TOP ACTIVATIONS
MAX = 0.091

sn ugg led her t eddy bear all the time .
the happiest frog in the whole wide world . <|endoftext|> Once
her daddy and the t eddy bear tight . She knew
and c uddled her t eddy bear . Her mom smiled
, and the blue t eddy bear closed their eyes and
was happy that her t eddy bear was fixed and that
man who shielded her t eddy bear from harm . <|endoftext|>
and disappeared with the t eddy bear . Amy was left
to hold up her t eddy bear . In the end
to leave his soft t eddy bear behind . But his
rot was the happiest he had ever been . <|endoftext|> Once
and made her the happiest bird in the world ! <|endoftext|>
each other and the t eddy bear . They were happy
and soon became the strongest bird in the sky . The
listen . She threw a tant rum in the store .
love my new modern t eddy bear !" <|endoftext|> Once upon
she had lost her t eddy bear . Her mom told
comfortably and happily in its bird c age . The boy
her feet and threw a tant rum . She grabbed the
She c uddled her t eddy bear and read her book

MIN ACTIVATIONS
MIN = -0.098

The end . <|endoftext|> Re becca was feeling anxious .
smiled and said , " Money is like coins that help
to help with the grocery shopping . Lily was excited to
if she wanted to go shopping . She nodded vigorously .
mom agreed to take him shopping for a new uniform .
to help with the grocery shopping . Lily was excited to
him if he did his homework . Tim my said no
Rachel would also teach Jack maths and stories . Jack loved
After they finished shopping , they went home .
could only watch TV for 30 minutes . Lily asked Mitt
" Do you really do maths in here ?" asked Amanda
mom if she could try yoga . Her mom said yes
Bob 's grandma took him shopping for clothes . She showed
After they completed their shopping , they went home and
mum if she could try yoga and her mum said yes
wall . It said " DO NOT T OU CH ".
who really wanted to try yoga . She was only three
red button that said " DO NOT PRESS ." She couldn
her mom my went grocery shopping . Laura was having lots
some food at the farm cafe . Lily and Sam were

INTERVAL 0.044 - 0.091
CONTAINS 1.117%

From then on , the whole family could come and play
happy . She hugged the fairy and thanked her for the
experience outside of shooting his arrow . He was excited to
. One day , the bird came to the palace and
was so happy to have returned the button and her dad

INTERVAL -0.004 - 0.044
CONTAINS 57.447%

the clipboard . " There , you are all fixed .
day of the delivery , Lily waited and waited . Finally
, determined not to be scared of the earthquake . <|endoftext|>
Lily 's mom my noticed that the f auc et was
big , brown t eddy bear that she loved very much

INTERVAL -0.051 - -0.004
CONTAINS 40.873%

, there was a big park with a large playground .
€ ™ t speak . But then the baker
and started looking at the pictures . He saw animals ,
not to turn over any sm elly worms , then he
and Ben are shocked . They see the toys . They

INTERVAL -0.098 - -0.051
CONTAINS 0.563%

smiled and whispered : � � € � � Once upon
the window . She tells them to spray the water on
mouth . One day , Jimmy was alone
a whip . It was very long and shiny . Sarah
mom took him to the library to pick out a new
Token sn
Token ID3013
Feature activation-6.15e-03

Pos logit contributions
 crying+0.032
 weeping+0.032
Ros+0.030
 bleeding+0.030
apped+0.030
Neg logit contributions
 stalls-0.038
 stall-0.036
 fare-0.036
 hike-0.035
 banks-0.035
Tokenugg
Token ID6837
Feature activation+1.66e-02

Pos logit contributions
 stalls+0.013
 stall+0.013
 hike+0.012
 banks+0.012
 fare+0.012
Neg logit contributions
 weeping-0.007
 crying-0.007
 bleeding-0.007
Ros-0.007
 blood-0.007
Tokenled
Token ID992
Feature activation+2.43e-02

Pos logit contributions
Ros+0.023
 weeping+0.023
 crying+0.023
ir+0.022
 bleeding+0.022
Neg logit contributions
 stalls-0.032
 stall-0.029
 flavors-0.029
 banks-0.029
 hike-0.029
Token her
Token ID607
Feature activation+2.72e-02

Pos logit contributions
 weeping+0.028
 crying+0.027
Ros+0.027
 bleeding+0.027
ir+0.027
Neg logit contributions
 stalls-0.051
 stall-0.049
 hike-0.049
 flavors-0.048
 banks-0.048
Token t
Token ID256
Feature activation+4.57e-02

Pos logit contributions
 weeping+0.029
 crying+0.029
 bleeding+0.028
Ros+0.028
ir+0.027
Neg logit contributions
 stalls-0.060
 stall-0.057
 hike-0.057
 flavors-0.056
 banks-0.056
Tokeneddy
Token ID21874
Feature activation+9.06e-02

Pos logit contributions
 weeping+0.044
 Emily+0.042
 crying+0.041
 sorry+0.039
 blood+0.038
Neg logit contributions
 fare-0.103
 belts-0.100
cases-0.099
 stalls-0.098
 flavors-0.097
Token bear
Token ID6842
Feature activation+4.32e-02

Pos logit contributions
Ros+0.129
ir+0.128
 weeping+0.128
apped+0.126
 crying+0.125
Neg logit contributions
 stalls-0.159
 hike-0.155
 stall-0.153
 banks-0.153
 fare-0.147
Token all
Token ID477
Feature activation+3.96e-02

Pos logit contributions
 weeping+0.048
 crying+0.047
 bleeding+0.046
Ros+0.045
ir+0.044
Neg logit contributions
 stalls-0.094
 flavors-0.090
 stall-0.088
 hike-0.088
 fare-0.087
Token the
Token ID262
Feature activation+2.25e-02

Pos logit contributions
 weeping+0.057
Ros+0.056
 crying+0.056
 bleeding+0.056
ir+0.055
Neg logit contributions
 stalls-0.072
 hike-0.068
 stall-0.068
 banks-0.067
 flavors-0.067
Token time
Token ID640
Feature activation+1.31e-03

Pos logit contributions
 weeping+0.045
 crying+0.044
 bleeding+0.044
Ros+0.044
ir+0.043
Neg logit contributions
 stalls-0.028
 stall-0.026
 flavors-0.026
 hike-0.026
 banks-0.026
Token.
Token ID13
Feature activation+2.12e-02

Pos logit contributions
 crying+0.002
 weeping+0.002
 bleeding+0.002
 streaming+0.002
ir+0.002
Neg logit contributions
 stalls-0.002
 banks-0.002
 fare-0.002
 hike-0.002
 stall-0.002
Token the
Token ID262
Feature activation+2.53e-02

Pos logit contributions
 weeping+0.031
 crying+0.030
 bleeding+0.029
Ros+0.028
ir+0.028
Neg logit contributions
 stalls-0.056
 hike-0.054
 banks-0.053
 stall-0.053
 fare-0.053
Token happiest
Token ID49414
Feature activation+6.77e-02

Pos logit contributions
 weeping+0.030
Ros+0.029
 crying+0.029
 bleeding+0.029
 microphone+0.027
Neg logit contributions
 stalls-0.053
 stall-0.050
cases-0.050
 banks-0.049
 hike-0.049
Token frog
Token ID21264
Feature activation+6.08e-02

Pos logit contributions
 weeping+0.095
 crying+0.093
Ros+0.092
 bleeding+0.091
ir+0.090
Neg logit contributions
 stalls-0.127
 stall-0.119
 hike-0.119
 banks-0.118
 flavors-0.117
Token in
Token ID287
Feature activation+4.18e-02

Pos logit contributions
 weeping+0.055
 crying+0.054
 bleeding+0.051
ir+0.048
apped+0.048
Neg logit contributions
 stalls-0.145
 fare-0.140
 flavors-0.139
 banks-0.139
 hike-0.138
Token the
Token ID262
Feature activation+5.85e-02

Pos logit contributions
Ros+0.067
 bleeding+0.065
 weeping+0.063
apped+0.062
 crying+0.062
Neg logit contributions
 stalls-0.070
 stall-0.067
 hike-0.067
 banks-0.064
 flavors-0.063
Token whole
Token ID2187
Feature activation+9.06e-02

Pos logit contributions
 bleeding+0.098
apped+0.098
 weeping+0.097
Ros+0.097
 crying+0.096
Neg logit contributions
 stalls-0.092
 hike-0.089
 banks-0.087
 stall-0.086
 flavors-0.082
Token wide
Token ID3094
Feature activation+4.95e-02

Pos logit contributions
 bleeding+0.154
 weeping+0.150
 crying+0.148
 blood+0.147
Ros+0.147
Neg logit contributions
 stalls-0.148
 hike-0.139
 stall-0.139
 banks-0.134
 flavors-0.130
Token world
Token ID995
Feature activation+2.09e-02

Pos logit contributions
 weeping+0.077
 crying+0.075
ir+0.075
Ros+0.075
 bleeding+0.074
Neg logit contributions
 stalls-0.084
 hike-0.078
 stall-0.078
 flavors-0.078
 fare-0.077
Token.
Token ID13
Feature activation+3.17e-02

Pos logit contributions
 weeping+0.026
 crying+0.026
ir+0.024
apped+0.023
 streaming+0.023
Neg logit contributions
 stalls-0.041
 banks-0.040
 flavors-0.040
 stall-0.039
 list-0.039
Token<|endoftext|>
Token ID50256
Feature activation+2.59e-02

Pos logit contributions
 bleeding+0.051
 weeping+0.050
 crying+0.049
Ros+0.048
 streaming+0.048
Neg logit contributions
 stalls-0.052
 stall-0.050
 hike-0.049
 banks-0.048
 flavors-0.047
TokenOnce
Token ID7454
Feature activation+3.58e-03

Pos logit contributions
 weeping+0.039
 bleeding+0.038
Ros+0.038
 crying+0.037
 streaming+0.036
Neg logit contributions
 stalls-0.046
 stall-0.044
 hike-0.043
 banks-0.042
 flavors-0.042
Token her
Token ID607
Feature activation+2.29e-02

Pos logit contributions
w+0.083
 bleeding+0.081
itt+0.080
apped+0.080
 shattered+0.078
Neg logit contributions
 list-0.058
 shelves-0.056
 stalls-0.056
 hike-0.055
 experiments-0.054
Token daddy
Token ID35822
Feature activation+6.60e-02

Pos logit contributions
ir+0.027
Ros+0.026
 weeping+0.025
 bleeding+0.025
apped+0.025
Neg logit contributions
 stalls-0.047
 stall-0.046
 hike-0.045
 flavors-0.045
 list-0.045
Token and
Token ID290
Feature activation+1.51e-02

Pos logit contributions
 weeping+0.086
 crying+0.084
 bleeding+0.083
Ros+0.082
ir+0.080
Neg logit contributions
 stalls-0.130
 hike-0.124
 banks-0.122
 flavors-0.122
 stall-0.122
Token the
Token ID262
Feature activation+3.37e-02

Pos logit contributions
ir+0.027
Ros+0.026
 weeping+0.025
 bleeding+0.025
el+0.024
Neg logit contributions
 stalls-0.023
 flavors-0.021
 hike-0.021
 stall-0.021
 list-0.021
Token t
Token ID256
Feature activation+4.71e-02

Pos logit contributions
ir+0.061
apped+0.059
Ros+0.056
apping+0.055
 weeping+0.055
Neg logit contributions
 stalls-0.050
 flavors-0.048
 stall-0.048
 hike-0.048
 fare-0.046
Tokeneddy
Token ID21874
Feature activation+9.03e-02

Pos logit contributions
 Emily+0.052
rys+0.048
 microphone+0.047
 weeping+0.046
 boom+0.045
Neg logit contributions
 fare-0.092
 flavors-0.091
 belts-0.090
 stall-0.090
 stalls-0.090
Token bear
Token ID6842
Feature activation+4.86e-02

Pos logit contributions
ir+0.158
apping+0.151
Ros+0.149
apped+0.148
itt+0.147
Neg logit contributions
 stalls-0.124
 banks-0.121
 stall-0.120
 bank-0.119
 flavors-0.119
Token tight
Token ID5381
Feature activation+4.50e-03

Pos logit contributions
 weeping+0.080
 crying+0.079
 bleeding+0.078
Ros+0.077
ir+0.075
Neg logit contributions
 stalls-0.080
 hike-0.075
 stall-0.074
 banks-0.073
 flavors-0.073
Token.
Token ID13
Feature activation+2.28e-02

Pos logit contributions
 crying+0.007
 weeping+0.007
 bleeding+0.007
 streaming+0.007
apped+0.006
Neg logit contributions
 stalls-0.008
 stall-0.008
 hike-0.008
 fare-0.007
 flavors-0.007
Token She
Token ID1375
Feature activation+1.50e-02

Pos logit contributions
ir+0.036
 bleeding+0.034
 weeping+0.033
Ros+0.033
 microphone+0.033
Neg logit contributions
 stalls-0.037
 stall-0.035
 hike-0.034
 list-0.034
 banks-0.034
Token knew
Token ID2993
Feature activation+1.81e-02

Pos logit contributions
 weeping+0.025
ir+0.025
Ros+0.025
 bleeding+0.024
 crying+0.024
Neg logit contributions
 stalls-0.023
 hike-0.022
 flavors-0.022
 stall-0.022
 banks-0.022
Token and
Token ID290
Feature activation+2.77e-03

Pos logit contributions
Ros+0.051
ir+0.050
 weeping+0.049
 bleeding+0.048
 crying+0.048
Neg logit contributions
 stalls-0.052
 stall-0.049
 hike-0.048
 flavors-0.047
 banks-0.046
Token c
Token ID269
Feature activation+9.67e-03

Pos logit contributions
ir+0.004
 weeping+0.004
Ros+0.004
 crying+0.004
 bleeding+0.004
Neg logit contributions
 stalls-0.005
 stall-0.004
 hike-0.004
 flavors-0.004
 banks-0.004
Tokenuddled
Token ID32745
Feature activation+1.84e-02

Pos logit contributions
 weeping+0.017
Ros+0.016
 crying+0.016
 Emily+0.016
 Ros+0.016
Neg logit contributions
 stalls-0.015
 belts-0.014
 flavors-0.014
cases-0.014
 stall-0.014
Token her
Token ID607
Feature activation+3.20e-02

Pos logit contributions
 weeping+0.021
 crying+0.020
 bleeding+0.020
Ros+0.020
ir+0.019
Neg logit contributions
 stalls-0.040
 hike-0.038
 stall-0.038
 banks-0.037
 flavors-0.037
Token t
Token ID256
Feature activation+3.68e-02

Pos logit contributions
ir+0.050
Ros+0.050
apped+0.048
 weeping+0.047
 streaming+0.046
Neg logit contributions
 stalls-0.054
 stall-0.052
 flavors-0.051
 hike-0.051
 fare-0.051
Tokeneddy
Token ID21874
Feature activation+8.97e-02

Pos logit contributions
 Emily+0.045
 weeping+0.041
 Ros+0.041
 streaming+0.040
 Po+0.038
Neg logit contributions
 belts-0.075
 fare-0.073
cases-0.072
belt-0.069
 flavors-0.069
Token bear
Token ID6842
Feature activation+4.56e-02

Pos logit contributions
Ros+0.143
ir+0.143
apped+0.139
apping+0.135
 weeping+0.134
Neg logit contributions
 stalls-0.141
 stall-0.137
 banks-0.135
 bank-0.132
 hike-0.132
Token.
Token ID13
Feature activation+1.90e-03

Pos logit contributions
 weeping+0.057
 crying+0.057
 bleeding+0.056
Ros+0.053
ir+0.052
Neg logit contributions
 stalls-0.092
 flavors-0.088
 hike-0.087
 stall-0.086
 banks-0.085
Token Her
Token ID2332
Feature activation+1.05e-02

Pos logit contributions
ir+0.002
 weeping+0.002
 bleeding+0.002
Ros+0.002
 crying+0.002
Neg logit contributions
 stalls-0.004
 stall-0.004
 hike-0.003
 banks-0.003
 flavors-0.003
Token mom
Token ID1995
Feature activation+3.49e-02

Pos logit contributions
ir+0.017
Ros+0.016
apped+0.016
 streaming+0.015
 weeping+0.015
Neg logit contributions
 flavors-0.018
 stalls-0.017
 stall-0.017
 belts-0.017
 hike-0.017
Token smiled
Token ID13541
Feature activation+2.77e-02

Pos logit contributions
 weeping+0.047
 crying+0.047
 bleeding+0.046
Ros+0.046
ir+0.044
Neg logit contributions
 stalls-0.067
 hike-0.063
 banks-0.063
 stall-0.063
 fare-0.062
Token,
Token ID11
Feature activation+1.79e-02

Pos logit contributions
 weeping+0.032
 crying+0.032
 bleeding+0.030
Ros+0.029
apped+0.029
Neg logit contributions
 stalls-0.044
 fare-0.043
 stall-0.042
 flavors-0.042
 hike-0.042
Token and
Token ID290
Feature activation+8.61e-03

Pos logit contributions
Ros+0.026
 weeping+0.026
ir+0.026
 bleeding+0.026
 crying+0.026
Neg logit contributions
 stalls-0.032
 hike-0.030
 stall-0.030
 flavors-0.029
 fare-0.029
Token the
Token ID262
Feature activation+1.39e-02

Pos logit contributions
ir+0.012
Ros+0.012
 bleeding+0.011
 blood+0.011
apped+0.011
Neg logit contributions
 stalls-0.016
 hike-0.016
 stall-0.015
 banks-0.015
 flavors-0.015
Token blue
Token ID4171
Feature activation+3.80e-02

Pos logit contributions
 weeping+0.019
 crying+0.019
 bleeding+0.019
Ros+0.019
ir+0.018
Neg logit contributions
 stalls-0.027
 hike-0.025
 stall-0.025
 banks-0.025
 flavors-0.025
Token t
Token ID256
Feature activation+3.79e-02

Pos logit contributions
Ros+0.053
 weeping+0.052
ir+0.051
 crying+0.051
 bleeding+0.050
Neg logit contributions
 stalls-0.068
 stall-0.066
 hike-0.065
 banks-0.065
 flavors-0.065
Tokeneddy
Token ID21874
Feature activation+8.94e-02

Pos logit contributions
 Emily+0.045
 weeping+0.040
 crying+0.040
rys+0.039
 Po+0.038
Neg logit contributions
 fare-0.075
 belts-0.074
cases-0.073
 flavors-0.073
 stalls-0.072
Token bear
Token ID6842
Feature activation+4.21e-02

Pos logit contributions
Ros+0.135
ir+0.135
 weeping+0.131
apped+0.131
apping+0.128
Neg logit contributions
 stalls-0.145
 flavors-0.138
 banks-0.138
 hike-0.137
 stall-0.136
Token closed
Token ID4838
Feature activation-5.96e-03

Pos logit contributions
 weeping+0.063
 crying+0.063
 bleeding+0.061
Ros+0.060
ir+0.060
Neg logit contributions
 stalls-0.074
 stall-0.070
 flavors-0.069
 hike-0.069
 banks-0.068
Token their
Token ID511
Feature activation-1.60e-02

Pos logit contributions
 stalls+0.011
 stall+0.011
 banks+0.010
 hike+0.010
 flavors+0.010
Neg logit contributions
Ros-0.008
 weeping-0.008
 bleeding-0.008
ir-0.008
 crying-0.008
Token eyes
Token ID2951
Feature activation+1.04e-02

Pos logit contributions
 stalls+0.041
 stall+0.039
 hike+0.039
 banks+0.039
 flavors+0.039
Neg logit contributions
 weeping-0.011
Ros-0.011
 crying-0.011
 bleeding-0.011
ir-0.010
Token and
Token ID290
Feature activation+1.35e-03

Pos logit contributions
 weeping+0.015
 crying+0.015
 bleeding+0.015
Ros+0.015
ir+0.015
Neg logit contributions
 stalls-0.018
 hike-0.017
 stall-0.017
 flavors-0.017
 banks-0.017
Token was
Token ID373
Feature activation+3.27e-02

Pos logit contributions
 weeping+0.047
 crying+0.046
Ros+0.046
 bleeding+0.045
ir+0.044
Neg logit contributions
 stalls-0.053
 stall-0.050
 hike-0.050
 fare-0.049
 banks-0.049
Token happy
Token ID3772
Feature activation+5.41e-02

Pos logit contributions
ir+0.047
Ros+0.046
 weeping+0.046
 bleeding+0.045
el+0.045
Neg logit contributions
 stalls-0.058
 stall-0.055
 hike-0.054
 banks-0.054
 flavors-0.054
Token that
Token ID326
Feature activation+6.35e-02

Pos logit contributions
ir+0.111
itt+0.102
el+0.102
Ros+0.100
ternally+0.100
Neg logit contributions
 hike-0.065
 stalls-0.062
 stall-0.060
 flavors-0.059
 list-0.055
Token her
Token ID607
Feature activation+2.74e-02

Pos logit contributions
 weeping+0.089
ir+0.089
Ros+0.088
 bleeding+0.087
 crying+0.087
Neg logit contributions
 stalls-0.116
 hike-0.111
 stall-0.109
 flavors-0.108
 banks-0.107
Token t
Token ID256
Feature activation+4.66e-02

Pos logit contributions
Ros+0.039
ir+0.039
 weeping+0.039
 crying+0.038
apped+0.038
Neg logit contributions
 stalls-0.049
 hike-0.047
 stall-0.047
 flavors-0.047
 banks-0.046
Tokeneddy
Token ID21874
Feature activation+8.91e-02

Pos logit contributions
 Emily+0.057
 sorry+0.053
rys+0.053
 boom+0.052
 weeping+0.051
Neg logit contributions
 belts-0.086
 flavors-0.085
 fare-0.084
cases-0.083
 stall-0.081
Token bear
Token ID6842
Feature activation+4.72e-02

Pos logit contributions
ir+0.145
apped+0.138
apping+0.135
aps+0.134
Ros+0.128
Neg logit contributions
 flavors-0.139
 stalls-0.138
 hike-0.133
 stall-0.132
 banks-0.131
Token was
Token ID373
Feature activation+4.06e-02

Pos logit contributions
 weeping+0.070
 crying+0.070
 bleeding+0.069
Ros+0.065
 streaming+0.064
Neg logit contributions
 stalls-0.089
 stall-0.083
 hike-0.082
 fare-0.081
 banks-0.081
Token fixed
Token ID5969
Feature activation+4.37e-02

Pos logit contributions
ir+0.062
 weeping+0.060
Ros+0.060
apped+0.059
el+0.059
Neg logit contributions
 stalls-0.066
 hike-0.063
 stall-0.063
 flavors-0.061
 fare-0.061
Token and
Token ID290
Feature activation+2.30e-02

Pos logit contributions
 weeping+0.065
Ros+0.064
 crying+0.064
 bleeding+0.063
ir+0.063
Neg logit contributions
 stalls-0.077
 stall-0.072
 hike-0.072
 flavors-0.071
 banks-0.071
Token that
Token ID326
Feature activation+4.73e-02

Pos logit contributions
ir+0.040
el+0.036
 weeping+0.036
Ros+0.036
apped+0.036
Neg logit contributions
 stalls-0.036
 hike-0.034
 flavors-0.034
 stall-0.032
 belts-0.032
Token man
Token ID582
Feature activation+3.72e-02

Pos logit contributions
 weeping+0.052
 crying+0.052
 bleeding+0.050
 blood+0.049
Ros+0.049
Neg logit contributions
 stalls-0.045
 banks-0.043
 hike-0.042
 stall-0.040
 fare-0.039
Token who
Token ID508
Feature activation+1.16e-02

Pos logit contributions
 weeping+0.043
 crying+0.043
 bleeding+0.041
Ros+0.038
 streaming+0.038
Neg logit contributions
 stalls-0.081
 banks-0.077
 flavors-0.077
 hike-0.077
 stall-0.076
Token shielded
Token ID45153
Feature activation+1.37e-02

Pos logit contributions
 weeping+0.020
 crying+0.019
Ros+0.019
 bleeding+0.019
ir+0.019
Neg logit contributions
 stalls-0.018
 hike-0.017
 stall-0.017
 banks-0.016
 flavors-0.016
Token her
Token ID607
Feature activation+2.29e-02

Pos logit contributions
apped+0.023
apping+0.023
 weeping+0.022
aps+0.021
itt+0.021
Neg logit contributions
 stalls-0.020
 flavors-0.018
 banks-0.018
 bank-0.018
 shelves-0.018
Token t
Token ID256
Feature activation+4.13e-02

Pos logit contributions
Ros+0.034
ir+0.034
apped+0.033
 weeping+0.033
 crying+0.032
Neg logit contributions
 stalls-0.041
 flavors-0.039
 banks-0.038
 stall-0.038
 fare-0.038
Tokeneddy
Token ID21874
Feature activation+8.89e-02

Pos logit contributions
 Emily+0.040
 weeping+0.039
 crying+0.036
 sorry+0.035
 Ros+0.035
Neg logit contributions
cases-0.090
 fare-0.090
 stalls-0.088
 flavors-0.087
 stall-0.087
Token bear
Token ID6842
Feature activation+4.40e-02

Pos logit contributions
apped+0.137
ir+0.134
Ros+0.134
apping+0.130
 weeping+0.129
Neg logit contributions
 stalls-0.147
 banks-0.145
 stall-0.142
 hike-0.142
 bank-0.141
Token from
Token ID422
Feature activation+2.71e-02

Pos logit contributions
 weeping+0.062
 crying+0.061
Ros+0.060
 bleeding+0.060
ir+0.059
Neg logit contributions
 stalls-0.082
 stall-0.077
 hike-0.077
 flavors-0.076
 banks-0.075
Token harm
Token ID4419
Feature activation+4.30e-02

Pos logit contributions
apped+0.041
Ros+0.040
ir+0.038
apping+0.037
aps+0.037
Neg logit contributions
 stalls-0.049
 stall-0.046
 banks-0.046
 hike-0.046
 bank-0.045
Token.
Token ID13
Feature activation+3.18e-02

Pos logit contributions
 weeping+0.062
Ros+0.061
 crying+0.061
 bleeding+0.061
ir+0.060
Neg logit contributions
 stalls-0.078
 stall-0.073
 hike-0.073
 banks-0.072
 flavors-0.072
Token<|endoftext|>
Token ID50256
Feature activation+1.69e-02

Pos logit contributions
 bleeding+0.051
ir+0.051
 weeping+0.050
 streaming+0.049
 microphone+0.049
Neg logit contributions
 stalls-0.050
 stall-0.049
 hike-0.047
 flavors-0.046
 fare-0.046
Token and
Token ID290
Feature activation+1.69e-03

Pos logit contributions
 stalls+0.033
 hike+0.032
 flavors+0.032
 stall+0.031
 fare+0.031
Neg logit contributions
 crying-0.021
 weeping-0.021
 bleeding-0.019
 streaming-0.018
ir-0.017
Token disappeared
Token ID12120
Feature activation+1.65e-02

Pos logit contributions
 weeping+0.003
 crying+0.003
 bleeding+0.003
Ros+0.003
ir+0.003
Neg logit contributions
 stalls-0.003
 stall-0.003
 hike-0.003
 flavors-0.003
 banks-0.003
Token with
Token ID351
Feature activation+1.17e-02

Pos logit contributions
 weeping+0.026
 crying+0.025
 bleeding+0.025
Ros+0.025
ir+0.025
Neg logit contributions
 stalls-0.028
 hike-0.027
 stall-0.027
 flavors-0.026
 banks-0.026
Token the
Token ID262
Feature activation+2.14e-02

Pos logit contributions
Ros+0.021
eddy+0.020
apping+0.020
apped+0.019
ooming+0.019
Neg logit contributions
 stalls-0.017
 stall-0.016
 flavors-0.016
 shelves-0.015
 hike-0.015
Token t
Token ID256
Feature activation+3.86e-02

Pos logit contributions
apped+0.037
ir+0.037
Ros+0.035
apping+0.034
aps+0.033
Neg logit contributions
 stalls-0.033
 flavors-0.033
 stall-0.033
 fare-0.032
 hike-0.031
Tokeneddy
Token ID21874
Feature activation+8.87e-02

Pos logit contributions
 weeping+0.041
 streaming+0.039
 crying+0.038
 Emily+0.038
 sorry+0.038
Neg logit contributions
cases-0.077
 stall-0.076
 flavors-0.075
 fare-0.075
 stalls-0.074
Token bear
Token ID6842
Feature activation+4.56e-02

Pos logit contributions
apping+0.154
apped+0.154
ir+0.151
Ros+0.148
aps+0.145
Neg logit contributions
 stalls-0.122
 stall-0.121
 flavors-0.120
 bank-0.118
 banks-0.118
Token.
Token ID13
Feature activation+1.44e-02

Pos logit contributions
 weeping+0.073
 crying+0.073
 bleeding+0.071
ir+0.068
 streaming+0.067
Neg logit contributions
 stalls-0.078
 flavors-0.073
 hike-0.073
 stall-0.072
 banks-0.071
Token Amy
Token ID14235
Feature activation+1.14e-02

Pos logit contributions
ir+0.024
 microphone+0.024
el+0.023
 weeping+0.023
 streaming+0.023
Neg logit contributions
 stalls-0.023
 stall-0.022
 hike-0.021
 banks-0.021
 flavors-0.020
Token was
Token ID373
Feature activation+1.65e-02

Pos logit contributions
Ros+0.019
ir+0.019
 weeping+0.019
 bleeding+0.019
 crying+0.018
Neg logit contributions
 stalls-0.017
 banks-0.016
 stall-0.016
 hike-0.016
 flavors-0.016
Token left
Token ID1364
Feature activation+3.68e-02

Pos logit contributions
ir+0.028
Ros+0.027
el+0.026
w+0.026
itt+0.025
Neg logit contributions
 stalls-0.024
 stall-0.024
 bank-0.023
 banks-0.023
 ranc-0.022
Token to
Token ID284
Feature activation+9.57e-03

Pos logit contributions
 stalls+0.002
 stall+0.001
 banks+0.001
 hike+0.001
nic+0.001
Neg logit contributions
 crying-0.001
 weeping-0.001
 bleeding-0.001
 streaming-0.001
el-0.001
Token hold
Token ID1745
Feature activation+3.99e-03

Pos logit contributions
 crying+0.014
 weeping+0.013
ir+0.013
 bleeding+0.013
 streaming+0.012
Neg logit contributions
 stalls-0.019
 stall-0.017
cases-0.017
 belts-0.016
 flavors-0.016
Token up
Token ID510
Feature activation-8.92e-04

Pos logit contributions
 weeping+0.005
Ros+0.005
 bleeding+0.005
 crying+0.005
ir+0.005
Neg logit contributions
 stalls-0.008
 stall-0.007
 flavors-0.007
 hike-0.007
 fare-0.007
Token her
Token ID607
Feature activation+3.30e-02

Pos logit contributions
 stalls+0.002
 hike+0.002
 stall+0.002
 banks+0.002
 flavors+0.002
Neg logit contributions
 weeping-0.001
 crying-0.001
Ros-0.001
 bleeding-0.001
ir-0.001
Token t
Token ID256
Feature activation+3.91e-02

Pos logit contributions
Ros+0.048
 weeping+0.048
ir+0.047
 crying+0.047
 bleeding+0.046
Neg logit contributions
 stalls-0.059
 stall-0.056
 hike-0.056
 flavors-0.055
 banks-0.055
Tokeneddy
Token ID21874
Feature activation+8.86e-02

Pos logit contributions
 weeping+0.035
 crying+0.033
 bleeding+0.031
 blood+0.030
 Emily+0.030
Neg logit contributions
 flavors-0.088
 fare-0.088
 stalls-0.088
 stall-0.088
 belts-0.087
Token bear
Token ID6842
Feature activation+4.17e-02

Pos logit contributions
ir+0.125
Ros+0.124
 weeping+0.124
apped+0.122
 crying+0.120
Neg logit contributions
 stalls-0.158
 stall-0.153
 hike-0.152
 banks-0.151
 flavors-0.148
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
 crying+0.057
 weeping+0.056
 bleeding+0.055
ir+0.053
el+0.053
Neg logit contributions
 stalls-0.081
 flavors-0.077
 hike-0.074
 fare-0.073
 stall-0.072
Token In
Token ID554
Feature activation-5.25e-03

Pos logit contributions
ir+0.024
 bleeding+0.024
 weeping+0.023
 blood+0.023
el+0.023
Neg logit contributions
 stalls-0.020
 banks-0.018
 hike-0.018
 stall-0.018
 list-0.018
Token the
Token ID262
Feature activation-1.14e-02

Pos logit contributions
 stalls+0.008
 stall+0.008
 hike+0.008
 list+0.008
 flavors+0.007
Neg logit contributions
Ros-0.008
 weeping-0.008
 bleeding-0.008
apped-0.008
ir-0.008
Token end
Token ID886
Feature activation+9.47e-03

Pos logit contributions
 stalls+0.018
 stall+0.017
 flavors+0.017
 hike+0.017
 fare+0.017
Neg logit contributions
 weeping-0.019
 crying-0.019
 bleeding-0.019
Ros-0.019
ir-0.018
Token to
Token ID284
Feature activation+2.50e-02

Pos logit contributions
Ros+0.027
ir+0.027
 weeping+0.026
 bleeding+0.026
apped+0.025
Neg logit contributions
 stalls-0.024
 hike-0.023
 stall-0.022
 banks-0.022
 flavors-0.022
Token leave
Token ID2666
Feature activation+3.44e-02

Pos logit contributions
Ros+0.045
apped+0.044
 weeping+0.044
ir+0.044
 bleeding+0.044
Neg logit contributions
 stalls-0.036
 hike-0.035
 banks-0.034
 stall-0.033
 fare-0.032
Token his
Token ID465
Feature activation+3.20e-02

Pos logit contributions
 weeping+0.048
 crying+0.048
 bleeding+0.047
Ros+0.046
ir+0.045
Neg logit contributions
 stalls-0.065
 hike-0.061
 stall-0.061
 banks-0.060
 flavors-0.060
Token soft
Token ID2705
Feature activation+4.22e-02

Pos logit contributions
Ros+0.049
ir+0.049
 weeping+0.049
 bleeding+0.047
 crying+0.047
Neg logit contributions
 stalls-0.055
 stall-0.052
 hike-0.052
 flavors-0.051
 banks-0.051
Token t
Token ID256
Feature activation+4.07e-02

Pos logit contributions
 weeping+0.063
 crying+0.062
Ros+0.062
 bleeding+0.061
ir+0.061
Neg logit contributions
 stalls-0.075
 hike-0.070
 stall-0.070
 banks-0.069
 flavors-0.069
Tokeneddy
Token ID21874
Feature activation+8.84e-02

Pos logit contributions
 Emily+0.037
 weeping+0.036
 streaming+0.034
 crying+0.033
 Ros+0.033
Neg logit contributions
 stall-0.090
 stalls-0.090
 fare-0.090
 flavors-0.089
cases-0.089
Token bear
Token ID6842
Feature activation+4.52e-02

Pos logit contributions
ir+0.130
Ros+0.130
apped+0.128
 weeping+0.125
 crying+0.121
Neg logit contributions
 stalls-0.153
 hike-0.148
 stall-0.147
 banks-0.146
 flavors-0.143
Token behind
Token ID2157
Feature activation+3.37e-03

Pos logit contributions
 weeping+0.050
 crying+0.049
 bleeding+0.048
Ros+0.046
ir+0.045
Neg logit contributions
 stalls-0.099
 flavors-0.094
 stall-0.093
 hike-0.093
 banks-0.092
Token.
Token ID13
Feature activation+2.18e-02

Pos logit contributions
 weeping+0.006
 crying+0.006
 bleeding+0.005
Ros+0.005
ir+0.005
Neg logit contributions
 stalls-0.005
 stall-0.005
 hike-0.005
 flavors-0.005
 banks-0.005
Token But
Token ID887
Feature activation+3.30e-03

Pos logit contributions
 weeping+0.031
 crying+0.031
 Ros+0.029
 bleeding+0.029
Ros+0.028
Neg logit contributions
 stalls-0.043
 stall-0.040
 flavors-0.039
 hike-0.039
 fare-0.038
Token his
Token ID465
Feature activation+7.04e-04

Pos logit contributions
 weeping+0.006
 crying+0.006
 bleeding+0.006
Ros+0.005
apped+0.005
Neg logit contributions
 stalls-0.005
 stall-0.005
 banks-0.005
 hike-0.005
 flavors-0.005
Tokenrot
Token ID10599
Feature activation+6.22e-02

Pos logit contributions
 weeping+0.058
 crying+0.057
 bleeding+0.055
 streaming+0.055
 Emily+0.053
Neg logit contributions
 stall-0.085
 stalls-0.085
 fare-0.084
nic-0.082
land-0.082
Token was
Token ID373
Feature activation+4.18e-02

Pos logit contributions
 weeping+0.096
 crying+0.094
Ros+0.092
 bleeding+0.092
ir+0.091
Neg logit contributions
 stalls-0.108
 hike-0.101
 stall-0.101
 banks-0.099
 fare-0.099
Token the
Token ID262
Feature activation+3.19e-02

Pos logit contributions
 weeping+0.056
 crying+0.055
Ros+0.055
 bleeding+0.054
ir+0.054
Neg logit contributions
 stalls-0.080
 hike-0.076
 stall-0.076
 banks-0.075
 flavors-0.075
Token happiest
Token ID49414
Feature activation+6.41e-02

Pos logit contributions
 weeping+0.045
 crying+0.044
Ros+0.044
 bleeding+0.043
ir+0.043
Neg logit contributions
 stalls-0.060
 hike-0.056
 stall-0.056
 banks-0.055
 flavors-0.055
Token he
Token ID339
Feature activation+6.47e-02

Pos logit contributions
 bleeding+0.085
 weeping+0.085
Ros+0.085
ir+0.083
 crying+0.082
Neg logit contributions
 stalls-0.121
 stall-0.117
 hike-0.113
 flavors-0.113
 fare-0.109
Token had
Token ID550
Feature activation+8.84e-02

Pos logit contributions
 weeping+0.080
 crying+0.079
 bleeding+0.076
Ros+0.074
ir+0.073
Neg logit contributions
 stalls-0.134
 banks-0.131
 fare-0.127
 flavors-0.127
 hike-0.125
Token ever
Token ID1683
Feature activation+5.78e-02

Pos logit contributions
 weeping+0.090
 crying+0.088
ir+0.086
Ros+0.086
 bleeding+0.085
Neg logit contributions
 stalls-0.199
 hike-0.190
 stall-0.189
 banks-0.189
 flavors-0.187
Token been
Token ID587
Feature activation+7.56e-02

Pos logit contributions
 crying+0.085
 weeping+0.085
 bleeding+0.081
ir+0.081
 streaming+0.077
Neg logit contributions
 stalls-0.106
 fare-0.103
 stall-0.102
 hike-0.101
 list-0.099
Token.
Token ID13
Feature activation+2.86e-02

Pos logit contributions
 weeping+0.111
 crying+0.109
Ros+0.108
 bleeding+0.108
ir+0.107
Neg logit contributions
 stalls-0.136
 hike-0.128
 stall-0.128
 flavors-0.126
 banks-0.126
Token<|endoftext|>
Token ID50256
Feature activation+1.45e-02

Pos logit contributions
 bleeding+0.044
 weeping+0.044
 crying+0.042
Ros+0.042
ir+0.042
Neg logit contributions
 stalls-0.049
 stall-0.047
 hike-0.045
 banks-0.045
 flavors-0.045
TokenOnce
Token ID7454
Feature activation-2.99e-03

Pos logit contributions
 weeping+0.021
 bleeding+0.020
Ros+0.020
 crying+0.020
ir+0.020
Neg logit contributions
 stalls-0.027
 stall-0.025
 hike-0.025
 banks-0.025
 flavors-0.024
Token and
Token ID290
Feature activation+1.11e-02

Pos logit contributions
 stalls+0.005
 hike+0.005
 banks+0.005
 stall+0.005
 fare+0.005
Neg logit contributions
 crying-0.003
 weeping-0.003
 bleeding-0.003
 streaming-0.003
ir-0.003
Token made
Token ID925
Feature activation+1.23e-02

Pos logit contributions
ir+0.017
 weeping+0.017
Ros+0.017
 bleeding+0.017
 crying+0.016
Neg logit contributions
 stalls-0.019
 hike-0.018
 flavors-0.018
 stall-0.018
 banks-0.017
Token her
Token ID607
Feature activation+7.36e-03

Pos logit contributions
 weeping+0.018
 bleeding+0.017
Ros+0.017
ir+0.017
 crying+0.017
Neg logit contributions
 stalls-0.022
 flavors-0.021
 hike-0.021
 stall-0.020
 banks-0.020
Token the
Token ID262
Feature activation+1.93e-02

Pos logit contributions
 weeping+0.012
Ros+0.012
 bleeding+0.012
 crying+0.011
ir+0.011
Neg logit contributions
 stalls-0.012
 hike-0.012
 stall-0.012
 banks-0.011
 flavors-0.011
Token happiest
Token ID49414
Feature activation+6.58e-02

Pos logit contributions
 streaming+0.031
Ros+0.031
 bleeding+0.031
 weeping+0.031
ir+0.031
Neg logit contributions
 stalls-0.031
 hike-0.030
 stall-0.030
 banks-0.030
 fare-0.029
Token bird
Token ID6512
Feature activation+8.78e-02

Pos logit contributions
Ros+0.089
 bleeding+0.086
ir+0.084
 weeping+0.084
apped+0.082
Neg logit contributions
 stalls-0.125
 stall-0.120
 hike-0.119
 flavors-0.116
 fare-0.113
Token in
Token ID287
Feature activation+2.93e-02

Pos logit contributions
 weeping+0.107
 crying+0.104
Ros+0.103
 bleeding+0.103
ir+0.101
Neg logit contributions
 stalls-0.180
 hike-0.170
 stall-0.170
 banks-0.168
 flavors-0.168
Token the
Token ID262
Feature activation+4.46e-02

Pos logit contributions
Ros+0.047
 bleeding+0.046
apped+0.045
 weeping+0.044
 crying+0.043
Neg logit contributions
 stalls-0.049
 stall-0.047
 hike-0.046
 flavors-0.044
 banks-0.043
Token world
Token ID995
Feature activation+2.26e-02

Pos logit contributions
apped+0.073
 bleeding+0.073
 streaming+0.073
Ros+0.072
ir+0.072
Neg logit contributions
 stalls-0.072
 hike-0.069
 stall-0.068
 banks-0.067
 bank-0.064
Token!
Token ID0
Feature activation+2.70e-02

Pos logit contributions
 crying+0.030
 weeping+0.030
ir+0.028
 streaming+0.028
 bleeding+0.028
Neg logit contributions
 stalls-0.044
 banks-0.042
 stall-0.041
 flavors-0.041
 hike-0.041
Token<|endoftext|>
Token ID50256
Feature activation+1.46e-02

Pos logit contributions
ir+0.047
 bleeding+0.047
 weeping+0.047
 crying+0.045
 blood+0.044
Neg logit contributions
 stalls-0.037
 stall-0.036
 fare-0.035
 hike-0.034
 list-0.034
Token each
Token ID1123
Feature activation+1.81e-02

Pos logit contributions
 crying+0.033
 weeping+0.033
 bleeding+0.032
 blood+0.030
Ros+0.030
Neg logit contributions
 stalls-0.065
 stall-0.064
 fare-0.063
 hike-0.062
 banks-0.061
Token other
Token ID584
Feature activation+1.52e-03

Pos logit contributions
 weeping+0.028
Ros+0.028
ir+0.028
 crying+0.028
 bleeding+0.027
Neg logit contributions
 stalls-0.031
 stall-0.029
 hike-0.029
 flavors-0.029
 banks-0.029
Token and
Token ID290
Feature activation-7.87e-04

Pos logit contributions
 crying+0.002
 weeping+0.002
 bleeding+0.002
 microphone+0.002
 streaming+0.002
Neg logit contributions
 stalls-0.003
 hike-0.003
 stall-0.003
 banks-0.003
 fare-0.003
Token the
Token ID262
Feature activation+2.35e-02

Pos logit contributions
 stalls+0.001
 hike+0.001
 stall+0.001
 flavors+0.001
 banks+0.001
Neg logit contributions
 weeping-0.001
Ros-0.001
ir-0.001
 crying-0.001
 bleeding-0.001
Token t
Token ID256
Feature activation+4.12e-02

Pos logit contributions
ir+0.039
apped+0.038
Ros+0.037
apping+0.036
 weeping+0.035
Neg logit contributions
 stalls-0.039
 stall-0.037
 flavors-0.037
 hike-0.037
 fare-0.036
Tokeneddy
Token ID21874
Feature activation+8.73e-02

Pos logit contributions
 Emily+0.051
rys+0.047
 weeping+0.046
 Ros+0.045
 Po+0.045
Neg logit contributions
cases-0.081
 fare-0.079
 belts-0.078
 stalls-0.076
 flavors-0.075
Token bear
Token ID6842
Feature activation+4.15e-02

Pos logit contributions
ir+0.148
Ros+0.143
apping+0.142
apped+0.139
aps+0.134
Neg logit contributions
 stalls-0.132
 flavors-0.127
 stall-0.125
 banks-0.125
 bank-0.124
Token.
Token ID13
Feature activation+3.55e-03

Pos logit contributions
 crying+0.063
 weeping+0.062
 bleeding+0.061
 microphone+0.056
 blood+0.055
Neg logit contributions
 stalls-0.076
 hike-0.071
 stall-0.070
 banks-0.070
 flavors-0.070
Token They
Token ID1119
Feature activation-2.48e-03

Pos logit contributions
 weeping+0.006
 bleeding+0.006
ir+0.006
 crying+0.006
Ros+0.006
Neg logit contributions
 stalls-0.005
 stall-0.005
 hike-0.005
 flavors-0.005
 banks-0.005
Token were
Token ID547
Feature activation+1.34e-02

Pos logit contributions
 stalls+0.004
 stall+0.004
 hike+0.004
 flavors+0.004
 banks+0.004
Neg logit contributions
 weeping-0.004
 crying-0.004
 bleeding-0.004
Ros-0.004
ir-0.004
Token happy
Token ID3772
Feature activation+4.55e-02

Pos logit contributions
Ros+0.021
ir+0.021
 weeping+0.019
el+0.019
apped+0.019
Neg logit contributions
 stalls-0.023
 stall-0.021
 hike-0.021
 flavors-0.021
 banks-0.021
Token and
Token ID290
Feature activation+7.82e-03

Pos logit contributions
 weeping+0.032
 crying+0.032
ir+0.031
Ros+0.031
 bleeding+0.031
Neg logit contributions
 stalls-0.039
 flavors-0.037
 hike-0.037
 fare-0.036
 stall-0.036
Token soon
Token ID2582
Feature activation+3.34e-02

Pos logit contributions
ir+0.013
Ros+0.013
 bleeding+0.013
 weeping+0.012
apped+0.012
Neg logit contributions
 stalls-0.012
 hike-0.012
 flavors-0.011
 banks-0.011
 stall-0.011
Token became
Token ID2627
Feature activation+4.37e-02

Pos logit contributions
 weeping+0.043
 crying+0.042
Ros+0.041
ir+0.040
 bleeding+0.040
Neg logit contributions
 stalls-0.068
 stall-0.065
 fare-0.064
 flavors-0.063
 banks-0.063
Token the
Token ID262
Feature activation+2.12e-02

Pos logit contributions
 weeping+0.060
 crying+0.059
Ros+0.058
ir+0.058
 bleeding+0.058
Neg logit contributions
 stalls-0.082
 stall-0.077
 hike-0.077
 flavors-0.077
 banks-0.076
Token strongest
Token ID12841
Feature activation+3.96e-02

Pos logit contributions
 weeping+0.033
Ros+0.033
ir+0.033
 crying+0.033
 bleeding+0.032
Neg logit contributions
 stalls-0.036
 hike-0.034
 stall-0.034
 flavors-0.033
 fare-0.033
Token bird
Token ID6512
Feature activation+8.73e-02

Pos logit contributions
Ros+0.057
apped+0.052
 weeping+0.051
 bleeding+0.051
ir+0.049
Neg logit contributions
 stalls-0.073
 stall-0.070
 flavors-0.068
 hike-0.068
 banks-0.066
Token in
Token ID287
Feature activation+4.18e-02

Pos logit contributions
Ros+0.111
 weeping+0.110
 bleeding+0.107
 crying+0.107
ir+0.106
Neg logit contributions
 stalls-0.169
 hike-0.160
 stall-0.160
 banks-0.157
 flavors-0.155
Token the
Token ID262
Feature activation+4.14e-02

Pos logit contributions
Ros+0.076
apped+0.072
 bleeding+0.071
ternally+0.068
 streaming+0.067
Neg logit contributions
 stalls-0.060
 stall-0.060
 hike-0.057
 banks-0.052
 flavors-0.051
Token sky
Token ID6766
Feature activation+4.94e-02

Pos logit contributions
apped+0.070
Ros+0.069
 streaming+0.068
 bleeding+0.068
 weeping+0.068
Neg logit contributions
 stalls-0.065
 hike-0.062
 stall-0.061
 banks-0.061
 bank-0.058
Token.
Token ID13
Feature activation+2.13e-02

Pos logit contributions
 weeping+0.085
 crying+0.083
 bleeding+0.082
Ros+0.082
ir+0.082
Neg logit contributions
 stalls-0.077
 hike-0.072
 stall-0.072
 banks-0.070
 flavors-0.070
Token The
Token ID383
Feature activation+5.97e-03

Pos logit contributions
 bleeding+0.031
 weeping+0.030
 crying+0.030
ir+0.030
Ros+0.030
Neg logit contributions
 stalls-0.038
 stall-0.036
 hike-0.036
 banks-0.036
 fare-0.035
Token listen
Token ID6004
Feature activation-1.44e-02

Pos logit contributions
 stalls+0.018
 stall+0.017
 hike+0.017
 banks+0.017
 flavors+0.017
Neg logit contributions
 weeping-0.022
 crying-0.021
Ros-0.021
 bleeding-0.021
ir-0.021
Token.
Token ID13
Feature activation-6.11e-03

Pos logit contributions
 stalls+0.029
 hike+0.028
 stall+0.028
 banks+0.028
 flavors+0.028
Neg logit contributions
 crying-0.019
 weeping-0.019
 bleeding-0.018
Ros-0.017
ir-0.017
Token She
Token ID1375
Feature activation-1.26e-02

Pos logit contributions
 stalls+0.010
 stall+0.010
 banks+0.009
 hike+0.009
 list+0.009
Neg logit contributions
ir-0.009
 bleeding-0.009
Ros-0.009
 microphone-0.009
 weeping-0.009
Token threw
Token ID9617
Feature activation-3.13e-03

Pos logit contributions
 stalls+0.023
 fare+0.022
 stall+0.022
 hike+0.022
 banks+0.021
Neg logit contributions
 crying-0.018
 bleeding-0.017
 weeping-0.017
 blood-0.017
Ros-0.016
Token a
Token ID257
Feature activation+2.14e-02

Pos logit contributions
 stalls+0.006
 stall+0.005
 flavors+0.005
 banks+0.005
 bank+0.005
Neg logit contributions
Ros-0.005
 weeping-0.005
apped-0.005
ir-0.005
apping-0.004
Token tant
Token ID24246
Feature activation+8.70e-02

Pos logit contributions
 weeping+0.031
Ros+0.031
 streaming+0.031
el+0.030
apped+0.030
Neg logit contributions
 stall-0.039
 stalls-0.038
 list-0.037
 bank-0.036
 ticket-0.036
Tokenrum
Token ID6582
Feature activation+3.16e-03

Pos logit contributions
 weeping+0.153
Ros+0.150
 bleeding+0.149
 crying+0.148
 microphone+0.147
Neg logit contributions
 flavors-0.112
 stalls-0.104
 stall-0.103
nic-0.102
 fare-0.100
Token in
Token ID287
Feature activation-7.62e-03

Pos logit contributions
Ros+0.006
ir+0.006
 weeping+0.005
 microphone+0.005
 streaming+0.005
Neg logit contributions
 stalls-0.005
 stall-0.004
 belts-0.004
 banks-0.004
 flavors-0.004
Token the
Token ID262
Feature activation+1.45e-02

Pos logit contributions
 stalls+0.012
 stall+0.012
 flavors+0.011
 shelves+0.011
 list+0.011
Neg logit contributions
apped-0.012
Ros-0.012
 weeping-0.012
apping-0.011
eddy-0.011
Token store
Token ID3650
Feature activation-4.00e-02

Pos logit contributions
Ros+0.022
ir+0.021
 weeping+0.021
 crying+0.021
 bleeding+0.020
Neg logit contributions
 stalls-0.026
 stall-0.025
 banks-0.025
 hike-0.024
 flavors-0.024
Token.
Token ID13
Feature activation-1.03e-02

Pos logit contributions
 hike+0.083
 flavors+0.081
 stalls+0.080
nic+0.080
 banks+0.078
Neg logit contributions
 crying-0.052
 weeping-0.048
 bleeding-0.046
apped-0.044
 blood-0.043
Token love
Token ID1842
Feature activation+4.32e-02

Pos logit contributions
 weeping+0.028
 crying+0.027
Ros+0.027
 bleeding+0.027
ir+0.026
Neg logit contributions
 stalls-0.041
 stall-0.039
 hike-0.038
 banks-0.038
 fare-0.037
Token my
Token ID616
Feature activation+1.27e-02

Pos logit contributions
apped+0.049
ir+0.049
Ros+0.048
 weeping+0.046
 bleeding+0.045
Neg logit contributions
 stalls-0.091
 flavors-0.088
 hike-0.088
 stall-0.085
 banks-0.083
Token new
Token ID649
Feature activation+3.68e-02

Pos logit contributions
 weeping+0.019
 crying+0.018
 bleeding+0.018
Ros+0.018
ir+0.018
Neg logit contributions
 stalls-0.023
 hike-0.022
 stall-0.022
 banks-0.021
 flavors-0.021
Token modern
Token ID3660
Feature activation+2.07e-02

Pos logit contributions
 weeping+0.049
 crying+0.048
Ros+0.047
 bleeding+0.047
ir+0.046
Neg logit contributions
 stalls-0.072
 hike-0.067
 stall-0.067
 banks-0.067
 flavors-0.066
Token t
Token ID256
Feature activation+3.28e-02

Pos logit contributions
 weeping+0.028
 crying+0.028
 bleeding+0.027
Ros+0.026
 microphone+0.026
Neg logit contributions
 stalls-0.040
 hike-0.038
 stall-0.038
 banks-0.037
 fare-0.037
Tokeneddy
Token ID21874
Feature activation+8.67e-02

Pos logit contributions
 weeping+0.024
 crying+0.023
ir+0.022
 streaming+0.022
 bleeding+0.022
Neg logit contributions
 stalls-0.079
 flavors-0.078
 fare-0.078
 stall-0.078
 belts-0.076
Token bear
Token ID6842
Feature activation+4.18e-02

Pos logit contributions
ir+0.116
Ros+0.115
 weeping+0.114
apped+0.112
 crying+0.111
Neg logit contributions
 stalls-0.162
 flavors-0.155
 stall-0.154
 hike-0.153
 banks-0.151
Token!"
Token ID2474
Feature activation-5.60e-03

Pos logit contributions
 crying+0.056
 weeping+0.056
 bleeding+0.055
 microphone+0.053
apped+0.052
Neg logit contributions
 stalls-0.081
 hike-0.078
 banks-0.076
 stall-0.074
 fare-0.074
Token<|endoftext|>
Token ID50256
Feature activation+6.97e-03

Pos logit contributions
 stalls+0.008
 stall+0.008
 list+0.007
 banks+0.007
 bank+0.007
Neg logit contributions
ir-0.010
w-0.009
 bleeding-0.009
ubby-0.009
el-0.009
TokenOnce
Token ID7454
Feature activation-3.03e-03

Pos logit contributions
 weeping+0.011
 bleeding+0.010
Ros+0.010
 crying+0.010
ir+0.010
Neg logit contributions
 stalls-0.012
 stall-0.012
 hike-0.012
 flavors-0.011
 banks-0.011
Token upon
Token ID2402
Feature activation+8.02e-03

Pos logit contributions
 hike+0.004
 stalls+0.004
 stall+0.004
 list+0.004
 flavors+0.004
Neg logit contributions
Ros-0.005
 bleeding-0.005
 streaming-0.005
 weeping-0.005
ir-0.005
Token she
Token ID673
Feature activation+1.42e-02

Pos logit contributions
ir+0.025
Ros+0.025
apped+0.024
 weeping+0.024
 bleeding+0.024
Neg logit contributions
 stalls-0.020
 stall-0.018
 hike-0.018
 flavors-0.017
 fare-0.017
Token had
Token ID550
Feature activation+2.76e-02

Pos logit contributions
ir+0.026
 weeping+0.026
Ros+0.025
apped+0.025
 bleeding+0.025
Neg logit contributions
 stalls-0.019
 hike-0.019
 banks-0.018
 flavors-0.018
 stall-0.018
Token lost
Token ID2626
Feature activation+1.61e-02

Pos logit contributions
 weeping+0.037
 crying+0.037
 bleeding+0.036
Ros+0.035
ir+0.034
Neg logit contributions
 stalls-0.054
 stall-0.051
 hike-0.051
 banks-0.050
 flavors-0.050
Token her
Token ID607
Feature activation+2.44e-02

Pos logit contributions
Ros+0.021
apped+0.020
ir+0.020
 weeping+0.020
 crying+0.019
Neg logit contributions
 stalls-0.033
 stall-0.031
 flavors-0.031
 banks-0.030
 hike-0.030
Token t
Token ID256
Feature activation+3.48e-02

Pos logit contributions
Ros+0.038
 weeping+0.038
ir+0.038
 crying+0.037
 bleeding+0.037
Neg logit contributions
 stalls-0.041
 stall-0.039
 hike-0.039
 flavors-0.038
 banks-0.038
Tokeneddy
Token ID21874
Feature activation+8.65e-02

Pos logit contributions
 Emily+0.035
 weeping+0.035
 sorry+0.034
 streaming+0.033
 crying+0.033
Neg logit contributions
 flavors-0.071
 stall-0.071
 stalls-0.071
cases-0.071
 fare-0.070
Token bear
Token ID6842
Feature activation+4.30e-02

Pos logit contributions
ir+0.131
apped+0.129
Ros+0.126
 weeping+0.123
apping+0.121
Neg logit contributions
 stalls-0.146
 stall-0.142
 flavors-0.141
 hike-0.139
 banks-0.138
Token.
Token ID13
Feature activation+2.08e-02

Pos logit contributions
 crying+0.060
 weeping+0.060
 bleeding+0.059
Ros+0.055
 microphone+0.055
Neg logit contributions
 stalls-0.083
 flavors-0.078
 hike-0.077
 stall-0.076
 banks-0.076
Token Her
Token ID2332
Feature activation+1.21e-02

Pos logit contributions
ir+0.034
 bleeding+0.032
 streaming+0.032
 microphone+0.032
el+0.032
Neg logit contributions
 stalls-0.033
 banks-0.032
 stall-0.031
 hike-0.031
 list-0.030
Token mom
Token ID1995
Feature activation+4.05e-02

Pos logit contributions
ir+0.021
Ros+0.021
apped+0.019
 streaming+0.019
apping+0.018
Neg logit contributions
 stalls-0.019
 flavors-0.019
 stall-0.019
 hike-0.018
 belts-0.018
Token told
Token ID1297
Feature activation-1.14e-02

Pos logit contributions
ir+0.075
Ros+0.071
aps+0.071
 weeping+0.071
apped+0.071
Neg logit contributions
 stalls-0.053
 stall-0.050
 hike-0.049
 bank-0.048
 flavors-0.048
Token comfortably
Token ID24341
Feature activation+5.12e-03

Pos logit contributions
 weeping+0.020
 crying+0.019
Ros+0.019
ir+0.019
 bleeding+0.019
Neg logit contributions
 stalls-0.042
 hike-0.040
 stall-0.040
 banks-0.040
 flavors-0.040
Token and
Token ID290
Feature activation+7.11e-03

Pos logit contributions
 weeping+0.006
 crying+0.006
ir+0.006
 bleeding+0.006
Ros+0.006
Neg logit contributions
 stalls-0.010
 flavors-0.010
 stall-0.010
 fare-0.010
 hike-0.010
Token happily
Token ID18177
Feature activation+2.69e-02

Pos logit contributions
ir+0.012
Ros+0.011
 weeping+0.011
 bleeding+0.011
apped+0.011
Neg logit contributions
 stalls-0.011
 hike-0.011
 flavors-0.011
 banks-0.011
 stall-0.011
Token in
Token ID287
Feature activation+6.56e-03

Pos logit contributions
 weeping+0.034
Ros+0.033
 crying+0.033
 bleeding+0.032
ir+0.032
Neg logit contributions
 stalls-0.054
 hike-0.051
 stall-0.051
 banks-0.050
 flavors-0.050
Token its
Token ID663
Feature activation+3.91e-02

Pos logit contributions
Ros+0.010
 weeping+0.010
 bleeding+0.010
apped+0.010
ir+0.010
Neg logit contributions
 stalls-0.011
 stall-0.011
 flavors-0.010
 hike-0.010
 banks-0.010
Token bird
Token ID6512
Feature activation+8.64e-02

Pos logit contributions
 weeping+0.054
 crying+0.053
 bleeding+0.053
Ros+0.052
ir+0.052
Neg logit contributions
 stalls-0.074
 hike-0.070
 stall-0.069
 flavors-0.068
 banks-0.068
Tokenc
Token ID66
Feature activation+2.78e-02

Pos logit contributions
 weeping+0.122
Ros+0.120
 bleeding+0.118
 crying+0.118
ir+0.116
Neg logit contributions
 stalls-0.156
 stall-0.148
 hike-0.147
 banks-0.146
 flavors-0.146
Tokenage
Token ID496
Feature activation+3.21e-02

Pos logit contributions
 weeping+0.051
 crying+0.050
 bleeding+0.050
Ros+0.049
 streaming+0.048
Neg logit contributions
 stalls-0.039
 banks-0.036
 stall-0.036
 hike-0.036
 flavors-0.036
Token.
Token ID13
Feature activation+1.76e-02

Pos logit contributions
 weeping+0.051
 crying+0.050
 bleeding+0.049
Ros+0.049
ir+0.048
Neg logit contributions
 stalls-0.054
 hike-0.051
 stall-0.051
 banks-0.050
 flavors-0.050
Token The
Token ID383
Feature activation+1.38e-02

Pos logit contributions
 weeping+0.024
 bleeding+0.024
 crying+0.024
Ros+0.023
ir+0.023
Neg logit contributions
 stalls-0.033
 hike-0.032
 stall-0.032
 banks-0.031
 flavors-0.031
Token boy
Token ID2933
Feature activation+4.50e-02

Pos logit contributions
Ros+0.020
 weeping+0.020
ir+0.019
 streaming+0.019
apped+0.019
Neg logit contributions
 stalls-0.025
 stall-0.024
 hike-0.024
 flavors-0.024
 banks-0.023
Token her
Token ID607
Feature activation+1.78e-02

Pos logit contributions
Ros+0.035
 weeping+0.034
ir+0.033
 bleeding+0.033
 crying+0.033
Neg logit contributions
 stalls-0.062
 stall-0.059
 hike-0.058
 banks-0.058
 flavors-0.058
Token feet
Token ID3625
Feature activation+4.81e-03

Pos logit contributions
Ros+0.025
ir+0.025
 weeping+0.024
 bleeding+0.023
apped+0.023
Neg logit contributions
 stalls-0.034
 stall-0.033
 hike-0.032
 banks-0.031
 flavors-0.031
Token and
Token ID290
Feature activation-9.06e-03

Pos logit contributions
 weeping+0.007
Ros+0.007
 crying+0.007
 bleeding+0.007
ir+0.007
Neg logit contributions
 stalls-0.009
 stall-0.008
 hike-0.008
 banks-0.008
 flavors-0.008
Token threw
Token ID9617
Feature activation-1.92e-03

Pos logit contributions
 stalls+0.016
 hike+0.015
 stall+0.015
 banks+0.014
 flavors+0.014
Neg logit contributions
 weeping-0.014
 crying-0.014
 bleeding-0.014
Ros-0.014
ir-0.013
Token a
Token ID257
Feature activation+1.84e-02

Pos logit contributions
 stalls+0.004
 stall+0.004
 flavors+0.004
 hike+0.004
 banks+0.004
Neg logit contributions
 weeping-0.002
Ros-0.002
 crying-0.002
 bleeding-0.002
ir-0.002
Token tant
Token ID24246
Feature activation+8.63e-02

Pos logit contributions
Ros+0.024
apped+0.023
 weeping+0.023
 streaming+0.022
ir+0.022
Neg logit contributions
 stall-0.037
 stalls-0.036
 bank-0.034
 list-0.034
 flavors-0.034
Tokenrum
Token ID6582
Feature activation+1.13e-03

Pos logit contributions
 weeping+0.146
Ros+0.145
 bleeding+0.144
 crying+0.143
 microphone+0.140
Neg logit contributions
 stalls-0.119
 flavors-0.118
 stall-0.116
 fare-0.112
nic-0.109
Token.
Token ID13
Feature activation-9.88e-04

Pos logit contributions
Ros+0.002
 weeping+0.002
ir+0.002
 crying+0.002
 bleeding+0.002
Neg logit contributions
 stalls-0.002
 stall-0.002
 hike-0.002
 banks-0.002
 flavors-0.002
Token She
Token ID1375
Feature activation-2.52e-03

Pos logit contributions
 stalls+0.002
 stall+0.002
 banks+0.002
 hike+0.002
 list+0.002
Neg logit contributions
 bleeding-0.001
Ros-0.001
 weeping-0.001
ir-0.001
 crying-0.001
Token grabbed
Token ID13176
Feature activation-8.76e-03

Pos logit contributions
 stalls+0.004
 hike+0.004
 stall+0.004
 flavors+0.004
 banks+0.004
Neg logit contributions
 weeping-0.004
Ros-0.004
ir-0.004
 crying-0.004
 bleeding-0.004
Token the
Token ID262
Feature activation-6.49e-03

Pos logit contributions
 stalls+0.015
 stall+0.014
 flavors+0.014
 banks+0.013
 hike+0.013
Neg logit contributions
 weeping-0.013
Ros-0.013
apping-0.013
apped-0.013
 bleeding-0.013
Token She
Token ID1375
Feature activation-2.31e-03

Pos logit contributions
 stalls+0.027
 stall+0.026
 hike+0.024
 list+0.023
 banks+0.023
Neg logit contributions
ir-0.028
Ros-0.025
ternally-0.025
el-0.025
 streaming-0.025
Token c
Token ID269
Feature activation+1.40e-02

Pos logit contributions
 stalls+0.005
 hike+0.004
 stall+0.004
 banks+0.004
 fare+0.004
Neg logit contributions
 weeping-0.003
 crying-0.003
 bleeding-0.003
Ros-0.003
apped-0.003
Tokenuddled
Token ID32745
Feature activation+2.08e-02

Pos logit contributions
 weeping+0.026
 crying+0.026
 Emily+0.026
Ros+0.026
 Ros+0.025
Neg logit contributions
 stalls-0.019
 belts-0.018
cases-0.018
 stall-0.017
 flavors-0.017
Token her
Token ID607
Feature activation+3.46e-02

Pos logit contributions
ir+0.029
Ros+0.027
el+0.026
 weeping+0.026
 streaming+0.026
Neg logit contributions
 stalls-0.040
 stall-0.038
 hike-0.036
 belts-0.036
 banks-0.035
Token t
Token ID256
Feature activation+4.16e-02

Pos logit contributions
ir+0.065
Ros+0.063
apped+0.061
itt+0.059
 streaming+0.059
Neg logit contributions
 stalls-0.048
 belts-0.048
 hike-0.048
 stall-0.047
 flavors-0.046
Tokeneddy
Token ID21874
Feature activation+8.61e-02

Pos logit contributions
 Emily+0.053
 streaming+0.051
 sorry+0.050
 weeping+0.050
 Ros+0.048
Neg logit contributions
 belts-0.081
cases-0.076
 fare-0.075
belt-0.074
 stalls-0.072
Token bear
Token ID6842
Feature activation+3.90e-02

Pos logit contributions
ir+0.148
Ros+0.145
apped+0.140
apping+0.138
itt+0.138
Neg logit contributions
 stalls-0.128
 stall-0.123
 banks-0.121
 hike-0.121
 belts-0.119
Token and
Token ID290
Feature activation-7.06e-04

Pos logit contributions
 weeping+0.053
 crying+0.051
 bleeding+0.051
Ros+0.051
ir+0.050
Neg logit contributions
 stalls-0.075
 hike-0.070
 stall-0.070
 flavors-0.070
 banks-0.069
Token read
Token ID1100
Feature activation-1.43e-02

Pos logit contributions
 stalls+0.001
 stall+0.001
 hike+0.001
 banks+0.001
 fare+0.001
Neg logit contributions
 weeping-0.001
 crying-0.001
 bleeding-0.001
Ros-0.001
 blood-0.001
Token her
Token ID607
Feature activation+4.97e-03

Pos logit contributions
 stalls+0.031
 hike+0.030
 stall+0.030
 banks+0.030
 flavors+0.029
Neg logit contributions
 weeping-0.015
 crying-0.015
 bleeding-0.015
Ros-0.015
ir-0.014
Token book
Token ID1492
Feature activation-1.34e-02

Pos logit contributions
ir+0.008
Ros+0.008
 weeping+0.008
apped+0.008
 crying+0.008
Neg logit contributions
 stalls-0.008
 hike-0.007
 flavors-0.007
 stall-0.007
 belts-0.007
Token
Token ID198
Feature activation+1.75e-02

Pos logit contributions
 weeping+0.028
 bleeding+0.027
ir+0.027
 crying+0.027
 streaming+0.026
Neg logit contributions
 stalls-0.033
 stall-0.031
 hike-0.030
 banks-0.030
 flavors-0.030
TokenThe
Token ID464
Feature activation+6.03e-03

Pos logit contributions
 weeping+0.026
 crying+0.025
 bleeding+0.025
ir+0.024
Ros+0.024
Neg logit contributions
 stalls-0.032
 stall-0.030
 hike-0.030
 flavors-0.029
 banks-0.029
Token end
Token ID886
Feature activation+2.84e-02

Pos logit contributions
 weeping+0.010
 crying+0.010
 bleeding+0.009
Ros+0.009
 microphone+0.009
Neg logit contributions
 stalls-0.010
 stall-0.009
 hike-0.009
 banks-0.009
 flavors-0.009
Token.
Token ID13
Feature activation+3.01e-02

Pos logit contributions
 crying+0.037
 bleeding+0.037
 weeping+0.036
 Emily+0.034
 Sad+0.034
Neg logit contributions
 stalls-0.059
 banks-0.057
 bank-0.055
 stall-0.054
 hike-0.054
Token<|endoftext|>
Token ID50256
Feature activation+1.07e-02

Pos logit contributions
 weeping+0.048
ir+0.048
 bleeding+0.047
 crying+0.046
Ros+0.046
Neg logit contributions
 stalls-0.049
 stall-0.047
 hike-0.045
 fare-0.045
 flavors-0.044
TokenRe
Token ID3041
Feature activation-9.77e-02

Pos logit contributions
 weeping+0.016
 bleeding+0.016
Ros+0.016
 crying+0.016
ir+0.016
Neg logit contributions
 stalls-0.019
 stall-0.018
 hike-0.017
 banks-0.017
 flavors-0.017
Tokenbecca
Token ID20627
Feature activation-4.23e-02

Pos logit contributions
 stalls+0.138
 fare+0.128
 banks+0.128
 hike+0.127
 stall+0.127
Neg logit contributions
ir-0.179
el-0.174
apped-0.172
Ros-0.171
 weeping-0.169
Token was
Token ID373
Feature activation+1.71e-02

Pos logit contributions
 fare+0.085
 stalls+0.083
 banks+0.080
 stall+0.080
 hike+0.079
Neg logit contributions
 crying-0.051
 weeping-0.051
el-0.051
ir-0.049
apped-0.048
Token feeling
Token ID4203
Feature activation+1.63e-02

Pos logit contributions
Ros+0.028
ir+0.027
 blood+0.025
 accidents+0.024
anish+0.024
Neg logit contributions
 stalls-0.027
 stall-0.026
 hike-0.024
 packed-0.024
 banks-0.023
Token anxious
Token ID18116
Feature activation-1.35e-02

Pos logit contributions
 weeping+0.021
Ros+0.021
ir+0.020
 crying+0.020
 bleeding+0.020
Neg logit contributions
 stalls-0.032
 stall-0.030
 hike-0.030
 banks-0.030
 flavors-0.029
Token.
Token ID13
Feature activation-6.68e-03

Pos logit contributions
 stalls+0.025
 stall+0.024
 hike+0.023
 flavors+0.023
 banks+0.023
Neg logit contributions
 weeping-0.021
 crying-0.021
 bleeding-0.019
 streaming-0.019
apped-0.019
Token smiled
Token ID13541
Feature activation+1.88e-02

Pos logit contributions
 weeping+0.005
 crying+0.005
Ros+0.005
 bleeding+0.005
ir+0.005
Neg logit contributions
 stalls-0.005
 hike-0.005
 stall-0.005
 banks-0.004
 flavors-0.004
Token and
Token ID290
Feature activation-9.48e-03

Pos logit contributions
 weeping+0.027
Ros+0.027
 crying+0.027
ir+0.027
 bleeding+0.027
Neg logit contributions
 stalls-0.034
 stall-0.032
 hike-0.032
 flavors-0.031
 banks-0.031
Token said
Token ID531
Feature activation-2.72e-02

Pos logit contributions
 stalls+0.018
 banks+0.017
 fare+0.017
 hike+0.017
 stall+0.017
Neg logit contributions
 crying-0.013
 weeping-0.013
 Ros-0.013
Ros-0.013
 bleeding-0.013
Token,
Token ID11
Feature activation-2.55e-02

Pos logit contributions
 stalls+0.052
 hike+0.050
 banks+0.049
 flavors+0.049
 fare+0.048
Neg logit contributions
 weeping-0.038
 crying-0.038
apped-0.037
 bleeding-0.036
Ros-0.036
Token "
Token ID366
Feature activation-6.16e-03

Pos logit contributions
 stalls+0.044
 hike+0.042
 flavors+0.042
 banks+0.041
 fare+0.041
Neg logit contributions
 weeping-0.042
 crying-0.042
 bleeding-0.040
Ros-0.039
apped-0.038
TokenMoney
Token ID26788
Feature activation-9.49e-02

Pos logit contributions
 stalls+0.014
 hike+0.014
 banks+0.014
 stall+0.014
 list+0.013
Neg logit contributions
Ros-0.006
 weeping-0.005
 bleeding-0.005
el-0.005
apped-0.005
Token is
Token ID318
Feature activation-2.06e-02

Pos logit contributions
 hike+0.210
 stalls+0.210
 banks+0.205
 stall+0.201
 hikes+0.194
Neg logit contributions
el-0.092
ing-0.085
 crying-0.084
 blood-0.084
 bleeding-0.084
Token like
Token ID588
Feature activation-8.32e-03

Pos logit contributions
 stalls+0.046
 hike+0.046
 banks+0.045
 stall+0.044
 fare+0.042
Neg logit contributions
 Ros-0.022
 crying-0.021
 weeping-0.021
 bleeding-0.020
 microphone-0.019
Token coins
Token ID10796
Feature activation-5.20e-02

Pos logit contributions
 stalls+0.017
 stall+0.016
 hike+0.016
 banks+0.016
 fare+0.016
Neg logit contributions
 weeping-0.010
 crying-0.010
 bleeding-0.010
Ros-0.010
 blood-0.010
Token that
Token ID326
Feature activation+2.01e-02

Pos logit contributions
 stalls+0.106
 hike+0.106
 fare+0.100
 banks+0.100
 stall+0.099
Neg logit contributions
 crying-0.060
 weeping-0.058
 bleeding-0.057
 microphone-0.055
el-0.054
Token help
Token ID1037
Feature activation-1.21e-02

Pos logit contributions
 crying+0.025
 weeping+0.025
 microphone+0.024
el+0.024
 bleeding+0.024
Neg logit contributions
 stalls-0.039
 hike-0.037
 stall-0.036
 banks-0.036
nic-0.035
Token to
Token ID284
Feature activation-5.32e-02

Pos logit contributions
 stalls+0.116
 banks+0.110
 flavors+0.108
 stall+0.107
 hike+0.107
Neg logit contributions
 weeping-0.073
 crying-0.073
Ros-0.070
 bleeding-0.068
apped-0.066
Token help
Token ID1037
Feature activation-2.07e-02

Pos logit contributions
 stalls+0.090
 stall+0.084
 hike+0.084
 banks+0.083
 flavors+0.083
Neg logit contributions
 weeping-0.084
 crying-0.082
Ros-0.082
 bleeding-0.082
ir-0.081
Token with
Token ID351
Feature activation-1.69e-02

Pos logit contributions
 stalls+0.034
 hike+0.032
 stall+0.032
 banks+0.031
 flavors+0.031
Neg logit contributions
 weeping-0.034
Ros-0.033
 crying-0.033
 bleeding-0.033
ir-0.032
Token the
Token ID262
Feature activation-3.06e-02

Pos logit contributions
 stalls+0.035
 stall+0.033
 hike+0.033
 banks+0.033
 flavors+0.032
Neg logit contributions
 weeping-0.021
 crying-0.020
 bleeding-0.020
Ros-0.020
ir-0.020
Token grocery
Token ID16918
Feature activation-8.15e-02

Pos logit contributions
 stalls+0.051
 hike+0.047
 stall+0.047
 banks+0.047
 flavors+0.046
Neg logit contributions
 weeping-0.049
 crying-0.049
 bleeding-0.048
Ros-0.048
ir-0.047
Token shopping
Token ID9735
Feature activation-9.44e-02

Pos logit contributions
land+0.161
 flavors+0.160
nic+0.157
 hikes+0.155
uries+0.153
Neg logit contributions
 crying-0.094
 together-0.089
ucky-0.078
 Ros-0.077
 weeping-0.075
Token.
Token ID13
Feature activation-4.94e-04

Pos logit contributions
 hike+0.208
nic+0.203
 banks+0.197
 hikes+0.197
 Street+0.195
Neg logit contributions
apped-0.094
 together-0.092
 crying-0.084
uck-0.082
el-0.077
Token Lily
Token ID20037
Feature activation-4.78e-03

Pos logit contributions
 stalls+0.001
 hike+0.001
 fare+0.001
 flavors+0.001
 stall+0.001
Neg logit contributions
 crying-0.000
Ros-0.000
 weeping-0.000
 bleeding-0.000
apped-0.000
Token was
Token ID373
Feature activation+6.27e-03

Pos logit contributions
 fare+0.010
 stalls+0.010
land+0.010
 stall+0.009
 banks+0.009
Neg logit contributions
 crying-0.006
uck-0.005
apped-0.005
Ros-0.005
el-0.005
Token excited
Token ID6568
Feature activation-1.29e-02

Pos logit contributions
 weeping+0.009
Ros+0.009
 crying+0.008
 bleeding+0.008
ir+0.008
Neg logit contributions
 stalls-0.012
 stall-0.011
 hike-0.011
 banks-0.011
 flavors-0.011
Token to
Token ID284
Feature activation-8.39e-03

Pos logit contributions
 stalls+0.027
 fare+0.026
 banks+0.026
 stall+0.026
 flavors+0.025
Neg logit contributions
 crying-0.016
 weeping-0.016
 bleeding-0.015
Ros-0.014
 Emily-0.014
Token if
Token ID611
Feature activation+1.77e-02

Pos logit contributions
 stalls+0.094
 banks+0.091
 flavors+0.091
 hike+0.088
 stall+0.087
Neg logit contributions
 crying-0.067
 weeping-0.066
 Emily-0.058
 Ros-0.058
 microphone-0.058
Token she
Token ID673
Feature activation-5.00e-02

Pos logit contributions
 crying+0.025
 weeping+0.025
Ros+0.024
 bleeding+0.023
 Ros+0.023
Neg logit contributions
 stalls-0.033
 banks-0.032
 stall-0.032
 fare-0.031
 flavors-0.031
Token wanted
Token ID2227
Feature activation-1.51e-02

Pos logit contributions
 stalls+0.084
 stall+0.078
 banks+0.078
 hike+0.076
 fare+0.074
Neg logit contributions
 weeping-0.086
 crying-0.085
 bleeding-0.082
Ros-0.081
apped-0.079
Token to
Token ID284
Feature activation-3.11e-02

Pos logit contributions
 stalls+0.028
 stall+0.027
 hike+0.027
 banks+0.026
 flavors+0.026
Neg logit contributions
 weeping-0.021
 crying-0.021
Ros-0.020
 bleeding-0.020
ir-0.020
Token go
Token ID467
Feature activation+4.16e-03

Pos logit contributions
 stalls+0.049
 hike+0.048
 flavors+0.047
 banks+0.046
 stall+0.046
Neg logit contributions
 weeping-0.051
 bleeding-0.050
 crying-0.049
Ros-0.049
ir-0.049
Token shopping
Token ID9735
Feature activation-9.34e-02

Pos logit contributions
 weeping+0.005
 crying+0.005
Ros+0.005
 bleeding+0.005
ir+0.005
Neg logit contributions
 stalls-0.008
 stall-0.008
 hike-0.008
 flavors-0.008
 banks-0.008
Token.
Token ID13
Feature activation-1.55e-02

Pos logit contributions
 hike+0.206
nic+0.205
 banks+0.203
 stalls+0.199
 pm+0.192
Neg logit contributions
 together-0.091
 crying-0.089
apped-0.080
 weeping-0.077
el-0.077
Token She
Token ID1375
Feature activation-2.11e-02

Pos logit contributions
 stalls+0.029
 hike+0.028
 flavors+0.028
 fare+0.028
 banks+0.027
Neg logit contributions
 weeping-0.022
 crying-0.022
Ros-0.021
 bleeding-0.020
apped-0.020
Token nodded
Token ID14464
Feature activation+9.09e-03

Pos logit contributions
 banks+0.043
 stalls+0.043
 fare+0.040
 experiments+0.039
 stall+0.039
Neg logit contributions
 crying-0.027
 weeping-0.026
apped-0.026
 Emily-0.025
Ros-0.025
Token vigorously
Token ID31609
Feature activation-3.32e-02

Pos logit contributions
 weeping+0.012
 crying+0.012
 bleeding+0.011
Ros+0.011
 streaming+0.011
Neg logit contributions
 stalls-0.018
 hike-0.018
 banks-0.018
 stall-0.017
 flavors-0.017
Token.
Token ID13
Feature activation-1.35e-02

Pos logit contributions
 stalls+0.074
 flavors+0.072
 hike+0.070
 banks+0.069
 stall+0.068
Neg logit contributions
 crying-0.042
 weeping-0.039
 bleeding-0.034
 streaming-0.033
apped-0.032
Token mom
Token ID1995
Feature activation+1.45e-02

Pos logit contributions
 stalls+0.034
 stall+0.032
 banks+0.032
 hike+0.032
 flavors+0.031
Neg logit contributions
 weeping-0.017
 crying-0.017
 bleeding-0.016
Ros-0.016
 microphone-0.015
Token agreed
Token ID4987
Feature activation-2.84e-02

Pos logit contributions
 crying+0.022
 weeping+0.021
 bleeding+0.021
 streaming+0.020
Ros+0.020
Neg logit contributions
 stalls-0.028
 banks-0.026
 stall-0.026
 hike-0.026
 flavors-0.025
Token to
Token ID284
Feature activation-2.75e-02

Pos logit contributions
 stalls+0.066
 banks+0.062
 stall+0.062
 hike+0.061
 fare+0.060
Neg logit contributions
 crying-0.034
 weeping-0.031
 bleeding-0.028
 streaming-0.027
 Emily-0.026
Token take
Token ID1011
Feature activation+1.46e-03

Pos logit contributions
 stalls+0.049
 stall+0.044
 banks+0.044
 flavors+0.043
 hike+0.042
Neg logit contributions
 crying-0.044
 weeping-0.044
 bleeding-0.042
ir-0.041
Ros-0.041
Token him
Token ID683
Feature activation-6.69e-03

Pos logit contributions
 weeping+0.002
 bleeding+0.002
Ros+0.002
ir+0.002
 crying+0.002
Neg logit contributions
 stalls-0.003
 hike-0.002
 stall-0.002
 goods-0.002
 fare-0.002
Token shopping
Token ID9735
Feature activation-9.31e-02

Pos logit contributions
 stalls+0.015
 flavors+0.015
 stall+0.015
 hike+0.014
 banks+0.014
Neg logit contributions
 crying-0.007
 weeping-0.007
 bleeding-0.006
 Ros-0.006
 microphone-0.006
Token for
Token ID329
Feature activation-1.40e-02

Pos logit contributions
 hike+0.208
nic+0.205
 banks+0.198
bread+0.192
 stalls+0.191
Neg logit contributions
 together-0.092
 crying-0.089
apped-0.083
el-0.072
 weeping-0.072
Token a
Token ID257
Feature activation-9.83e-03

Pos logit contributions
 stalls+0.029
 stall+0.027
 hike+0.027
 banks+0.027
 flavors+0.027
Neg logit contributions
 weeping-0.017
 crying-0.017
Ros-0.017
 bleeding-0.016
ir-0.016
Token new
Token ID649
Feature activation-1.39e-03

Pos logit contributions
 stalls+0.017
 banks+0.016
 flavors+0.016
 stall+0.014
 fare+0.014
Neg logit contributions
 crying-0.016
 weeping-0.016
Ros-0.015
 microphone-0.015
 bleeding-0.014
Token uniform
Token ID8187
Feature activation-1.54e-02

Pos logit contributions
 banks+0.003
 stalls+0.003
bread+0.002
 pm+0.002
 flavors+0.002
Neg logit contributions
 crying-0.002
 microphone-0.002
 weeping-0.002
Ros-0.002
 laser-0.002
Token.
Token ID13
Feature activation+6.86e-03

Pos logit contributions
 stalls+0.032
 hike+0.031
 banks+0.031
 flavors+0.030
 stall+0.030
Neg logit contributions
 crying-0.019
 weeping-0.018
apped-0.017
 bleeding-0.017
 Ros-0.016
Token to
Token ID284
Feature activation-5.57e-02

Pos logit contributions
 stalls+0.115
 banks+0.111
 flavors+0.108
 stall+0.107
 hike+0.107
Neg logit contributions
 crying-0.071
 weeping-0.071
Ros-0.067
 Ros-0.064
 bleeding-0.064
Token help
Token ID1037
Feature activation-2.10e-02

Pos logit contributions
 stalls+0.079
 hike+0.079
 list+0.077
 stall+0.077
 flavors+0.076
Neg logit contributions
 weeping-0.099
 bleeding-0.098
 crying-0.095
ir-0.094
 streaming-0.093
Token with
Token ID351
Feature activation-1.77e-02

Pos logit contributions
 stalls+0.038
 hike+0.036
 stall+0.036
 banks+0.035
 flavors+0.035
Neg logit contributions
 weeping-0.031
 bleeding-0.030
Ros-0.029
 crying-0.029
ir-0.029
Token the
Token ID262
Feature activation-2.93e-02

Pos logit contributions
 stalls+0.038
 stall+0.036
 hike+0.036
 banks+0.035
 flavors+0.035
Neg logit contributions
 weeping-0.020
 crying-0.020
Ros-0.020
 bleeding-0.020
ir-0.019
Token grocery
Token ID16918
Feature activation-8.00e-02

Pos logit contributions
 stalls+0.056
 hike+0.053
 stall+0.053
 flavors+0.053
 banks+0.052
Neg logit contributions
 weeping-0.039
Ros-0.039
ir-0.038
 bleeding-0.038
 crying-0.038
Token shopping
Token ID9735
Feature activation-9.27e-02

Pos logit contributions
land+0.149
nic+0.147
 flavors+0.146
 stalls+0.142
 hike+0.141
Neg logit contributions
 crying-0.104
 Ros-0.098
 Emily-0.096
 together-0.092
 microphone-0.091
Token.
Token ID13
Feature activation-2.87e-03

Pos logit contributions
 hike+0.199
nic+0.199
 banks+0.192
 Street+0.191
 hikes+0.187
Neg logit contributions
apped-0.097
 together-0.091
 crying-0.089
uck-0.084
el-0.084
Token Lily
Token ID20037
Feature activation-4.30e-03

Pos logit contributions
 stalls+0.007
 hike+0.006
 stall+0.006
 flavors+0.006
 banks+0.006
Neg logit contributions
 weeping-0.003
 crying-0.003
Ros-0.003
 bleeding-0.003
ir-0.003
Token was
Token ID373
Feature activation+5.76e-03

Pos logit contributions
 stalls+0.009
 fare+0.008
land+0.008
 stall+0.008
 banks+0.008
Neg logit contributions
 crying-0.006
Ros-0.006
 weeping-0.005
apped-0.005
el-0.005
Token excited
Token ID6568
Feature activation-1.36e-02

Pos logit contributions
 weeping+0.008
Ros+0.008
ir+0.008
 crying+0.008
 bleeding+0.008
Neg logit contributions
 stalls-0.011
 stall-0.010
 hike-0.010
 flavors-0.010
 banks-0.010
Token to
Token ID284
Feature activation-8.39e-03

Pos logit contributions
 stalls+0.028
 stall+0.026
 banks+0.026
 fare+0.026
 hike+0.026
Neg logit contributions
 crying-0.018
 weeping-0.018
 bleeding-0.017
Ros-0.016
ir-0.016
Token him
Token ID683
Feature activation-4.65e-02

Pos logit contributions
 stalls+0.039
 stall+0.037
 hike+0.037
 banks+0.037
 flavors+0.037
Neg logit contributions
 weeping-0.021
 crying-0.020
 bleeding-0.020
Ros-0.020
ir-0.019
Token if
Token ID611
Feature activation+1.36e-02

Pos logit contributions
 stalls+0.100
 banks+0.095
 flavors+0.094
 stall+0.093
 fare+0.092
Neg logit contributions
 crying-0.056
 weeping-0.056
 Ros-0.053
Ros-0.052
 Emily-0.051
Token he
Token ID339
Feature activation-5.32e-02

Pos logit contributions
 crying+0.024
 weeping+0.024
 Ros+0.024
Ros+0.024
 bleeding+0.023
Neg logit contributions
 stalls-0.021
 stall-0.020
 banks-0.020
 fare-0.020
 flavors-0.019
Token did
Token ID750
Feature activation-3.82e-02

Pos logit contributions
 stalls+0.095
 banks+0.088
 stall+0.088
 fare+0.086
 hike+0.084
Neg logit contributions
 crying-0.085
 weeping-0.084
 bleeding-0.083
Ros-0.081
apped-0.080
Token his
Token ID465
Feature activation-6.26e-02

Pos logit contributions
 stalls+0.094
 stall+0.090
 banks+0.089
 fare+0.086
 hike+0.086
Neg logit contributions
 crying-0.038
 bleeding-0.036
 weeping-0.034
 blood-0.033
 Ros-0.031
Token homework
Token ID26131
Feature activation-9.26e-02

Pos logit contributions
 stalls+0.111
 stall+0.102
 banks+0.096
 fare+0.092
nic+0.091
Neg logit contributions
 crying-0.103
 weeping-0.098
 bleeding-0.098
 microphone-0.093
 Ros-0.092
Token.
Token ID13
Feature activation-9.12e-05

Pos logit contributions
 stalls+0.193
 banks+0.183
 stall+0.181
 hike+0.178
 flavors+0.176
Neg logit contributions
 crying-0.117
 bleeding-0.108
 weeping-0.106
apped-0.099
 microphone-0.098
Token Tim
Token ID5045
Feature activation+1.91e-03

Pos logit contributions
 stalls+0.000
 fare+0.000
 stall+0.000
 hike+0.000
 banks+0.000
Neg logit contributions
 crying-0.000
 weeping-0.000
Ros-0.000
 Ros-0.000
 bleeding-0.000
Tokenmy
Token ID1820
Feature activation+3.96e-03

Pos logit contributions
 weeping+0.002
Ros+0.002
ir+0.002
 crying+0.002
 bleeding+0.002
Neg logit contributions
 stalls-0.004
 flavors-0.004
 hike-0.004
 stall-0.004
 banks-0.004
Token said
Token ID531
Feature activation-2.89e-02

Pos logit contributions
 crying+0.006
 weeping+0.006
 bleeding+0.006
Ros+0.005
apped+0.005
Neg logit contributions
 stalls-0.008
 fare-0.007
 stall-0.007
 banks-0.007
 hike-0.007
Token no
Token ID645
Feature activation-2.33e-02

Pos logit contributions
 stalls+0.051
 banks+0.048
 hike+0.047
 stall+0.047
 fare+0.046
Neg logit contributions
 crying-0.048
 weeping-0.047
 bleeding-0.047
Ros-0.045
apped-0.044
Token Rachel
Token ID15984
Feature activation-1.20e-02

Pos logit contributions
 stalls+0.038
 stall+0.036
 hike+0.036
 flavors+0.036
 banks+0.035
Neg logit contributions
 weeping-0.016
 bleeding-0.016
 crying-0.015
Ros-0.015
ir-0.015
Token would
Token ID561
Feature activation+1.32e-02

Pos logit contributions
 fare+0.023
 stalls+0.022
 stall+0.021
 goods+0.021
 list+0.021
Neg logit contributions
 crying-0.016
 weeping-0.016
 microphone-0.016
 Emily-0.015
el-0.015
Token also
Token ID635
Feature activation-2.71e-02

Pos logit contributions
 crying+0.019
 weeping+0.019
 bleeding+0.018
ir+0.017
 microphone+0.017
Neg logit contributions
 stalls-0.025
cases-0.023
 flavors-0.023
 stall-0.023
 fare-0.023
Token teach
Token ID4545
Feature activation-1.51e-02

Pos logit contributions
 stalls+0.048
 stall+0.044
 flavors+0.044
 fare+0.044
 banks+0.044
Neg logit contributions
 weeping-0.041
 crying-0.041
 bleeding-0.039
ir-0.038
 streaming-0.038
Token Jack
Token ID3619
Feature activation-2.15e-02

Pos logit contributions
 stalls+0.029
 hike+0.027
 flavors+0.027
 stall+0.027
 experiments+0.026
Neg logit contributions
 bleeding-0.020
 weeping-0.019
Ros-0.019
ir-0.019
apped-0.019
Token maths
Token ID47761
Feature activation-9.20e-02

Pos logit contributions
 stalls+0.046
 fare+0.045
 stall+0.044
 banks+0.044
 hike+0.043
Neg logit contributions
 crying-0.025
 weeping-0.024
 Emily-0.022
 bleeding-0.021
aps-0.021
Token and
Token ID290
Feature activation-1.57e-02

Pos logit contributions
 stalls+0.184
 ranc+0.183
 banks+0.182
 hike+0.181
 stall+0.179
Neg logit contributions
 crying-0.101
 microphone-0.098
 together-0.096
aps-0.096
 accidents-0.094
Token stories
Token ID3923
Feature activation-4.21e-02

Pos logit contributions
 stalls+0.027
 stall+0.026
 hike+0.025
 banks+0.025
 fare+0.025
Neg logit contributions
 weeping-0.024
 crying-0.024
Ros-0.024
 bleeding-0.023
apped-0.022
Token.
Token ID13
Feature activation-1.16e-02

Pos logit contributions
 stalls+0.087
 banks+0.086
 stall+0.082
 hike+0.082
 ranc+0.080
Neg logit contributions
 crying-0.050
 weeping-0.049
apped-0.048
 streaming-0.048
 bleeding-0.047
Token Jack
Token ID3619
Feature activation+4.10e-03

Pos logit contributions
 stalls+0.023
 stall+0.021
 hike+0.021
 banks+0.021
 flavors+0.021
Neg logit contributions
 weeping-0.015
 bleeding-0.015
 crying-0.015
ir-0.015
Ros-0.015
Token loved
Token ID6151
Feature activation+2.85e-02

Pos logit contributions
 weeping+0.006
 crying+0.006
apped+0.006
el+0.006
 microphone+0.006
Neg logit contributions
 stalls-0.007
 fare-0.007
 stall-0.007
 list-0.007
 hike-0.007
Token
Token ID198
Feature activation-1.05e-02

Pos logit contributions
 stalls+0.011
 experiments+0.011
 hike+0.011
 list+0.011
 banks+0.011
Neg logit contributions
ir-0.013
 bleeding-0.013
 weeping-0.012
w-0.012
 blood-0.012
Token
Token ID198
Feature activation-7.86e-04

Pos logit contributions
 hike+0.013
 stalls+0.012
 stall+0.012
books+0.011
 experiments+0.011
Neg logit contributions
 bleeding-0.021
 streaming-0.021
 weeping-0.021
 crying-0.020
 blood-0.019
TokenAfter
Token ID3260
Feature activation-2.24e-02

Pos logit contributions
 stalls+0.001
 hike+0.001
 stall+0.001
 banks+0.001
nic+0.001
Neg logit contributions
 weeping-0.001
 crying-0.001
 bleeding-0.001
 blood-0.001
 streaming-0.001
Token they
Token ID484
Feature activation-5.33e-02

Pos logit contributions
 stalls+0.042
 stall+0.039
 hike+0.039
 flavors+0.039
 banks+0.039
Neg logit contributions
 weeping-0.032
 crying-0.031
Ros-0.031
 bleeding-0.030
ir-0.030
Token finished
Token ID5201
Feature activation-2.10e-02

Pos logit contributions
 stalls+0.095
 banks+0.089
 fare+0.088
 flavors+0.088
 stall+0.087
Neg logit contributions
ir-0.077
 weeping-0.076
 crying-0.075
Ros-0.075
el-0.075
Token shopping
Token ID9735
Feature activation-9.15e-02

Pos logit contributions
 stalls+0.041
 fare+0.041
 flavors+0.040
 banks+0.040
nic+0.039
Neg logit contributions
 weeping-0.028
 crying-0.027
el-0.025
ir-0.025
 Ros-0.024
Token,
Token ID11
Feature activation+9.34e-03

Pos logit contributions
nic+0.185
 hike+0.180
 banks+0.179
 stalls+0.178
 fare+0.175
Neg logit contributions
 crying-0.110
apped-0.110
 weeping-0.104
el-0.100
 Ros-0.097
Token they
Token ID484
Feature activation-2.62e-02

Pos logit contributions
ir+0.012
ez+0.011
 bleeding+0.011
ivering+0.011
 weeping+0.011
Neg logit contributions
 experiments-0.019
 hike-0.019
 stalls-0.018
 shelves-0.017
 list-0.017
Token went
Token ID1816
Feature activation+9.41e-03

Pos logit contributions
 stalls+0.046
 fare+0.045
 flavors+0.044
 banks+0.044
 experiments+0.043
Neg logit contributions
apped-0.038
 weeping-0.038
 crying-0.038
Ros-0.037
ir-0.036
Token home
Token ID1363
Feature activation-1.34e-02

Pos logit contributions
 weeping+0.011
 crying+0.011
Ros+0.011
 bleeding+0.010
apped+0.010
Neg logit contributions
 stalls-0.020
 flavors-0.019
 hike-0.019
 banks-0.019
 stall-0.018
Token.
Token ID13
Feature activation+6.35e-03

Pos logit contributions
 stalls+0.025
 hike+0.024
 flavors+0.024
 banks+0.024
 stall+0.024
Neg logit contributions
 crying-0.019
 weeping-0.019
 bleeding-0.018
apped-0.017
Ros-0.017
Token could
Token ID714
Feature activation+9.17e-04

Pos logit contributions
 stalls+0.058
 hike+0.054
 stall+0.053
 flavors+0.053
 banks+0.052
Neg logit contributions
 weeping-0.074
Ros-0.073
 crying-0.073
 bleeding-0.072
ir-0.072
Token only
Token ID691
Feature activation-4.99e-02

Pos logit contributions
 weeping+0.001
 crying+0.001
 bleeding+0.001
Ros+0.001
ir+0.001
Neg logit contributions
 stalls-0.002
 stall-0.001
 hike-0.001
 flavors-0.001
 banks-0.001
Token watch
Token ID2342
Feature activation-3.08e-02

Pos logit contributions
 stalls+0.085
 stall+0.079
 hike+0.077
 banks+0.077
 flavors+0.077
Neg logit contributions
 weeping-0.079
 crying-0.078
 bleeding-0.077
Ros-0.075
ir-0.075
Token TV
Token ID3195
Feature activation-5.78e-02

Pos logit contributions
 stalls+0.059
 hike+0.056
 stall+0.056
 flavors+0.055
 banks+0.055
Neg logit contributions
 weeping-0.041
Ros-0.041
 crying-0.040
ir-0.040
 bleeding-0.040
Token for
Token ID329
Feature activation-1.75e-02

Pos logit contributions
 stalls+0.116
 stall+0.110
 flavors+0.109
 banks+0.109
 hike+0.106
Neg logit contributions
 crying-0.073
 weeping-0.071
 bleeding-0.069
 microphone-0.068
Ros-0.065
Token 30
Token ID1542
Feature activation-9.15e-02

Pos logit contributions
 stalls+0.030
 hike+0.029
 flavors+0.028
 stall+0.028
 banks+0.028
Neg logit contributions
 weeping-0.027
apped-0.026
Ros-0.026
 bleeding-0.026
 crying-0.026
Token minutes
Token ID2431
Feature activation-4.91e-02

Pos logit contributions
 banks+0.158
 fare+0.158
 stalls+0.157
 stall+0.155
 hike+0.153
Neg logit contributions
 crying-0.123
ir-0.121
el-0.120
 bleeding-0.120
itt-0.117
Token.
Token ID13
Feature activation-3.68e-03

Pos logit contributions
 stalls+0.086
 hike+0.082
 stall+0.082
 flavors+0.081
 banks+0.080
Neg logit contributions
 crying-0.075
 weeping-0.074
 bleeding-0.072
apped-0.071
ir-0.071
Token Lily
Token ID20037
Feature activation-6.76e-03

Pos logit contributions
 stalls+0.008
 hike+0.007
 stall+0.007
 banks+0.007
 flavors+0.007
Neg logit contributions
 weeping-0.004
 crying-0.004
Ros-0.004
 bleeding-0.004
ir-0.004
Token asked
Token ID1965
Feature activation-1.27e-02

Pos logit contributions
 stalls+0.013
 fare+0.013
 stall+0.013
 hike+0.012
 banks+0.012
Neg logit contributions
 crying-0.009
 weeping-0.009
 bleeding-0.009
Ros-0.009
apped-0.009
Token Mitt
Token ID16627
Feature activation-2.15e-02

Pos logit contributions
 stalls+0.030
 stall+0.028
 hike+0.028
 banks+0.028
 flavors+0.028
Neg logit contributions
 weeping-0.012
 crying-0.012
Ros-0.012
 bleeding-0.012
ir-0.012
Token"
Token ID1
Feature activation+2.31e-03

Pos logit contributions
Ros+0.023
ir+0.020
Sad+0.019
apped+0.019
Emily+0.019
Neg logit contributions
 flavors-0.028
 stalls-0.028
 list-0.028
 hike-0.028
 fare-0.028
TokenDo
Token ID5211
Feature activation-1.46e-02

Pos logit contributions
 weeping+0.003
 crying+0.003
 bleeding+0.003
ir+0.003
Ros+0.003
Neg logit contributions
 stalls-0.005
 stall-0.005
 hike-0.005
 flavors-0.005
 banks-0.004
Token you
Token ID345
Feature activation-3.90e-02

Pos logit contributions
 stalls+0.041
 stall+0.040
 banks+0.039
 fare+0.039
 bank+0.038
Neg logit contributions
Ros-0.007
ir-0.006
 crying-0.006
 Ros-0.005
 weeping-0.005
Token really
Token ID1107
Feature activation-4.93e-02

Pos logit contributions
 stalls+0.071
 stall+0.067
 belts+0.064
 flavors+0.063
 fare+0.063
Neg logit contributions
ir-0.062
 weeping-0.060
 crying-0.058
Ros-0.057
 Emily-0.056
Token do
Token ID466
Feature activation-2.65e-02

Pos logit contributions
 stalls+0.100
 stall+0.096
 belts+0.095
cases+0.094
 fare+0.092
Neg logit contributions
 weeping-0.065
 crying-0.065
ir-0.060
 bleeding-0.058
Ros-0.055
Token maths
Token ID47761
Feature activation-9.14e-02

Pos logit contributions
 stalls+0.057
 stall+0.053
 hike+0.053
 banks+0.052
 flavors+0.052
Neg logit contributions
 weeping-0.031
 crying-0.031
Ros-0.030
 microphone-0.029
ir-0.029
Token in
Token ID287
Feature activation+5.43e-03

Pos logit contributions
 stalls+0.193
 stall+0.186
 banks+0.184
 hike+0.184
 fare+0.181
Neg logit contributions
 crying-0.097
 microphone-0.096
 weeping-0.094
 Ros-0.093
 Emily-0.091
Token here
Token ID994
Feature activation-1.59e-02

Pos logit contributions
 weeping+0.007
 crying+0.007
Ros+0.007
ir+0.006
 bleeding+0.006
Neg logit contributions
 stalls-0.011
 hike-0.010
 stall-0.010
 banks-0.010
 flavors-0.010
Token?"
Token ID1701
Feature activation+9.26e-03

Pos logit contributions
 stalls+0.033
 hike+0.030
 banks+0.030
 fare+0.030
 stall+0.030
Neg logit contributions
 weeping-0.019
 crying-0.019
Ros-0.018
ir-0.018
 bleeding-0.018
Token asked
Token ID1965
Feature activation+1.34e-02

Pos logit contributions
 weeping+0.016
 crying+0.016
Ros+0.015
 bleeding+0.015
ir+0.015
Neg logit contributions
 stalls-0.014
 hike-0.014
 stall-0.013
 flavors-0.013
 fare-0.013
Token Amanda
Token ID23040
Feature activation-6.60e-03

Pos logit contributions
 weeping+0.019
 bleeding+0.019
Ros+0.018
 crying+0.018
ir+0.018
Neg logit contributions
 stalls-0.025
 stall-0.023
 hike-0.023
 flavors-0.023
 banks-0.023
Token mom
Token ID1995
Feature activation-7.94e-03

Pos logit contributions
 stalls+0.077
 stall+0.073
 hike+0.073
 banks+0.073
 flavors+0.072
Neg logit contributions
 weeping-0.043
 crying-0.042
Ros-0.042
 bleeding-0.041
ir-0.040
Token if
Token ID611
Feature activation+9.67e-03

Pos logit contributions
 stalls+0.017
 banks+0.017
 flavors+0.017
 stall+0.017
 hike+0.017
Neg logit contributions
 weeping-0.009
 crying-0.009
Ros-0.008
 bleeding-0.008
 Ros-0.008
Token she
Token ID673
Feature activation-5.51e-02

Pos logit contributions
 crying+0.013
 weeping+0.013
 bleeding+0.012
 Ros+0.012
Ros+0.012
Neg logit contributions
 stalls-0.019
 banks-0.019
 stall-0.018
 fare-0.018
 flavors-0.018
Token could
Token ID714
Feature activation-5.95e-03

Pos logit contributions
 stalls+0.110
 banks+0.104
 fare+0.103
 stall+0.101
 hike+0.099
Neg logit contributions
 crying-0.074
apped-0.074
 weeping-0.074
Ros-0.072
 Ros-0.070
Token try
Token ID1949
Feature activation-2.50e-02

Pos logit contributions
 stalls+0.011
 stall+0.010
 flavors+0.010
 banks+0.010
 fare+0.010
Neg logit contributions
 weeping-0.009
 crying-0.008
Ros-0.008
ir-0.008
el-0.008
Token yoga
Token ID20351
Feature activation-9.13e-02

Pos logit contributions
 stalls+0.049
 stall+0.047
 hike+0.047
 fare+0.046
 banks+0.046
Neg logit contributions
 weeping-0.033
 crying-0.032
 bleeding-0.031
Ros-0.031
ir-0.031
Token.
Token ID13
Feature activation-7.25e-03

Pos logit contributions
 stalls+0.198
 fare+0.193
 banks+0.189
 ranc+0.185
 hike+0.184
Neg logit contributions
 together-0.093
 crying-0.092
 microphone-0.092
aps-0.091
 Emily-0.090
Token Her
Token ID2332
Feature activation-1.54e-02

Pos logit contributions
 stalls+0.014
 stall+0.014
 hike+0.014
 flavors+0.013
 banks+0.013
Neg logit contributions
 weeping-0.009
 bleeding-0.009
 crying-0.009
Ros-0.009
ir-0.009
Token mom
Token ID1995
Feature activation+1.19e-02

Pos logit contributions
 stalls+0.032
 stall+0.030
 hike+0.030
 flavors+0.030
 banks+0.030
Neg logit contributions
 weeping-0.019
Ros-0.018
 crying-0.018
 bleeding-0.018
ir-0.018
Token said
Token ID531
Feature activation-3.25e-02

Pos logit contributions
 crying+0.019
 weeping+0.019
Ros+0.019
 bleeding+0.019
apped+0.018
Neg logit contributions
 stalls-0.021
 banks-0.019
 stall-0.019
 hike-0.019
 fare-0.019
Token yes
Token ID3763
Feature activation-5.19e-02

Pos logit contributions
 stalls+0.057
 banks+0.055
 hike+0.053
 fare+0.052
 stall+0.052
Neg logit contributions
 crying-0.054
 weeping-0.053
 bleeding-0.052
apped-0.050
 microphone-0.049
Token Bob
Token ID5811
Feature activation-5.28e-03

Pos logit contributions
 stalls+0.001
 hike+0.001
 stall+0.001
 flavors+0.001
 banks+0.001
Neg logit contributions
 weeping-0.001
 crying-0.001
Ros-0.001
 bleeding-0.001
ir-0.001
Token's
Token ID338
Feature activation-1.28e-02

Pos logit contributions
 stalls+0.009
 fare+0.009
 stall+0.009
 flavors+0.009
 hike+0.009
Neg logit contributions
 crying-0.008
 weeping-0.008
el-0.008
 bleeding-0.007
Ros-0.007
Token grandma
Token ID49890
Feature activation+1.19e-02

Pos logit contributions
 stalls+0.024
 stall+0.023
 hike+0.023
 flavors+0.023
 banks+0.023
Neg logit contributions
 weeping-0.017
Ros-0.017
 bleeding-0.017
 crying-0.017
ir-0.017
Token took
Token ID1718
Feature activation+1.05e-02

Pos logit contributions
 crying+0.017
 weeping+0.017
 Emily+0.016
 streaming+0.016
 Ros+0.016
Neg logit contributions
 stalls-0.022
 fare-0.022
 banks-0.022
 stall-0.021
 hike-0.021
Token him
Token ID683
Feature activation-1.11e-02

Pos logit contributions
 weeping+0.013
Ros+0.012
 crying+0.012
 bleeding+0.012
ir+0.012
Neg logit contributions
 stalls-0.022
 hike-0.021
 stall-0.020
 flavors-0.020
 banks-0.020
Token shopping
Token ID9735
Feature activation-9.10e-02

Pos logit contributions
 stalls+0.023
 flavors+0.022
 stall+0.022
 banks+0.021
 hike+0.021
Neg logit contributions
 crying-0.013
 weeping-0.013
 bleeding-0.013
Ros-0.013
ir-0.012
Token for
Token ID329
Feature activation-1.84e-02

Pos logit contributions
nic+0.205
 hike+0.203
 banks+0.192
 Street+0.190
 pm+0.187
Neg logit contributions
 together-0.083
apped-0.082
 crying-0.082
el-0.077
 microphone-0.070
Token clothes
Token ID8242
Feature activation-3.71e-02

Pos logit contributions
 stalls+0.041
 banks+0.039
 stall+0.039
 hike+0.039
 flavors+0.038
Neg logit contributions
 crying-0.019
 weeping-0.018
Ros-0.018
 bleeding-0.018
ir-0.017
Token.
Token ID13
Feature activation+8.79e-04

Pos logit contributions
 stalls+0.077
 hike+0.075
 flavors+0.075
nic+0.074
 stall+0.073
Neg logit contributions
 crying-0.045
 weeping-0.042
 bleeding-0.041
 streaming-0.040
el-0.039
Token She
Token ID1375
Feature activation-1.30e-02

Pos logit contributions
 crying+0.001
 weeping+0.001
 Ros+0.001
apped+0.001
 streaming+0.001
Neg logit contributions
 stalls-0.002
 flavors-0.002
 hike-0.002
 belts-0.002
 fare-0.002
Token showed
Token ID3751
Feature activation-3.11e-03

Pos logit contributions
 stalls+0.024
 fare+0.024
 banks+0.024
 hike+0.023
 stall+0.023
Neg logit contributions
 crying-0.018
apped-0.017
 weeping-0.017
Ros-0.017
 Emily-0.016
Token
Token ID198
Feature activation-3.85e-03

Pos logit contributions
 stalls+0.018
 hike+0.017
 stall+0.017
 banks+0.016
 list+0.016
Neg logit contributions
 weeping-0.021
 streaming-0.021
 bleeding-0.021
 crying-0.020
ir-0.020
TokenAfter
Token ID3260
Feature activation-1.62e-02

Pos logit contributions
 stalls+0.007
 stall+0.007
 hike+0.007
 fare+0.006
 flavors+0.006
Neg logit contributions
Ros-0.006
 weeping-0.005
 bleeding-0.005
 crying-0.005
ir-0.005
Token they
Token ID484
Feature activation-5.52e-02

Pos logit contributions
 stalls+0.028
 hike+0.026
 stall+0.026
 banks+0.026
 flavors+0.026
Neg logit contributions
 weeping-0.025
Ros-0.025
 crying-0.024
 bleeding-0.024
ir-0.024
Token completed
Token ID5668
Feature activation-2.84e-02

Pos logit contributions
 stalls+0.095
 experiments+0.090
 banks+0.090
 fare+0.087
 flavors+0.087
Neg logit contributions
el-0.083
ir-0.081
 streaming-0.080
apped-0.079
 crying-0.077
Token their
Token ID511
Feature activation-5.16e-02

Pos logit contributions
 stalls+0.060
 hike+0.057
 flavors+0.057
 fare+0.056
 banks+0.056
Neg logit contributions
 weeping-0.033
 crying-0.033
 bleeding-0.031
ir-0.031
Ros-0.030
Token shopping
Token ID9735
Feature activation-9.07e-02

Pos logit contributions
 stalls+0.068
 fare+0.066
land+0.063
 belts+0.061
nic+0.060
Neg logit contributions
el-0.097
 crying-0.095
 weeping-0.095
 microphone-0.090
 bleeding-0.086
Token,
Token ID11
Feature activation+5.76e-03

Pos logit contributions
nic+0.206
 hike+0.185
uries+0.183
 fare+0.181
land+0.181
Neg logit contributions
 crying-0.092
el-0.091
apped-0.089
 heart-0.083
 weeping-0.082
Token they
Token ID484
Feature activation-2.79e-02

Pos logit contributions
ir+0.008
 weeping+0.008
 bleeding+0.008
el+0.008
Ros+0.007
Neg logit contributions
 stalls-0.011
 hike-0.010
 stall-0.010
 banks-0.010
 list-0.010
Token went
Token ID1816
Feature activation+1.34e-02

Pos logit contributions
 stalls+0.048
 fare+0.047
 flavors+0.047
 banks+0.046
 experiments+0.046
Neg logit contributions
apped-0.040
Ros-0.040
 weeping-0.040
 crying-0.040
ir-0.039
Token home
Token ID1363
Feature activation-1.06e-02

Pos logit contributions
 crying+0.015
 weeping+0.014
apped+0.013
 bleeding+0.013
Ros+0.013
Neg logit contributions
 flavors-0.029
 stalls-0.029
 banks-0.028
 hike-0.028
 fare-0.027
Token and
Token ID290
Feature activation+2.05e-03

Pos logit contributions
 stalls+0.022
 hike+0.021
 flavors+0.021
 banks+0.021
 fare+0.020
Neg logit contributions
 crying-0.014
 weeping-0.013
 bleeding-0.012
apped-0.012
Ros-0.012
Token mum
Token ID25682
Feature activation-2.00e-02

Pos logit contributions
 stalls+0.091
 banks+0.087
 flavors+0.085
 fare+0.085
 hike+0.085
Neg logit contributions
 weeping-0.042
 crying-0.041
Ros-0.038
 bleeding-0.038
 microphone-0.037
Token if
Token ID611
Feature activation+1.60e-02

Pos logit contributions
 stalls+0.039
 flavors+0.039
 banks+0.039
 fare+0.038
 hike+0.038
Neg logit contributions
 crying-0.026
 weeping-0.026
 Emily-0.023
 Ros-0.023
el-0.023
Token she
Token ID673
Feature activation-5.25e-02

Pos logit contributions
 crying+0.021
 weeping+0.020
 Ros+0.019
 bleeding+0.019
Ros+0.019
Neg logit contributions
 stalls-0.032
 banks-0.032
 fare-0.031
 stall-0.031
 flavors-0.031
Token could
Token ID714
Feature activation-3.79e-03

Pos logit contributions
 stalls+0.107
 banks+0.101
 fare+0.099
 stall+0.098
 hike+0.096
Neg logit contributions
 crying-0.069
 weeping-0.068
apped-0.068
 bleeding-0.065
el-0.065
Token try
Token ID1949
Feature activation-2.24e-02

Pos logit contributions
 stalls+0.007
 stall+0.007
 banks+0.007
 belts+0.006
 flavors+0.006
Neg logit contributions
 crying-0.005
 weeping-0.005
ir-0.005
 bleeding-0.005
Ros-0.005
Token yoga
Token ID20351
Feature activation-9.04e-02

Pos logit contributions
 stalls+0.045
 fare+0.043
 stall+0.043
 banks+0.042
 hike+0.042
Neg logit contributions
 crying-0.029
 weeping-0.029
 bleeding-0.027
 blood-0.026
ir-0.026
Token and
Token ID290
Feature activation-2.48e-02

Pos logit contributions
 stalls+0.192
 fare+0.187
 banks+0.183
 ranc+0.181
 hike+0.181
Neg logit contributions
 crying-0.095
 together-0.094
 microphone-0.093
 Emily-0.093
aps-0.092
Token her
Token ID607
Feature activation-1.56e-02

Pos logit contributions
 stalls+0.048
 stall+0.046
 fare+0.045
 banks+0.044
 hike+0.043
Neg logit contributions
 crying-0.035
 weeping-0.034
 bleeding-0.033
 Emily-0.032
Ros-0.031
Token mum
Token ID25682
Feature activation+8.91e-03

Pos logit contributions
 stalls+0.034
 banks+0.032
 fare+0.032
 stall+0.031
 hike+0.031
Neg logit contributions
 crying-0.018
 weeping-0.018
 bleeding-0.017
 microphone-0.017
apped-0.016
Token said
Token ID531
Feature activation-2.85e-02

Pos logit contributions
 crying+0.011
 Emily+0.011
apped+0.010
 weeping+0.010
 bleeding+0.010
Neg logit contributions
 stalls-0.019
 banks-0.019
 hike-0.018
 stall-0.018
 fare-0.018
Token yes
Token ID3763
Feature activation-5.55e-02

Pos logit contributions
 stalls+0.052
 banks+0.050
 hike+0.049
 stall+0.047
 fare+0.047
Neg logit contributions
 crying-0.046
 weeping-0.045
 bleeding-0.044
apped-0.043
 microphone-0.042
Token wall
Token ID3355
Feature activation-8.08e-03

Pos logit contributions
 stalls+0.034
 hike+0.033
 flavors+0.032
 stall+0.031
 banks+0.031
Neg logit contributions
 crying-0.033
 bleeding-0.033
 weeping-0.033
 microphone-0.033
Ros-0.032
Token.
Token ID13
Feature activation+5.02e-03

Pos logit contributions
 stalls+0.015
 hike+0.015
 flavors+0.015
 banks+0.014
 fare+0.014
Neg logit contributions
 crying-0.011
 weeping-0.011
 bleeding-0.010
 microphone-0.010
 streaming-0.010
Token It
Token ID632
Feature activation+3.89e-03

Pos logit contributions
apped+0.004
 crying+0.004
uck+0.004
aps+0.004
 Ow+0.004
Neg logit contributions
 experiments-0.012
 hike-0.012
 fare-0.012
 Street-0.012
nic-0.012
Token said
Token ID531
Feature activation-3.44e-02

Pos logit contributions
Ros+0.005
 crying+0.005
 weeping+0.005
 bleeding+0.005
 Ros+0.005
Neg logit contributions
 stalls-0.007
 stall-0.007
 hike-0.007
 fare-0.007
 banks-0.007
Token "
Token ID366
Feature activation-1.08e-02

Pos logit contributions
 stalls+0.060
 hike+0.056
 banks+0.056
 flavors+0.056
 stall+0.056
Neg logit contributions
 weeping-0.054
 crying-0.053
Ros-0.052
 bleeding-0.051
ir-0.051
TokenDO
Token ID18227
Feature activation-9.04e-02

Pos logit contributions
 stalls+0.020
 stall+0.019
 hike+0.018
 flavors+0.018
 fare+0.018
Neg logit contributions
 weeping-0.016
 crying-0.015
 bleeding-0.015
Ros-0.015
ir-0.015
Token NOT
Token ID5626
Feature activation-4.15e-02

Pos logit contributions
 stalls+0.191
 flavors+0.185
 stall+0.183
 hike+0.177
 fare+0.177
Neg logit contributions
ir-0.107
Ros-0.104
 streaming-0.100
 bleeding-0.099
 weeping-0.097
Token T
Token ID309
Feature activation-1.59e-02

Pos logit contributions
 stalls+0.076
 hike+0.071
 stall+0.071
 flavors+0.071
 banks+0.070
Neg logit contributions
 weeping-0.059
 crying-0.058
Ros-0.058
 bleeding-0.057
ir-0.057
TokenOU
Token ID2606
Feature activation-1.85e-02

Pos logit contributions
 stalls+0.030
 banks+0.028
 flavors+0.028
 stall+0.028
 hike+0.028
Neg logit contributions
ir-0.023
 weeping-0.022
 crying-0.021
apped-0.021
Ros-0.021
TokenCH
Token ID3398
Feature activation-2.04e-02

Pos logit contributions
 stalls+0.033
 hike+0.032
 stall+0.032
 banks+0.031
 flavors+0.031
Neg logit contributions
 weeping-0.027
 crying-0.026
Ros-0.026
 bleeding-0.026
ir-0.026
Token".
Token ID1911
Feature activation-1.24e-02

Pos logit contributions
 stalls+0.036
 stall+0.035
 hike+0.035
 banks+0.034
 fare+0.034
Neg logit contributions
 weeping-0.029
Ros-0.029
 crying-0.028
 bleeding-0.028
ir-0.027
Token who
Token ID508
Feature activation-7.99e-03

Pos logit contributions
 weeping+0.048
Ros+0.048
 bleeding+0.047
ir+0.047
 streaming+0.046
Neg logit contributions
 stalls-0.057
 stall-0.055
 hike-0.053
 banks-0.052
 flavors-0.052
Token really
Token ID1107
Feature activation-1.71e-02

Pos logit contributions
 stalls+0.011
 stall+0.010
 hike+0.010
 flavors+0.010
 banks+0.010
Neg logit contributions
 weeping-0.015
Ros-0.015
ir-0.014
 crying-0.014
 bleeding-0.014
Token wanted
Token ID2227
Feature activation+1.02e-02

Pos logit contributions
 stalls+0.027
 stall+0.025
 hike+0.025
 banks+0.025
 flavors+0.025
Neg logit contributions
 weeping-0.029
 crying-0.028
Ros-0.028
 bleeding-0.028
ir-0.028
Token to
Token ID284
Feature activation-1.66e-02

Pos logit contributions
 crying+0.013
 weeping+0.013
 bleeding+0.012
 blood+0.012
Ros+0.012
Neg logit contributions
 stalls-0.021
 banks-0.020
 stall-0.020
 fare-0.020
 hike-0.020
Token try
Token ID1949
Feature activation-1.49e-02

Pos logit contributions
 stalls+0.030
cases+0.027
 stall+0.027
 fare+0.027
 belts+0.027
Neg logit contributions
 crying-0.024
 weeping-0.024
Ros-0.024
 microphone-0.024
 Ros-0.023
Token yoga
Token ID20351
Feature activation-9.01e-02

Pos logit contributions
 stalls+0.031
 fare+0.031
 banks+0.030
 stall+0.030
 flavors+0.029
Neg logit contributions
 crying-0.017
 weeping-0.017
 blood-0.016
 bleeding-0.016
 microphone-0.015
Token.
Token ID13
Feature activation+5.28e-03

Pos logit contributions
 stalls+0.197
 fare+0.195
 flavors+0.187
 banks+0.186
 stall+0.182
Neg logit contributions
 together-0.094
 microphone-0.092
aps-0.090
 crying-0.089
 Emily-0.086
Token She
Token ID1375
Feature activation-1.33e-02

Pos logit contributions
 weeping+0.004
Ros+0.004
 Ros+0.004
 crying+0.004
apped+0.004
Neg logit contributions
 stalls-0.014
 hike-0.013
 flavors-0.013
 banks-0.013
 fare-0.013
Token was
Token ID373
Feature activation+1.05e-02

Pos logit contributions
 fare+0.027
 stalls+0.026
 stall+0.025
 goods+0.024
 banks+0.024
Neg logit contributions
 crying-0.016
Ros-0.016
 bleeding-0.016
el-0.016
apped-0.016
Token only
Token ID691
Feature activation-2.33e-02

Pos logit contributions
 weeping+0.014
Ros+0.014
 crying+0.014
ir+0.014
 bleeding+0.014
Neg logit contributions
 stalls-0.020
 stall-0.019
 hike-0.019
 banks-0.019
 flavors-0.019
Token three
Token ID1115
Feature activation-2.07e-02

Pos logit contributions
 stalls+0.055
 banks+0.053
 hike+0.053
 flavors+0.052
 stall+0.052
Neg logit contributions
 crying-0.022
 weeping-0.022
 bleeding-0.021
apped-0.020
Ros-0.019
Token red
Token ID2266
Feature activation-2.61e-03

Pos logit contributions
Ros+0.024
 bleeding+0.023
 weeping+0.023
ir+0.023
apped+0.023
Neg logit contributions
 stalls-0.027
 stall-0.026
 hike-0.026
 flavors-0.025
 list-0.025
Token button
Token ID4936
Feature activation-1.19e-02

Pos logit contributions
 stalls+0.005
 stall+0.004
 hike+0.004
 flavors+0.004
 banks+0.004
Neg logit contributions
 weeping-0.004
Ros-0.004
 bleeding-0.004
 crying-0.004
ir-0.004
Token that
Token ID326
Feature activation+1.38e-02

Pos logit contributions
 stalls+0.026
 hike+0.025
 stall+0.025
 banks+0.024
 flavors+0.024
Neg logit contributions
 weeping-0.013
 crying-0.013
Ros-0.012
 bleeding-0.012
ir-0.012
Token said
Token ID531
Feature activation-2.35e-02

Pos logit contributions
 weeping+0.022
 crying+0.022
Ros+0.022
 bleeding+0.021
ir+0.021
Neg logit contributions
 stalls-0.023
 stall-0.022
 hike-0.022
 banks-0.021
 flavors-0.021
Token "
Token ID366
Feature activation-9.61e-03

Pos logit contributions
 stalls+0.038
 hike+0.035
 banks+0.035
 stall+0.035
 flavors+0.035
Neg logit contributions
 weeping-0.041
 crying-0.040
Ros-0.039
 bleeding-0.039
ir-0.039
TokenDO
Token ID18227
Feature activation-9.00e-02

Pos logit contributions
 stalls+0.017
 flavors+0.016
 fare+0.016
 stall+0.016
 hike+0.016
Neg logit contributions
 weeping-0.015
 crying-0.014
 bleeding-0.014
 blood-0.013
apped-0.013
Token NOT
Token ID5626
Feature activation-4.31e-02

Pos logit contributions
 stalls+0.186
 flavors+0.176
 stall+0.175
 hike+0.171
 fare+0.169
Neg logit contributions
ir-0.117
Ros-0.111
 bleeding-0.111
 weeping-0.111
 streaming-0.109
Token PRESS
Token ID44310
Feature activation-6.62e-02

Pos logit contributions
 stalls+0.080
 stall+0.076
 hike+0.076
 banks+0.075
 flavors+0.074
Neg logit contributions
 weeping-0.060
Ros-0.059
 crying-0.059
 bleeding-0.059
ir-0.057
Token."
Token ID526
Feature activation-1.22e-02

Pos logit contributions
 stalls+0.110
 flavors+0.103
 hike+0.102
 stall+0.101
 fare+0.100
Neg logit contributions
ir-0.105
 weeping-0.105
apped-0.103
Ros-0.103
 crying-0.102
Token She
Token ID1375
Feature activation-1.47e-02

Pos logit contributions
 stalls+0.025
 stall+0.024
 hike+0.024
 banks+0.024
 flavors+0.024
Neg logit contributions
 weeping-0.015
Ros-0.014
 bleeding-0.014
ir-0.014
 crying-0.014
Token couldn
Token ID3521
Feature activation+4.56e-02

Pos logit contributions
 stalls+0.026
 fare+0.024
 stall+0.024
 hike+0.024
 banks+0.023
Neg logit contributions
Ros-0.023
 crying-0.023
 weeping-0.023
 bleeding-0.022
apped-0.021
Token her
Token ID607
Feature activation-7.73e-04

Pos logit contributions
 stalls+0.046
 hike+0.044
 stall+0.044
 flavors+0.043
 banks+0.043
Neg logit contributions
 weeping-0.021
Ros-0.021
 bleeding-0.020
ir-0.020
 crying-0.020
Token mom
Token ID1995
Feature activation+1.88e-02

Pos logit contributions
 stalls+0.001
 stall+0.001
 hike+0.001
 list+0.001
 flavors+0.001
Neg logit contributions
ir-0.001
Ros-0.001
 weeping-0.001
 bleeding-0.001
 streaming-0.001
Tokenmy
Token ID1820
Feature activation+7.51e-03

Pos logit contributions
 weeping+0.035
ir+0.035
 bleeding+0.035
Ros+0.035
 blood+0.034
Neg logit contributions
 stalls-0.025
 hike-0.024
 stall-0.023
 banks-0.022
 flavors-0.022
Token went
Token ID1816
Feature activation+1.61e-02

Pos logit contributions
 weeping+0.014
 bleeding+0.014
Ros+0.014
 crying+0.013
ir+0.013
Neg logit contributions
 stalls-0.011
 hike-0.010
 stall-0.010
 banks-0.010
 flavors-0.009
Token grocery
Token ID16918
Feature activation-8.32e-02

Pos logit contributions
Ros+0.024
 weeping+0.024
ir+0.024
 bleeding+0.024
 crying+0.023
Neg logit contributions
 stalls-0.028
 hike-0.027
 stall-0.027
 banks-0.026
 fare-0.025
Token shopping
Token ID9735
Feature activation-8.97e-02

Pos logit contributions
 flavors+0.181
nic+0.174
 fare+0.174
land+0.172
uries+0.168
Neg logit contributions
 crying-0.089
 together-0.079
 weeping-0.074
Ros-0.071
 Ros-0.070
Token.
Token ID13
Feature activation-1.44e-02

Pos logit contributions
nic+0.193
 hike+0.192
 flavors+0.186
 fare+0.186
 banks+0.184
Neg logit contributions
apped-0.093
 crying-0.093
 together-0.091
 weeping-0.080
aps-0.080
Token Laura
Token ID16753
Feature activation-2.10e-02

Pos logit contributions
 flavors+0.027
 stalls+0.027
 fare+0.027
 banks+0.027
 hike+0.027
Neg logit contributions
 weeping-0.018
 crying-0.018
apped-0.017
 Emily-0.017
Ros-0.016
Token was
Token ID373
Feature activation+5.07e-03

Pos logit contributions
 fare+0.046
 flavors+0.044
 stalls+0.043
land+0.043
 experiments+0.041
Neg logit contributions
 crying-0.025
 weeping-0.022
apped-0.022
 Emily-0.022
el-0.021
Token having
Token ID1719
Feature activation+2.78e-03

Pos logit contributions
 crying+0.006
 weeping+0.006
 bleeding+0.006
apped+0.005
Ros+0.005
Neg logit contributions
 stalls-0.011
 flavors-0.011
 hike-0.011
 banks-0.010
 fare-0.010
Token lots
Token ID6041
Feature activation-3.68e-02

Pos logit contributions
 crying+0.003
 weeping+0.003
Ros+0.003
 bleeding+0.003
 Ros+0.003
Neg logit contributions
 stalls-0.006
 stall-0.006
 fare-0.006
 banks-0.006
 hike-0.006
Token some
Token ID617
Feature activation-4.77e-02

Pos logit contributions
 stalls+0.121
 hike+0.119
 stall+0.117
 list+0.115
 banks+0.115
Neg logit contributions
 crying-0.060
 blood-0.060
 Ros-0.058
el-0.058
 weeping-0.058
Token food
Token ID2057
Feature activation-4.59e-02

Pos logit contributions
 stalls+0.079
cases+0.078
 hike+0.077
 fare+0.076
 stall+0.075
Neg logit contributions
 microphone-0.068
el-0.067
 blood-0.067
 weeping-0.066
 bleeding-0.066
Token at
Token ID379
Feature activation-7.25e-03

Pos logit contributions
 stalls+0.101
 hike+0.100
 banks+0.100
 fare+0.099
cases+0.098
Neg logit contributions
 together-0.040
 Ros-0.038
 microphone-0.036
ucky-0.035
 crying-0.035
Token the
Token ID262
Feature activation-1.19e-02

Pos logit contributions
 stalls+0.015
 flavors+0.014
 hike+0.014
 fare+0.014
 stall+0.014
Neg logit contributions
 weeping-0.008
 crying-0.008
ir-0.008
el-0.008
 bleeding-0.008
Token farm
Token ID5318
Feature activation-4.68e-02

Pos logit contributions
cases+0.016
 flavors+0.014
 belts+0.014
nic+0.014
books+0.014
Neg logit contributions
 microphone-0.021
 crying-0.020
 weeping-0.020
 bleeding-0.020
el-0.019
Token cafe
Token ID26725
Feature activation-8.97e-02

Pos logit contributions
cases+0.101
 fare+0.100
nic+0.100
 hike+0.098
 flavors+0.097
Neg logit contributions
 together-0.039
 crying-0.036
 microphone-0.034
 Emily-0.031
 Ros-0.031
Token.
Token ID13
Feature activation-7.76e-03

Pos logit contributions
 hike+0.216
 banks+0.213
 fare+0.205
 quest+0.205
nic+0.203
Neg logit contributions
 together-0.066
 on-0.050
 in-0.049
ing-0.047
ful-0.046
Token Lily
Token ID20037
Feature activation-1.52e-03

Pos logit contributions
 stalls+0.014
 stall+0.013
 hike+0.013
 banks+0.012
 flavors+0.012
Neg logit contributions
 weeping-0.012
 bleeding-0.012
ir-0.012
 crying-0.011
 streaming-0.011
Token and
Token ID290
Feature activation-2.50e-02

Pos logit contributions
 fare+0.004
 experiments+0.004
land+0.004
 stalls+0.004
 banks+0.004
Neg logit contributions
 crying-0.001
 Emily-0.001
apped-0.001
 weeping-0.001
el-0.001
Token Sam
Token ID3409
Feature activation-2.45e-02

Pos logit contributions
 stalls+0.066
 fare+0.063
 stall+0.063
 banks+0.062
 flavors+0.062
Neg logit contributions
 crying-0.017
 weeping-0.016
el-0.016
ir-0.016
 Emily-0.016
Token were
Token ID547
Feature activation+2.48e-04

Pos logit contributions
land+0.056
 experiments+0.055
 rink+0.053
 fare+0.053
 pm+0.052
Neg logit contributions
 both-0.022
apped-0.018
el-0.018
 wiped-0.016
 together-0.015
Token From
Token ID3574
Feature activation-1.94e-02

Pos logit contributions
 bleeding+0.025
ir+0.025
 weeping+0.025
Ros+0.024
 crying+0.024
Neg logit contributions
 stalls-0.025
 stall-0.024
 hike-0.022
 banks-0.022
 flavors-0.022
Token then
Token ID788
Feature activation+1.55e-02

Pos logit contributions
 stalls+0.035
 hike+0.033
 stall+0.033
 banks+0.033
 fare+0.033
Neg logit contributions
 weeping-0.028
 crying-0.028
Ros-0.027
 bleeding-0.027
ir-0.027
Token on
Token ID319
Feature activation-1.23e-02

Pos logit contributions
 weeping+0.018
 crying+0.017
Ros+0.017
 blood+0.016
 bleeding+0.016
Neg logit contributions
 stalls-0.032
 hike-0.031
 stall-0.031
 list-0.030
 fare-0.030
Token,
Token ID11
Feature activation+2.21e-02

Pos logit contributions
 stalls+0.021
 hike+0.019
 stall+0.019
 flavors+0.019
 banks+0.019
Neg logit contributions
 weeping-0.020
 crying-0.019
Ros-0.019
 bleeding-0.019
ir-0.019
Token the
Token ID262
Feature activation+1.09e-02

Pos logit contributions
 bleeding+0.038
 weeping+0.038
ir+0.037
Ros+0.037
 crying+0.036
Neg logit contributions
 stalls-0.034
 hike-0.032
 stall-0.032
 flavors-0.031
 banks-0.030
Token whole
Token ID2187
Feature activation+5.06e-02

Pos logit contributions
 weeping+0.014
ir+0.013
Ros+0.013
 bleeding+0.013
 crying+0.013
Neg logit contributions
 stalls-0.022
 hike-0.021
 stall-0.021
 flavors-0.021
 banks-0.021
Token family
Token ID1641
Feature activation+3.55e-02

Pos logit contributions
 bleeding+0.090
 weeping+0.089
Ros+0.089
ir+0.088
 crying+0.087
Neg logit contributions
 stalls-0.079
 hike-0.074
 stall-0.074
 flavors-0.071
 list-0.069
Token could
Token ID714
Feature activation+4.51e-02

Pos logit contributions
 weeping+0.057
 crying+0.056
Ros+0.056
 bleeding+0.055
ir+0.055
Neg logit contributions
 stalls-0.059
 stall-0.055
 hike-0.055
 banks-0.054
 flavors-0.054
Token come
Token ID1282
Feature activation+2.30e-02

Pos logit contributions
 weeping+0.067
 crying+0.066
Ros+0.065
ir+0.064
apped+0.063
Neg logit contributions
 stalls-0.081
 stall-0.076
 flavors-0.074
 fare-0.073
 banks-0.073
Token and
Token ID290
Feature activation+4.06e-03

Pos logit contributions
 weeping+0.029
 bleeding+0.029
ir+0.029
Ros+0.029
 crying+0.028
Neg logit contributions
 stalls-0.046
 hike-0.044
 stall-0.043
 banks-0.043
 fare-0.043
Token play
Token ID711
Feature activation+2.34e-02

Pos logit contributions
 weeping+0.006
 crying+0.006
Ros+0.006
 bleeding+0.006
ir+0.006
Neg logit contributions
 stalls-0.007
 stall-0.007
 hike-0.007
 flavors-0.007
 banks-0.007
Token happy
Token ID3772
Feature activation+4.42e-02

Pos logit contributions
 stalls+0.030
 stall+0.029
 flavors+0.028
 banks+0.027
 fare+0.027
Neg logit contributions
Ros-0.031
ir-0.030
 bleeding-0.029
 blood-0.029
 streaming-0.029
Token.
Token ID13
Feature activation+2.40e-02

Pos logit contributions
 weeping+0.064
 crying+0.063
 bleeding+0.060
Ros+0.059
 streaming+0.058
Neg logit contributions
 stalls-0.083
 hike-0.079
 stall-0.079
 banks-0.078
 fare-0.078
Token She
Token ID1375
Feature activation+8.02e-03

Pos logit contributions
 weeping+0.033
 bleeding+0.033
 crying+0.032
ir+0.032
Ros+0.032
Neg logit contributions
 stalls-0.045
 stall-0.043
 hike-0.042
 banks-0.041
 flavors-0.041
Token hugged
Token ID41130
Feature activation+3.82e-02

Pos logit contributions
 weeping+0.013
 crying+0.012
Ros+0.012
 bleeding+0.012
ir+0.012
Neg logit contributions
 stalls-0.014
 stall-0.013
 hike-0.013
 banks-0.012
 flavors-0.012
Token the
Token ID262
Feature activation+2.06e-02

Pos logit contributions
 bleeding+0.071
aps+0.070
Ros+0.069
apped+0.069
apping+0.069
Neg logit contributions
 stalls-0.049
 stall-0.048
 hike-0.046
 banks-0.045
 bank-0.044
Token fairy
Token ID25607
Feature activation+5.96e-02

Pos logit contributions
ir+0.025
apped+0.024
Ros+0.024
 bleeding+0.023
 weeping+0.023
Neg logit contributions
 stalls-0.042
 stall-0.041
 hike-0.040
 banks-0.040
 flavors-0.039
Token and
Token ID290
Feature activation+6.11e-03

Pos logit contributions
 weeping+0.082
 crying+0.081
Ros+0.080
 bleeding+0.080
ir+0.079
Neg logit contributions
 stalls-0.112
 hike-0.105
 stall-0.105
 flavors-0.104
 banks-0.104
Token thanked
Token ID26280
Feature activation+3.36e-02

Pos logit contributions
 weeping+0.010
Ros+0.010
ir+0.010
 bleeding+0.010
 crying+0.010
Neg logit contributions
 stalls-0.010
 hike-0.009
 stall-0.009
 flavors-0.009
 banks-0.009
Token her
Token ID607
Feature activation-5.02e-03

Pos logit contributions
apped+0.056
 bleeding+0.056
Ros+0.053
 weeping+0.052
aps+0.052
Neg logit contributions
 stalls-0.052
 hike-0.050
 stall-0.049
 flavors-0.047
 banks-0.046
Token for
Token ID329
Feature activation-2.73e-03

Pos logit contributions
 stalls+0.009
 hike+0.009
 stall+0.009
 flavors+0.009
 banks+0.009
Neg logit contributions
 weeping-0.007
 crying-0.007
Ros-0.007
 bleeding-0.007
ir-0.007
Token the
Token ID262
Feature activation+1.00e-02

Pos logit contributions
 stalls+0.005
 stall+0.005
 hike+0.004
 fare+0.004
 flavors+0.004
Neg logit contributions
ir-0.004
Ros-0.004
apped-0.004
el-0.004
ucky-0.004
Token experience
Token ID1998
Feature activation-7.13e-03

Pos logit contributions
 crying+0.002
 weeping+0.002
 bleeding+0.002
 streaming+0.002
Ros+0.002
Neg logit contributions
 stalls-0.003
 stall-0.002
 banks-0.002
 fare-0.002
 flavors-0.002
Token outside
Token ID2354
Feature activation+3.34e-03

Pos logit contributions
 stalls+0.015
 banks+0.015
 stall+0.015
 fare+0.014
 hike+0.014
Neg logit contributions
 crying-0.008
 weeping-0.008
 bleeding-0.008
 streaming-0.007
 microphone-0.007
Token of
Token ID286
Feature activation+3.03e-02

Pos logit contributions
 weeping+0.005
 crying+0.005
 bleeding+0.004
 streaming+0.004
ir+0.004
Neg logit contributions
 stalls-0.006
 stall-0.006
 hike-0.006
 banks-0.006
 flavors-0.006
Token shooting
Token ID4395
Feature activation-9.58e-03

Pos logit contributions
 weeping+0.041
 crying+0.041
 bleeding+0.040
Ros+0.039
ir+0.039
Neg logit contributions
 stalls-0.058
 stall-0.055
 hike-0.055
 banks-0.054
 flavors-0.054
Token his
Token ID465
Feature activation+1.75e-02

Pos logit contributions
 stalls+0.018
 stall+0.017
 hike+0.017
 banks+0.017
 flavors+0.017
Neg logit contributions
 weeping-0.013
 crying-0.013
 bleeding-0.012
Ros-0.012
ir-0.012
Token arrow
Token ID15452
Feature activation+4.74e-02

Pos logit contributions
 weeping+0.028
Ros+0.027
ir+0.027
 crying+0.027
 bleeding+0.027
Neg logit contributions
 stalls-0.029
 hike-0.027
 stall-0.027
 banks-0.027
 flavors-0.027
Token.
Token ID13
Feature activation+2.08e-02

Pos logit contributions
 weeping+0.065
 crying+0.065
 bleeding+0.063
 streaming+0.061
Ros+0.059
Neg logit contributions
 stalls-0.091
 flavors-0.087
 stall-0.085
 fare-0.084
 hike-0.084
Token He
Token ID679
Feature activation+7.36e-03

Pos logit contributions
 weeping+0.029
 crying+0.028
Ros+0.028
 bleeding+0.028
apped+0.027
Neg logit contributions
 stalls-0.040
 hike-0.037
 stall-0.037
 banks-0.037
 flavors-0.037
Token was
Token ID373
Feature activation+2.06e-02

Pos logit contributions
 weeping+0.011
 crying+0.011
 bleeding+0.011
Ros+0.011
apped+0.010
Neg logit contributions
 stalls-0.013
 stall-0.012
 fare-0.012
 banks-0.012
 hike-0.012
Token excited
Token ID6568
Feature activation+3.48e-03

Pos logit contributions
 weeping+0.028
Ros+0.027
 crying+0.027
 bleeding+0.027
ir+0.026
Neg logit contributions
 stalls-0.040
 stall-0.037
 hike-0.037
 banks-0.037
 flavors-0.037
Token to
Token ID284
Feature activation+9.03e-03

Pos logit contributions
 crying+0.004
 weeping+0.004
 bleeding+0.004
 streaming+0.004
el+0.004
Neg logit contributions
 stalls-0.007
 stall-0.007
 banks-0.007
 fare-0.007
 flavors-0.007
Token.
Token ID13
Feature activation+1.12e-02

Pos logit contributions
 crying+0.017
 weeping+0.016
 Emily+0.015
el+0.015
ir+0.015
Neg logit contributions
 flavors-0.025
 stalls-0.025
 fare-0.025
 hike-0.025
 banks-0.024
Token One
Token ID1881
Feature activation-3.86e-02

Pos logit contributions
 weeping+0.015
Ros+0.015
 crying+0.015
 Ros+0.014
 Emily+0.014
Neg logit contributions
 stalls-0.021
 flavors-0.020
 hike-0.020
 fare-0.020
 banks-0.020
Token day
Token ID1110
Feature activation+1.00e-02

Pos logit contributions
 stalls+0.064
 banks+0.059
 stall+0.059
 flavors+0.059
 hike+0.058
Neg logit contributions
 weeping-0.063
 crying-0.062
 bleeding-0.061
Ros-0.061
ir-0.060
Token,
Token ID11
Feature activation+2.77e-03

Pos logit contributions
 weeping+0.017
Ros+0.017
 crying+0.017
 bleeding+0.016
ir+0.016
Neg logit contributions
 stalls-0.016
 hike-0.015
 stall-0.015
 flavors-0.015
 banks-0.015
Token the
Token ID262
Feature activation-3.67e-03

Pos logit contributions
 weeping+0.005
ir+0.005
Ros+0.005
 bleeding+0.005
 streaming+0.005
Neg logit contributions
 stalls-0.004
 stall-0.004
 hike-0.004
 flavors-0.004
 banks-0.004
Token bird
Token ID6512
Feature activation+4.67e-02

Pos logit contributions
 stalls+0.006
 stall+0.006
 hike+0.006
 flavors+0.006
 fare+0.006
Neg logit contributions
 streaming-0.006
ir-0.006
 weeping-0.006
Ros-0.006
 bleeding-0.006
Token came
Token ID1625
Feature activation+3.08e-02

Pos logit contributions
 weeping+0.077
 crying+0.075
Ros+0.075
 bleeding+0.075
ir+0.074
Neg logit contributions
 stalls-0.076
 stall-0.071
 hike-0.071
 banks-0.069
 flavors-0.069
Token to
Token ID284
Feature activation-2.32e-03

Pos logit contributions
Ros+0.059
 bleeding+0.059
ir+0.059
 blood+0.058
itt+0.058
Neg logit contributions
 stalls-0.037
 hike-0.036
 stall-0.035
 banks-0.034
 bank-0.031
Token the
Token ID262
Feature activation-7.49e-03

Pos logit contributions
 stalls+0.004
 hike+0.004
 stall+0.004
 flavors+0.004
 banks+0.004
Neg logit contributions
 weeping-0.003
 bleeding-0.003
Ros-0.003
 crying-0.003
apped-0.003
Token palace
Token ID20562
Feature activation-2.09e-02

Pos logit contributions
 stalls+0.011
 hike+0.011
 stall+0.011
 flavors+0.011
 banks+0.010
Neg logit contributions
 weeping-0.013
 crying-0.013
Ros-0.013
 bleeding-0.013
ir-0.012
Token and
Token ID290
Feature activation+3.83e-03

Pos logit contributions
 stalls+0.042
 flavors+0.042
 hike+0.040
 fare+0.040
nic+0.040
Neg logit contributions
 crying-0.026
 weeping-0.026
apped-0.024
 bleeding-0.024
ir-0.023
Token was
Token ID373
Feature activation+2.22e-02

Pos logit contributions
 crying+0.040
 weeping+0.040
Ros+0.038
 bleeding+0.038
apped+0.038
Neg logit contributions
 stalls-0.057
 fare-0.053
 hike-0.053
 stall-0.053
 banks-0.052
Token so
Token ID523
Feature activation-1.37e-02

Pos logit contributions
Ros+0.033
ir+0.032
 weeping+0.032
 bleeding+0.031
 crying+0.031
Neg logit contributions
 stalls-0.039
 stall-0.037
 hike-0.036
 banks-0.035
 flavors-0.035
Token happy
Token ID3772
Feature activation+4.26e-02

Pos logit contributions
 stalls+0.020
 stall+0.019
 flavors+0.018
 list+0.018
 fare+0.017
Neg logit contributions
Ros-0.024
ir-0.024
 blood-0.023
el-0.023
 bleeding-0.023
Token to
Token ID284
Feature activation+3.39e-02

Pos logit contributions
 weeping+0.059
 crying+0.059
 bleeding+0.056
Ros+0.055
ir+0.054
Neg logit contributions
 stalls-0.082
 stall-0.077
 hike-0.077
 banks-0.077
 fare-0.076
Token have
Token ID423
Feature activation+4.73e-02

Pos logit contributions
 bleeding+0.058
 weeping+0.057
Ros+0.056
apped+0.056
 streaming+0.056
Neg logit contributions
 hike-0.048
 stalls-0.047
 stall-0.046
 banks-0.046
 fare-0.046
Token returned
Token ID4504
Feature activation+4.62e-02

Pos logit contributions
 weeping+0.064
 crying+0.063
Ros+0.062
 bleeding+0.061
ir+0.060
Neg logit contributions
 stalls-0.091
 stall-0.087
 hike-0.087
 banks-0.086
 fare-0.085
Token the
Token ID262
Feature activation+2.95e-02

Pos logit contributions
 crying+0.057
 weeping+0.057
 bleeding+0.054
Ros+0.053
 Ros+0.052
Neg logit contributions
 stalls-0.094
 hike-0.092
 banks-0.089
 flavors-0.089
 stall-0.089
Token button
Token ID4936
Feature activation+2.86e-02

Pos logit contributions
 weeping+0.049
Ros+0.049
ir+0.048
 bleeding+0.047
 crying+0.047
Neg logit contributions
 stalls-0.047
 hike-0.044
 stall-0.044
 flavors-0.044
 fare-0.043
Token and
Token ID290
Feature activation+1.83e-02

Pos logit contributions
 weeping+0.040
 crying+0.039
 bleeding+0.038
Ros+0.038
ir+0.037
Neg logit contributions
 stalls-0.054
 hike-0.051
 flavors-0.051
 stall-0.050
 banks-0.050
Token her
Token ID607
Feature activation+2.61e-02

Pos logit contributions
 weeping+0.029
Ros+0.029
 crying+0.029
 bleeding+0.028
ir+0.028
Neg logit contributions
 stalls-0.031
 hike-0.029
 stall-0.029
 flavors-0.028
 banks-0.028
Token dad
Token ID9955
Feature activation+4.72e-02

Pos logit contributions
 weeping+0.037
Ros+0.036
 crying+0.036
 bleeding+0.036
ir+0.035
Neg logit contributions
 stalls-0.049
 stall-0.046
 hike-0.046
 flavors-0.045
 banks-0.045
Token the
Token ID262
Feature activation+3.84e-03

Pos logit contributions
 stalls+0.001
 hike+0.001
 banks+0.001
 fare+0.001
 stall+0.001
Neg logit contributions
 bleeding-0.001
 weeping-0.001
 crying-0.001
Ros-0.001
ir-0.001
Token clipboard
Token ID47999
Feature activation-1.68e-02

Pos logit contributions
Ros+0.006
ir+0.006
apped+0.006
 weeping+0.006
 crying+0.006
Neg logit contributions
 stalls-0.006
 stall-0.006
 hike-0.006
 banks-0.006
 flavors-0.006
Token.
Token ID13
Feature activation-1.43e-02

Pos logit contributions
 stalls+0.034
 hike+0.034
 banks+0.032
 flavors+0.031
 fare+0.031
Neg logit contributions
 crying-0.022
 weeping-0.021
 bleeding-0.020
el-0.020
 streaming-0.020
Token "
Token ID366
Feature activation+1.91e-03

Pos logit contributions
 stalls+0.026
 stall+0.024
 hike+0.024
 flavors+0.024
 banks+0.024
Neg logit contributions
 weeping-0.021
 crying-0.020
 bleeding-0.020
Ros-0.020
ir-0.020
TokenThere
Token ID1858
Feature activation+2.97e-02

Pos logit contributions
 weeping+0.003
 crying+0.003
ir+0.002
 bleeding+0.002
apped+0.002
Neg logit contributions
 stalls-0.004
 stall-0.003
 flavors-0.003
 hike-0.003
 fare-0.003
Token,
Token ID11
Feature activation-2.37e-03

Pos logit contributions
 crying+0.039
 weeping+0.039
el+0.038
 bleeding+0.037
ir+0.037
Neg logit contributions
 stalls-0.057
 stall-0.054
 hike-0.054
 fare-0.053
 banks-0.053
Token you
Token ID345
Feature activation+1.04e-03

Pos logit contributions
 stalls+0.005
 hike+0.004
 stall+0.004
 flavors+0.004
 fare+0.004
Neg logit contributions
 weeping-0.003
Ros-0.003
ir-0.003
 crying-0.003
 bleeding-0.003
Token are
Token ID389
Feature activation+4.66e-03

Pos logit contributions
 weeping+0.002
Ros+0.002
 crying+0.002
 bleeding+0.002
ir+0.002
Neg logit contributions
 stalls-0.002
 hike-0.002
 banks-0.002
 fare-0.001
 stall-0.001
Token all
Token ID477
Feature activation+2.95e-02

Pos logit contributions
 weeping+0.006
 crying+0.006
 bleeding+0.006
Ros+0.006
ir+0.006
Neg logit contributions
 stalls-0.009
 hike-0.009
 stall-0.009
 banks-0.009
 flavors-0.009
Token fixed
Token ID5969
Feature activation+2.57e-02

Pos logit contributions
Ros+0.039
ir+0.038
 weeping+0.038
apped+0.037
 crying+0.037
Neg logit contributions
 stalls-0.056
 stall-0.053
 flavors-0.053
 hike-0.053
 banks-0.052
Token.
Token ID13
Feature activation+7.01e-03

Pos logit contributions
 bleeding+0.036
 weeping+0.035
 crying+0.035
el+0.035
ir+0.034
Neg logit contributions
 stalls-0.048
 hike-0.046
 stall-0.046
 fare-0.044
 banks-0.044
Token day
Token ID1110
Feature activation+7.71e-03

Pos logit contributions
 stalls+0.048
 flavors+0.046
 appet+0.044
nic+0.043
 banks+0.043
Neg logit contributions
el-0.056
 crying-0.056
 weeping-0.053
 microphone-0.052
 bleeding-0.052
Token of
Token ID286
Feature activation+8.33e-05

Pos logit contributions
 weeping+0.007
 crying+0.007
Ros+0.007
 bleeding+0.007
ir+0.007
Neg logit contributions
 stalls-0.018
 stall-0.017
 hike-0.017
 flavors-0.017
 banks-0.017
Token the
Token ID262
Feature activation-1.85e-02

Pos logit contributions
Ros+0.000
 bleeding+0.000
apped+0.000
ir+0.000
 weeping+0.000
Neg logit contributions
 hike-0.000
 stalls-0.000
 stall-0.000
 list-0.000
 flavors-0.000
Token delivery
Token ID7585
Feature activation-3.56e-02

Pos logit contributions
 stalls+0.025
 banks+0.024
 flavors+0.023
nic+0.022
 belts+0.022
Neg logit contributions
 crying-0.033
el-0.032
 weeping-0.032
 microphone-0.031
 bleeding-0.030
Token,
Token ID11
Feature activation+2.95e-04

Pos logit contributions
 banks+0.057
 stalls+0.056
 flavors+0.054
nic+0.053
 hike+0.053
Neg logit contributions
 crying-0.057
 weeping-0.054
 microphone-0.053
el-0.053
apped-0.052
Token Lily
Token ID20037
Feature activation+1.15e-02

Pos logit contributions
ir+0.000
 weeping+0.000
 bleeding+0.000
el+0.000
 crying+0.000
Neg logit contributions
 stalls-0.001
 hike-0.001
 stall-0.001
 banks-0.001
 flavors-0.001
Token waited
Token ID13488
Feature activation-3.29e-02

Pos logit contributions
 crying+0.017
 weeping+0.016
Ros+0.016
 bleeding+0.015
el+0.015
Neg logit contributions
 fare-0.021
 stalls-0.021
 flavors-0.021
 stall-0.020
 hike-0.020
Token and
Token ID290
Feature activation-2.04e-02

Pos logit contributions
 stalls+0.063
 flavors+0.061
 stall+0.060
 hike+0.060
 fare+0.060
Neg logit contributions
 weeping-0.045
 crying-0.044
 bleeding-0.042
ir-0.041
 streaming-0.041
Token waited
Token ID13488
Feature activation-3.71e-02

Pos logit contributions
 stalls+0.029
 stall+0.028
 banks+0.027
 hike+0.027
 flavors+0.027
Neg logit contributions
 weeping-0.037
 crying-0.037
Ros-0.036
 bleeding-0.036
ir-0.035
Token.
Token ID13
Feature activation-1.06e-02

Pos logit contributions
 stalls+0.067
 flavors+0.064
 stall+0.063
 hike+0.062
 fare+0.062
Neg logit contributions
 weeping-0.055
 crying-0.054
 bleeding-0.053
ir-0.052
Ros-0.051
Token Finally
Token ID9461
Feature activation-2.17e-02

Pos logit contributions
 stalls+0.017
 stall+0.016
 hike+0.016
 banks+0.016
 fare+0.015
Neg logit contributions
 weeping-0.017
 bleeding-0.017
ir-0.017
 crying-0.017
Ros-0.017
Token,
Token ID11
Feature activation+8.26e-03

Pos logit contributions
 weeping+0.037
Ros+0.037
 crying+0.036
 bleeding+0.035
ir+0.035
Neg logit contributions
 stalls-0.070
 stall-0.066
 hike-0.066
 flavors-0.065
 banks-0.065
Token determined
Token ID5295
Feature activation-9.06e-03

Pos logit contributions
 weeping+0.013
Ros+0.013
 crying+0.013
 bleeding+0.013
ir+0.013
Neg logit contributions
 stalls-0.014
 hike-0.013
 stall-0.013
 banks-0.012
 flavors-0.012
Token not
Token ID407
Feature activation+2.51e-02

Pos logit contributions
 stalls+0.021
 stall+0.020
 flavors+0.020
 hike+0.020
 banks+0.019
Neg logit contributions
 crying-0.010
 weeping-0.009
 bleeding-0.009
 blood-0.009
 streaming-0.008
Token to
Token ID284
Feature activation+7.52e-03

Pos logit contributions
 weeping+0.034
 crying+0.034
 bleeding+0.032
Ros+0.031
 streaming+0.030
Neg logit contributions
 stalls-0.050
 stall-0.047
 hike-0.046
 flavors-0.046
 banks-0.046
Token be
Token ID307
Feature activation+2.88e-02

Pos logit contributions
 weeping+0.012
 crying+0.012
 bleeding+0.011
Ros+0.011
ir+0.011
Neg logit contributions
 stalls-0.013
 stall-0.012
 hike-0.012
 flavors-0.012
 banks-0.012
Token scared
Token ID12008
Feature activation+6.24e-03

Pos logit contributions
 weeping+0.034
Ros+0.034
 crying+0.033
ir+0.033
 bleeding+0.033
Neg logit contributions
 stalls-0.059
 stall-0.056
 hike-0.056
 flavors-0.055
 banks-0.055
Token of
Token ID286
Feature activation+3.94e-02

Pos logit contributions
Ros+0.009
 weeping+0.009
ir+0.009
apped+0.009
 bleeding+0.009
Neg logit contributions
 stalls-0.011
 hike-0.010
 stall-0.010
 banks-0.010
 flavors-0.010
Token the
Token ID262
Feature activation+2.39e-02

Pos logit contributions
 weeping+0.052
 crying+0.051
 bleeding+0.050
Ros+0.050
ir+0.050
Neg logit contributions
 stalls-0.077
 hike-0.073
 stall-0.073
 banks-0.072
 flavors-0.071
Token earthquake
Token ID16295
Feature activation+3.30e-02

Pos logit contributions
Ros+0.036
apped+0.035
ir+0.033
 streaming+0.033
 weeping+0.032
Neg logit contributions
 stalls-0.043
 hike-0.042
 banks-0.041
 stall-0.040
 flavors-0.040
Token.
Token ID13
Feature activation+2.73e-02

Pos logit contributions
 crying+0.048
 weeping+0.047
 bleeding+0.045
 streaming+0.043
ir+0.041
Neg logit contributions
 stalls-0.063
 hike-0.059
 flavors-0.059
 stall-0.059
 banks-0.058
Token<|endoftext|>
Token ID50256
Feature activation+1.46e-02

Pos logit contributions
 weeping+0.037
 crying+0.036
Ros+0.036
 bleeding+0.035
ir+0.035
Neg logit contributions
 stalls-0.053
 hike-0.050
 stall-0.050
 flavors-0.049
 banks-0.049
Token Lily
Token ID20037
Feature activation+6.26e-03

Pos logit contributions
 stalls+0.002
 hike+0.002
 stall+0.002
 flavors+0.002
 list+0.002
Neg logit contributions
ir-0.002
 weeping-0.002
 streaming-0.002
aps-0.002
ez-0.002
Token's
Token ID338
Feature activation-8.02e-03

Pos logit contributions
 weeping+0.010
 crying+0.010
Ros+0.010
 bleeding+0.010
ir+0.009
Neg logit contributions
 stalls-0.010
 fare-0.010
 stall-0.010
 hike-0.010
 flavors-0.010
Token mom
Token ID1995
Feature activation+2.41e-02

Pos logit contributions
 stalls+0.013
 stall+0.013
 hike+0.013
 banks+0.013
 flavors+0.013
Neg logit contributions
ir-0.013
Ros-0.012
 weeping-0.012
 streaming-0.011
 bleeding-0.011
Tokenmy
Token ID1820
Feature activation+1.48e-02

Pos logit contributions
 weeping+0.038
Ros+0.037
ir+0.037
 bleeding+0.037
 crying+0.037
Neg logit contributions
 stalls-0.040
 hike-0.038
 stall-0.037
 flavors-0.037
 banks-0.037
Token noticed
Token ID6810
Feature activation+4.43e-03

Pos logit contributions
 crying+0.022
 weeping+0.022
Ros+0.022
 bleeding+0.021
 Ros+0.020
Neg logit contributions
 stalls-0.027
 fare-0.026
 stall-0.025
 banks-0.025
 flavors-0.025
Token that
Token ID326
Feature activation+1.80e-02

Pos logit contributions
apped+0.008
ir+0.008
aps+0.008
 weeping+0.008
apping+0.008
Neg logit contributions
 stalls-0.006
 flavors-0.006
 hike-0.005
 experiments-0.005
 stall-0.005
Token the
Token ID262
Feature activation-4.86e-03

Pos logit contributions
ir+0.024
Ros+0.023
apped+0.023
 weeping+0.023
el+0.022
Neg logit contributions
 stalls-0.035
 hike-0.033
 flavors-0.033
 stall-0.033
 banks-0.031
Token f
Token ID277
Feature activation-3.27e-02

Pos logit contributions
 stalls+0.009
 banks+0.009
 flavors+0.008
 bank+0.008
 stall+0.008
Neg logit contributions
ir-0.007
Ros-0.007
apped-0.007
 weeping-0.007
 Emily-0.006
Tokenauc
Token ID14272
Feature activation-3.33e-02

Pos logit contributions
 stalls+0.048
cases+0.048
 banks+0.047
 flavors+0.044
 experiments+0.043
Neg logit contributions
 weeping-0.055
 crying-0.054
Ros-0.054
 bleeding-0.053
 Ros-0.052
Tokenet
Token ID316
Feature activation-3.04e-03

Pos logit contributions
 stalls+0.039
 banks+0.035
 flavors+0.035
 hike+0.035
 stall+0.035
Neg logit contributions
 weeping-0.070
Ros-0.069
 crying-0.068
 bleeding-0.067
ir-0.067
Token was
Token ID373
Feature activation+4.97e-03

Pos logit contributions
 stalls+0.005
 stall+0.005
 hike+0.005
 flavors+0.005
 banks+0.005
Neg logit contributions
 weeping-0.005
 crying-0.005
 bleeding-0.005
Ros-0.005
ir-0.005
Token big
Token ID1263
Feature activation+2.03e-02

Pos logit contributions
ir+0.034
apped+0.033
ez+0.032
umbled+0.032
 streaming+0.032
Neg logit contributions
 stall-0.028
 list-0.027
 stalls-0.026
 hike-0.026
 ticket-0.024
Token,
Token ID11
Feature activation+2.69e-03

Pos logit contributions
Ros+0.033
ir+0.032
 weeping+0.032
 bleeding+0.032
apped+0.031
Neg logit contributions
 stalls-0.033
 stall-0.032
 hike-0.032
 flavors-0.031
 banks-0.031
Token brown
Token ID7586
Feature activation+7.25e-03

Pos logit contributions
ir+0.004
apped+0.004
 weeping+0.004
Ros+0.004
 crying+0.004
Neg logit contributions
 stalls-0.004
 stall-0.004
 hike-0.004
 flavors-0.004
 fare-0.004
Token t
Token ID256
Feature activation+9.16e-03

Pos logit contributions
Ros+0.011
ir+0.011
 streaming+0.011
 weeping+0.011
apped+0.011
Neg logit contributions
 stalls-0.012
 stall-0.012
 hike-0.012
 bank-0.011
 banks-0.011
Tokeneddy
Token ID21874
Feature activation+6.16e-02

Pos logit contributions
 weeping+0.009
 Emily+0.009
 streaming+0.009
 bleeding+0.009
 crying+0.009
Neg logit contributions
 stall-0.020
 stalls-0.020
 belts-0.019
 flavors-0.019
cases-0.019
Token bear
Token ID6842
Feature activation+1.81e-02

Pos logit contributions
ir+0.111
apped+0.105
apping+0.104
Ros+0.104
 streaming+0.103
Neg logit contributions
 stall-0.081
 stalls-0.080
 bank-0.079
 flavors-0.078
 belts-0.076
Token that
Token ID326
Feature activation+3.16e-02

Pos logit contributions
 weeping+0.022
 crying+0.022
 bleeding+0.021
Ros+0.021
ir+0.021
Neg logit contributions
 stalls-0.037
 hike-0.036
 flavors-0.035
 stall-0.035
 banks-0.035
Token she
Token ID673
Feature activation+1.88e-02

Pos logit contributions
 weeping+0.047
 crying+0.047
Ros+0.045
 bleeding+0.045
 Ros+0.045
Neg logit contributions
 stalls-0.059
 banks-0.056
 fare-0.055
 stall-0.055
 hike-0.055
Token loved
Token ID6151
Feature activation+2.83e-02

Pos logit contributions
 weeping+0.029
 crying+0.029
Ros+0.029
 bleeding+0.028
ir+0.027
Neg logit contributions
 stalls-0.033
 stall-0.031
 hike-0.031
 fare-0.031
 banks-0.030
Token very
Token ID845
Feature activation-4.71e-02

Pos logit contributions
 weeping+0.037
 crying+0.036
 bleeding+0.035
Ros+0.035
ir+0.033
Neg logit contributions
 stalls-0.057
 stall-0.054
 hike-0.054
 banks-0.054
 fare-0.053
Token much
Token ID881
Feature activation+1.09e-03

Pos logit contributions
 stalls+0.092
 stall+0.088
 hike+0.087
 flavors+0.086
 banks+0.086
Neg logit contributions
 weeping-0.062
 crying-0.060
 bleeding-0.060
Ros-0.059
ir-0.058
Token,
Token ID11
Feature activation-6.49e-05

Pos logit contributions
 stalls+0.011
 hike+0.010
 stall+0.010
 banks+0.009
 flavors+0.009
Neg logit contributions
Ros-0.015
 weeping-0.015
 bleeding-0.015
 streaming-0.015
ir-0.015
Token there
Token ID612
Feature activation+7.06e-03

Pos logit contributions
 hike+0.000
 stall+0.000
 stalls+0.000
 banks+0.000
 flavors+0.000
Neg logit contributions
Ros-0.000
 streaming-0.000
ir-0.000
 weeping-0.000
 microphone-0.000
Token was
Token ID373
Feature activation+1.93e-02

Pos logit contributions
 weeping+0.012
Ros+0.012
 crying+0.012
 bleeding+0.012
ir+0.011
Neg logit contributions
 stalls-0.011
 hike-0.010
 stall-0.010
 flavors-0.010
 banks-0.010
Token a
Token ID257
Feature activation+1.28e-02

Pos logit contributions
 weeping+0.034
 streaming+0.033
ir+0.033
 bleeding+0.032
Ros+0.032
Neg logit contributions
 stalls-0.028
 hike-0.027
 stall-0.026
 flavors-0.026
 banks-0.024
Token big
Token ID1263
Feature activation+1.98e-02

Pos logit contributions
 streaming+0.019
ir+0.019
apped+0.019
el+0.018
 weeping+0.018
Neg logit contributions
 stall-0.022
 stalls-0.022
 hike-0.022
 list-0.020
 flavors-0.020
Token park
Token ID3952
Feature activation-2.90e-02

Pos logit contributions
Ros+0.034
 streaming+0.033
ir+0.033
apped+0.032
 weeping+0.032
Neg logit contributions
 stalls-0.031
 stall-0.030
 hike-0.030
 flavors-0.028
 list-0.028
Token with
Token ID351
Feature activation-6.68e-03

Pos logit contributions
 stalls+0.071
 banks+0.068
 flavors+0.068
 fare+0.068
 hike+0.067
Neg logit contributions
 crying-0.024
apped-0.024
 weeping-0.023
Ros-0.022
 bleeding-0.021
Token a
Token ID257
Feature activation+1.15e-02

Pos logit contributions
 stalls+0.013
 flavors+0.012
 stall+0.012
 hike+0.012
 banks+0.011
Neg logit contributions
 weeping-0.010
 crying-0.009
 bleeding-0.009
ir-0.009
Ros-0.009
Token large
Token ID1588
Feature activation+1.07e-02

Pos logit contributions
 weeping+0.017
ir+0.017
 streaming+0.017
 bleeding+0.017
Ros+0.017
Neg logit contributions
 stalls-0.020
 stall-0.019
 hike-0.019
 flavors-0.019
 banks-0.018
Token playground
Token ID24817
Feature activation-3.52e-02

Pos logit contributions
 weeping+0.018
 crying+0.018
ir+0.017
 bleeding+0.017
 microphone+0.017
Neg logit contributions
 stalls-0.016
 fare-0.015
 banks-0.015
cases-0.015
 flavors-0.015
Token.
Token ID13
Feature activation+8.24e-03

Pos logit contributions
 hike+0.070
 stalls+0.069
 fare+0.068
 flavors+0.067
 banks+0.066
Neg logit contributions
apped-0.045
 crying-0.043
 weeping-0.040
aps-0.039
 bleeding-0.039
Token€
Token ID26391
Feature activation-3.91e-02

Pos logit contributions
 stalls+0.111
 stall+0.098
 belts+0.095
 flavors+0.094
cases+0.093
Neg logit contributions
ir-0.116
el-0.115
uck-0.113
w-0.110
Ros-0.110
Tokenâ„¢
Token ID8151
Feature activation+2.70e-03

Pos logit contributions
 stalls+0.078
 flavors+0.073
 stall+0.072
 banks+0.072
 hike+0.071
Neg logit contributions
 crying-0.050
Ros-0.047
apped-0.047
 weeping-0.046
aps-0.046
Tokent
Token ID83
Feature activation-1.17e-04

Pos logit contributions
 crying+0.004
 weeping+0.004
ir+0.004
 bleeding+0.004
Ros+0.004
Neg logit contributions
 stalls-0.005
 hike-0.005
 flavors-0.005
 banks-0.005
 stall-0.005
Token speak
Token ID2740
Feature activation-1.68e-02

Pos logit contributions
 stalls+0.000
 stall+0.000
 flavors+0.000
 belts+0.000
 banks+0.000
Neg logit contributions
 weeping-0.000
 crying-0.000
ir-0.000
 bleeding-0.000
Ros-0.000
Token.
Token ID13
Feature activation+5.03e-03

Pos logit contributions
 stalls+0.033
 hike+0.032
 stall+0.031
 banks+0.030
 flavors+0.030
Neg logit contributions
 weeping-0.023
 crying-0.023
 bleeding-0.022
 streaming-0.021
ir-0.020
Token
Token ID198
Feature activation-7.99e-03

Pos logit contributions
 weeping+0.007
 crying+0.007
Ros+0.007
 bleeding+0.007
 Ros+0.007
Neg logit contributions
 stalls-0.009
 hike-0.009
 flavors-0.009
 stall-0.009
 banks-0.009
Token
Token ID198
Feature activation-6.74e-04

Pos logit contributions
 stalls+0.016
 hike+0.015
 stall+0.015
 flavors+0.015
 banks+0.015
Neg logit contributions
 weeping-0.010
Ros-0.010
 crying-0.010
 bleeding-0.010
ir-0.010
TokenBut
Token ID1537
Feature activation-1.01e-02

Pos logit contributions
 stalls+0.001
 flavors+0.001
 hike+0.001
 stall+0.001
 fare+0.001
Neg logit contributions
Ros-0.001
ir-0.001
 bleeding-0.001
 crying-0.001
 weeping-0.001
Token then
Token ID788
Feature activation+9.80e-03

Pos logit contributions
 stalls+0.020
 flavors+0.018
 stall+0.018
 hike+0.018
 fare+0.018
Neg logit contributions
 crying-0.014
 weeping-0.014
 bleeding-0.014
Ros-0.013
 Ros-0.013
Token the
Token ID262
Feature activation-3.80e-03

Pos logit contributions
 crying+0.014
 Ros+0.014
 weeping+0.014
 bleeding+0.013
Ros+0.013
Neg logit contributions
 stalls-0.018
 stall-0.017
 fare-0.017
 banks-0.017
 flavors-0.017
Token baker
Token ID46412
Feature activation-3.85e-04

Pos logit contributions
 stalls+0.006
 stall+0.006
 hike+0.006
 flavors+0.006
 fare+0.006
Neg logit contributions
ir-0.006
Ros-0.006
 weeping-0.006
 streaming-0.006
apped-0.006
Token and
Token ID290
Feature activation-1.77e-02

Pos logit contributions
 stalls+0.031
 flavors+0.030
 hike+0.030
 fare+0.030
 banks+0.029
Neg logit contributions
 crying-0.018
 weeping-0.018
apped-0.017
 bleeding-0.017
Ros-0.017
Token started
Token ID2067
Feature activation-1.27e-03

Pos logit contributions
 stalls+0.029
 stall+0.027
 hike+0.027
 banks+0.027
 flavors+0.027
Neg logit contributions
 weeping-0.029
 crying-0.028
Ros-0.028
 bleeding-0.028
ir-0.028
Token looking
Token ID2045
Feature activation-2.65e-02

Pos logit contributions
 stalls+0.003
 fare+0.002
 banks+0.002
 flavors+0.002
 stall+0.002
Neg logit contributions
 crying-0.002
 weeping-0.002
Ros-0.001
 bleeding-0.001
 streaming-0.001
Token at
Token ID379
Feature activation+6.11e-03

Pos logit contributions
 stalls+0.058
 fare+0.057
 hike+0.057
 flavors+0.057
 stall+0.056
Neg logit contributions
 weeping-0.026
 crying-0.026
apped-0.026
 bleeding-0.024
ir-0.024
Token the
Token ID262
Feature activation-1.29e-02

Pos logit contributions
Ros+0.009
 weeping+0.009
 bleeding+0.008
 crying+0.008
ir+0.008
Neg logit contributions
 stalls-0.011
 stall-0.011
 flavors-0.011
 hike-0.011
 banks-0.011
Token pictures
Token ID5986
Feature activation-2.72e-02

Pos logit contributions
 stalls+0.023
 stall+0.021
 hike+0.021
 flavors+0.021
 banks+0.021
Neg logit contributions
 weeping-0.019
 crying-0.019
Ros-0.019
 bleeding-0.019
ir-0.019
Token.
Token ID13
Feature activation-7.77e-03

Pos logit contributions
 stalls+0.052
 hike+0.051
 banks+0.051
 ranc+0.049
 fare+0.049
Neg logit contributions
 crying-0.036
 weeping-0.035
apped-0.035
 bleeding-0.034
 streaming-0.034
Token He
Token ID679
Feature activation-1.33e-02

Pos logit contributions
 stalls+0.016
 hike+0.015
 banks+0.015
nic+0.015
 flavors+0.015
Neg logit contributions
apped-0.009
 crying-0.008
 Emily-0.008
 Ros-0.008
 weeping-0.008
Token saw
Token ID2497
Feature activation+2.60e-03

Pos logit contributions
 stalls+0.024
 fare+0.023
 stall+0.022
 hike+0.022
 banks+0.022
Neg logit contributions
 weeping-0.019
 crying-0.019
Ros-0.019
apped-0.018
 bleeding-0.018
Token animals
Token ID4695
Feature activation-1.83e-02

Pos logit contributions
 weeping+0.004
ir+0.004
Ros+0.003
 bleeding+0.003
 crying+0.003
Neg logit contributions
 stalls-0.005
 stall-0.005
 flavors-0.005
 hike-0.005
 banks-0.005
Token,
Token ID11
Feature activation-1.49e-02

Pos logit contributions
 stalls+0.035
 fare+0.035
 flavors+0.034
 hike+0.034
 banks+0.034
Neg logit contributions
 crying-0.024
 weeping-0.024
 bleeding-0.022
 streaming-0.022
 microphone-0.022
Token not
Token ID407
Feature activation-9.75e-03

Pos logit contributions
 crying+0.007
 weeping+0.007
 bleeding+0.007
 blood+0.007
Ros+0.007
Neg logit contributions
 stalls-0.010
 stall-0.009
 hike-0.009
 banks-0.009
 fare-0.009
Token to
Token ID284
Feature activation-1.80e-02

Pos logit contributions
 stalls+0.019
 stall+0.018
 hike+0.018
 fare+0.018
 flavors+0.018
Neg logit contributions
 weeping-0.013
 crying-0.013
 bleeding-0.013
Ros-0.012
ir-0.012
Token turn
Token ID1210
Feature activation-1.22e-02

Pos logit contributions
 stalls+0.031
 stall+0.028
 hike+0.028
 flavors+0.028
 fare+0.027
Neg logit contributions
 weeping-0.029
 crying-0.029
 bleeding-0.028
Ros-0.028
ir-0.028
Token over
Token ID625
Feature activation+2.48e-02

Pos logit contributions
 stalls+0.023
 hike+0.022
 banks+0.022
 flavors+0.022
 stall+0.022
Neg logit contributions
Ros-0.016
apped-0.016
 weeping-0.016
ir-0.015
 bleeding-0.015
Token any
Token ID597
Feature activation-3.45e-03

Pos logit contributions
 weeping+0.031
 crying+0.031
 bleeding+0.030
ir+0.030
 blood+0.029
Neg logit contributions
 stalls-0.049
 hike-0.047
 stall-0.047
 fare-0.047
 banks-0.046
Token sm
Token ID895
Feature activation-2.02e-02

Pos logit contributions
 stalls+0.006
 hike+0.005
 stall+0.005
 banks+0.005
 flavors+0.005
Neg logit contributions
Ros-0.006
 weeping-0.006
 crying-0.005
 bleeding-0.005
ir-0.005
Tokenelly
Token ID6148
Feature activation-1.56e-02

Pos logit contributions
 stalls+0.045
 flavors+0.043
 stall+0.042
 banks+0.042
 hike+0.042
Neg logit contributions
 weeping-0.021
 bleeding-0.021
Ros-0.021
 crying-0.021
 microphone-0.020
Token worms
Token ID29174
Feature activation-8.73e-03

Pos logit contributions
 fare+0.027
 stalls+0.027
 stall+0.026
cases+0.026
nic+0.026
Neg logit contributions
 weeping-0.022
 crying-0.022
 blood-0.022
 bleeding-0.021
 bug-0.021
Token,
Token ID11
Feature activation-6.11e-03

Pos logit contributions
 stalls+0.018
 flavors+0.018
 fare+0.017
 stall+0.017
 belts+0.017
Neg logit contributions
 crying-0.011
 weeping-0.011
 bleeding-0.010
ir-0.010
 streaming-0.010
Token then
Token ID788
Feature activation+2.47e-02

Pos logit contributions
 stalls+0.010
 stall+0.009
 hike+0.009
 fare+0.009
 flavors+0.009
Neg logit contributions
 weeping-0.010
 crying-0.010
 bleeding-0.010
Ros-0.010
 blood-0.010
Token he
Token ID339
Feature activation+6.73e-03

Pos logit contributions
 weeping+0.042
 crying+0.042
 bleeding+0.041
Ros+0.040
 blood+0.039
Neg logit contributions
 stalls-0.039
 stall-0.037
 fare-0.036
 hike-0.036
 flavors-0.035
Token and
Token ID290
Feature activation-2.47e-02

Pos logit contributions
 crying+0.011
 weeping+0.010
 bleeding+0.010
 blood+0.010
apped+0.010
Neg logit contributions
 stalls-0.019
 hike-0.018
 fare-0.018
 banks-0.018
 flavors-0.018
Token Ben
Token ID3932
Feature activation-8.05e-04

Pos logit contributions
 stalls+0.063
 flavors+0.061
 fare+0.060
 banks+0.060
 stall+0.059
Neg logit contributions
 crying-0.018
 weeping-0.018
ir-0.017
 blood-0.016
 bleeding-0.016
Token are
Token ID389
Feature activation+4.24e-03

Pos logit contributions
 stalls+0.002
 flavors+0.002
 stall+0.002
 hike+0.001
 banks+0.001
Neg logit contributions
 crying-0.001
 weeping-0.001
 bleeding-0.001
ir-0.001
 blood-0.001
Token shocked
Token ID11472
Feature activation+1.63e-02

Pos logit contributions
 weeping+0.005
Ros+0.005
 crying+0.005
ir+0.005
 bleeding+0.005
Neg logit contributions
 stalls-0.008
 stall-0.008
 hike-0.008
 flavors-0.008
 banks-0.008
Token.
Token ID13
Feature activation-2.08e-03

Pos logit contributions
Ros+0.032
 weeping+0.032
ir+0.031
 bleeding+0.031
 crying+0.030
Neg logit contributions
 stalls-0.021
 stall-0.019
 flavors-0.019
 hike-0.019
 banks-0.018
Token They
Token ID1119
Feature activation-1.41e-02

Pos logit contributions
 stalls+0.004
 hike+0.004
 stall+0.004
 flavors+0.004
 banks+0.004
Neg logit contributions
 weeping-0.003
 crying-0.003
Ros-0.003
 bleeding-0.003
apped-0.003
Token see
Token ID766
Feature activation-3.30e-03

Pos logit contributions
 stalls+0.026
 stall+0.024
 flavors+0.024
 hike+0.023
 fare+0.023
Neg logit contributions
 weeping-0.021
 crying-0.021
 bleeding-0.020
ir-0.020
 blood-0.020
Token the
Token ID262
Feature activation+2.22e-03

Pos logit contributions
 stalls+0.006
 stall+0.006
 hike+0.006
 banks+0.006
 flavors+0.006
Neg logit contributions
Ros-0.005
ir-0.005
 weeping-0.005
 streaming-0.004
 bleeding-0.004
Token toys
Token ID14958
Feature activation-1.99e-02

Pos logit contributions
 weeping+0.003
 crying+0.003
 bleeding+0.003
Ros+0.003
ir+0.003
Neg logit contributions
 stalls-0.004
 hike-0.004
 stall-0.004
 banks-0.004
 flavors-0.004
Token.
Token ID13
Feature activation-9.72e-03

Pos logit contributions
 stalls+0.043
 flavors+0.041
 hike+0.041
 banks+0.041
nic+0.040
Neg logit contributions
 crying-0.023
 weeping-0.023
 bleeding-0.022
 blood-0.021
 microphone-0.020
Token They
Token ID1119
Feature activation-1.56e-02

Pos logit contributions
 stalls+0.018
 hike+0.017
 stall+0.017
 flavors+0.017
 banks+0.017
Neg logit contributions
 weeping-0.013
 crying-0.013
Ros-0.013
 bleeding-0.013
ir-0.013
Token smiled
Token ID13541
Feature activation+2.66e-02

Pos logit contributions
 stalls+0.037
 fare+0.037
 banks+0.035
 products+0.035
 stall+0.035
Neg logit contributions
 crying-0.025
Ros-0.025
 weeping-0.025
apped-0.024
 Emily-0.024
Token and
Token ID290
Feature activation-1.10e-02

Pos logit contributions
 weeping+0.036
 crying+0.035
Ros+0.035
 bleeding+0.035
ir+0.034
Neg logit contributions
 stalls-0.051
 hike-0.048
 stall-0.048
 banks-0.048
 flavors-0.048
Token whispered
Token ID25029
Feature activation-2.41e-02

Pos logit contributions
 stalls+0.019
 fare+0.018
 stall+0.018
 banks+0.018
 hike+0.018
Neg logit contributions
 crying-0.017
 weeping-0.017
apped-0.017
Ros-0.016
 bleeding-0.016
Token:
Token ID25
Feature activation-2.25e-02

Pos logit contributions
 stalls+0.047
 flavors+0.045
 banks+0.045
 hike+0.045
 stall+0.044
Neg logit contributions
 weeping-0.033
 crying-0.033
 bleeding-0.031
apped-0.031
Ros-0.030
Token �
Token ID6184
Feature activation-6.24e-02

Pos logit contributions
 stalls+0.035
 flavors+0.033
 banks+0.033
 hike+0.032
 fare+0.032
Neg logit contributions
 weeping-0.039
 crying-0.039
Ros-0.038
 bleeding-0.037
ir-0.037
Token�
Token ID95
Feature activation-5.23e-02

Pos logit contributions
 stalls+0.107
 flavors+0.100
 banks+0.093
 fare+0.092
cases+0.092
Neg logit contributions
 weeping-0.100
 crying-0.095
ir-0.095
Ros-0.092
 Ros-0.090
Token€
Token ID26391
Feature activation-1.34e-02

Pos logit contributions
 stalls+0.100
cases+0.094
 flavors+0.092
 belts+0.091
 hike+0.091
Neg logit contributions
Ros-0.070
 weeping-0.069
 crying-0.066
 bleeding-0.064
 Emily-0.063
Token�
Token ID129
Feature activation-4.12e-02

Pos logit contributions
 stalls+0.024
 banks+0.023
 fare+0.023
 flavors+0.023
 stall+0.022
Neg logit contributions
Ros-0.020
 crying-0.019
 weeping-0.019
 bleeding-0.018
 Ros-0.018
Token�
Token ID241
Feature activation-1.11e-02

Pos logit contributions
 stalls+0.063
 flavors+0.061
cases+0.057
 fare+0.056
 belts+0.055
Neg logit contributions
 weeping-0.074
ir-0.073
 crying-0.071
el-0.068
 bleeding-0.065
TokenOnce
Token ID7454
Feature activation-4.83e-03

Pos logit contributions
 stalls+0.024
 banks+0.023
 hike+0.023
 stall+0.023
 list+0.022
Neg logit contributions
 crying-0.012
 weeping-0.012
Ros-0.012
 bleeding-0.011
ir-0.011
Token upon
Token ID2402
Feature activation-8.76e-04

Pos logit contributions
 stalls+0.008
 stall+0.008
 hike+0.008
 banks+0.008
 flavors+0.008
Neg logit contributions
 weeping-0.007
 crying-0.007
Ros-0.007
 bleeding-0.007
ir-0.007
Token the
Token ID262
Feature activation+1.96e-03

Pos logit contributions
 stalls+0.041
 hike+0.039
 stall+0.038
 flavors+0.037
 banks+0.037
Neg logit contributions
ir-0.039
Ros-0.038
apped-0.036
el-0.036
 weeping-0.035
Token window
Token ID4324
Feature activation+4.12e-03

Pos logit contributions
 weeping+0.003
 crying+0.003
Ros+0.003
 bleeding+0.003
ir+0.003
Neg logit contributions
 stalls-0.003
 hike-0.003
 stall-0.003
 flavors-0.003
 banks-0.003
Token.
Token ID13
Feature activation-2.00e-03

Pos logit contributions
 weeping+0.006
 crying+0.006
 bleeding+0.006
ir+0.006
el+0.006
Neg logit contributions
 stalls-0.007
 hike-0.007
 stall-0.007
 flavors-0.007
 banks-0.007
Token She
Token ID1375
Feature activation-8.01e-03

Pos logit contributions
 stalls+0.003
 stall+0.003
 fare+0.003
 banks+0.003
 hike+0.003
Neg logit contributions
ir-0.003
 weeping-0.003
 streaming-0.003
 bleeding-0.003
Ros-0.003
Token tells
Token ID4952
Feature activation-2.59e-02

Pos logit contributions
 stalls+0.013
 stall+0.013
 hike+0.013
 banks+0.012
 fare+0.012
Neg logit contributions
 weeping-0.013
 crying-0.013
 bleeding-0.012
 blood-0.012
Ros-0.012
Token them
Token ID606
Feature activation-5.75e-02

Pos logit contributions
 stalls+0.051
 hike+0.049
 stall+0.048
 banks+0.048
 flavors+0.048
Neg logit contributions
 weeping-0.033
Ros-0.033
 bleeding-0.032
 crying-0.032
ir-0.032
Token to
Token ID284
Feature activation-4.26e-02

Pos logit contributions
 stalls+0.122
 stall+0.121
 fare+0.115
 banks+0.115
 hike+0.115
Neg logit contributions
 crying-0.067
 weeping-0.065
 bleeding-0.062
 blood-0.061
 microphone-0.057
Token spray
Token ID11662
Feature activation-3.44e-02

Pos logit contributions
 stalls+0.068
 hike+0.063
 stall+0.063
 banks+0.062
 flavors+0.062
Neg logit contributions
 weeping-0.071
 crying-0.070
Ros-0.070
 bleeding-0.069
ir-0.069
Token the
Token ID262
Feature activation+2.75e-03

Pos logit contributions
 stalls+0.060
 banks+0.057
 flavors+0.057
 stall+0.056
 hike+0.056
Neg logit contributions
 weeping-0.053
Ros-0.053
 crying-0.050
 bleeding-0.049
ir-0.049
Token water
Token ID1660
Feature activation-4.03e-03

Pos logit contributions
 weeping+0.004
 crying+0.004
 bleeding+0.004
Ros+0.004
ir+0.004
Neg logit contributions
 stalls-0.005
 hike-0.004
 stall-0.004
 flavors-0.004
 banks-0.004
Token on
Token ID319
Feature activation+2.51e-04

Pos logit contributions
 stalls+0.008
 hike+0.007
 stall+0.007
 flavors+0.007
 banks+0.007
Neg logit contributions
 weeping-0.006
 crying-0.005
 bleeding-0.005
Ros-0.005
ir-0.005
Token mouth
Token ID5422
Feature activation-1.07e-02

Pos logit contributions
 weeping+0.014
 crying+0.014
 bleeding+0.013
Ros+0.013
ir+0.013
Neg logit contributions
 stalls-0.024
 hike-0.023
 stall-0.023
 flavors-0.023
 banks-0.023
Token.
Token ID13
Feature activation-2.08e-03

Pos logit contributions
 stalls+0.023
 hike+0.022
 fare+0.022
nic+0.022
 flavors+0.021
Neg logit contributions
 crying-0.012
 bleeding-0.012
 weeping-0.012
 streaming-0.011
apped-0.011
Token
Token ID220
Feature activation+6.42e-04

Pos logit contributions
 fare+0.004
 stalls+0.004
 hike+0.004
nic+0.004
 belts+0.004
Neg logit contributions
 crying-0.002
 weeping-0.002
apped-0.002
ucky-0.002
 Emily-0.002
Token
Token ID198
Feature activation-5.57e-03

Pos logit contributions
 weeping+0.001
 crying+0.001
Ros+0.001
 bleeding+0.001
apped+0.001
Neg logit contributions
 stalls-0.001
 hike-0.001
 stall-0.001
 flavors-0.001
 banks-0.001
Token
Token ID198
Feature activation-6.97e-03

Pos logit contributions
 stalls+0.013
 banks+0.013
 flavors+0.013
 hike+0.013
 fare+0.013
Neg logit contributions
Ros-0.005
 crying-0.004
apped-0.004
 weeping-0.004
aps-0.004
TokenOne
Token ID3198
Feature activation-5.12e-02

Pos logit contributions
 fare+0.013
 banks+0.013
 experiments+0.013
 bank+0.013
 hike+0.013
Neg logit contributions
apping-0.008
Ros-0.008
ucky-0.007
apped-0.007
 crying-0.007
Token day
Token ID1110
Feature activation-5.24e-03

Pos logit contributions
 stalls+0.088
 banks+0.082
 stall+0.081
 fare+0.081
 flavors+0.081
Neg logit contributions
 weeping-0.081
 crying-0.080
 bleeding-0.079
ir-0.077
Ros-0.077
Token,
Token ID11
Feature activation-1.46e-02

Pos logit contributions
 fare+0.011
 banks+0.011
 stalls+0.011
 hike+0.010
 stall+0.010
Neg logit contributions
 crying-0.006
apped-0.006
 bleeding-0.005
 weeping-0.005
 streaming-0.005
Token Jimmy
Token ID12963
Feature activation-2.99e-03

Pos logit contributions
 stalls+0.030
 fare+0.029
 banks+0.029
 hike+0.028
 stall+0.028
Neg logit contributions
 crying-0.017
 bleeding-0.017
 weeping-0.017
apped-0.016
 streaming-0.016
Token was
Token ID373
Feature activation+1.85e-02

Pos logit contributions
 stalls+0.005
 fare+0.005
 flavors+0.005
 stall+0.005
 hike+0.005
Neg logit contributions
 crying-0.005
 weeping-0.004
 bleeding-0.004
Ros-0.004
apped-0.004
Token alone
Token ID3436
Feature activation+2.28e-02

Pos logit contributions
Ros+0.023
ir+0.023
el+0.022
 weeping+0.021
 bleeding+0.021
Neg logit contributions
 stalls-0.037
 stall-0.035
 hike-0.035
 banks-0.034
 bank-0.033
Token a
Token ID257
Feature activation-1.03e-02

Pos logit contributions
 stalls+0.059
 flavors+0.056
 stall+0.056
 hike+0.056
 banks+0.055
Neg logit contributions
 weeping-0.036
 crying-0.035
 bleeding-0.035
Ros-0.034
ir-0.034
Token whip
Token ID22972
Feature activation-2.65e-02

Pos logit contributions
 stalls+0.019
 stall+0.018
 hike+0.018
 flavors+0.017
 banks+0.017
Neg logit contributions
 weeping-0.015
Ros-0.014
 crying-0.014
ir-0.014
 bleeding-0.014
Token.
Token ID13
Feature activation-3.02e-03

Pos logit contributions
 stalls+0.050
 flavors+0.048
 hike+0.047
 banks+0.046
 stall+0.045
Neg logit contributions
 crying-0.039
 weeping-0.039
 bleeding-0.038
 microphone-0.036
 streaming-0.036
Token It
Token ID632
Feature activation+1.84e-03

Pos logit contributions
 stalls+0.007
 stall+0.006
 hike+0.006
 flavors+0.006
 banks+0.006
Neg logit contributions
 weeping-0.003
 crying-0.003
 bleeding-0.003
Ros-0.003
ir-0.003
Token was
Token ID373
Feature activation-6.98e-03

Pos logit contributions
 bleeding+0.003
 weeping+0.003
 crying+0.003
Ros+0.003
 streaming+0.002
Neg logit contributions
 stalls-0.003
 stall-0.003
 fare-0.003
 hike-0.003
 banks-0.003
Token very
Token ID845
Feature activation-5.10e-02

Pos logit contributions
 stalls+0.015
 hike+0.014
 banks+0.014
 stall+0.014
 fare+0.014
Neg logit contributions
 weeping-0.008
 bleeding-0.008
 crying-0.008
Ros-0.008
 streaming-0.007
Token long
Token ID890
Feature activation-1.85e-02

Pos logit contributions
 stalls+0.101
 fare+0.096
 banks+0.095
 hike+0.095
 stall+0.093
Neg logit contributions
 weeping-0.068
 crying-0.067
 bleeding-0.067
ir-0.064
Ros-0.064
Token and
Token ID290
Feature activation-1.37e-02

Pos logit contributions
 stalls+0.042
 fare+0.041
 hike+0.040
 flavors+0.040
 stall+0.040
Neg logit contributions
 crying-0.020
 weeping-0.019
 bleeding-0.018
 streaming-0.017
 blood-0.016
Token shiny
Token ID22441
Feature activation-2.10e-02

Pos logit contributions
 stalls+0.026
 hike+0.025
 banks+0.025
 stall+0.025
 fare+0.025
Neg logit contributions
 bleeding-0.019
 weeping-0.019
 crying-0.019
Ros-0.018
 streaming-0.017
Token.
Token ID13
Feature activation+1.57e-03

Pos logit contributions
 stalls+0.040
 hike+0.040
 fare+0.038
 stall+0.038
 banks+0.037
Neg logit contributions
 bleeding-0.029
 weeping-0.029
 crying-0.029
 streaming-0.028
el-0.028
Token Sarah
Token ID10490
Feature activation-9.84e-03

Pos logit contributions
 weeping+0.002
ir+0.002
 bleeding+0.002
 crying+0.002
Ros+0.002
Neg logit contributions
 stalls-0.003
 hike-0.003
 stall-0.003
 flavors-0.003
 banks-0.003
Token mom
Token ID1995
Feature activation+1.99e-02

Pos logit contributions
 stalls+0.023
 stall+0.022
 hike+0.022
 flavors+0.021
 fare+0.021
Neg logit contributions
ir-0.016
 weeping-0.016
Ros-0.016
 bleeding-0.015
 crying-0.015
Token took
Token ID1718
Feature activation+7.28e-03

Pos logit contributions
 crying+0.029
 weeping+0.029
Ros+0.028
 bleeding+0.028
el+0.026
Neg logit contributions
 stalls-0.037
 banks-0.035
 stall-0.035
 fare-0.035
 hike-0.035
Token him
Token ID683
Feature activation-1.92e-02

Pos logit contributions
 weeping+0.012
Ros+0.011
 bleeding+0.011
ir+0.011
apped+0.011
Neg logit contributions
 stalls-0.012
 hike-0.012
 goods-0.011
 stall-0.011
 banks-0.011
Token to
Token ID284
Feature activation-3.24e-02

Pos logit contributions
 stalls+0.036
 stall+0.034
 hike+0.034
 flavors+0.034
 banks+0.034
Neg logit contributions
 weeping-0.027
 crying-0.026
Ros-0.026
 bleeding-0.026
ir-0.025
Token the
Token ID262
Feature activation-2.93e-02

Pos logit contributions
 stalls+0.070
 banks+0.066
 stall+0.066
 flavors+0.066
 fare+0.065
Neg logit contributions
 crying-0.036
 weeping-0.036
 bleeding-0.034
Ros-0.034
 blood-0.033
Token library
Token ID5888
Feature activation-6.78e-02

Pos logit contributions
 stalls+0.035
 banks+0.033
 flavors+0.032
 hike+0.031
 fare+0.031
Neg logit contributions
 weeping-0.059
 crying-0.059
Ros-0.058
 bleeding-0.058
 microphone-0.057
Token to
Token ID284
Feature activation-2.60e-02

Pos logit contributions
 flavors+0.157
 banks+0.151
 stalls+0.148
 hike+0.146
 flavours+0.143
Neg logit contributions
 crying-0.067
apped-0.061
 together-0.059
apping-0.058
aps-0.055
Token pick
Token ID2298
Feature activation-2.16e-02

Pos logit contributions
 stalls+0.045
 banks+0.043
 flavors+0.042
cases+0.042
 belts+0.041
Neg logit contributions
 crying-0.040
 weeping-0.039
Ros-0.039
 Ros-0.038
 bleeding-0.037
Token out
Token ID503
Feature activation-3.29e-02

Pos logit contributions
 stalls+0.046
 hike+0.045
 banks+0.044
 stall+0.044
 fare+0.043
Neg logit contributions
 weeping-0.023
 crying-0.023
 bleeding-0.022
Ros-0.022
el-0.022
Token a
Token ID257
Feature activation-8.75e-03

Pos logit contributions
 stalls+0.069
 hike+0.069
 banks+0.068
 stall+0.067
 fare+0.066
Neg logit contributions
 weeping-0.033
 crying-0.033
 bleeding-0.032
 blood-0.032
el-0.032
Token new
Token ID649
Feature activation+5.23e-04

Pos logit contributions
 stalls+0.016
 banks+0.016
 hike+0.015
 flavors+0.015
 fare+0.015
Neg logit contributions
 weeping-0.012
 crying-0.012
 bleeding-0.011
Ros-0.011
 microphone-0.011