TOP ACTIVATIONS
MAX = 11.086

grandma 's house for safe keeping . When
I folded up those paper towels for our nap kins ?
eat the y ummy ins ides . From that day on
to see what all the fuss was about . " What
to measure how big the maze was so she knew exactly
to measure how deep the hole was . So , he
to the direction of the leash . Little
it to me for safe keeping . He told me to
special powers that can heal injuries . With my powers ,
, learning all about how flour is made . <|endoftext|> Once
dry and cozy p aj amas and sn ugg led in
in any of the cup boards . The pan
proceeded to guess what the shield was for . Was it
on wearing his bright red boots , so his mum had
you carry a l ily pad too ?" The
put on your p aj amas . Then we can read
no matter how big the threat seems , it is always
a few days when the renovations were done . Everyone could
stopped . The colorful rain drops stopped falling from the sky
on a leash . The leash was a long , red

MIN ACTIVATIONS
MIN = -15.920

helped the old lady fix the fence .
" Maybe we can share the ball with him . Maybe
you . I will fix the pengu in . Then you
him if he would share the pie with her .
little boy to put away the hose and come inside .
why he had fixed her the screen . "
They discuss how to share the sausage . They
did not want to share the cricket . He wanted to
and Daddy helped Olive fix the toy . They were gentle
! You need to share the ball ." Buddy and Wh
Look . I have fixed the important roof . Let 's
. They decided to share the ball and play together .
smiled as he started pushing the yacht across the lake .
she would try to fix the train . She said they
going to help us fix the screen ." The
should be able to fix the dress ."
learned to be careful of the crabs . <|endoftext|> Ben and
let me help you carry the ac orn home ? I
Tim and his dad fixed the raft . They used different
impatient . We can share the mel on ." Tom said

INTERVAL 4.334 - 11.086
CONTAINS 4.609%

was loud and sm elly . There were machines that moved
cat wanted to go home . But the cat did not
table just like his friends . <|endoftext|> Once , there were
sand and ate y ummy snacks . They even saw a
upon a man with a bag full of things . The

INTERVAL -2.417 - 4.334
CONTAINS 77.521%

The cat and the kids played all day . They had
. " Let 's go back to the pond
She praised them and helped them measure the new radio .
in front of the tree and quickly the cat came down
Julia realised what was missing - a friend to share lunch

INTERVAL -9.168 - -2.417
CONTAINS 12.889%

rain got heavier , and the wind got stronger . The
the island . There were lots of new things to explore
blue frog . He was always very dirty and messy .
Joe decided to wander around the house . He went up
clearing . There were lots of tasty fruits and nuts scattered

INTERVAL -15.920 - -9.168
CONTAINS 4.981%

fly . The girl reminded the little bird that its wing
the house were scared of the noises . One
them . Lily had a pink doll with long hair and
help her search for a good hiding spot for her candy
through the sea and soon the oy ster was free !
Token grandma
Token ID49890
Feature activation+2.25e+00

Pos logit contributions
 ey+0.137
 pastor+0.133
Ta+0.131
 priest+0.131
 orchestra+0.131
Neg logit contributions
es-0.120
 they-0.117
 could-0.115
 with-0.109
bed-0.108
Token's
Token ID338
Feature activation-9.32e+00

Pos logit contributions
icked+0.036
es+0.032
ched+0.025
iled+0.024
ed+0.023
Neg logit contributions
 city-0.051
 opera-0.050
 airport-0.050
 hotel-0.050
 orchestra-0.049
Token house
Token ID2156
Feature activation+4.99e+00

Pos logit contributions
 orchestra+0.126
 thrill+0.124
Ta+0.124
 priest+0.123
 pastor+0.120
Neg logit contributions
es-0.176
 could-0.164
 they-0.164
 with-0.161
 in-0.161
Token for
Token ID329
Feature activation+3.35e-01

Pos logit contributions
icked+0.065
es+0.065
iled+0.044
 could+0.043
owed+0.043
Neg logit contributions
 city-0.122
 hotel-0.120
 opera-0.120
 talent-0.119
 orchestra-0.119
Token safe
Token ID3338
Feature activation+1.48e+00

Pos logit contributions
es+0.005
icked+0.005
 could+0.004
iled+0.004
 they+0.004
Neg logit contributions
 city-0.007
 hotel-0.007
 orchestra-0.007
 mirror-0.007
 opera-0.007
Tokenkeeping
Token ID19934
Feature activation+1.10e+01

Pos logit contributions
icked+0.024
es+0.023
iled+0.017
 could+0.017
owed+0.017
Neg logit contributions
 city-0.033
 hotel-0.032
 opera-0.032
 talent-0.032
 orchestra-0.031
Token.
Token ID13
Feature activation+4.57e+00

Pos logit contributions
icked+0.163
es+0.154
 Remy+0.117
iled+0.113
 could+0.108
Neg logit contributions
 city-0.248
 hotel-0.242
 opera-0.241
 talent-0.240
 mirror-0.239
Token
Token ID220
Feature activation+1.37e+00

Pos logit contributions
es+0.072
icked+0.072
owed+0.050
ched+0.050
iled+0.050
Neg logit contributions
 city-0.104
 hotel-0.100
 playground-0.099
 classroom-0.098
 mirror-0.098
Token
Token ID198
Feature activation+1.00e+00

Pos logit contributions
icked+0.022
es+0.022
iled+0.017
owed+0.017
ched+0.016
Neg logit contributions
 city-0.026
 park-0.025
 classroom-0.024
 mirror-0.024
 playground-0.024
Token
Token ID198
Feature activation+1.27e+00

Pos logit contributions
icked+0.017
es+0.016
owed+0.013
iled+0.012
 neither+0.012
Neg logit contributions
 city-0.021
 playground-0.020
 talent-0.020
 park-0.020
 hotel-0.020
TokenWhen
Token ID2215
Feature activation+2.15e+00

Pos logit contributions
icked+0.022
es+0.022
owed+0.017
 could+0.016
iled+0.016
Neg logit contributions
 city-0.026
 talent-0.025
 orchestra-0.025
 opera-0.025
 hotel-0.025
Token I
Token ID314
Feature activation+3.38e+00

Pos logit contributions
es+0.004
icked+0.004
iled+0.003
 could+0.003
owed+0.003
Neg logit contributions
 city-0.005
 talent-0.005
 opera-0.005
 hotel-0.005
 orchestra-0.005
Token folded
Token ID24650
Feature activation+3.53e+00

Pos logit contributions
es+0.062
icked+0.061
iled+0.047
owed+0.046
 Remy+0.045
Neg logit contributions
 talent-0.065
 city-0.065
 mirror-0.064
 hotel-0.064
 opera-0.064
Token up
Token ID510
Feature activation+3.16e+00

Pos logit contributions
es+0.050
icked+0.049
iled+0.035
 could+0.034
owed+0.034
Neg logit contributions
 city-0.083
 talent-0.082
 hotel-0.082
 opera-0.082
 orchestra-0.081
Token those
Token ID883
Feature activation-1.99e+00

Pos logit contributions
es+0.046
icked+0.044
 could+0.033
iled+0.032
owed+0.031
Neg logit contributions
 city-0.073
 hotel-0.071
 orchestra-0.071
 opera-0.071
 talent-0.071
Token paper
Token ID3348
Feature activation+5.04e+00

Pos logit contributions
 city+0.031
 orchestra+0.030
 hotel+0.030
 talent+0.029
 airport+0.029
Neg logit contributions
es-0.044
icked-0.041
 could-0.036
owed-0.034
ed-0.034
Token towels
Token ID38494
Feature activation+1.04e+01

Pos logit contributions
icked+0.072
es+0.070
iled+0.051
owed+0.049
 could+0.049
Neg logit contributions
 city-0.117
 talent-0.116
 opera-0.115
 hotel-0.115
 mirror-0.115
Token for
Token ID329
Feature activation+2.24e+00

Pos logit contributions
icked+0.160
es+0.153
 Remy+0.118
iled+0.115
owed+0.111
Neg logit contributions
 city-0.231
 mirror-0.230
 hotel-0.229
 airport-0.227
 opera-0.226
Token our
Token ID674
Feature activation-6.88e+00

Pos logit contributions
es+0.033
icked+0.032
 could+0.024
iled+0.023
ched+0.022
Neg logit contributions
 city-0.051
 hotel-0.050
 orchestra-0.050
 opera-0.050
 talent-0.050
Token nap
Token ID25422
Feature activation+3.87e+00

Pos logit contributions
 orchestra+0.083
 city+0.081
 opera+0.078
 airport+0.078
 mirror+0.077
Neg logit contributions
es-0.158
icked-0.151
 could-0.141
 they-0.137
 on-0.129
Tokenkins
Token ID5331
Feature activation+7.54e+00

Pos logit contributions
icked+0.064
es+0.060
iled+0.047
owed+0.046
ched+0.045
Neg logit contributions
 hotel-0.083
 city-0.082
 talent-0.082
 opera-0.080
 airport-0.079
Token?
Token ID30
Feature activation+6.66e+00

Pos logit contributions
icked+0.122
es+0.114
 Remy+0.088
iled+0.087
owed+0.086
Neg logit contributions
 talent-0.163
 city-0.162
 hotel-0.161
 mirror-0.159
 airport-0.159
Token eat
Token ID4483
Feature activation+4.78e+00

Pos logit contributions
es+0.010
icked+0.010
 could+0.008
iled+0.008
owed+0.008
Neg logit contributions
 city-0.009
 hotel-0.009
 orchestra-0.009
 talent-0.009
 opera-0.009
Token the
Token ID262
Feature activation-8.24e+00

Pos logit contributions
icked+0.077
es+0.074
iled+0.057
owed+0.057
ched+0.054
Neg logit contributions
 city-0.104
 hotel-0.102
 mirror-0.102
 opera-0.102
 talent-0.102
Token y
Token ID331
Feature activation-2.17e+00

Pos logit contributions
 orchestra+0.128
 city+0.117
 apartment+0.117
 talent+0.116
 passenger+0.116
Neg logit contributions
es-0.159
icked-0.148
 on-0.143
 could-0.141
 they-0.137
Tokenummy
Token ID13513
Feature activation-4.26e+00

Pos logit contributions
 city+0.038
 hotel+0.038
 talent+0.037
 opera+0.037
 mirror+0.037
Neg logit contributions
es-0.044
icked-0.044
 could-0.035
owed-0.034
ched-0.034
Token ins
Token ID1035
Feature activation+1.18e+00

Pos logit contributions
 city+0.061
 orchestra+0.060
 talent+0.059
 opera+0.059
 airport+0.059
Neg logit contributions
es-0.098
icked-0.096
 could-0.080
iled-0.079
owed-0.079
Tokenides
Token ID1460
Feature activation+1.03e+01

Pos logit contributions
es+0.019
icked+0.018
 could+0.014
iled+0.013
ched+0.013
Neg logit contributions
 city-0.026
 talent-0.025
 hotel-0.025
 orchestra-0.025
 opera-0.025
Token.
Token ID13
Feature activation+5.13e+00

Pos logit contributions
icked+0.157
es+0.143
 Remy+0.110
iled+0.109
ched+0.105
Neg logit contributions
 city-0.227
 mirror-0.223
 hotel-0.222
 opera-0.220
 airport-0.219
Token From
Token ID3574
Feature activation+6.66e-01

Pos logit contributions
icked+0.082
es+0.081
owed+0.060
ched+0.059
iled+0.058
Neg logit contributions
 city-0.115
 hotel-0.108
 former-0.108
 mirror-0.107
 orchestra-0.107
Token that
Token ID326
Feature activation-1.95e-01

Pos logit contributions
es+0.010
icked+0.009
iled+0.007
owed+0.007
uts+0.006
Neg logit contributions
 talent-0.016
 opera-0.016
 city-0.016
 airport-0.015
 mirror-0.015
Token day
Token ID1110
Feature activation+3.01e+00

Pos logit contributions
 city+0.002
 hotel+0.002
 talent+0.002
 opera+0.002
 mirror+0.002
Neg logit contributions
icked-0.005
es-0.005
owed-0.005
iled-0.005
ched-0.004
Token on
Token ID319
Feature activation+1.93e+00

Pos logit contributions
icked+0.042
owed+0.040
uts+0.039
es+0.038
cr+0.035
Neg logit contributions
 audience-0.059
 mirror-0.058
 opera-0.058
 city-0.057
 pride-0.056
Token to
Token ID284
Feature activation+8.18e-01

Pos logit contributions
es+0.018
icked+0.018
 could+0.012
iled+0.012
owed+0.012
Neg logit contributions
 city-0.035
 hotel-0.034
 talent-0.034
 opera-0.034
 orchestra-0.034
Token see
Token ID766
Feature activation+2.88e+00

Pos logit contributions
es+0.014
icked+0.014
 could+0.011
 they+0.011
iled+0.011
Neg logit contributions
 city-0.016
 talent-0.016
 hotel-0.016
 orchestra-0.016
 airport-0.016
Token what
Token ID644
Feature activation+1.60e+00

Pos logit contributions
es+0.041
icked+0.040
 could+0.029
iled+0.028
owed+0.028
Neg logit contributions
 city-0.067
 hotel-0.066
 talent-0.066
 orchestra-0.066
 opera-0.066
Token all
Token ID477
Feature activation+1.40e+00

Pos logit contributions
icked+0.025
es+0.023
iled+0.018
owed+0.018
bed+0.017
Neg logit contributions
 city-0.037
 talent-0.036
 opera-0.036
 hotel-0.036
 mirror-0.036
Token the
Token ID262
Feature activation-5.67e+00

Pos logit contributions
es+0.023
icked+0.023
iled+0.017
owed+0.017
 could+0.017
Neg logit contributions
 city-0.030
 opera-0.029
 hotel-0.029
 talent-0.029
 orchestra-0.029
Token fuss
Token ID34297
Feature activation+1.01e+01

Pos logit contributions
 orchestra+0.077
 hotel+0.075
 airport+0.073
 passenger+0.070
 opera+0.070
Neg logit contributions
es-0.123
 could-0.110
icked-0.109
 they-0.105
 on-0.105
Token was
Token ID373
Feature activation-1.72e-02

Pos logit contributions
icked+0.151
es+0.145
 Remy+0.123
iled+0.116
ched+0.106
Neg logit contributions
 city-0.216
 opera-0.213
 mirror-0.212
 hotel-0.210
 orchestra-0.209
Token about
Token ID546
Feature activation+2.17e+00

Pos logit contributions
 city+0.000
 opera+0.000
 talent+0.000
 orchestra+0.000
 hotel+0.000
Neg logit contributions
es-0.000
icked-0.000
iled-0.000
owed-0.000
 could-0.000
Token.
Token ID13
Feature activation+5.19e+00

Pos logit contributions
icked+0.029
es+0.028
iled+0.019
owed+0.019
ched+0.019
Neg logit contributions
 city-0.053
 talent-0.053
 opera-0.052
 hotel-0.052
 orchestra-0.052
Token "
Token ID366
Feature activation+1.15e+00

Pos logit contributions
icked+0.080
es+0.079
owed+0.059
ched+0.058
iled+0.056
Neg logit contributions
 city-0.118
 mirror-0.112
 classroom-0.112
 orchestra-0.111
 hotel-0.111
TokenWhat
Token ID2061
Feature activation+3.01e+00

Pos logit contributions
es+0.017
icked+0.016
 could+0.013
iled+0.013
ched+0.012
Neg logit contributions
 hotel-0.026
 city-0.026
 airport-0.025
 talent-0.025
 opera-0.025
Token to
Token ID284
Feature activation+1.98e-01

Pos logit contributions
 city+0.048
 hotel+0.047
 talent+0.047
 orchestra+0.047
 opera+0.047
Neg logit contributions
es-0.054
icked-0.053
 could-0.043
iled-0.042
owed-0.042
Token measure
Token ID3953
Feature activation+4.85e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.003
 they+0.003
iled+0.003
Neg logit contributions
 city-0.004
 hotel-0.004
 orchestra-0.004
 talent-0.004
 opera-0.004
Token how
Token ID703
Feature activation+2.31e+00

Pos logit contributions
icked+0.079
es+0.078
iled+0.058
 could+0.058
owed+0.057
Neg logit contributions
 city-0.104
 opera-0.101
 talent-0.101
 hotel-0.101
 orchestra-0.100
Token big
Token ID1263
Feature activation+2.56e+00

Pos logit contributions
es+0.044
icked+0.044
owed+0.034
iled+0.034
ched+0.033
Neg logit contributions
 city-0.044
 talent-0.043
 mirror-0.042
 hotel-0.042
 opera-0.042
Token the
Token ID262
Feature activation-7.84e+00

Pos logit contributions
icked+0.041
es+0.039
owed+0.030
iled+0.029
ched+0.028
Neg logit contributions
 city-0.058
 opera-0.056
 mirror-0.056
 hotel-0.055
 talent-0.055
Token maze
Token ID31237
Feature activation+1.00e+01

Pos logit contributions
 orchestra+0.098
 talent+0.093
 hotel+0.092
 opera+0.090
 airport+0.090
Neg logit contributions
es-0.177
icked-0.162
 could-0.158
 they-0.153
 on-0.150
Token was
Token ID373
Feature activation+1.31e+00

Pos logit contributions
icked+0.157
es+0.146
iled+0.117
 Remy+0.113
owed+0.109
Neg logit contributions
 city-0.225
 opera-0.219
 airport-0.216
 hotel-0.216
 talent-0.213
Token so
Token ID523
Feature activation-8.86e-01

Pos logit contributions
icked+0.019
es+0.019
iled+0.013
owed+0.013
ched+0.013
Neg logit contributions
 city-0.031
 opera-0.030
 talent-0.030
 hotel-0.030
 airport-0.030
Token she
Token ID673
Feature activation+2.74e+00

Pos logit contributions
 city+0.011
 hotel+0.010
 orchestra+0.010
 opera+0.010
 talent+0.010
Neg logit contributions
es-0.023
icked-0.023
 could-0.019
iled-0.019
owed-0.019
Token knew
Token ID2993
Feature activation+4.07e+00

Pos logit contributions
icked+0.033
es+0.033
iled+0.022
 Remy+0.022
owed+0.021
Neg logit contributions
 city-0.067
 talent-0.067
 mirror-0.067
 opera-0.066
 orchestra-0.066
Token exactly
Token ID3446
Feature activation+1.27e+00

Pos logit contributions
icked+0.053
es+0.052
iled+0.038
owed+0.035
ched+0.035
Neg logit contributions
 city-0.099
 talent-0.098
 opera-0.098
 mirror-0.097
 airport-0.096
Token to
Token ID284
Feature activation-3.57e-01

Pos logit contributions
es+0.050
icked+0.050
iled+0.037
owed+0.036
ched+0.034
Neg logit contributions
 talent-0.077
 opera-0.076
 city-0.076
 hotel-0.074
 mirror-0.074
Token measure
Token ID3953
Feature activation+4.25e+00

Pos logit contributions
 city+0.006
 hotel+0.006
 orchestra+0.006
 talent+0.006
 opera+0.006
Neg logit contributions
es-0.007
icked-0.007
 could-0.005
iled-0.005
owed-0.005
Token how
Token ID703
Feature activation+2.18e+00

Pos logit contributions
icked+0.069
es+0.069
iled+0.052
owed+0.051
ched+0.051
Neg logit contributions
 city-0.091
 opera-0.089
 talent-0.088
 hotel-0.088
 mirror-0.088
Token deep
Token ID2769
Feature activation+4.58e+00

Pos logit contributions
es+0.044
icked+0.044
owed+0.035
iled+0.034
ched+0.033
Neg logit contributions
 city-0.039
 talent-0.038
 hotel-0.037
 opera-0.037
 mirror-0.037
Token the
Token ID262
Feature activation-7.49e+00

Pos logit contributions
icked+0.076
es+0.070
owed+0.056
iled+0.053
o+0.053
Neg logit contributions
 city-0.101
 opera-0.098
 mirror-0.097
 hotel-0.097
 talent-0.096
Token hole
Token ID7604
Feature activation+9.94e+00

Pos logit contributions
 orchestra+0.091
 talent+0.088
 former+0.085
 opera+0.085
 hotel+0.084
Neg logit contributions
es-0.168
icked-0.156
 could-0.155
 they-0.147
 on-0.146
Token was
Token ID373
Feature activation+1.20e+00

Pos logit contributions
icked+0.142
es+0.137
iled+0.109
 Remy+0.101
o+0.097
Neg logit contributions
 hotel-0.219
 city-0.219
 opera-0.218
 mirror-0.218
 airport-0.212
Token.
Token ID13
Feature activation+5.34e+00

Pos logit contributions
icked+0.019
es+0.019
iled+0.013
owed+0.013
o+0.013
Neg logit contributions
 city-0.027
 opera-0.026
 hotel-0.026
 talent-0.026
 airport-0.026
Token So
Token ID1406
Feature activation+3.44e+00

Pos logit contributions
icked+0.089
es+0.086
owed+0.067
ched+0.064
iled+0.062
Neg logit contributions
 city-0.116
 classroom-0.110
 mirror-0.110
 playground-0.109
 park-0.108
Token,
Token ID11
Feature activation+3.78e+00

Pos logit contributions
icked+0.053
es+0.048
ched+0.038
iled+0.037
owed+0.036
Neg logit contributions
 city-0.080
 hotel-0.077
 opera-0.076
 mirror-0.076
 airport-0.075
Token he
Token ID339
Feature activation+2.17e+00

Pos logit contributions
icked+0.054
es+0.049
owed+0.036
iled+0.035
ched+0.034
Neg logit contributions
 city-0.092
 opera-0.090
 hotel-0.089
 talent-0.089
 orchestra-0.088
Token to
Token ID284
Feature activation+1.98e-01

Pos logit contributions
icked+0.046
es+0.042
iled+0.035
ched+0.033
hot+0.033
Neg logit contributions
 talent-0.058
 city-0.058
 opera-0.057
 hotel-0.057
 mirror-0.056
Token the
Token ID262
Feature activation-6.63e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.003
 they+0.003
iled+0.003
Neg logit contributions
 city-0.004
 hotel-0.004
 orchestra-0.004
 talent-0.004
 opera-0.004
Token direction
Token ID4571
Feature activation+5.50e+00

Pos logit contributions
 orchestra+0.073
 tant+0.071
 talent+0.066
 former+0.064
 hotel+0.059
Neg logit contributions
es-0.146
 on-0.137
icked-0.137
 could-0.136
 they-0.135
Token of
Token ID286
Feature activation-1.14e+00

Pos logit contributions
icked+0.079
es+0.076
iled+0.053
ched+0.052
owed+0.052
Neg logit contributions
 city-0.131
 opera-0.129
 hotel-0.127
 mirror-0.127
 talent-0.127
Token the
Token ID262
Feature activation-9.06e+00

Pos logit contributions
 hotel+0.025
 city+0.025
 orchestra+0.025
 talent+0.024
 former+0.024
Neg logit contributions
es-0.017
icked-0.016
 they-0.013
 could-0.012
 Remy-0.012
Token leash
Token ID40529
Feature activation+9.91e+00

Pos logit contributions
 orchestra+0.099
 former+0.098
 talent+0.095
 tant+0.093
 hotel+0.090
Neg logit contributions
es-0.197
 on-0.184
 could-0.181
 they-0.177
 in-0.175
Token.
Token ID13
Feature activation+5.32e+00

Pos logit contributions
icked+0.146
es+0.138
iled+0.105
 Remy+0.099
owed+0.097
Neg logit contributions
 city-0.227
 airport-0.223
 opera-0.220
 hotel-0.219
 orchestra-0.217
Token
Token ID220
Feature activation+2.19e+00

Pos logit contributions
icked+0.083
es+0.082
owed+0.061
ched+0.059
iled+0.058
Neg logit contributions
 city-0.121
 orchestra-0.114
 hotel-0.114
 playground-0.114
 classroom-0.114
Token
Token ID198
Feature activation+1.84e+00

Pos logit contributions
icked+0.037
es+0.035
iled+0.028
owed+0.027
bed+0.027
Neg logit contributions
 city-0.042
 park-0.040
 mirror-0.039
 hotel-0.039
 classroom-0.039
Token
Token ID198
Feature activation+1.80e+00

Pos logit contributions
icked+0.025
es+0.024
owed+0.017
iled+0.016
 could+0.016
Neg logit contributions
 city-0.045
 talent-0.044
 hotel-0.044
 orchestra-0.044
 opera-0.043
TokenLittle
Token ID22253
Feature activation-2.42e-01

Pos logit contributions
es+0.031
icked+0.031
 could+0.023
iled+0.023
owed+0.023
Neg logit contributions
 city-0.037
 talent-0.036
 hotel-0.036
 opera-0.036
 orchestra-0.036
Token it
Token ID340
Feature activation+1.32e+00

Pos logit contributions
es+0.013
icked+0.013
 could+0.008
ed+0.007
ched+0.007
Neg logit contributions
 city-0.035
 orchestra-0.035
 hotel-0.034
 talent-0.034
 airport-0.034
Token to
Token ID284
Feature activation+1.03e+00

Pos logit contributions
icked+0.014
es+0.014
 could+0.009
o+0.008
ed+0.008
Neg logit contributions
 city-0.036
 orchestra-0.036
 hotel-0.035
 airport-0.035
 opera-0.034
Token me
Token ID502
Feature activation+2.15e+00

Pos logit contributions
es+0.015
icked+0.014
 they+0.011
 could+0.011
 Remy+0.010
Neg logit contributions
 city-0.023
 talent-0.023
 hotel-0.023
 orchestra-0.023
 tant-0.022
Token for
Token ID329
Feature activation+1.46e-02

Pos logit contributions
icked+0.029
es+0.029
iled+0.020
ched+0.020
 could+0.020
Neg logit contributions
 city-0.051
 opera-0.051
 hotel-0.050
 talent-0.050
 orchestra-0.050
Token safe
Token ID3338
Feature activation+2.77e-01

Pos logit contributions
es+0.000
icked+0.000
 they+0.000
 could+0.000
 Remy+0.000
Neg logit contributions
 city-0.000
 hotel-0.000
 orchestra-0.000
 airport-0.000
 mirror-0.000
Tokenkeeping
Token ID19934
Feature activation+9.88e+00

Pos logit contributions
es+0.005
icked+0.005
 could+0.003
iled+0.003
owed+0.003
Neg logit contributions
 city-0.006
 talent-0.006
 orchestra-0.006
 hotel-0.006
 opera-0.006
Token.
Token ID13
Feature activation+3.83e+00

Pos logit contributions
icked+0.142
es+0.136
 Remy+0.100
iled+0.099
 could+0.097
Neg logit contributions
 city-0.225
 opera-0.222
 hotel-0.221
 talent-0.221
 mirror-0.219
Token He
Token ID679
Feature activation+1.70e+00

Pos logit contributions
es+0.055
icked+0.055
owed+0.043
iled+0.040
ched+0.039
Neg logit contributions
 city-0.087
 mirror-0.085
 museum-0.084
 orchestra-0.084
 opera-0.084
Token told
Token ID1297
Feature activation+1.90e+00

Pos logit contributions
es+0.025
icked+0.025
 could+0.019
iled+0.018
owed+0.018
Neg logit contributions
 city-0.039
 hotel-0.038
 orchestra-0.038
 opera-0.038
 talent-0.038
Token me
Token ID502
Feature activation+2.43e+00

Pos logit contributions
es+0.028
icked+0.027
 they+0.021
 could+0.020
 Remy+0.019
Neg logit contributions
 city-0.044
 talent-0.043
 orchestra-0.043
 former-0.042
 hotel-0.042
Token to
Token ID284
Feature activation-5.25e-01

Pos logit contributions
icked+0.032
es+0.031
iled+0.022
ched+0.021
owed+0.021
Neg logit contributions
 city-0.059
 opera-0.058
 hotel-0.058
 talent-0.058
 mirror-0.057
Token special
Token ID2041
Feature activation-4.60e+00

Pos logit contributions
es+0.009
icked+0.009
iled+0.007
owed+0.007
 could+0.007
Neg logit contributions
 city-0.011
 talent-0.010
 opera-0.010
 hotel-0.010
 orchestra-0.010
Token powers
Token ID5635
Feature activation+4.91e+00

Pos logit contributions
 city+0.073
 hotel+0.071
 orchestra+0.070
 airport+0.070
 opera+0.067
Neg logit contributions
es-0.101
icked-0.098
 could-0.083
ched-0.079
owed-0.079
Token that
Token ID326
Feature activation+1.49e+00

Pos logit contributions
icked+0.065
es+0.063
iled+0.044
owed+0.043
ched+0.042
Neg logit contributions
 talent-0.118
 city-0.118
 opera-0.117
 hotel-0.116
 mirror-0.115
Token can
Token ID460
Feature activation+1.78e+00

Pos logit contributions
es+0.019
icked+0.019
iled+0.013
owed+0.013
 could+0.012
Neg logit contributions
 city-0.037
 opera-0.036
 talent-0.036
 hotel-0.036
 orchestra-0.036
Token heal
Token ID12035
Feature activation+3.45e+00

Pos logit contributions
icked+0.035
es+0.035
iled+0.028
owed+0.027
o+0.027
Neg logit contributions
 city-0.031
 opera-0.031
 talent-0.031
 hotel-0.030
 orchestra-0.030
Token injuries
Token ID6821
Feature activation+9.88e+00

Pos logit contributions
icked+0.058
es+0.053
iled+0.042
owed+0.041
ched+0.040
Neg logit contributions
 opera-0.073
 city-0.073
 talent-0.072
 orchestra-0.071
 hotel-0.070
Token.
Token ID13
Feature activation+5.06e+00

Pos logit contributions
icked+0.170
es+0.149
 Remy+0.129
iled+0.125
 Carl+0.120
Neg logit contributions
 city-0.200
 opera-0.199
 airport-0.196
 orchestra-0.195
 mirror-0.191
Token With
Token ID2080
Feature activation+6.38e-01

Pos logit contributions
es+0.071
icked+0.070
owed+0.049
iled+0.047
ched+0.047
Neg logit contributions
 city-0.119
 talent-0.117
 orchestra-0.117
 opera-0.116
 mirror-0.116
Token my
Token ID616
Feature activation-8.58e+00

Pos logit contributions
es+0.011
icked+0.011
 could+0.008
iled+0.008
ched+0.008
Neg logit contributions
 city-0.013
 hotel-0.013
 orchestra-0.013
 airport-0.013
 opera-0.013
Token powers
Token ID5635
Feature activation+4.61e+00

Pos logit contributions
 orchestra+0.146
 city+0.144
 opera+0.141
 airport+0.140
 hotel+0.137
Neg logit contributions
es-0.174
icked-0.174
 could-0.155
 on-0.141
 Remy-0.141
Token,
Token ID11
Feature activation+3.11e+00

Pos logit contributions
icked+0.060
es+0.059
iled+0.041
 Remy+0.039
owed+0.039
Neg logit contributions
 talent-0.110
 city-0.110
 opera-0.108
 mirror-0.108
 hotel-0.107
Token,
Token ID11
Feature activation+1.92e+00

Pos logit contributions
icked+0.096
es+0.094
iled+0.067
owed+0.064
 could+0.063
Neg logit contributions
 city-0.166
 talent-0.164
 opera-0.163
 hotel-0.162
 orchestra-0.162
Token learning
Token ID4673
Feature activation+2.67e+00

Pos logit contributions
icked+0.027
es+0.027
iled+0.019
owed+0.019
ched+0.018
Neg logit contributions
 city-0.045
 opera-0.044
 talent-0.044
 orchestra-0.044
 hotel-0.044
Token all
Token ID477
Feature activation+3.55e-01

Pos logit contributions
es+0.042
icked+0.039
 could+0.031
 they+0.029
owed+0.028
Neg logit contributions
 city-0.060
 hotel-0.060
 orchestra-0.059
 airport-0.058
 mirror-0.057
Token about
Token ID546
Feature activation+1.81e+00

Pos logit contributions
es+0.006
icked+0.006
 could+0.005
ched+0.004
 they+0.004
Neg logit contributions
 city-0.007
 hotel-0.007
 orchestra-0.007
 airport-0.007
 mirror-0.007
Token how
Token ID703
Feature activation+1.50e+00

Pos logit contributions
es+0.032
icked+0.028
 they+0.025
 could+0.025
 Remy+0.023
Neg logit contributions
 hotel-0.036
 orchestra-0.036
 airport-0.035
 city-0.035
 mirror-0.034
Token flour
Token ID10601
Feature activation+9.87e+00

Pos logit contributions
icked+0.029
es+0.029
owed+0.022
iled+0.022
ched+0.022
Neg logit contributions
 city-0.028
 talent-0.027
 opera-0.027
 hotel-0.027
 orchestra-0.027
Token is
Token ID318
Feature activation+1.40e+00

Pos logit contributions
icked+0.156
es+0.141
iled+0.115
 Remy+0.111
owed+0.110
Neg logit contributions
 mirror-0.208
 city-0.206
 hotel-0.203
 opera-0.202
 talent-0.200
Token made
Token ID925
Feature activation+5.00e+00

Pos logit contributions
es+0.024
icked+0.024
iled+0.018
owed+0.018
ched+0.017
Neg logit contributions
 city-0.028
 talent-0.028
 opera-0.028
 hotel-0.027
 airport-0.027
Token.
Token ID13
Feature activation+5.59e+00

Pos logit contributions
icked+0.075
es+0.073
iled+0.053
owed+0.052
o+0.050
Neg logit contributions
 city-0.114
 opera-0.112
 talent-0.111
 hotel-0.111
 airport-0.110
Token<|endoftext|>
Token ID50256
Feature activation+3.42e+00

Pos logit contributions
icked+0.086
es+0.085
owed+0.064
ched+0.062
iled+0.061
Neg logit contributions
 city-0.126
 orchestra-0.120
 mirror-0.119
 talent-0.119
 museum-0.119
TokenOnce
Token ID7454
Feature activation-1.57e+00

Pos logit contributions
icked+0.063
es+0.057
owed+0.051
 neither+0.049
ppers+0.048
Neg logit contributions
 city-0.069
 park-0.067
 museum-0.066
 hotel-0.065
 mirror-0.065
Token dry
Token ID5894
Feature activation+5.00e-01

Pos logit contributions
es+0.045
icked+0.042
 could+0.033
 they+0.032
iled+0.031
Neg logit contributions
 orchestra-0.064
 city-0.064
 talent-0.063
 hotel-0.063
 opera-0.063
Token and
Token ID290
Feature activation+8.32e-01

Pos logit contributions
es+0.013
icked+0.013
 could+0.011
iled+0.011
owed+0.011
Neg logit contributions
 city-0.006
 hotel-0.006
 talent-0.006
 opera-0.006
 orchestra-0.006
Token cozy
Token ID37438
Feature activation-4.49e-01

Pos logit contributions
es+0.019
icked+0.019
iled+0.015
 could+0.015
owed+0.015
Neg logit contributions
 city-0.013
 hotel-0.012
 talent-0.012
 opera-0.012
 orchestra-0.012
Token p
Token ID279
Feature activation+1.13e+00

Pos logit contributions
 city+0.007
 hotel+0.007
 talent+0.007
 opera+0.007
 orchestra+0.007
Neg logit contributions
es-0.010
icked-0.010
iled-0.008
 could-0.008
owed-0.008
Tokenaj
Token ID1228
Feature activation+5.45e+00

Pos logit contributions
es+0.026
icked+0.026
iled+0.021
owed+0.021
 could+0.021
Neg logit contributions
 city-0.017
 opera-0.016
 orchestra-0.016
 talent-0.016
 hotel-0.016
Tokenamas
Token ID17485
Feature activation+9.87e+00

Pos logit contributions
icked+0.105
es+0.097
iled+0.083
 Remy+0.081
 could+0.080
Neg logit contributions
 city-0.096
 mirror-0.096
 airport-0.094
 orchestra-0.094
 hotel-0.094
Token and
Token ID290
Feature activation+1.55e+00

Pos logit contributions
icked+0.148
es+0.141
 Remy+0.110
iled+0.104
 Carl+0.101
Neg logit contributions
 city-0.220
 hotel-0.218
 airport-0.218
 opera-0.217
 mirror-0.215
Token sn
Token ID3013
Feature activation+2.58e+00

Pos logit contributions
es+0.031
icked+0.031
iled+0.024
owed+0.024
 could+0.024
Neg logit contributions
 city-0.027
 hotel-0.027
 opera-0.027
 talent-0.027
 orchestra-0.026
Tokenugg
Token ID6837
Feature activation+2.65e+00

Pos logit contributions
es+0.052
icked+0.051
 could+0.041
 Remy+0.040
iled+0.039
Neg logit contributions
 hotel-0.044
 mirror-0.044
 city-0.044
 opera-0.043
 former-0.043
Tokenled
Token ID992
Feature activation+3.69e+00

Pos logit contributions
icked+0.029
es+0.027
iled+0.019
o+0.017
 Remy+0.017
Neg logit contributions
 hotel-0.070
 opera-0.070
 mirror-0.070
 city-0.069
 talent-0.069
Token in
Token ID287
Feature activation+1.08e+00

Pos logit contributions
es+0.047
icked+0.047
iled+0.032
 could+0.032
owed+0.031
Neg logit contributions
 city-0.091
 hotel-0.090
 opera-0.090
 talent-0.090
 orchestra-0.089
Token in
Token ID287
Feature activation+1.90e+00

Pos logit contributions
es+0.028
icked+0.028
iled+0.019
 could+0.019
owed+0.019
Neg logit contributions
 city-0.049
 hotel-0.048
 talent-0.048
 opera-0.048
 orchestra-0.048
Token any
Token ID597
Feature activation-3.82e+00

Pos logit contributions
es+0.031
icked+0.029
 could+0.023
iled+0.022
 they+0.021
Neg logit contributions
 city-0.040
 orchestra-0.040
 hotel-0.040
 opera-0.039
 airport-0.039
Token of
Token ID286
Feature activation-1.03e-01

Pos logit contributions
 orchestra+0.050
 city+0.048
 airport+0.048
 priest+0.047
 opera+0.047
Neg logit contributions
es-0.087
icked-0.080
 could-0.076
 they-0.072
 on-0.072
Token the
Token ID262
Feature activation-7.98e+00

Pos logit contributions
 city+0.002
 hotel+0.002
 orchestra+0.002
 airport+0.002
 opera+0.002
Neg logit contributions
es-0.002
icked-0.002
 they-0.001
 could-0.001
 Remy-0.001
Token cup
Token ID6508
Feature activation+6.15e+00

Pos logit contributions
 orchestra+0.089
 tant+0.083
 priest+0.082
 talent+0.078
 opera+0.077
Neg logit contributions
es-0.172
 on-0.165
 could-0.158
 in-0.157
icked-0.156
Tokenboards
Token ID12821
Feature activation+9.67e+00

Pos logit contributions
icked+0.109
es+0.103
iled+0.084
 Remy+0.082
owed+0.080
Neg logit contributions
 city-0.122
 hotel-0.121
 mirror-0.121
 opera-0.120
 airport-0.119
Token.
Token ID13
Feature activation+5.73e+00

Pos logit contributions
icked+0.151
es+0.142
 Remy+0.109
iled+0.109
owed+0.102
Neg logit contributions
 city-0.214
 hotel-0.211
 airport-0.210
 opera-0.209
 mirror-0.205
Token
Token ID198
Feature activation+1.92e+00

Pos logit contributions
icked+0.087
es+0.085
owed+0.064
ched+0.061
iled+0.060
Neg logit contributions
 city-0.132
 orchestra-0.125
 mirror-0.125
 classroom-0.125
 playground-0.125
Token
Token ID198
Feature activation+1.77e+00

Pos logit contributions
icked+0.026
es+0.025
owed+0.017
iled+0.017
 could+0.017
Neg logit contributions
 city-0.047
 talent-0.046
 hotel-0.046
 opera-0.046
 orchestra-0.046
TokenThe
Token ID464
Feature activation-9.96e+00

Pos logit contributions
es+0.029
icked+0.029
 could+0.022
iled+0.022
owed+0.021
Neg logit contributions
 city-0.037
 hotel-0.037
 talent-0.036
 opera-0.036
 orchestra-0.036
Token pan
Token ID3425
Feature activation+6.57e+00

Pos logit contributions
 orchestra+0.132
 tant+0.124
 former+0.124
 talent+0.123
 cockpit+0.122
Neg logit contributions
es-0.186
 could-0.185
 they-0.185
 on-0.177
icked-0.177
Token proceeded
Token ID18814
Feature activation+4.50e+00

Pos logit contributions
icked+0.032
es+0.031
iled+0.024
owed+0.023
ched+0.022
Neg logit contributions
 city-0.049
 opera-0.048
 orchestra-0.048
 hotel-0.048
 airport-0.048
Token to
Token ID284
Feature activation+2.02e-01

Pos logit contributions
es+0.056
icked+0.056
iled+0.038
owed+0.037
ched+0.036
Neg logit contributions
 city-0.112
 talent-0.111
 opera-0.110
 airport-0.110
 mirror-0.109
Token guess
Token ID4724
Feature activation+3.19e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.003
 they+0.003
 Remy+0.003
Neg logit contributions
 orchestra-0.004
 city-0.004
 hotel-0.004
 tant-0.004
 talent-0.004
Token what
Token ID644
Feature activation+1.13e+00

Pos logit contributions
icked+0.045
es+0.044
iled+0.033
ched+0.031
 could+0.031
Neg logit contributions
 city-0.074
 opera-0.074
 talent-0.073
 hotel-0.073
 orchestra-0.072
Token the
Token ID262
Feature activation-8.82e+00

Pos logit contributions
icked+0.016
es+0.015
iled+0.011
owed+0.011
ched+0.011
Neg logit contributions
 city-0.027
 talent-0.027
 opera-0.027
 hotel-0.026
 mirror-0.026
Token shield
Token ID7614
Feature activation+9.67e+00

Pos logit contributions
 apartment+0.114
 tant+0.108
 hotel+0.108
 orchestra+0.104
 former+0.102
Neg logit contributions
es-0.196
 could-0.184
 on-0.175
icked-0.175
 he-0.168
Token was
Token ID373
Feature activation+4.86e-01

Pos logit contributions
icked+0.145
es+0.139
iled+0.107
 Remy+0.100
owed+0.099
Neg logit contributions
 city-0.216
 opera-0.211
 hotel-0.211
 talent-0.211
 airport-0.210
Token for
Token ID329
Feature activation+3.06e+00

Pos logit contributions
icked+0.007
es+0.007
iled+0.005
owed+0.005
ched+0.005
Neg logit contributions
 city-0.011
 opera-0.011
 talent-0.011
 hotel-0.011
 orchestra-0.011
Token.
Token ID13
Feature activation+4.33e+00

Pos logit contributions
icked+0.048
es+0.043
ched+0.035
iled+0.034
o+0.033
Neg logit contributions
 opera-0.067
 talent-0.066
 city-0.065
 orchestra-0.063
 museum-0.063
Token Was
Token ID8920
Feature activation+1.62e+00

Pos logit contributions
es+0.066
icked+0.065
owed+0.049
iled+0.049
ched+0.048
Neg logit contributions
 city-0.098
 playground-0.093
 mirror-0.092
 zoo-0.092
 classroom-0.092
Token it
Token ID340
Feature activation+5.98e-01

Pos logit contributions
icked+0.021
es+0.020
iled+0.015
owed+0.014
o+0.013
Neg logit contributions
 opera-0.039
 city-0.039
 talent-0.039
 airport-0.038
 hotel-0.038
Token on
Token ID319
Feature activation+2.15e+00

Pos logit contributions
es+0.005
icked+0.005
iled+0.003
owed+0.003
ched+0.003
Neg logit contributions
 city-0.010
 talent-0.009
 opera-0.009
 hotel-0.009
 mirror-0.009
Token wearing
Token ID5762
Feature activation+3.47e+00

Pos logit contributions
es+0.031
icked+0.030
 could+0.022
iled+0.021
 they+0.021
Neg logit contributions
 city-0.049
 hotel-0.049
 orchestra-0.049
 talent-0.049
 opera-0.048
Token his
Token ID465
Feature activation-5.32e+00

Pos logit contributions
es+0.060
icked+0.057
 could+0.046
 they+0.043
ed+0.043
Neg logit contributions
 city-0.071
 hotel-0.071
 airport-0.069
 orchestra-0.069
 former-0.068
Token bright
Token ID6016
Feature activation-5.86e+00

Pos logit contributions
 city+0.069
 orchestra+0.068
 hotel+0.068
 former+0.066
 airport+0.066
Neg logit contributions
es-0.121
icked-0.116
 could-0.106
 on-0.105
o-0.101
Token red
Token ID2266
Feature activation-3.62e+00

Pos logit contributions
 former+0.075
 city+0.073
 hotel+0.072
 priest+0.071
 orchestra+0.070
Neg logit contributions
es-0.133
icked-0.130
 on-0.115
 could-0.114
owed-0.111
Token boots
Token ID14412
Feature activation+9.66e+00

Pos logit contributions
 orchestra+0.049
 city+0.049
 former+0.048
 hotel+0.048
 opera+0.047
Neg logit contributions
es-0.083
icked-0.081
 could-0.070
bed-0.069
 on-0.068
Token,
Token ID11
Feature activation+2.53e+00

Pos logit contributions
icked+0.143
es+0.138
 Remy+0.103
iled+0.103
owed+0.096
Neg logit contributions
 city-0.218
 airport-0.214
 opera-0.214
 hotel-0.211
 orchestra-0.211
Token so
Token ID523
Feature activation+2.83e+00

Pos logit contributions
es+0.046
icked+0.045
 could+0.036
iled+0.035
owed+0.034
Neg logit contributions
 city-0.049
 hotel-0.048
 talent-0.048
 orchestra-0.048
 opera-0.048
Token his
Token ID465
Feature activation-5.11e+00

Pos logit contributions
icked+0.032
es+0.032
iled+0.021
owed+0.021
ched+0.020
Neg logit contributions
 city-0.075
 hotel-0.073
 opera-0.073
 talent-0.072
 orchestra-0.072
Token mum
Token ID25682
Feature activation+4.86e+00

Pos logit contributions
 city+0.088
 orchestra+0.087
 hotel+0.084
 opera+0.084
 museum+0.084
Neg logit contributions
icked-0.090
es-0.089
 could-0.084
 they-0.078
bed-0.077
Token had
Token ID550
Feature activation+9.13e-01

Pos logit contributions
icked+0.077
es+0.074
iled+0.056
 Remy+0.056
uts+0.056
Neg logit contributions
 mirror-0.102
 opera-0.100
 city-0.100
 orchestra-0.100
 airport-0.100
Token you
Token ID345
Feature activation+3.61e+00

Pos logit contributions
es+0.004
icked+0.004
iled+0.003
 could+0.003
owed+0.003
Neg logit contributions
 city-0.007
 talent-0.006
 opera-0.006
 hotel-0.006
 orchestra-0.006
Token carry
Token ID3283
Feature activation+3.75e+00

Pos logit contributions
es+0.059
icked+0.058
iled+0.044
owed+0.042
o+0.041
Neg logit contributions
 city-0.076
 talent-0.075
 opera-0.075
 hotel-0.075
 orchestra-0.074
Token a
Token ID257
Feature activation-7.12e+00

Pos logit contributions
es+0.063
icked+0.063
 could+0.047
iled+0.047
owed+0.047
Neg logit contributions
 city-0.078
 hotel-0.076
 talent-0.076
 opera-0.076
 orchestra-0.076
Token l
Token ID300
Feature activation+1.00e+00

Pos logit contributions
 orchestra+0.116
 opera+0.110
 city+0.106
 former+0.102
 airport+0.101
Neg logit contributions
 on-0.121
 could-0.121
 in-0.114
 they-0.111
es-0.110
Tokenily
Token ID813
Feature activation+3.92e+00

Pos logit contributions
es+0.016
icked+0.015
iled+0.012
 could+0.012
 they+0.011
Neg logit contributions
 talent-0.021
 city-0.021
 opera-0.021
 mirror-0.021
 hotel-0.021
Token pad
Token ID14841
Feature activation+9.59e+00

Pos logit contributions
icked+0.070
es+0.067
iled+0.052
owed+0.051
 Remy+0.050
Neg logit contributions
 city-0.077
 opera-0.076
 talent-0.076
 mirror-0.075
 airport-0.075
Token too
Token ID1165
Feature activation+2.26e+00

Pos logit contributions
icked+0.128
es+0.127
 Remy+0.105
iled+0.098
 Carl+0.091
Neg logit contributions
 mirror-0.203
 talent-0.201
 bus-0.199
 city-0.199
 playground-0.198
Token?"
Token ID1701
Feature activation+4.31e+00

Pos logit contributions
es+0.042
icked+0.041
iled+0.031
owed+0.031
 could+0.031
Neg logit contributions
 city-0.044
 hotel-0.043
 talent-0.043
 opera-0.043
 orchestra-0.042
Token
Token ID220
Feature activation+2.05e+00

Pos logit contributions
icked+0.070
es+0.069
owed+0.051
iled+0.050
ched+0.049
Neg logit contributions
 mirror-0.094
 talent-0.092
 city-0.092
 opera-0.090
 orchestra-0.090
Token
Token ID198
Feature activation+2.42e+00

Pos logit contributions
icked+0.033
es+0.032
iled+0.025
owed+0.025
 could+0.023
Neg logit contributions
 city-0.041
 mirror-0.041
 talent-0.040
 park-0.039
 classroom-0.039
TokenThe
Token ID464
Feature activation-9.50e+00

Pos logit contributions
icked+0.033
es+0.032
owed+0.022
iled+0.022
 could+0.022
Neg logit contributions
 city-0.058
 talent-0.057
 mirror-0.057
 opera-0.057
 orchestra-0.057
Token put
Token ID1234
Feature activation+2.82e+00

Pos logit contributions
 city+0.004
 hotel+0.004
 talent+0.004
 mirror+0.004
 opera+0.004
Neg logit contributions
es-0.007
icked-0.007
iled-0.005
owed-0.005
ched-0.005
Token on
Token ID319
Feature activation+1.87e+00

Pos logit contributions
es+0.041
icked+0.040
 could+0.029
iled+0.029
owed+0.028
Neg logit contributions
 city-0.065
 hotel-0.064
 talent-0.064
 opera-0.064
 orchestra-0.064
Token your
Token ID534
Feature activation-6.84e+00

Pos logit contributions
es+0.028
icked+0.026
 they+0.021
 could+0.020
 Remy+0.020
Neg logit contributions
 orchestra-0.041
 city-0.041
 airport-0.040
 hotel-0.040
 talent-0.040
Token p
Token ID279
Feature activation-2.00e+00

Pos logit contributions
 city+0.092
 orchestra+0.091
 talent+0.087
 opera+0.086
 airport+0.085
Neg logit contributions
es-0.151
icked-0.148
 could-0.132
 on-0.126
 they-0.125
Tokenaj
Token ID1228
Feature activation+3.14e+00

Pos logit contributions
 city+0.038
 orchestra+0.038
 opera+0.037
 talent+0.036
 airport+0.036
Neg logit contributions
icked-0.037
es-0.037
owed-0.031
o-0.030
iled-0.029
Tokenamas
Token ID17485
Feature activation+9.54e+00

Pos logit contributions
icked+0.059
es+0.056
iled+0.046
 could+0.046
ched+0.045
Neg logit contributions
 city-0.058
 mirror-0.058
 talent-0.058
 airport-0.057
 hotel-0.057
Token.
Token ID13
Feature activation+4.99e+00

Pos logit contributions
icked+0.148
es+0.139
 Remy+0.103
iled+0.102
owed+0.101
Neg logit contributions
 city-0.212
 airport-0.208
 hotel-0.208
 opera-0.208
 mirror-0.206
Token Then
Token ID3244
Feature activation+3.72e+00

Pos logit contributions
icked+0.073
es+0.071
owed+0.053
ched+0.052
ed+0.051
Neg logit contributions
 city-0.116
 mirror-0.113
 orchestra-0.113
 hotel-0.112
 opera-0.111
Token we
Token ID356
Feature activation+3.30e+00

Pos logit contributions
icked+0.053
es+0.049
iled+0.041
owed+0.038
ched+0.038
Neg logit contributions
 hotel-0.089
 mirror-0.089
 city-0.087
 opera-0.086
 airport-0.085
Token can
Token ID460
Feature activation+1.76e+00

Pos logit contributions
es+0.057
icked+0.057
iled+0.043
 could+0.043
owed+0.043
Neg logit contributions
 city-0.067
 hotel-0.066
 opera-0.065
 talent-0.065
 orchestra-0.065
Token read
Token ID1100
Feature activation+5.14e+00

Pos logit contributions
icked+0.038
es+0.037
iled+0.030
owed+0.030
ed+0.030
Neg logit contributions
 hotel-0.029
 city-0.029
 mirror-0.029
 airport-0.029
 opera-0.029
Token no
Token ID645
Feature activation-2.98e+00

Pos logit contributions
icked+0.026
es+0.025
iled+0.018
owed+0.018
ched+0.017
Neg logit contributions
 city-0.041
 opera-0.040
 talent-0.040
 hotel-0.040
 mirror-0.039
Token matter
Token ID2300
Feature activation+4.32e+00

Pos logit contributions
 orchestra+0.044
 airport+0.043
 city+0.043
 hotel+0.042
 priest+0.041
Neg logit contributions
es-0.065
icked-0.060
 could-0.057
 they-0.056
 on-0.052
Token how
Token ID703
Feature activation+2.12e+00

Pos logit contributions
icked+0.066
es+0.063
ched+0.048
iled+0.048
ed+0.047
Neg logit contributions
 opera-0.097
 city-0.096
 talent-0.095
 orchestra-0.092
 mirror-0.092
Token big
Token ID1263
Feature activation+2.02e+00

Pos logit contributions
icked+0.050
es+0.049
owed+0.041
iled+0.040
ched+0.040
Neg logit contributions
 city-0.030
 talent-0.029
 hotel-0.029
 opera-0.028
 mirror-0.028
Token the
Token ID262
Feature activation-8.70e+00

Pos logit contributions
icked+0.032
es+0.030
owed+0.022
iled+0.022
ched+0.021
Neg logit contributions
 city-0.046
 opera-0.045
 talent-0.045
 hotel-0.045
 airport-0.045
Token threat
Token ID2372
Feature activation+9.52e+00

Pos logit contributions
 passenger+0.107
 orchestra+0.104
 apartment+0.098
 cockpit+0.097
 tant+0.097
Neg logit contributions
es-0.190
 could-0.177
icked-0.170
 on-0.167
 they-0.167
Token seems
Token ID2331
Feature activation+4.12e+00

Pos logit contributions
icked+0.156
es+0.149
iled+0.117
owed+0.112
 Remy+0.108
Neg logit contributions
 city-0.202
 opera-0.201
 talent-0.196
 mirror-0.195
 hotel-0.195
Token,
Token ID11
Feature activation+3.30e+00

Pos logit contributions
icked+0.067
es+0.064
owed+0.046
iled+0.045
bed+0.044
Neg logit contributions
 city-0.089
 talent-0.089
 opera-0.088
 hotel-0.086
 mirror-0.085
Token it
Token ID340
Feature activation+4.59e+00

Pos logit contributions
icked+0.043
es+0.041
owed+0.029
iled+0.027
ched+0.026
Neg logit contributions
 city-0.084
 opera-0.082
 talent-0.081
 orchestra-0.081
 airport-0.081
Token is
Token ID318
Feature activation-3.92e-01

Pos logit contributions
icked+0.076
es+0.073
iled+0.055
owed+0.054
 Remy+0.053
Neg logit contributions
 city-0.096
 opera-0.095
 talent-0.094
 hotel-0.093
 orchestra-0.093
Token always
Token ID1464
Feature activation-8.19e-01

Pos logit contributions
 city+0.008
 talent+0.008
 opera+0.008
 hotel+0.008
 orchestra+0.008
Neg logit contributions
es-0.007
icked-0.007
iled-0.005
 could-0.005
owed-0.005
Token a
Token ID257
Feature activation-7.52e+00

Pos logit contributions
 orchestra+0.011
 city+0.011
 talent+0.011
 mirror+0.011
 opera+0.011
Neg logit contributions
es-0.006
icked-0.006
 they-0.004
 could-0.004
iled-0.004
Token few
Token ID1178
Feature activation-3.21e+00

Pos logit contributions
 ey+0.103
 orchestra+0.101
 former+0.093
 tant+0.090
 prope+0.090
Neg logit contributions
 they-0.145
 there-0.142
es-0.137
 could-0.133
 on-0.132
Token days
Token ID1528
Feature activation+4.63e+00

Pos logit contributions
 orchestra+0.052
 former+0.051
 mirror+0.051
 talent+0.050
 city+0.050
Neg logit contributions
es-0.066
icked-0.061
 could-0.053
 they-0.053
 on-0.051
Token when
Token ID618
Feature activation+2.76e+00

Pos logit contributions
icked+0.070
es+0.066
ched+0.048
iled+0.046
 Remy+0.046
Neg logit contributions
 city-0.109
 airport-0.106
 hotel-0.105
 opera-0.105
 talent-0.103
Token the
Token ID262
Feature activation-1.03e+01

Pos logit contributions
icked+0.025
es+0.024
iled+0.013
owed+0.013
ched+0.012
Neg logit contributions
 city-0.080
 hotel-0.078
 opera-0.078
 talent-0.078
 orchestra-0.078
Token renovations
Token ID44481
Feature activation+9.50e+00

Pos logit contributions
 tant+0.133
 talent+0.121
 orchestra+0.121
 ey+0.119
 former+0.116
Neg logit contributions
es-0.199
icked-0.197
 they-0.196
 on-0.193
 could-0.189
Token were
Token ID547
Feature activation-1.29e+00

Pos logit contributions
icked+0.129
es+0.124
 Remy+0.100
iled+0.096
ched+0.092
Neg logit contributions
 city-0.223
 hotel-0.217
 airport-0.213
 opera-0.212
 museum-0.209
Token done
Token ID1760
Feature activation+3.04e+00

Pos logit contributions
 city+0.025
 orchestra+0.024
 hotel+0.024
 talent+0.024
 opera+0.024
Neg logit contributions
es-0.024
icked-0.024
 could-0.019
iled-0.018
owed-0.018
Token.
Token ID13
Feature activation+5.25e+00

Pos logit contributions
icked+0.043
es+0.041
iled+0.029
owed+0.028
 could+0.028
Neg logit contributions
 city-0.072
 talent-0.071
 opera-0.071
 hotel-0.071
 orchestra-0.070
Token Everyone
Token ID11075
Feature activation+2.46e+00

Pos logit contributions
icked+0.075
es+0.074
owed+0.053
ched+0.052
iled+0.051
Neg logit contributions
 city-0.126
 hotel-0.122
 orchestra-0.120
 airport-0.120
 opera-0.120
Token could
Token ID714
Feature activation+1.10e+00

Pos logit contributions
icked+0.037
es+0.037
iled+0.027
 Remy+0.027
owed+0.026
Neg logit contributions
 city-0.054
 opera-0.054
 talent-0.053
 hotel-0.053
 orchestra-0.053
Token stopped
Token ID5025
Feature activation+3.47e+00

Pos logit contributions
icked+0.038
es+0.038
iled+0.027
 Remy+0.027
uts+0.026
Neg logit contributions
 mirror-0.065
 opera-0.063
 city-0.063
 airport-0.063
 hotel-0.063
Token.
Token ID13
Feature activation+5.19e+00

Pos logit contributions
es+0.046
icked+0.045
 could+0.031
iled+0.031
owed+0.030
Neg logit contributions
 city-0.085
 hotel-0.083
 talent-0.083
 opera-0.083
 orchestra-0.083
Token The
Token ID383
Feature activation-9.32e+00

Pos logit contributions
icked+0.079
es+0.077
owed+0.057
ched+0.056
iled+0.055
Neg logit contributions
 city-0.119
 mirror-0.114
 orchestra-0.114
 classroom-0.113
 playground-0.112
Token colorful
Token ID20239
Feature activation-5.92e+00

Pos logit contributions
 orchestra+0.113
 former+0.111
 tant+0.109
 talent+0.106
 hotel+0.104
Neg logit contributions
es-0.191
icked-0.181
 could-0.179
 on-0.177
 they-0.174
Token rain
Token ID6290
Feature activation+6.42e+00

Pos logit contributions
 former+0.071
 orchestra+0.064
 tant+0.064
 city+0.063
 hotel+0.063
Neg logit contributions
es-0.140
icked-0.131
 could-0.124
 they-0.124
 on-0.118
Tokendrops
Token ID49253
Feature activation+9.49e+00

Pos logit contributions
es+0.096
icked+0.096
iled+0.069
owed+0.066
 could+0.066
Neg logit contributions
 city-0.146
 opera-0.143
 hotel-0.143
 orchestra-0.142
 talent-0.141
Token stopped
Token ID5025
Feature activation+3.21e+00

Pos logit contributions
icked+0.144
es+0.143
 Remy+0.105
iled+0.105
owed+0.100
Neg logit contributions
 city-0.210
 opera-0.206
 hotel-0.206
 airport-0.204
 mirror-0.204
Token falling
Token ID7463
Feature activation+4.58e+00

Pos logit contributions
es+0.047
icked+0.047
 could+0.033
iled+0.032
owed+0.032
Neg logit contributions
 city-0.075
 opera-0.073
 hotel-0.073
 talent-0.073
 orchestra-0.072
Token from
Token ID422
Feature activation+8.01e-01

Pos logit contributions
icked+0.065
es+0.064
iled+0.045
owed+0.044
 could+0.042
Neg logit contributions
 city-0.109
 opera-0.106
 hotel-0.106
 airport-0.105
 orchestra-0.105
Token the
Token ID262
Feature activation-8.32e+00

Pos logit contributions
es+0.010
icked+0.009
 they+0.007
 could+0.007
 Remy+0.006
Neg logit contributions
 talent-0.019
 orchestra-0.019
 hotel-0.019
 city-0.019
 tant-0.019
Token sky
Token ID6766
Feature activation+5.29e+00

Pos logit contributions
 talent+0.095
 orchestra+0.089
 tant+0.089
 former+0.085
 opera+0.083
Neg logit contributions
es-0.181
 could-0.172
 they-0.172
icked-0.170
 on-0.169
Token on
Token ID319
Feature activation+1.42e+00

Pos logit contributions
icked+0.080
es+0.076
owed+0.053
iled+0.053
 could+0.051
Neg logit contributions
 city-0.128
 hotel-0.126
 opera-0.125
 airport-0.124
 talent-0.123
Token a
Token ID257
Feature activation-5.05e+00

Pos logit contributions
es+0.019
icked+0.018
 could+0.014
iled+0.013
 they+0.013
Neg logit contributions
 city-0.033
 orchestra-0.033
 talent-0.033
 hotel-0.033
 opera-0.033
Token leash
Token ID40529
Feature activation+8.23e+00

Pos logit contributions
 ey+0.073
 former+0.073
 pride+0.071
 amaz+0.069
 tant+0.069
Neg logit contributions
 they-0.090
es-0.089
 could-0.086
 Remy-0.085
ches-0.085
Token.
Token ID13
Feature activation+5.14e+00

Pos logit contributions
icked+0.124
es+0.120
iled+0.089
 Remy+0.083
owed+0.083
Neg logit contributions
 city-0.186
 airport-0.183
 opera-0.182
 hotel-0.181
 bus-0.178
Token The
Token ID383
Feature activation-9.22e+00

Pos logit contributions
icked+0.081
es+0.080
owed+0.062
ched+0.059
iled+0.057
Neg logit contributions
 city-0.116
 playground-0.109
 classroom-0.109
 orchestra-0.109
 hotel-0.109
Token leash
Token ID40529
Feature activation+9.46e+00

Pos logit contributions
 former+0.132
 orchestra+0.132
 talent+0.131
 tant+0.128
 hotel+0.127
Neg logit contributions
es-0.178
 could-0.167
icked-0.166
 Remy-0.165
 they-0.163
Token was
Token ID373
Feature activation-7.39e-01

Pos logit contributions
icked+0.141
es+0.138
iled+0.103
owed+0.099
 Remy+0.098
Neg logit contributions
 city-0.211
 hotel-0.207
 airport-0.206
 mirror-0.206
 opera-0.205
Token a
Token ID257
Feature activation-4.83e+00

Pos logit contributions
 city+0.012
 opera+0.012
 talent+0.012
 hotel+0.012
 orchestra+0.012
Neg logit contributions
es-0.016
icked-0.015
owed-0.012
iled-0.012
 could-0.012
Token long
Token ID890
Feature activation-3.33e+00

Pos logit contributions
 orchestra+0.078
 opera+0.077
 ey+0.077
 former+0.077
 city+0.076
Neg logit contributions
 could-0.084
 they-0.082
 on-0.078
es-0.074
 there-0.073
Token,
Token ID11
Feature activation+8.92e-01

Pos logit contributions
 city+0.067
 hotel+0.065
 talent+0.065
 opera+0.065
 orchestra+0.065
Neg logit contributions
es-0.058
icked-0.057
iled-0.044
 could-0.044
owed-0.043
Token red
Token ID2266
Feature activation-3.21e+00

Pos logit contributions
es+0.019
icked+0.019
 could+0.015
iled+0.015
ched+0.015
Neg logit contributions
 city-0.015
 hotel-0.014
 talent-0.014
 opera-0.014
 orchestra-0.014
Token helped
Token ID4193
Feature activation+1.58e+00

Pos logit contributions
es+0.004
icked+0.004
uts+0.003
 Remy+0.003
owed+0.003
Neg logit contributions
 city-0.005
 airport-0.005
 talent-0.005
 hotel-0.005
 mirror-0.005
Token the
Token ID262
Feature activation-1.27e+01

Pos logit contributions
icked+0.032
es+0.030
iled+0.025
uts+0.025
ched+0.025
Neg logit contributions
 city-0.027
 opera-0.027
 hotel-0.026
 church-0.026
 mirror-0.026
Token old
Token ID1468
Feature activation-9.27e+00

Pos logit contributions
 orchestra+0.147
 talent+0.145
 cockpit+0.138
 former+0.138
 mirror+0.137
Neg logit contributions
es-0.309
icked-0.307
 they-0.287
 could-0.277
owed-0.267
Token lady
Token ID10846
Feature activation+1.61e+00

Pos logit contributions
 talent+0.086
 orchestra+0.085
 former+0.084
 city+0.082
 hotel+0.081
Neg logit contributions
es-0.258
icked-0.251
 they-0.222
owed-0.218
 could-0.218
Token fix
Token ID4259
Feature activation-7.63e-01

Pos logit contributions
icked+0.031
es+0.030
 Remy+0.024
iled+0.024
ched+0.023
Neg logit contributions
 city-0.029
 mirror-0.028
 hotel-0.028
 talent-0.028
 opera-0.028
Token the
Token ID262
Feature activation-1.59e+01

Pos logit contributions
 city+0.017
 hotel+0.017
 opera+0.017
 talent+0.016
 orchestra+0.016
Neg logit contributions
icked-0.012
es-0.012
iled-0.009
 could-0.009
owed-0.008
Token fence
Token ID13990
Feature activation+3.84e+00

Pos logit contributions
 orchestra+0.223
 talent+0.210
 mirror+0.205
 actor+0.202
 priest+0.202
Neg logit contributions
es-0.358
icked-0.350
 could-0.315
 they-0.311
iled-0.305
Token.
Token ID13
Feature activation+2.66e+00

Pos logit contributions
icked+0.062
es+0.058
 Remy+0.044
iled+0.043
ched+0.042
Neg logit contributions
 city-0.085
 hotel-0.081
 airport-0.081
 playground-0.080
 mirror-0.080
Token
Token ID220
Feature activation+6.46e-01

Pos logit contributions
es+0.047
icked+0.046
ppers+0.040
ched+0.038
iled+0.038
Neg logit contributions
 city-0.051
 park-0.048
 playground-0.047
 zoo-0.047
 hotel-0.047
Token
Token ID198
Feature activation-3.08e-01

Pos logit contributions
icked+0.011
es+0.011
iled+0.010
ched+0.009
ppers+0.009
Neg logit contributions
 city-0.011
 park-0.011
 new-0.010
 playground-0.010
 classroom-0.010
Token
Token ID198
Feature activation-4.58e-01

Pos logit contributions
 city+0.006
 park+0.006
 talent+0.006
 playground+0.006
 hotel+0.006
Neg logit contributions
icked-0.006
es-0.005
owed-0.004
 neither-0.004
iled-0.004
Token "
Token ID366
Feature activation+3.33e-01

Pos logit contributions
icked+0.040
es+0.038
owed+0.034
ched+0.032
ung+0.031
Neg logit contributions
 mirror-0.037
 opera-0.036
 bus-0.036
 woman-0.035
 zoo-0.035
TokenMaybe
Token ID13300
Feature activation+1.50e+00

Pos logit contributions
icked+0.007
es+0.006
owed+0.005
iled+0.005
 could+0.005
Neg logit contributions
 city-0.006
 mirror-0.006
 talent-0.006
 opera-0.006
 airport-0.006
Token we
Token ID356
Feature activation+1.10e+00

Pos logit contributions
icked+0.023
es+0.022
ched+0.018
iled+0.017
owed+0.017
Neg logit contributions
 city-0.035
 opera-0.034
 museum-0.034
 airport-0.034
 talent-0.033
Token can
Token ID460
Feature activation-2.53e-01

Pos logit contributions
es+0.019
icked+0.019
owed+0.015
iled+0.015
 Remy+0.014
Neg logit contributions
 city-0.022
 talent-0.021
 hotel-0.021
 opera-0.021
 airport-0.021
Token share
Token ID2648
Feature activation+1.81e+00

Pos logit contributions
 city+0.004
 talent+0.004
 opera+0.004
 hotel+0.004
 mirror+0.004
Neg logit contributions
icked-0.005
es-0.005
iled-0.004
owed-0.004
 could-0.004
Token the
Token ID262
Feature activation-1.53e+01

Pos logit contributions
es+0.027
icked+0.026
 could+0.019
iled+0.019
 they+0.018
Neg logit contributions
 city-0.042
 hotel-0.041
 orchestra-0.041
 airport-0.041
 talent-0.041
Token ball
Token ID2613
Feature activation+2.69e+00

Pos logit contributions
 priest+0.304
 orchestra+0.298
 former+0.292
 hotel+0.290
 opera+0.289
Neg logit contributions
es-0.250
icked-0.242
 they-0.224
 could-0.222
 on-0.217
Token with
Token ID351
Feature activation+4.04e-01

Pos logit contributions
icked+0.039
es+0.036
owed+0.027
ched+0.026
 could+0.026
Neg logit contributions
 city-0.063
 opera-0.062
 talent-0.062
 hotel-0.061
 mirror-0.060
Token him
Token ID683
Feature activation+1.67e+00

Pos logit contributions
es+0.006
icked+0.006
iled+0.004
 could+0.004
owed+0.004
Neg logit contributions
 city-0.009
 talent-0.009
 opera-0.009
 hotel-0.009
 orchestra-0.009
Token.
Token ID13
Feature activation+3.03e+00

Pos logit contributions
icked+0.027
es+0.024
ched+0.020
owed+0.019
iled+0.019
Neg logit contributions
 city-0.035
 opera-0.034
 hotel-0.034
 talent-0.033
 airport-0.033
Token Maybe
Token ID6674
Feature activation+1.45e+00

Pos logit contributions
es+0.046
icked+0.045
owed+0.036
ched+0.034
iled+0.032
Neg logit contributions
 city-0.067
 museum-0.065
 opera-0.064
 orchestra-0.064
 mirror-0.064
Token you
Token ID345
Feature activation+1.13e+00

Pos logit contributions
es+0.021
icked+0.020
owed+0.015
iled+0.014
ched+0.014
Neg logit contributions
 city-0.035
 opera-0.035
 talent-0.035
 hotel-0.034
 mirror-0.034
Token.
Token ID13
Feature activation+2.50e+00

Pos logit contributions
icked+0.019
es+0.017
ched+0.015
owed+0.014
iled+0.014
Neg logit contributions
 opera-0.023
 talent-0.023
 city-0.023
 mistake-0.023
 orchestra-0.022
Token I
Token ID314
Feature activation+1.41e+00

Pos logit contributions
es+0.041
icked+0.039
owed+0.031
ched+0.030
iled+0.029
Neg logit contributions
 talent-0.053
 city-0.053
 mirror-0.053
 opera-0.053
 museum-0.052
Token will
Token ID481
Feature activation+2.72e-01

Pos logit contributions
es+0.028
icked+0.028
 could+0.024
iled+0.023
o+0.023
Neg logit contributions
 city-0.025
 airport-0.025
 orchestra-0.025
 hotel-0.025
 opera-0.024
Token fix
Token ID4259
Feature activation+1.63e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.004
iled+0.003
 they+0.003
Neg logit contributions
 city-0.006
 hotel-0.006
 airport-0.006
 orchestra-0.006
 opera-0.005
Token the
Token ID262
Feature activation-1.52e+01

Pos logit contributions
es+0.025
icked+0.023
 they+0.017
 could+0.016
 Remy+0.016
Neg logit contributions
 city-0.038
 orchestra-0.037
 hotel-0.037
 talent-0.037
 airport-0.037
Token pengu
Token ID38373
Feature activation-3.49e+00

Pos logit contributions
 orchestra+0.239
 former+0.237
 classrooms+0.224
 talent+0.223
 doorway+0.219
Neg logit contributions
es-0.275
 on-0.250
 with-0.243
 we-0.238
 in-0.237
Tokenin
Token ID259
Feature activation+3.50e-02

Pos logit contributions
 city+0.071
 orchestra+0.069
 airport+0.068
 talent+0.068
 hotel+0.068
Neg logit contributions
es-0.064
icked-0.061
iled-0.047
 could-0.047
owed-0.046
Token.
Token ID13
Feature activation+2.84e+00

Pos logit contributions
icked+0.000
es+0.000
 could+0.000
ched+0.000
iled+0.000
Neg logit contributions
 city-0.001
 hotel-0.001
 opera-0.001
 talent-0.001
 orchestra-0.001
Token Then
Token ID3244
Feature activation+2.51e+00

Pos logit contributions
es+0.043
icked+0.040
owed+0.031
iled+0.029
ched+0.029
Neg logit contributions
 city-0.065
 talent-0.064
 mirror-0.064
 opera-0.064
 orchestra-0.063
Token you
Token ID345
Feature activation+1.74e+00

Pos logit contributions
icked+0.045
es+0.045
uts+0.036
ched+0.036
iled+0.036
Neg logit contributions
 mirror-0.051
 talent-0.050
 new-0.049
 opera-0.049
 memory-0.049
Token him
Token ID683
Feature activation+7.81e-01

Pos logit contributions
es+0.033
icked+0.033
iled+0.021
 could+0.021
owed+0.020
Neg logit contributions
 city-0.074
 opera-0.073
 hotel-0.073
 talent-0.072
 orchestra-0.072
Token if
Token ID611
Feature activation+2.30e+00

Pos logit contributions
icked+0.011
es+0.010
iled+0.008
ched+0.008
owed+0.007
Neg logit contributions
 opera-0.018
 mirror-0.018
 city-0.018
 talent-0.018
 hotel-0.018
Token he
Token ID339
Feature activation+2.35e+00

Pos logit contributions
icked+0.029
es+0.027
iled+0.020
owed+0.020
ched+0.019
Neg logit contributions
 opera-0.059
 city-0.058
 talent-0.058
 airport-0.058
 hotel-0.058
Token would
Token ID561
Feature activation+1.05e+00

Pos logit contributions
es+0.026
icked+0.026
 Remy+0.018
iled+0.018
bed+0.017
Neg logit contributions
 city-0.057
 mirror-0.057
 hotel-0.056
 talent-0.056
 museum-0.056
Token share
Token ID2648
Feature activation+1.89e+00

Pos logit contributions
es+0.020
icked+0.019
iled+0.015
 could+0.015
owed+0.015
Neg logit contributions
 city-0.020
 hotel-0.020
 talent-0.019
 opera-0.019
 orchestra-0.019
Token the
Token ID262
Feature activation-1.52e+01

Pos logit contributions
es+0.028
icked+0.027
 could+0.020
iled+0.020
owed+0.019
Neg logit contributions
 city-0.044
 orchestra-0.043
 hotel-0.043
 talent-0.043
 opera-0.043
Token pie
Token ID2508
Feature activation+1.44e+00

Pos logit contributions
 orchestra+0.286
 opera+0.252
 classrooms+0.251
 former+0.249
 hotel+0.246
Neg logit contributions
icked-0.275
es-0.268
 they-0.248
 there-0.243
 could-0.241
Token with
Token ID351
Feature activation+6.87e-01

Pos logit contributions
icked+0.021
es+0.020
ched+0.015
iled+0.015
owed+0.014
Neg logit contributions
 city-0.033
 talent-0.032
 opera-0.032
 hotel-0.032
 mirror-0.032
Token her
Token ID607
Feature activation-5.17e+00

Pos logit contributions
es+0.008
icked+0.007
 could+0.005
iled+0.004
 they+0.004
Neg logit contributions
 city-0.018
 hotel-0.018
 orchestra-0.018
 talent-0.018
 opera-0.018
Token.
Token ID13
Feature activation+2.00e+00

Pos logit contributions
 city+0.131
 hotel+0.129
 orchestra+0.129
 opera+0.129
 talent+0.129
Neg logit contributions
es-0.065
icked-0.064
 could-0.044
iled-0.043
owed-0.042
Token
Token ID198
Feature activation+7.95e-01

Pos logit contributions
icked+0.029
es+0.028
owed+0.021
iled+0.020
ched+0.020
Neg logit contributions
 city-0.048
 talent-0.047
 mirror-0.046
 opera-0.046
 classroom-0.046
Token little
Token ID1310
Feature activation-6.32e+00

Pos logit contributions
 former+0.204
reating+0.187
 tant+0.181
opl+0.177
 orchestra+0.175
Neg logit contributions
 could-0.231
es-0.231
 there-0.228
 they-0.225
 in-0.216
Token boy
Token ID2933
Feature activation+1.34e+00

Pos logit contributions
 former+0.078
 talent+0.077
 orchestra+0.076
 city+0.073
 opera+0.072
Neg logit contributions
es-0.155
icked-0.149
 could-0.137
 they-0.134
ed-0.133
Token to
Token ID284
Feature activation-1.38e+00

Pos logit contributions
icked+0.019
es+0.018
iled+0.014
ched+0.014
uts+0.014
Neg logit contributions
 opera-0.030
 city-0.030
 hotel-0.030
 mirror-0.030
 talent-0.029
Token put
Token ID1234
Feature activation-5.75e-01

Pos logit contributions
 city+0.026
 hotel+0.026
 orchestra+0.026
 talent+0.026
 airport+0.025
Neg logit contributions
icked-0.025
es-0.025
 could-0.020
 they-0.019
iled-0.019
Token away
Token ID1497
Feature activation+1.24e+00

Pos logit contributions
 city+0.013
 mirror+0.013
 opera+0.013
 talent+0.013
 hotel+0.013
Neg logit contributions
es-0.009
icked-0.009
iled-0.006
owed-0.006
 could-0.006
Token the
Token ID262
Feature activation-1.51e+01

Pos logit contributions
es+0.021
icked+0.021
 could+0.015
iled+0.015
owed+0.015
Neg logit contributions
 city-0.026
 hotel-0.026
 opera-0.026
 talent-0.026
 orchestra-0.026
Token hose
Token ID31489
Feature activation+4.04e+00

Pos logit contributions
 former+0.256
 orchestra+0.248
 hotel+0.242
 talent+0.241
 city+0.240
Neg logit contributions
icked-0.287
es-0.286
 they-0.256
 could-0.253
 on-0.250
Token and
Token ID290
Feature activation+1.39e-01

Pos logit contributions
icked+0.061
es+0.060
iled+0.042
ched+0.041
owed+0.041
Neg logit contributions
 city-0.092
 hotel-0.089
 opera-0.089
 mirror-0.089
 airport-0.089
Token come
Token ID1282
Feature activation+7.84e-01

Pos logit contributions
es+0.002
icked+0.002
iled+0.002
owed+0.002
ched+0.002
Neg logit contributions
 city-0.003
 opera-0.003
 hotel-0.003
 talent-0.003
 orchestra-0.003
Token inside
Token ID2641
Feature activation+3.18e+00

Pos logit contributions
icked+0.017
es+0.016
ched+0.013
uts+0.013
iled+0.013
Neg logit contributions
 city-0.013
 mirror-0.012
 opera-0.012
 hotel-0.012
 playground-0.012
Token.
Token ID13
Feature activation+3.56e+00

Pos logit contributions
icked+0.044
es+0.043
iled+0.030
owed+0.030
ched+0.029
Neg logit contributions
 city-0.077
 hotel-0.075
 opera-0.075
 talent-0.075
 mirror-0.074
Token why
Token ID1521
Feature activation-3.60e-01

Pos logit contributions
icked+0.044
es+0.041
ched+0.031
iled+0.031
owed+0.029
Neg logit contributions
 city-0.076
 mirror-0.075
 hotel-0.075
 opera-0.075
 talent-0.075
Token he
Token ID339
Feature activation+1.52e+00

Pos logit contributions
 city+0.010
 hotel+0.009
 opera+0.009
 airport+0.009
 mirror+0.009
Neg logit contributions
icked-0.004
es-0.004
iled-0.003
ched-0.003
owed-0.003
Token had
Token ID550
Feature activation-8.62e-01

Pos logit contributions
icked+0.025
es+0.024
iled+0.017
 Remy+0.017
uts+0.016
Neg logit contributions
 city-0.033
 mirror-0.032
 talent-0.032
 hotel-0.031
 airport-0.031
Token fixed
Token ID5969
Feature activation-4.97e-01

Pos logit contributions
 city+0.020
 orchestra+0.020
 hotel+0.020
 opera+0.019
 former+0.019
Neg logit contributions
icked-0.012
es-0.012
 could-0.010
 they-0.009
ched-0.009
Token her
Token ID607
Feature activation-9.61e+00

Pos logit contributions
 city+0.013
 hotel+0.013
 talent+0.013
 orchestra+0.013
 opera+0.013
Neg logit contributions
es-0.006
icked-0.005
 could-0.003
iled-0.003
owed-0.003
Token the
Token ID262
Feature activation-1.51e+01

Pos logit contributions
 former+0.112
 city+0.111
 orchestra+0.108
 passenger+0.107
 opera+0.104
Neg logit contributions
es-0.242
icked-0.225
 could-0.203
 they-0.197
 on-0.193
Token screen
Token ID3159
Feature activation+4.24e+00

Pos logit contributions
 former+0.247
 priest+0.211
 salon+0.205
 tant+0.197
 passenger+0.191
Neg logit contributions
es-0.291
 on-0.290
 could-0.269
 in-0.267
 onto-0.265
Token.
Token ID13
Feature activation+2.95e+00

Pos logit contributions
icked+0.061
es+0.059
iled+0.043
ched+0.042
 Remy+0.041
Neg logit contributions
 city-0.100
 hotel-0.098
 opera-0.098
 talent-0.097
 mirror-0.097
Token
Token ID198
Feature activation-3.30e-01

Pos logit contributions
es+0.048
icked+0.048
iled+0.035
owed+0.035
ched+0.035
Neg logit contributions
 city-0.063
 talent-0.062
 hotel-0.062
 opera-0.062
 orchestra-0.062
Token
Token ID198
Feature activation-2.12e-03

Pos logit contributions
 city+0.008
 talent+0.008
 hotel+0.008
 opera+0.008
 orchestra+0.008
Neg logit contributions
es-0.004
icked-0.004
 could-0.003
iled-0.003
owed-0.003
Token"
Token ID1
Feature activation-1.38e-01

Pos logit contributions
 city+0.000
 hotel+0.000
 orchestra+0.000
 opera+0.000
 talent+0.000
Neg logit contributions
es-0.000
icked-0.000
 could-0.000
iled-0.000
owed-0.000
Token They
Token ID1119
Feature activation+2.38e-01

Pos logit contributions
es+0.038
icked+0.036
owed+0.027
ched+0.025
iled+0.025
Neg logit contributions
 city-0.067
 hotel-0.063
 talent-0.063
 playground-0.062
 museum-0.062
Token discuss
Token ID2112
Feature activation+2.88e+00

Pos logit contributions
icked+0.004
es+0.004
 could+0.003
iled+0.003
 they+0.003
Neg logit contributions
 hotel-0.005
 orchestra-0.005
 talent-0.005
 city-0.005
 opera-0.005
Token how
Token ID703
Feature activation-3.98e-01

Pos logit contributions
icked+0.041
es+0.041
iled+0.029
 could+0.028
owed+0.028
Neg logit contributions
 city-0.068
 opera-0.066
 hotel-0.066
 talent-0.066
 orchestra-0.066
Token to
Token ID284
Feature activation-2.13e+00

Pos logit contributions
 city+0.009
 talent+0.009
 opera+0.009
 hotel+0.009
 mirror+0.009
Neg logit contributions
es-0.006
icked-0.006
owed-0.004
iled-0.004
ched-0.004
Token share
Token ID2648
Feature activation+1.98e+00

Pos logit contributions
 orchestra+0.046
 city+0.046
 hotel+0.045
 airport+0.045
 opera+0.045
Neg logit contributions
icked-0.032
es-0.032
 could-0.026
 they-0.025
iled-0.025
Token the
Token ID262
Feature activation-1.50e+01

Pos logit contributions
es+0.026
icked+0.025
 could+0.018
iled+0.017
owed+0.017
Neg logit contributions
 city-0.049
 orchestra-0.048
 hotel-0.048
 opera-0.048
 talent-0.048
Token sausage
Token ID31889
Feature activation+2.12e+00

Pos logit contributions
 orchestra+0.314
 opera+0.309
 city+0.304
 airport+0.299
 hotel+0.297
Neg logit contributions
es-0.243
icked-0.231
 could-0.190
 they-0.185
iled-0.184
Token.
Token ID13
Feature activation+2.88e+00

Pos logit contributions
icked+0.033
es+0.029
owed+0.023
ched+0.022
iled+0.022
Neg logit contributions
 talent-0.047
 city-0.046
 hotel-0.045
 opera-0.045
 mistake-0.045
Token
Token ID198
Feature activation+6.47e-01

Pos logit contributions
es+0.042
icked+0.041
owed+0.032
ched+0.029
iled+0.029
Neg logit contributions
 city-0.070
 talent-0.067
 hotel-0.066
 playground-0.065
 opera-0.065
Token
Token ID198
Feature activation+9.59e-01

Pos logit contributions
icked+0.009
es+0.009
owed+0.007
iled+0.006
ched+0.006
Neg logit contributions
 city-0.015
 talent-0.015
 hotel-0.015
 opera-0.015
 mirror-0.015
TokenThey
Token ID2990
Feature activation+6.09e-01

Pos logit contributions
icked+0.014
es+0.014
owed+0.011
ched+0.009
ed+0.009
Neg logit contributions
 city-0.022
 talent-0.021
 hotel-0.021
 playground-0.021
 opera-0.021
Token did
Token ID750
Feature activation+5.04e-01

Pos logit contributions
 city+0.002
 talent+0.002
 hotel+0.002
 opera+0.002
 orchestra+0.002
Neg logit contributions
es-0.001
icked-0.001
iled-0.001
owed-0.001
 could-0.001
Token not
Token ID407
Feature activation-2.90e+00

Pos logit contributions
es+0.007
icked+0.007
iled+0.004
owed+0.004
 could+0.004
Neg logit contributions
 city-0.012
 talent-0.012
 opera-0.012
 hotel-0.012
 orchestra-0.012
Token want
Token ID765
Feature activation+3.55e-01

Pos logit contributions
 city+0.059
 hotel+0.058
 airport+0.058
 orchestra+0.057
 opera+0.057
Neg logit contributions
es-0.050
icked-0.050
 could-0.039
iled-0.039
ched-0.038
Token to
Token ID284
Feature activation-3.29e+00

Pos logit contributions
es+0.003
icked+0.003
 could+0.002
 they+0.002
iled+0.002
Neg logit contributions
 city-0.010
 orchestra-0.010
 hotel-0.010
 airport-0.010
 opera-0.010
Token share
Token ID2648
Feature activation+1.00e+00

Pos logit contributions
 city+0.069
 hotel+0.067
 orchestra+0.067
 airport+0.067
 opera+0.066
Neg logit contributions
es-0.055
icked-0.055
 could-0.042
iled-0.042
owed-0.040
Token the
Token ID262
Feature activation-1.50e+01

Pos logit contributions
icked+0.014
es+0.014
iled+0.010
owed+0.009
 could+0.009
Neg logit contributions
 city-0.024
 opera-0.024
 hotel-0.023
 talent-0.023
 mirror-0.023
Token cricket
Token ID18836
Feature activation+7.31e-01

Pos logit contributions
 city+0.323
 hotel+0.315
 airport+0.315
 priest+0.314
 former+0.311
Neg logit contributions
es-0.239
icked-0.215
 Remy-0.192
 Carl-0.181
ched-0.179
Token.
Token ID13
Feature activation+2.04e+00

Pos logit contributions
icked+0.011
es+0.010
owed+0.008
 could+0.008
iled+0.007
Neg logit contributions
 city-0.017
 opera-0.016
 talent-0.016
 hotel-0.016
 mirror-0.016
Token He
Token ID679
Feature activation+2.24e-01

Pos logit contributions
icked+0.029
es+0.029
owed+0.021
iled+0.021
ched+0.020
Neg logit contributions
 city-0.048
 mirror-0.047
 hotel-0.046
 talent-0.046
 opera-0.046
Token wanted
Token ID2227
Feature activation+4.05e-01

Pos logit contributions
es+0.003
icked+0.003
 could+0.002
iled+0.002
ched+0.002
Neg logit contributions
 city-0.006
 hotel-0.005
 orchestra-0.005
 opera-0.005
 talent-0.005
Token to
Token ID284
Feature activation-3.04e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.002
 they+0.002
 Remy+0.002
Neg logit contributions
 city-0.012
 orchestra-0.012
 hotel-0.011
 airport-0.011
 former-0.011
Token and
Token ID290
Feature activation-3.58e+00

Pos logit contributions
icked+0.024
es+0.021
othe+0.019
 neither+0.019
owed+0.018
Neg logit contributions
 city-0.024
 airport-0.023
 park-0.023
 opera-0.022
 mirror-0.022
Token Daddy
Token ID26926
Feature activation+1.35e+00

Pos logit contributions
 city+0.068
 talent+0.067
 hotel+0.067
 opera+0.067
 mirror+0.066
Neg logit contributions
es-0.069
icked-0.068
owed-0.053
iled-0.053
ched-0.052
Token helped
Token ID4193
Feature activation+2.13e+00

Pos logit contributions
icked+0.021
es+0.021
uts+0.016
 Remy+0.016
ched+0.016
Neg logit contributions
 city-0.028
 airport-0.027
 hotel-0.027
 mirror-0.027
 opera-0.027
Token Olive
Token ID30012
Feature activation+4.33e-01

Pos logit contributions
icked+0.043
es+0.040
uts+0.035
owed+0.034
umbered+0.032
Neg logit contributions
 new-0.036
 other-0.036
 park-0.035
 tant-0.034
 opera-0.034
Token fix
Token ID4259
Feature activation-7.87e-03

Pos logit contributions
icked+0.009
es+0.009
 Remy+0.007
iled+0.007
owed+0.007
Neg logit contributions
 mirror-0.007
 talent-0.007
 city-0.006
 park-0.006
 hotel-0.006
Token the
Token ID262
Feature activation-1.50e+01

Pos logit contributions
 city+0.000
 hotel+0.000
 talent+0.000
 orchestra+0.000
 opera+0.000
Neg logit contributions
es-0.000
icked-0.000
 could-0.000
iled-0.000
owed-0.000
Token toy
Token ID13373
Feature activation-1.30e+00

Pos logit contributions
 orchestra+0.238
 pastor+0.220
 priest+0.218
 opera+0.209
 audience+0.209
Neg logit contributions
icked-0.314
es-0.310
 could-0.284
 on-0.273
ed-0.270
Token.
Token ID13
Feature activation+3.07e+00

Pos logit contributions
 city+0.028
 hotel+0.026
 airport+0.026
 opera+0.026
 talent+0.026
Neg logit contributions
icked-0.022
es-0.021
iled-0.017
owed-0.017
ched-0.016
Token They
Token ID1119
Feature activation+4.56e-01

Pos logit contributions
es+0.057
icked+0.052
owed+0.051
ppers+0.047
mar+0.043
Neg logit contributions
 playground-0.056
 park-0.056
 city-0.056
 zoo-0.056
 elderly-0.056
Token were
Token ID547
Feature activation-2.79e+00

Pos logit contributions
es+0.008
icked+0.008
iled+0.006
owed+0.006
ched+0.006
Neg logit contributions
 city-0.009
 hotel-0.009
 talent-0.009
 airport-0.009
 mirror-0.009
Token gentle
Token ID10296
Feature activation+3.87e-02

Pos logit contributions
 city+0.042
 talent+0.041
 opera+0.041
 orchestra+0.040
 hotel+0.040
Neg logit contributions
es-0.060
icked-0.060
owed-0.050
iled-0.049
ched-0.049
Token!
Token ID0
Feature activation+3.19e+00

Pos logit contributions
icked+0.040
es+0.037
owed+0.029
iled+0.027
umbered+0.026
Neg logit contributions
 city-0.060
 opera-0.060
 mistake-0.059
 talent-0.059
 mirror-0.058
Token You
Token ID921
Feature activation+1.11e+00

Pos logit contributions
es+0.048
icked+0.047
owed+0.038
mar+0.035
ched+0.034
Neg logit contributions
 city-0.068
 zoo-0.064
 museum-0.064
 opera-0.064
 orchestra-0.064
Token need
Token ID761
Feature activation-3.17e-02

Pos logit contributions
es+0.018
icked+0.018
owed+0.013
iled+0.013
 Remy+0.012
Neg logit contributions
 city-0.023
 talent-0.023
 opera-0.023
 hotel-0.023
 airport-0.022
Token to
Token ID284
Feature activation-2.16e+00

Pos logit contributions
 city+0.001
 hotel+0.001
 orchestra+0.001
 airport+0.001
 opera+0.001
Neg logit contributions
es-0.000
icked-0.000
 could-0.000
iled-0.000
ched-0.000
Token share
Token ID2648
Feature activation+2.18e+00

Pos logit contributions
 hotel+0.047
 city+0.046
 orchestra+0.046
 airport+0.046
 opera+0.044
Neg logit contributions
icked-0.032
es-0.032
iled-0.025
 could-0.025
ched-0.024
Token the
Token ID262
Feature activation-1.50e+01

Pos logit contributions
es+0.031
icked+0.029
 could+0.021
iled+0.021
ched+0.020
Neg logit contributions
 city-0.052
 hotel-0.051
 orchestra-0.051
 talent-0.050
 airport-0.050
Token ball
Token ID2613
Feature activation+2.93e+00

Pos logit contributions
 orchestra+0.242
 priest+0.241
 opera+0.239
 hotel+0.233
 city+0.232
Neg logit contributions
es-0.308
icked-0.302
 could-0.263
iled-0.255
 they-0.254
Token."
Token ID526
Feature activation+3.65e+00

Pos logit contributions
icked+0.039
es+0.037
owed+0.026
 could+0.025
iled+0.025
Neg logit contributions
 city-0.072
 opera-0.071
 talent-0.070
 hotel-0.070
 orchestra-0.069
Token Buddy
Token ID36896
Feature activation+2.49e+00

Pos logit contributions
icked+0.061
owed+0.050
es+0.049
bled+0.049
ppers+0.047
Neg logit contributions
 zoo-0.070
 city-0.070
 park-0.070
 playground-0.066
 new-0.066
Token and
Token ID290
Feature activation-2.90e+00

Pos logit contributions
icked+0.045
es+0.042
owed+0.033
iled+0.032
umbered+0.032
Neg logit contributions
 mistake-0.044
 city-0.044
 talent-0.044
 playground-0.044
 park-0.044
Token Wh
Token ID854
Feature activation-2.03e+00

Pos logit contributions
 city+0.060
 hotel+0.058
 bus+0.058
 opera+0.058
 airport+0.058
Neg logit contributions
icked-0.048
es-0.047
owed-0.037
iled-0.037
ched-0.036
TokenLook
Token ID8567
Feature activation+1.19e+00

Pos logit contributions
es+0.007
icked+0.007
 could+0.005
iled+0.005
ched+0.005
Neg logit contributions
 city-0.008
 hotel-0.008
 talent-0.008
 opera-0.008
 orchestra-0.008
Token.
Token ID13
Feature activation+3.43e+00

Pos logit contributions
icked+0.018
es+0.017
ched+0.014
iled+0.013
uts+0.013
Neg logit contributions
 opera-0.026
 mirror-0.026
 city-0.025
 hotel-0.025
 airport-0.025
Token I
Token ID314
Feature activation+7.76e-01

Pos logit contributions
es+0.046
icked+0.045
ched+0.034
owed+0.032
iled+0.031
Neg logit contributions
 city-0.083
 museum-0.081
 mirror-0.079
 hotel-0.079
 opera-0.079
Token have
Token ID423
Feature activation-1.10e+00

Pos logit contributions
es+0.012
icked+0.011
iled+0.008
owed+0.008
 could+0.008
Neg logit contributions
 city-0.018
 hotel-0.017
 talent-0.017
 opera-0.017
 orchestra-0.017
Token fixed
Token ID5969
Feature activation-6.72e-02

Pos logit contributions
 city+0.023
 orchestra+0.023
 hotel+0.023
 opera+0.023
 talent+0.022
Neg logit contributions
icked-0.018
es-0.018
 could-0.014
ched-0.013
iled-0.013
Token the
Token ID262
Feature activation-1.50e+01

Pos logit contributions
 city+0.001
 hotel+0.001
 talent+0.001
 orchestra+0.001
 opera+0.001
Neg logit contributions
es-0.001
icked-0.001
 could-0.001
iled-0.001
owed-0.001
Token important
Token ID1593
Feature activation-1.24e+01

Pos logit contributions
 talent+0.230
 actor+0.224
 circus+0.223
 tant+0.221
reating+0.217
Neg logit contributions
 could-0.279
 in-0.265
 on-0.263
 there-0.261
 they-0.256
Token roof
Token ID9753
Feature activation+4.11e+00

Pos logit contributions
 orchestra+0.157
 talent+0.151
 circus+0.146
 audience+0.145
Ta+0.145
Neg logit contributions
es-0.250
 could-0.246
 in-0.245
 they-0.242
 there-0.241
Token.
Token ID13
Feature activation+3.77e+00

Pos logit contributions
icked+0.060
es+0.058
iled+0.042
 Remy+0.041
ched+0.041
Neg logit contributions
 city-0.096
 hotel-0.094
 opera-0.094
 airport-0.093
 orchestra-0.092
Token Let
Token ID3914
Feature activation+2.99e+00

Pos logit contributions
es+0.052
icked+0.050
owed+0.036
ched+0.035
iled+0.035
Neg logit contributions
 city-0.091
 orchestra-0.088
 opera-0.088
 museum-0.087
 talent-0.087
Token's
Token ID338
Feature activation-1.95e+00

Pos logit contributions
icked+0.048
es+0.046
iled+0.036
owed+0.034
uts+0.033
Neg logit contributions
 city-0.066
 opera-0.066
 airport-0.065
 hotel-0.064
 mirror-0.064
Token.
Token ID13
Feature activation+3.05e+00

Pos logit contributions
icked+0.011
es+0.010
ched+0.008
iled+0.008
owed+0.008
Neg logit contributions
 city-0.015
 opera-0.015
 hotel-0.015
 mirror-0.014
 airport-0.014
Token They
Token ID1119
Feature activation+6.53e-01

Pos logit contributions
icked+0.050
es+0.048
owed+0.042
ched+0.039
iled+0.037
Neg logit contributions
 city-0.062
 zoo-0.058
 former-0.058
 park-0.057
 classroom-0.056
Token decided
Token ID3066
Feature activation+2.77e+00

Pos logit contributions
icked+0.009
es+0.009
 could+0.007
iled+0.006
owed+0.006
Neg logit contributions
 city-0.016
 hotel-0.016
 orchestra-0.016
 talent-0.016
 opera-0.016
Token to
Token ID284
Feature activation-9.67e-01

Pos logit contributions
icked+0.042
es+0.041
owed+0.030
iled+0.030
ched+0.027
Neg logit contributions
 city-0.064
 talent-0.063
 opera-0.063
 airport-0.062
 hotel-0.062
Token share
Token ID2648
Feature activation+1.96e+00

Pos logit contributions
 city+0.021
 orchestra+0.021
 hotel+0.021
 airport+0.020
 opera+0.020
Neg logit contributions
icked-0.015
es-0.015
 could-0.012
iled-0.011
 they-0.011
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
es+0.028
icked+0.027
 could+0.020
iled+0.019
 they+0.019
Neg logit contributions
 city-0.046
 orchestra-0.046
 hotel-0.046
 talent-0.045
 airport-0.045
Token ball
Token ID2613
Feature activation+2.34e+00

Pos logit contributions
 priest+0.273
 orchestra+0.270
 opera+0.266
 hotel+0.258
 airport+0.257
Neg logit contributions
es-0.262
icked-0.258
 they-0.232
 could-0.231
 Remy-0.222
Token and
Token ID290
Feature activation-1.37e+00

Pos logit contributions
icked+0.026
es+0.025
 could+0.016
owed+0.016
iled+0.016
Neg logit contributions
 city-0.062
 talent-0.061
 opera-0.060
 hotel-0.060
 orchestra-0.060
Token play
Token ID711
Feature activation+1.17e+00

Pos logit contributions
 city+0.029
 hotel+0.028
 orchestra+0.028
 opera+0.028
 airport+0.028
Neg logit contributions
es-0.023
icked-0.022
 could-0.018
 they-0.017
owed-0.016
Token together
Token ID1978
Feature activation+2.43e+00

Pos logit contributions
icked+0.013
es+0.012
owed+0.008
iled+0.008
o+0.007
Neg logit contributions
 talent-0.032
 opera-0.031
 city-0.031
 orchestra-0.030
 mirror-0.030
Token.
Token ID13
Feature activation+3.52e+00

Pos logit contributions
icked+0.039
es+0.037
owed+0.029
iled+0.028
 Remy+0.027
Neg logit contributions
 talent-0.052
 city-0.052
 opera-0.052
 hotel-0.050
 mirror-0.050
Token smiled
Token ID13541
Feature activation+4.45e-01

Pos logit contributions
icked+0.036
es+0.035
owed+0.025
iled+0.025
uts+0.025
Neg logit contributions
 city-0.049
 hotel-0.047
 airport-0.047
 talent-0.047
 opera-0.047
Token as
Token ID355
Feature activation+9.33e-01

Pos logit contributions
icked+0.007
es+0.007
ched+0.006
owed+0.005
iled+0.005
Neg logit contributions
 city-0.009
 opera-0.009
 mirror-0.009
 talent-0.009
 airport-0.009
Token he
Token ID339
Feature activation+1.17e+00

Pos logit contributions
es+0.012
icked+0.011
owed+0.008
iled+0.008
ched+0.007
Neg logit contributions
 city-0.024
 hotel-0.024
 airport-0.024
 opera-0.024
 mirror-0.024
Token started
Token ID2067
Feature activation+4.16e-01

Pos logit contributions
es+0.018
icked+0.018
iled+0.013
 could+0.013
owed+0.013
Neg logit contributions
 city-0.026
 hotel-0.025
 talent-0.025
 opera-0.025
 orchestra-0.025
Token pushing
Token ID7796
Feature activation+1.26e+00

Pos logit contributions
icked+0.006
es+0.006
owed+0.004
iled+0.004
 could+0.004
Neg logit contributions
 city-0.010
 talent-0.010
 opera-0.010
 hotel-0.010
 orchestra-0.010
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
es+0.018
icked+0.018
 could+0.013
iled+0.013
ched+0.012
Neg logit contributions
 city-0.030
 hotel-0.029
 talent-0.029
 orchestra-0.029
 opera-0.029
Token yacht
Token ID43207
Feature activation+3.68e+00

Pos logit contributions
 former+0.274
 talent+0.222
 orchestra+0.220
 opera+0.218
 library+0.217
Neg logit contributions
 they-0.288
icked-0.284
 could-0.278
es-0.274
ed-0.267
Token across
Token ID1973
Feature activation-9.59e-04

Pos logit contributions
icked+0.052
es+0.050
iled+0.035
owed+0.035
 Remy+0.034
Neg logit contributions
 city-0.087
 opera-0.085
 hotel-0.084
 airport-0.084
 talent-0.083
Token the
Token ID262
Feature activation-1.03e+01

Pos logit contributions
 orchestra+0.000
 talent+0.000
 former+0.000
 city+0.000
 opera+0.000
Neg logit contributions
es-0.000
icked-0.000
 could-0.000
 they-0.000
iled-0.000
Token lake
Token ID13546
Feature activation+4.21e+00

Pos logit contributions
 former+0.125
 orchestra+0.118
 talent+0.114
 tant+0.114
 fascination+0.113
Neg logit contributions
 could-0.220
 they-0.217
 on-0.215
 in-0.215
 there-0.209
Token.
Token ID13
Feature activation+2.95e+00

Pos logit contributions
icked+0.067
es+0.064
owed+0.047
ched+0.046
iled+0.046
Neg logit contributions
 city-0.094
 hotel-0.091
 opera-0.090
 playground-0.089
 airport-0.089
Token she
Token ID673
Feature activation+2.92e+00

Pos logit contributions
icked+0.013
es+0.013
iled+0.008
owed+0.007
ched+0.007
Neg logit contributions
 opera-0.045
 city-0.045
 talent-0.044
 mirror-0.044
 hotel-0.044
Token would
Token ID561
Feature activation+1.29e+00

Pos logit contributions
icked+0.048
es+0.048
iled+0.036
owed+0.035
ched+0.035
Neg logit contributions
 city-0.060
 talent-0.059
 opera-0.059
 hotel-0.058
 mirror-0.058
Token try
Token ID1949
Feature activation+2.13e+00

Pos logit contributions
es+0.025
icked+0.025
 could+0.020
iled+0.020
owed+0.020
Neg logit contributions
 city-0.023
 hotel-0.023
 orchestra-0.023
 opera-0.023
 airport-0.023
Token to
Token ID284
Feature activation-2.30e+00

Pos logit contributions
es+0.026
icked+0.025
 could+0.017
iled+0.017
owed+0.016
Neg logit contributions
 city-0.055
 hotel-0.054
 orchestra-0.054
 opera-0.053
 talent-0.053
Token fix
Token ID4259
Feature activation+1.68e+00

Pos logit contributions
 hotel+0.051
 city+0.049
 airport+0.048
 orchestra+0.048
 tant+0.048
Neg logit contributions
es-0.034
icked-0.034
 could-0.028
 they-0.027
owed-0.026
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
es+0.024
icked+0.022
 could+0.015
 they+0.015
iled+0.015
Neg logit contributions
 city-0.040
 orchestra-0.040
 talent-0.040
 hotel-0.040
 airport-0.039
Token train
Token ID4512
Feature activation+2.30e+00

Pos logit contributions
 orchestra+0.248
 former+0.235
 priest+0.232
 actor+0.229
 opera+0.227
Neg logit contributions
es-0.278
icked-0.268
 they-0.265
 on-0.249
 there-0.245
Token.
Token ID13
Feature activation+2.72e+00

Pos logit contributions
icked+0.031
es+0.031
iled+0.022
 could+0.021
owed+0.021
Neg logit contributions
 city-0.056
 opera-0.054
 hotel-0.054
 talent-0.054
 orchestra-0.054
Token She
Token ID1375
Feature activation+5.72e-01

Pos logit contributions
es+0.036
icked+0.036
owed+0.025
iled+0.025
ched+0.024
Neg logit contributions
 city-0.068
 talent-0.066
 mirror-0.066
 opera-0.065
 hotel-0.065
Token said
Token ID531
Feature activation+1.54e+00

Pos logit contributions
es+0.008
icked+0.008
iled+0.005
 could+0.005
owed+0.005
Neg logit contributions
 city-0.014
 talent-0.014
 hotel-0.013
 opera-0.013
 orchestra-0.013
Token they
Token ID484
Feature activation+2.66e+00

Pos logit contributions
icked+0.014
es+0.014
iled+0.008
owed+0.008
ched+0.007
Neg logit contributions
 opera-0.044
 city-0.044
 talent-0.043
 mirror-0.043
 hotel-0.043
Token going
Token ID1016
Feature activation+2.00e+00

Pos logit contributions
 city+0.053
 orchestra+0.052
 hotel+0.052
 talent+0.052
 opera+0.051
Neg logit contributions
es-0.045
icked-0.045
 could-0.036
iled-0.035
ched-0.033
Token to
Token ID284
Feature activation-1.13e+00

Pos logit contributions
icked+0.026
es+0.026
owed+0.018
iled+0.017
ched+0.016
Neg logit contributions
 city-0.050
 opera-0.049
 hotel-0.049
 airport-0.049
 talent-0.049
Token help
Token ID1037
Feature activation+2.45e+00

Pos logit contributions
 tant+0.022
 orchestra+0.022
 city+0.022
 hotel+0.021
 mirror+0.021
Neg logit contributions
es-0.017
icked-0.016
 they-0.015
 could-0.014
 Remy-0.014
Token us
Token ID514
Feature activation-2.47e-01

Pos logit contributions
icked+0.039
es+0.037
iled+0.027
owed+0.027
ched+0.027
Neg logit contributions
 city-0.055
 opera-0.054
 hotel-0.054
 talent-0.054
 orchestra-0.053
Token fix
Token ID4259
Feature activation+1.52e+00

Pos logit contributions
 city+0.005
 hotel+0.005
 opera+0.005
 talent+0.005
 airport+0.005
Neg logit contributions
icked-0.005
es-0.004
iled-0.003
owed-0.003
 could-0.003
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
es+0.022
icked+0.021
 could+0.015
iled+0.015
ched+0.014
Neg logit contributions
 city-0.036
 orchestra-0.035
 talent-0.035
 hotel-0.035
 opera-0.035
Token screen
Token ID3159
Feature activation+4.51e+00

Pos logit contributions
 former+0.215
 orchestra+0.190
reating+0.189
 priest+0.186
 tant+0.177
Neg logit contributions
 on-0.295
 could-0.280
 in-0.279
es-0.276
 there-0.272
Token."
Token ID526
Feature activation+3.46e+00

Pos logit contributions
icked+0.066
es+0.063
iled+0.046
ched+0.045
owed+0.044
Neg logit contributions
 city-0.106
 hotel-0.103
 opera-0.102
 airport-0.102
 talent-0.101
Token
Token ID198
Feature activation+1.29e-01

Pos logit contributions
icked+0.061
es+0.058
owed+0.046
ched+0.045
iled+0.044
Neg logit contributions
 city-0.072
 mirror-0.071
 talent-0.068
 hotel-0.067
 playground-0.067
Token
Token ID198
Feature activation+1.13e-01

Pos logit contributions
icked+0.002
es+0.002
owed+0.001
iled+0.001
 could+0.001
Neg logit contributions
 city-0.003
 talent-0.003
 hotel-0.003
 opera-0.003
 orchestra-0.003
TokenThe
Token ID464
Feature activation-1.20e+01

Pos logit contributions
es+0.002
icked+0.002
iled+0.001
 could+0.001
ched+0.001
Neg logit contributions
 city-0.002
 hotel-0.002
 opera-0.002
 talent-0.002
 orchestra-0.002
Token should
Token ID815
Feature activation+1.81e-01

Pos logit contributions
es+0.022
icked+0.021
iled+0.016
 could+0.016
owed+0.016
Neg logit contributions
 city-0.028
 hotel-0.027
 talent-0.027
 opera-0.027
 orchestra-0.027
Token be
Token ID307
Feature activation-3.27e+00

Pos logit contributions
es+0.003
icked+0.003
 could+0.003
iled+0.003
owed+0.003
Neg logit contributions
 city-0.003
 hotel-0.003
 talent-0.003
 orchestra-0.003
 opera-0.003
Token able
Token ID1498
Feature activation-3.21e-01

Pos logit contributions
 talent+0.050
 city+0.050
 opera+0.050
 orchestra+0.049
 mirror+0.048
Neg logit contributions
es-0.072
icked-0.072
owed-0.059
iled-0.059
 could-0.058
Token to
Token ID284
Feature activation-3.04e+00

Pos logit contributions
 talent+0.007
 opera+0.007
 city+0.007
 hotel+0.007
 new+0.007
Neg logit contributions
icked-0.005
es-0.005
owed-0.004
iled-0.004
ched-0.003
Token fix
Token ID4259
Feature activation+1.30e+00

Pos logit contributions
 city+0.056
 hotel+0.055
 orchestra+0.055
 mirror+0.055
 talent+0.054
Neg logit contributions
es-0.056
icked-0.054
 could-0.046
ched-0.044
 they-0.043
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
es+0.020
icked+0.019
 could+0.014
iled+0.014
ched+0.014
Neg logit contributions
 city-0.029
 hotel-0.029
 orchestra-0.028
 talent-0.028
 opera-0.028
Token dress
Token ID6576
Feature activation+3.48e+00

Pos logit contributions
 priest+0.220
 tant+0.212
 passenger+0.208
 city+0.206
 orchestra+0.206
Neg logit contributions
es-0.295
 on-0.276
icked-0.271
 could-0.268
 there-0.262
Token."
Token ID526
Feature activation+3.65e+00

Pos logit contributions
icked+0.051
es+0.047
iled+0.036
ched+0.034
 Remy+0.034
Neg logit contributions
 city-0.081
 opera-0.081
 hotel-0.080
 airport-0.079
 orchestra-0.079
Token
Token ID220
Feature activation+6.64e-01

Pos logit contributions
icked+0.059
es+0.057
owed+0.045
iled+0.044
ched+0.041
Neg logit contributions
 mirror-0.076
 talent-0.075
 city-0.073
 orchestra-0.072
 former-0.071
Token
Token ID198
Feature activation+2.30e-01

Pos logit contributions
icked+0.012
es+0.011
iled+0.009
owed+0.009
pper+0.009
Neg logit contributions
 new-0.011
 mirror-0.011
 former-0.011
 classroom-0.011
 mistake-0.011
Token
Token ID198
Feature activation+1.21e-01

Pos logit contributions
icked+0.004
es+0.003
owed+0.003
iled+0.002
 neither+0.002
Neg logit contributions
 talent-0.005
 city-0.005
 mirror-0.005
 orchestra-0.005
 opera-0.005
Token learned
Token ID4499
Feature activation-4.96e-01

Pos logit contributions
icked+0.005
es+0.005
 could+0.004
 they+0.003
 suggested+0.003
Neg logit contributions
 orchestra-0.007
 opera-0.007
 city-0.007
 hotel-0.007
 mirror-0.007
Token to
Token ID284
Feature activation-2.68e+00

Pos logit contributions
 city+0.015
 orchestra+0.015
 hotel+0.015
 airport+0.014
 mirror+0.014
Neg logit contributions
icked-0.004
es-0.004
 they-0.003
 could-0.002
 together-0.002
Token be
Token ID307
Feature activation-3.24e+00

Pos logit contributions
 hotel+0.065
 airport+0.065
 orchestra+0.064
 city+0.064
 tant+0.063
Neg logit contributions
icked-0.031
es-0.030
 could-0.025
 they-0.025
iled-0.024
Token careful
Token ID8161
Feature activation+2.15e+00

Pos logit contributions
 city+0.048
 hotel+0.047
 orchestra+0.046
 airport+0.046
 opera+0.045
Neg logit contributions
es-0.076
icked-0.075
 could-0.063
iled-0.062
ched-0.061
Token of
Token ID286
Feature activation-3.02e+00

Pos logit contributions
icked+0.036
es+0.032
 Remy+0.026
oped+0.025
owed+0.024
Neg logit contributions
 opera-0.044
 talent-0.044
 city-0.044
 mistake-0.043
 mirror-0.042
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
 orchestra+0.058
 city+0.056
 hotel+0.055
 airport+0.054
 talent+0.054
Neg logit contributions
es-0.057
icked-0.054
 could-0.045
 they-0.045
 Remy-0.043
Token crabs
Token ID49172
Feature activation+2.57e+00

Pos logit contributions
 orchestra+0.276
 passenger+0.274
 apartment+0.260
 former+0.257
 channels+0.255
Neg logit contributions
 could-0.237
 there-0.235
 they-0.228
 on-0.219
 it-0.216
Token.
Token ID13
Feature activation+2.68e+00

Pos logit contributions
icked+0.042
es+0.036
owed+0.029
ched+0.029
iled+0.029
Neg logit contributions
 city-0.056
 opera-0.055
 hotel-0.054
 airport-0.054
 talent-0.054
Token<|endoftext|>
Token ID50256
Feature activation+1.19e+00

Pos logit contributions
es+0.037
icked+0.037
owed+0.028
ched+0.025
iled+0.025
Neg logit contributions
 city-0.065
 talent-0.063
 hotel-0.062
 opera-0.062
 mirror-0.061
TokenBen
Token ID11696
Feature activation+2.25e+00

Pos logit contributions
icked+0.024
es+0.022
owed+0.021
iled+0.019
 neither+0.019
Neg logit contributions
 city-0.022
 park-0.021
 museum-0.021
 hotel-0.021
 mirror-0.021
Token and
Token ID290
Feature activation-1.12e+00

Pos logit contributions
icked+0.041
es+0.039
owed+0.032
 neither+0.031
othe+0.031
Neg logit contributions
 city-0.044
 park-0.043
 mirror-0.043
 museum-0.042
 hotel-0.042
Token let
Token ID1309
Feature activation+1.29e+00

Pos logit contributions
 city+0.003
 hotel+0.003
 opera+0.003
 mirror+0.003
 talent+0.003
Neg logit contributions
es-0.004
icked-0.004
iled-0.003
owed-0.003
ed-0.003
Token me
Token ID502
Feature activation-2.48e-01

Pos logit contributions
icked+0.027
es+0.027
iled+0.022
owed+0.022
o+0.021
Neg logit contributions
 opera-0.022
 city-0.022
 airport-0.021
 talent-0.021
 mirror-0.021
Token help
Token ID1037
Feature activation+1.39e+00

Pos logit contributions
 city+0.004
 talent+0.004
 hotel+0.004
 mirror+0.004
 opera+0.004
Neg logit contributions
es-0.006
icked-0.005
iled-0.004
ed-0.004
owed-0.004
Token you
Token ID345
Feature activation-1.05e+00

Pos logit contributions
icked+0.031
es+0.029
ed+0.024
owed+0.024
iled+0.024
Neg logit contributions
 opera-0.024
 talent-0.023
 mirror-0.023
 city-0.023
 orchestra-0.022
Token carry
Token ID3283
Feature activation+5.97e-01

Pos logit contributions
 city+0.018
 hotel+0.017
 talent+0.017
 mirror+0.017
 opera+0.017
Neg logit contributions
icked-0.022
es-0.022
iled-0.017
ed-0.017
owed-0.017
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
icked+0.009
es+0.009
owed+0.007
iled+0.007
 could+0.007
Neg logit contributions
 city-0.013
 talent-0.013
 opera-0.013
 hotel-0.013
 mirror-0.013
Token ac
Token ID936
Feature activation-4.52e+00

Pos logit contributions
 orchestra+0.311
 opera+0.300
 city+0.290
 hotel+0.288
 talent+0.287
Neg logit contributions
es-0.238
icked-0.235
 could-0.223
 on-0.194
ched-0.192
Tokenorn
Token ID1211
Feature activation+1.85e+00

Pos logit contributions
 city+0.076
 mirror+0.075
 talent+0.074
 hotel+0.073
 airport+0.073
Neg logit contributions
icked-0.092
es-0.090
owed-0.073
 they-0.072
iled-0.072
Token home
Token ID1363
Feature activation+3.85e+00

Pos logit contributions
icked+0.034
es+0.032
iled+0.025
owed+0.024
ched+0.023
Neg logit contributions
 city-0.037
 hotel-0.037
 talent-0.037
 airport-0.036
 mirror-0.036
Token?
Token ID30
Feature activation+4.26e+00

Pos logit contributions
icked+0.064
es+0.060
iled+0.046
ched+0.046
 Remy+0.044
Neg logit contributions
 city-0.083
 hotel-0.082
 talent-0.081
 mirror-0.081
 opera-0.081
Token I
Token ID314
Feature activation+9.99e-01

Pos logit contributions
icked+0.061
es+0.060
owed+0.045
led+0.045
mar+0.044
Neg logit contributions
 mirror-0.090
 city-0.089
 talent-0.088
 orchestra-0.088
 former-0.088
Token Tim
Token ID5045
Feature activation+1.13e+00

Pos logit contributions
es+0.028
icked+0.026
owed+0.021
ched+0.021
ppy+0.020
Neg logit contributions
 city-0.041
 playground-0.039
 museum-0.039
 park-0.038
 beach-0.038
Token and
Token ID290
Feature activation-3.12e+00

Pos logit contributions
icked+0.019
es+0.019
 Remy+0.015
owed+0.015
iled+0.014
Neg logit contributions
 city-0.022
 airport-0.021
 mirror-0.021
 talent-0.021
 park-0.021
Token his
Token ID465
Feature activation-9.91e+00

Pos logit contributions
 city+0.070
 orchestra+0.070
 talent+0.069
 hotel+0.069
 airport+0.069
Neg logit contributions
es-0.047
icked-0.047
 could-0.035
 they-0.034
o-0.034
Token dad
Token ID9955
Feature activation+1.38e+00

Pos logit contributions
 former+0.184
 orchestra+0.182
 opera+0.177
 hotel+0.172
 tant+0.171
Neg logit contributions
icked-0.176
es-0.168
 they-0.163
 could-0.159
 on-0.143
Token fixed
Token ID5969
Feature activation-3.70e+00

Pos logit contributions
es+0.023
icked+0.022
iled+0.017
owed+0.016
 Remy+0.016
Neg logit contributions
 city-0.029
 talent-0.028
 hotel-0.028
 opera-0.028
 airport-0.028
Token the
Token ID262
Feature activation-1.49e+01

Pos logit contributions
 city+0.086
 orchestra+0.085
 talent+0.084
 hotel+0.084
 opera+0.084
Neg logit contributions
es-0.054
icked-0.053
 could-0.038
iled-0.038
ched-0.036
Token raft
Token ID31812
Feature activation+3.95e+00

Pos logit contributions
 orchestra+0.299
 priest+0.288
 opera+0.287
 former+0.281
 audience+0.280
Neg logit contributions
icked-0.226
 they-0.212
 could-0.212
es-0.209
 on-0.200
Token.
Token ID13
Feature activation+2.15e+00

Pos logit contributions
icked+0.061
es+0.058
 Remy+0.046
iled+0.045
owed+0.043
Neg logit contributions
 city-0.083
 hotel-0.079
 airport-0.079
 model-0.079
 museum-0.079
Token They
Token ID1119
Feature activation+4.01e-01

Pos logit contributions
es+0.032
icked+0.029
owed+0.022
ched+0.021
iled+0.021
Neg logit contributions
 city-0.053
 museum-0.050
 playground-0.050
 hotel-0.049
 opera-0.048
Token used
Token ID973
Feature activation-8.23e-01

Pos logit contributions
es+0.007
icked+0.007
 could+0.005
iled+0.005
owed+0.005
Neg logit contributions
 city-0.009
 orchestra-0.008
 talent-0.008
 hotel-0.008
 opera-0.008
Token different
Token ID1180
Feature activation-7.61e+00

Pos logit contributions
 orchestra+0.019
 city+0.019
 hotel+0.019
 opera+0.019
 airport+0.019
Neg logit contributions
icked-0.012
es-0.012
 could-0.009
 they-0.009
ched-0.008
Token impatient
Token ID33440
Feature activation+2.24e+00

Pos logit contributions
 city+0.044
 hotel+0.043
 orchestra+0.043
 airport+0.042
 former+0.041
Neg logit contributions
icked-0.042
es-0.042
 could-0.033
o-0.032
 they-0.032
Token.
Token ID13
Feature activation+3.26e+00

Pos logit contributions
icked+0.038
es+0.036
owed+0.027
 Remy+0.027
ched+0.027
Neg logit contributions
 talent-0.048
 city-0.047
 opera-0.047
 mirror-0.046
 orchestra-0.046
Token We
Token ID775
Feature activation+1.91e+00

Pos logit contributions
es+0.050
icked+0.046
owed+0.037
ched+0.036
iled+0.034
Neg logit contributions
 mirror-0.073
 city-0.072
 opera-0.071
 talent-0.070
 orchestra-0.070
Token can
Token ID460
Feature activation+7.71e-01

Pos logit contributions
icked+0.027
es+0.027
 could+0.020
iled+0.019
ched+0.018
Neg logit contributions
 city-0.045
 orchestra-0.045
 hotel-0.045
 opera-0.044
 talent-0.044
Token share
Token ID2648
Feature activation+2.73e+00

Pos logit contributions
es+0.014
icked+0.014
iled+0.011
owed+0.011
ched+0.010
Neg logit contributions
 city-0.015
 talent-0.015
 opera-0.015
 hotel-0.015
 mirror-0.015
Token the
Token ID262
Feature activation-1.48e+01

Pos logit contributions
es+0.036
icked+0.035
 could+0.024
iled+0.024
 they+0.023
Neg logit contributions
 city-0.067
 hotel-0.067
 orchestra-0.067
 talent-0.066
 airport-0.066
Token mel
Token ID7758
Feature activation-2.10e+00

Pos logit contributions
 orchestra+0.340
 opera+0.326
 former+0.324
 hotel+0.320
 apartment+0.308
Neg logit contributions
 could-0.180
es-0.179
icked-0.179
o-0.178
cr-0.175
Tokenon
Token ID261
Feature activation+1.03e+00

Pos logit contributions
 city+0.046
 mirror+0.045
 talent+0.045
 museum+0.044
 opera+0.044
Neg logit contributions
icked-0.031
es-0.029
owed-0.022
ched-0.022
 could-0.021
Token."
Token ID526
Feature activation+2.91e+00

Pos logit contributions
icked+0.014
es+0.013
 could+0.009
iled+0.009
ched+0.009
Neg logit contributions
 city-0.025
 talent-0.024
 opera-0.024
 hotel-0.024
 orchestra-0.024
Token Tom
Token ID4186
Feature activation+1.85e+00

Pos logit contributions
es+0.043
icked+0.042
owed+0.032
iled+0.031
ched+0.030
Neg logit contributions
 city-0.069
 mirror-0.067
 talent-0.067
 hotel-0.066
 opera-0.066
Token said
Token ID531
Feature activation+1.89e+00

Pos logit contributions
icked+0.030
es+0.029
iled+0.023
ched+0.021
owed+0.021
Neg logit contributions
 mirror-0.037
 opera-0.037
 orchestra-0.037
 talent-0.037
 city-0.036
Token was
Token ID373
Feature activation-2.48e+00

Pos logit contributions
es+0.059
icked+0.058
 could+0.044
iled+0.041
owed+0.041
Neg logit contributions
 city-0.094
 talent-0.093
 orchestra-0.093
 hotel-0.092
 opera-0.092
Token loud
Token ID7812
Feature activation+2.33e+00

Pos logit contributions
 city+0.046
 talent+0.046
 orchestra+0.046
 hotel+0.045
 mirror+0.045
Neg logit contributions
icked-0.046
es-0.045
 could-0.038
iled-0.036
 they-0.035
Token and
Token ID290
Feature activation-1.65e+00

Pos logit contributions
icked+0.040
es+0.036
owed+0.029
 could+0.028
iled+0.028
Neg logit contributions
 city-0.050
 opera-0.050
 orchestra-0.048
 airport-0.048
 hotel-0.048
Token sm
Token ID895
Feature activation-1.21e-02

Pos logit contributions
 city+0.023
 talent+0.023
 orchestra+0.022
 hotel+0.022
 opera+0.022
Neg logit contributions
es-0.038
icked-0.038
 could-0.033
 they-0.032
iled-0.031
Tokenelly
Token ID6148
Feature activation+1.04e+00

Pos logit contributions
 city+0.000
 opera+0.000
 talent+0.000
 hotel+0.000
 orchestra+0.000
Neg logit contributions
es-0.000
icked-0.000
 could-0.000
owed-0.000
iled-0.000
Token.
Token ID13
Feature activation+4.40e+00

Pos logit contributions
icked+0.016
es+0.015
ched+0.012
owed+0.012
iled+0.011
Neg logit contributions
 city-0.023
 opera-0.023
 hotel-0.023
 airport-0.022
 mirror-0.022
Token There
Token ID1318
Feature activation+3.28e+00

Pos logit contributions
es+0.056
icked+0.055
owed+0.038
iled+0.038
ched+0.036
Neg logit contributions
 city-0.111
 hotel-0.108
 opera-0.108
 orchestra-0.108
 talent-0.107
Token were
Token ID547
Feature activation-8.79e-01

Pos logit contributions
icked+0.055
es+0.050
iled+0.043
ched+0.040
owed+0.039
Neg logit contributions
 city-0.071
 hotel-0.070
 airport-0.068
 opera-0.067
 museum-0.065
Token machines
Token ID8217
Feature activation+3.72e+00

Pos logit contributions
 orchestra+0.016
 talent+0.016
 city+0.015
 mirror+0.015
 hotel+0.015
Neg logit contributions
es-0.017
icked-0.017
 could-0.014
 they-0.014
 neither-0.013
Token that
Token ID326
Feature activation+1.73e+00

Pos logit contributions
icked+0.050
es+0.049
owed+0.034
iled+0.033
 could+0.033
Neg logit contributions
 city-0.090
 opera-0.088
 hotel-0.088
 airport-0.088
 talent-0.087
Token moved
Token ID3888
Feature activation+2.71e+00

Pos logit contributions
es+0.026
icked+0.026
iled+0.019
owed+0.019
 could+0.019
Neg logit contributions
 city-0.039
 hotel-0.038
 opera-0.038
 talent-0.037
 orchestra-0.037
Token cat
Token ID3797
Feature activation+5.33e+00

Pos logit contributions
 former+0.217
 orchestra+0.215
 cockpit+0.210
 talent+0.206
 priest+0.205
Neg logit contributions
es-0.144
 could-0.129
 on-0.128
 Buddy-0.127
icked-0.127
Token wanted
Token ID2227
Feature activation+2.31e+00

Pos logit contributions
icked+0.091
es+0.084
iled+0.066
owed+0.065
uts+0.064
Neg logit contributions
 city-0.107
 hotel-0.105
 mirror-0.105
 opera-0.104
 airport-0.104
Token to
Token ID284
Feature activation-1.12e+00

Pos logit contributions
icked+0.030
es+0.030
iled+0.021
owed+0.021
ched+0.019
Neg logit contributions
 city-0.057
 hotel-0.056
 opera-0.056
 talent-0.056
 mirror-0.056
Token go
Token ID467
Feature activation+1.49e+00

Pos logit contributions
 city+0.020
 hotel+0.019
 talent+0.019
 opera+0.019
 orchestra+0.019
Neg logit contributions
es-0.023
icked-0.023
 could-0.018
iled-0.018
owed-0.018
Token home
Token ID1363
Feature activation+3.43e+00

Pos logit contributions
icked+0.031
es+0.031
owed+0.024
iled+0.024
 could+0.023
Neg logit contributions
 hotel-0.027
 city-0.027
 opera-0.026
 mirror-0.026
 talent-0.026
Token.
Token ID13
Feature activation+4.77e+00

Pos logit contributions
icked+0.048
es+0.045
ched+0.032
iled+0.032
 could+0.031
Neg logit contributions
 city-0.083
 hotel-0.083
 opera-0.082
 talent-0.081
 airport-0.081
Token But
Token ID887
Feature activation+2.95e+00

Pos logit contributions
icked+0.088
es+0.088
owed+0.070
ppers+0.068
ched+0.067
Neg logit contributions
 city-0.093
 playground-0.088
 park-0.088
 mirror-0.087
 classroom-0.086
Token the
Token ID262
Feature activation-1.08e+01

Pos logit contributions
icked+0.060
es+0.056
uts+0.047
iled+0.046
ched+0.046
Neg logit contributions
 city-0.055
 airport-0.053
 opera-0.053
 hotel-0.053
 park-0.052
Token cat
Token ID3797
Feature activation+5.28e+00

Pos logit contributions
 orchestra+0.221
 former+0.217
 talent+0.211
 hotel+0.211
 city+0.211
Neg logit contributions
es-0.162
icked-0.145
 could-0.142
 on-0.137
 Remy-0.136
Token did
Token ID750
Feature activation+2.13e+00

Pos logit contributions
icked+0.091
es+0.083
uts+0.066
iled+0.065
hot+0.063
Neg logit contributions
 park-0.100
 city-0.098
 hotel-0.097
 mirror-0.097
 bus-0.097
Token not
Token ID407
Feature activation-2.01e+00

Pos logit contributions
icked+0.039
es+0.036
ed+0.029
o+0.028
iled+0.028
Neg logit contributions
 mistake-0.044
 opera-0.041
 talent-0.041
 city-0.041
 hotel-0.040
Token table
Token ID3084
Feature activation+5.42e+00

Pos logit contributions
 orchestra+0.075
 tant+0.072
 talent+0.071
 city+0.068
 opera+0.067
Neg logit contributions
es-0.150
icked-0.141
 could-0.140
 on-0.137
 in-0.133
Token just
Token ID655
Feature activation+1.67e+00

Pos logit contributions
icked+0.071
es+0.071
iled+0.048
 could+0.048
owed+0.047
Neg logit contributions
 city-0.132
 hotel-0.130
 talent-0.130
 opera-0.129
 orchestra-0.129
Token like
Token ID588
Feature activation+1.62e+00

Pos logit contributions
es+0.024
icked+0.024
iled+0.017
 could+0.017
owed+0.017
Neg logit contributions
 city-0.038
 opera-0.038
 talent-0.038
 hotel-0.038
 orchestra-0.038
Token his
Token ID465
Feature activation-6.37e+00

Pos logit contributions
es+0.026
icked+0.025
 could+0.019
iled+0.019
 they+0.018
Neg logit contributions
 city-0.035
 orchestra-0.034
 hotel-0.034
 talent-0.034
 opera-0.034
Token friends
Token ID2460
Feature activation+4.35e+00

Pos logit contributions
 orchestra+0.090
 city+0.089
 opera+0.086
 airport+0.085
 hotel+0.084
Neg logit contributions
es-0.144
icked-0.139
 could-0.123
 they-0.117
owed-0.115
Token.
Token ID13
Feature activation+5.24e+00

Pos logit contributions
icked+0.065
es+0.062
iled+0.045
owed+0.045
ched+0.044
Neg logit contributions
 city-0.100
 mirror-0.098
 hotel-0.097
 talent-0.097
 opera-0.097
Token<|endoftext|>
Token ID50256
Feature activation+3.60e+00

Pos logit contributions
icked+0.078
es+0.078
owed+0.057
iled+0.055
ched+0.055
Neg logit contributions
 city-0.122
 hotel-0.117
 orchestra-0.116
 mirror-0.116
 airport-0.116
TokenOnce
Token ID7454
Feature activation-1.16e+00

Pos logit contributions
icked+0.068
es+0.060
owed+0.055
 neither+0.053
ppers+0.052
Neg logit contributions
 city-0.071
 park-0.070
 museum-0.068
 hotel-0.067
 mirror-0.067
Token,
Token ID11
Feature activation+6.06e-02

Pos logit contributions
 city+0.019
 hotel+0.018
 mirror+0.018
 talent+0.018
 opera+0.018
Neg logit contributions
es-0.025
icked-0.025
iled-0.020
ched-0.020
owed-0.020
Token there
Token ID612
Feature activation+1.80e+00

Pos logit contributions
es+0.001
icked+0.001
uts+0.001
ched+0.001
owed+0.001
Neg logit contributions
 city-0.001
 mirror-0.001
 hotel-0.001
 opera-0.001
 airport-0.001
Token were
Token ID547
Feature activation-3.27e+00

Pos logit contributions
icked+0.032
es+0.030
iled+0.024
owed+0.024
uts+0.023
Neg logit contributions
 city-0.038
 hotel-0.037
 park-0.036
 mirror-0.036
 playground-0.036
Token sand
Token ID6450
Feature activation+2.93e+00

Pos logit contributions
 orchestra+0.104
 former+0.102
reating+0.097
 mistake+0.097
 talent+0.096
Neg logit contributions
 could-0.132
icked-0.132
 on-0.128
 in-0.122
 onto-0.120
Token and
Token ID290
Feature activation+7.38e-01

Pos logit contributions
es+0.039
icked+0.039
 could+0.027
iled+0.026
owed+0.026
Neg logit contributions
 city-0.071
 talent-0.070
 opera-0.070
 orchestra-0.070
 hotel-0.070
Token ate
Token ID15063
Feature activation+2.23e+00

Pos logit contributions
es+0.012
icked+0.012
iled+0.009
 could+0.009
owed+0.009
Neg logit contributions
 city-0.016
 hotel-0.016
 opera-0.016
 talent-0.016
 orchestra-0.016
Token y
Token ID331
Feature activation-9.79e-01

Pos logit contributions
es+0.041
icked+0.040
iled+0.031
 could+0.031
owed+0.031
Neg logit contributions
 city-0.044
 hotel-0.043
 talent-0.043
 opera-0.042
 orchestra-0.042
Tokenummy
Token ID13513
Feature activation-3.95e+00

Pos logit contributions
 city+0.016
 talent+0.016
 hotel+0.016
 mirror+0.015
 opera+0.015
Neg logit contributions
es-0.021
icked-0.020
 could-0.017
owed-0.016
ched-0.016
Token snacks
Token ID28487
Feature activation+5.70e+00

Pos logit contributions
 city+0.054
 hotel+0.053
 talent+0.052
 opera+0.052
 orchestra+0.052
Neg logit contributions
es-0.094
icked-0.094
 could-0.077
iled-0.077
owed-0.077
Token.
Token ID13
Feature activation+4.74e+00

Pos logit contributions
icked+0.102
es+0.093
owed+0.075
ched+0.075
umbered+0.075
Neg logit contributions
 beach-0.108
 hotel-0.107
 city-0.107
 playground-0.105
 opera-0.104
Token They
Token ID1119
Feature activation+1.24e+00

Pos logit contributions
es+0.074
icked+0.073
owed+0.056
iled+0.052
ched+0.052
Neg logit contributions
 city-0.108
 playground-0.103
 hotel-0.102
 classroom-0.101
 orchestra-0.100
Token even
Token ID772
Feature activation-9.20e-01

Pos logit contributions
es+0.017
icked+0.016
 could+0.012
iled+0.011
owed+0.011
Neg logit contributions
 city-0.030
 talent-0.030
 opera-0.030
 orchestra-0.030
 hotel-0.030
Token saw
Token ID2497
Feature activation+1.72e+00

Pos logit contributions
 city+0.023
 orchestra+0.022
 talent+0.022
 opera+0.022
 hotel+0.022
Neg logit contributions
es-0.012
icked-0.012
 could-0.009
owed-0.008
iled-0.008
Token a
Token ID257
Feature activation-8.85e+00

Pos logit contributions
es+0.030
icked+0.029
 could+0.024
 they+0.023
ed+0.022
Neg logit contributions
 orchestra-0.036
 city-0.035
 talent-0.035
 former-0.034
 hotel-0.034
Token upon
Token ID2402
Feature activation+2.25e+00

Pos logit contributions
icked+0.067
es+0.066
iled+0.049
owed+0.049
 Remy+0.046
Neg logit contributions
 city-0.084
 mirror-0.084
 hotel-0.082
 airport-0.082
 opera-0.082
Token a
Token ID257
Feature activation-8.72e+00

Pos logit contributions
es+0.028
icked+0.027
 could+0.021
 they+0.020
ched+0.019
Neg logit contributions
 orchestra-0.056
 city-0.055
 hotel-0.054
 talent-0.054
 tant-0.054
Token man
Token ID582
Feature activation+3.18e+00

Pos logit contributions
 former+0.126
 orchestra+0.121
 opera+0.116
 talent+0.114
 alphabet+0.111
Neg logit contributions
es-0.172
 could-0.171
 on-0.159
 they-0.157
 there-0.154
Token with
Token ID351
Feature activation-1.25e+00

Pos logit contributions
icked+0.051
es+0.050
 could+0.037
iled+0.036
owed+0.036
Neg logit contributions
 city-0.069
 opera-0.068
 hotel-0.067
 mirror-0.067
 orchestra-0.067
Token a
Token ID257
Feature activation-8.34e+00

Pos logit contributions
 hotel+0.030
 city+0.030
 orchestra+0.030
 former+0.029
 talent+0.029
Neg logit contributions
icked-0.018
es-0.018
 could-0.013
ched-0.012
ed-0.012
Token bag
Token ID6131
Feature activation+4.70e+00

Pos logit contributions
 former+0.122
 opera+0.122
 orchestra+0.117
 hotel+0.113
 city+0.113
Neg logit contributions
es-0.174
icked-0.161
 could-0.161
 on-0.155
iled-0.150
Token full
Token ID1336
Feature activation-2.11e+00

Pos logit contributions
es+0.075
icked+0.075
 Remy+0.058
owed+0.054
 Carl+0.053
Neg logit contributions
 city-0.098
 airport-0.097
 mirror-0.097
 bus-0.095
 opera-0.094
Token of
Token ID286
Feature activation-7.70e+00

Pos logit contributions
 city+0.052
 orchestra+0.051
 talent+0.051
 hotel+0.051
 opera+0.051
Neg logit contributions
icked-0.027
es-0.027
 could-0.019
iled-0.018
owed-0.018
Token things
Token ID1243
Feature activation+3.33e+00

Pos logit contributions
 orchestra+0.128
 former+0.126
 city+0.123
 airport+0.120
 opera+0.119
Neg logit contributions
es-0.152
icked-0.145
 could-0.128
 on-0.126
 they-0.125
Token.
Token ID13
Feature activation+4.41e+00

Pos logit contributions
icked+0.050
es+0.050
 Remy+0.037
owed+0.037
iled+0.036
Neg logit contributions
 talent-0.073
 mirror-0.072
 opera-0.072
 city-0.072
 hotel-0.070
Token The
Token ID383
Feature activation-1.07e+01

Pos logit contributions
icked+0.071
es+0.069
owed+0.052
iled+0.051
ched+0.050
Neg logit contributions
 city-0.099
 mirror-0.096
 hotel-0.095
 playground-0.094
 museum-0.094
TokenThe
Token ID464
Feature activation-1.23e+01

Pos logit contributions
icked+0.001
es+0.001
owed+0.000
iled+0.000
 could+0.000
Neg logit contributions
 city-0.001
 opera-0.001
 talent-0.001
 hotel-0.001
 orchestra-0.001
Token cat
Token ID3797
Feature activation+3.44e+00

Pos logit contributions
 former+0.247
 tant+0.223
 cockpit+0.220
 prope+0.217
 reality+0.214
Neg logit contributions
es-0.209
 they-0.197
 could-0.195
 on-0.194
 Sam-0.188
Token and
Token ID290
Feature activation-4.20e+00

Pos logit contributions
icked+0.061
es+0.057
iled+0.046
owed+0.045
uts+0.043
Neg logit contributions
 city-0.071
 opera-0.069
 airport-0.069
 hotel-0.068
 park-0.068
Token the
Token ID262
Feature activation-1.29e+01

Pos logit contributions
 former+0.106
 hotel+0.105
 city+0.104
 talent+0.104
 orchestra+0.103
Neg logit contributions
es-0.052
icked-0.050
 could-0.037
 they-0.034
 Remy-0.034
Token kids
Token ID3988
Feature activation+2.45e+00

Pos logit contributions
 former+0.246
 cockpit+0.221
 prope+0.214
 airport+0.211
 reality+0.209
Neg logit contributions
es-0.236
 could-0.225
 Sam-0.223
 they-0.217
 on-0.214
Token played
Token ID2826
Feature activation+1.67e-01

Pos logit contributions
es+0.038
icked+0.038
iled+0.028
 Remy+0.028
ched+0.027
Neg logit contributions
 city-0.054
 opera-0.053
 hotel-0.052
 orchestra-0.052
 airport-0.052
Token all
Token ID477
Feature activation-2.85e+00

Pos logit contributions
icked+0.003
es+0.003
iled+0.002
owed+0.002
 could+0.002
Neg logit contributions
 city-0.004
 talent-0.004
 opera-0.004
 hotel-0.004
 orchestra-0.004
Token day
Token ID1110
Feature activation-2.32e-02

Pos logit contributions
 city+0.040
 hotel+0.039
 orchestra+0.039
 airport+0.038
 talent+0.038
Neg logit contributions
es-0.067
icked-0.066
 could-0.056
iled-0.055
owed-0.054
Token.
Token ID13
Feature activation+2.92e+00

Pos logit contributions
 city+0.001
 talent+0.001
 hotel+0.000
 opera+0.000
 orchestra+0.000
Neg logit contributions
icked-0.000
es-0.000
owed-0.000
ched-0.000
iled-0.000
Token They
Token ID1119
Feature activation+6.59e-02

Pos logit contributions
es+0.042
icked+0.040
owed+0.031
iled+0.030
ched+0.029
Neg logit contributions
 city-0.071
 opera-0.068
 hotel-0.067
 talent-0.067
 orchestra-0.067
Token had
Token ID550
Feature activation-1.76e+00

Pos logit contributions
es+0.001
icked+0.001
iled+0.001
owed+0.001
ched+0.001
Neg logit contributions
 city-0.002
 hotel-0.001
 opera-0.001
 talent-0.001
 orchestra-0.001
Token.
Token ID13
Feature activation+2.94e+00

Pos logit contributions
 city+0.001
 opera+0.001
 talent+0.001
 hotel+0.001
 mirror+0.001
Neg logit contributions
icked-0.001
es-0.001
ched-0.001
owed-0.001
iled-0.001
Token
Token ID198
Feature activation+8.69e-01

Pos logit contributions
es+0.042
icked+0.040
owed+0.030
ched+0.028
iled+0.028
Neg logit contributions
 city-0.073
 hotel-0.070
 talent-0.070
 opera-0.069
 airport-0.069
Token
Token ID198
Feature activation+1.14e+00

Pos logit contributions
es+0.015
icked+0.015
owed+0.011
iled+0.011
 could+0.011
Neg logit contributions
 city-0.018
 hotel-0.017
 talent-0.017
 opera-0.017
 orchestra-0.017
Token"
Token ID1
Feature activation+1.95e-01

Pos logit contributions
icked+0.017
es+0.017
owed+0.012
 could+0.012
iled+0.011
Neg logit contributions
 city-0.027
 opera-0.026
 talent-0.026
 hotel-0.026
 orchestra-0.026
TokenLet
Token ID5756
Feature activation+1.99e+00

Pos logit contributions
es+0.004
icked+0.004
 could+0.003
iled+0.003
ched+0.003
Neg logit contributions
 city-0.004
 hotel-0.004
 talent-0.004
 orchestra-0.004
 opera-0.004
Token's
Token ID338
Feature activation-1.66e+00

Pos logit contributions
icked+0.033
es+0.032
owed+0.025
iled+0.025
uts+0.023
Neg logit contributions
 city-0.043
 opera-0.043
 airport-0.042
 hotel-0.042
 mirror-0.042
Token go
Token ID467
Feature activation+1.66e+00

Pos logit contributions
 orchestra+0.032
 city+0.032
 talent+0.032
 hotel+0.032
 opera+0.031
Neg logit contributions
icked-0.029
es-0.029
 could-0.023
iled-0.022
owed-0.022
Token back
Token ID736
Feature activation+1.35e+00

Pos logit contributions
icked+0.034
es+0.034
iled+0.026
ched+0.026
owed+0.026
Neg logit contributions
 city-0.030
 hotel-0.029
 opera-0.028
 airport-0.028
 mirror-0.028
Token to
Token ID284
Feature activation-1.14e+00

Pos logit contributions
icked+0.022
es+0.021
ched+0.016
iled+0.016
owed+0.016
Neg logit contributions
 city-0.031
 hotel-0.030
 opera-0.030
 airport-0.029
 talent-0.029
Token the
Token ID262
Feature activation-1.17e+01

Pos logit contributions
 tant+0.027
 orchestra+0.027
 talent+0.027
 city+0.027
 mirror+0.027
Neg logit contributions
es-0.014
icked-0.014
 Remy-0.010
 they-0.010
 could-0.010
Token pond
Token ID16723
Feature activation+3.73e+00

Pos logit contributions
 tant+0.196
 orchestra+0.162
 channels+0.151
 actor+0.147
 former+0.147
Neg logit contributions
 they-0.245
 we-0.225
 there-0.223
 on-0.221
 Remy-0.216
Token She
Token ID1375
Feature activation+9.39e-01

Pos logit contributions
es+0.040
icked+0.040
iled+0.027
owed+0.027
ched+0.026
Neg logit contributions
 city-0.081
 opera-0.079
 talent-0.079
 hotel-0.079
 orchestra-0.079
Token praised
Token ID15342
Feature activation+3.19e+00

Pos logit contributions
es+0.012
icked+0.012
 could+0.009
iled+0.008
owed+0.008
Neg logit contributions
 city-0.024
 hotel-0.023
 orchestra-0.023
 opera-0.023
 talent-0.023
Token them
Token ID606
Feature activation+1.31e+00

Pos logit contributions
es+0.049
icked+0.047
 they+0.034
 could+0.034
owed+0.033
Neg logit contributions
 city-0.073
 hotel-0.072
 orchestra-0.071
 airport-0.071
 talent-0.071
Token and
Token ID290
Feature activation-4.61e-01

Pos logit contributions
es+0.016
icked+0.015
 could+0.010
iled+0.010
owed+0.010
Neg logit contributions
 city-0.034
 hotel-0.033
 talent-0.033
 orchestra-0.033
 opera-0.033
Token helped
Token ID4193
Feature activation+2.46e+00

Pos logit contributions
 city+0.011
 hotel+0.011
 airport+0.010
 orchestra+0.010
 opera+0.010
Neg logit contributions
es-0.007
icked-0.007
 could-0.005
 they-0.005
owed-0.005
Token them
Token ID606
Feature activation+7.51e-01

Pos logit contributions
es+0.040
icked+0.039
 could+0.029
owed+0.029
iled+0.029
Neg logit contributions
 city-0.053
 hotel-0.052
 orchestra-0.052
 talent-0.052
 airport-0.051
Token measure
Token ID3953
Feature activation+2.69e+00

Pos logit contributions
icked+0.014
es+0.014
iled+0.011
ched+0.011
 Remy+0.011
Neg logit contributions
 city-0.014
 talent-0.014
 opera-0.013
 mirror-0.013
 hotel-0.013
Token the
Token ID262
Feature activation-1.10e+01

Pos logit contributions
es+0.040
icked+0.039
 could+0.028
iled+0.028
owed+0.028
Neg logit contributions
 city-0.062
 hotel-0.060
 talent-0.060
 orchestra-0.060
 opera-0.060
Token new
Token ID649
Feature activation-9.98e+00

Pos logit contributions
 former+0.144
 priest+0.140
 hotel+0.132
 mall+0.115
 orchestra+0.114
Neg logit contributions
 on-0.243
es-0.240
 in-0.219
 against-0.217
 could-0.214
Token radio
Token ID5243
Feature activation+5.35e+00

Pos logit contributions
 former+0.134
 priest+0.123
 hotel+0.122
 tant+0.111
 mall+0.111
Neg logit contributions
es-0.252
 on-0.210
 they-0.200
icked-0.198
 together-0.197
Token.
Token ID13
Feature activation+3.45e+00

Pos logit contributions
icked+0.078
es+0.073
iled+0.053
 Remy+0.053
ched+0.053
Neg logit contributions
 city-0.125
 opera-0.122
 talent-0.121
 hotel-0.121
 airport-0.120
Token in
Token ID287
Feature activation-2.81e-01

Pos logit contributions
icked+0.046
es+0.045
iled+0.030
owed+0.030
 Remy+0.029
Neg logit contributions
 city-0.091
 opera-0.091
 orchestra-0.090
 talent-0.090
 mirror-0.089
Token front
Token ID2166
Feature activation+1.14e+00

Pos logit contributions
 airport+0.007
 hotel+0.007
 talent+0.007
 orchestra+0.007
 opera+0.007
Neg logit contributions
es-0.003
icked-0.003
 on-0.002
 Fred-0.002
 they-0.002
Token of
Token ID286
Feature activation+6.34e-01

Pos logit contributions
es+0.013
icked+0.012
 could+0.009
iled+0.008
 they+0.008
Neg logit contributions
 city-0.029
 hotel-0.029
 talent-0.029
 former-0.029
 orchestra-0.029
Token the
Token ID262
Feature activation-8.23e+00

Pos logit contributions
es+0.007
icked+0.006
 they+0.005
 could+0.005
 Remy+0.004
Neg logit contributions
 talent-0.016
 orchestra-0.016
 city-0.016
 former-0.016
 hotel-0.016
Token tree
Token ID5509
Feature activation+5.51e+00

Pos logit contributions
 former+0.095
 ey+0.082
 priest+0.082
Ta+0.081
 talent+0.080
Neg logit contributions
es-0.181
 on-0.174
 in-0.169
 could-0.162
 they-0.162
Token and
Token ID290
Feature activation+1.41e+00

Pos logit contributions
icked+0.076
es+0.074
iled+0.051
 could+0.051
ched+0.051
Neg logit contributions
 city-0.132
 hotel-0.129
 opera-0.128
 orchestra-0.128
 talent-0.128
Token quickly
Token ID2952
Feature activation+2.97e+00

Pos logit contributions
es+0.020
icked+0.020
iled+0.014
 could+0.014
owed+0.014
Neg logit contributions
 city-0.033
 hotel-0.032
 opera-0.032
 talent-0.032
 orchestra-0.032
Token the
Token ID262
Feature activation-1.05e+01

Pos logit contributions
es+0.044
icked+0.043
iled+0.031
 could+0.031
owed+0.031
Neg logit contributions
 city-0.067
 hotel-0.066
 talent-0.066
 opera-0.066
 orchestra-0.066
Token cat
Token ID3797
Feature activation+4.99e+00

Pos logit contributions
 former+0.127
 talent+0.117
 ey+0.115
 hotel+0.115
 airport+0.114
Neg logit contributions
es-0.226
 on-0.210
 in-0.204
 they-0.202
 could-0.200
Token came
Token ID1625
Feature activation+2.11e+00

Pos logit contributions
icked+0.073
es+0.069
 Remy+0.050
iled+0.050
owed+0.048
Neg logit contributions
 city-0.112
 opera-0.110
 hotel-0.109
 orchestra-0.109
 airport-0.108
Token down
Token ID866
Feature activation+3.20e+00

Pos logit contributions
icked+0.035
es+0.035
owed+0.026
 could+0.025
iled+0.025
Neg logit contributions
 city-0.044
 opera-0.043
 mirror-0.042
 orchestra-0.042
 hotel-0.042
Token Julia
Token ID22300
Feature activation+1.54e+00

Pos logit contributions
icked+0.021
es+0.020
iled+0.016
o+0.015
ched+0.015
Neg logit contributions
 city-0.031
 opera-0.031
 hotel-0.030
 talent-0.030
 mirror-0.030
Token realised
Token ID19169
Feature activation+1.03e+00

Pos logit contributions
icked+0.028
es+0.026
 Remy+0.021
owed+0.021
iled+0.021
Neg logit contributions
 airport-0.029
 mirror-0.028
 city-0.028
 hotel-0.028
 talent-0.027
Token what
Token ID644
Feature activation-5.09e-01

Pos logit contributions
icked+0.013
es+0.013
iled+0.009
owed+0.009
o+0.009
Neg logit contributions
 opera-0.025
 city-0.025
 mirror-0.025
 airport-0.025
 talent-0.025
Token was
Token ID373
Feature activation-2.38e+00

Pos logit contributions
 talent+0.012
 mirror+0.012
 opera+0.012
 city+0.012
 museum+0.011
Neg logit contributions
icked-0.008
es-0.007
owed-0.006
iled-0.006
ched-0.005
Token missing
Token ID4814
Feature activation+1.85e+00

Pos logit contributions
 opera+0.038
 talent+0.038
 city+0.037
 mirror+0.036
 hotel+0.036
Neg logit contributions
es-0.052
icked-0.052
iled-0.043
owed-0.043
o-0.041
Token -
Token ID532
Feature activation+1.05e+00

Pos logit contributions
icked+0.031
es+0.029
iled+0.024
o+0.023
 Remy+0.023
Neg logit contributions
 mirror-0.039
 opera-0.038
 city-0.038
 hotel-0.037
 airport-0.037
Token a
Token ID257
Feature activation-9.37e+00

Pos logit contributions
es+0.013
icked+0.013
 could+0.009
iled+0.008
owed+0.008
Neg logit contributions
 city-0.027
 hotel-0.026
 talent-0.026
 orchestra-0.026
 opera-0.026
Token friend
Token ID1545
Feature activation+1.80e+00

Pos logit contributions
 orchestra+0.142
 former+0.132
 opera+0.132
 city+0.128
 cockpit+0.128
Neg logit contributions
 could-0.181
es-0.175
 they-0.169
 on-0.166
icked-0.165
Token to
Token ID284
Feature activation-2.82e+00

Pos logit contributions
icked+0.029
es+0.027
owed+0.020
uts+0.020
 Remy+0.019
Neg logit contributions
 city-0.039
 opera-0.039
 mirror-0.039
 airport-0.038
 hotel-0.038
Token share
Token ID2648
Feature activation+1.17e+00

Pos logit contributions
 city+0.052
 hotel+0.050
 orchestra+0.050
 airport+0.050
 opera+0.049
Neg logit contributions
es-0.055
icked-0.054
 could-0.043
iled-0.042
owed-0.042
Token lunch
Token ID9965
Feature activation+2.73e+00

Pos logit contributions
icked+0.018
es+0.018
iled+0.013
 could+0.013
owed+0.012
Neg logit contributions
 city-0.026
 talent-0.026
 opera-0.026
 hotel-0.026
 orchestra-0.026
Token rain
Token ID6290
Feature activation+5.03e+00

Pos logit contributions
 former+0.150
 tant+0.148
 perspective+0.144
 orchestra+0.142
 amaz+0.137
Neg logit contributions
 on-0.174
 they-0.169
 in-0.167
es-0.160
 could-0.159
Token got
Token ID1392
Feature activation-3.49e-01

Pos logit contributions
icked+0.079
es+0.077
 Remy+0.058
iled+0.057
owed+0.054
Neg logit contributions
 city-0.111
 hotel-0.110
 opera-0.109
 airport-0.108
 orchestra-0.106
Token heavier
Token ID20140
Feature activation+5.03e-01

Pos logit contributions
 city+0.006
 talent+0.006
 orchestra+0.006
 hotel+0.006
 opera+0.006
Neg logit contributions
es-0.007
icked-0.007
 could-0.006
iled-0.006
owed-0.006
Token,
Token ID11
Feature activation+1.90e+00

Pos logit contributions
es+0.007
icked+0.007
iled+0.005
 could+0.005
owed+0.005
Neg logit contributions
 city-0.012
 talent-0.012
 hotel-0.012
 orchestra-0.012
 opera-0.012
Token and
Token ID290
Feature activation+8.35e-01

Pos logit contributions
icked+0.028
es+0.028
iled+0.020
owed+0.020
 could+0.020
Neg logit contributions
 city-0.044
 opera-0.043
 hotel-0.042
 airport-0.042
 orchestra-0.042
Token the
Token ID262
Feature activation-9.09e+00

Pos logit contributions
es+0.013
icked+0.013
iled+0.010
owed+0.010
 could+0.010
Neg logit contributions
 city-0.018
 hotel-0.018
 opera-0.018
 talent-0.018
 orchestra-0.018
Token wind
Token ID2344
Feature activation+7.35e+00

Pos logit contributions
 talent+0.118
 orchestra+0.115
 former+0.109
 tant+0.109
 fascination+0.103
Neg logit contributions
es-0.185
icked-0.180
 could-0.179
 they-0.179
 on-0.175
Token got
Token ID1392
Feature activation+3.11e-01

Pos logit contributions
icked+0.125
es+0.122
 Remy+0.104
 Carl+0.093
iled+0.092
Neg logit contributions
 hotel-0.142
 opera-0.140
 city-0.140
 mirror-0.137
 airport-0.137
Token stronger
Token ID7387
Feature activation+1.48e+00

Pos logit contributions
es+0.006
icked+0.006
iled+0.005
owed+0.005
 could+0.005
Neg logit contributions
 city-0.005
 hotel-0.005
 opera-0.005
 talent-0.005
 airport-0.005
Token.
Token ID13
Feature activation+4.88e+00

Pos logit contributions
icked+0.022
es+0.021
owed+0.015
iled+0.014
 could+0.014
Neg logit contributions
 city-0.035
 hotel-0.034
 opera-0.034
 airport-0.034
 talent-0.034
Token The
Token ID383
Feature activation-9.74e+00

Pos logit contributions
es+0.068
icked+0.067
iled+0.047
owed+0.046
 could+0.046
Neg logit contributions
 city-0.118
 hotel-0.115
 opera-0.115
 orchestra-0.114
 talent-0.114
Token the
Token ID262
Feature activation-9.82e+00

Pos logit contributions
es+0.000
icked+0.000
 could+0.000
 they+0.000
iled+0.000
Neg logit contributions
 orchestra-0.000
 city-0.000
 talent-0.000
 opera-0.000
 mirror-0.000
Token island
Token ID7022
Feature activation+3.72e+00

Pos logit contributions
 tant+0.175
 orchestra+0.160
 ey+0.159
 hairst+0.150
 talent+0.149
Neg logit contributions
 could-0.177
 they-0.165
 on-0.159
 in-0.155
 there-0.154
Token.
Token ID13
Feature activation+3.82e+00

Pos logit contributions
es+0.049
icked+0.048
iled+0.033
 could+0.032
owed+0.032
Neg logit contributions
 city-0.092
 hotel-0.090
 opera-0.090
 talent-0.090
 orchestra-0.089
Token There
Token ID1318
Feature activation+3.25e+00

Pos logit contributions
es+0.063
icked+0.056
ppers+0.045
owed+0.045
ched+0.045
Neg logit contributions
 city-0.084
 playground-0.083
 classroom-0.081
 hotel-0.079
 park-0.079
Token were
Token ID547
Feature activation-1.28e+00

Pos logit contributions
icked+0.057
es+0.050
iled+0.045
o+0.042
ched+0.041
Neg logit contributions
 city-0.068
 hotel-0.067
 airport-0.065
 opera-0.064
 park-0.063
Token lots
Token ID6041
Feature activation-4.08e+00

Pos logit contributions
 orchestra+0.024
 talent+0.024
 tant+0.023
 mirror+0.023
 opera+0.023
Neg logit contributions
es-0.023
icked-0.023
 could-0.020
 they-0.019
ed-0.018
Token of
Token ID286
Feature activation-5.74e+00

Pos logit contributions
 former+0.117
 talent+0.114
 mirror+0.114
 city+0.114
 orchestra+0.113
Neg logit contributions
es-0.037
icked-0.029
 they-0.022
 could-0.021
ched-0.017
Token new
Token ID649
Feature activation-6.37e+00

Pos logit contributions
 salon+0.095
 former+0.087
 mirror+0.084
 classroom+0.084
 orchestra+0.084
Neg logit contributions
es-0.109
 could-0.104
 they-0.104
 on-0.104
 there-0.100
Token things
Token ID1243
Feature activation+3.92e+00

Pos logit contributions
 mirror+0.096
 orchestra+0.096
 former+0.092
 tant+0.092
 opera+0.091
Neg logit contributions
es-0.143
icked-0.124
 they-0.117
 could-0.116
ed-0.110
Token to
Token ID284
Feature activation-1.50e+00

Pos logit contributions
icked+0.060
es+0.057
owed+0.041
 Remy+0.041
iled+0.041
Neg logit contributions
 talent-0.088
 opera-0.088
 city-0.087
 hotel-0.086
 mirror-0.086
Token explore
Token ID7301
Feature activation+3.47e+00

Pos logit contributions
 city+0.030
 tant+0.030
 orchestra+0.030
 mirror+0.029
 hotel+0.029
Neg logit contributions
es-0.025
icked-0.024
 could-0.020
 they-0.019
owed-0.018
Token blue
Token ID4171
Feature activation-5.37e+00

Pos logit contributions
 former+0.099
 orchestra+0.090
 talent+0.088
 airport+0.085
 city+0.084
Neg logit contributions
es-0.151
icked-0.141
 they-0.136
iled-0.136
o-0.135
Token frog
Token ID21264
Feature activation+3.21e+00

Pos logit contributions
 former+0.084
 orchestra+0.084
 talent+0.083
 tant+0.078
 city+0.078
Neg logit contributions
es-0.108
icked-0.106
o-0.092
 they-0.091
 could-0.091
Token.
Token ID13
Feature activation+2.44e+00

Pos logit contributions
icked+0.062
es+0.058
bled+0.049
ched+0.048
ed+0.047
Neg logit contributions
 talent-0.057
 city-0.056
 opera-0.056
 zoo-0.054
 park-0.054
Token He
Token ID679
Feature activation-3.28e-01

Pos logit contributions
es+0.047
icked+0.045
ched+0.038
owed+0.037
ppers+0.036
Neg logit contributions
 city-0.049
 museum-0.047
 zoo-0.045
 park-0.045
 universe-0.045
Token was
Token ID373
Feature activation-3.22e+00

Pos logit contributions
 talent+0.005
 city+0.005
 hotel+0.005
 museum+0.005
 opera+0.005
Neg logit contributions
icked-0.007
es-0.006
owed-0.005
 neither-0.005
uts-0.005
Token always
Token ID1464
Feature activation-3.10e+00

Pos logit contributions
 city+0.049
 opera+0.048
 talent+0.048
 hotel+0.047
 mirror+0.047
Neg logit contributions
es-0.073
icked-0.071
owed-0.059
iled-0.059
 could-0.058
Token very
Token ID845
Feature activation-5.18e+00

Pos logit contributions
 talent+0.049
 opera+0.048
 city+0.048
 mirror+0.048
 orchestra+0.047
Neg logit contributions
icked-0.068
es-0.068
owed-0.055
ched-0.054
 could-0.053
Token dirty
Token ID11841
Feature activation+1.17e+00

Pos logit contributions
 city+0.073
 orchestra+0.070
 hotel+0.070
 airport+0.070
 opera+0.070
Neg logit contributions
icked-0.123
es-0.123
 could-0.102
iled-0.102
owed-0.100
Token and
Token ID290
Feature activation-1.19e-01

Pos logit contributions
icked+0.019
es+0.017
owed+0.013
iled+0.013
ched+0.013
Neg logit contributions
 city-0.026
 hotel-0.026
 opera-0.025
 mirror-0.025
 airport-0.025
Token messy
Token ID24097
Feature activation+5.86e-01

Pos logit contributions
 city+0.002
 hotel+0.002
 mirror+0.002
 talent+0.002
 opera+0.002
Neg logit contributions
es-0.002
icked-0.002
iled-0.002
owed-0.002
ched-0.002
Token.
Token ID13
Feature activation+3.49e+00

Pos logit contributions
icked+0.010
es+0.009
iled+0.008
owed+0.007
ver+0.007
Neg logit contributions
 hotel-0.012
 city-0.012
 opera-0.012
 mirror-0.012
 rooms-0.012
Token Joe
Token ID5689
Feature activation+1.31e+00

Pos logit contributions
 city+0.103
 orchestra+0.103
 talent+0.103
 opera+0.102
 hotel+0.102
Neg logit contributions
es-0.063
icked-0.061
iled-0.045
 could-0.045
 Remy-0.043
Token decided
Token ID3066
Feature activation+2.25e+00

Pos logit contributions
icked+0.023
es+0.023
uts+0.018
owed+0.018
othe+0.018
Neg logit contributions
 city-0.027
 hotel-0.026
 park-0.026
 airport-0.026
 mirror-0.025
Token to
Token ID284
Feature activation-1.23e+00

Pos logit contributions
icked+0.029
es+0.029
owed+0.019
iled+0.019
ched+0.017
Neg logit contributions
 city-0.058
 talent-0.057
 opera-0.056
 airport-0.056
 hotel-0.056
Token wander
Token ID27776
Feature activation+1.49e+00

Pos logit contributions
 city+0.021
 hotel+0.021
 opera+0.020
 talent+0.020
 airport+0.020
Neg logit contributions
icked-0.026
es-0.026
iled-0.021
owed-0.021
ed-0.020
Token around
Token ID1088
Feature activation+1.22e+00

Pos logit contributions
es+0.019
icked+0.018
 could+0.013
ched+0.012
iled+0.012
Neg logit contributions
 city-0.037
 orchestra-0.036
 hotel-0.036
 talent-0.036
 opera-0.036
Token the
Token ID262
Feature activation-8.35e+00

Pos logit contributions
es+0.017
icked+0.017
 could+0.012
iled+0.012
owed+0.012
Neg logit contributions
 city-0.028
 hotel-0.028
 orchestra-0.028
 talent-0.028
 opera-0.028
Token house
Token ID2156
Feature activation+4.70e+00

Pos logit contributions
 tant+0.129
 apology+0.117
 thrill+0.106
 humor+0.096
 orchestra+0.095
Neg logit contributions
 there-0.175
 with-0.170
 in-0.169
 could-0.168
es-0.162
Token.
Token ID13
Feature activation+3.02e+00

Pos logit contributions
icked+0.066
es+0.064
iled+0.045
ched+0.043
owed+0.043
Neg logit contributions
 city-0.114
 hotel-0.112
 opera-0.111
 airport-0.111
 talent-0.110
Token He
Token ID679
Feature activation+1.11e-01

Pos logit contributions
es+0.052
icked+0.049
ched+0.037
owed+0.035
iled+0.035
Neg logit contributions
 city-0.070
 playground-0.066
 hotel-0.066
 museum-0.065
 beach-0.064
Token went
Token ID1816
Feature activation+8.37e-01

Pos logit contributions
es+0.002
icked+0.002
iled+0.001
 could+0.001
owed+0.001
Neg logit contributions
 city-0.002
 hotel-0.002
 talent-0.002
 opera-0.002
 orchestra-0.002
Token up
Token ID510
Feature activation+3.44e-01

Pos logit contributions
es+0.011
icked+0.011
iled+0.007
 could+0.007
owed+0.007
Neg logit contributions
 city-0.021
 hotel-0.020
 opera-0.020
 talent-0.020
 orchestra-0.020
Token clearing
Token ID17304
Feature activation+1.19e+00

Pos logit contributions
 orchestra+0.145
 opera+0.138
 former+0.135
 tant+0.134
 talent+0.127
Neg logit contributions
 could-0.191
es-0.185
 there-0.180
 on-0.177
 they-0.175
Token.
Token ID13
Feature activation+3.46e+00

Pos logit contributions
icked+0.019
es+0.018
 Remy+0.014
ched+0.013
iled+0.013
Neg logit contributions
 city-0.027
 hotel-0.026
 opera-0.026
 airport-0.026
 mirror-0.025
Token There
Token ID1318
Feature activation+2.52e+00

Pos logit contributions
es+0.054
icked+0.052
ppers+0.042
ched+0.040
iled+0.040
Neg logit contributions
 city-0.075
 playground-0.072
 zoo-0.070
 bus-0.069
 park-0.069
Token were
Token ID547
Feature activation-2.58e+00

Pos logit contributions
icked+0.049
iled+0.041
es+0.040
ppers+0.040
uts+0.037
Neg logit contributions
 city-0.045
 park-0.044
 airport-0.043
 hotel-0.043
 new-0.042
Token lots
Token ID6041
Feature activation-5.16e+00

Pos logit contributions
 orchestra+0.045
 talent+0.044
 tant+0.043
 opera+0.043
 city+0.043
Neg logit contributions
es-0.052
icked-0.051
 could-0.044
 they-0.041
ed-0.040
Token of
Token ID286
Feature activation-7.35e+00

Pos logit contributions
 former+0.139
 talent+0.138
 city+0.138
 orchestra+0.138
 airport+0.137
Neg logit contributions
es-0.056
icked-0.048
 could-0.034
iled-0.031
 they-0.030
Token tasty
Token ID25103
Feature activation-6.83e+00

Pos logit contributions
 orchestra+0.111
 tant+0.108
 former+0.107
 priest+0.107
 airport+0.106
Neg logit contributions
es-0.162
icked-0.146
 could-0.139
 on-0.132
 Remy-0.130
Token fruits
Token ID15921
Feature activation+2.78e+00

Pos logit contributions
 city+0.104
 hotel+0.101
 talent+0.100
 opera+0.100
 orchestra+0.098
Neg logit contributions
icked-0.154
es-0.153
 could-0.125
iled-0.125
owed-0.125
Token and
Token ID290
Feature activation-1.61e+00

Pos logit contributions
icked+0.053
es+0.046
owed+0.042
ched+0.039
iled+0.038
Neg logit contributions
 park-0.052
 city-0.052
 beach-0.051
 playground-0.050
 airport-0.049
Token nuts
Token ID14380
Feature activation+3.31e+00

Pos logit contributions
 city+0.025
 orchestra+0.025
 talent+0.025
 opera+0.025
 hotel+0.025
Neg logit contributions
es-0.035
icked-0.035
 could-0.029
iled-0.028
owed-0.028
Token scattered
Token ID16830
Feature activation+1.19e+00

Pos logit contributions
icked+0.055
es+0.051
 Remy+0.042
owed+0.041
iled+0.041
Neg logit contributions
 park-0.063
 city-0.063
 beach-0.063
 playground-0.062
 hotel-0.062
Token fly
Token ID6129
Feature activation+1.23e+00

Pos logit contributions
 city+0.045
 orchestra+0.044
 hotel+0.043
 talent+0.043
 opera+0.043
Neg logit contributions
es-0.051
icked-0.050
 could-0.040
iled-0.040
owed-0.039
Token.
Token ID13
Feature activation+3.39e+00

Pos logit contributions
icked+0.019
es+0.019
iled+0.014
owed+0.013
 could+0.013
Neg logit contributions
 city-0.028
 talent-0.027
 opera-0.027
 hotel-0.027
 orchestra-0.027
Token The
Token ID383
Feature activation-1.21e+01

Pos logit contributions
es+0.050
icked+0.049
iled+0.035
ched+0.034
owed+0.033
Neg logit contributions
 city-0.081
 opera-0.078
 airport-0.078
 talent-0.078
 orchestra-0.077
Token girl
Token ID2576
Feature activation+1.89e+00

Pos logit contributions
 orchestra+0.213
 tant+0.205
 classrooms+0.202
 apartment+0.201
 channels+0.199
Neg logit contributions
 could-0.211
 there-0.209
 they-0.205
 in-0.200
icked-0.199
Token reminded
Token ID14516
Feature activation+1.06e+00

Pos logit contributions
icked+0.031
es+0.030
uts+0.023
iled+0.023
hot+0.022
Neg logit contributions
 city-0.039
 opera-0.038
 airport-0.038
 hotel-0.038
 mirror-0.037
Token the
Token ID262
Feature activation-1.16e+01

Pos logit contributions
es+0.017
icked+0.017
 could+0.012
iled+0.012
owed+0.012
Neg logit contributions
 city-0.023
 hotel-0.023
 talent-0.023
 orchestra-0.023
 opera-0.023
Token little
Token ID1310
Feature activation-7.19e+00

Pos logit contributions
 tant+0.193
 orchestra+0.191
 apartment+0.182
 classrooms+0.181
 former+0.178
Neg logit contributions
 could-0.211
icked-0.210
 there-0.205
es-0.204
 they-0.194
Token bird
Token ID6512
Feature activation+2.86e+00

Pos logit contributions
 orchestra+0.108
 former+0.108
 talent+0.107
 tant+0.102
 city+0.101
Neg logit contributions
icked-0.148
es-0.147
 could-0.138
 they-0.133
 there-0.130
Token that
Token ID326
Feature activation+1.15e-01

Pos logit contributions
icked+0.039
es+0.039
iled+0.028
owed+0.026
ched+0.026
Neg logit contributions
 opera-0.067
 talent-0.067
 city-0.067
 hotel-0.066
 mirror-0.065
Token its
Token ID663
Feature activation-1.21e+01

Pos logit contributions
icked+0.002
es+0.002
iled+0.001
owed+0.001
ched+0.001
Neg logit contributions
 city-0.003
 opera-0.003
 hotel-0.003
 airport-0.003
 talent-0.003
Token wing
Token ID8539
Feature activation+3.69e+00

Pos logit contributions
 orchestra+0.228
 hotel+0.221
 museum+0.218
 city+0.217
 opera+0.214
Neg logit contributions
icked-0.218
es-0.209
 could-0.184
 Remy-0.170
 they-0.170
Token the
Token ID262
Feature activation-9.57e+00

Pos logit contributions
 former+0.026
 orchestra+0.026
 talent+0.026
 mirror+0.025
 playground+0.025
Neg logit contributions
es-0.013
icked-0.012
 could-0.010
 they-0.009
iled-0.008
Token house
Token ID2156
Feature activation+5.76e+00

Pos logit contributions
 former+0.110
 orchestra+0.110
 circus+0.108
 amusement+0.102
 apology+0.102
Neg logit contributions
es-0.223
 could-0.219
 there-0.210
 on-0.205
 in-0.202
Token were
Token ID547
Feature activation-3.01e+00

Pos logit contributions
es+0.081
icked+0.081
iled+0.057
ched+0.055
owed+0.054
Neg logit contributions
 city-0.136
 opera-0.133
 hotel-0.133
 talent-0.133
 orchestra-0.132
Token scared
Token ID12008
Feature activation+1.56e+00

Pos logit contributions
 city+0.053
 hotel+0.052
 talent+0.051
 orchestra+0.051
 opera+0.051
Neg logit contributions
es-0.061
icked-0.060
 could-0.048
iled-0.048
owed-0.047
Token of
Token ID286
Feature activation-1.80e+00

Pos logit contributions
icked+0.022
es+0.022
owed+0.016
iled+0.015
ched+0.015
Neg logit contributions
 city-0.037
 opera-0.037
 hotel-0.036
 orchestra-0.036
 talent-0.036
Token the
Token ID262
Feature activation-1.12e+01

Pos logit contributions
 talent+0.043
 orchestra+0.043
 city+0.043
 mirror+0.043
 hotel+0.043
Neg logit contributions
es-0.023
icked-0.022
 could-0.016
 they-0.016
iled-0.015
Token noises
Token ID26782
Feature activation+4.58e+00

Pos logit contributions
 circus+0.146
 former+0.143
 passenger+0.141
 fascination+0.138
 amaz+0.137
Neg logit contributions
 could-0.235
es-0.232
 there-0.231
 on-0.219
 in-0.213
Token.
Token ID13
Feature activation+4.29e+00

Pos logit contributions
icked+0.083
es+0.074
owed+0.063
o+0.058
umbered+0.057
Neg logit contributions
 opera-0.095
 city-0.094
 hotel-0.090
 airport-0.090
 orchestra-0.089
Token
Token ID198
Feature activation+6.09e-01

Pos logit contributions
icked+0.070
es+0.070
ched+0.053
owed+0.053
iled+0.052
Neg logit contributions
 city-0.098
 museum-0.091
 hotel-0.091
 opera-0.090
 zoo-0.089
Token
Token ID198
Feature activation+8.99e-01

Pos logit contributions
icked+0.010
es+0.009
owed+0.007
iled+0.007
ed+0.007
Neg logit contributions
 city-0.014
 talent-0.013
 hotel-0.013
 opera-0.013
 museum-0.013
TokenOne
Token ID3198
Feature activation-2.62e+00

Pos logit contributions
es+0.018
icked+0.017
iled+0.014
 could+0.014
ched+0.014
Neg logit contributions
 city-0.016
 hotel-0.015
 orchestra-0.015
 mirror-0.015
 talent-0.015
Token them
Token ID606
Feature activation+1.26e+00

Pos logit contributions
 city+0.002
 orchestra+0.002
 hotel+0.002
 talent+0.002
 airport+0.002
Neg logit contributions
es-0.002
icked-0.002
 could-0.001
 they-0.001
iled-0.001
Token.
Token ID13
Feature activation+3.74e+00

Pos logit contributions
icked+0.017
es+0.016
iled+0.012
ched+0.012
 could+0.012
Neg logit contributions
 city-0.030
 opera-0.030
 talent-0.030
 hotel-0.030
 orchestra-0.029
Token Lily
Token ID20037
Feature activation+1.74e+00

Pos logit contributions
es+0.068
icked+0.066
owed+0.054
iled+0.050
ched+0.050
Neg logit contributions
 city-0.079
 hotel-0.077
 mirror-0.075
 classroom-0.075
 park-0.074
Token had
Token ID550
Feature activation-1.23e+00

Pos logit contributions
icked+0.029
es+0.028
owed+0.021
iled+0.021
ched+0.021
Neg logit contributions
 city-0.037
 airport-0.036
 hotel-0.035
 talent-0.035
 mirror-0.035
Token a
Token ID257
Feature activation-8.87e+00

Pos logit contributions
 city+0.027
 orchestra+0.027
 hotel+0.027
 opera+0.026
 mirror+0.026
Neg logit contributions
es-0.019
icked-0.019
 could-0.015
 they-0.014
iled-0.013
Token pink
Token ID11398
Feature activation-9.61e+00

Pos logit contributions
reating+0.129
 orchestra+0.127
 former+0.124
 angles+0.118
 stations+0.118
Neg logit contributions
 they-0.175
 on-0.167
 could-0.163
es-0.161
 there-0.155
Token doll
Token ID3654
Feature activation+1.89e+00

Pos logit contributions
 former+0.126
 city+0.121
 orchestra+0.120
 talent+0.116
 priest+0.113
Neg logit contributions
es-0.219
icked-0.204
 on-0.191
bed-0.186
 they-0.186
Token with
Token ID351
Feature activation-6.84e-02

Pos logit contributions
icked+0.027
es+0.025
ched+0.019
owed+0.018
iled+0.018
Neg logit contributions
 city-0.045
 opera-0.045
 talent-0.044
 airport-0.044
 hotel-0.044
Token long
Token ID890
Feature activation-4.39e+00

Pos logit contributions
 hotel+0.001
 city+0.001
 mistake+0.001
 orchestra+0.001
 apartment+0.001
Neg logit contributions
es-0.001
icked-0.001
 could-0.001
ed-0.001
 they-0.001
Token hair
Token ID4190
Feature activation+1.88e+00

Pos logit contributions
 city+0.061
 hotel+0.060
 opera+0.059
 talent+0.059
 orchestra+0.059
Neg logit contributions
es-0.106
icked-0.103
 could-0.086
iled-0.085
ched-0.084
Token and
Token ID290
Feature activation-3.62e-01

Pos logit contributions
icked+0.028
es+0.027
iled+0.020
owed+0.020
ched+0.019
Neg logit contributions
 city-0.043
 opera-0.042
 mirror-0.042
 talent-0.042
 airport-0.042
Token help
Token ID1037
Feature activation+1.68e+00

Pos logit contributions
 city+0.034
 orchestra+0.033
 hotel+0.033
 airport+0.033
 opera+0.033
Neg logit contributions
icked-0.033
es-0.033
 could-0.026
owed-0.025
iled-0.025
Token her
Token ID607
Feature activation-3.56e+00

Pos logit contributions
icked+0.039
es+0.036
o+0.030
iled+0.030
ched+0.030
Neg logit contributions
 opera-0.027
 mirror-0.025
 talent-0.025
 fashion-0.024
 city-0.024
Token search
Token ID2989
Feature activation+2.39e+00

Pos logit contributions
 mirror+0.054
 classroom+0.054
 mistake+0.053
 city+0.053
 hotel+0.053
Neg logit contributions
icked-0.081
es-0.075
iled-0.067
ched-0.064
ed-0.063
Token for
Token ID329
Feature activation-1.65e+00

Pos logit contributions
es+0.035
icked+0.035
iled+0.027
o+0.025
 Remy+0.024
Neg logit contributions
 city-0.054
 hotel-0.053
 opera-0.052
 mirror-0.052
 airport-0.052
Token a
Token ID257
Feature activation-1.15e+01

Pos logit contributions
 city+0.037
 hotel+0.036
 talent+0.036
 opera+0.036
 mirror+0.036
Neg logit contributions
es-0.025
icked-0.025
iled-0.018
owed-0.018
 could-0.018
Token good
Token ID922
Feature activation-9.86e+00

Pos logit contributions
 orchestra+0.191
 opera+0.184
 city+0.169
 airport+0.168
 talent+0.163
Neg logit contributions
es-0.236
 could-0.228
icked-0.219
 they-0.208
 on-0.205
Token hiding
Token ID11816
Feature activation-5.18e+00

Pos logit contributions
 orchestra+0.147
 opera+0.139
 city+0.139
 airport+0.136
 cockpit+0.134
Neg logit contributions
es-0.217
icked-0.211
 could-0.183
ed-0.178
iled-0.177
Token spot
Token ID4136
Feature activation+2.05e+00

Pos logit contributions
 orchestra+0.056
 city+0.053
 talent+0.053
 opera+0.052
 mirror+0.050
Neg logit contributions
es-0.137
icked-0.133
 could-0.116
 Remy-0.111
iled-0.111
Token for
Token ID329
Feature activation-6.36e-01

Pos logit contributions
icked+0.031
es+0.031
iled+0.023
ched+0.023
 Remy+0.023
Neg logit contributions
 city-0.044
 hotel-0.044
 opera-0.043
 airport-0.043
 talent-0.042
Token her
Token ID607
Feature activation-8.01e+00

Pos logit contributions
 city+0.013
 hotel+0.013
 talent+0.013
 opera+0.013
 orchestra+0.013
Neg logit contributions
es-0.011
icked-0.011
iled-0.008
 could-0.008
owed-0.008
Token candy
Token ID18550
Feature activation+9.84e-01

Pos logit contributions
 city+0.145
 hotel+0.141
 orchestra+0.141
 opera+0.141
 talent+0.141
Neg logit contributions
es-0.157
icked-0.155
 could-0.124
iled-0.122
owed-0.121
Token through
Token ID832
Feature activation-5.89e-01

Pos logit contributions
es+0.014
icked+0.014
 could+0.009
iled+0.009
owed+0.009
Neg logit contributions
 city-0.026
 hotel-0.026
 talent-0.026
 opera-0.026
 orchestra-0.025
Token the
Token ID262
Feature activation-1.15e+01

Pos logit contributions
 orchestra+0.015
 city+0.015
 hotel+0.015
 opera+0.015
 talent+0.015
Neg logit contributions
es-0.007
icked-0.007
 could-0.005
 they-0.005
iled-0.005
Token sea
Token ID5417
Feature activation+1.82e+00

Pos logit contributions
 orchestra+0.156
 opera+0.151
 salon+0.151
 former+0.147
 apartment+0.147
Neg logit contributions
 could-0.229
 on-0.222
icked-0.215
 in-0.211
 there-0.210
Token and
Token ID290
Feature activation-3.31e-02

Pos logit contributions
es+0.022
icked+0.022
iled+0.014
 could+0.014
owed+0.014
Neg logit contributions
 city-0.047
 hotel-0.046
 talent-0.046
 opera-0.046
 orchestra-0.046
Token soon
Token ID2582
Feature activation+7.44e-02

Pos logit contributions
 city+0.001
 hotel+0.001
 opera+0.001
 orchestra+0.001
 talent+0.001
Neg logit contributions
icked-0.001
es-0.000
iled-0.000
owed-0.000
uts-0.000
Token the
Token ID262
Feature activation-1.26e+01

Pos logit contributions
icked+0.001
es+0.001
iled+0.001
owed+0.001
hot+0.001
Neg logit contributions
 hotel-0.002
 city-0.002
 airport-0.002
 beach-0.002
 opera-0.002
Token oy
Token ID35104
Feature activation-2.71e+00

Pos logit contributions
 apartment+0.222
 orchestra+0.217
 opera+0.205
 salon+0.202
 tant+0.201
Neg logit contributions
 could-0.221
icked-0.217
 in-0.211
 on-0.211
 there-0.208
Tokenster
Token ID1706
Feature activation+2.90e+00

Pos logit contributions
 mirror+0.045
 city+0.045
 hotel+0.045
 talent+0.044
 airport+0.044
Neg logit contributions
icked-0.054
es-0.054
owed-0.043
iled-0.043
ched-0.043
Token was
Token ID373
Feature activation-2.68e+00

Pos logit contributions
icked+0.043
es+0.042
iled+0.031
owed+0.030
 Remy+0.029
Neg logit contributions
 city-0.065
 hotel-0.064
 airport-0.064
 talent-0.063
 opera-0.063
Token free
Token ID1479
Feature activation+7.02e-02

Pos logit contributions
 city+0.051
 hotel+0.050
 opera+0.050
 talent+0.050
 orchestra+0.050
Neg logit contributions
es-0.050
icked-0.049
owed-0.038
iled-0.038
 could-0.037
Token!
Token ID0
Feature activation+3.01e+00

Pos logit contributions
icked+0.001
es+0.001
owed+0.001
iled+0.001
 Remy+0.001
Neg logit contributions
 city-0.002
 hotel-0.002
 opera-0.002
 mirror-0.002
 airport-0.002