iddles+0.044
mus+0.044
Neg logit contributions
 she-0.062
 they-0.054
 he-0.054
 day-0.053
 stair-0.051
Token number
Token ID1271
Feature activation+2.40e+00

Pos logit contributions
ps+0.061
ches+0.056
mus+0.054
wy+0.052
uable+0.051
Neg logit contributions
 she-0.049
 day-0.047
 stair-0.047
 hallway-0.045
 weekend-0.043
Token okay
Token ID8788
Feature activation+5.69e-01

Pos logit contributions
ps+0.133
iddles+0.116
aff+0.115
rier+0.115
ss+0.114
Neg logit contributions
 day-0.096
 sunny-0.088
 she-0.082
 weekend-0.076
 summer-0.075
Token if
Token ID611
Feature activation+2.38e-01

Pos logit contributions
ps+0.027
ss+0.022
ches+0.021
iddles+0.021
mus+0.021
Neg logit contributions
 she-0.022
 day-0.020
 weekend-0.019
 they-0.019
 stair-0.019
Token you
Token ID345
Feature activation+2.77e+00

Pos logit contributions
ps+0.013
ss+0.012
ches+0.012
 teeth+0.011
mus+0.011
Neg logit contributions
 she-0.006
 day-0.005
 weekend-0.005
 stair-0.005
 hallway-0.005
Token can
Token ID460
Feature activation+2.44e+00

Pos logit contributions
ps+0.154
iddles+0.145
rier+0.142
ss+0.140
mus+0.140
Neg logit contributions
 wander-0.063
 day-0.063
 explore-0.062
 trip-0.058
 collect-0.055
Token't
Token ID470
Feature activation+1.69e+00

Pos logit contributions
ps+0.132
iddles+0.117
 teeth+0.117
ss+0.116
rier+0.115
Neg logit contributions
 explore-0.063
 day-0.062
 wander-0.062
 she-0.060
 visit-0.057
Token whistle
Token ID16121
Feature activation+4.82e+00

Pos logit contributions
ps+0.088
ss+0.076
iddles+0.076
 teeth+0.074
wy+0.073
Neg logit contributions
 she-0.050
 day-0.049
 explore-0.047
 wander-0.047
 he-0.045
Token yet
Token ID1865
Feature activation+2.99e+00

Pos logit contributions
ps+0.268
wy+0.233
aned+0.232
ented+0.231
iddles+0.230
Neg logit contributions
 she-0.152
 he-0.143
 day-0.142
 they-0.138
 today-0.131
Token.
Token ID13
Feature activation+5.02e-01

Pos logit contributions
ps+0.149
iddles+0.130
ss+0.129
wy+0.127
mus+0.118
Neg logit contributions
 she-0.118
 they-0.106
 day-0.102
 he-0.101
 we-0.094
Token It
Token ID632
Feature activation+1.81e+00

Pos logit contributions
ps+0.027
ss+0.024
ches+0.023
 teeth+0.023
mus+0.023
Neg logit contributions
 she-0.013
 day-0.012
 weekend-0.012
 stair-0.012
 hallway-0.012
Token's
Token ID338
Feature activation+2.40e+00

Pos logit contributions
ps+0.086
iddles+0.077
mus+0.076
ss+0.073
wy+0.071
Neg logit contributions
 she-0.061
 day-0.056
 they-0.054
 he-0.051
 wander-0.050
Token interesting
Token ID3499
Feature activation+2.36e+00

Pos logit contributions
ps+0.110
ss+0.096
iddles+0.096
aff+0.094
rier+0.094
Neg logit contributions
 day-0.097
 sunny-0.086
 she-0.085
 weekend-0.079
 they-0.076
Token you
Token ID345
Feature activation+1.78e+00

Pos logit contributions
 hallway+0.039
 weekend+0.036
 stair+0.035
 driveway+0.033
 holiday+0.032
Neg logit contributions
ps-0.056
ss-0.055
ches-0.052
 teeth-0.051
mus-0.051
Token teach
Token ID4545
Feature activation+2.44e+00

Pos logit contributions
ps+0.098
ss+0.086
iddles+0.085
 teeth+0.084
wy+0.083
Neg logit contributions
 she-0.049
 explore-0.047
 day-0.045
 he-0.045
 wander-0.044
Token me
Token ID502
Feature activation+1.98e+00

Pos logit contributions
ps+0.117
rier+0.107
wy+0.104
ealing+0.101
ows+0.099
Neg logit contributions
 she-0.092
 day-0.085
 explore-0.078
 they-0.075
 visit-0.075
Token how
Token ID703
Feature activation+1.45e+00

Pos logit contributions
ps+0.098
ss+0.087
wy+0.082
mus+0.082
rier+0.081
Neg logit contributions
 she-0.062
 day-0.062
 he-0.058
 of-0.057
 weekend-0.057
Token to
Token ID284
Feature activation+1.78e+00

Pos logit contributions
ps+0.089
ss+0.075
iddles+0.075
wy+0.074
ches+0.073
Neg logit contributions
 she-0.037
 day-0.036
 weekend-0.032
 they-0.031
 he-0.031
Token whistle
Token ID16121
Feature activation+4.82e+00

Pos logit contributions
ps+0.092
ss+0.085
 teeth+0.082
iddles+0.082
ussy+0.080
Neg logit contributions
 day-0.049
 she-0.049
 explore-0.049
 wander-0.047
 he-0.044
Token too
Token ID1165
Feature activation+2.76e+00

Pos logit contributions
ps+0.283
ented+0.246
ussy+0.246
wy+0.246
aned+0.242
Neg logit contributions
 day-0.149
 he-0.128
 sunny-0.123
 she-0.120
 today-0.120
Token?"
Token ID1701
Feature activation+4.77e-01

Pos logit contributions
ps+0.167
 wounds+0.136
iddles+0.135
wy+0.135
ss+0.130
Neg logit contributions
 day-0.084
 of-0.073
 weekend-0.072
 sunny-0.071
 he-0.070
Token
Token ID220
Feature activation+1.57e+00

Pos logit contributions
ps+0.026
ss+0.022
ches+0.021
 teeth+0.021
mus+0.021
Neg logit contributions
 she-0.015
 day-0.013
 weekend-0.013
 they-0.012
 stair-0.012
Token
Token ID198
Feature activation+1.80e+00

Pos logit contributions
ps+0.093
ss+0.078
 teeth+0.078
mus+0.077
wy+0.077
Neg logit contributions
 she-0.040
 day-0.034
 they-0.033
 weekend-0.033
 he-0.031
Token
Token ID198
Feature activation+6.00e-01

Pos logit contributions
ps+0.125
 teeth+0.106
wy+0.103
rier+0.103
ss+0.101
Neg logit contributions
 she-0.042
 they-0.032
 day-0.030
 he-0.030
 weekend-0.027
Token "
Token ID366
Feature activation-6.95e-01

Pos logit contributions
ps+0.084
wy+0.076
ches+0.074
ws+0.073
 unnecessary+0.072
Neg logit contributions
 she-0.058
 they-0.052
 he-0.051
 day-0.045
 stair-0.042
TokenBut
Token ID1537
Feature activation-3.92e-01

Pos logit contributions
 weekend+0.019
 stair+0.019
 hallway+0.019
 holiday+0.019
 she+0.019
Neg logit contributions
ps-0.033
ss-0.032
mus-0.030
ches-0.029
 teeth-0.028
Token I
Token ID314
Feature activation+4.86e-01

Pos logit contributions
 weekend+0.010
 hallway+0.010
 she+0.010
 stair+0.009
 day+0.009
Neg logit contributions
ps-0.021
ss-0.019
 teeth-0.018
ches-0.018
mus-0.018
Token like
Token ID588
Feature activation+1.65e+00

Pos logit contributions
ps+0.026
ss+0.022
mus+0.021
iddles+0.021
ches+0.021
Neg logit contributions
 she-0.014
 day-0.014
 weekend-0.013
 stair-0.012
 hallway-0.012
Token to
Token ID284
Feature activation+8.38e-01

Pos logit contributions
ps+0.081
ows+0.067
ealing+0.066
rier+0.066
iddles+0.066
Neg logit contributions
 she-0.062
 he-0.057
 day-0.057
 they-0.055
 weekend-0.050
Token chew
Token ID34722
Feature activation+4.81e+00

Pos logit contributions
ps+0.047
ss+0.041
iddles+0.041
mus+0.040
ches+0.040
Neg logit contributions
 she-0.021
 day-0.020
 explore-0.019
 wander-0.018
 weekend-0.018
Token gum
Token ID27142
Feature activation+2.74e+00

Pos logit contributions
ps+0.286
iddles+0.257
ows+0.252
amus+0.250
aned+0.241
Neg logit contributions
 he-0.131
 she-0.126
 day-0.108
 they-0.107
 today-0.090
Token,
Token ID11
Feature activation-8.21e-01

Pos logit contributions
ps+0.124
ss+0.108
iddles+0.108
 unnecessary+0.107
ealing+0.104
Neg logit contributions
 she-0.105
 day-0.099
 he-0.095
 they-0.094
 today-0.088
Token it
Token ID340
Feature activation+1.61e+00

Pos logit contributions
 hallway+0.026
 weekend+0.026
 she+0.024
days+0.024
 day+0.024
Neg logit contributions
ps-0.037
ss-0.034
 teeth-0.033
ches-0.032
mus-0.031
Token's
Token ID338
Feature activation+2.37e+00

Pos logit contributions
ps+0.072
iddles+0.065
mus+0.062
rier+0.059
ags+0.059
Neg logit contributions
 she-0.058
 day-0.054
 he-0.052
 they-0.049
 stair-0.048
Token so
Token ID523
Feature activation+1.87e+00

Pos logit contributions
ps+0.109
iddles+0.100
rier+0.096
rots+0.096
ino+0.093
Neg logit contributions
 day-0.096
 sunny-0.089
 she-0.083
 he-0.078
 summer-0.078
Token pieces
Token ID5207
Feature activation-4.78e-01

Pos logit contributions
ps+0.039
ss+0.034
mus+0.032
ches+0.031
iddles+0.031
Neg logit contributions
 day-0.035
 she-0.035
 stair-0.034
 weekend-0.034
 hallway-0.033
Token to
Token ID284
Feature activation+1.04e+00

Pos logit contributions
 hallway+0.015
 weekend+0.015
 she+0.015
 stair+0.015
 day+0.015
Neg logit contributions
ps-0.022
ss-0.019
ches-0.019
 teeth-0.018
mus-0.018
Token try
Token ID1949
Feature activation+3.50e+00

Pos logit contributions
ps+0.055
ss+0.050
 teeth+0.048
iddles+0.047
ches+0.047
Neg logit contributions
 she-0.028
 explore-0.026
 day-0.026
 particular-0.025
 stair-0.025
Token and
Token ID290
Feature activation-7.42e-01

Pos logit contributions
ps+0.183
wy+0.164
ss+0.156
ealing+0.155
aned+0.153
Neg logit contributions
 she-0.112
 they-0.106
 day-0.094
 he-0.093
 stair-0.090
Token check
Token ID2198
Feature activation+2.89e+00

Pos logit contributions
 weekend+0.025
 hallway+0.025
 stair+0.025
 she+0.024
 day+0.024
Neg logit contributions
ps-0.033
ss-0.029
 teeth-0.028
ches-0.028
mus-0.027
Tokenmate
Token ID9830
Feature activation+4.75e+00

Pos logit contributions
ps+0.141
 teeth+0.124
ss+0.123
ache+0.122
ksh+0.120
Neg logit contributions
 she-0.096
 they-0.091
 day-0.082
 he-0.079
 stair-0.078
Token your
Token ID534
Feature activation+2.15e+00

Pos logit contributions
ps+0.220
 infected+0.192
obb+0.189
wy+0.188
aned+0.187
Neg logit contributions
 she-0.167
 they-0.163
 he-0.145
 we-0.132
 day-0.131
Token opponent
Token ID6125
Feature activation+2.59e+00

Pos logit contributions
ps+0.096
mus+0.087
wy+0.084
ches+0.083
ache+0.082
Neg logit contributions
 day-0.080
 stair-0.071
 holiday-0.069
 hill-0.069
 trip-0.069
Token."
Token ID526
Feature activation-5.90e-01

Pos logit contributions
ps+0.102
ss+0.093
iddles+0.086
mus+0.086
 teeth+0.081
Neg logit contributions
 she-0.103
 day-0.098
 they-0.094
 weekend-0.093
 he-0.092
Token
Token ID198
Feature activation+3.18e-01

Pos logit contributions
 hallway+0.021
 weekend+0.021
 stair+0.021
 day+0.020
 she+0.020
Neg logit contributions
ps-0.024
ss-0.022
 teeth-0.021
mus-0.020
ches-0.020
Token
Token ID198
Feature activation-5.98e-01

Pos logit contributions
ps+0.016
ss+0.014
ches+0.014
 teeth+0.014
mus+0.013
Neg logit contributions
 she-0.010
 weekend-0.009
 day-0.009
 stair-0.009
 hallway-0.009
Token that
Token ID326
Feature activation+1.08e+00

Pos logit contributions
ps+0.008
ss+0.007
ches+0.006
 teeth+0.006
mus+0.006
Neg logit contributions
 she-0.005
 weekend-0.005
 day-0.005
 stair-0.005
 hallway-0.005
Token said
Token ID531
Feature activation+2.06e+00

Pos logit contributions
ps+0.054
ss+0.047
ches+0.045
mus+0.045
ealing+0.045
Neg logit contributions
 she-0.037
 day-0.034
 weekend-0.032
 they-0.032
 stair-0.032
Token "
Token ID366
Feature activation+2.00e+00

Pos logit contributions
ps+0.104
wy+0.099
ches+0.095
ksh+0.090
 unnecessary+0.089
Neg logit contributions
 she-0.076
 they-0.065
 he-0.061
 stair-0.056
 day-0.055
TokenDo
Token ID5211
Feature activation+3.88e-01

Pos logit contributions
ps+0.101
 teeth+0.086
ealing+0.084
ches+0.083
rier+0.081
Neg logit contributions
 she-0.065
 day-0.064
 weekend-0.063
 particular-0.059
days-0.058
Token not
Token ID407
Feature activation+7.44e-01

Pos logit contributions
ps+0.017
ss+0.014
mus+0.014
ches+0.014
 teeth+0.014
Neg logit contributions
 she-0.015
 day-0.014
 stair-0.014
 weekend-0.014
 they-0.013
Token write
Token ID3551
Feature activation+4.74e+00

Pos logit contributions
ps+0.039
ches+0.034
ss+0.033
iddles+0.032
wy+0.032
Neg logit contributions
 she-0.022
 day-0.022
 stair-0.020
 weekend-0.020
 wander-0.019
Token on
Token ID319
Feature activation+2.87e+00

Pos logit contributions
ps+0.232
ksh+0.217
 wounds+0.209
 unnecessary+0.209
wy+0.205
Neg logit contributions
 she-0.145
 today-0.121
 day-0.119
 they-0.116
 of-0.107
Token the
Token ID262
Feature activation+2.03e+00

Pos logit contributions
ps+0.160
ss+0.139
iddles+0.138
ealing+0.137
wy+0.136
Neg logit contributions
 she-0.079
 stair-0.069
 day-0.068
 they-0.064
 holiday-0.062
Token rail
Token ID6787
Feature activation+1.45e+00

Pos logit contributions
ps+0.126
iddles+0.119
wy+0.119
ussy+0.117
ealing+0.116
Neg logit contributions
 day-0.042
 stair-0.040
 hallway-0.038
 hill-0.036
 playground-0.034
Token".
Token ID1911
Feature activation+6.88e-01

Pos logit contributions
ps+0.052
 teeth+0.042
ss+0.042
mus+0.040
iddles+0.040
Neg logit contributions
 she-0.065
 day-0.063
 weekend-0.062
 stair-0.062
 hallway-0.060
Token She
Token ID1375
Feature activation+1.76e+00

Pos logit contributions
ps+0.040
ss+0.035
ches+0.035
wy+0.034
mus+0.034
Neg logit contributions
 she-0.018
 stair-0.015
 day-0.015
 weekend-0.015
 hallway-0.015
Token bird
Token ID6512
Feature activation+5.00e-01

Pos logit contributions
ps+0.027
ches+0.024
ss+0.023
mus+0.023
 teeth+0.022
Neg logit contributions
 she-0.019
 day-0.018
 weekend-0.018
 stair-0.017
 hallway-0.017
Token said
Token ID531
Feature activation+4.98e-01

Pos logit contributions
ps+0.027
ss+0.024
iddles+0.023
 teeth+0.023
mus+0.022
Neg logit contributions
 she-0.014
 day-0.013
 hallway-0.012
 weekend-0.012
 stair-0.012
Token "
Token ID366
Feature activation+2.48e-01

Pos logit contributions
ps+0.027
ss+0.023
ches+0.023
iddles+0.023
wy+0.022
Neg logit contributions
 she-0.015
 day-0.013
 they-0.013
 he-0.013
 stair-0.013
Tokent
Token ID83
Feature activation-1.07e+00

Pos logit contributions
ps+0.013
ss+0.012
 teeth+0.011
ches+0.011
mus+0.011
Neg logit contributions
 she-0.007
 weekend-0.006
 day-0.006
 hallway-0.006
 stair-0.006
Tokenweet
Token ID7277
Feature activation+1.83e+00

Pos logit contributions
 weekend+0.042
 hallway+0.041
 she+0.040
 driveway+0.040
 particular+0.039
Neg logit contributions
ps-0.040
 teeth-0.036
ss-0.036
mus-0.032
ches-0.032
Token tweet
Token ID6126
Feature activation+4.73e+00

Pos logit contributions
ps+0.082
iddles+0.076
wy+0.075
 teeth+0.075
mus+0.074
Neg logit contributions
 she-0.059
 day-0.055
 they-0.051
 stair-0.049
 wander-0.048
Token"
Token ID1
Feature activation+2.76e+00

Pos logit contributions
ps+0.285
 wounds+0.282
ags+0.260
ws+0.255
wy+0.254
Neg logit contributions
 she-0.114
 of-0.112
 day-0.107
 they-0.094
where-0.093
Token and
Token ID290
Feature activation+1.20e+00

Pos logit contributions
ps+0.150
ealing+0.124
wy+0.123
ws+0.123
mus+0.123
Neg logit contributions
 she-0.092
 day-0.086
 he-0.081
 weekend-0.076
 they-0.072
Token flew
Token ID13112
Feature activation+1.11e+00

Pos logit contributions
ps+0.059
ss+0.054
ches+0.053
iddles+0.052
wy+0.051
Neg logit contributions
 she-0.041
 they-0.036
 he-0.035
 day-0.033
 stair-0.032
Token away
Token ID1497
Feature activation+6.89e-01

Pos logit contributions
ps+0.044
ss+0.038
wy+0.037
iddles+0.037
ealing+0.037
Neg logit contributions
 she-0.047
 day-0.044
 he-0.042
 they-0.040
 weekend-0.040
Token.
Token ID13
Feature activation+1.48e+00

Pos logit contributions
ps+0.030
ss+0.028
ches+0.027
wy+0.026
ealing+0.025
Neg logit contributions
 she-0.026
 day-0.024
 stair-0.023
 he-0.023
 hallway-0.023
Token songs
Token ID7259
Feature activation+1.80e+00

Pos logit contributions
ps+0.189
wy+0.179
ss+0.175
aned+0.166
ussy+0.165
Neg logit contributions
 she-0.099
 they-0.086
 day-0.084
 sunny-0.077
 today-0.075
Token you
Token ID345
Feature activation+4.54e+00

Pos logit contributions
ps+0.092
ss+0.085
 teeth+0.082
iddles+0.079
wy+0.078
Neg logit contributions
 she-0.059
 day-0.052
 they-0.050
 weekend-0.048
 he-0.046
Token know
Token ID760
Feature activation+2.39e+00

Pos logit contributions
ps+0.245
rier+0.238
iddles+0.232
ss+0.231
 infected+0.228
Neg logit contributions
 day-0.121
 she-0.118
 they-0.102
 visit-0.100
 explore-0.099
Token and
Token ID290
Feature activation+1.72e-01

Pos logit contributions
ps+0.114
ss+0.104
ealing+0.098
wy+0.097
iddles+0.097
Neg logit contributions
 she-0.087
 they-0.078
 day-0.077
 he-0.074
 holiday-0.071
Token songs
Token ID7259
Feature activation+2.40e+00

Pos logit contributions
ps+0.008
ss+0.007
ches+0.006
 teeth+0.006
mus+0.006
Neg logit contributions
 she-0.006
 weekend-0.006
 day-0.006
 stair-0.006
 hallway-0.006
Token you
Token ID345
Feature activation+4.72e+00

Pos logit contributions
ps+0.109
ss+0.100
 teeth+0.097
iddles+0.094
wy+0.094
Neg logit contributions
 she-0.093
 day-0.084
 they-0.083
 he-0.076
 weekend-0.075
Token make
Token ID787
Feature activation+2.74e+00

Pos logit contributions
rier+0.288
ss+0.269
ps+0.268
iddles+0.264
 infected+0.264
Neg logit contributions
 day-0.093
 explore-0.087
 visit-0.082
 have-0.081
 she-0.078
Token up
Token ID510
Feature activation+1.87e+00

Pos logit contributions
ps+0.131
wy+0.127
ss+0.122
uable+0.116
ache+0.115
Neg logit contributions
 she-0.093
 day-0.088
 they-0.080
 holiday-0.079
 he-0.075
Token."
Token ID526
Feature activation-7.63e-01

Pos logit contributions
ps+0.080
ss+0.075
wy+0.069
ches+0.068
uable+0.067
Neg logit contributions
 she-0.073
 day-0.073
 weekend-0.067
 sunny-0.066
 they-0.066
Token<|endoftext|>
Token ID50256
Feature activation+6.98e-01

Pos logit contributions
 hallway+0.031
 weekend+0.031
 stair+0.031
 holiday+0.029
 exploration+0.029
Neg logit contributions
ps-0.027
ss-0.025
 teeth-0.024
mus-0.022
ches-0.022
TokenL
Token ID43
Feature activation+2.32e-02

Pos logit contributions
ps+0.034
ss+0.029
 teeth+0.028
ches+0.028
mus+0.027
Neg logit contributions
 she-0.023
 day-0.023
 weekend-0.022
 stair-0.021
 hallway-0.021
Token like
Token ID588
Feature activation+1.04e+00

Pos logit contributions
ps+0.026
ss+0.024
ches+0.022
mus+0.022
 teeth+0.022
Neg logit contributions
 she-0.025
 day-0.023
 they-0.022
 weekend-0.022
 he-0.021
Token "
Token ID366
Feature activation+1.01e+00

Pos logit contributions
ps+0.045
ss+0.040
ches+0.040
wy+0.038
mus+0.038
Neg logit contributions
 she-0.040
 day-0.037
 they-0.036
 weekend-0.036
 holiday-0.034
Tokent
Token ID83
Feature activation+4.28e-01

Pos logit contributions
ps+0.047
 teeth+0.040
ches+0.039
ss+0.039
iddles+0.038
Neg logit contributions
 she-0.035
 day-0.034
 weekend-0.033
days-0.032
 particular-0.031
Tokenweet
Token ID7277
Feature activation+2.84e+00

Pos logit contributions
ps+0.021
ss+0.018
 teeth+0.018
ches+0.018
iddles+0.017
Neg logit contributions
 she-0.014
 day-0.013
 weekend-0.013
 stair-0.013
 hallway-0.012
Token,
Token ID11
Feature activation+5.00e-01

Pos logit contributions
ussy+0.119
ps+0.119
iddles+0.118
wy+0.114
 teeth+0.110
Neg logit contributions
 she-0.095
 day-0.093
 they-0.083
where-0.082
 wander-0.081
Token tweet
Token ID6126
Feature activation+4.71e+00

Pos logit contributions
ps+0.024
ss+0.021
iddles+0.021
ches+0.021
mus+0.020
Neg logit contributions
 she-0.016
 day-0.015
 stair-0.015
 weekend-0.015
 they-0.014
Token,
Token ID11
Feature activation+5.57e-01

Pos logit contributions
ps+0.277
 wounds+0.252
wy+0.244
asy+0.242
iddles+0.242
Neg logit contributions
 day-0.128
 of-0.123
 she-0.117
where-0.113
 they-0.105
Token tweet
Token ID6126
Feature activation+4.62e+00

Pos logit contributions
ps+0.024
ss+0.020
iddles+0.020
ches+0.020
 teeth+0.020
Neg logit contributions
 she-0.021
 day-0.020
 weekend-0.020
 stair-0.020
 hallway-0.019
Token".
Token ID1911
Feature activation+5.17e-01

Pos logit contributions
ps+0.274
 wounds+0.246
wy+0.242
asy+0.242
ussy+0.236
Neg logit contributions
 day-0.122
 of-0.118
 she-0.113
where-0.108
 they-0.101
Token Tim
Token ID5045
Feature activation+2.23e+00

Pos logit contributions
ps+0.027
ss+0.024
ches+0.023
 teeth+0.023
mus+0.023
Neg logit contributions
 she-0.016
 day-0.014
 they-0.014
 stair-0.013
 weekend-0.013
Token and
Token ID290
Feature activation+8.74e-02

Pos logit contributions
rier+0.123
mus+0.120
 unnecessary+0.120
ps+0.118
uable+0.116
Neg logit contributions
 she-0.045
 sandbox-0.040
 hill-0.038
 day-0.038
 slide-0.038
Token book
Token ID1492
Feature activation+1.89e+00

Pos logit contributions
ps+0.033
ss+0.028
iddles+0.027
 teeth+0.027
mus+0.027
Neg logit contributions
 she-0.024
 stair-0.023
 day-0.023
 weekend-0.023
 hallway-0.023
Token.
Token ID13
Feature activation+9.54e-01

Pos logit contributions
ps+0.091
ss+0.079
 teeth+0.076
ealing+0.075
iddles+0.075
Neg logit contributions
 she-0.069
 day-0.063
 weekend-0.062
 stair-0.060
 they-0.059
Token But
Token ID887
Feature activation-7.51e-02

Pos logit contributions
ps+0.048
ches+0.042
ss+0.041
 teeth+0.039
mus+0.039
Neg logit contributions
 she-0.029
 weekend-0.026
 day-0.026
 stair-0.026
 hallway-0.026
Token you
Token ID345
Feature activation+1.09e+00

Pos logit contributions
 she+0.002
 weekend+0.002
 hallway+0.002
 stair+0.002
 day+0.002
Neg logit contributions
ps-0.004
ss-0.003
 teeth-0.003
ches-0.003
mus-0.003
Token cannot
Token ID2314
Feature activation+4.37e-01

Pos logit contributions
ps+0.069
ss+0.063
rier+0.063
iddles+0.061
mus+0.061
Neg logit contributions
 she-0.018
 day-0.017
 explore-0.015
 weekend-0.015
 they-0.015
Token write
Token ID3551
Feature activation+4.70e+00

Pos logit contributions
ps+0.022
ss+0.020
ches+0.019
 teeth+0.019
mus+0.019
Neg logit contributions
 she-0.013
 day-0.012
 weekend-0.012
 stair-0.012
 hallway-0.012
Token on
Token ID319
Feature activation+3.20e+00

Pos logit contributions
ksh+0.246
ps+0.231
aned+0.225
ussy+0.221
igg+0.219
Neg logit contributions
 she-0.138
 they-0.114
 he-0.113
 today-0.107
 there-0.100
Token things
Token ID1243
Feature activation+1.72e+00

Pos logit contributions
ps+0.178
ealing+0.160
ss+0.158
aned+0.158
ussy+0.157
Neg logit contributions
 she-0.087
 day-0.071
 they-0.070
 he-0.067
 stair-0.067
Token that
Token ID326
Feature activation+5.52e-01

Pos logit contributions
ps+0.092
ss+0.083
 teeth+0.081
iddles+0.080
mus+0.078
Neg logit contributions
 she-0.056
 they-0.047
 day-0.045
 he-0.044
 stair-0.041
Token are
Token ID389
Feature activation+2.44e+00

Pos logit contributions
ps+0.028
ss+0.024
ches+0.024
mus+0.024
iddles+0.023
Neg logit contributions
 she-0.017
 day-0.016
 weekend-0.015
 stair-0.015
 hallway-0.015
Token not
Token ID407
Feature activation+2.23e+00

Pos logit contributions
ps+0.112
ealing+0.107
ss+0.106
iddles+0.104
mus+0.103
Neg logit contributions
 she-0.083
 day-0.078
 they-0.078
 he-0.071
 sunny-0.069
Token and
Token ID290
Feature activation-2.73e-02

Pos logit contributions
ps+0.033
ss+0.030
mus+0.028
 teeth+0.028
iddles+0.028
Neg logit contributions
 she-0.025
 day-0.022
 weekend-0.022
 they-0.021
 stair-0.021
Token says
Token ID1139
Feature activation+1.54e+00

Pos logit contributions
 she+0.001
 weekend+0.001
 day+0.001
 stair+0.001
 hallway+0.001
Neg logit contributions
ps-0.001
ss-0.001
ches-0.001
 teeth-0.001
mus-0.001
Token "
Token ID366
Feature activation+5.96e-01

Pos logit contributions
ps+0.089
wy+0.080
ches+0.079
ss+0.077
iddles+0.076
Neg logit contributions
 she-0.047
 they-0.040
 he-0.037
 day-0.033
 stair-0.033
Tokent
Token ID83
Feature activation-1.29e-01

Pos logit contributions
ps+0.029
ss+0.025
 teeth+0.024
ches+0.024
iddles+0.023
Neg logit contributions
 she-0.019
 weekend-0.019
 day-0.019
 hallway-0.018
 stair-0.017
Tokenweet
Token ID7277
Feature activation+2.33e+00

Pos logit contributions
 she+0.005
 weekend+0.005
 hallway+0.004
 stair+0.004
 day+0.004
Neg logit contributions
ps-0.006
ss-0.005
ches-0.005
 teeth-0.005
mus-0.005
Token tweet
Token ID6126
Feature activation+4.70e+00

Pos logit contributions
ps+0.095
ussy+0.089
wy+0.089
 infected+0.088
iddles+0.087
Neg logit contributions
 she-0.079
 day-0.074
 they-0.072
 wander-0.068
 explore-0.067
Token".
Token ID1911
Feature activation+8.42e-01

Pos logit contributions
ps+0.248
 wounds+0.245
wy+0.220
ags+0.218
ws+0.213
Neg logit contributions
 she-0.132
 of-0.131
 day-0.129
 they-0.128
where-0.116
Token Ben
Token ID3932
Feature activation+2.30e+00

Pos logit contributions
ps+0.045
ss+0.040
ches+0.039
wy+0.038
mus+0.038
Neg logit contributions
 she-0.027
 they-0.023
 stair-0.022
 day-0.022
 weekend-0.022
Token is
Token ID318
Feature activation+1.73e+00

Pos logit contributions
rier+0.096
ps+0.096
ss+0.094
mus+0.094
iddles+0.092
Neg logit contributions
 she-0.078
 they-0.070
 wander-0.066
 day-0.066
 he-0.065
Token a
Token ID257
Feature activation+6.34e-01

Pos logit contributions
ss+0.068
ps+0.068
ches+0.066
uable+0.065
mus+0.064
Neg logit contributions
 she-0.072
 day-0.061
 sunny-0.061
 they-0.061
 he-0.060
Token lion
Token ID18744
Feature activation-3.05e-01

Pos logit contributions
ps+0.025
ss+0.021
ches+0.020
iddles+0.020
mus+0.020
Neg logit contributions
 she-0.027
 day-0.027
 weekend-0.026
 hallway-0.025
 stair-0.025
Token should
Token ID815
Feature activation-1.31e+00

Pos logit contributions
ps+0.033
ss+0.029
mus+0.029
iddles+0.029
 teeth+0.028
Neg logit contributions
 she-0.017
 day-0.016
 they-0.015
 weekend-0.015
 stair-0.014
Token be
Token ID307
Feature activation+1.89e+00

Pos logit contributions
 hallway+0.060
 stair+0.059
 weekend+0.059
days+0.055
 particular+0.054
Neg logit contributions
ps-0.050
ss-0.040
 teeth-0.036
ches-0.036
uable-0.033
Token compassionate
Token ID32533
Feature activation+9.57e-01

Pos logit contributions
ches+0.076
ps+0.071
iddles+0.071
osaurus+0.071
oni+0.069
Neg logit contributions
 she-0.075
 they-0.069
 day-0.069
 he-0.065
 sunny-0.064
Token.
Token ID13
Feature activation-4.90e-01

Pos logit contributions
ps+0.039
ches+0.034
mus+0.034
ss+0.034
iddles+0.032
Neg logit contributions
 she-0.037
 day-0.036
 weekend-0.033
 stair-0.033
 they-0.032
Token Comp
Token ID3082
Feature activation+2.46e+00

Pos logit contributions
 weekend+0.018
 stair+0.017
 hallway+0.017
 she+0.017
 holiday+0.017
Neg logit contributions
ps-0.021
ss-0.019
 teeth-0.018
mus-0.017
ches-0.017
Tokenassion
Token ID11857
Feature activation+4.66e+00

Pos logit contributions
ps+0.098
 teeth+0.095
rier+0.079
ze+0.076
 degrees+0.076
Neg logit contributions
ite-0.084
boat-0.079
 weekend-0.077
 day-0.077
 she-0.076
Tokenate
Token ID378
Feature activation+1.21e+00

Pos logit contributions
ps+0.225
ache+0.207
ches+0.207
 unnecessary+0.204
 infected+0.202
Neg logit contributions
 day-0.131
 of-0.121
where-0.099
 they-0.096
 hill-0.094
Token means
Token ID1724
Feature activation+4.93e-01

Pos logit contributions
ps+0.051
mus+0.045
ss+0.044
iddles+0.042
 teeth+0.042
Neg logit contributions
 she-0.047
 day-0.044
 they-0.041
 stair-0.040
 weekend-0.040
Token you
Token ID345
Feature activation+2.00e+00

Pos logit contributions
ps+0.025
ss+0.023
ches+0.022
mus+0.022
 teeth+0.021
Neg logit contributions
 she-0.015
 day-0.014
 weekend-0.013
 stair-0.013
 they-0.013
Token care
Token ID1337
Feature activation+2.50e+00

Pos logit contributions
ps+0.104
rier+0.099
mus+0.098
iddles+0.098
ss+0.093
Neg logit contributions
 she-0.057
 day-0.054
 they-0.054
 he-0.048
 explore-0.045
Token about
Token ID546
Feature activation+2.13e+00

Pos logit contributions
ps+0.125
wy+0.116
mus+0.114
ss+0.114
 teeth+0.111
Neg logit contributions
 she-0.074
 day-0.072
 of-0.069
 they-0.064
 he-0.062
Token that
Token ID326
Feature activation+1.17e-01

Pos logit contributions
days+0.068
 hallway+0.066
 stair+0.065
 wander+0.065
 weekend+0.065
Neg logit contributions
ps-0.069
ss-0.058
ches-0.056
 louder-0.056
 teeth-0.054
Token said
Token ID531
Feature activation+5.57e-01

Pos logit contributions
ps+0.005
ss+0.004
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.005
 weekend-0.004
 day-0.004
 hallway-0.004
 stair-0.004
Token "
Token ID366
Feature activation+9.10e-01

Pos logit contributions
ps+0.024
ss+0.021
ches+0.021
mus+0.020
 teeth+0.020
Neg logit contributions
 she-0.021
 day-0.019
 weekend-0.019
 stair-0.019
 they-0.019
TokenDo
Token ID5211
Feature activation-1.60e-01

Pos logit contributions
ps+0.046
 teeth+0.039
ss+0.039
ches+0.038
iddles+0.037
Neg logit contributions
 she-0.028
 weekend-0.028
 day-0.027
 hallway-0.026
 particular-0.026
Token not
Token ID407
Feature activation-6.45e-02

Pos logit contributions
 she+0.006
 hallway+0.006
 weekend+0.006
 stair+0.006
 day+0.006
Neg logit contributions
ps-0.006
ss-0.006
ches-0.005
 teeth-0.005
mus-0.005
Token whistle
Token ID16121
Feature activation+4.62e+00

Pos logit contributions
 she+0.002
 weekend+0.002
 hallway+0.002
 stair+0.002
 day+0.002
Neg logit contributions
ps-0.003
ss-0.003
 teeth-0.003
ches-0.003
mus-0.003
Token here
Token ID994
Feature activation+2.68e+00

Pos logit contributions
ps+0.234
 wounds+0.206
 infected+0.201
iddles+0.199
wy+0.198
Neg logit contributions
 day-0.161
 they-0.147
 she-0.140
 he-0.139
 today-0.138
Token".
Token ID1911
Feature activation-4.84e-01

Pos logit contributions
ps+0.123
ss+0.106
 teeth+0.100
iddles+0.100
wy+0.098
Neg logit contributions
 she-0.101
 day-0.100
 weekend-0.099
 they-0.093
 holiday-0.092
Token They
Token ID1119
Feature activation+3.03e-01

Pos logit contributions
 weekend+0.015
 she+0.015
 day+0.015
 stair+0.015
 hallway+0.015
Neg logit contributions
ps-0.022
ss-0.020
 teeth-0.019
ches-0.019
mus-0.019
Token wondered
Token ID14028
Feature activation+1.97e-01

Pos logit contributions
ps+0.016
ss+0.014
ches+0.014
 teeth+0.014
mus+0.013
Neg logit contributions
 she-0.009
 weekend-0.008
 day-0.008
 hallway-0.008
 stair-0.008
Token why
Token ID1521
Feature activation+9.51e-01

Pos logit contributions
ps+0.009
ss+0.008
ches+0.008
 teeth+0.008
mus+0.008
Neg logit contributions
 she-0.007
 weekend-0.006
 day-0.006
 stair-0.006
 hallway-0.006
Tokent
Token ID83
Feature activation+4.28e-01

Pos logit contributions
ps+0.047
 teeth+0.040
ches+0.039
ss+0.039
iddles+0.038
Neg logit contributions
 she-0.035
 day-0.034
 weekend-0.033
days-0.032
 particular-0.031
Tokenweet
Token ID7277
Feature activation+2.84e+00

Pos logit contributions
ps+0.021
ss+0.018
 teeth+0.018
ches+0.018
iddles+0.017
Neg logit contributions
 she-0.014
 day-0.013
 weekend-0.013
 stair-0.013
 hallway-0.012
Token,
Token ID11
Feature activation+5.00e-01

Pos logit contributions
ussy+0.119
ps+0.119
iddles+0.118
wy+0.114
 teeth+0.110
Neg logit contributions
 she-0.095
 day-0.093
 they-0.083
where-0.082
 wander-0.081
Token tweet
Token ID6126
Feature activation+4.71e+00

Pos logit contributions
ps+0.024
ss+0.021
iddles+0.021
ches+0.021
mus+0.020
Neg logit contributions
 she-0.016
 day-0.015
 stair-0.015
 weekend-0.015
 they-0.014
Token,
Token ID11
Feature activation+5.57e-01

Pos logit contributions
ps+0.277
 wounds+0.252
wy+0.244
asy+0.242
iddles+0.242
Neg logit contributions
 day-0.128
 of-0.123
 she-0.117
where-0.113
 they-0.105
Token tweet
Token ID6126
Feature activation+4.62e+00

Pos logit contributions
ps+0.024
ss+0.020
iddles+0.020
ches+0.020
 teeth+0.020
Neg logit contributions
 she-0.021
 day-0.020
 weekend-0.020
 stair-0.020
 hallway-0.019
Token".
Token ID1911
Feature activation+5.17e-01

Pos logit contributions
ps+0.274
 wounds+0.246
wy+0.242
asy+0.242
ussy+0.236
Neg logit contributions
 day-0.122
 of-0.118
 she-0.113
where-0.108
 they-0.101
Token Tim
Token ID5045
Feature activation+2.23e+00

Pos logit contributions
ps+0.027
ss+0.024
ches+0.023
 teeth+0.023
mus+0.023
Neg logit contributions
 she-0.016
 day-0.014
 they-0.014
 stair-0.013
 weekend-0.013
Token and
Token ID290
Feature activation+8.74e-02

Pos logit contributions
rier+0.123
mus+0.120
 unnecessary+0.120
ps+0.118
uable+0.116
Neg logit contributions
 she-0.045
 sandbox-0.040
 hill-0.038
 day-0.038
 slide-0.038
Token Lily
Token ID20037
Feature activation+8.33e-01

Pos logit contributions
ps+0.004
ss+0.004
ches+0.003
 teeth+0.003
mus+0.003
Neg logit contributions
 she-0.003
 weekend-0.003
 day-0.003
 stair-0.003
 hallway-0.003
Token think
Token ID892
Feature activation+4.23e-01

Pos logit contributions
ps+0.044
ss+0.040
mus+0.039
iddles+0.039
ches+0.039
Neg logit contributions
 she-0.022
 day-0.020
 explore-0.019
 wander-0.019
 stair-0.019
Token a
Token ID257
Feature activation+1.46e+00

Pos logit contributions
ps+0.051
ss+0.046
ches+0.045
mus+0.044
ealing+0.044
Neg logit contributions
 she-0.034
 day-0.028
 he-0.028
 they-0.028
 weekend-0.026
Token new
Token ID649
Feature activation+3.75e-01

Pos logit contributions
ps+0.068
ches+0.054
iddles+0.054
ss+0.054
mus+0.054
Neg logit contributions
 she-0.057
 day-0.055
 stair-0.054
 hallway-0.053
 weekend-0.051
Token word
Token ID1573
Feature activation+7.53e-01

Pos logit contributions
ps+0.017
ss+0.015
ches+0.014
mus+0.014
 teeth+0.014
Neg logit contributions
 she-0.013
 day-0.013
 stair-0.013
 weekend-0.013
 hallway-0.012
Token -
Token ID532
Feature activation+6.18e-01

Pos logit contributions
ps+0.033
ss+0.029
ches+0.028
mus+0.028
 teeth+0.028
Neg logit contributions
 she-0.028
 day-0.025
 stair-0.025
 weekend-0.025
 hallway-0.024
Token "
Token ID366
Feature activation+8.54e-01

Pos logit contributions
ps+0.027
ss+0.023
ches+0.022
mus+0.022
 teeth+0.021
Neg logit contributions
 she-0.024
 day-0.022
 stair-0.021
 weekend-0.021
 hallway-0.021
Tokenignore
Token ID46430
Feature activation+4.61e+00

Pos logit contributions
ps+0.043
 teeth+0.037
ss+0.036
ches+0.036
iddles+0.035
Neg logit contributions
 she-0.026
 weekend-0.025
 day-0.024
 stair-0.023
 hallway-0.023
Token".
Token ID1911
Feature activation+5.26e-02

Pos logit contributions
ps+0.224
ksh+0.204
ags+0.196
ussy+0.196
ows+0.193
Neg logit contributions
 she-0.138
 day-0.137
 they-0.127
where-0.122
 stair-0.113
Token She
Token ID1375
Feature activation+1.87e+00

Pos logit contributions
ps+0.003
ss+0.003
ches+0.002
 teeth+0.002
mus+0.002
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token learned
Token ID4499
Feature activation+1.16e+00

Pos logit contributions
ps+0.104
rier+0.096
 unnecessary+0.093
ags+0.093
ss+0.091
Neg logit contributions
 she-0.055
 they-0.045
 he-0.044
 hallway-0.040
 day-0.039
Token that
Token ID326
Feature activation-1.49e+00

Pos logit contributions
ps+0.070
wy+0.064
ss+0.063
ches+0.062
ealing+0.061
Neg logit contributions
 she-0.030
 they-0.022
 he-0.022
 day-0.021
 stair-0.018
Token it
Token ID340
Feature activation+1.87e+00

Pos logit contributions
 hallway+0.051
 driveway+0.050
 weekend+0.049
 sandbox+0.047
loo+0.046
Neg logit contributions
ps-0.061
 teeth-0.058
ss-0.056
mus-0.052
 louder-0.051
Token Anna
Token ID11735
Feature activation+1.70e+00

Pos logit contributions
ss+0.100
ps+0.097
wy+0.097
ealing+0.096
ches+0.094
Neg logit contributions
 she-0.057
 Jane-0.044
 stair-0.042
 they-0.042
 particular-0.042
Token.
Token ID13
Feature activation+2.37e+00

Pos logit contributions
ps+0.080
wy+0.072
ss+0.069
iddles+0.068
ches+0.067
Neg logit contributions
 she-0.063
 day-0.052
 he-0.051
 they-0.051
days-0.047
Token "
Token ID366
Feature activation+1.41e+00

Pos logit contributions
ps+0.152
wy+0.150
 unnecessary+0.141
ache+0.134
ss+0.133
Neg logit contributions
 she-0.071
 they-0.058
 Jane-0.049
 he-0.046
 stair-0.040
TokenCan
Token ID6090
Feature activation+4.66e-01

Pos logit contributions
ps+0.083
 teeth+0.074
ches+0.071
ss+0.071
ealing+0.069
Neg logit contributions
 she-0.036
 weekend-0.029
 day-0.029
 stair-0.028
 particular-0.028
Token you
Token ID345
Feature activation+1.98e+00

Pos logit contributions
ps+0.031
ss+0.028
iddles+0.027
ches+0.027
 teeth+0.027
Neg logit contributions
 she-0.008
 day-0.006
 stair-0.006
 weekend-0.006
 they-0.006
Token say
Token ID910
Feature activation+4.60e+00

Pos logit contributions
ps+0.118
iddles+0.103
rier+0.103
ss+0.102
wy+0.101
Neg logit contributions
 she-0.052
 explore-0.046
 wander-0.042
 particular-0.042
 day-0.040
Token cat
Token ID3797
Feature activation+3.72e+00

Pos logit contributions
wy+0.270
ps+0.263
aned+0.245
ished+0.242
ksh+0.239
Neg logit contributions
 she-0.145
 they-0.122
 he-0.112
 day-0.099
 today-0.095
Token?"
Token ID1701
Feature activation+1.19e+00

Pos logit contributions
ps+0.207
rier+0.173
aned+0.170
wy+0.168
obb+0.165
Neg logit contributions
 she-0.127
 they-0.110
 day-0.106
 he-0.101
 weekend-0.097
Token
Token ID198
Feature activation+2.34e+00

Pos logit contributions
ps+0.073
ss+0.064
wy+0.064
 unnecessary+0.061
ches+0.061
Neg logit contributions
 she-0.037
 they-0.028
 he-0.027
 day-0.026
 weekend-0.024
Token
Token ID198
Feature activation+6.07e-01

Pos logit contributions
ps+0.168
wy+0.146
rier+0.145
 teeth+0.144
ss+0.140
Neg logit contributions
 she-0.047
 they-0.035
 day-0.034
 he-0.029
 weekend-0.029
TokenAnna
Token ID31160
Feature activation+2.66e+00

Pos logit contributions
ps+0.033
ss+0.029
 teeth+0.029
ches+0.027
iddles+0.027
Neg logit contributions
 she-0.018
 hallway-0.016
 stair-0.016
 weekend-0.016
 day-0.015
Token.
Token ID13
Feature activation+1.96e-01

Pos logit contributions
ps+0.034
ss+0.030
ches+0.029
mus+0.029
 teeth+0.029
Neg logit contributions
 she-0.034
 they-0.030
 day-0.030
 stair-0.029
 he-0.029
Token "
Token ID366
Feature activation-2.16e-01

Pos logit contributions
ps+0.010
ss+0.009
ches+0.008
 teeth+0.008
mus+0.008
Neg logit contributions
 she-0.006
 weekend-0.006
 stair-0.006
 day-0.006
 hallway-0.006
TokenMaybe
Token ID13300
Feature activation+1.81e-01

Pos logit contributions
 she+0.006
 weekend+0.006
 hallway+0.006
 stair+0.006
 day+0.006
Neg logit contributions
ps-0.011
ss-0.010
mus-0.009
ches-0.009
 teeth-0.009
Token you
Token ID345
Feature activation+1.36e+00

Pos logit contributions
ps+0.011
ss+0.010
ches+0.009
 teeth+0.009
mus+0.009
Neg logit contributions
 she-0.004
 weekend-0.003
 day-0.003
 stair-0.003
 hallway-0.003
Token can
Token ID460
Feature activation+9.99e-01

Pos logit contributions
ps+0.071
ss+0.063
iddles+0.061
mus+0.060
rier+0.060
Neg logit contributions
 day-0.037
 she-0.036
 explore-0.035
 weekend-0.035
 they-0.034
Token cut
Token ID2005
Feature activation+4.58e+00

Pos logit contributions
ps+0.052
ss+0.047
ches+0.046
mus+0.046
 teeth+0.045
Neg logit contributions
 she-0.028
 day-0.026
 explore-0.025
 weekend-0.024
 wander-0.024
Token it
Token ID340
Feature activation+2.36e+00

Pos logit contributions
iddles+0.281
ps+0.277
ows+0.261
ache+0.253
lic+0.253
Neg logit contributions
 he-0.083
 they-0.081
 she-0.078
 day-0.071
 we-0.052
Token in
Token ID287
Feature activation+1.41e+00

Pos logit contributions
ps+0.120
iddles+0.112
ss+0.112
ealing+0.109
 unnecessary+0.107
Neg logit contributions
 she-0.071
 day-0.066
 they-0.065
 he-0.063
 weekend-0.056
Token half
Token ID2063
Feature activation+1.43e+00

Pos logit contributions
ps+0.066
ches+0.061
ss+0.059
iddles+0.058
mus+0.057
Neg logit contributions
 she-0.047
 day-0.046
 stair-0.043
 weekend-0.042
 holiday-0.041
Token,
Token ID11
Feature activation-3.35e-01

Pos logit contributions
ps+0.066
ss+0.057
iddles+0.056
mus+0.056
ches+0.055
Neg logit contributions
 she-0.051
 day-0.049
 they-0.047
 he-0.046
 weekend-0.045
Token or
Token ID393
Feature activation+3.97e-01

Pos logit contributions
 weekend+0.011
 hallway+0.011
 she+0.011
 stair+0.011
 day+0.011
Neg logit contributions
ps-0.015
ss-0.013
 teeth-0.013
ches-0.012
mus-0.012
Token and
Token ID290
Feature activation+2.05e-01

Pos logit contributions
 weekend+0.017
 hallway+0.017
 she+0.017
 stair+0.017
 day+0.017
Neg logit contributions
ps-0.017
ss-0.014
 teeth-0.014
ches-0.013
mus-0.013
Token had
Token ID550
Feature activation+1.02e+00

Pos logit contributions
ps+0.010
ss+0.009
ches+0.009
 teeth+0.009
mus+0.008
Neg logit contributions
 she-0.006
 weekend-0.006
 day-0.006
 stair-0.006
 hallway-0.006
Token the
Token ID262
Feature activation+1.26e+00

Pos logit contributions
ps+0.040
ss+0.036
ches+0.034
mus+0.033
ealing+0.033
Neg logit contributions
 she-0.041
 day-0.038
 weekend-0.038
 hallway-0.037
 holiday-0.037
Token word
Token ID1573
Feature activation+1.29e+00

Pos logit contributions
ps+0.052
ss+0.047
ches+0.045
mus+0.044
iddles+0.043
Neg logit contributions
 day-0.049
 she-0.048
 weekend-0.047
 stair-0.045
 hallway-0.045
Token "
Token ID366
Feature activation+3.15e+00

Pos logit contributions
ps+0.057
ss+0.049
iddles+0.048
ches+0.047
mus+0.047
Neg logit contributions
 she-0.050
 day-0.045
 stair-0.044
 they-0.044
 weekend-0.043
Tokenw
Token ID86
Feature activation+4.58e+00

Pos logit contributions
ps+0.154
 teeth+0.128
ealing+0.123
rier+0.122
 wounds+0.121
Neg logit contributions
 she-0.108
 day-0.108
 weekend-0.105
days-0.103
 particular-0.103
Tokenelcome
Token ID9571
Feature activation+2.90e+00

Pos logit contributions
 smarter+0.179
 unnecessary+0.178
ps+0.175
iddles+0.166
 stronger+0.165
Neg logit contributions
 she-0.148
 day-0.139
 they-0.133
 Jane-0.127
 stair-0.127
Token"
Token ID1
Feature activation+2.92e+00

Pos logit contributions
ps+0.151
iddles+0.123
 teeth+0.120
ussy+0.117
wy+0.117
Neg logit contributions
 she-0.093
 day-0.091
 weekend-0.084
 they-0.080
 stair-0.079
Token written
Token ID3194
Feature activation+3.66e+00

Pos logit contributions
ps+0.158
 teeth+0.128
ches+0.127
 unnecessary+0.127
ealing+0.125
Neg logit contributions
 she-0.102
 day-0.089
 weekend-0.086
 they-0.083
 he-0.081
Token in
Token ID287
Feature activation+1.12e+00

Pos logit contributions
ps+0.178
 unnecessary+0.173
 wounds+0.158
iddles+0.158
ws+0.157
Neg logit contributions
 she-0.123
 he-0.101
 they-0.101
 of-0.094
 day-0.093
Token big
Token ID1263
Feature activation+1.12e+00

Pos logit contributions
ps+0.051
ches+0.045
ss+0.044
iddles+0.044
mus+0.043
Neg logit contributions
 she-0.039
 day-0.036
 stair-0.035
 hallway-0.035
 particular-0.034
Token played
Token ID2826
Feature activation+1.24e+00

Pos logit contributions
 she+0.011
 hallway+0.011
 stair+0.011
 weekend+0.011
 day+0.010
Neg logit contributions
ps-0.019
ss-0.017
ches-0.016
 teeth-0.016
mus-0.016
Token a
Token ID257
Feature activation-1.02e-01

Pos logit contributions
ps+0.063
ss+0.058
ache+0.057
uable+0.056
rier+0.055
Neg logit contributions
 she-0.037
 day-0.036
 weekend-0.034
 they-0.033
 sunny-0.033
Token game
Token ID983
Feature activation-1.95e+00

Pos logit contributions
 she+0.003
 weekend+0.003
 hallway+0.003
 stair+0.003
 day+0.003
Neg logit contributions
ps-0.005
ss-0.005
ches-0.005
 teeth-0.005
mus-0.005
Token called
Token ID1444
Feature activation+2.08e+00

Pos logit contributions
days+0.067
 hallway+0.067
 driveway+0.067
 holiday+0.066
 stair+0.066
Neg logit contributions
ps-0.078
ches-0.069
ss-0.064
iddles-0.062
ows-0.061
Token "
Token ID366
Feature activation+1.81e+00

Pos logit contributions
ps+0.097
ss+0.081
ches+0.081
wy+0.079
mus+0.078
Neg logit contributions
 she-0.072
 they-0.068
 weekend-0.066
 day-0.066
 he-0.066
Tokenshoot
Token ID30408
Feature activation+4.57e+00

Pos logit contributions
ps+0.086
 teeth+0.073
uable+0.070
ealing+0.069
ches+0.067
Neg logit contributions
 day-0.063
 weekend-0.063
 she-0.061
 stair-0.060
 hallway-0.059
Token the
Token ID262
Feature activation+2.99e+00

Pos logit contributions
ps+0.222
 wounds+0.186
ags+0.180
iddles+0.177
rier+0.173
Neg logit contributions
 she-0.148
 day-0.146
 he-0.142
where-0.134
 of-0.132
Token w
Token ID266
Feature activation+3.27e+00

Pos logit contributions
ps+0.135
uable+0.123
mus+0.113
ws+0.112
ealing+0.111
Neg logit contributions
 day-0.110
 stair-0.104
 she-0.103
 playground-0.103
 hallway-0.103
Tokenaffle
Token ID30697
Feature activation+2.50e+00

Pos logit contributions
ps+0.132
 louder+0.123
uable+0.120
 unnecessary+0.114
 infected+0.114
Neg logit contributions
loo-0.104
 day-0.102
 she-0.101
 stair-0.097
 weekend-0.096
Token."
Token ID526
Feature activation+1.43e-01

Pos logit contributions
ps+0.116
 unnecessary+0.103
iddles+0.098
 teeth+0.097
ss+0.097
Neg logit contributions
 she-0.086
 stair-0.082
 day-0.081
 weekend-0.077
 they-0.076
Token In
Token ID554
Feature activation-1.16e+00

Pos logit contributions
ps+0.008
ss+0.007
ches+0.007
 teeth+0.007
mus+0.007
Neg logit contributions
 she-0.004
 weekend-0.003
 day-0.003
 hallway-0.003
 stair-0.003
Token
Token ID198
Feature activation+6.52e-01

Pos logit contributions
ps+0.058
ches+0.050
ss+0.048
wy+0.047
 teeth+0.047
Neg logit contributions
 she-0.031
 they-0.026
 weekend-0.026
 day-0.026
 stair-0.026
Token
Token ID198
Feature activation-9.15e-01

Pos logit contributions
ps+0.035
ss+0.030
ches+0.030
 teeth+0.030
mus+0.029
Neg logit contributions
 she-0.018
 day-0.017
 weekend-0.017
 stair-0.016
 hallway-0.016
Token"
Token ID1
Feature activation+4.65e-01

Pos logit contributions
 weekend+0.041
 day+0.040
 holiday+0.040
 hallway+0.040
 stair+0.040
Neg logit contributions
ps-0.029
ss-0.027
mus-0.026
ches-0.024
ealing-0.022
TokenS
Token ID50
Feature activation+7.38e-01

Pos logit contributions
ps+0.026
ches+0.023
ss+0.023
 teeth+0.022
iddles+0.021
Neg logit contributions
 she-0.012
 weekend-0.011
 day-0.011
 stair-0.011
 hallway-0.011
Tokenss
Token ID824
Feature activation+4.22e+00

Pos logit contributions
ps+0.037
 teeth+0.031
ches+0.031
mus+0.030
ss+0.030
Neg logit contributions
 she-0.022
 day-0.021
 weekend-0.021
 stair-0.020
 holiday-0.019
Tokenss
Token ID824
Feature activation+4.55e+00

Pos logit contributions
ps+0.159
rier+0.149
ags+0.146
 wounds+0.143
 unnecessary+0.135
Neg logit contributions
 she-0.160
 he-0.152
 day-0.145
 they-0.145
ite-0.142
Token,"
Token ID553
Feature activation-1.12e+00

Pos logit contributions
ps+0.156
 wounds+0.147
rier+0.146
ags+0.145
iddles+0.134
Neg logit contributions
 she-0.188
 he-0.177
 they-0.169
 day-0.166
ite-0.162
Token Tom
Token ID4186
Feature activation+1.86e+00

Pos logit contributions
days+0.052
 hallway+0.051
 weekend+0.051
 stair+0.049
where+0.048
Neg logit contributions
ps-0.037
 teeth-0.034
ss-0.031
 louder-0.027
mus-0.027
Token says
Token ID1139
Feature activation+1.44e+00

Pos logit contributions
ps+0.079
rier+0.078
 unnecessary+0.076
ss+0.074
mus+0.074
Neg logit contributions
 she-0.069
 he-0.059
 they-0.059
 day-0.058
 Jane-0.055
Token.
Token ID13
Feature activation+1.13e+00

Pos logit contributions
ps+0.067
wy+0.063
ss+0.063
ches+0.062
uable+0.062
Neg logit contributions
 she-0.052
 they-0.046
 he-0.045
 day-0.043
 Jane-0.039
Token
Token ID198
Feature activation+8.07e-01

Pos logit contributions
ps+0.060
wy+0.053
ches+0.052
 unnecessary+0.052
ss+0.051
Neg logit contributions
 she-0.038
 he-0.034
 they-0.034
 day-0.030
 stair-0.030
Token happy
Token ID3772
Feature activation-2.33e+00

Pos logit contributions
 stair+0.025
 hallway+0.025
 she+0.024
 weekend+0.024
 driveway+0.023
Neg logit contributions
ps-0.018
ss-0.015
 teeth-0.014
mus-0.014
ches-0.013
Token!
Token ID0
Feature activation-1.10e+00

Pos logit contributions
 hallway+0.114
 driveway+0.111
 weekend+0.107
 particular+0.107
 stair+0.107
Neg logit contributions
ps-0.064
 teeth-0.052
ss-0.048
 louder-0.048
."-0.045
Token<|endoftext|>
Token ID50256
Feature activation-4.89e-02

Pos logit contributions
 weekend+0.042
 hallway+0.042
 stair+0.042
 holiday+0.040
 sandbox+0.040
Neg logit contributions
ps-0.040
ss-0.036
 teeth-0.035
ches-0.035
mus-0.033
TokenOn
Token ID2202
Feature activation-3.54e+00

Pos logit contributions
 she+0.002
 weekend+0.002
 stair+0.002
 hallway+0.002
 day+0.002
Neg logit contributions
ps-0.002
ss-0.002
ches-0.002
 teeth-0.002
mus-0.002
Token a
Token ID257
Feature activation-3.85e+00

Pos logit contributions
days+0.199
 policy+0.194
spe+0.186
 stair+0.182
 hallway+0.181
Neg logit contributions
ps-0.062
ches-0.055
 better-0.038
 teeth-0.034
 a-0.029
Token sunny
Token ID27737
Feature activation-6.39e+00

Pos logit contributions
stack+0.125
 policy+0.118
iked+0.114
keeping+0.114
loo+0.110
Neg logit contributions
ches-0.172
 funny-0.153
 loud-0.147
 hard-0.144
 silly-0.144
Token day
Token ID1110
Feature activation-3.35e+00

Pos logit contributions
 policy+0.243
oker+0.225
spe+0.224
otion+0.218
 haven+0.217
Neg logit contributions
 me-0.199
ches-0.168
 w-0.157
 nice-0.156
 fur-0.156
Token,
Token ID11
Feature activation-2.32e+00

Pos logit contributions
stack+0.185
 driveway+0.182
spe+0.171
 sandbox+0.169
days+0.166
Neg logit contributions
 funny-0.046
 silly-0.045
 teeth-0.045
ches-0.044
 loud-0.044
Token a
Token ID257
Feature activation+1.76e-01

Pos logit contributions
 driveway+0.120
 hallway+0.113
loo+0.110
 stair+0.109
 hill+0.108
Neg logit contributions
ps-0.053
 teeth-0.047
mus-0.043
 funny-0.041
 louder-0.040
Token little
Token ID1310
Feature activation-1.63e+00

Pos logit contributions
ps+0.007
ss+0.006
ches+0.005
 teeth+0.005
mus+0.005
Neg logit contributions
 she-0.007
 weekend-0.007
 day-0.007
 stair-0.007
 hallway-0.007
Token girl
Token ID2576
Feature activation-3.27e+00

Pos logit contributions
 hallway+0.058
days+0.057
 weekend+0.056
 stair+0.055
 driveway+0.055
Neg logit contributions
ps-0.056
ss-0.055
mus-0.052
ches-0.051
 teeth-0.050
Token forever
Token ID8097
Feature activation-5.40e-01

Pos logit contributions
 hallway+0.041
 weekend+0.041
 stair+0.041
 driveway+0.039
 particular+0.039
Neg logit contributions
ps-0.036
ss-0.030
ches-0.029
 teeth-0.028
mus-0.027
Token.
Token ID13
Feature activation-1.06e+00

Pos logit contributions
 stair+0.023
 weekend+0.023
 hallway+0.023
 she+0.023
 particular+0.022
Neg logit contributions
ps-0.019
ss-0.016
 teeth-0.015
ches-0.015
mus-0.015
Token<|endoftext|>
Token ID50256
Feature activation+3.27e-01

Pos logit contributions
 hallway+0.041
 stair+0.040
 weekend+0.040
 holiday+0.038
 sandbox+0.038
Neg logit contributions
ps-0.040
ss-0.036
ches-0.034
 teeth-0.033
iddles-0.033
TokenOn
Token ID2202
Feature activation-3.50e+00

Pos logit contributions
ps+0.015
ss+0.013
 teeth+0.013
ches+0.013
mus+0.012
Neg logit contributions
 she-0.011
 weekend-0.011
 day-0.011
 hallway-0.010
 stair-0.010
Token a
Token ID257
Feature activation-3.66e+00

Pos logit contributions
days+0.188
 policy+0.183
 stair+0.173
 hallway+0.173
loo+0.172
Neg logit contributions
ps-0.068
ches-0.062
 better-0.048
 teeth-0.043
 harder-0.043
Token sunny
Token ID27737
Feature activation-6.17e+00

Pos logit contributions
stack+0.115
 policy+0.107
loo+0.102
iked+0.101
keeping+0.100
Neg logit contributions
ches-0.167
 funny-0.147
 loud-0.142
 hard-0.141
 silly-0.140
Token day
Token ID1110
Feature activation-3.53e+00

Pos logit contributions
 policy+0.242
oker+0.223
 Sorry+0.217
 haven+0.216
spe+0.216
Neg logit contributions
 me-0.179
ches-0.154
 nice-0.149
 hard-0.148
 big-0.146
Token,
Token ID11
Feature activation-2.13e+00

Pos logit contributions
stack+0.198
 driveway+0.191
spe+0.175
 sandbox+0.173
field+0.172
Neg logit contributions
 funny-0.052
 silly-0.049
 loud-0.046
 teeth-0.045
 louder-0.044
Token a
Token ID257
Feature activation+2.39e-01

Pos logit contributions
 driveway+0.105
 hallway+0.100
loo+0.097
 stair+0.096
days+0.095
Neg logit contributions
ps-0.051
 teeth-0.044
mus-0.043
 louder-0.041
ss-0.040
Token mom
Token ID1995
Feature activation-1.96e+00

Pos logit contributions
ps+0.009
ss+0.008
ches+0.007
 teeth+0.007
mus+0.007
Neg logit contributions
 she-0.010
 weekend-0.010
 day-0.010
 stair-0.010
 hallway-0.010
Tokenmy
Token ID1820
Feature activation-1.21e+00

Pos logit contributions
 stair+0.093
stack+0.091
 exploration+0.089
 cliff+0.086
 hallway+0.086
Neg logit contributions
ss-0.059
ps-0.057
ches-0.045
ache-0.044
iddles-0.044
Token happy
Token ID3772
Feature activation-1.04e+00

Pos logit contributions
 she+0.001
 weekend+0.001
 stair+0.001
 hallway+0.001
 day+0.001
Neg logit contributions
ps-0.001
ss-0.001
ches-0.001
 teeth-0.001
mus-0.001
Token.
Token ID13
Feature activation-7.83e-01

Pos logit contributions
 hallway+0.051
 stair+0.050
 weekend+0.050
 particular+0.050
 driveway+0.049
Neg logit contributions
ps-0.032
 teeth-0.026
ss-0.022
mus-0.021
iddles-0.021
Token<|endoftext|>
Token ID50256
Feature activation+4.81e-01

Pos logit contributions
 stair+0.030
 hallway+0.030
 weekend+0.030
 holiday+0.029
 sandbox+0.028
Neg logit contributions
ps-0.029
ss-0.026
 teeth-0.025
ches-0.025
mus-0.024
TokenOn
Token ID2202
Feature activation-3.08e+00

Pos logit contributions
ps+0.024
ss+0.021
 teeth+0.020
ches+0.020
mus+0.020
Neg logit contributions
 she-0.015
 day-0.014
 weekend-0.014
 hallway-0.014
 stair-0.014
Token a
Token ID257
Feature activation-3.30e+00

Pos logit contributions
days+0.164
 policy+0.153
 hallway+0.148
 stair+0.148
 sandbox+0.147
Neg logit contributions
ps-0.070
ches-0.066
 teeth-0.047
 better-0.043
 harder-0.040
Token sunny
Token ID27737
Feature activation-5.96e+00

Pos logit contributions
stack+0.110
 policy+0.104
iked+0.098
days+0.098
loo+0.098
Neg logit contributions
ches-0.148
 funny-0.120
 loud-0.116
 sharp-0.115
 hard-0.115
Token day
Token ID1110
Feature activation-3.10e+00

Pos logit contributions
 policy+0.222
spe+0.212
 haven+0.197
bek+0.195
 Sorry+0.195
Neg logit contributions
 me-0.182
ches-0.175
p-0.165
 fur-0.153
--0.148
Token,
Token ID11
Feature activation-1.92e+00

Pos logit contributions
 driveway+0.176
stack+0.174
spe+0.162
 sandbox+0.162
days+0.159
Neg logit contributions
ps-0.043
 teeth-0.041
ches-0.040
 funny-0.038
 silly-0.037
Token a
Token ID257
Feature activation+2.47e-01

Pos logit contributions
 driveway+0.096
 hallway+0.092
 stair+0.089
loo+0.087
 sandbox+0.086
Neg logit contributions
ps-0.047
 teeth-0.041
mus-0.039
ss-0.037
 funny-0.035
Token little
Token ID1310
Feature activation-1.40e+00

Pos logit contributions
ps+0.010
ss+0.008
ches+0.008
 teeth+0.008
mus+0.007
Neg logit contributions
 she-0.010
 day-0.010
 weekend-0.010
 stair-0.010
 hallway-0.010
Token girl
Token ID2576
Feature activation-3.21e+00

Pos logit contributions
 hallway+0.047
days+0.047
 weekend+0.045
 she+0.045
 driveway+0.045
Neg logit contributions
ps-0.050
ss-0.050
ches-0.047
mus-0.047
 teeth-0.045
Token after
Token ID706
Feature activation-1.28e-01

Pos logit contributions
ps+0.005
ss+0.005
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.003
 weekend-0.003
 day-0.003
 hallway-0.002
 stair-0.002
Token!
Token ID0
Feature activation-1.05e+00

Pos logit contributions
 she+0.005
 weekend+0.005
 hallway+0.005
 stair+0.005
 day+0.005
Neg logit contributions
ps-0.005
ss-0.004
ches-0.004
 teeth-0.004
mus-0.004
Token<|endoftext|>
Token ID50256
Feature activation+4.12e-01

Pos logit contributions
 hallway+0.041
 stair+0.041
 weekend+0.040
 sandbox+0.039
 she+0.039
Neg logit contributions
ps-0.037
ss-0.035
ches-0.033
 teeth-0.032
mus-0.031
TokenOn
Token ID2202
Feature activation-3.22e+00

Pos logit contributions
ps+0.020
ss+0.017
 teeth+0.017
ches+0.017
mus+0.017
Neg logit contributions
 she-0.013
 day-0.013
 weekend-0.013
 hallway-0.012
 stair-0.012
Token a
Token ID257
Feature activation-2.93e+00

Pos logit contributions
days+0.156
 policy+0.148
 hallway+0.144
 driveway+0.143
 stair+0.143
Neg logit contributions
ps-0.084
ches-0.082
 teeth-0.060
 better-0.059
 harder-0.053
Token sunny
Token ID27737
Feature activation-5.68e+00

Pos logit contributions
stack+0.097
 policy+0.092
loo+0.090
iked+0.087
days+0.084
Neg logit contributions
ches-0.128
 funny-0.105
 silly-0.101
 loud-0.100
 sharp-0.099
Token day
Token ID1110
Feature activation-3.25e+00

Pos logit contributions
 policy+0.196
 awaited+0.170
oker+0.170
spe+0.169
line+0.168
Neg logit contributions
 me-0.185
ches-0.174
 fur-0.154
p-0.152
 nice-0.151
Token,
Token ID11
Feature activation-2.06e+00

Pos logit contributions
stack+0.182
 driveway+0.181
 hallway+0.167
 sandbox+0.165
spe+0.164
Neg logit contributions
ps-0.046
 teeth-0.039
 silly-0.037
 good-0.037
 funny-0.037
Token Billy
Token ID15890
Feature activation-1.32e+00

Pos logit contributions
 driveway+0.099
 hallway+0.094
loo+0.092
 stair+0.091
 sandbox+0.088
Neg logit contributions
ps-0.055
mus-0.046
 teeth-0.046
ss-0.043
ches-0.041
Token was
Token ID373
Feature activation-7.83e-01

Pos logit contributions
 stair+0.051
 weekend+0.048
days+0.046
 particular+0.046
 hallway+0.046
Neg logit contributions
ps-0.053
ches-0.048
ss-0.046
 teeth-0.044
mus-0.042
Token out
Token ID503
Feature activation-1.36e+00

Pos logit contributions
 stair+0.031
 weekend+0.030
 she+0.030
days+0.029
 particular+0.029
Neg logit contributions
ps-0.030
ss-0.025
 teeth-0.025
mus-0.024
ches-0.024
Token together
Token ID1978
Feature activation+3.07e-01

Pos logit contributions
ps+0.001
ss+0.000
ches+0.000
 teeth+0.000
mus+0.000
Neg logit contributions
 she-0.001
 weekend-0.001
 stair-0.001
 day-0.001
 hallway-0.001
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
ps+0.012
ss+0.010
ches+0.010
 teeth+0.009
mus+0.009
Neg logit contributions
 she-0.013
 day-0.012
 weekend-0.012
 stair-0.012
 hallway-0.012
Token<|endoftext|>
Token ID50256
Feature activation+3.19e-01

Pos logit contributions
 she+0.039
 stair+0.039
 hallway+0.039
 weekend+0.038
 wander+0.037
Neg logit contributions
ps-0.036
ss-0.033
 teeth-0.031
ches-0.030
mus-0.029
TokenOn
Token ID2202
Feature activation-3.23e+00

Pos logit contributions
ps+0.016
ss+0.014
ches+0.013
 teeth+0.013
mus+0.013
Neg logit contributions
 she-0.010
 weekend-0.009
 day-0.009
 hallway-0.009
 stair-0.009
Token a
Token ID257
Feature activation-3.05e+00

Pos logit contributions
days+0.164
 policy+0.159
 hallway+0.153
 stair+0.152
 driveway+0.151
Neg logit contributions
ps-0.072
ches-0.065
 better-0.051
 teeth-0.051
 harder-0.046
Token sunny
Token ID27737
Feature activation-5.65e+00

Pos logit contributions
stack+0.099
 policy+0.094
loo+0.091
days+0.090
iked+0.088
Neg logit contributions
ches-0.133
 funny-0.111
 loud-0.107
 silly-0.105
 sharp-0.105
Token day
Token ID1110
Feature activation-2.89e+00

Pos logit contributions
 policy+0.203
spe+0.183
 awaited+0.178
 haven+0.177
oker+0.176
Neg logit contributions
 me-0.180
ches-0.166
p-0.148
 hard-0.146
 nice-0.146
Token,
Token ID11
Feature activation-2.03e+00

Pos logit contributions
 driveway+0.166
stack+0.161
 sandbox+0.153
 hallway+0.153
spe+0.148
Neg logit contributions
 teeth-0.040
ps-0.038
ches-0.034
 funny-0.034
 silly-0.032
Token Joe
Token ID5689
Feature activation-2.11e+00

Pos logit contributions
 driveway+0.087
 hallway+0.086
 stair+0.081
days+0.079
 occasion+0.079
Neg logit contributions
ps-0.064
 teeth-0.057
mus-0.053
ss-0.053
 louder-0.051
Token decided
Token ID3066
Feature activation-1.69e+00

Pos logit contributions
 stair+0.090
 hallway+0.086
stack+0.085
 weekend+0.084
 particular+0.083
Neg logit contributions
ps-0.070
ches-0.067
 teeth-0.061
ss-0.058
mus-0.055
Token he
Token ID339
Feature activation-2.74e-01

Pos logit contributions
days+0.067
 hallway+0.066
 driveway+0.063
 stair+0.062
 particular+0.062
Neg logit contributions
ps-0.061
 teeth-0.055
ss-0.055
 louder-0.053
ches-0.051
Token best
Token ID1266
Feature activation-6.46e-01

Pos logit contributions
 stair+0.020
 hallway+0.020
 she+0.020
 weekend+0.020
 day+0.019
Neg logit contributions
ps-0.025
ss-0.022
ches-0.020
mus-0.020
 teeth-0.020
Token friends
Token ID2460
Feature activation-1.05e+00

Pos logit contributions
 hallway+0.031
 weekend+0.031
 stair+0.030
 she+0.030
 day+0.029
Neg logit contributions
ps-0.021
ss-0.017
 teeth-0.017
iddles-0.015
ches-0.015
Token.
Token ID13
Feature activation-8.42e-01

Pos logit contributions
 hallway+0.054
 stair+0.054
 weekend+0.053
 driveway+0.051
 particular+0.051
Neg logit contributions
ps-0.029
ss-0.022
ches-0.021
 teeth-0.021
mus-0.020
Token On
Token ID1550
Feature activation-2.44e+00

Pos logit contributions
 she+0.026
 hallway+0.026
 stair+0.026
 weekend+0.025
 holiday+0.025
Neg logit contributions
ps-0.037
ss-0.034
iddles-0.032
 teeth-0.032
ches-0.032
Token a
Token ID257
Feature activation-2.43e+00

Pos logit contributions
des+0.100
 hallway+0.098
days+0.097
 driveway+0.095
 stair+0.095
Neg logit contributions
ps-0.080
ches-0.063
 teeth-0.060
oni-0.056
ss-0.055
Token sunny
Token ID27737
Feature activation-5.54e+00

Pos logit contributions
 policy+0.083
stack+0.082
loo+0.078
days+0.078
iked+0.075
Neg logit contributions
ches-0.094
 funny-0.077
 loud-0.073
 silly-0.072
ss-0.072
Token afternoon
Token ID6672
Feature activation-3.97e+00

Pos logit contributions
 haven+0.166
 policy+0.162
 she+0.155
 awaited+0.154
oker+0.153
Neg logit contributions
 me-0.190
ches-0.189
ep-0.175
p-0.168
 fur-0.167
Token,
Token ID11
Feature activation-2.30e+00

Pos logit contributions
stack+0.223
 driveway+0.223
 hill+0.210
field+0.209
boat+0.205
Neg logit contributions
ps-0.040
 teeth-0.037
 louder-0.037
ers-0.036
 funny-0.032
Token they
Token ID484
Feature activation-7.53e-01

Pos logit contributions
 hallway+0.062
 driveway+0.060
loo+0.058
 occasion+0.057
stack+0.056
Neg logit contributions
 louder-0.103
ps-0.099
ss-0.096
 funny-0.092
 loud-0.092
Token decided
Token ID3066
Feature activation-1.60e+00

Pos logit contributions
 stair+0.023
 hallway+0.022
 she+0.022
 weekend+0.022
 day+0.021
Neg logit contributions
ps-0.036
ss-0.031
ches-0.031
 teeth-0.030
mus-0.030
Token to
Token ID284
Feature activation-2.00e+00

Pos logit contributions
 hallway+0.071
loo+0.067
days+0.067
 stair+0.066
 driveway+0.066
Neg logit contributions
ps-0.052
ss-0.046
 teeth-0.045
 louder-0.045
ches-0.042
Token<|endoftext|>
Token ID50256
Feature activation+4.04e-01

Pos logit contributions
 weekend+0.029
 hallway+0.029
 stair+0.029
 holiday+0.029
 she+0.029
Neg logit contributions
ps-0.030
ss-0.028
 teeth-0.026
ches-0.026
mus-0.026
TokenOn
Token ID2202
Feature activation-3.30e+00

Pos logit contributions
ps+0.019
ss+0.017
 teeth+0.016
ches+0.016
mus+0.016
Neg logit contributions
 she-0.013
 day-0.013
 weekend-0.013
 hallway-0.013
 stair-0.013
Token a
Token ID257
Feature activation-3.62e+00

Pos logit contributions
days+0.183
 policy+0.174
spe+0.170
loo+0.168
 hallway+0.166
Neg logit contributions
ps-0.058
ches-0.057
 better-0.042
 teeth-0.038
 harder-0.036
Token bright
Token ID6016
Feature activation-4.18e+00

Pos logit contributions
stack+0.112
 policy+0.107
iked+0.107
loo+0.103
keeping+0.102
Neg logit contributions
ches-0.165
 funny-0.144
 loud-0.139
 sharp-0.138
 silly-0.138
Token and
Token ID290
Feature activation-1.53e+00

Pos logit contributions
 policy+0.137
stack+0.136
days+0.133
keeping+0.131
Have+0.121
Neg logit contributions
ches-0.156
 funny-0.144
 me-0.143
 hard-0.140
 nice-0.139
Token sunny
Token ID27737
Feature activation-5.45e+00

Pos logit contributions
 hallway+0.027
 driveway+0.026
 stair+0.026
days+0.025
loo+0.024
Neg logit contributions
ps-0.080
mus-0.076
ss-0.075
 louder-0.074
 teeth-0.074
Token day
Token ID1110
Feature activation-2.36e+00

Pos logit contributions
 policy+0.195
spe+0.185
bek+0.183
 haven+0.182
path+0.180
Neg logit contributions
 me-0.152
p-0.138
--0.135
ches-0.133
 fur-0.131
Token,
Token ID11
Feature activation-2.22e+00

Pos logit contributions
 driveway+0.098
stack+0.097
days+0.094
 sandbox+0.093
 hallway+0.092
Neg logit contributions
ps-0.073
 teeth-0.070
ches-0.068
mus-0.063
ss-0.062
Token Jimmy
Token ID12963
Feature activation-1.93e+00

Pos logit contributions
 driveway+0.091
 hallway+0.087
loo+0.084
 weekend+0.083
 stair+0.082
Neg logit contributions
ps-0.070
 teeth-0.066
mus-0.064
ss-0.061
 louder-0.056
Token was
Token ID373
Feature activation-6.73e-01

Pos logit contributions
 stair+0.074
 weekend+0.071
 particular+0.068
stack+0.068
 occasion+0.066
Neg logit contributions
ps-0.076
 teeth-0.070
ches-0.068
ss-0.066
mus-0.060
Token so
Token ID523
Feature activation-1.25e+00

Pos logit contributions
 stair+0.021
 she+0.020
 weekend+0.020
 particular+0.020
 hallway+0.019
Neg logit contributions
ps-0.032
ss-0.028
 teeth-0.027
mus-0.027
ches-0.027
Token<|endoftext|>
Token ID50256
Feature activation+1.89e-02

Pos logit contributions
 she+0.009
 stair+0.009
 weekend+0.009
 hallway+0.009
 day+0.008
Neg logit contributions
ps-0.012
ss-0.011
 teeth-0.010
ches-0.010
mus-0.010
TokenOn
Token ID2202
Feature activation-3.25e+00

Pos logit contributions
ps+0.001
ss+0.001
ches+0.001
 teeth+0.001
mus+0.001
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token a
Token ID257
Feature activation-3.72e+00

Pos logit contributions
days+0.176
 policy+0.167
 stair+0.163
spe+0.160
 hallway+0.159
Neg logit contributions
ps-0.063
ches-0.058
 teeth-0.046
 better-0.044
 harder-0.038
Token bright
Token ID6016
Feature activation-4.13e+00

Pos logit contributions
stack+0.115
iked+0.107
 policy+0.107
keeping+0.104
days+0.098
Neg logit contributions
ches-0.169
 funny-0.151
 loud-0.146
 hard-0.145
 silly-0.144
Token and
Token ID290
Feature activation-1.38e+00

Pos logit contributions
stack+0.134
 policy+0.132
keeping+0.126
days+0.124
Have+0.118
Neg logit contributions
ches-0.152
 me-0.146
 funny-0.143
 nice-0.141
 hard-0.139
Token sunny
Token ID27737
Feature activation-5.44e+00

Pos logit contributions
 hallway+0.024
 stair+0.024
 driveway+0.022
days+0.021
 she+0.020
Neg logit contributions
ps-0.075
mus-0.071
ss-0.070
 teeth-0.070
 louder-0.069
Token day
Token ID1110
Feature activation-2.15e+00

Pos logit contributions
 policy+0.201
spe+0.199
oker+0.187
bek+0.186
 contain+0.179
Neg logit contributions
 me-0.161
p-0.139
--0.136
 fur-0.136
 nice-0.130
Token,
Token ID11
Feature activation-2.17e+00

Pos logit contributions
 driveway+0.076
stack+0.075
days+0.073
 stair+0.072
 hallway+0.072
Neg logit contributions
ps-0.081
 teeth-0.079
ches-0.075
mus-0.073
ss-0.072
Token a
Token ID257
Feature activation+5.70e-01

Pos logit contributions
 driveway+0.078
 hallway+0.076
 stair+0.075
days+0.071
 weekend+0.070
Neg logit contributions
ps-0.078
 teeth-0.076
mus-0.073
ss-0.069
 louder-0.066
Token little
Token ID1310
Feature activation-1.29e+00

Pos logit contributions
ps+0.024
ss+0.020
ches+0.019
 teeth+0.019
iddles+0.019
Neg logit contributions
 she-0.023
 day-0.022
 weekend-0.022
 stair-0.022
 hallway-0.022
Token boy
Token ID2933
Feature activation-2.44e+00

Pos logit contributions
 hallway+0.043
days+0.042
 weekend+0.042
 she+0.041
 stair+0.041
Neg logit contributions
ps-0.049
ss-0.047
 teeth-0.044
mus-0.044
ches-0.043
Token.
Token ID13
Feature activation-4.82e-01

Pos logit contributions
 weekend+0.027
 she+0.027
 hallway+0.027
 stair+0.027
 day+0.026
Neg logit contributions
ps-0.022
 teeth-0.019
ss-0.018
ches-0.017
mus-0.017
Token
Token ID198
Feature activation-2.64e-01

Pos logit contributions
 she+0.018
 weekend+0.018
 hallway+0.018
 stair+0.018
 day+0.017
Neg logit contributions
ps-0.018
ss-0.017
 teeth-0.016
mus-0.016
ches-0.016
Token
Token ID198
Feature activation-7.77e-01

Pos logit contributions
 she+0.009
 weekend+0.009
 hallway+0.009
 stair+0.009
 day+0.008
Neg logit contributions
ps-0.012
ss-0.010
ches-0.010
 teeth-0.010
mus-0.010
TokenOn
Token ID2202
Feature activation-2.28e+00

Pos logit contributions
 she+0.027
 weekend+0.026
 holiday+0.025
 stair+0.025
 day+0.025
Neg logit contributions
ps-0.030
ss-0.029
mus-0.029
ches-0.028
iddles-0.026
Token a
Token ID257
Feature activation-2.44e+00

Pos logit contributions
days+0.115
des+0.112
 sandbox+0.109
 policy+0.107
 driveway+0.106
Neg logit contributions
ps-0.059
oni-0.039
ches-0.039
 teeth-0.038
 bigger-0.036
Token sunny
Token ID27737
Feature activation-5.33e+00

Pos logit contributions
stack+0.088
 policy+0.087
days+0.081
iked+0.080
packs+0.080
Neg logit contributions
ches-0.090
 funny-0.080
 sharp-0.076
 loud-0.073
 silly-0.073
Token day
Token ID1110
Feature activation-2.46e+00

Pos logit contributions
spe+0.185
 policy+0.180
 haven+0.180
otion+0.163
 awaited+0.163
Neg logit contributions
 me-0.177
p-0.169
ches-0.158
 fur-0.145
 w-0.143
Token,
Token ID11
Feature activation+1.75e-02

Pos logit contributions
 driveway+0.112
 sandbox+0.103
 hallway+0.102
 stair+0.101
stack+0.100
Neg logit contributions
ps-0.068
ss-0.061
 louder-0.058
mus-0.058
 teeth-0.057
Token a
Token ID257
Feature activation+7.06e-01

Pos logit contributions
ps+0.001
ss+0.001
ches+0.001
 teeth+0.001
mus+0.001
Neg logit contributions
 she-0.001
 weekend-0.001
 hallway-0.001
 stair-0.001
 day-0.001
Token little
Token ID1310
Feature activation-1.30e+00

Pos logit contributions
ps+0.031
ss+0.025
ches+0.024
 teeth+0.024
iddles+0.023
Neg logit contributions
 she-0.029
 day-0.028
 weekend-0.027
 stair-0.027
 hallway-0.027
Token girl
Token ID2576
Feature activation-2.97e+00

Pos logit contributions
 hallway+0.047
 driveway+0.046
days+0.046
 weekend+0.045
 stair+0.044
Neg logit contributions
ps-0.047
ss-0.045
rier-0.042
 louder-0.042
 teeth-0.042
Token.
Token ID13
Feature activation-8.25e-01

Pos logit contributions
ps+0.054
ss+0.046
mus+0.045
iddles+0.045
 teeth+0.044
Neg logit contributions
 she-0.048
 day-0.045
 weekend-0.042
 he-0.041
 they-0.041
Token
Token ID198
Feature activation-5.36e-01

Pos logit contributions
 hallway+0.031
 stair+0.031
 weekend+0.030
days+0.030
 holiday+0.030
Neg logit contributions
ps-0.030
ss-0.029
 teeth-0.027
iddles-0.026
ches-0.026
Token
Token ID198
Feature activation-1.30e+00

Pos logit contributions
 she+0.019
 hallway+0.019
 stair+0.019
 weekend+0.018
 day+0.018
Neg logit contributions
ps-0.022
ss-0.020
ches-0.019
 teeth-0.019
mus-0.019
TokenOn
Token ID2202
Feature activation-2.00e+00

Pos logit contributions
 hallway+0.055
 stair+0.054
days+0.053
 holiday+0.052
 she+0.050
Neg logit contributions
ss-0.043
ps-0.040
ches-0.038
ealing-0.036
mus-0.036
Token a
Token ID257
Feature activation-1.81e+00

Pos logit contributions
days+0.093
 hallway+0.088
des+0.087
 stair+0.086
 sandbox+0.084
Neg logit contributions
ps-0.064
oni-0.047
ches-0.046
 teeth-0.044
 louder-0.042
Token sunny
Token ID27737
Feature activation-5.19e+00

Pos logit contributions
stack+0.064
loo+0.063
days+0.062
 policy+0.060
 she+0.058
Neg logit contributions
ches-0.069
ss-0.057
ps-0.056
mus-0.054
 funny-0.054
Token day
Token ID1110
Feature activation-2.81e+00

Pos logit contributions
 haven+0.173
spe+0.165
 policy+0.163
days+0.155
 Sorry+0.154
Neg logit contributions
ches-0.170
p-0.166
 me-0.154
ep-0.148
ps-0.144
Token,
Token ID11
Feature activation-1.54e+00

Pos logit contributions
 driveway+0.125
 hallway+0.116
 stair+0.115
 attic+0.115
stack+0.113
Neg logit contributions
ps-0.081
ss-0.078
 louder-0.075
p-0.068
mus-0.068
Token Sally
Token ID25737
Feature activation-6.17e-01

Pos logit contributions
 hallway+0.052
days+0.048
 driveway+0.048
 stair+0.045
 occasion+0.042
Neg logit contributions
ps-0.067
ss-0.065
 louder-0.061
 teeth-0.060
oni-0.060
Token and
Token ID290
Feature activation-1.28e-01

Pos logit contributions
 weekend+0.023
 stair+0.023
 she+0.023
 hallway+0.022
 day+0.022
Neg logit contributions
ps-0.025
ss-0.022
ches-0.021
 teeth-0.020
mus-0.020
Token Tom
Token ID4186
Feature activation-5.88e-02

Pos logit contributions
 she+0.004
 weekend+0.004
 hallway+0.004
 stair+0.004
 day+0.004
Neg logit contributions
ps-0.006
ss-0.005
 teeth-0.005
ches-0.005
mus-0.005
Token beach
Token ID10481
Feature activation-1.77e+00

Pos logit contributions
 she+0.016
 hallway+0.014
 stair+0.014
days+0.014
 weekend+0.014
Neg logit contributions
ps-0.043
ss-0.040
ches-0.039
mus-0.038
 teeth-0.038
Token.
Token ID13
Feature activation-1.89e+00

Pos logit contributions
 hallway+0.086
 stair+0.085
 particular+0.084
 weekend+0.083
 she+0.081
Neg logit contributions
ps-0.043
ches-0.036
ss-0.035
iddles-0.033
mus-0.033
Token It
Token ID632
Feature activation-1.22e+00

Pos logit contributions
boat+0.057
 hallway+0.055
 weekend+0.055
 day+0.054
 holiday+0.054
Neg logit contributions
ps-0.072
 teeth-0.071
ss-0.070
iddles-0.070
ches-0.064
Token was
Token ID373
Feature activation-7.18e-01

Pos logit contributions
 stair+0.045
 weekend+0.043
 hallway+0.042
 sandbox+0.042
 driveway+0.041
Neg logit contributions
ps-0.046
ches-0.043
ss-0.039
 teeth-0.038
mus-0.035
Token a
Token ID257
Feature activation-1.87e+00

Pos logit contributions
 stair+0.032
 hallway+0.031
 she+0.030
days+0.030
 driveway+0.030
Neg logit contributions
ps-0.024
ss-0.019
 teeth-0.019
ches-0.018
mus-0.018
Token sunny
Token ID27737
Feature activation-5.19e+00

Pos logit contributions
 she+0.090
days+0.090
path+0.088
 hallway+0.087
 stair+0.087
Neg logit contributions
ps-0.036
ss-0.036
 funny-0.032
 teeth-0.031
 smarter-0.029
Token day
Token ID1110
Feature activation-2.36e+00

Pos logit contributions
iked+0.158
spe+0.149
path+0.147
Over+0.145
osity+0.144
Neg logit contributions
 nice-0.154
 me-0.151
 good-0.151
."-0.150
--0.147
Token and
Token ID290
Feature activation-2.01e+00

Pos logit contributions
 sandbox+0.110
 stair+0.109
 hallway+0.107
 driveway+0.105
 particular+0.104
Neg logit contributions
ps-0.062
ss-0.048
ches-0.047
 teeth-0.045
 louder-0.044
Token Kate
Token ID16693
Feature activation-3.08e-01

Pos logit contributions
 driveway+0.048
 hallway+0.045
 stair+0.043
 policy+0.042
boat+0.042
Neg logit contributions
ps-0.097
 teeth-0.091
ss-0.087
ches-0.086
 louder-0.085
Token couldn
Token ID3521
Feature activation-1.01e+00

Pos logit contributions
 she+0.009
 stair+0.009
 weekend+0.009
 hallway+0.009
 day+0.009
Neg logit contributions
ps-0.015
ss-0.013
ches-0.013
 teeth-0.013
mus-0.012
Token't
Token ID470
Feature activation-1.26e+00

Pos logit contributions
 sandbox+0.037
 weekend+0.037
 hallway+0.037
 stair+0.036
 day+0.035
Neg logit contributions
ps-0.041
ches-0.036
ss-0.035
mus-0.032
 teeth-0.032
Token to
Token ID284
Feature activation-2.07e-01

Pos logit contributions
ps+0.019
ss+0.017
ches+0.017
mus+0.016
 teeth+0.016
Neg logit contributions
 she-0.011
 day-0.010
 weekend-0.010
 stair-0.010
 hallway-0.010
Token explore
Token ID7301
Feature activation+2.13e-01

Pos logit contributions
 she+0.004
 weekend+0.004
 hallway+0.004
 stair+0.004
 day+0.004
Neg logit contributions
ps-0.012
ss-0.011
 teeth-0.011
ches-0.011
mus-0.010
Token.
Token ID13
Feature activation-1.84e+00

Pos logit contributions
ps+0.009
ss+0.008
ches+0.007
 teeth+0.007
mus+0.007
Neg logit contributions
 she-0.008
 day-0.008
 weekend-0.008
 stair-0.008
 hallway-0.008
Token On
Token ID1550
Feature activation-2.93e+00

Pos logit contributions
boat+0.077
 hallway+0.073
days+0.070
 stair+0.070
 exploration+0.070
Neg logit contributions
 teeth-0.049
ps-0.049
ss-0.048
iddles-0.047
ches-0.041
Token one
Token ID530
Feature activation-2.50e+00

Pos logit contributions
days+0.137
des+0.136
 policy+0.127
 sandbox+0.127
 contain+0.126
Neg logit contributions
ps-0.075
ches-0.056
 hard-0.054
 teeth-0.052
 bad-0.045
Token sunny
Token ID27737
Feature activation-5.13e+00

Pos logit contributions
 awaited+0.057
 driveway+0.054
spe+0.053
 harbour+0.053
 hallway+0.053
Neg logit contributions
 bigger-0.100
ches-0.099
ss-0.099
 teeth-0.098
 louder-0.098
Token day
Token ID1110
Feature activation-3.06e+00

Pos logit contributions
spe+0.182
 contain+0.180
stack+0.167
Yep+0.163
 haven+0.163
Neg logit contributions
ags-0.158
--0.156
ps-0.154
 fur-0.152
 teeth-0.149
Token,
Token ID11
Feature activation-3.13e+00

Pos logit contributions
 driveway+0.189
stack+0.186
 harbour+0.182
 sandbox+0.178
field+0.178
Neg logit contributions
ps-0.038
 teeth-0.034
 good-0.029
ss-0.029
 bad-0.022
Token she
Token ID673
Feature activation-1.44e+00

Pos logit contributions
 driveway+0.104
boat+0.100
 harbour+0.095
 hallway+0.092
 voyage+0.090
Neg logit contributions
 funny-0.123
 teeth-0.121
ps-0.120
 loud-0.117
 good-0.116
Token went
Token ID1816
Feature activation-1.22e+00

Pos logit contributions
 stair+0.053
 particular+0.051
stack+0.051
 weekend+0.050
days+0.050
Neg logit contributions
ps-0.053
 teeth-0.049
ches-0.048
ss-0.047
iddles-0.042
Token on
Token ID319
Feature activation-2.12e-01

Pos logit contributions
days+0.053
 weekend+0.052
 stair+0.051
 particular+0.050
spe+0.050
Neg logit contributions
ps-0.041
ches-0.035
ss-0.030
iddles-0.030
 louder-0.030
Token and
Token ID290
Feature activation-1.66e+00

Pos logit contributions
 she+0.019
 weekend+0.019
 hallway+0.019
 stair+0.019
 day+0.019
Neg logit contributions
ps-0.018
ss-0.015
ches-0.015
 teeth-0.014
mus-0.014
Token Patrick
Token ID9925
Feature activation+5.89e-01

Pos logit contributions
 stair+0.059
 hallway+0.058
 weekend+0.056
 driveway+0.055
 hill+0.053
Neg logit contributions
ps-0.072
 teeth-0.062
ss-0.060
iddles-0.057
mus-0.057
Token.
Token ID13
Feature activation-3.58e-01

Pos logit contributions
ps+0.024
ss+0.021
ches+0.019
mus+0.019
 teeth+0.019
Neg logit contributions
 she-0.024
 day-0.023
 weekend-0.022
 they-0.022
 hallway-0.021
Token On
Token ID1550
Feature activation-1.95e+00

Pos logit contributions
 she+0.011
 stair+0.011
 weekend+0.010
 day+0.010
 hallway+0.010
Neg logit contributions
ps-0.017
ss-0.015
 teeth-0.015
ches-0.015
iddles-0.014
Token a
Token ID257
Feature activation-2.11e+00

Pos logit contributions
days+0.085
des+0.084
 hallway+0.083
 stair+0.083
 she+0.081
Neg logit contributions
ps-0.062
ches-0.047
 teeth-0.047
oni-0.045
mus-0.041
Token sunny
Token ID27737
Feature activation-5.12e+00

Pos logit contributions
stack+0.086
 policy+0.086
days+0.083
 discovering+0.082
 she+0.082
Neg logit contributions
ches-0.065
 funny-0.050
ps-0.050
ss-0.049
 loud-0.048
Token day
Token ID1110
Feature activation-2.48e+00

Pos logit contributions
 awaited+0.151
 policy+0.150
des+0.146
 haven+0.145
spe+0.143
Neg logit contributions
ches-0.173
 me-0.170
ps-0.164
 fur-0.162
ep-0.162
Token,
Token ID11
Feature activation-2.21e+00

Pos logit contributions
 driveway+0.124
 sandbox+0.119
stack+0.118
 hill+0.115
 stair+0.114
Neg logit contributions
ps-0.061
 teeth-0.057
ches-0.052
ss-0.050
es-0.046
Token they
Token ID484
Feature activation-6.61e-01

Pos logit contributions
stack+0.065
 driveway+0.062
 occasion+0.061
 hallway+0.060
 stair+0.057
Neg logit contributions
 teeth-0.091
 louder-0.090
 funny-0.090
ss-0.089
ps-0.088
Token decided
Token ID3066
Feature activation-1.44e+00

Pos logit contributions
 stair+0.020
 hallway+0.019
 she+0.019
 weekend+0.019
 particular+0.018
Neg logit contributions
ps-0.031
ss-0.028
ches-0.027
 teeth-0.027
mus-0.026
Token to
Token ID284
Feature activation-1.62e+00

Pos logit contributions
 hallway+0.067
 particular+0.065
days+0.065
 stair+0.063
 driveway+0.062
Neg logit contributions
ps-0.045
 teeth-0.039
ss-0.039
 louder-0.037
ches-0.035
Token to
Token ID284
Feature activation-1.38e+00

Pos logit contributions
 stair+0.065
 particular+0.065
 hallway+0.065
 sandbox+0.065
days+0.063
Neg logit contributions
ps-0.053
ches-0.046
ss-0.042
mus-0.039
iddles-0.039
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 hallway+0.054
loo+0.053
 she+0.053
 stair+0.053
 weekend+0.052
Neg logit contributions
ps-0.048
 teeth-0.040
ss-0.038
ches-0.038
mus-0.037
Token park
Token ID3952
Feature activation-2.23e+00

Pos logit contributions
 she+0.021
days+0.020
 weekend+0.019
 hallway+0.019
 stair+0.018
Neg logit contributions
ps-0.053
ss-0.049
ches-0.047
mus-0.046
 teeth-0.046
Token on
Token ID319
Feature activation-1.37e+00

Pos logit contributions
 hallway+0.129
 hurricane+0.127
 stair+0.126
loo+0.125
 particular+0.124
Neg logit contributions
ps-0.025
ss-0.017
ches-0.016
ows-0.015
 teeth-0.015
Token a
Token ID257
Feature activation-2.17e+00

Pos logit contributions
 hallway+0.062
loo+0.061
days+0.061
 weekend+0.060
 stair+0.060
Neg logit contributions
ps-0.038
 teeth-0.031
ches-0.029
ss-0.027
mus-0.026
Token sunny
Token ID27737
Feature activation-5.11e+00

Pos logit contributions
days+0.073
stack+0.073
 policy+0.072
 she+0.071
loo+0.070
Neg logit contributions
ches-0.077
ss-0.067
 funny-0.067
ps-0.064
 bigger-0.064
Token day
Token ID1110
Feature activation-2.32e+00

Pos logit contributions
 policy+0.162
otion+0.146
 awaited+0.143
 haven+0.138
 she+0.134
Neg logit contributions
 me-0.183
ches-0.161
 nice-0.154
p-0.153
."-0.149
Token.
Token ID13
Feature activation-4.10e-01

Pos logit contributions
stack+0.132
 stair+0.131
loo+0.129
 hallway+0.128
 sandbox+0.128
Neg logit contributions
ps-0.032
ss-0.025
ches-0.024
 louder-0.021
mus-0.019
Token Mack
Token ID16640
Feature activation+3.18e-01

Pos logit contributions
 she+0.013
 weekend+0.013
 hallway+0.012
 stair+0.012
 day+0.012
Neg logit contributions
ps-0.019
ss-0.017
ches-0.016
 teeth-0.016
mus-0.016
Tokenenzie
Token ID26389
Feature activation-7.40e-01

Pos logit contributions
ps+0.015
ss+0.013
 teeth+0.012
mus+0.012
ches+0.012
Neg logit contributions
 she-0.011
 weekend-0.011
 day-0.011
 hallway-0.010
 stair-0.010
Token loved
Token ID6151
Feature activation-9.48e-01

Pos logit contributions
 weekend+0.022
 she+0.022
 stair+0.022
 hallway+0.021
 day+0.021
Neg logit contributions
ps-0.034
ss-0.031
ches-0.030
 teeth-0.030
mus-0.029
Token in
Token ID287
Feature activation-2.85e+00

Pos logit contributions
loo+0.094
 policy+0.093
 hallway+0.093
path+0.093
 stair+0.092
Neg logit contributions
ps-0.051
 louder-0.042
mus-0.038
ches-0.036
 teeth-0.035
Token the
Token ID262
Feature activation-2.61e+00

Pos logit contributions
loo+0.169
storm+0.169
boat+0.165
oker+0.159
path+0.159
Neg logit contributions
ps-0.030
 teeth-0.026
 louder-0.020
 loud-0.019
 funny-0.015
Token park
Token ID3952
Feature activation-3.06e+00

Pos logit contributions
days+0.089
 policy+0.089
 contain+0.081
iked+0.080
 awaited+0.078
Neg logit contributions
 funny-0.085
 bigger-0.085
ps-0.083
 louder-0.082
 better-0.082
Token on
Token ID319
Feature activation-1.77e+00

Pos logit contributions
loo+0.198
 hurricane+0.192
 stair+0.191
 awaited+0.185
 policy+0.185
Neg logit contributions
ps-0.022
 louder-0.015
 teeth-0.010
.-0.008
 softly-0.007
Token a
Token ID257
Feature activation-2.40e+00

Pos logit contributions
days+0.091
loo+0.088
boat+0.086
 weekend+0.086
 occasion+0.084
Neg logit contributions
ps-0.040
 teeth-0.035
iddles-0.025
ches-0.025
mus-0.024
Token sunny
Token ID27737
Feature activation-5.09e+00

Pos logit contributions
stack+0.090
 policy+0.089
days+0.089
loo+0.087
iked+0.083
Neg logit contributions
ches-0.077
 funny-0.069
 bigger-0.065
 loud-0.064
 better-0.064
Token day
Token ID1110
Feature activation-2.49e+00

Pos logit contributions
 policy+0.170
otion+0.154
 awaited+0.149
oker+0.149
 haven+0.147
Neg logit contributions
 me-0.184
 nice-0.153
 fur-0.151
p-0.150
ep-0.147
Token.
Token ID13
Feature activation-9.44e-01

Pos logit contributions
stack+0.150
loo+0.147
 stair+0.145
days+0.143
 sandbox+0.141
Neg logit contributions
ps-0.025
 louder-0.021
ss-0.018
ches-0.015
mus-0.014
Token It
Token ID632
Feature activation-1.20e+00

Pos logit contributions
 stair+0.030
 day+0.029
 weekend+0.029
 hallway+0.029
 she+0.029
Neg logit contributions
ps-0.040
ss-0.036
 teeth-0.036
ches-0.035
mus-0.035
Token was
Token ID373
Feature activation-7.83e-01

Pos logit contributions
 weekend+0.043
 holiday+0.042
 stair+0.042
days+0.042
 occasion+0.042
Neg logit contributions
ps-0.045
ches-0.043
ss-0.040
 teeth-0.039
mus-0.036
Token a
Token ID257
Feature activation-1.88e+00

Pos logit contributions
 stair+0.036
 she+0.036
days+0.036
 hallway+0.035
 weekend+0.035
Neg logit contributions
ps-0.024
 teeth-0.019
ss-0.019
mus-0.017
ches-0.017
Token.
Token ID13
Feature activation-1.92e+00

Pos logit contributions
ps+0.005
ss+0.004
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.004
 weekend-0.004
 day-0.004
 stair-0.004
 hallway-0.004
Token On
Token ID1550
Feature activation-3.00e+00

Pos logit contributions
 she+0.080
 stair+0.076
days+0.075
 hallway+0.075
 sandbox+0.074
Neg logit contributions
ps-0.056
 teeth-0.055
iddles-0.048
ss-0.048
mus-0.046
Token a
Token ID257
Feature activation-2.22e+00

Pos logit contributions
days+0.123
yt+0.117
 sandbox+0.115
des+0.115
 policy+0.113
Neg logit contributions
ps-0.093
 teeth-0.075
ches-0.075
mus-0.065
 hard-0.064
Token warm
Token ID5814
Feature activation-4.17e+00

Pos logit contributions
 policy+0.061
stack+0.061
days+0.057
loo+0.057
iked+0.053
Neg logit contributions
ches-0.100
 funny-0.085
ss-0.084
 loud-0.083
mus-0.083
Token and
Token ID290
Feature activation-1.26e+00

Pos logit contributions
days+0.127
 policy+0.124
Have+0.124
We+0.116
spe+0.116
Neg logit contributions
ches-0.150
 me-0.140
 fur-0.129
ps-0.129
 co-0.128
Token sunny
Token ID27737
Feature activation-5.08e+00

Pos logit contributions
 stair+0.030
 hallway+0.030
 she+0.030
days+0.029
 driveway+0.029
Neg logit contributions
ps-0.058
 teeth-0.055
ss-0.054
mus-0.053
ches-0.052
Token day
Token ID1110
Feature activation-2.59e+00

Pos logit contributions
 policy+0.161
 haven+0.150
oker+0.148
spe+0.148
 contain+0.147
Neg logit contributions
 me-0.162
ches-0.139
 nice-0.139
--0.138
 fur-0.134
Token,
Token ID11
Feature activation-2.72e+00

Pos logit contributions
stack+0.117
 stair+0.115
 driveway+0.114
 sandbox+0.114
 hallway+0.111
Neg logit contributions
ps-0.070
 teeth-0.067
 louder-0.059
mus-0.058
ss-0.056
Token John
Token ID1757
Feature activation-4.16e-01

Pos logit contributions
 hallway+0.083
 wander+0.081
ite+0.081
 stair+0.081
stack+0.080
Neg logit contributions
 teeth-0.103
ps-0.100
 louder-0.096
 loud-0.093
 funny-0.093
Token and
Token ID290
Feature activation-1.50e+00

Pos logit contributions
 she+0.012
 stair+0.012
 weekend+0.012
 hallway+0.012
 day+0.011
Neg logit contributions
ps-0.020
ss-0.018
ches-0.018
 teeth-0.018
mus-0.017
Token his
Token ID465
Feature activation+5.17e-01

Pos logit contributions
 hallway+0.062
 stair+0.061
path+0.060
 sandbox+0.060
stack+0.060
Neg logit contributions
 teeth-0.044
ps-0.044
ss-0.037
uable-0.034
iddles-0.034
Token<|endoftext|>
Token ID50256
Feature activation-2.09e-02

Pos logit contributions
 stair+0.028
 weekend+0.028
 hallway+0.028
 she+0.027
 holiday+0.027
Neg logit contributions
ps-0.032
ss-0.029
 teeth-0.027
ches-0.026
mus-0.026
TokenMark
Token ID9704
Feature activation-4.96e-01

Pos logit contributions
 she+0.001
 weekend+0.001
 stair+0.001
 day+0.001
 hallway+0.001
Neg logit contributions
ps-0.001
ss-0.001
ches-0.001
 teeth-0.001
mus-0.001
Token was
Token ID373
Feature activation-8.90e-01

Pos logit contributions
 she+0.016
 stair+0.016
 weekend+0.016
 hallway+0.016
 day+0.015
Neg logit contributions
ps-0.023
ss-0.020
ches-0.020
 teeth-0.019
mus-0.019
Token very
Token ID845
Feature activation-1.03e+00

Pos logit contributions
 stair+0.037
 she+0.036
 hallway+0.036
 weekend+0.035
days+0.035
Neg logit contributions
ps-0.034
 teeth-0.026
ss-0.026
mus-0.025
ches-0.025
Token excited
Token ID6568
Feature activation-2.04e+00

Pos logit contributions
 stair+0.049
 hallway+0.048
 she+0.046
 weekend+0.046
 driveway+0.046
Neg logit contributions
ps-0.030
 teeth-0.024
ss-0.024
mus-0.022
ealing-0.022
Token because
Token ID780
Feature activation-5.07e+00

Pos logit contributions
 hallway+0.101
 sandbox+0.100
 stair+0.099
 particular+0.099
days+0.097
Neg logit contributions
ps-0.051
 teeth-0.042
ches-0.039
ss-0.039
iddles-0.036
Token today
Token ID1909
Feature activation-2.96e+00

Pos logit contributions
 spine+0.179
 distance+0.169
stack+0.168
ugg+0.160
 occasion+0.160
Neg logit contributions
 teeth-0.144
 too-0.140
 loud-0.139
 a-0.134
 bigger-0.127
Token he
Token ID339
Feature activation-1.03e+00

Pos logit contributions
 driveway+0.122
stack+0.122
days+0.122
 stair+0.120
 hallway+0.120
Neg logit contributions
 teeth-0.082
ps-0.073
ss-0.070
ches-0.066
oni-0.064
Token got
Token ID1392
Feature activation-3.37e-01

Pos logit contributions
 stair+0.032
 weekend+0.031
 she+0.030
 hallway+0.030
 sandbox+0.029
Neg logit contributions
ps-0.046
ches-0.041
ss-0.041
 teeth-0.040
mus-0.038
Token to
Token ID284
Feature activation-2.26e+00

Pos logit contributions
 stair+0.012
 hallway+0.011
 she+0.011
 weekend+0.011
 day+0.011
Neg logit contributions
ps-0.015
ss-0.013
 teeth-0.012
mus-0.012
ches-0.012
Token go
Token ID467
Feature activation-1.20e+00

Pos logit contributions
loo+0.071
 weekend+0.069
 hallway+0.069
 she+0.068
 stair+0.068
Neg logit contributions
ps-0.097
 teeth-0.085
ss-0.076
mus-0.073
ches-0.070
Token
Token ID220
Feature activation+8.37e-01

Pos logit contributions
 she+0.007
 weekend+0.007
 hallway+0.007
 stair+0.007
 day+0.007
Neg logit contributions
ps-0.010
ss-0.009
 teeth-0.009
ches-0.009
mus-0.009
Token
Token ID198
Feature activation+1.08e-01

Pos logit contributions
ps+0.045
ss+0.038
ches+0.037
 teeth+0.037
mus+0.037
Neg logit contributions
 she-0.024
 weekend-0.022
 day-0.022
 stair-0.021
 hallway-0.020
Token
Token ID198
Feature activation-1.21e+00

Pos logit contributions
ps+0.005
ss+0.005
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.003
 weekend-0.003
 day-0.003
 stair-0.003
 hallway-0.003
TokenOn
Token ID2202
Feature activation-2.36e+00

Pos logit contributions
 she+0.047
 stair+0.046
 hallway+0.046
stack+0.046
 sandbox+0.044
Neg logit contributions
ss-0.040
ps-0.038
ches-0.037
mus-0.037
iddles-0.034
Token one
Token ID530
Feature activation-2.34e+00

Pos logit contributions
days+0.112
 sandbox+0.108
des+0.103
 hallway+0.103
 stair+0.102
Neg logit contributions
ps-0.068
oni-0.052
ches-0.051
 teeth-0.049
 harder-0.047
Token sunny
Token ID27737
Feature activation-5.07e+00

Pos logit contributions
 hallway+0.055
 driveway+0.053
 stair+0.051
 awaited+0.051
days+0.050
Neg logit contributions
 teeth-0.105
ps-0.103
ss-0.100
ches-0.098
 better-0.095
Token day
Token ID1110
Feature activation-2.87e+00

Pos logit contributions
spe+0.165
 haven+0.151
Have+0.142
stack+0.139
line+0.137
Neg logit contributions
 teeth-0.187
ps-0.181
p-0.178
 fur-0.177
--0.173
Token,
Token ID11
Feature activation-1.80e+00

Pos logit contributions
 driveway+0.164
stack+0.164
 sandbox+0.162
 stair+0.154
field+0.153
Neg logit contributions
ps-0.042
 teeth-0.041
ss-0.036
mus-0.032
 good-0.031
Token Jack
Token ID3619
Feature activation-6.94e-02

Pos logit contributions
 hallway+0.067
 driveway+0.064
days+0.063
 wander+0.059
loo+0.059
Neg logit contributions
ps-0.066
 teeth-0.064
 louder-0.063
ss-0.061
oni-0.058
Token's
Token ID338
Feature activation-3.66e-01

Pos logit contributions
 she+0.002
 weekend+0.002
 stair+0.002
 hallway+0.002
 day+0.002
Neg logit contributions
ps-0.003
ss-0.003
ches-0.003
 teeth-0.003
mus-0.003
Token mom
Token ID1995
Feature activation-1.11e-01

Pos logit contributions
 she+0.014
 weekend+0.014
 hallway+0.014
 stair+0.014
 day+0.013
Neg logit contributions
ps-0.014
 teeth-0.013
ss-0.013
ches-0.011
mus-0.011
TokenEverybody
Token ID28172
Feature activation-1.88e+00

Pos logit contributions
ps+0.018
ss+0.016
 teeth+0.015
ches+0.015
mus+0.015
Neg logit contributions
 she-0.011
 day-0.010
 weekend-0.010
 hallway-0.010
 stair-0.010
Token was
Token ID373
Feature activation-6.13e-01

Pos logit contributions
 hallway+0.079
 stair+0.078
 particular+0.076
 weekend+0.076
 driveway+0.075
Neg logit contributions
ps-0.065
ss-0.056
ches-0.051
mus-0.049
 teeth-0.049
Token outside
Token ID2354
Feature activation-2.68e+00

Pos logit contributions
 she+0.018
 stair+0.017
 hallway+0.017
 weekend+0.016
 particular+0.016
Neg logit contributions
ps-0.031
ss-0.027
 teeth-0.026
mus-0.026
iddles-0.025
Token on
Token ID319
Feature activation-8.50e-01

Pos logit contributions
boat+0.111
loo+0.106
 occasion+0.104
stack+0.104
path+0.103
Neg logit contributions
ps-0.094
 teeth-0.079
mus-0.071
 louder-0.070
ss-0.067
Token a
Token ID257
Feature activation-2.15e+00

Pos logit contributions
 weekend+0.035
days+0.035
 hallway+0.035
 she+0.034
 stair+0.033
Neg logit contributions
ps-0.030
 teeth-0.025
ss-0.023
mus-0.022
ches-0.022
Token sunny
Token ID27737
Feature activation-5.06e+00

Pos logit contributions
 policy+0.073
stack+0.072
days+0.072
iked+0.072
 she+0.069
Neg logit contributions
ches-0.074
 funny-0.066
ss-0.064
 loud-0.063
 better-0.062
Token day
Token ID1110
Feature activation-1.96e+00

Pos logit contributions
 policy+0.166
oker+0.155
iked+0.154
spe+0.150
bars+0.149
Neg logit contributions
 me-0.181
 fur-0.149
p-0.146
."-0.145
.-0.144
Token when
Token ID618
Feature activation-1.90e+00

Pos logit contributions
 stair+0.097
 particular+0.097
 weekend+0.096
stack+0.096
 hallway+0.094
Neg logit contributions
ps-0.047
 teeth-0.036
ss-0.035
mus-0.032
 louder-0.031
Token suddenly
Token ID6451
Feature activation-1.03e+00

Pos logit contributions
 hallway+0.054
 occasion+0.054
 driveway+0.053
 weekend+0.052
days+0.051
Neg logit contributions
ps-0.081
 teeth-0.075
 louder-0.073
mus-0.072
ss-0.068
Token Jack
Token ID3619
Feature activation+2.84e-01

Pos logit contributions
 weekend+0.036
days+0.034
 hallway+0.034
 stair+0.034
 day+0.033
Neg logit contributions
ps-0.044
 teeth-0.038
ss-0.037
ches-0.036
mus-0.035
Token noticed
Token ID6810
Feature activation-6.90e-01

Pos logit contributions
ps+0.015
ss+0.013
 teeth+0.012
mus+0.012
ches+0.012
Neg logit contributions
 she-0.008
 hallway-0.008
 weekend-0.008
 day-0.008
 stair-0.008
Token garden
Token ID11376
Feature activation-2.32e+00

Pos logit contributions
 awaited+0.092
boat+0.089
Jane+0.089
 contain+0.084
des+0.084
Neg logit contributions
 teeth-0.094
ps-0.090
 ears-0.076
 fur-0.075
 hair-0.071
Token.
Token ID13
Feature activation-8.86e-01

Pos logit contributions
 weekend+0.097
 awaited+0.097
days+0.095
loo+0.095
 stair+0.095
Neg logit contributions
ps-0.070
 teeth-0.059
mus-0.055
ss-0.055
ches-0.052
Token It
Token ID632
Feature activation-1.11e+00

Pos logit contributions
 weekend+0.024
 holiday+0.023
 day+0.023
 she+0.023
days+0.023
Neg logit contributions
ps-0.041
 teeth-0.038
ss-0.038
mus-0.036
iddles-0.036
Token was
Token ID373
Feature activation-9.14e-01

Pos logit contributions
 weekend+0.041
 holiday+0.039
 stair+0.038
 driveway+0.038
 occasion+0.037
Neg logit contributions
ps-0.043
ches-0.040
 teeth-0.037
ss-0.037
mus-0.034
Token a
Token ID257
Feature activation-1.73e+00

Pos logit contributions
 stair+0.039
days+0.039
 she+0.038
 weekend+0.038
 hallway+0.038
Neg logit contributions
ps-0.030
 teeth-0.025
ss-0.024
mus-0.023
ches-0.022
Token sunny
Token ID27737
Feature activation-5.05e+00

Pos logit contributions
days+0.072
 she+0.071
 weekend+0.069
iked+0.068
path+0.068
Neg logit contributions
ss-0.045
ps-0.043
 teeth-0.042
 funny-0.040
mus-0.040
Token day
Token ID1110
Feature activation-2.13e+00

Pos logit contributions
iked+0.172
 contain+0.156
spe+0.149
des+0.148
path+0.141
Neg logit contributions
 me-0.155
 nice-0.147
."-0.144
 too-0.143
 sweet-0.142
Token and
Token ID290
Feature activation-1.52e+00

Pos logit contributions
 stair+0.099
 sandbox+0.099
 hallway+0.099
stack+0.098
 driveway+0.098
Neg logit contributions
ps-0.053
 teeth-0.043
 louder-0.041
ss-0.040
ches-0.039
Token the
Token ID262
Feature activation-9.41e-01

Pos logit contributions
days+0.045
loo+0.044
 driveway+0.044
 hallway+0.044
 weekend+0.044
Neg logit contributions
ps-0.064
 teeth-0.059
ss-0.056
ches-0.055
 louder-0.055
Token grass
Token ID8701
Feature activation-1.66e+00

Pos logit contributions
 she+0.035
days+0.034
 weekend+0.033
 hallway+0.033
 stair+0.033
Neg logit contributions
ps-0.036
ss-0.030
 teeth-0.029
ches-0.028
iddles-0.027
Token was
Token ID373
Feature activation-4.93e-01

Pos logit contributions
 weekend+0.059
 holiday+0.056
 driveway+0.055
 hallway+0.055
 explore+0.054
Neg logit contributions
ps-0.067
ches-0.057
ss-0.057
 teeth-0.055
ows-0.051
Token Please
Token ID4222
Feature activation+1.38e+00

Pos logit contributions
 she+0.010
 weekend+0.010
 hallway+0.010
 stair+0.010
 day+0.010
Neg logit contributions
ps-0.020
ss-0.018
 teeth-0.017
ches-0.017
mus-0.017
Token just
Token ID655
Feature activation-4.99e-01

Pos logit contributions
ps+0.072
iddles+0.065
 teeth+0.064
mus+0.064
rier+0.063
Neg logit contributions
 she-0.038
 explore-0.035
 day-0.033
 wander-0.033
days-0.031
Token lend
Token ID22096
Feature activation+2.54e+00

Pos logit contributions
 she+0.016
 weekend+0.016
 hallway+0.016
 stair+0.016
 day+0.015
Neg logit contributions
ps-0.024
ss-0.021
 teeth-0.020
ches-0.020
mus-0.019
Token it
Token ID340
Feature activation+1.97e+00

Pos logit contributions
ps+0.127
iddles+0.117
rier+0.115
ache+0.110
ussy+0.109
Neg logit contributions
 she-0.083
 day-0.073
 he-0.068
 they-0.067
 explore-0.066
Token to
Token ID284
Feature activation+1.45e+00

Pos logit contributions
ps+0.100
iddles+0.091
ss+0.090
mus+0.089
rier+0.084
Neg logit contributions
 she-0.058
 day-0.056
 weekend-0.050
 they-0.049
 explore-0.048
Token me
Token ID502
Feature activation+2.23e+00

Pos logit contributions
ps+0.059
ss+0.050
mus+0.047
ches+0.047
iddles+0.046
Neg logit contributions
 she-0.061
 day-0.058
 explore-0.055
 weekend-0.055
 holiday-0.054
Token".
Token ID1911
Feature activation-6.16e-01

Pos logit contributions
ps+0.097
ss+0.083
iddles+0.081
 teeth+0.079
mus+0.078
Neg logit contributions
 she-0.085
 day-0.080
 they-0.075
 explore-0.074
 weekend-0.074
Token
Token ID220
Feature activation+1.60e+00

Pos logit contributions
 weekend+0.021
 stair+0.020
 day+0.020
 hallway+0.020
 she+0.020
Neg logit contributions
ps-0.027
ss-0.024
 teeth-0.023
ches-0.022
mus-0.022
Token
Token ID198
Feature activation+2.14e-01

Pos logit contributions
ps+0.094
 teeth+0.080
wy+0.079
ches+0.079
ss+0.079
Neg logit contributions
 she-0.040
 they-0.033
 he-0.032
 weekend-0.032
 day-0.032
Token
Token ID198
Feature activation-1.12e+00

Pos logit contributions
ps+0.011
ss+0.010
 teeth+0.010
ches+0.010
mus+0.009
Neg logit contributions
 she-0.006
 weekend-0.006
 day-0.006
 hallway-0.005
 stair-0.005
TokenBut
Token ID1537
Feature activation-1.78e+00

Pos logit contributions
 weekend+0.039
 holiday+0.038
 stair+0.038
 day+0.038
 hallway+0.038
Neg logit contributions
ps-0.045
ss-0.044
mus-0.041
ches-0.040
ealing-0.038
Token ringing
Token ID32333
Feature activation+1.78e+00

Pos logit contributions
 weekend+0.040
 stair+0.039
 particular+0.039
days+0.039
 she+0.038
Neg logit contributions
ps-0.051
ss-0.040
ches-0.040
 teeth-0.039
mus-0.038
Token the
Token ID262
Feature activation+9.92e-01

Pos logit contributions
ps+0.081
wy+0.074
ss+0.073
ches+0.072
mus+0.069
Neg logit contributions
 she-0.061
 they-0.058
 day-0.058
 stair-0.055
 he-0.055
Token bell
Token ID8966
Feature activation+3.59e-02

Pos logit contributions
ps+0.044
ss+0.040
mus+0.039
ches+0.039
iddles+0.038
Neg logit contributions
 day-0.035
 weekend-0.034
 hallway-0.034
 stair-0.033
 holiday-0.032
Token and
Token ID290
Feature activation+2.26e-01

Pos logit contributions
ps+0.001
ss+0.001
ches+0.001
 teeth+0.001
mus+0.001
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token making
Token ID1642
Feature activation+1.55e+00

Pos logit contributions
ps+0.011
ss+0.009
ches+0.009
 teeth+0.009
mus+0.009
Neg logit contributions
 she-0.007
 weekend-0.007
 day-0.007
 hallway-0.007
 stair-0.007
Token loud
Token ID7812
Feature activation+2.19e+00

Pos logit contributions
ps+0.064
ss+0.057
wy+0.056
ches+0.055
mus+0.053
Neg logit contributions
 she-0.060
 day-0.058
 they-0.056
 holiday-0.055
 weekend-0.055
Token noises
Token ID26782
Feature activation+7.70e-01

Pos logit contributions
ps+0.069
wy+0.068
ches+0.061
ss+0.061
ussy+0.058
Neg logit contributions
 she-0.101
 day-0.100
 hallway-0.097
 they-0.097
 stair-0.095
Token.
Token ID13
Feature activation+1.86e-01

Pos logit contributions
ps+0.031
ss+0.027
 teeth+0.026
ches+0.026
mus+0.025
Neg logit contributions
 she-0.031
 day-0.029
 weekend-0.028
 stair-0.028
 they-0.028
Token They
Token ID1119
Feature activation+1.79e-01

Pos logit contributions
ps+0.009
ss+0.008
ches+0.008
 teeth+0.008
mus+0.007
Neg logit contributions
 she-0.006
 weekend-0.006
 day-0.006
 hallway-0.006
 stair-0.006
Token had
Token ID550
Feature activation+8.93e-01

Pos logit contributions
ps+0.010
ss+0.009
ches+0.008
 teeth+0.008
mus+0.008
Neg logit contributions
 she-0.005
 weekend-0.005
 day-0.004
 hallway-0.004
 stair-0.004
Token so
Token ID523
Feature activation+6.38e-01

Pos logit contributions
ps+0.038
ss+0.035
ches+0.033
mus+0.033
rier+0.032
Neg logit contributions
 she-0.033
 day-0.032
 weekend-0.031
 hallway-0.031
 holiday-0.030
Token puzzle
Token ID15027
Feature activation-2.62e+00

Pos logit contributions
ps+0.009
ss+0.007
ches+0.007
 teeth+0.007
mus+0.007
Neg logit contributions
 she-0.010
 weekend-0.009
 day-0.009
 stair-0.009
 hallway-0.009
Token that
Token ID326
Feature activation-9.85e-01

Pos logit contributions
days+0.130
 wander+0.121
 weekend+0.120
 particular+0.117
 Suz+0.115
Neg logit contributions
ps-0.075
ches-0.065
ers-0.057
oni-0.056
 stronger-0.054
Token he
Token ID339
Feature activation+1.37e+00

Pos logit contributions
 stair+0.031
 weekend+0.030
 hallway+0.030
days+0.030
 she+0.029
Neg logit contributions
ps-0.042
ss-0.039
 teeth-0.039
ches-0.036
mus-0.035
Token wanted
Token ID2227
Feature activation+1.27e+00

Pos logit contributions
ps+0.085
rier+0.077
ss+0.073
ealing+0.072
mus+0.071
Neg logit contributions
 she-0.028
 day-0.028
 he-0.024
 weekend-0.024
 hallway-0.023
Token to
Token ID284
Feature activation-3.36e-01

Pos logit contributions
ps+0.077
ss+0.069
ealing+0.069
wy+0.067
ches+0.067
Neg logit contributions
 she-0.030
 day-0.027
 he-0.026
 weekend-0.024
 they-0.024
Token put
Token ID1234
Feature activation+2.27e+00

Pos logit contributions
 she+0.010
 weekend+0.010
 hallway+0.010
 stair+0.010
 day+0.009
Neg logit contributions
ps-0.016
ss-0.015
 teeth-0.014
ches-0.014
mus-0.014
Token together
Token ID1978
Feature activation+1.11e-01

Pos logit contributions
ps+0.136
ache+0.131
ss+0.128
ches+0.125
ealing+0.125
Neg logit contributions
 day-0.053
 he-0.049
 she-0.049
 they-0.040
 weekend-0.038
Token.
Token ID13
Feature activation+1.91e-01

Pos logit contributions
ps+0.004
ss+0.004
ches+0.003
 teeth+0.003
mus+0.003
Neg logit contributions
 she-0.005
 weekend-0.004
 day-0.004
 stair-0.004
 hallway-0.004
Token He
Token ID679
Feature activation+2.72e-01

Pos logit contributions
ps+0.010
ss+0.009
ches+0.008
mus+0.008
 teeth+0.008
Neg logit contributions
 she-0.006
 weekend-0.005
 day-0.005
 stair-0.005
 hallway-0.005
Token tried
Token ID3088
Feature activation+1.06e+00

Pos logit contributions
ps+0.014
ss+0.012
ches+0.011
mus+0.011
 teeth+0.011
Neg logit contributions
 she-0.008
 weekend-0.008
 day-0.008
 hallway-0.008
 stair-0.008
Token and
Token ID290
Feature activation-2.60e-01

Pos logit contributions
ps+0.058
ss+0.051
wy+0.050
ches+0.049
ealing+0.049
Neg logit contributions
 she-0.031
 day-0.028
 stair-0.027
 he-0.027
 they-0.027
Token You
Token ID921
Feature activation+1.23e+00

Pos logit contributions
ps+0.001
ss+0.001
ches+0.001
 teeth+0.001
mus+0.001
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token have
Token ID423
Feature activation+9.69e-01

Pos logit contributions
ps+0.066
iddles+0.060
rier+0.058
ss+0.058
mus+0.058
Neg logit contributions
 she-0.033
 day-0.030
 explore-0.029
 wander-0.029
 they-0.028
Token sharp
Token ID7786
Feature activation+2.36e+00

Pos logit contributions
ps+0.048
ss+0.043
mus+0.042
iddles+0.042
ches+0.042
Neg logit contributions
 she-0.030
 day-0.028
 weekend-0.026
 they-0.026
 stair-0.025
Token teeth
Token ID9941
Feature activation+2.47e+00

Pos logit contributions
ussy+0.055
ss+0.051
rier+0.050
mus+0.048
oni+0.048
Neg logit contributions
 she-0.120
 day-0.120
 explore-0.112
 he-0.109
 wander-0.108
Token and
Token ID290
Feature activation+7.26e-01

Pos logit contributions
ps+0.117
ss+0.107
iddles+0.107
ealing+0.102
mus+0.101
Neg logit contributions
 she-0.081
 day-0.074
 of-0.067
 he-0.067
 today-0.066
Token claws
Token ID28421
Feature activation+2.80e+00

Pos logit contributions
ps+0.028
ss+0.024
iddles+0.023
ches+0.023
mus+0.023
Neg logit contributions
 she-0.031
 day-0.029
 weekend-0.028
 stair-0.028
 hallway-0.027
Token.
Token ID13
Feature activation+8.40e-01

Pos logit contributions
ps+0.127
iddles+0.121
ss+0.116
wy+0.116
mus+0.115
Neg logit contributions
 she-0.096
 day-0.083
 of-0.082
 he-0.081
 they-0.076
Token You
Token ID921
Feature activation+1.39e+00

Pos logit contributions
ps+0.044
ss+0.038
ches+0.038
iddles+0.036
 teeth+0.036
Neg logit contributions
 she-0.024
 day-0.021
 weekend-0.021
 hallway-0.021
 stair-0.021
Token want
Token ID765
Feature activation+1.78e+00

Pos logit contributions
ps+0.075
iddles+0.069
rier+0.069
ss+0.067
mus+0.066
Neg logit contributions
 she-0.036
 day-0.032
 explore-0.032
 wander-0.031
 they-0.030
Token to
Token ID284
Feature activation+9.50e-01

Pos logit contributions
ps+0.111
ealing+0.103
rier+0.101
ss+0.101
wy+0.100
Neg logit contributions
 she-0.040
 day-0.037
 explore-0.031
 they-0.031
 holiday-0.030
Token hurt
Token ID5938
Feature activation+1.92e+00

Pos logit contributions
ps+0.055
ss+0.050
rier+0.049
iddles+0.049
mus+0.048
Neg logit contributions
 she-0.022
 explore-0.020
 wander-0.020
 day-0.020
 hallway-0.018
Token You
Token ID921
Feature activation+9.76e-01

Pos logit contributions
ps+0.025
ss+0.022
ches+0.022
iddles+0.021
mus+0.021
Neg logit contributions
 she-0.010
 weekend-0.009
 stair-0.009
 day-0.009
 hallway-0.009
Token should
Token ID815
Feature activation-6.23e-01

Pos logit contributions
ps+0.046
iddles+0.039
ss+0.039
mus+0.038
rier+0.037
Neg logit contributions
 she-0.034
 day-0.031
 stair-0.030
 explore-0.030
 wander-0.030
Token come
Token ID1282
Feature activation+1.02e+00

Pos logit contributions
 weekend+0.020
 hallway+0.020
 stair+0.019
 she+0.019
 day+0.019
Neg logit contributions
ps-0.030
ss-0.026
 teeth-0.025
ches-0.025
mus-0.023
Token down
Token ID866
Feature activation+1.37e+00

Pos logit contributions
ps+0.045
ss+0.040
ches+0.038
wy+0.038
iddles+0.038
Neg logit contributions
 she-0.038
 day-0.036
 explore-0.035
 stair-0.034
 weekend-0.033
Token.
Token ID13
Feature activation+6.13e-01

Pos logit contributions
ps+0.059
ss+0.053
ches+0.051
iddles+0.050
 teeth+0.050
Neg logit contributions
 she-0.055
 day-0.051
 weekend-0.050
 they-0.049
 stair-0.049
Token Mom
Token ID11254
Feature activation+2.22e+00

Pos logit contributions
ps+0.033
ss+0.029
ches+0.029
iddles+0.028
 teeth+0.027
Neg logit contributions
 she-0.016
 day-0.015
 weekend-0.015
 stair-0.015
 hallway-0.015
Token said
Token ID531
Feature activation+7.54e-01

Pos logit contributions
ps+0.109
rier+0.099
mus+0.096
 unnecessary+0.096
iddles+0.095
Neg logit contributions
 she-0.074
 day-0.066
 he-0.061
 they-0.061
 weekend-0.056
Token not
Token ID407
Feature activation+2.45e-01

Pos logit contributions
ps+0.042
ss+0.038
iddles+0.037
ches+0.037
mus+0.036
Neg logit contributions
 she-0.020
 day-0.017
 they-0.017
 weekend-0.016
 stair-0.016
Token to
Token ID284
Feature activation+1.15e+00

Pos logit contributions
ps+0.012
ss+0.011
ches+0.011
mus+0.010
 teeth+0.010
Neg logit contributions
 she-0.007
 day-0.007
 weekend-0.007
 stair-0.007
 hallway-0.007
Token climb
Token ID12080
Feature activation+2.40e+00

Pos logit contributions
ps+0.066
iddles+0.058
ches+0.058
ss+0.056
mus+0.056
Neg logit contributions
 she-0.031
 day-0.029
 wander-0.027
 explore-0.025
 stair-0.025
Token the
Token ID262
Feature activation+8.73e-01

Pos logit contributions
ps+0.129
ss+0.123
iddles+0.121
ealing+0.120
uable+0.116
Neg logit contributions
 she-0.071
 day-0.063
 they-0.061
 stair-0.056
 he-0.056
Token,
Token ID11
Feature activation+8.39e-01

Pos logit contributions
ps+0.013
ss+0.012
ches+0.011
 teeth+0.011
mus+0.011
Neg logit contributions
 she-0.008
 day-0.007
 weekend-0.007
 stair-0.007
 hallway-0.007
Token "
Token ID366
Feature activation-1.46e+00

Pos logit contributions
ps+0.037
ss+0.032
ches+0.031
 teeth+0.031
wy+0.030
Neg logit contributions
 she-0.033
 they-0.029
 day-0.029
 weekend-0.029
 stair-0.029
TokenLet
Token ID5756
Feature activation+6.38e-01

Pos logit contributions
 stair+0.050
 hallway+0.048
 holiday+0.046
 explore+0.046
 hall+0.046
Neg logit contributions
ps-0.060
ss-0.059
mus-0.055
 louder-0.053
ches-0.052
Token your
Token ID534
Feature activation+3.71e-01

Pos logit contributions
ps+0.029
iddles+0.025
mus+0.025
ss+0.024
 teeth+0.024
Neg logit contributions
 she-0.024
 day-0.022
 weekend-0.022
 they-0.021
 stair-0.020
Token shadow
Token ID9082
Feature activation-7.63e-01

Pos logit contributions
ps+0.018
ss+0.016
ches+0.016
mus+0.016
iddles+0.015
Neg logit contributions
 she-0.012
 day-0.011
 stair-0.011
 hallway-0.011
 weekend-0.011
Token go
Token ID467
Feature activation-7.10e-01

Pos logit contributions
 weekend+0.025
 hallway+0.025
 stair+0.025
 she+0.024
 day+0.024
Neg logit contributions
ps-0.035
ss-0.030
ches-0.029
 teeth-0.029
mus-0.028
Token on
Token ID319
Feature activation+1.98e-01

Pos logit contributions
 stair+0.030
 weekend+0.030
 hallway+0.030
 she+0.030
 day+0.029
Neg logit contributions
ps-0.025
ches-0.021
ss-0.021
mus-0.019
 teeth-0.018
Token a
Token ID257
Feature activation-3.90e-01

Pos logit contributions
ps+0.008
ss+0.007
ches+0.007
 teeth+0.007
mus+0.007
Neg logit contributions
 she-0.007
 weekend-0.007
 day-0.007
 stair-0.007
 hallway-0.007
Token walk
Token ID2513
Feature activation+2.66e-01

Pos logit contributions
 she+0.010
 weekend+0.010
 hallway+0.009
 stair+0.009
 day+0.009
Neg logit contributions
ps-0.020
ches-0.018
ss-0.018
 teeth-0.017
mus-0.017
Token.
Token ID13
Feature activation-4.88e-01

Pos logit contributions
ps+0.011
ss+0.009
 teeth+0.009
ches+0.009
mus+0.009
Neg logit contributions
 she-0.010
 weekend-0.010
 day-0.010
 stair-0.010
 hallway-0.010
Token I
Token ID314
Feature activation-1.88e-01

Pos logit contributions
 she+0.015
 weekend+0.014
 hallway+0.014
 stair+0.014
 day+0.014
Neg logit contributions
ps-0.024
ss-0.021
 teeth-0.020
ches-0.020
mus-0.020
TokenL
Token ID43
Feature activation-8.17e-01

Pos logit contributions
 stair+0.013
 hallway+0.013
 weekend+0.013
 she+0.013
 holiday+0.012
Neg logit contributions
ps-0.023
ss-0.022
mus-0.020
ches-0.020
 teeth-0.020
Tokenily
Token ID813
Feature activation+2.60e-01

Pos logit contributions
 hallway+0.038
 stair+0.037
 weekend+0.036
 she+0.036
 driveway+0.036
Neg logit contributions
ps-0.029
ss-0.025
ches-0.022
iddles-0.021
mus-0.020
Token,
Token ID11
Feature activation-1.39e+00

Pos logit contributions
ps+0.013
ss+0.011
ches+0.011
mus+0.011
 teeth+0.011
Neg logit contributions
 she-0.008
 day-0.007
 weekend-0.007
 stair-0.007
 hallway-0.007
Token I
Token ID314
Feature activation+1.22e-01

Pos logit contributions
 hallway+0.053
 weekend+0.049
 driveway+0.049
 stair+0.046
 particular+0.046
Neg logit contributions
ps-0.060
ss-0.054
 teeth-0.050
ches-0.047
uable-0.046
Token like
Token ID588
Feature activation+2.18e-01

Pos logit contributions
ps+0.006
ss+0.006
ches+0.005
 teeth+0.005
mus+0.005
Neg logit contributions
 she-0.004
 weekend-0.003
 day-0.003
 hallway-0.003
 stair-0.003
Token you
Token ID345
Feature activation+1.65e+00

Pos logit contributions
ps+0.010
ss+0.009
ches+0.008
mus+0.008
 teeth+0.008
Neg logit contributions
 she-0.008
 weekend-0.007
 day-0.007
 stair-0.007
 hallway-0.007
Token a
Token ID257
Feature activation+7.69e-01

Pos logit contributions
ps+0.078
ss+0.070
iddles+0.069
mus+0.068
rier+0.067
Neg logit contributions
 she-0.057
 day-0.057
 weekend-0.052
 they-0.050
days-0.049
Token lot
Token ID1256
Feature activation+2.22e-01

Pos logit contributions
ps+0.036
ches+0.030
ss+0.029
mus+0.029
iddles+0.028
Neg logit contributions
 she-0.028
 day-0.028
 weekend-0.027
 stair-0.027
 holiday-0.026
Token,
Token ID11
Feature activation-5.96e-01

Pos logit contributions
ps+0.011
ss+0.010
 teeth+0.010
ches+0.009
mus+0.009
Neg logit contributions
 she-0.007
 weekend-0.006
 day-0.006
 stair-0.006
 hallway-0.006
Token but
Token ID475
Feature activation-1.35e+00

Pos logit contributions
 hallway+0.022
 weekend+0.021
 stair+0.021
 she+0.021
 day+0.021
Neg logit contributions
ps-0.025
ss-0.022
 teeth-0.021
ches-0.021
mus-0.020
Token I
Token ID314
Feature activation+3.20e-01

Pos logit contributions
 hallway+0.044
 driveway+0.041
 stair+0.040
 weekend+0.040
days+0.039
Neg logit contributions
ps-0.063
ss-0.055
 teeth-0.055
uable-0.051
ches-0.050
Token.
Token ID13
Feature activation-5.56e-01

Pos logit contributions
ps+0.000
ss+0.000
ches+0.000
 teeth+0.000
mus+0.000
Neg logit contributions
 she-0.000
 weekend-0.000
 stair-0.000
 hallway-0.000
 day-0.000
Token One
Token ID1881
Feature activation-2.18e+00

Pos logit contributions
 weekend+0.015
 hallway+0.015
 she+0.014
 stair+0.014
 holiday+0.014
Neg logit contributions
ps-0.027
ss-0.025
 teeth-0.024
mus-0.024
ches-0.023
Token day
Token ID1110
Feature activation-1.33e+00

Pos logit contributions
stack+0.048
 driveway+0.045
 sandbox+0.042
 hallway+0.041
 stair+0.041
Neg logit contributions
 teeth-0.102
ss-0.101
 funny-0.100
ps-0.099
p-0.095
Token she
Token ID673
Feature activation-1.52e+00

Pos logit contributions
 driveway+0.063
 hallway+0.063
 weekend+0.062
 occasion+0.058
 stair+0.058
Neg logit contributions
 teeth-0.036
ps-0.035
ss-0.034
mus-0.032
ches-0.028
Token found
Token ID1043
Feature activation+5.62e-01

Pos logit contributions
 stair+0.057
stack+0.057
 hallway+0.052
 exploration+0.052
 weekend+0.052
Neg logit contributions
ps-0.056
 teeth-0.051
ches-0.051
ss-0.050
mus-0.046
Token a
Token ID257
Feature activation-2.70e-02

Pos logit contributions
ps+0.023
ss+0.021
ches+0.020
mus+0.019
 teeth+0.019
Neg logit contributions
 she-0.023
 day-0.021
 weekend-0.021
 stair-0.020
 hallway-0.020
Token zip
Token ID19974
Feature activation-1.86e+00

Pos logit contributions
 she+0.001
 weekend+0.001
 day+0.001
 hallway+0.001
 stair+0.001
Neg logit contributions
ps-0.001
ss-0.001
ches-0.001
 teeth-0.001
mus-0.001
Token at
Token ID379
Feature activation+7.31e-02

Pos logit contributions
days+0.077
 weekend+0.075
 holiday+0.075
 hallway+0.073
Jane+0.072
Neg logit contributions
ps-0.065
ches-0.056
ss-0.053
 teeth-0.048
rier-0.046
Token the
Token ID262
Feature activation+2.73e-02

Pos logit contributions
ps+0.003
ss+0.003
ches+0.003
 teeth+0.003
mus+0.003
Neg logit contributions
 she-0.002
 weekend-0.002
 day-0.002
 stair-0.002
 hallway-0.002
Token beach
Token ID10481
Feature activation-1.96e+00

Pos logit contributions
ps+0.001
ss+0.001
ches+0.001
 teeth+0.001
mus+0.001
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token.
Token ID13
Feature activation-6.20e-01

Pos logit contributions
 particular+0.102
 weekend+0.101
days+0.101
 hallway+0.100
 holiday+0.100
Neg logit contributions
ps-0.037
 teeth-0.030
ss-0.029
oni-0.025
ches-0.024
Token and
Token ID290
Feature activation-7.09e-01

Pos logit contributions
ps+0.079
ches+0.070
ss+0.069
wy+0.067
mus+0.067
Neg logit contributions
 she-0.045
 day-0.040
 they-0.035
 stair-0.035
 hallway-0.034
Token very
Token ID845
Feature activation+6.00e-01

Pos logit contributions
 weekend+0.027
 stair+0.027
 hallway+0.027
 driveway+0.027
 holiday+0.026
Neg logit contributions
ps-0.027
ss-0.023
 teeth-0.023
mus-0.022
ches-0.021
Token attractive
Token ID10966
Feature activation+1.24e-01

Pos logit contributions
ps+0.023
ss+0.019
ches+0.018
mus+0.017
iddles+0.017
Neg logit contributions
 she-0.027
 day-0.026
 weekend-0.025
 stair-0.025
 hallway-0.025
Token.
Token ID13
Feature activation+1.01e+00

Pos logit contributions
ps+0.005
ss+0.004
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.005
 weekend-0.005
 day-0.005
 stair-0.005
 hallway-0.005
Token One
Token ID1881
Feature activation-1.78e+00

Pos logit contributions
ps+0.060
ss+0.052
mus+0.049
wy+0.049
rier+0.049
Neg logit contributions
 she-0.027
 day-0.023
 weekend-0.022
 particular-0.022
 hallway-0.021
Token day
Token ID1110
Feature activation+8.19e-01

Pos logit contributions
stack+0.033
 driveway+0.031
 sandbox+0.031
 stair+0.028
 hallway+0.028
Neg logit contributions
ss-0.088
 teeth-0.088
ps-0.086
 funny-0.085
ches-0.083
Token she
Token ID673
Feature activation-2.08e+00

Pos logit contributions
ps+0.045
ss+0.039
iddles+0.038
ches+0.037
 teeth+0.037
Neg logit contributions
 she-0.024
 day-0.021
 they-0.020
 stair-0.020
 weekend-0.020
Token was
Token ID373
Feature activation-1.72e-01

Pos logit contributions
stack+0.095
 stair+0.090
 exploration+0.084
 particular+0.082
 weekend+0.081
Neg logit contributions
ps-0.062
 teeth-0.062
ches-0.057
ss-0.057
ze-0.048
Token in
Token ID287
Feature activation-2.34e+00

Pos logit contributions
 she+0.004
 weekend+0.004
 stair+0.004
 hallway+0.003
 day+0.003
Neg logit contributions
ps-0.010
ss-0.009
 teeth-0.009
ches-0.009
mus-0.009
Token her
Token ID607
Feature activation-1.56e+00

Pos logit contributions
boat+0.128
ite+0.121
loo+0.120
castle+0.118
days+0.118
Neg logit contributions
ps-0.054
 teeth-0.049
 loud-0.035
 x-0.035
 bad-0.034
Token bedroom
Token ID14043
Feature activation-2.49e+00

Pos logit contributions
boat+0.049
days+0.048
 contain+0.048
 weekend+0.047
Jane+0.047
Neg logit contributions
 teeth-0.064
ps-0.063
ss-0.055
iddles-0.051
 funny-0.049
Token toys
Token ID14958
Feature activation-1.44e+00

Pos logit contributions
 she+0.028
 weekend+0.028
 stair+0.028
 hallway+0.027
 day+0.027
Neg logit contributions
ps-0.025
 teeth-0.021
ss-0.021
ches-0.020
mus-0.019
Token in
Token ID287
Feature activation-8.80e-01

Pos logit contributions
 hallway+0.063
 stair+0.062
 weekend+0.062
 particular+0.060
 sandbox+0.059
Neg logit contributions
ps-0.047
ches-0.038
ss-0.037
 teeth-0.036
iddles-0.034
Token the
Token ID262
Feature activation-8.45e-01

Pos logit contributions
 weekend+0.033
 hallway+0.032
 stair+0.031
 sandbox+0.031
 day+0.031
Neg logit contributions
ps-0.034
 teeth-0.029
ss-0.028
mus-0.027
ches-0.027
Token attic
Token ID46113
Feature activation-1.68e+00

Pos logit contributions
 she+0.016
days+0.015
 weekend+0.015
 they+0.015
 stair+0.014
Neg logit contributions
ps-0.046
ss-0.043
 teeth-0.042
ches-0.041
mus-0.040
Token and
Token ID290
Feature activation-2.42e+00

Pos logit contributions
 weekend+0.080
 hallway+0.079
 stair+0.079
 holiday+0.079
 particular+0.079
Neg logit contributions
ps-0.043
ches-0.033
ss-0.033
 teeth-0.031
rier-0.029
Token she
Token ID673
Feature activation-7.16e-02

Pos logit contributions
loo+0.085
 stair+0.082
 weekend+0.080
 hallway+0.077
boat+0.075
Neg logit contributions
ps-0.097
 teeth-0.091
ches-0.079
 smarter-0.075
 better-0.074
Token put
Token ID1234
Feature activation+1.56e+00

Pos logit contributions
 she+0.002
 weekend+0.002
 stair+0.002
 day+0.002
 hallway+0.002
Neg logit contributions
ps-0.003
ss-0.003
ches-0.003
 teeth-0.003
mus-0.003
Token them
Token ID606
Feature activation+1.17e-01

Pos logit contributions
ps+0.065
ss+0.060
 teeth+0.054
ealing+0.054
ache+0.054
Neg logit contributions
 she-0.064
 day-0.058
 they-0.056
 weekend-0.056
 holiday-0.054
Token in
Token ID287
Feature activation-8.91e-02

Pos logit contributions
ps+0.005
ss+0.005
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.004
 weekend-0.004
 day-0.004
 stair-0.004
 hallway-0.004
Token a
Token ID257
Feature activation-4.97e-01

Pos logit contributions
 she+0.003
 weekend+0.003
 stair+0.003
 hallway+0.003
 day+0.003
Neg logit contributions
ps-0.004
ss-0.003
 teeth-0.003
ches-0.003
mus-0.003
Token different
Token ID1180
Feature activation-9.78e-01

Pos logit contributions
 she+0.020
 weekend+0.019
 stair+0.018
 day+0.018
 hallway+0.018
Neg logit contributions
ps-0.018
ss-0.016
ches-0.016
 teeth-0.015
mus-0.015
Token and
Token ID290
Feature activation-8.26e-01

Pos logit contributions
 weekend+0.068
 hallway+0.068
 stair+0.068
 day+0.066
 particular+0.066
Neg logit contributions
ps-0.063
ss-0.054
 teeth-0.053
ches-0.050
mus-0.049
Token the
Token ID262
Feature activation-6.07e-01

Pos logit contributions
 stair+0.022
 hallway+0.021
 weekend+0.021
 day+0.020
 she+0.020
Neg logit contributions
ps-0.042
ss-0.038
 teeth-0.037
mus-0.036
ches-0.036
Token house
Token ID2156
Feature activation-1.18e+00

Pos logit contributions
 she+0.018
 weekend+0.017
 stair+0.017
 hallway+0.017
 day+0.016
Neg logit contributions
ps-0.029
ss-0.025
 teeth-0.025
mus-0.024
iddles-0.024
Token was
Token ID373
Feature activation+8.78e-02

Pos logit contributions
 stair+0.036
 weekend+0.036
 she+0.035
 hallway+0.035
 particular+0.034
Neg logit contributions
ps-0.055
ss-0.048
 teeth-0.046
ches-0.045
mus-0.043
Token very
Token ID845
Feature activation+7.26e-02

Pos logit contributions
ps+0.004
ss+0.003
ches+0.003
 teeth+0.003
mus+0.003
Neg logit contributions
 she-0.003
 weekend-0.003
 day-0.003
 hallway-0.003
 stair-0.003
Token busy
Token ID8179
Feature activation-1.65e+00

Pos logit contributions
ps+0.003
ss+0.002
ches+0.002
 teeth+0.002
mus+0.002
Neg logit contributions
 she-0.003
 weekend-0.003
 day-0.003
 stair-0.003
 hallway-0.003
Token.
Token ID13
Feature activation+6.77e-01

Pos logit contributions
 hallway+0.078
 stair+0.078
 weekend+0.078
 sandbox+0.077
 she+0.077
Neg logit contributions
ps-0.047
 teeth-0.035
ss-0.033
mus-0.032
ches-0.032
Token Everyone
Token ID11075
Feature activation-1.47e+00

Pos logit contributions
ps+0.036
ss+0.031
ches+0.031
wy+0.030
mus+0.030
Neg logit contributions
 she-0.019
 weekend-0.018
 day-0.018
 hallway-0.018
 stair-0.018
Token knew
Token ID2993
Feature activation-8.12e-01

Pos logit contributions
 stair+0.055
 hallway+0.053
 weekend+0.052
 driveway+0.052
 sandbox+0.051
Neg logit contributions
ps-0.059
ches-0.049
ss-0.049
 teeth-0.047
mus-0.045
Token they
Token ID484
Feature activation-4.74e-01

Pos logit contributions
 hallway+0.024
 stair+0.024
 weekend+0.023
 day+0.022
 she+0.022
Neg logit contributions
ps-0.040
ss-0.035
 teeth-0.034
ches-0.033
mus-0.032
Token could
Token ID714
Feature activation-1.19e+00

Pos logit contributions
 stair+0.013
 she+0.013
 hallway+0.012
 weekend+0.012
 day+0.012
Neg logit contributions
ps-0.024
ss-0.022
ches-0.021
 teeth-0.021
mus-0.020
Token We
Token ID775
Feature activation+1.10e-01

Pos logit contributions
ps+0.030
ss+0.026
ches+0.025
 teeth+0.025
mus+0.025
Neg logit contributions
 she-0.016
 weekend-0.015
 day-0.015
 hallway-0.015
 stair-0.014
Token need
Token ID761
Feature activation+2.81e-01

Pos logit contributions
ps+0.005
ss+0.005
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.004
 weekend-0.003
 day-0.003
 stair-0.003
 hallway-0.003
Token to
Token ID284
Feature activation-6.42e-01

Pos logit contributions
ps+0.014
ss+0.012
ches+0.012
 teeth+0.012
mus+0.012
Neg logit contributions
 she-0.009
 weekend-0.008
 day-0.008
 hallway-0.008
 stair-0.008
Token go
Token ID467
Feature activation-7.28e-01

Pos logit contributions
 she+0.020
 stair+0.019
 weekend+0.019
 hallway+0.019
 day+0.018
Neg logit contributions
ps-0.031
ss-0.027
ches-0.026
 teeth-0.026
mus-0.025
Token home
Token ID1363
Feature activation+1.66e+00

Pos logit contributions
 stair+0.027
 she+0.027
 hallway+0.026
 weekend+0.026
 particular+0.026
Neg logit contributions
ps-0.030
ches-0.025
ss-0.025
 teeth-0.023
mus-0.023
Token and
Token ID290
Feature activation-2.82e+00

Pos logit contributions
ps+0.080
ss+0.075
ches+0.069
 teeth+0.069
ealing+0.068
Neg logit contributions
 she-0.055
 day-0.053
 weekend-0.053
 they-0.050
 holiday-0.049
Token rest
Token ID1334
Feature activation+8.02e-01

Pos logit contributions
 stair+0.105
 hallway+0.099
 particular+0.099
loo+0.096
 weekend+0.093
Neg logit contributions
ps-0.105
 teeth-0.095
 better-0.093
ches-0.089
 louder-0.088
Token."
Token ID526
Feature activation-1.31e-02

Pos logit contributions
ps+0.028
ss+0.025
iddles+0.023
mus+0.022
 teeth+0.022
Neg logit contributions
 she-0.037
 day-0.036
 weekend-0.035
 they-0.034
 stair-0.034
Token
Token ID220
Feature activation+1.56e+00

Pos logit contributions
 she+0.000
 weekend+0.000
 stair+0.000
 day+0.000
 hallway+0.000
Neg logit contributions
ps-0.001
ss-0.001
ches-0.001
 teeth-0.001
mus-0.001
Token
Token ID198
Feature activation+6.62e-01

Pos logit contributions
ps+0.092
ss+0.080
wy+0.080
mus+0.079
 teeth+0.078
Neg logit contributions
 she-0.038
 they-0.032
 day-0.030
 he-0.029
 weekend-0.028
Token
Token ID198
Feature activation-2.64e-01

Pos logit contributions
ps+0.037
ss+0.032
 teeth+0.031
ches+0.030
iddles+0.030
Neg logit contributions
 she-0.019
 day-0.017
 weekend-0.017
 hallway-0.016
 they-0.016
Token man
Token ID582
Feature activation-1.18e+00

Pos logit contributions
ps+0.063
ss+0.053
ches+0.050
mus+0.050
 teeth+0.050
Neg logit contributions
 she-0.042
 day-0.041
 stair-0.038
 weekend-0.038
 hallway-0.037
Token.
Token ID13
Feature activation+4.79e-02

Pos logit contributions
 stair+0.042
 weekend+0.040
 hallway+0.040
 holiday+0.038
 she+0.038
Neg logit contributions
ps-0.049
ss-0.046
mus-0.044
ches-0.043
 teeth-0.043
Token Every
Token ID3887
Feature activation-1.08e+00

Pos logit contributions
ps+0.002
ss+0.002
ches+0.002
 teeth+0.002
mus+0.002
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token day
Token ID1110
Feature activation-6.46e-01

Pos logit contributions
 she+0.012
 hallway+0.011
 stair+0.011
 sandbox+0.010
 driveway+0.010
Neg logit contributions
ps-0.067
ss-0.064
ches-0.063
 teeth-0.063
mus-0.061
Token he
Token ID339
Feature activation+4.65e-01

Pos logit contributions
 hallway+0.019
 weekend+0.018
 stair+0.018
 driveway+0.017
days+0.017
Neg logit contributions
ps-0.030
ss-0.029
 teeth-0.028
mus-0.027
ches-0.026
Token would
Token ID561
Feature activation-1.65e+00

Pos logit contributions
ps+0.026
ss+0.023
 teeth+0.022
mus+0.022
ches+0.021
Neg logit contributions
 she-0.012
 day-0.011
 weekend-0.011
 hallway-0.011
 stair-0.011
Token take
Token ID1011
Feature activation+1.74e+00

Pos logit contributions
 weekend+0.048
 stair+0.047
 she+0.047
 hallway+0.046
 policy+0.045
Neg logit contributions
ps-0.077
ches-0.068
 teeth-0.065
ss-0.065
mus-0.064
Token a
Token ID257
Feature activation-9.05e-01

Pos logit contributions
ps+0.089
rier+0.079
ss+0.078
ches+0.077
ussy+0.077
Neg logit contributions
 day-0.056
 she-0.054
 holiday-0.053
 he-0.050
 summer-0.049
Token big
Token ID1263
Feature activation-1.41e+00

Pos logit contributions
 she+0.021
 hallway+0.020
 weekend+0.020
 stair+0.019
days+0.018
Neg logit contributions
ps-0.046
ches-0.044
ss-0.044
 teeth-0.042
mus-0.041
Token bucket
Token ID19236
Feature activation-2.80e+00

Pos logit contributions
 she+0.041
 hallway+0.038
 weekend+0.038
days+0.037
Jane+0.036
Neg logit contributions
ps-0.064
ches-0.060
 teeth-0.059
ss-0.059
mus-0.055
Token outside
Token ID2354
Feature activation-2.19e+00

Pos logit contributions
 hallway+0.101
 stair+0.098
days+0.093
 sandbox+0.093
 particular+0.091
Neg logit contributions
ches-0.103
ps-0.103
oni-0.092
ss-0.089
 stronger-0.085
Token monkey
Token ID21657
Feature activation+1.99e+00

Pos logit contributions
ps+0.026
ches+0.023
ss+0.023
mus+0.022
 teeth+0.022
Neg logit contributions
 day-0.016
 she-0.016
 stair-0.015
 weekend-0.015
 hallway-0.015
Token yelled
Token ID22760
Feature activation+2.80e+00

Pos logit contributions
ps+0.113
iddles+0.098
rier+0.098
 unnecessary+0.096
mus+0.096
Neg logit contributions
 she-0.055
 day-0.048
 he-0.046
 they-0.041
 stair-0.041
Token back
Token ID736
Feature activation+9.85e-01

Pos logit contributions
wy+0.174
ps+0.165
iddles+0.161
ymes+0.156
ws+0.155
Neg logit contributions
 she-0.069
 he-0.061
 they-0.060
 of-0.042
 exploring-0.042
Token,
Token ID11
Feature activation+1.03e+00

Pos logit contributions
ps+0.054
wy+0.047
ches+0.047
ss+0.047
mus+0.045
Neg logit contributions
 she-0.029
 day-0.026
 stair-0.024
 they-0.024
 he-0.024
Token "
Token ID366
Feature activation-8.25e-01

Pos logit contributions
ps+0.046
ss+0.041
ches+0.040
wy+0.040
iddles+0.040
Neg logit contributions
 she-0.040
 they-0.035
 he-0.034
 day-0.034
 hallway-0.033
TokenBut
Token ID1537
Feature activation-7.52e-01

Pos logit contributions
 weekend+0.022
 stair+0.022
 hallway+0.022
 holiday+0.022
 she+0.021
Neg logit contributions
ps-0.041
ss-0.040
mus-0.037
ches-0.036
 teeth-0.035
Token why
Token ID1521
Feature activation+5.43e-01

Pos logit contributions
 weekend+0.020
 hallway+0.020
 stair+0.019
 she+0.019
 day+0.019
Neg logit contributions
ps-0.039
ss-0.035
 teeth-0.034
ches-0.032
mus-0.032
Token?
Token ID30
Feature activation+4.59e-01

Pos logit contributions
ps+0.026
ss+0.022
mus+0.022
ches+0.022
iddles+0.021
Neg logit contributions
 she-0.018
 day-0.017
 stair-0.017
 weekend-0.016
 hallway-0.016
Token I
Token ID314
Feature activation-9.60e-02

Pos logit contributions
ps+0.026
ss+0.023
ches+0.022
 teeth+0.022
mus+0.022
Neg logit contributions
 she-0.011
 stair-0.010
 weekend-0.010
 hallway-0.010
 day-0.010
Token'm
Token ID1101
Feature activation+1.05e+00

Pos logit contributions
 she+0.003
 weekend+0.003
 stair+0.003
 hallway+0.003
 day+0.003
Neg logit contributions
ps-0.005
ss-0.004
ches-0.004
 teeth-0.004
mus-0.004
Token trying
Token ID2111
Feature activation+1.22e+00

Pos logit contributions
ps+0.043
ss+0.038
ches+0.036
iddles+0.036
mus+0.036
Neg logit contributions
 she-0.043
 day-0.039
 stair-0.037
 weekend-0.037
 hallway-0.037
Token Lily
Token ID20037
Feature activation-9.50e-02

Pos logit contributions
 weekend+0.017
 hallway+0.017
 stair+0.016
 she+0.016
 day+0.016
Neg logit contributions
ps-0.020
 teeth-0.018
ss-0.018
ches-0.017
iddles-0.016
Token want
Token ID765
Feature activation+1.23e+00

Pos logit contributions
 she+0.003
 weekend+0.003
 stair+0.003
 hallway+0.003
 day+0.003
Neg logit contributions
ps-0.005
ss-0.004
ches-0.004
 teeth-0.004
mus-0.004
Token to
Token ID284
Feature activation-1.26e-01

Pos logit contributions
ps+0.076
ss+0.070
wy+0.069
ches+0.069
ealing+0.068
Neg logit contributions
 she-0.028
 they-0.027
 day-0.025
 weekend-0.024
 holiday-0.023
Token play
Token ID711
Feature activation+1.87e+00

Pos logit contributions
 she+0.003
 weekend+0.003
 hallway+0.003
 stair+0.003
 day+0.003
Neg logit contributions
ps-0.007
ss-0.006
 teeth-0.006
ches-0.006
mus-0.006
Token the
Token ID262
Feature activation+1.09e+00

Pos logit contributions
ps+0.115
ss+0.108
ache+0.104
ches+0.104
wy+0.104
Neg logit contributions
 she-0.043
 they-0.040
 day-0.039
 sunny-0.036
 he-0.036
Token puzzle
Token ID15027
Feature activation-8.61e-01

Pos logit contributions
ps+0.063
ss+0.057
 teeth+0.055
ches+0.055
ealing+0.055
Neg logit contributions
 weekend-0.027
 day-0.027
 stair-0.026
 she-0.024
 holiday-0.023
Token game
Token ID983
Feature activation-6.96e-01

Pos logit contributions
 hallway+0.034
 she+0.033
 weekend+0.033
 stair+0.033
 driveway+0.032
Neg logit contributions
ps-0.032
ches-0.027
ss-0.026
 teeth-0.025
mus-0.025
Token.
Token ID13
Feature activation+7.47e-01

Pos logit contributions
 hallway+0.031
 weekend+0.030
 stair+0.030
 she+0.030
 holiday+0.029
Neg logit contributions
ps-0.023
ss-0.019
ches-0.018
 teeth-0.018
mus-0.017
Token
Token ID198
Feature activation+5.38e-01

Pos logit contributions
ps+0.046
ches+0.039
ss+0.039
wy+0.038
mus+0.038
Neg logit contributions
 she-0.018
 they-0.016
 day-0.015
 weekend-0.015
 stair-0.014
Token
Token ID198
Feature activation-1.10e+00

Pos logit contributions
ps+0.028
ss+0.025
 teeth+0.024
ches+0.024
mus+0.024
Neg logit contributions
 she-0.015
 day-0.015
 weekend-0.014
 hallway-0.014
 stair-0.014
TokenThey
Token ID2990
Feature activation-7.05e-01

Pos logit contributions
 stair+0.048
 holiday+0.048
 weekend+0.048
days+0.047
 she+0.047
Neg logit contributions
ps-0.034
ss-0.033
ches-0.030
mus-0.030
ealing-0.026
Token to
Token ID284
Feature activation-8.51e-01

Pos logit contributions
 she+0.034
 weekend+0.034
 stair+0.034
 hallway+0.033
 particular+0.033
Neg logit contributions
ps-0.038
ches-0.031
 teeth-0.031
ss-0.030
iddles-0.029
Token the
Token ID262
Feature activation-8.68e-01

Pos logit contributions
 weekend+0.030
 stair+0.029
 hallway+0.029
 she+0.029
 day+0.029
Neg logit contributions
ps-0.037
 teeth-0.030
ss-0.029
mus-0.027
ches-0.027
Token poppy
Token ID47867
Feature activation-1.94e+00

Pos logit contributions
 she+0.022
 weekend+0.019
days+0.019
 day+0.018
 stair+0.018
Neg logit contributions
ps-0.046
ss-0.040
 teeth-0.039
ches-0.039
mus-0.038
Token in
Token ID287
Feature activation-1.34e+00

Pos logit contributions
 weekend+0.074
 particular+0.073
 hallway+0.070
 holiday+0.070
 stair+0.069
Neg logit contributions
ps-0.071
ches-0.064
 teeth-0.058
ss-0.057
mus-0.055
Token the
Token ID262
Feature activation-1.10e+00

Pos logit contributions
 weekend+0.054
 stair+0.054
loo+0.052
 sandbox+0.052
 hallway+0.051
Neg logit contributions
ps-0.050
 teeth-0.040
ss-0.037
ches-0.033
mus-0.033
Token sunny
Token ID27737
Feature activation-3.69e+00

Pos logit contributions
 she+0.028
days+0.028
 weekend+0.025
 stair+0.025
 they+0.025
Neg logit contributions
ps-0.055
ss-0.049
ches-0.048
 teeth-0.047
mus-0.045
Token field
Token ID2214
Feature activation-2.49e+00

Pos logit contributions
 she+0.112
path+0.109
 awaited+0.106
Not+0.105
days+0.104
Neg logit contributions
ps-0.126
 teeth-0.116
 co-0.111
ers-0.110
 me-0.110
Token.
Token ID13
Feature activation-1.67e+00

Pos logit contributions
 stair+0.122
 hallway+0.120
 weekend+0.120
 particular+0.119
 holiday+0.119
Neg logit contributions
ps-0.065
ches-0.047
ss-0.044
 teeth-0.043
ows-0.041
Token She
Token ID1375
Feature activation-8.41e-02

Pos logit contributions
 weekend+0.059
 stair+0.058
 day+0.058
 hallway+0.057
 trip+0.057
Neg logit contributions
ps-0.060
ss-0.057
 teeth-0.056
mus-0.053
ches-0.051
Token put
Token ID1234
Feature activation+1.50e-01

Pos logit contributions
 she+0.003
 weekend+0.003
 stair+0.002
 day+0.002
 hallway+0.002
Neg logit contributions
ps-0.004
ss-0.004
ches-0.003
 teeth-0.003
mus-0.003
Token the
Token ID262
Feature activation-3.98e-01

Pos logit contributions
ps+0.007
ss+0.006
ches+0.006
 teeth+0.006
mus+0.006
Neg logit contributions
 she-0.005
 weekend-0.005
 day-0.005
 stair-0.005
 hallway-0.005
Token,
Token ID11
Feature activation-2.65e+00

Pos logit contributions
 hallway+0.080
 wander+0.078
 sandbox+0.077
 occasion+0.077
 weekend+0.077
Neg logit contributions
ps-0.063
ches-0.055
ss-0.054
 teeth-0.050
oni-0.048
Token Mike
Token ID4995
Feature activation+6.16e-01

Pos logit contributions
 hallway+0.080
 wander+0.076
 distance+0.075
stack+0.075
 driveway+0.075
Neg logit contributions
ps-0.107
 teeth-0.103
ss-0.099
 good-0.092
 louder-0.091
Token saw
Token ID2497
Feature activation+1.42e-01

Pos logit contributions
ps+0.032
ss+0.028
 teeth+0.027
mus+0.026
ches+0.026
Neg logit contributions
 she-0.018
 day-0.017
 weekend-0.017
 hallway-0.017
 stair-0.016
Token a
Token ID257
Feature activation+1.31e-01

Pos logit contributions
ps+0.006
ss+0.005
ches+0.005
 teeth+0.005
mus+0.005
Neg logit contributions
 she-0.006
 weekend-0.006
 day-0.005
 stair-0.005
 hallway-0.005
Token big
Token ID1263
Feature activation-1.39e+00

Pos logit contributions
ps+0.005
ss+0.004
ches+0.004
 teeth+0.004
mus+0.004
Neg logit contributions
 she-0.006
 weekend-0.005
 day-0.005
 stair-0.005
 hallway-0.005
Token container
Token ID9290
Feature activation-4.00e+00

Pos logit contributions
 weekend+0.051
 she+0.051
days+0.050
 hallway+0.047
 wander+0.046
Neg logit contributions
ps-0.054
 teeth-0.049
ss-0.047
ches-0.043
mus-0.043
Token of
Token ID286
Feature activation-5.10e-01

Pos logit contributions
 wander+0.144
days+0.142
 weekend+0.141
 holiday+0.136
 explore+0.135
Neg logit contributions
ps-0.148
ers-0.134
oni-0.131
ows-0.129
 bigger-0.124
Token cream
Token ID8566
Feature activation-1.94e+00

Pos logit contributions
 she+0.020
 weekend+0.020
 stair+0.020
 hallway+0.020
 day+0.019
Neg logit contributions
ps-0.020
ss-0.016
 teeth-0.016
mus-0.015
ches-0.015
Token!
Token ID0
Feature activation-1.44e+00

Pos logit contributions
days+0.078
 wander+0.077
 weekend+0.076
 explore+0.074
 exploration+0.074
Neg logit contributions
ps-0.070
oni-0.061
ss-0.061
 teeth-0.057
rier-0.054
Token
Token ID220
Feature activation+8.22e-01

Pos logit contributions
 weekend+0.054
 day+0.052
 exploration+0.052
 wander+0.051
 stair+0.051
Neg logit contributions
ps-0.052
 teeth-0.051
ss-0.049
mus-0.044
ealing-0.041
Token
Token ID198
Feature activation-5.26e-01

Pos logit contributions
ps+0.044
ss+0.038
ches+0.038
mus+0.037
 teeth+0.036
Neg logit contributions
 she-0.023
 weekend-0.021
 day-0.020
 stair-0.020
 hallway-0.020
Token to
Token ID284
Feature activation-1.25e+00

Pos logit contributions
 she+0.010
 hallway+0.010
 weekend+0.010
 stair+0.010
 day+0.009
Neg logit contributions
ps-0.015
ss-0.013
 teeth-0.013
mus-0.012
ches-0.012
Token go
Token ID467
Feature activation-1.21e+00

Pos logit contributions
 weekend+0.034
 hallway+0.032
 she+0.031
 stair+0.031
 day+0.029
Neg logit contributions
ps-0.063
 teeth-0.057
ches-0.054
ss-0.053
mus-0.053
Token to
Token ID284
Feature activation-1.86e-01

Pos logit contributions
 stair+0.051
 hallway+0.050
 weekend+0.048
stack+0.048
 sandbox+0.048
Neg logit contributions
ps-0.040
ches-0.035
ss-0.034
 teeth-0.032
mus-0.032
Token the
Token ID262
Feature activation+2.53e-01

Pos logit contributions
 she+0.007
 weekend+0.007
 hallway+0.006
 stair+0.006
 day+0.006
Neg logit contributions
ps-0.008
ss-0.007
 teeth-0.007
ches-0.007
mus-0.006
Token beach
Token ID10481
Feature activation-1.82e+00

Pos logit contributions
ps+0.016
ss+0.015
ches+0.015
 teeth+0.014
mus+0.014
Neg logit contributions
 she-0.004
 weekend-0.004
 day-0.004
 stair-0.004
 hallway-0.004
Token and
Token ID290
Feature activation-3.61e+00

Pos logit contributions
 hallway+0.079
 stair+0.078
 driveway+0.076
 weekend+0.076
 policy+0.075
Neg logit contributions
ps-0.053
 teeth-0.045
ches-0.043
ss-0.043
iddles-0.043
Token play
Token ID711
Feature activation-1.11e+00

Pos logit contributions
 policy+0.164
loo+0.144
oker+0.131
olt+0.131
umm+0.130
Neg logit contributions
 teeth-0.095
 better-0.093
 funny-0.093
 w-0.092
ps-0.091
Token with
Token ID351
Feature activation-5.12e-01

Pos logit contributions
 hallway+0.047
 stair+0.046
 driveway+0.045
 weekend+0.045
 she+0.043
Neg logit contributions
ps-0.040
ss-0.032
mus-0.031
ches-0.031
 teeth-0.030
Token the
Token ID262
Feature activation-8.68e-01

Pos logit contributions
 weekend+0.019
 hallway+0.018
 driveway+0.017
 stair+0.017
 day+0.017
Neg logit contributions
ps-0.022
 teeth-0.019
ss-0.018
ches-0.017
mus-0.017
Token sand
Token ID6450
Feature activation-1.91e+00

Pos logit contributions
 she+0.031
 hallway+0.030
 stair+0.029
days+0.029
 weekend+0.029
Neg logit contributions
ps-0.037
ss-0.031
 teeth-0.031
iddles-0.028
rier-0.028
Token.
Token ID13
Feature activation-8.45e-01

Pos logit contributions
 hallway+0.089
 weekend+0.087
where+0.086
 particular+0.086
 stair+0.083
Neg logit contributions
ps-0.062
 teeth-0.049
ss-0.047
oni-0.045
iddles-0.044
Token She
Token ID1375
Feature activation+7.97e-01

Pos logit contributions
ps+0.035
ss+0.030
ches+0.030
 teeth+0.029
mus+0.029
Neg logit contributions
 she-0.013
 stair-0.012
 weekend-0.012
 day-0.012
 hallway-0.011
Token only
Token ID691
Feature activation-1.08e-01

Pos logit contributions
ps+0.041
ss+0.037
mus+0.035
ches+0.035
iddles+0.035
Neg logit contributions
 she-0.023
 day-0.021
 hallway-0.021
 weekend-0.020
 stair-0.020
Token saw
Token ID2497
Feature activation+4.77e-01

Pos logit contributions
 she+0.003
 weekend+0.003
 stair+0.003
 hallway+0.003
 day+0.003
Neg logit contributions
ps-0.005
ss-0.005
 teeth-0.005
ches-0.005
mus-0.005
Token a
Token ID257
Feature activation+1.63e-01

Pos logit contributions
ps+0.020
ss+0.017
ches+0.016
mus+0.016
 teeth+0.016
Neg logit contributions
 she-0.019
 weekend-0.018
 day-0.018
 stair-0.017
 hallway-0.017
Token dusty
Token ID36972
Feature activation-1.52e+00

Pos logit contributions
ps+0.007
ss+0.006
ches+0.005
 teeth+0.005
mus+0.005
Neg logit contributions
 she-0.006
 weekend-0.006
 stair-0.006
 day-0.006
 hallway-0.006
Token yard
Token ID12699
Feature activation-3.59e+00

Pos logit contributions
 she+0.049
 weekend+0.047
days+0.047
 they+0.045
where+0.044
Neg logit contributions
ps-0.066
 teeth-0.059
ss-0.056
ches-0.053
rier-0.052
Token and
Token ID290
Feature activation-2.26e+00

Pos logit contributions
days+0.166
des+0.165
 weekend+0.157
 occasion+0.156
Coming+0.155
Neg logit contributions
ps-0.091
ers-0.072
ows-0.067
ss-0.064
 bigger-0.063
Token a
Token ID257
Feature activation-1.05e+00

Pos logit contributions
 weekend+0.094
 particular+0.092
loo+0.091
boat+0.089
ite+0.088
Neg logit contributions
ps-0.082
 teeth-0.076
 louder-0.060
 horns-0.056
 words-0.056
Token broken
Token ID5445
Feature activation-1.27e+00

Pos logit contributions
 weekend+0.037
 she+0.037
 day+0.035
days+0.034
 particular+0.034
Neg logit contributions
ps-0.043
ss-0.038
 teeth-0.038
mus-0.035
ches-0.035
Token fence
Token ID13990
Feature activation-2.29e+00

Pos logit contributions
 she+0.036
 weekend+0.035
days+0.035
 day+0.033
 particular+0.032
Neg logit contributions
ps-0.062
 teeth-0.056
ss-0.053
ches-0.050
rier-0.050
Token.
Token ID13
Feature activation+4.00e-01

Pos logit contributions
 weekend+0.116
days+0.113
 particular+0.112
 holiday+0.111
 occasion+0.109
Neg logit contributions
ps-0.062
ches-0.046
 teeth-0.043
ss-0.042
ows-0.038
Token Tim
Token ID5045
Feature activation+3.67e-02

Pos logit contributions
 she+0.021
 weekend+0.021
 hallway+0.020
 stair+0.020
 holiday+0.019
Neg logit contributions
ps-0.034
ss-0.032
 teeth-0.031
mus-0.030
iddles-0.029
Tokenmy
Token ID1820
Feature activation-1.29e+00

Pos logit contributions
ps+0.002
ss+0.002
ches+0.002
 teeth+0.002
mus+0.002
Neg logit contributions
 she-0.001
 weekend-0.001
 day-0.001
 stair-0.001
 hallway-0.001
Token was
Token ID373
Feature activation-3.63e-01

Pos logit contributions
 stair+0.056
 weekend+0.054
 she+0.053
 hallway+0.052
 particular+0.052
Neg logit contributions
ps-0.042
 teeth-0.039
ss-0.038
ches-0.036
mus-0.034
Token very
Token ID845
Feature activation+3.01e-01

Pos logit contributions
 she+0.018
 stair+0.018
 weekend+0.018
 hallway+0.018
 day+0.017
Neg logit contributions
ps-0.010
ss-0.009
 teeth-0.008
mus-0.008
ches-0.007
Token excited
Token ID6568
Feature activation-1.89e+00

Pos logit contributions
ps+0.011
ss+0.009
ches+0.009
 teeth+0.009
mus+0.009
Neg logit contributions
 she-0.013
 day-0.013
 weekend-0.013
 hallway-0.012
 stair-0.012
Token because
Token ID780
Feature activation-3.60e+00

Pos logit contributions
stack+0.088
 hallway+0.087
 sandbox+0.087
 particular+0.087
 stair+0.085
Neg logit contributions
 teeth-0.048
ps-0.047
ss-0.041
ches-0.040
iddles-0.039
Token he
Token ID339
Feature activation+2.80e-01

Pos logit contributions
stack+0.142
 wander+0.139
 contain+0.136
 perspective+0.135
field+0.134
Neg logit contributions
 better-0.109
 teeth-0.106
 too-0.106
 stronger-0.105
 bigger-0.105
Token was
Token ID373
Feature activation-2.31e-01

Pos logit contributions
ps+0.015
ss+0.013
 teeth+0.013
ches+0.013
mus+0.013
Neg logit contributions
 she-0.008
 day-0.007
 weekend-0.007
 hallway-0.007
 stair-0.007
Token going
Token ID1016
Feature activation+5.16e-01

Pos logit contributions
 she+0.007
 stair+0.006
 hallway+0.006
 weekend+0.006
 day+0.006
Neg logit contributions
ps-0.012
ss-0.010
 teeth-0.010
mus-0.010
ches-0.010
Token on
Token ID319
Feature activation+2.94e-01

Pos logit contributions
ps+0.027
ss+0.024
 teeth+0.023
ches+0.023
mus+0.023
Neg logit contributions
 she-0.015
 weekend-0.014
 day-0.014
 hallway-0.013
 stair-0.013
Token a
Token ID257
Feature activation-9.90e-01

Pos logit contributions
ps+0.012
ss+0.010
ches+0.010
 teeth+0.009
mus+0.009
Neg logit contributions
 she-0.012
 weekend-0.011
 day-0.011
 hallway-0.011
 stair-0.011