TOP ACTIVATIONS
MAX = 0.099

" You are welcome , Anna . You are a good
" You are right , Anna . The book is not
" You are welcome , Jill . You are the best
" You 're welcome , Anna . I love you .
love you , too , Anna and Ben . You are
says , " Sure , Anna . You can call grandma
" You 're welcome , Lily . I 'm proud of
" You 're welcome , Lily and Ben . I 'm
" You 're welcome , Lily . I love you too
I love you too , Anna and Ben . You are
" That 's good , Anna and Ben . I am
" You are welcome , Anna and Ben !" Mom says
" You 're welcome , Sara and Ben . You are
see what we can come up with ." They
" You 're welcome , Lily . We 're glad you
. " Thank you , Anna . You are a good
, " Thank you , Lily . You are such a
, " Thank you , Lily , thank you , Ben
smiled . They had come up with a creative way to
said , " OK , Lily . I will do that

MIN ACTIVATIONS
MIN = -0.128

delicious vegetables and bright - col oured flowers .
tried to jump on the ostr ich 's back .
, they realized that the ostr ich was very messy .
saved my life ," the ostr ich says .
a strange , raspberry - col oured object hiding under a
its clean feathers . The ostr ich was friendly and Lily
as he opened the ch ocol ates and started eating them
the ostr ich . The ostr ich had a long neck
big , beautiful rainbow - col oured butterfly ! The little
feathers . Ben thought the ostr ich was very original .
his friends dropped her lip bal m on the ground .
couldn 't see what the ostr ich was hiding , but
, she saw that the pr une had disappeared ! She
the ostr ich . The ostr ich did not like them
out the bread . The ostr ich came closer and pe
fence . He puts the ostr ich in the box .
She made sure that the m ule was always nice and
, mom my ! The ostr ich is so fat !"
por cup ine ! The por cup ine qu illed and
the ostr ich . The ostr ich sees them and gets

INTERVAL 0.042 - 0.099
CONTAINS 2.809%

He looked wealthy . He smiled when he saw Tom in
. <|endoftext|> Tom loves to play ball with his dad .
felt warm and safe once again . Jack
And we have to be quiet ." They ran
day , Bob o woke up early . He saw a

INTERVAL -0.015 - 0.042
CONTAINS 73.423%

's coming to visit us tomorrow and we need to pick
to give Sam any of his meat . He thinks Sam
put the stones together and suddenly , the forest lit up
went away , and she had more energy to play with
and th orns , but she never gave up . After

INTERVAL -0.071 - -0.015
CONTAINS 23.644%

spark ly cape . One day Lucy wanted to
The pig replied , " Thank you , Lily !
She picked up Fl uffy and gave her a big hug
, because if they do , they 're more likely to
to play . The rain made him wet , but he

INTERVAL -0.128 - -0.071
CONTAINS 0.125%

. It was a big re ind eer !
But , the c actus had a friend -
ate them together . The p ears were sweet and juicy
day , he lost his er aser while he was drawing
But one day , she accidentally knocked over some furniture .
Token "
Token ID366
Feature activation+2.02e-02

Pos logit contributions
 day+0.014
 good+0.012
 time+0.012
 end+0.012
 team+0.012
Neg logit contributions
iqu-0.049
ridge-0.049
umes-0.047
yles-0.046
rol-0.045
TokenYou
Token ID1639
Feature activation+3.66e-02

Pos logit contributions
,+0.037
 day+0.036
 time+0.035
 end+0.035
 team+0.034
Neg logit contributions
ridge-0.042
iqu-0.041
yles-0.040
umes-0.038
cher-0.038
Token are
Token ID389
Feature activation+3.72e-02

Pos logit contributions
 day+0.064
 afternoon+0.059
 girl+0.057
 time+0.056
 their+0.055
Neg logit contributions
iqu-0.081
ridge-0.080
vell-0.077
umes-0.077
yles-0.074
Token welcome
Token ID7062
Feature activation+6.20e-02

Pos logit contributions
 day+0.043
 team+0.040
 girl+0.039
 moment+0.038
 afternoon+0.038
Neg logit contributions
vell-0.096
iqu-0.094
ridge-0.092
umes-0.092
cher-0.090
Token,
Token ID11
Feature activation+2.09e-02

Pos logit contributions
 day+0.043
 afternoon+0.037
 end+0.030
 team+0.030
 boy+0.025
Neg logit contributions
ridge-0.198
iqu-0.197
umes-0.188
angs-0.186
paste-0.183
Token Anna
Token ID11735
Feature activation+9.88e-02

Pos logit contributions
 day+0.024
 afternoon+0.020
 team+0.019
 town+0.019
,+0.018
Neg logit contributions
ridge-0.056
iqu-0.056
yles-0.055
umes-0.054
paste-0.053
Token.
Token ID13
Feature activation+5.53e-02

Pos logit contributions
 day+0.121
 team+0.120
 afternoon+0.118
 time+0.104
 town+0.103
Neg logit contributions
ridge-0.231
ache-0.217
paste-0.217
ker-0.215
pled-0.215
Token You
Token ID921
Feature activation+3.36e-02

Pos logit contributions
 day+0.109
 end+0.107
 team+0.105
 good+0.105
 time+0.104
Neg logit contributions
ridge-0.106
iqu-0.106
umes-0.098
yles-0.097
rol-0.096
Token are
Token ID389
Feature activation+3.09e-02

Pos logit contributions
 day+0.054
 afternoon+0.049
 girl+0.049
 boy+0.047
 time+0.046
Neg logit contributions
ridge-0.077
iqu-0.076
vell-0.074
umes-0.073
cher-0.072
Token a
Token ID257
Feature activation+5.43e-02

Pos logit contributions
 day+0.033
 team+0.030
 moment+0.029
 girl+0.029
 afternoon+0.029
Neg logit contributions
iqu-0.084
vell-0.084
ridge-0.083
umes-0.082
cher-0.080
Token good
Token ID922
Feature activation+5.15e-02

Pos logit contributions
 day+0.043
 They+0.041
 afternoon+0.040
 time+0.035
,+0.032
Neg logit contributions
vell-0.146
iqu-0.142
ache-0.139
ridge-0.139
rou-0.136
Token "
Token ID366
Feature activation+6.15e-03

Pos logit contributions
 day+0.007
 good+0.006
 time+0.005
 end+0.005
 team+0.005
Neg logit contributions
iqu-0.021
ridge-0.021
umes-0.020
yles-0.020
angs-0.019
TokenYou
Token ID1639
Feature activation+2.55e-02

Pos logit contributions
,+0.011
 day+0.011
 time+0.010
 end+0.010
 good+0.010
Neg logit contributions
ridge-0.013
iqu-0.013
yles-0.013
umes-0.013
paste-0.012
Token are
Token ID389
Feature activation+3.01e-02

Pos logit contributions
 day+0.033
 good+0.031
,+0.031
 time+0.029
 end+0.029
Neg logit contributions
iqu-0.073
ridge-0.070
umes-0.068
yles-0.067
rol-0.067
Token right
Token ID826
Feature activation+5.94e-02

Pos logit contributions
 day+0.030
 team+0.027
 girl+0.026
 boy+0.025
 afternoon+0.024
Neg logit contributions
iqu-0.089
ridge-0.088
vell-0.086
umes-0.085
angs-0.082
Token,
Token ID11
Feature activation+1.78e-02

Pos logit contributions
 day+0.029
 team+0.020
 afternoon+0.017
 good+0.017
 girl+0.017
Neg logit contributions
iqu-0.212
ridge-0.205
umes-0.200
yles-0.197
angs-0.197
Token Anna
Token ID11735
Feature activation+9.33e-02

Pos logit contributions
 day+0.021
 afternoon+0.017
 team+0.017
 town+0.015
 end+0.015
Neg logit contributions
ridge-0.048
iqu-0.048
yles-0.047
umes-0.047
paste-0.045
Token.
Token ID13
Feature activation+4.55e-02

Pos logit contributions
 team+0.104
 day+0.103
 afternoon+0.099
 moment+0.089
 town+0.088
Neg logit contributions
ridge-0.233
iqu-0.223
ker-0.218
paste-0.217
ache-0.216
Token The
Token ID383
Feature activation+1.83e-02

Pos logit contributions
 day+0.078
 good+0.074
,+0.072
 end+0.072
 time+0.071
Neg logit contributions
iqu-0.107
ridge-0.105
umes-0.099
yles-0.098
rol-0.097
Token book
Token ID1492
Feature activation-1.34e-02

Pos logit contributions
 They+0.031
 their+0.031
,+0.029
 day+0.029
 good+0.028
Neg logit contributions
ridge-0.037
iqu-0.036
vell-0.036
cher-0.034
yles-0.033
Token is
Token ID318
Feature activation-1.68e-03

Pos logit contributions
iqu+0.039
ridge+0.039
angs+0.037
umes+0.037
paste+0.036
Neg logit contributions
 day-0.013
 girl-0.011
 good-0.011
 team-0.011
 afternoon-0.011
Token not
Token ID407
Feature activation-1.31e-03

Pos logit contributions
iqu+0.005
ridge+0.005
yles+0.005
umes+0.005
rol+0.005
Neg logit contributions
 day-0.002
 good-0.002
,-0.002
 time-0.002
 and-0.002
Token "
Token ID366
Feature activation+1.54e-02

Pos logit contributions
 day+0.009
 good+0.008
,+0.008
 time+0.007
 end+0.007
Neg logit contributions
iqu-0.037
ridge-0.036
umes-0.034
yles-0.034
rol-0.034
TokenYou
Token ID1639
Feature activation+3.41e-02

Pos logit contributions
 day+0.025
,+0.025
 good+0.024
 end+0.023
 team+0.023
Neg logit contributions
iqu-0.037
ridge-0.036
yles-0.034
umes-0.034
rol-0.033
Token are
Token ID389
Feature activation+3.49e-02

Pos logit contributions
 day+0.064
 good+0.060
,+0.059
 girl+0.058
 time+0.058
Neg logit contributions
iqu-0.077
ridge-0.074
umes-0.070
yles-0.069
angs-0.068
Token welcome
Token ID7062
Feature activation+5.89e-02

Pos logit contributions
 day+0.037
 team+0.035
 girl+0.033
 boy+0.032
 moment+0.031
Neg logit contributions
iqu-0.098
ridge-0.097
vell-0.094
umes-0.093
yles-0.089
Token,
Token ID11
Feature activation+1.75e-02

Pos logit contributions
 day+0.025
 team+0.015
 afternoon+0.015
 end+0.013
 good+0.012
Neg logit contributions
iqu-0.215
ridge-0.208
umes-0.204
angs-0.200
yles-0.198
Token Jill
Token ID22772
Feature activation+9.17e-02

Pos logit contributions
 day+0.029
 good+0.026
 team+0.026
,+0.026
 time+0.026
Neg logit contributions
iqu-0.042
ridge-0.039
umes-0.038
yles-0.038
angs-0.037
Token.
Token ID13
Feature activation+5.27e-02

Pos logit contributions
 team+0.098
 day+0.092
 town+0.088
 afternoon+0.087
 end+0.083
Neg logit contributions
ridge-0.237
iqu-0.230
ker-0.227
ache-0.225
rou-0.222
Token You
Token ID921
Feature activation+2.91e-02

Pos logit contributions
 day+0.093
 good+0.089
 team+0.088
 end+0.087
,+0.087
Neg logit contributions
iqu-0.118
ridge-0.116
umes-0.108
yles-0.107
rol-0.106
Token are
Token ID389
Feature activation+2.85e-02

Pos logit contributions
 day+0.042
 good+0.038
,+0.037
 girl+0.037
 time+0.037
Neg logit contributions
iqu-0.076
ridge-0.075
umes-0.071
yles-0.070
vell-0.070
Token the
Token ID262
Feature activation+5.47e-02

Pos logit contributions
 day+0.030
 team+0.026
 girl+0.026
,+0.025
 time+0.025
Neg logit contributions
iqu-0.084
ridge-0.083
umes-0.080
vell-0.079
yles-0.078
Token best
Token ID1266
Feature activation+5.63e-02

Pos logit contributions
 day+0.043
,+0.039
 good+0.035
 their+0.034
 and+0.034
Neg logit contributions
iqu-0.179
ridge-0.173
umes-0.168
yles-0.165
vell-0.164
Token "
Token ID366
Feature activation+1.12e-02

Pos logit contributions
 day+0.006
 time+0.005
 good+0.005
 end+0.005
 team+0.005
Neg logit contributions
ridge-0.018
iqu-0.018
umes-0.018
yles-0.017
rings-0.017
TokenYou
Token ID1639
Feature activation+2.86e-02

Pos logit contributions
,+0.020
 day+0.019
 good+0.018
 team+0.018
 end+0.018
Neg logit contributions
ridge-0.025
iqu-0.024
yles-0.024
umes-0.023
paste-0.023
Token're
Token ID821
Feature activation+2.91e-02

Pos logit contributions
 day+0.047
 good+0.043
,+0.043
 girl+0.042
 time+0.042
Neg logit contributions
iqu-0.069
ridge-0.068
umes-0.065
yles-0.064
angs-0.063
Token welcome
Token ID7062
Feature activation+5.88e-02

Pos logit contributions
 day+0.028
 team+0.027
 girl+0.026
 boy+0.025
 moment+0.023
Neg logit contributions
iqu-0.083
ridge-0.083
umes-0.080
vell-0.080
yles-0.077
Token,
Token ID11
Feature activation+1.98e-02

Pos logit contributions
 day+0.040
 afternoon+0.033
 team+0.029
 end+0.025
 boy+0.024
Neg logit contributions
iqu-0.191
ridge-0.187
umes-0.183
angs-0.178
rings-0.175
Token Anna
Token ID11735
Feature activation+9.07e-02

Pos logit contributions
 day+0.023
 afternoon+0.020
 team+0.019
 town+0.019
 They+0.018
Neg logit contributions
ridge-0.051
iqu-0.051
yles-0.050
umes-0.050
paste-0.049
Token.
Token ID13
Feature activation+5.33e-02

Pos logit contributions
 day+0.087
 team+0.084
 afternoon+0.082
 end+0.072
 time+0.071
Neg logit contributions
ridge-0.246
ker-0.229
ache-0.228
umes-0.228
iqu-0.227
Token I
Token ID314
Feature activation+2.32e-02

Pos logit contributions
 day+0.098
 good+0.094
 team+0.093
 end+0.093
,+0.092
Neg logit contributions
iqu-0.113
ridge-0.113
umes-0.105
yles-0.104
rol-0.103
Token love
Token ID1842
Feature activation+1.83e-02

Pos logit contributions
 day+0.030
 good+0.027
 girl+0.026
,+0.026
 team+0.026
Neg logit contributions
iqu-0.063
ridge-0.062
umes-0.059
yles-0.058
rol-0.058
Token you
Token ID345
Feature activation+4.51e-02

Pos logit contributions
 day+0.030
 good+0.028
,+0.028
 end+0.027
 girl+0.027
Neg logit contributions
iqu-0.044
ridge-0.043
yles-0.041
umes-0.041
angs-0.039
Token.
Token ID13
Feature activation+4.21e-02

Pos logit contributions
 day+0.055
 girl+0.052
 boy+0.051
 afternoon+0.050
 moment+0.048
Neg logit contributions
ridge-0.114
umes-0.109
rings-0.103
cher-0.103
iqu-0.102
Token love
Token ID1842
Feature activation+1.95e-02

Pos logit contributions
 day+0.052
 team+0.048
 afternoon+0.047
 girl+0.047
 time+0.045
Neg logit contributions
iqu-0.100
ridge-0.099
umes-0.094
cher-0.092
yles-0.091
Token you
Token ID345
Feature activation+4.97e-02

Pos logit contributions
 day+0.033
 good+0.032
 girl+0.031
 end+0.031
,+0.031
Neg logit contributions
iqu-0.045
ridge-0.044
yles-0.042
umes-0.041
rol-0.040
Token,
Token ID11
Feature activation+7.17e-03

Pos logit contributions
 day+0.075
 afternoon+0.069
 moment+0.067
 girl+0.066
 boy+0.065
Neg logit contributions
ridge-0.113
umes-0.106
bar-0.102
iqu-0.102
ker-0.101
Token too
Token ID1165
Feature activation+5.93e-02

Pos logit contributions
 day+0.009
 afternoon+0.008
 good+0.007
 time+0.007
,+0.007
Neg logit contributions
iqu-0.019
ridge-0.019
yles-0.018
umes-0.018
ocol-0.017
Token,
Token ID11
Feature activation+1.65e-02

Pos logit contributions
 day+0.040
 afternoon+0.032
 girl+0.031
 end+0.028
 team+0.028
Neg logit contributions
iqu-0.196
ridge-0.184
umes-0.183
angs-0.181
yles-0.180
Token Anna
Token ID11735
Feature activation+8.99e-02

Pos logit contributions
 day+0.022
 afternoon+0.019
 town+0.018
 good+0.018
 team+0.017
Neg logit contributions
ridge-0.041
iqu-0.040
yles-0.040
umes-0.039
rings-0.038
Token and
Token ID290
Feature activation+3.41e-02

Pos logit contributions
 day+0.085
 team+0.083
 afternoon+0.081
 end+0.074
 good+0.072
Neg logit contributions
ridge-0.239
ache-0.227
paste-0.227
ker-0.225
cher-0.223
Token Ben
Token ID3932
Feature activation+6.33e-02

Pos logit contributions
 day+0.042
 time+0.037
 good+0.037
 end+0.036
,+0.035
Neg logit contributions
ridge-0.095
iqu-0.095
yles-0.091
umes-0.090
rings-0.087
Token.
Token ID13
Feature activation+4.87e-02

Pos logit contributions
 team+0.082
 day+0.080
 afternoon+0.079
 moment+0.073
 end+0.072
Neg logit contributions
ridge-0.137
ache-0.128
cher-0.127
angs-0.124
paste-0.123
Token You
Token ID921
Feature activation+2.66e-02

Pos logit contributions
 day+0.081
 good+0.077
,+0.075
 end+0.075
 team+0.075
Neg logit contributions
iqu-0.112
ridge-0.111
umes-0.104
yles-0.104
rol-0.103
Token are
Token ID389
Feature activation+2.96e-02

Pos logit contributions
 day+0.043
 afternoon+0.040
 girl+0.040
 their+0.038
 boy+0.038
Neg logit contributions
ridge-0.058
vell-0.057
iqu-0.056
umes-0.055
cher-0.055
Token says
Token ID1139
Feature activation-1.81e-02

Pos logit contributions
 day+0.014
 good+0.013
,+0.013
 time+0.012
 end+0.012
Neg logit contributions
iqu-0.029
ridge-0.028
umes-0.027
yles-0.027
rol-0.026
Token,
Token ID11
Feature activation-4.06e-04

Pos logit contributions
iqu+0.067
ridge+0.066
umes+0.064
yles+0.063
angs+0.063
Neg logit contributions
 day-0.006
 end-0.003
 afternoon-0.003
 time-0.003
 girl-0.003
Token "
Token ID366
Feature activation-1.22e-02

Pos logit contributions
iqu+0.001
ridge+0.001
umes+0.001
yles+0.001
rol+0.001
Neg logit contributions
 day-0.000
 good-0.000
,-0.000
 time-0.000
 end-0.000
TokenSure
Token ID19457
Feature activation+2.89e-02

Pos logit contributions
iqu+0.028
ridge+0.027
umes+0.026
yles+0.026
rol+0.026
Neg logit contributions
 day-0.021
,-0.021
 good-0.020
 time-0.020
 end-0.020
Token,
Token ID11
Feature activation+5.85e-03

Pos logit contributions
 day+0.015
 team+0.013
 afternoon+0.012
 They+0.011
 end+0.010
Neg logit contributions
iqu-0.094
ridge-0.094
umes-0.089
angs-0.088
rings-0.088
Token Anna
Token ID11735
Feature activation+8.98e-02

Pos logit contributions
 day+0.007
 good+0.006
 end+0.006
 team+0.006
,+0.006
Neg logit contributions
iqu-0.016
ridge-0.016
umes-0.016
yles-0.015
rol-0.015
Token.
Token ID13
Feature activation+3.08e-02

Pos logit contributions
 team+0.091
 day+0.091
 end+0.087
 afternoon+0.085
 girl+0.083
Neg logit contributions
ridge-0.232
iqu-0.230
ker-0.219
angs-0.217
cher-0.217
Token You
Token ID921
Feature activation+1.75e-02

Pos logit contributions
 day+0.045
 good+0.043
,+0.041
 time+0.041
 end+0.041
Neg logit contributions
iqu-0.080
ridge-0.078
umes-0.075
yles-0.073
rol-0.073
Token can
Token ID460
Feature activation-1.77e-02

Pos logit contributions
 day+0.023
 good+0.021
,+0.021
 time+0.020
 girl+0.020
Neg logit contributions
iqu-0.049
ridge-0.048
umes-0.046
yles-0.045
rol-0.045
Token call
Token ID869
Feature activation-1.40e-02

Pos logit contributions
iqu+0.051
ridge+0.049
umes+0.048
yles+0.048
rol+0.047
Neg logit contributions
 day-0.022
,-0.021
 good-0.021
 and-0.020
 end-0.020
Token grandma
Token ID49890
Feature activation+1.53e-02

Pos logit contributions
iqu+0.042
ridge+0.040
umes+0.039
yles+0.038
rol+0.038
Neg logit contributions
 day-0.016
 good-0.015
,-0.015
 time-0.014
 end-0.014
Token "
Token ID366
Feature activation+1.37e-02

Pos logit contributions
 day+0.008
 good+0.007
 time+0.006
,+0.006
 end+0.006
Neg logit contributions
iqu-0.029
ridge-0.028
umes-0.027
yles-0.027
rol-0.026
TokenYou
Token ID1639
Feature activation+3.27e-02

Pos logit contributions
,+0.024
 day+0.024
 good+0.023
 time+0.023
 team+0.023
Neg logit contributions
iqu-0.029
ridge-0.029
yles-0.028
umes-0.028
rol-0.027
Token're
Token ID821
Feature activation+3.21e-02

Pos logit contributions
 day+0.067
,+0.062
 good+0.062
 girl+0.061
 end+0.061
Neg logit contributions
iqu-0.067
ridge-0.064
umes-0.061
yles-0.060
angs-0.059
Token welcome
Token ID7062
Feature activation+5.84e-02

Pos logit contributions
 day+0.032
 team+0.031
 girl+0.030
 boy+0.029
 afternoon+0.026
Neg logit contributions
iqu-0.091
ridge-0.089
vell-0.088
umes-0.088
rol-0.085
Token,
Token ID11
Feature activation+2.44e-02

Pos logit contributions
 day+0.032
 afternoon+0.024
 team+0.021
 end+0.018
 boy+0.016
Neg logit contributions
iqu-0.204
ridge-0.195
umes-0.193
angs-0.190
rings-0.186
Token Lily
Token ID20037
Feature activation+8.96e-02

Pos logit contributions
 day+0.025
 afternoon+0.020
 good+0.020
,+0.020
 team+0.020
Neg logit contributions
iqu-0.072
ridge-0.070
umes-0.069
yles-0.068
bar-0.066
Token.
Token ID13
Feature activation+5.23e-02

Pos logit contributions
 day+0.096
 team+0.093
 afternoon+0.089
 town+0.086
 time+0.085
Neg logit contributions
ridge-0.231
iqu-0.221
ker-0.219
ache-0.214
ous-0.213
Token I
Token ID314
Feature activation+2.45e-02

Pos logit contributions
 day+0.098
 team+0.096
 end+0.095
 good+0.095
 time+0.094
Neg logit contributions
iqu-0.106
ridge-0.104
umes-0.100
yles-0.097
rol-0.096
Token'm
Token ID1101
Feature activation+1.85e-02

Pos logit contributions
 day+0.036
 good+0.033
 girl+0.032
 team+0.032
,+0.032
Neg logit contributions
iqu-0.062
ridge-0.061
umes-0.058
yles-0.057
rol-0.057
Token proud
Token ID6613
Feature activation+5.74e-02

Pos logit contributions
 day+0.022
 good+0.021
,+0.020
 time+0.019
 and+0.019
Neg logit contributions
iqu-0.055
ridge-0.053
umes-0.051
yles-0.051
angs-0.050
Token of
Token ID286
Feature activation+2.18e-02

Pos logit contributions
 day+0.082
 team+0.077
 good+0.077
 end+0.076
 girl+0.075
Neg logit contributions
iqu-0.141
ridge-0.134
umes-0.130
yles-0.129
angs-0.128
Token "
Token ID366
Feature activation+1.79e-02

Pos logit contributions
 day+0.007
 good+0.006
,+0.006
 time+0.006
 end+0.006
Neg logit contributions
iqu-0.028
ridge-0.027
umes-0.026
yles-0.026
rol-0.026
TokenYou
Token ID1639
Feature activation+3.44e-02

Pos logit contributions
,+0.033
 day+0.032
 time+0.031
 end+0.031
 good+0.031
Neg logit contributions
ridge-0.038
iqu-0.037
yles-0.036
umes-0.035
paste-0.034
Token're
Token ID821
Feature activation+3.80e-02

Pos logit contributions
 day+0.069
 afternoon+0.063
 girl+0.062
,+0.062
 team+0.062
Neg logit contributions
iqu-0.069
ridge-0.067
umes-0.065
vell-0.064
angs-0.062
Token welcome
Token ID7062
Feature activation+6.12e-02

Pos logit contributions
 day+0.041
 team+0.040
 girl+0.039
 boy+0.038
 moment+0.036
Neg logit contributions
iqu-0.101
vell-0.100
ridge-0.099
umes-0.098
cher-0.094
Token,
Token ID11
Feature activation+2.45e-02

Pos logit contributions
 day+0.040
 afternoon+0.033
 team+0.028
 end+0.025
 boy+0.022
Neg logit contributions
iqu-0.203
ridge-0.196
umes-0.192
angs-0.189
ocol-0.184
Token Lily
Token ID20037
Feature activation+8.93e-02

Pos logit contributions
 day+0.030
 afternoon+0.025
,+0.025
 time+0.024
 good+0.024
Neg logit contributions
iqu-0.067
ridge-0.064
umes-0.063
yles-0.063
bar-0.061
Token and
Token ID290
Feature activation+4.24e-02

Pos logit contributions
 team+0.086
 day+0.082
 afternoon+0.078
 town+0.074
 their+0.073
Neg logit contributions
ker-0.228
ridge-0.227
pled-0.221
ache-0.221
paste-0.218
Token Ben
Token ID3932
Feature activation+7.15e-02

Pos logit contributions
 day+0.052
 time+0.047
 good+0.046
 end+0.045
,+0.044
Neg logit contributions
iqu-0.119
ridge-0.117
yles-0.112
umes-0.112
rings-0.107
Token.
Token ID13
Feature activation+5.25e-02

Pos logit contributions
 team+0.099
 day+0.087
 afternoon+0.085
 town+0.082
 end+0.081
Neg logit contributions
ridge-0.153
iqu-0.150
pled-0.149
angs-0.143
ache-0.142
Token I
Token ID314
Feature activation+2.61e-02

Pos logit contributions
 team+0.102
 day+0.102
 end+0.101
 time+0.100
 good+0.097
Neg logit contributions
ridge-0.099
iqu-0.098
umes-0.092
rings-0.090
rol-0.089
Token'm
Token ID1101
Feature activation+2.28e-02

Pos logit contributions
 day+0.036
 team+0.034
 girl+0.032
 afternoon+0.032
 time+0.032
Neg logit contributions
iqu-0.064
ridge-0.063
umes-0.059
cher-0.058
rol-0.058
Token "
Token ID366
Feature activation+1.67e-02

Pos logit contributions
 day+0.006
 good+0.005
,+0.005
 time+0.005
 end+0.005
Neg logit contributions
iqu-0.025
ridge-0.025
umes-0.024
yles-0.024
rol-0.023
TokenYou
Token ID1639
Feature activation+3.54e-02

Pos logit contributions
,+0.031
 day+0.030
 time+0.029
 team+0.029
 end+0.029
Neg logit contributions
iqu-0.034
ridge-0.034
yles-0.033
umes-0.032
rol-0.031
Token're
Token ID821
Feature activation+3.50e-02

Pos logit contributions
 day+0.076
 afternoon+0.070
 girl+0.069
,+0.069
 time+0.068
Neg logit contributions
iqu-0.063
ridge-0.063
umes-0.060
vell-0.059
yles-0.058
Token welcome
Token ID7062
Feature activation+5.72e-02

Pos logit contributions
 day+0.037
 team+0.036
 girl+0.035
 boy+0.034
 moment+0.032
Neg logit contributions
iqu-0.093
ridge-0.092
vell-0.092
umes-0.091
rol-0.087
Token,
Token ID11
Feature activation+2.06e-02

Pos logit contributions
 day+0.033
 afternoon+0.025
 team+0.023
 end+0.019
 boy+0.016
Neg logit contributions
iqu-0.198
ridge-0.190
umes-0.187
angs-0.183
rings-0.180
Token Lily
Token ID20037
Feature activation+8.92e-02

Pos logit contributions
 day+0.019
,+0.016
 good+0.016
 time+0.015
 team+0.015
Neg logit contributions
iqu-0.064
ridge-0.062
umes-0.061
yles-0.060
rol-0.059
Token.
Token ID13
Feature activation+5.15e-02

Pos logit contributions
 team+0.095
 day+0.091
 afternoon+0.087
 town+0.084
 time+0.080
Neg logit contributions
ridge-0.219
ker-0.219
iqu-0.214
ache-0.213
pled-0.209
Token I
Token ID314
Feature activation+2.63e-02

Pos logit contributions
 day+0.096
 end+0.095
 team+0.095
 time+0.094
 good+0.093
Neg logit contributions
iqu-0.102
ridge-0.101
umes-0.095
yles-0.094
rings-0.093
Token love
Token ID1842
Feature activation+1.57e-02

Pos logit contributions
 day+0.037
 team+0.034
 girl+0.033
 time+0.033
 afternoon+0.033
Neg logit contributions
iqu-0.063
ridge-0.063
umes-0.060
yles-0.058
rol-0.058
Token you
Token ID345
Feature activation+4.37e-02

Pos logit contributions
 day+0.025
 good+0.023
,+0.023
 time+0.022
 girl+0.022
Neg logit contributions
iqu-0.041
ridge-0.039
umes-0.038
yles-0.037
rol-0.037
Token too
Token ID1165
Feature activation+5.36e-02

Pos logit contributions
 day+0.063
 afternoon+0.061
 girl+0.060
 time+0.059
 boy+0.059
Neg logit contributions
ridge-0.096
umes-0.091
rings-0.088
cher-0.086
ents-0.084
TokenI
Token ID40
Feature activation+2.80e-02

Pos logit contributions
,+0.010
 day+0.010
 good+0.009
 time+0.009
 end+0.009
Neg logit contributions
iqu-0.013
ridge-0.013
yles-0.012
umes-0.012
paste-0.011
Token love
Token ID1842
Feature activation+1.50e-02

Pos logit contributions
 day+0.037
 good+0.034
,+0.033
 girl+0.033
 time+0.032
Neg logit contributions
iqu-0.077
ridge-0.074
umes-0.071
yles-0.071
angs-0.069
Token you
Token ID345
Feature activation+4.39e-02

Pos logit contributions
 day+0.023
 good+0.022
,+0.022
 time+0.021
 girl+0.021
Neg logit contributions
iqu-0.039
ridge-0.038
umes-0.036
yles-0.036
rol-0.035
Token too
Token ID1165
Feature activation+5.42e-02

Pos logit contributions
 day+0.068
 girl+0.062
 afternoon+0.061
 boy+0.060
 moment+0.059
Neg logit contributions
ridge-0.100
iqu-0.096
umes-0.095
bar-0.092
angs-0.090
Token,
Token ID11
Feature activation+8.80e-03

Pos logit contributions
 day+0.031
 girl+0.027
 afternoon+0.024
 boy+0.023
 team+0.021
Neg logit contributions
iqu-0.182
angs-0.171
umes-0.170
ridge-0.169
ocol-0.166
Token Anna
Token ID11735
Feature activation+8.90e-02

Pos logit contributions
 day+0.011
 afternoon+0.009
 town+0.008
 good+0.008
,+0.008
Neg logit contributions
ridge-0.023
iqu-0.023
yles-0.023
umes-0.022
paste-0.021
Token and
Token ID290
Feature activation+3.22e-02

Pos logit contributions
 day+0.081
 team+0.076
 afternoon+0.076
 good+0.068
 their+0.066
Neg logit contributions
ridge-0.243
paste-0.231
umes-0.231
ache-0.230
ker-0.229
Token Ben
Token ID3932
Feature activation+6.22e-02

Pos logit contributions
 day+0.036
 good+0.032
,+0.031
 time+0.030
 end+0.030
Neg logit contributions
iqu-0.096
ridge-0.094
umes-0.091
yles-0.091
angs-0.088
Token.
Token ID13
Feature activation+4.28e-02

Pos logit contributions
 day+0.081
 team+0.080
 afternoon+0.080
 town+0.072
 moment+0.072
Neg logit contributions
ridge-0.135
ache-0.127
angs-0.125
paste-0.125
iqu-0.124
Token You
Token ID921
Feature activation+1.83e-02

Pos logit contributions
 day+0.068
 good+0.064
,+0.062
 end+0.062
 team+0.062
Neg logit contributions
iqu-0.103
ridge-0.101
umes-0.096
yles-0.095
ocol-0.093
Token are
Token ID389
Feature activation+2.23e-02

Pos logit contributions
 day+0.026
 good+0.023
,+0.023
 girl+0.023
 time+0.022
Neg logit contributions
iqu-0.048
ridge-0.047
umes-0.045
yles-0.044
angs-0.044
Token "
Token ID366
Feature activation+1.27e-02

Pos logit contributions
 day+0.002
 good+0.001
 time+0.001
 end+0.001
 team+0.001
Neg logit contributions
iqu-0.005
ridge-0.005
umes-0.005
yles-0.005
rol-0.005
TokenThat
Token ID2504
Feature activation+4.08e-02

Pos logit contributions
,+0.024
 day+0.023
 time+0.023
 end+0.022
 good+0.022
Neg logit contributions
iqu-0.026
ridge-0.026
yles-0.025
umes-0.024
paste-0.024
Token's
Token ID338
Feature activation+2.95e-02

Pos logit contributions
 day+0.043
 good+0.042
 team+0.042
 their+0.041
 time+0.040
Neg logit contributions
iqu-0.109
ridge-0.107
bar-0.099
ous-0.098
umes-0.098
Token good
Token ID922
Feature activation+4.81e-02

Pos logit contributions
 team+0.033
 girl+0.029
 day+0.028
 boy+0.027
 town+0.026
Neg logit contributions
iqu-0.079
vell-0.074
umes-0.072
ridge-0.071
yles-0.068
Token,
Token ID11
Feature activation+1.08e-02

Pos logit contributions
 day+0.035
 afternoon+0.033
 end+0.031
 time+0.030
 town+0.029
Neg logit contributions
iqu-0.139
angs-0.135
ous-0.133
ridge-0.132
paste-0.131
Token Anna
Token ID11735
Feature activation+8.90e-02

Pos logit contributions
 day+0.014
 afternoon+0.011
 town+0.010
 team+0.010
 end+0.010
Neg logit contributions
iqu-0.028
ridge-0.028
yles-0.028
umes-0.027
paste-0.026
Token and
Token ID290
Feature activation+3.94e-02

Pos logit contributions
 team+0.091
 day+0.085
 afternoon+0.082
 town+0.076
 end+0.072
Neg logit contributions
ridge-0.228
ker-0.223
paste-0.223
pled-0.221
ache-0.220
Token Ben
Token ID3932
Feature activation+6.77e-02

Pos logit contributions
 day+0.051
 time+0.045
 end+0.044
 good+0.043
,+0.043
Neg logit contributions
iqu-0.107
ridge-0.105
yles-0.102
umes-0.101
rings-0.097
Token.
Token ID13
Feature activation+4.39e-02

Pos logit contributions
 team+0.097
 afternoon+0.089
 day+0.089
 town+0.082
 sunny+0.082
Neg logit contributions
pled-0.136
ridge-0.135
iqu-0.135
ache-0.134
angs-0.127
Token I
Token ID314
Feature activation+2.77e-02

Pos logit contributions
 day+0.080
 good+0.076
 time+0.076
 end+0.076
 team+0.074
Neg logit contributions
iqu-0.097
ridge-0.094
umes-0.089
yles-0.088
angs-0.086
Token am
Token ID716
Feature activation+2.24e-02

Pos logit contributions
 day+0.039
 afternoon+0.035
 time+0.034
 team+0.034
 girl+0.033
Neg logit contributions
iqu-0.070
ridge-0.068
umes-0.065
yles-0.063
cher-0.063
Token"
Token ID1
Feature activation-1.18e-02

Pos logit contributions
 day+0.015
 time+0.014
 good+0.014
 team+0.014
,+0.013
Neg logit contributions
iqu-0.035
ridge-0.035
umes-0.035
yles-0.034
cher-0.034
TokenYou
Token ID1639
Feature activation+2.29e-02

Pos logit contributions
iqu+0.028
ridge+0.028
yles+0.027
umes+0.027
rol+0.026
Neg logit contributions
 day-0.019
,-0.019
 good-0.018
 time-0.017
 team-0.017
Token are
Token ID389
Feature activation+2.85e-02

Pos logit contributions
 day+0.037
 good+0.035
,+0.034
 girl+0.033
 time+0.033
Neg logit contributions
iqu-0.057
ridge-0.055
umes-0.053
yles-0.052
rol-0.051
Token welcome
Token ID7062
Feature activation+5.63e-02

Pos logit contributions
 day+0.027
 girl+0.024
 team+0.024
 moment+0.023
 boy+0.023
Neg logit contributions
iqu-0.083
ridge-0.083
umes-0.082
vell-0.081
yles-0.078
Token,
Token ID11
Feature activation+6.16e-03

Pos logit contributions
 day+0.036
 afternoon+0.028
 end+0.027
 team+0.023
 good+0.022
Neg logit contributions
iqu-0.188
ridge-0.187
umes-0.181
angs-0.176
yles-0.175
Token Anna
Token ID11735
Feature activation+8.90e-02

Pos logit contributions
 day+0.007
 afternoon+0.006
,+0.006
 town+0.006
 moment+0.006
Neg logit contributions
iqu-0.016
ridge-0.016
yles-0.016
umes-0.016
ocol-0.015
Token and
Token ID290
Feature activation+3.58e-02

Pos logit contributions
 day+0.066
 team+0.065
 afternoon+0.057
 good+0.057
 end+0.055
Neg logit contributions
ridge-0.263
iqu-0.254
ker-0.251
ache-0.250
umes-0.249
Token Ben
Token ID3932
Feature activation+6.38e-02

Pos logit contributions
 day+0.044
 good+0.039
 time+0.039
,+0.038
 end+0.037
Neg logit contributions
iqu-0.101
ridge-0.099
yles-0.096
umes-0.094
angs-0.092
Token!"
Token ID2474
Feature activation+2.81e-02

Pos logit contributions
 team+0.083
 day+0.074
 afternoon+0.071
 their+0.070
 end+0.069
Neg logit contributions
ridge-0.147
ache-0.144
iqu-0.142
ker-0.140
pled-0.136
Token Mom
Token ID11254
Feature activation-1.17e-03

Pos logit contributions
 day+0.041
 afternoon+0.040
 team+0.039
 time+0.037
 whole+0.037
Neg logit contributions
iqu-0.058
yles-0.056
rings-0.056
ents-0.054
ridge-0.052
Token says
Token ID1139
Feature activation+5.77e-03

Pos logit contributions
iqu+0.003
ridge+0.003
umes+0.003
yles+0.003
rol+0.003
Neg logit contributions
 day-0.002
 good-0.002
,-0.001
 time-0.001
 team-0.001
Token "
Token ID366
Feature activation+1.63e-02

Pos logit contributions
 day+0.009
 time+0.008
 good+0.008
 end+0.007
 girl+0.007
Neg logit contributions
iqu-0.027
ridge-0.027
umes-0.026
yles-0.026
rol-0.025
TokenYou
Token ID1639
Feature activation+3.78e-02

Pos logit contributions
 day+0.029
,+0.029
 good+0.027
 time+0.027
 end+0.027
Neg logit contributions
ridge-0.035
iqu-0.035
yles-0.034
umes-0.033
rol-0.033
Token're
Token ID821
Feature activation+3.76e-02

Pos logit contributions
 day+0.076
 girl+0.070
 afternoon+0.069
,+0.069
 team+0.068
Neg logit contributions
iqu-0.073
ridge-0.073
umes-0.070
vell-0.069
yles-0.068
Token welcome
Token ID7062
Feature activation+6.28e-02

Pos logit contributions
 day+0.040
 girl+0.039
 team+0.039
 boy+0.038
 moment+0.035
Neg logit contributions
ridge-0.100
iqu-0.099
vell-0.097
umes-0.097
rol-0.093
Token,
Token ID11
Feature activation+2.29e-02

Pos logit contributions
 day+0.042
 afternoon+0.034
 team+0.030
 end+0.026
 boy+0.024
Neg logit contributions
iqu-0.205
ridge-0.199
umes-0.197
angs-0.191
ocol-0.188
Token Sara
Token ID24799
Feature activation+8.88e-02

Pos logit contributions
 day+0.033
 afternoon+0.028
 team+0.028
,+0.027
 good+0.027
Neg logit contributions
iqu-0.058
ridge-0.056
yles-0.055
umes-0.055
ocol-0.053
Token and
Token ID290
Feature activation+4.58e-02

Pos logit contributions
 day+0.075
 team+0.073
 afternoon+0.069
 end+0.067
 time+0.065
Neg logit contributions
ridge-0.246
ker-0.234
ache-0.234
paste-0.227
umes-0.227
Token Ben
Token ID3932
Feature activation+7.36e-02

Pos logit contributions
 day+0.052
 good+0.046
 time+0.045
,+0.045
 end+0.043
Neg logit contributions
iqu-0.135
ridge-0.133
umes-0.127
yles-0.127
angs-0.123
Token.
Token ID13
Feature activation+5.25e-02

Pos logit contributions
 team+0.095
 day+0.089
 afternoon+0.087
 town+0.084
 end+0.083
Neg logit contributions
ridge-0.164
iqu-0.149
pled-0.148
ache-0.146
rou-0.146
Token You
Token ID921
Feature activation+2.97e-02

Pos logit contributions
 day+0.100
 end+0.097
 team+0.097
 good+0.096
 time+0.096
Neg logit contributions
ridge-0.105
iqu-0.104
umes-0.097
rol-0.096
yles-0.095
Token are
Token ID389
Feature activation+3.06e-02

Pos logit contributions
 day+0.046
 girl+0.041
 afternoon+0.041
,+0.040
 good+0.040
Neg logit contributions
ridge-0.072
iqu-0.071
umes-0.068
vell-0.067
yles-0.066
Token see
Token ID766
Feature activation+4.18e-03

Pos logit contributions
 day+0.006
 good+0.005
,+0.005
 time+0.005
 end+0.005
Neg logit contributions
iqu-0.017
ridge-0.016
umes-0.016
yles-0.016
rol-0.016
Token what
Token ID644
Feature activation-3.08e-03

Pos logit contributions
 day+0.006
,+0.005
 good+0.005
 and+0.005
 time+0.005
Neg logit contributions
iqu-0.012
ridge-0.011
umes-0.011
yles-0.011
rol-0.011
Token we
Token ID356
Feature activation-4.15e-03

Pos logit contributions
yles+0.007
iqu+0.007
ridge+0.007
umes+0.007
rings+0.006
Neg logit contributions
 end-0.004
 day-0.004
 girl-0.004
 afternoon-0.004
,-0.004
Token can
Token ID460
Feature activation-2.05e-02

Pos logit contributions
iqu+0.012
ridge+0.011
umes+0.011
yles+0.011
rol+0.011
Neg logit contributions
 day-0.005
 good-0.005
,-0.005
 girl-0.005
 time-0.005
Token come
Token ID1282
Feature activation+1.13e-02

Pos logit contributions
iqu+0.061
ridge+0.059
umes+0.057
yles+0.057
rol+0.056
Neg logit contributions
 day-0.023
 good-0.021
,-0.021
 end-0.020
 time-0.020
Token up
Token ID510
Feature activation+8.87e-02

Pos logit contributions
 day+0.009
 good+0.008
 girl+0.008
 end+0.008
 time+0.008
Neg logit contributions
iqu-0.036
ridge-0.035
umes-0.035
yles-0.034
angs-0.034
Token with
Token ID351
Feature activation+3.15e-02

Pos logit contributions
 end+0.082
 day+0.081
 girl+0.079
 good+0.072
 boy+0.070
Neg logit contributions
iqu-0.242
umes-0.242
yles-0.241
ridge-0.239
vell-0.228
Token."
Token ID526
Feature activation+1.09e-02

Pos logit contributions
 day+0.039
 girl+0.036
 good+0.035
 boy+0.034
 end+0.034
Neg logit contributions
iqu-0.083
ridge-0.083
yles-0.082
umes-0.081
ocol-0.076
Token
Token ID198
Feature activation+1.38e-03

Pos logit contributions
 end+0.012
 team+0.011
 day+0.011
 good+0.010
 afternoon+0.009
Neg logit contributions
ridge-0.030
rol-0.029
iqu-0.029
yles-0.028
umes-0.028
Token
Token ID198
Feature activation-7.90e-03

Pos logit contributions
 team+0.001
 end+0.001
 day+0.001
 girl+0.001
 good+0.001
Neg logit contributions
iqu-0.004
ridge-0.004
rol-0.004
yles-0.004
umes-0.003
TokenThey
Token ID2990
Feature activation+3.91e-02

Pos logit contributions
iqu+0.014
umes+0.014
rol+0.014
ridge+0.014
yles+0.013
Neg logit contributions
 day-0.015
 team-0.014
 end-0.013
 good-0.013
,-0.013
Token "
Token ID366
Feature activation+8.77e-03

Pos logit contributions
 day+0.006
,+0.005
 good+0.005
 and+0.004
 girl+0.004
Neg logit contributions
iqu-0.028
ridge-0.027
umes-0.026
yles-0.026
angs-0.026
TokenYou
Token ID1639
Feature activation+3.09e-02

Pos logit contributions
,+0.016
 day+0.016
 time+0.015
 end+0.015
 good+0.015
Neg logit contributions
iqu-0.019
ridge-0.019
yles-0.018
umes-0.018
rol-0.018
Token're
Token ID821
Feature activation+2.99e-02

Pos logit contributions
 day+0.061
 afternoon+0.056
 time+0.055
 girl+0.054
 end+0.054
Neg logit contributions
iqu-0.063
ridge-0.062
umes-0.059
vell-0.059
yles-0.057
Token welcome
Token ID7062
Feature activation+6.03e-02

Pos logit contributions
 day+0.032
 team+0.029
 girl+0.029
 boy+0.027
 afternoon+0.027
Neg logit contributions
iqu-0.083
ridge-0.082
vell-0.080
umes-0.080
rol-0.078
Token,
Token ID11
Feature activation+2.11e-02

Pos logit contributions
 day+0.036
 afternoon+0.028
 end+0.025
 team+0.024
 time+0.019
Neg logit contributions
iqu-0.206
ridge-0.200
umes-0.196
angs-0.190
rol-0.189
Token Lily
Token ID20037
Feature activation+8.86e-02

Pos logit contributions
 day+0.021
,+0.017
 end+0.017
 time+0.017
 good+0.017
Neg logit contributions
iqu-0.062
ridge-0.060
umes-0.059
yles-0.059
rol-0.057
Token.
Token ID13
Feature activation+4.97e-02

Pos logit contributions
 team+0.083
 day+0.081
 afternoon+0.073
 town+0.072
 time+0.072
Neg logit contributions
ridge-0.238
ker-0.231
iqu-0.229
ache-0.225
ous-0.224
Token We
Token ID775
Feature activation+2.88e-02

Pos logit contributions
 day+0.091
 end+0.089
 time+0.089
 good+0.087
 team+0.086
Neg logit contributions
iqu-0.105
ridge-0.104
rol-0.097
umes-0.096
yles-0.096
Token're
Token ID821
Feature activation+2.74e-02

Pos logit contributions
 day+0.039
 girl+0.036
 team+0.036
 new+0.036
 boy+0.036
Neg logit contributions
iqu-0.070
ridge-0.069
umes-0.067
rol-0.066
yles-0.064
Token glad
Token ID9675
Feature activation+4.24e-02

Pos logit contributions
 day+0.037
 end+0.033
 team+0.033
 time+0.033
 girl+0.033
Neg logit contributions
iqu-0.073
ridge-0.071
umes-0.069
yles-0.069
rol-0.068
Token you
Token ID345
Feature activation+2.23e-02

Pos logit contributions
 day+0.077
 end+0.073
 good+0.071
 afternoon+0.071
 time+0.071
Neg logit contributions
iqu-0.091
ridge-0.088
yles-0.084
umes-0.082
rings-0.080
Token.
Token ID13
Feature activation+4.56e-02

Pos logit contributions
 day+0.007
 time+0.006
 afternoon+0.006
 end+0.006
 team+0.006
Neg logit contributions
iqu-0.029
ridge-0.028
umes-0.027
angs-0.027
rings-0.027
Token "
Token ID366
Feature activation+1.69e-02

Pos logit contributions
 day+0.064
 afternoon+0.060
 end+0.058
 time+0.056
 team+0.054
Neg logit contributions
iqu-0.110
ridge-0.109
yles-0.104
paste-0.104
umes-0.104
TokenThank
Token ID10449
Feature activation+1.87e-02

Pos logit contributions
 day+0.028
,+0.027
 good+0.027
 time+0.026
 end+0.026
Neg logit contributions
iqu-0.040
ridge-0.039
yles-0.037
umes-0.037
rol-0.036
Token you
Token ID345
Feature activation+6.89e-02

Pos logit contributions
 day+0.036
 time+0.034
 good+0.034
 team+0.033
,+0.033
Neg logit contributions
ridge-0.037
iqu-0.037
yles-0.034
ocol-0.034
umes-0.034
Token,
Token ID11
Feature activation+2.30e-02

Pos logit contributions
 day+0.078
 afternoon+0.074
 time+0.073
 moment+0.068
 team+0.066
Neg logit contributions
ridge-0.180
ous-0.168
angs-0.165
rol-0.165
iqu-0.165
Token Anna
Token ID11735
Feature activation+8.85e-02

Pos logit contributions
 day+0.028
 good+0.023
,+0.023
 team+0.023
 afternoon+0.023
Neg logit contributions
iqu-0.063
ridge-0.063
yles-0.059
umes-0.059
ocol-0.058
Token.
Token ID13
Feature activation+4.94e-02

Pos logit contributions
 day+0.116
 team+0.116
 afternoon+0.113
 time+0.111
 moment+0.104
Neg logit contributions
ridge-0.214
cher-0.195
ache-0.190
paste-0.190
ents-0.184
Token You
Token ID921
Feature activation+2.52e-02

Pos logit contributions
 day+0.090
 good+0.085
,+0.084
 time+0.082
 end+0.082
Neg logit contributions
iqu-0.111
ridge-0.108
umes-0.102
yles-0.101
rol-0.101
Token are
Token ID389
Feature activation+2.70e-02

Pos logit contributions
 day+0.035
,+0.031
 good+0.031
 girl+0.031
 time+0.030
Neg logit contributions
iqu-0.068
ridge-0.066
umes-0.063
yles-0.062
angs-0.062
Token a
Token ID257
Feature activation+5.53e-02

Pos logit contributions
 day+0.028
 team+0.024
 girl+0.024
 time+0.023
,+0.023
Neg logit contributions
iqu-0.079
ridge-0.079
umes-0.075
vell-0.074
yles-0.074
Token good
Token ID922
Feature activation+4.86e-02

Pos logit contributions
 day+0.037
 They+0.035
 afternoon+0.034
 time+0.030
 Wednesday+0.027
Neg logit contributions
ridge-0.161
vell-0.161
iqu-0.159
ache-0.156
cher-0.152
Token,
Token ID11
Feature activation+7.92e-03

Pos logit contributions
iqu+0.016
ridge+0.016
umes+0.016
paste+0.015
ous+0.015
Neg logit contributions
 day-0.003
 afternoon-0.002
 end-0.002
 team-0.002
 town-0.002
Token "
Token ID366
Feature activation+5.44e-03

Pos logit contributions
 day+0.007
 good+0.006
,+0.006
 time+0.005
 end+0.005
Neg logit contributions
iqu-0.026
ridge-0.025
umes-0.024
yles-0.024
rol-0.024
TokenThank
Token ID10449
Feature activation+1.69e-02

Pos logit contributions
,+0.008
 day+0.008
 good+0.008
 end+0.008
 team+0.008
Neg logit contributions
iqu-0.013
ridge-0.013
yles-0.013
umes-0.012
rol-0.012
Token you
Token ID345
Feature activation+6.67e-02

Pos logit contributions
 day+0.033
 boy+0.031
 team+0.031
 good+0.031
 girl+0.030
Neg logit contributions
iqu-0.033
ridge-0.032
umes-0.031
yles-0.030
ocol-0.030
Token,
Token ID11
Feature activation+1.97e-02

Pos logit contributions
 day+0.053
 afternoon+0.046
 time+0.044
 town+0.043
 boy+0.041
Neg logit contributions
ridge-0.198
iqu-0.197
umes-0.196
ous-0.193
ker-0.187
Token Lily
Token ID20037
Feature activation+8.84e-02

Pos logit contributions
 day+0.018
 good+0.016
,+0.016
 time+0.015
 end+0.015
Neg logit contributions
iqu-0.063
ridge-0.061
umes-0.059
yles-0.058
rol-0.058
Token.
Token ID13
Feature activation+3.88e-02

Pos logit contributions
 team+0.093
 day+0.089
 town+0.085
 time+0.085
 afternoon+0.082
Neg logit contributions
ridge-0.232
iqu-0.224
rou-0.217
ous-0.215
ker-0.213
Token You
Token ID921
Feature activation+2.64e-02

Pos logit contributions
 day+0.062
 good+0.059
,+0.058
 time+0.057
 end+0.057
Neg logit contributions
iqu-0.096
ridge-0.093
umes-0.089
yles-0.088
rol-0.087
Token are
Token ID389
Feature activation+2.88e-02

Pos logit contributions
 day+0.043
 girl+0.039
,+0.039
 good+0.039
 team+0.038
Neg logit contributions
iqu-0.063
ridge-0.062
umes-0.059
vell-0.058
yles-0.058
Token such
Token ID884
Feature activation+4.39e-02

Pos logit contributions
 day+0.035
 team+0.032
 girl+0.031
 boy+0.030
 time+0.030
Neg logit contributions
iqu-0.077
ridge-0.077
umes-0.075
vell-0.073
yles-0.073
Token a
Token ID257
Feature activation+3.96e-02

Pos logit contributions
 day+0.034
,+0.030
 end+0.028
 girl+0.028
 team+0.027
Neg logit contributions
iqu-0.143
ridge-0.137
umes-0.135
yles-0.132
vell-0.132
Token,
Token ID11
Feature activation+1.08e-02

Pos logit contributions
 day+0.004
 afternoon+0.003
 place+0.003
 end+0.003
 girl+0.003
Neg logit contributions
ridge-0.015
iqu-0.015
paste-0.015
umes-0.015
rings-0.014
Token "
Token ID366
Feature activation+1.66e-02

Pos logit contributions
 day+0.011
 good+0.009
 time+0.009
 end+0.009
 girl+0.009
Neg logit contributions
iqu-0.033
ridge-0.033
umes-0.032
yles-0.031
rol-0.030
TokenThank
Token ID10449
Feature activation+2.37e-02

Pos logit contributions
,+0.031
 day+0.029
 time+0.029
 end+0.028
 good+0.028
Neg logit contributions
ridge-0.034
yles-0.034
iqu-0.034
umes-0.032
cher-0.031
Token you
Token ID345
Feature activation+7.09e-02

Pos logit contributions
 day+0.050
 place+0.047
 afternoon+0.046
 boy+0.046
 team+0.046
Neg logit contributions
iqu-0.039
ridge-0.039
ocol-0.038
umes-0.037
yles-0.036
Token,
Token ID11
Feature activation+1.95e-02

Pos logit contributions
 day+0.060
 afternoon+0.059
 moment+0.046
 town+0.046
 time+0.045
Neg logit contributions
umes-0.195
ridge-0.194
iqu-0.192
ker-0.191
ous-0.190
Token Lily
Token ID20037
Feature activation+8.80e-02

Pos logit contributions
 day+0.021
 good+0.018
 time+0.017
,+0.017
 afternoon+0.017
Neg logit contributions
iqu-0.057
ridge-0.055
umes-0.054
yles-0.054
rol-0.052
Token,
Token ID11
Feature activation+2.15e-02

Pos logit contributions
 day+0.098
 afternoon+0.098
 team+0.096
 town+0.090
 sunny+0.085
Neg logit contributions
ker-0.207
ridge-0.202
pled-0.198
ous-0.195
ache-0.195
Token thank
Token ID5875
Feature activation-2.73e-04

Pos logit contributions
 day+0.028
 girl+0.025
 time+0.025
,+0.025
 good+0.025
Neg logit contributions
iqu-0.056
ridge-0.055
umes-0.054
yles-0.053
ocol-0.051
Token you
Token ID345
Feature activation+5.71e-02

Pos logit contributions
ridge+0.000
iqu+0.000
umes+0.000
yles+0.000
ocol+0.000
Neg logit contributions
 day-0.001
 good-0.001
 boy-0.001
 girl-0.001
 afternoon-0.001
Token,
Token ID11
Feature activation+2.13e-02

Pos logit contributions
 day+0.054
 afternoon+0.053
 time+0.046
 town+0.045
 girl+0.045
Neg logit contributions
umes-0.149
iqu-0.148
ridge-0.147
ous-0.146
cher-0.145
Token Ben
Token ID3932
Feature activation+6.99e-02

Pos logit contributions
 day+0.025
 afternoon+0.021
 time+0.021
,+0.021
 good+0.021
Neg logit contributions
iqu-0.059
ridge-0.057
umes-0.057
yles-0.056
rol-0.054
Token smiled
Token ID13541
Feature activation+5.05e-02

Pos logit contributions
 day+0.037
 time+0.036
 end+0.035
 good+0.034
 team+0.032
Neg logit contributions
iqu-0.072
ridge-0.070
yles-0.067
umes-0.067
rol-0.066
Token.
Token ID13
Feature activation+2.06e-02

Pos logit contributions
 day+0.050
 time+0.045
 afternoon+0.042
 end+0.041
 team+0.037
Neg logit contributions
iqu-0.148
ous-0.139
vell-0.137
umes-0.137
ridge-0.136
Token They
Token ID1119
Feature activation+4.51e-02

Pos logit contributions
 day+0.020
 end+0.019
 girl+0.016
 team+0.016
 time+0.016
Neg logit contributions
umes-0.061
yles-0.061
iqu-0.061
rol-0.060
ridge-0.060
Token had
Token ID550
Feature activation+1.15e-02

Pos logit contributions
 day+0.062
 good+0.057
 time+0.056
 girl+0.055
,+0.054
Neg logit contributions
iqu-0.120
ridge-0.116
umes-0.111
rol-0.110
yles-0.109
Token come
Token ID1282
Feature activation+1.14e-02

Pos logit contributions
 day+0.015
 girl+0.014
 end+0.014
 team+0.013
,+0.013
Neg logit contributions
ridge-0.030
iqu-0.030
umes-0.028
rol-0.028
yles-0.028
Token up
Token ID510
Feature activation+8.77e-02

Pos logit contributions
 day+0.011
 good+0.009
 time+0.009
 girl+0.009
 end+0.009
Neg logit contributions
iqu-0.035
ridge-0.034
umes-0.034
yles-0.033
rol-0.033
Token with
Token ID351
Feature activation+2.73e-02

Pos logit contributions
 day+0.089
 girl+0.088
 end+0.085
 boy+0.082
 good+0.080
Neg logit contributions
iqu-0.244
ridge-0.233
yles-0.229
umes-0.227
rol-0.226
Token a
Token ID257
Feature activation+2.67e-02

Pos logit contributions
 day+0.023
 girl+0.020
 team+0.019
 boy+0.018
 end+0.018
Neg logit contributions
iqu-0.086
ridge-0.083
yles-0.081
umes-0.081
rol-0.078
Token creative
Token ID7325
Feature activation+2.95e-02

Pos logit contributions
 day+0.026
 They+0.024
 end+0.023
,+0.023
 their+0.022
Neg logit contributions
iqu-0.077
ridge-0.073
umes-0.073
vell-0.070
bar-0.069
Token way
Token ID835
Feature activation+3.71e-02

Pos logit contributions
 They+0.028
+0.028
 their+0.027
 day+0.025
 many+0.025
Neg logit contributions
iqu-0.072
bar-0.069
ker-0.068
ridge-0.067
ocol-0.066
Token to
Token ID284
Feature activation+5.60e-04

Pos logit contributions
 day+0.040
 good+0.038
 girl+0.036
 end+0.035
 team+0.035
Neg logit contributions
iqu-0.110
ridge-0.105
umes-0.103
rol-0.102
yles-0.101
Token said
Token ID531
Feature activation-9.56e-04

Pos logit contributions
 day+0.057
 end+0.052
 time+0.051
 afternoon+0.050
 good+0.050
Neg logit contributions
iqu-0.075
ridge-0.072
umes-0.069
paste-0.067
cher-0.066
Token,
Token ID11
Feature activation+1.57e-02

Pos logit contributions
ridge+0.003
iqu+0.003
paste+0.003
bar+0.003
cher+0.003
Neg logit contributions
 day-0.001
 afternoon-0.001
 time-0.000
 girl-0.000
 place-0.000
Token "
Token ID366
Feature activation+1.43e-02

Pos logit contributions
 day+0.016
 time+0.014
 good+0.014
 end+0.014
 girl+0.013
Neg logit contributions
ridge-0.047
iqu-0.046
umes-0.045
yles-0.044
bar-0.043
TokenOK
Token ID11380
Feature activation+6.08e-02

Pos logit contributions
,+0.024
 day+0.023
 time+0.023
 end+0.022
 good+0.021
Neg logit contributions
ridge-0.032
iqu-0.031
yles-0.030
cher-0.030
bar-0.030
Token,
Token ID11
Feature activation+1.49e-02

Pos logit contributions
 afternoon+0.059
 day+0.057
 town+0.048
 country+0.045
 their+0.043
Neg logit contributions
iqu-0.149
ridge-0.148
bar-0.144
ker-0.143
umes-0.142
Token Lily
Token ID20037
Feature activation+8.76e-02

Pos logit contributions
 day+0.016
 good+0.014
,+0.014
 time+0.014
 end+0.013
Neg logit contributions
iqu-0.045
ridge-0.044
umes-0.042
yles-0.042
rol-0.041
Token.
Token ID13
Feature activation+5.39e-02

Pos logit contributions
 afternoon+0.093
 day+0.088
 town+0.083
 their+0.080
 team+0.077
Neg logit contributions
iqu-0.207
ker-0.207
ridge-0.206
ache-0.202
ous-0.202
Token I
Token ID314
Feature activation+3.20e-02

Pos logit contributions
 day+0.107
 time+0.105
 and+0.100
 good+0.100
 end+0.100
Neg logit contributions
ridge-0.103
iqu-0.100
bar-0.094
cher-0.093
rings-0.091
Token will
Token ID481
Feature activation+2.34e-02

Pos logit contributions
 day+0.048
 girl+0.044
 afternoon+0.044
 their+0.043
 boy+0.041
Neg logit contributions
iqu-0.076
ridge-0.073
umes-0.071
angs-0.068
cher-0.068
Token do
Token ID466
Feature activation+4.78e-02

Pos logit contributions
 day+0.026
 good+0.022
 girl+0.022
,+0.021
 boy+0.021
Neg logit contributions
iqu-0.070
ridge-0.067
umes-0.065
yles-0.064
angs-0.063
Token that
Token ID326
Feature activation+2.50e-02

Pos logit contributions
 day+0.067
 girl+0.061
 team+0.059
 boy+0.058
 time+0.058
Neg logit contributions
iqu-0.129
ridge-0.125
yles-0.120
umes-0.119
angs-0.115
Token delicious
Token ID12625
Feature activation-1.20e-02

Pos logit contributions
iqu+0.047
ridge+0.045
umes+0.043
yles+0.043
rol+0.043
Neg logit contributions
 day-0.019
,-0.017
 good-0.017
 end-0.016
 team-0.016
Token vegetables
Token ID13701
Feature activation-4.24e-03

Pos logit contributions
iqu+0.032
ridge+0.031
umes+0.029
yles+0.029
ocol+0.029
Neg logit contributions
 day-0.016
 end-0.015
 good-0.015
 team-0.015
 girl-0.014
Token and
Token ID290
Feature activation-3.23e-02

Pos logit contributions
iqu+0.016
ridge+0.016
yles+0.015
umes+0.015
rol+0.015
Neg logit contributions
 day-0.002
,-0.001
 good-0.001
 and-0.001
 time-0.001
Token bright
Token ID6016
Feature activation+2.54e-03

Pos logit contributions
iqu+0.087
ridge+0.084
umes+0.080
yles+0.080
rol+0.079
Neg logit contributions
 day-0.046
 good-0.043
,-0.043
 time-0.041
 end-0.041
Token-
Token ID12
Feature activation-6.67e-02

Pos logit contributions
 day+0.003
 good+0.003
 end+0.003
 team+0.003
 girl+0.003
Neg logit contributions
iqu-0.007
ridge-0.007
yles-0.006
umes-0.006
ocol-0.006
Tokencol
Token ID4033
Feature activation-1.28e-01

Pos logit contributions
iqu+0.131
ridge+0.119
yles+0.113
umes+0.113
angs+0.108
Neg logit contributions
 day-0.153
 good-0.143
,-0.141
 and-0.139
 girl-0.137
Tokenoured
Token ID8167
Feature activation-7.85e-03

Pos logit contributions
cher+0.182
iqu+0.155
 outage+0.155
angs+0.154
paste+0.152
Neg logit contributions
,-0.297
 and-0.285
 happily-0.269
 loves-0.263
 They-0.261
Token flowers
Token ID12734
Feature activation-6.26e-03

Pos logit contributions
iqu+0.019
ridge+0.019
umes+0.018
yles+0.018
ocol+0.018
Neg logit contributions
 day-0.012
 end-0.011
 good-0.011
 time-0.011
,-0.011
Token.
Token ID13
Feature activation-1.24e-02

Pos logit contributions
iqu+0.024
ridge+0.023
yles+0.023
umes+0.023
rol+0.023
Neg logit contributions
 day-0.002
,-0.002
 and-0.001
 good-0.001
 time-0.001
Token
Token ID220
Feature activation-1.59e-02

Pos logit contributions
ridge+0.035
iqu+0.035
bar+0.034
yles+0.034
rol+0.033
Neg logit contributions
 good-0.014
 day-0.013
 end-0.012
 girl-0.012
,-0.012
Token
Token ID198
Feature activation-1.15e-02

Pos logit contributions
iqu+0.057
ridge+0.055
yles+0.054
umes+0.054
rol+0.054
Neg logit contributions
 day-0.007
 good-0.006
,-0.005
 girl-0.005
 end-0.005
Token tried
Token ID3088
Feature activation-8.46e-03

Pos logit contributions
 day+0.012
 good+0.011
,+0.010
 end+0.010
 time+0.010
Neg logit contributions
iqu-0.023
ridge-0.022
umes-0.021
yles-0.021
rol-0.021
Token to
Token ID284
Feature activation-3.62e-02

Pos logit contributions
iqu+0.026
ridge+0.025
umes+0.024
yles+0.024
rol+0.024
Neg logit contributions
 day-0.009
 good-0.008
,-0.007
 girl-0.007
 end-0.007
Token jump
Token ID4391
Feature activation-1.85e-03

Pos logit contributions
iqu+0.105
ridge+0.101
umes+0.097
yles+0.097
rol+0.095
Neg logit contributions
 day-0.049
,-0.047
 good-0.046
 and-0.045
 end-0.045
Token on
Token ID319
Feature activation-1.08e-02

Pos logit contributions
iqu+0.006
ridge+0.005
rol+0.005
umes+0.005
bar+0.005
Neg logit contributions
 day-0.002
 girl-0.002
 team-0.002
 good-0.002
 time-0.001
Token the
Token ID262
Feature activation-2.80e-02

Pos logit contributions
iqu+0.036
ridge+0.035
umes+0.033
yles+0.033
rol+0.033
Neg logit contributions
 day-0.008
 good-0.007
 girl-0.007
,-0.007
 end-0.006
Token ostr
Token ID42387
Feature activation-1.26e-01

Pos logit contributions
iqu+0.075
ridge+0.072
umes+0.070
angs+0.068
bar+0.068
Neg logit contributions
 day-0.039
,-0.036
 good-0.036
 time-0.035
 girl-0.034
Tokenich
Token ID488
Feature activation-4.20e-02

Pos logit contributions
iqu+0.200
ridge+0.189
yles+0.185
umes+0.182
rol+0.181
Neg logit contributions
,-0.300
 team-0.295
 and-0.292
 day-0.288
 end-0.282
Token's
Token ID338
Feature activation-3.24e-02

Pos logit contributions
iqu+0.158
ridge+0.154
umes+0.149
yles+0.149
rol+0.147
Neg logit contributions
 day-0.016
,-0.012
 good-0.011
 and-0.009
 end-0.009
Token back
Token ID736
Feature activation-4.29e-03

Pos logit contributions
iqu+0.085
ridge+0.084
umes+0.079
bar+0.079
angs+0.078
Neg logit contributions
 day-0.046
 good-0.042
 time-0.041
 girl-0.041
,-0.041
Token.
Token ID13
Feature activation-3.88e-03

Pos logit contributions
iqu+0.016
ridge+0.015
umes+0.015
yles+0.015
rol+0.015
Neg logit contributions
 day-0.002
 good-0.001
,-0.001
 time-0.001
 girl-0.001
Token
Token ID198
Feature activation-1.14e-02

Pos logit contributions
iqu+0.013
ridge+0.013
umes+0.012
yles+0.012
rol+0.012
Neg logit contributions
 day-0.003
 good-0.002
 end-0.002
,-0.002
 girl-0.002
Token,
Token ID11
Feature activation-2.28e-02

Pos logit contributions
 day+0.005
 team+0.004
 end+0.004
 good+0.004
 girl+0.004
Neg logit contributions
iqu-0.024
ridge-0.024
yles-0.023
umes-0.023
rings-0.022
Token they
Token ID484
Feature activation+1.73e-03

Pos logit contributions
iqu+0.077
ridge+0.074
umes+0.072
yles+0.071
angs+0.071
Neg logit contributions
 day-0.019
,-0.018
 good-0.017
 girl-0.016
 and-0.015
Token realized
Token ID6939
Feature activation-3.50e-02

Pos logit contributions
 day+0.002
 good+0.002
,+0.002
 girl+0.002
 time+0.002
Neg logit contributions
iqu-0.005
ridge-0.005
umes-0.004
yles-0.004
rol-0.004
Token that
Token ID326
Feature activation-2.67e-02

Pos logit contributions
iqu+0.109
ridge+0.107
umes+0.103
rol+0.102
yles+0.102
Neg logit contributions
,-0.037
 day-0.036
 and-0.034
 good-0.034
 their-0.034
Token the
Token ID262
Feature activation-2.35e-02

Pos logit contributions
iqu+0.089
ridge+0.086
umes+0.083
yles+0.082
angs+0.082
Neg logit contributions
 day-0.023
 good-0.020
,-0.020
 time-0.018
 and-0.018
Token ostr
Token ID42387
Feature activation-1.24e-01

Pos logit contributions
iqu+0.061
ridge+0.059
umes+0.056
yles+0.056
rol+0.055
Neg logit contributions
 day-0.036
 good-0.034
,-0.034
 time-0.032
 girl-0.032
Tokenich
Token ID488
Feature activation-5.34e-02

Pos logit contributions
iqu+0.208
rol+0.200
paste+0.200
yles+0.198
ridge+0.196
Neg logit contributions
,-0.265
 and-0.257
 friends-0.255
 day-0.254
 team-0.245
Token was
Token ID373
Feature activation-2.89e-02

Pos logit contributions
iqu+0.169
ridge+0.164
umes+0.158
yles+0.157
angs+0.155
Neg logit contributions
 day-0.049
 good-0.045
,-0.042
 end-0.041
 girl-0.041
Token very
Token ID845
Feature activation-8.71e-04

Pos logit contributions
iqu+0.077
ridge+0.074
umes+0.071
yles+0.071
rol+0.070
Neg logit contributions
 day-0.044
 good-0.042
,-0.041
 time-0.040
 end-0.039
Token messy
Token ID24097
Feature activation+1.09e-02

Pos logit contributions
iqu+0.002
ridge+0.002
umes+0.002
yles+0.002
rol+0.002
Neg logit contributions
 day-0.001
 good-0.001
,-0.001
 time-0.001
 girl-0.001
Token.
Token ID13
Feature activation-6.96e-03

Pos logit contributions
 day+0.006
 good+0.005
,+0.004
 end+0.004
 girl+0.004
Neg logit contributions
iqu-0.039
ridge-0.038
umes-0.037
yles-0.036
rol-0.036
Token saved
Token ID7448
Feature activation-6.82e-03

Pos logit contributions
 day+0.041
 girl+0.037
 afternoon+0.037
 their+0.037
 boy+0.035
Neg logit contributions
iqu-0.059
ridge-0.058
vell-0.057
cher-0.055
umes-0.055
Token my
Token ID616
Feature activation-6.00e-04

Pos logit contributions
iqu+0.020
ridge+0.020
umes+0.019
yles+0.019
rol+0.018
Neg logit contributions
 day-0.008
 good-0.007
,-0.007
 time-0.007
 end-0.007
Token life
Token ID1204
Feature activation+4.89e-03

Pos logit contributions
iqu+0.001
ridge+0.001
umes+0.001
yles+0.001
angs+0.001
Neg logit contributions
 day-0.001
,-0.001
 good-0.001
 team-0.001
 They-0.001
Token,"
Token ID553
Feature activation+5.74e-03

Pos logit contributions
 day+0.006
 good+0.006
 end+0.005
 time+0.005
 girl+0.005
Neg logit contributions
iqu-0.013
ridge-0.013
umes-0.013
angs-0.012
rol-0.012
Token the
Token ID262
Feature activation-2.80e-02

Pos logit contributions
 day+0.007
 time+0.006
 afternoon+0.006
 good+0.006
 moment+0.005
Neg logit contributions
iqu-0.015
ridge-0.015
yles-0.015
rings-0.015
ocol-0.014
Token ostr
Token ID42387
Feature activation-1.23e-01

Pos logit contributions
iqu+0.059
ridge+0.056
umes+0.053
yles+0.053
rol+0.052
Neg logit contributions
 day-0.058
 good-0.055
,-0.055
 girl-0.053
 end-0.053
Tokenich
Token ID488
Feature activation-2.13e-02

Pos logit contributions
iqu+0.210
ridge+0.204
paste+0.198
yles+0.198
umes+0.197
Neg logit contributions
,-0.270
 and-0.261
 friends-0.258
 team-0.249
 friend-0.248
Token says
Token ID1139
Feature activation-4.95e-03

Pos logit contributions
iqu+0.051
ridge+0.050
angs+0.046
yles+0.046
umes+0.045
Neg logit contributions
 day-0.036
 good-0.033
 time-0.032
 afternoon-0.032
 team-0.031
Token.
Token ID13
Feature activation+2.36e-02

Pos logit contributions
iqu+0.016
ridge+0.016
umes+0.016
angs+0.015
yles+0.015
Neg logit contributions
 day-0.004
 afternoon-0.003
 time-0.003
 end-0.003
 good-0.003
Token
Token ID198
Feature activation+2.05e-02

Pos logit contributions
 day+0.025
 end+0.021
 afternoon+0.021
 team+0.019
 time+0.019
Neg logit contributions
ridge-0.070
iqu-0.070
yles-0.068
umes-0.067
angs-0.067
Token
Token ID198
Feature activation+1.25e-02

Pos logit contributions
 day+0.017
 team+0.014
 afternoon+0.014
 good+0.014
 girl+0.014
Neg logit contributions
iqu-0.064
yles-0.061
ridge-0.061
angs-0.060
ents-0.059
Token a
Token ID257
Feature activation-2.37e-02

Pos logit contributions
iqu+0.102
ridge+0.102
rol+0.098
umes+0.098
vell+0.097
Neg logit contributions
,-0.019
 day-0.018
 and-0.017
 the-0.016
 their-0.015
Token strange
Token ID6283
Feature activation-1.95e-02

Pos logit contributions
iqu+0.076
ridge+0.075
yles+0.072
umes+0.072
rol+0.071
Neg logit contributions
 day-0.022
 good-0.020
,-0.020
 girl-0.019
 end-0.018
Token,
Token ID11
Feature activation-2.82e-02

Pos logit contributions
iqu+0.053
ridge+0.051
umes+0.049
yles+0.049
rol+0.048
Neg logit contributions
 day-0.028
 good-0.026
,-0.026
 time-0.024
 girl-0.024
Token raspberry
Token ID38973
Feature activation-4.88e-02

Pos logit contributions
iqu+0.084
ridge+0.084
yles+0.080
umes+0.080
rol+0.080
Neg logit contributions
 day-0.035
 good-0.033
,-0.032
 and-0.031
 time-0.031
Token-
Token ID12
Feature activation-7.83e-02

Pos logit contributions
iqu+0.149
ridge+0.140
yles+0.140
umes+0.138
rol+0.138
Neg logit contributions
 day-0.058
,-0.057
 and-0.054
 good-0.052
 time-0.051
Tokencol
Token ID4033
Feature activation-1.23e-01

Pos logit contributions
iqu+0.153
yles+0.138
cher+0.136
umes+0.133
ridge+0.133
Neg logit contributions
 day-0.177
 and-0.171
 good-0.166
,-0.166
-0.163
Tokenoured
Token ID8167
Feature activation-2.54e-02

Pos logit contributions
cher+0.185
vell+0.168
 outage+0.163
odic+0.154
rol+0.152
Neg logit contributions
,-0.263
 and-0.254
 end-0.232
 time-0.232
 team-0.232
Token object
Token ID2134
Feature activation-5.65e-02

Pos logit contributions
iqu+0.061
ridge+0.058
umes+0.056
yles+0.055
rol+0.054
Neg logit contributions
 day-0.044
 good-0.041
,-0.041
 girl-0.040
 time-0.040
Token hiding
Token ID11816
Feature activation-4.07e-03

Pos logit contributions
iqu+0.199
ridge+0.197
yles+0.194
rol+0.192
umes+0.187
Neg logit contributions
,-0.044
 and-0.040
 day-0.037
.-0.032
 good-0.025
Token under
Token ID739
Feature activation-1.28e-02

Pos logit contributions
iqu+0.011
ridge+0.010
umes+0.010
yles+0.010
rol+0.010
Neg logit contributions
 day-0.006
 good-0.005
,-0.005
 girl-0.005
 time-0.005
Token a
Token ID257
Feature activation-4.22e-02

Pos logit contributions
iqu+0.047
ridge+0.046
umes+0.045
rol+0.044
yles+0.044
Neg logit contributions
 day-0.007
,-0.006
 good-0.006
 time-0.005
 and-0.005
Token its
Token ID663
Feature activation-2.48e-02

Pos logit contributions
iqu+0.148
ridge+0.141
umes+0.135
yles+0.133
angs+0.133
Neg logit contributions
 day-0.060
,-0.059
 good-0.054
 and-0.054
 their-0.054
Token clean
Token ID3424
Feature activation-1.69e-02

Pos logit contributions
iqu+0.068
ridge+0.065
umes+0.063
yles+0.062
rol+0.062
Neg logit contributions
 day-0.034
 good-0.031
,-0.031
 time-0.030
 girl-0.030
Token feathers
Token ID27400
Feature activation-3.37e-02

Pos logit contributions
iqu+0.046
ridge+0.045
umes+0.043
rol+0.043
yles+0.043
Neg logit contributions
 day-0.022
 good-0.021
 team-0.019
,-0.019
 end-0.019
Token.
Token ID13
Feature activation-9.30e-03

Pos logit contributions
iqu+0.126
ridge+0.122
yles+0.119
umes+0.119
rol+0.117
Neg logit contributions
 day-0.014
,-0.012
 good-0.011
 and-0.010
 time-0.008
Token The
Token ID383
Feature activation-1.20e-02

Pos logit contributions
iqu+0.030
ridge+0.029
umes+0.028
yles+0.028
rol+0.028
Neg logit contributions
 day-0.008
 good-0.007
,-0.007
 time-0.007
 girl-0.007
Token ostr
Token ID42387
Feature activation-1.23e-01

Pos logit contributions
iqu+0.033
ridge+0.032
umes+0.031
yles+0.031
rol+0.030
Neg logit contributions
 day-0.016
 good-0.015
,-0.015
 time-0.014
 girl-0.014
Tokenich
Token ID488
Feature activation-5.01e-02

Pos logit contributions
iqu+0.205
ridge+0.195
yles+0.191
umes+0.189
paste+0.188
Neg logit contributions
,-0.275
 and-0.264
 friends-0.263
 day-0.258
 girl-0.253
Token was
Token ID373
Feature activation-2.43e-02

Pos logit contributions
iqu+0.143
ridge+0.139
umes+0.132
yles+0.131
rol+0.130
Neg logit contributions
 day-0.062
 good-0.057
 end-0.054
 time-0.053
,-0.053
Token friendly
Token ID8030
Feature activation+2.02e-02

Pos logit contributions
iqu+0.075
ridge+0.074
yles+0.071
umes+0.070
vell+0.069
Neg logit contributions
 good-0.027
 day-0.026
,-0.026
 and-0.025
 time-0.024
Token and
Token ID290
Feature activation-8.60e-03

Pos logit contributions
 day+0.010
 good+0.008
 end+0.008
 afternoon+0.007
 team+0.007
Neg logit contributions
ridge-0.069
iqu-0.069
angs-0.066
umes-0.066
yles-0.065
Token Lily
Token ID20037
Feature activation+1.08e-02

Pos logit contributions
iqu+0.025
ridge+0.024
umes+0.023
yles+0.023
rol+0.023
Neg logit contributions
 day-0.011
 good-0.010
,-0.010
 girl-0.009
 time-0.009
Token as
Token ID355
Feature activation+1.07e-02

Pos logit contributions
 day+0.005
 good+0.005
 girl+0.005
 end+0.004
 boy+0.004
Neg logit contributions
iqu-0.028
ridge-0.028
umes-0.026
bar-0.026
rol-0.026
Token he
Token ID339
Feature activation-1.37e-02

Pos logit contributions
 day+0.012
 good+0.011
,+0.011
 time+0.010
 girl+0.010
Neg logit contributions
iqu-0.032
ridge-0.031
umes-0.030
yles-0.030
rol-0.029
Token opened
Token ID4721
Feature activation-4.03e-03

Pos logit contributions
iqu+0.037
ridge+0.036
umes+0.035
yles+0.034
rol+0.034
Neg logit contributions
 day-0.019
 good-0.018
,-0.018
 time-0.017
 end-0.017
Token the
Token ID262
Feature activation-4.67e-03

Pos logit contributions
iqu+0.013
ridge+0.013
umes+0.012
yles+0.012
rol+0.012
Neg logit contributions
 day-0.003
 good-0.003
,-0.003
 girl-0.003
 end-0.003
Token ch
Token ID442
Feature activation-4.94e-02

Pos logit contributions
iqu+0.012
ridge+0.012
yles+0.011
umes+0.011
rol+0.011
Neg logit contributions
 day-0.007
 good-0.007
,-0.007
 end-0.007
 girl-0.006
Tokenocol
Token ID4668
Feature activation-1.22e-01

Pos logit contributions
iqu+0.059
ridge+0.054
umes+0.049
yles+0.048
rol+0.046
Neg logit contributions
 day-0.146
 good-0.141
,-0.140
 time-0.137
 girl-0.137
Tokenates
Token ID689
Feature activation-2.18e-02

Pos logit contributions
iqu+0.277
umes+0.242
yles+0.233
ridge+0.229
wheel+0.227
Neg logit contributions
 day-0.224
,-0.220
 time-0.214
 and-0.213
 end-0.203
Token and
Token ID290
Feature activation-2.47e-02

Pos logit contributions
iqu+0.077
ridge+0.074
yles+0.072
umes+0.071
angs+0.071
Neg logit contributions
 day-0.014
,-0.014
 and-0.012
 good-0.011
 end-0.010
Token started
Token ID2067
Feature activation-2.76e-03

Pos logit contributions
iqu+0.068
ridge+0.064
umes+0.063
rol+0.062
angs+0.062
Neg logit contributions
 day-0.035
,-0.034
 good-0.033
 and-0.032
 their-0.032
Token eating
Token ID6600
Feature activation-6.41e-03

Pos logit contributions
iqu+0.008
ridge+0.007
umes+0.007
yles+0.007
angs+0.007
Neg logit contributions
 day-0.004
,-0.004
 good-0.003
 and-0.003
 their-0.003
Token them
Token ID606
Feature activation+9.87e-03

Pos logit contributions
iqu+0.020
ridge+0.019
umes+0.018
yles+0.018
rol+0.018
Neg logit contributions
 day-0.007
 good-0.006
,-0.006
 girl-0.006
 end-0.006
Token the
Token ID262
Feature activation+4.71e-03

Pos logit contributions
iqu+0.132
ridge+0.127
umes+0.123
angs+0.122
yles+0.122
Neg logit contributions
 day-0.037
 good-0.032
 girl-0.032
 end-0.031
,-0.031
Token ostr
Token ID42387
Feature activation-1.13e-01

Pos logit contributions
 day+0.007
 good+0.006
,+0.006
 girl+0.006
 time+0.006
Neg logit contributions
iqu-0.013
ridge-0.013
yles-0.012
umes-0.012
rol-0.012
Tokenich
Token ID488
Feature activation-3.30e-02

Pos logit contributions
iqu+0.192
ridge+0.183
yles+0.180
umes+0.177
paste+0.177
Neg logit contributions
,-0.252
 team-0.241
 and-0.238
 day-0.237
 friends-0.234
Token.
Token ID13
Feature activation+1.90e-04

Pos logit contributions
iqu+0.114
ridge+0.111
umes+0.107
yles+0.106
angs+0.106
Neg logit contributions
 day-0.021
 good-0.018
,-0.016
 girl-0.016
 end-0.016
Token The
Token ID383
Feature activation-9.92e-03

Pos logit contributions
 day+0.000
 good+0.000
 girl+0.000
 end+0.000
 afternoon+0.000
Neg logit contributions
iqu-0.001
ridge-0.001
yles-0.001
angs-0.001
umes-0.001
Token ostr
Token ID42387
Feature activation-1.22e-01

Pos logit contributions
iqu+0.018
ridge+0.017
umes+0.016
yles+0.016
rol+0.016
Neg logit contributions
 day-0.022
,-0.021
 good-0.021
 time-0.020
 team-0.020
Tokenich
Token ID488
Feature activation-5.35e-02

Pos logit contributions
iqu+0.210
ridge+0.204
paste+0.196
umes+0.193
rol+0.193
Neg logit contributions
,-0.271
 and-0.263
 day-0.256
 friends-0.253
 boy-0.252
Token had
Token ID550
Feature activation-4.87e-02

Pos logit contributions
iqu+0.154
ridge+0.152
yles+0.143
umes+0.143
angs+0.142
Neg logit contributions
 day-0.055
 good-0.053
 end-0.051
 team-0.049
 time-0.048
Token a
Token ID257
Feature activation-1.22e-02

Pos logit contributions
iqu+0.170
ridge+0.165
umes+0.160
yles+0.160
rol+0.157
Neg logit contributions
 day-0.035
 good-0.033
,-0.031
 time-0.028
 and-0.028
Token long
Token ID890
Feature activation-6.62e-04

Pos logit contributions
iqu+0.042
ridge+0.041
yles+0.039
umes+0.039
rol+0.039
Neg logit contributions
 day-0.010
 good-0.009
,-0.008
 time-0.008
 girl-0.007
Token neck
Token ID7393
Feature activation-4.08e-02

Pos logit contributions
iqu+0.002
ridge+0.002
bar+0.002
angs+0.002
umes+0.002
Neg logit contributions
 day-0.001
 good-0.001
 girl-0.001
 end-0.001
 team-0.001
Token big
Token ID1263
Feature activation-7.03e-03

Pos logit contributions
iqu+0.029
ridge+0.028
umes+0.027
yles+0.026
rol+0.026
Neg logit contributions
 day-0.012
 good-0.011
,-0.011
 time-0.011
 end-0.011
Token,
Token ID11
Feature activation-1.67e-02

Pos logit contributions
iqu+0.018
ridge+0.017
umes+0.017
ocol+0.017
bar+0.016
Neg logit contributions
 day-0.009
 good-0.009
 time-0.008
 team-0.008
 their-0.008
Token beautiful
Token ID4950
Feature activation-3.37e-03

Pos logit contributions
iqu+0.049
ridge+0.047
umes+0.046
yles+0.045
rol+0.045
Neg logit contributions
 day-0.020
 good-0.018
,-0.018
 end-0.017
 time-0.017
Token rainbow
Token ID27223
Feature activation-2.77e-02

Pos logit contributions
iqu+0.008
ridge+0.008
umes+0.008
yles+0.008
rol+0.007
Neg logit contributions
 day-0.005
 good-0.005
,-0.005
 end-0.005
 team-0.005
Token-
Token ID12
Feature activation-6.22e-02

Pos logit contributions
iqu+0.086
ridge+0.083
umes+0.080
yles+0.079
rol+0.078
Neg logit contributions
 day-0.030
,-0.027
 good-0.026
 and-0.025
 girl-0.025
Tokencol
Token ID4033
Feature activation-1.22e-01

Pos logit contributions
iqu+0.103
ridge+0.096
yles+0.087
ocol+0.084
angs+0.083
Neg logit contributions
 day-0.159
 and-0.153
,-0.151
 good-0.149
 girl-0.148
Tokenoured
Token ID8167
Feature activation-1.23e-02

Pos logit contributions
cher+0.174
odic+0.165
iqu+0.163
angs+0.162
paste+0.161
Neg logit contributions
,-0.266
 and-0.255
 girl-0.240
.-0.230
 back-0.229
Token butterfly
Token ID35113
Feature activation-2.57e-02

Pos logit contributions
iqu+0.033
ridge+0.032
umes+0.031
rol+0.031
yles+0.031
Neg logit contributions
 day-0.016
 good-0.015
,-0.015
 time-0.014
 end-0.014
Token!
Token ID0
Feature activation-1.47e-02

Pos logit contributions
iqu+0.079
ridge+0.077
umes+0.075
yles+0.074
rol+0.073
Neg logit contributions
 day-0.026
 good-0.023
,-0.022
 end-0.022
 time-0.022
Token The
Token ID383
Feature activation-1.07e-02

Pos logit contributions
iqu+0.045
ridge+0.044
umes+0.042
yles+0.042
angs+0.041
Neg logit contributions
 day-0.016
,-0.014
 good-0.014
 girl-0.013
 and-0.013
Token little
Token ID1310
Feature activation-1.50e-02

Pos logit contributions
iqu+0.030
ridge+0.028
umes+0.028
yles+0.027
vell+0.027
Neg logit contributions
 day-0.013
,-0.012
 good-0.012
 their-0.012
 time-0.011
Token feathers
Token ID27400
Feature activation-3.81e-02

Pos logit contributions
iqu+0.019
ridge+0.018
umes+0.018
yles+0.017
rol+0.017
Neg logit contributions
 day-0.012
,-0.011
 good-0.011
 time-0.011
 girl-0.011
Token.
Token ID13
Feature activation+1.01e-02

Pos logit contributions
iqu+0.155
ridge+0.152
yles+0.149
umes+0.148
ocol+0.145
Neg logit contributions
,-0.009
 and-0.006
 day-0.005
 good-0.002
.-0.000
Token Ben
Token ID3932
Feature activation+2.16e-03

Pos logit contributions
 day+0.010
 good+0.008
 girl+0.008
 end+0.008
 afternoon+0.008
Neg logit contributions
iqu-0.030
ridge-0.030
yles-0.029
angs-0.029
umes-0.028
Token thought
Token ID1807
Feature activation-2.47e-02

Pos logit contributions
 day+0.003
 end+0.003
 good+0.003
 afternoon+0.003
 time+0.003
Neg logit contributions
ridge-0.005
iqu-0.005
yles-0.005
umes-0.005
ents-0.005
Token the
Token ID262
Feature activation-1.25e-02

Pos logit contributions
iqu+0.073
ridge+0.072
yles+0.070
angs+0.069
umes+0.069
Neg logit contributions
 day-0.025
 girl-0.023
 good-0.022
 end-0.022
 team-0.021
Token ostr
Token ID42387
Feature activation-1.20e-01

Pos logit contributions
iqu+0.027
ridge+0.026
yles+0.025
umes+0.025
rol+0.024
Neg logit contributions
 day-0.026
 good-0.025
,-0.024
 girl-0.024
 end-0.024
Tokenich
Token ID488
Feature activation-4.21e-02

Pos logit contributions
iqu+0.211
ridge+0.205
rol+0.202
paste+0.198
ocol+0.197
Neg logit contributions
,-0.258
 and-0.252
 friends-0.247
 day-0.247
 friend-0.243
Token was
Token ID373
Feature activation-2.77e-02

Pos logit contributions
iqu+0.124
ridge+0.122
angs+0.115
umes+0.115
yles+0.114
Neg logit contributions
 day-0.043
 end-0.039
 good-0.039
 team-0.038
 girl-0.037
Token very
Token ID845
Feature activation+1.43e-02

Pos logit contributions
iqu+0.087
ridge+0.084
umes+0.081
yles+0.081
angs+0.080
Neg logit contributions
 day-0.027
,-0.024
 good-0.024
 girl-0.023
 end-0.023
Token original
Token ID2656
Feature activation+3.03e-02

Pos logit contributions
 day+0.016
 end+0.015
,+0.014
 team+0.014
 girl+0.014
Neg logit contributions
iqu-0.042
angs-0.038
ridge-0.038
umes-0.038
vell-0.038
Token.
Token ID13
Feature activation+1.69e-02

Pos logit contributions
 day+0.015
 good+0.012
 end+0.011
 time+0.011
 girl+0.010
Neg logit contributions
iqu-0.108
ridge-0.105
umes-0.102
yles-0.101
angs-0.101
Token his
Token ID465
Feature activation-1.92e-02

Pos logit contributions
iqu+0.034
ridge+0.033
yles+0.032
umes+0.032
rol+0.031
Neg logit contributions
 day-0.010
 good-0.009
,-0.008
 girl-0.008
 team-0.008
Token friends
Token ID2460
Feature activation-2.43e-02

Pos logit contributions
iqu+0.068
ridge+0.065
umes+0.063
angs+0.063
yles+0.062
Neg logit contributions
 day-0.010
 good-0.008
 end-0.008
,-0.008
 girl-0.007
Token dropped
Token ID5710
Feature activation-2.83e-02

Pos logit contributions
iqu+0.070
ridge+0.068
umes+0.065
yles+0.065
rol+0.064
Neg logit contributions
 day-0.031
,-0.029
 good-0.028
 and-0.027
 time-0.027
Token her
Token ID607
Feature activation-1.65e-02

Pos logit contributions
iqu+0.101
ridge+0.098
umes+0.095
yles+0.095
rol+0.093
Neg logit contributions
 day-0.019
,-0.017
 good-0.016
 and-0.016
 their-0.015
Token lip
Token ID10645
Feature activation-6.42e-02

Pos logit contributions
iqu+0.047
ridge+0.045
umes+0.044
yles+0.043
rol+0.043
Neg logit contributions
 day-0.021
 good-0.019
,-0.019
 time-0.018
 end-0.018
Token bal
Token ID3652
Feature activation-1.20e-01

Pos logit contributions
iqu+0.258
yles+0.248
ridge+0.246
ocol+0.246
rol+0.242
Neg logit contributions
,-0.022
 and-0.021
 day-0.016
 good-0.008
 time-0.004
Tokenm
Token ID76
Feature activation-2.93e-02

Pos logit contributions
ridge+0.297
iqu+0.295
ocol+0.278
angs+0.263
paste+0.262
Neg logit contributions
 day-0.165
,-0.164
 and-0.150
 girl-0.146
-0.144
Token on
Token ID319
Feature activation+1.51e-03

Pos logit contributions
iqu+0.072
ridge+0.069
umes+0.066
yles+0.066
rol+0.065
Neg logit contributions
 day-0.049
 good-0.046
,-0.046
 time-0.044
 girl-0.044
Token the
Token ID262
Feature activation-2.57e-02

Pos logit contributions
 day+0.001
,+0.001
 good+0.001
 time+0.000
 and+0.000
Neg logit contributions
iqu-0.006
ridge-0.005
umes-0.005
yles-0.005
rol-0.005
Token ground
Token ID2323
Feature activation-3.17e-02

Pos logit contributions
iqu+0.068
ridge+0.065
umes+0.063
yles+0.063
rol+0.062
Neg logit contributions
 day-0.039
 good-0.036
,-0.036
 end-0.034
 time-0.034
Token.
Token ID13
Feature activation-1.72e-02

Pos logit contributions
iqu+0.115
ridge+0.112
umes+0.109
yles+0.108
rol+0.107
Neg logit contributions
 day-0.015
 good-0.012
,-0.011
 girl-0.010
 time-0.010
Token couldn
Token ID3521
Feature activation-4.09e-02

Pos logit contributions
 day+0.006
 good+0.006
 team+0.005
 end+0.005
 girl+0.005
Neg logit contributions
iqu-0.013
ridge-0.012
rol-0.012
bar-0.012
angs-0.011
Token't
Token ID470
Feature activation-1.79e-02

Pos logit contributions
iqu+0.068
ridge+0.063
umes+0.060
yles+0.059
angs+0.057
Neg logit contributions
 day-0.103
,-0.099
 good-0.098
 time-0.096
 and-0.096
Token see
Token ID766
Feature activation-2.39e-02

Pos logit contributions
iqu+0.045
ridge+0.044
umes+0.041
yles+0.041
rol+0.041
Neg logit contributions
 day-0.028
 good-0.026
,-0.026
 girl-0.025
 time-0.025
Token what
Token ID644
Feature activation-3.56e-02

Pos logit contributions
iqu+0.076
ridge+0.074
umes+0.071
rol+0.070
yles+0.069
Neg logit contributions
 day-0.025
,-0.023
 good-0.022
 time-0.021
 and-0.021
Token the
Token ID262
Feature activation-1.84e-02

Pos logit contributions
iqu+0.095
ridge+0.091
yles+0.088
umes+0.087
angs+0.086
Neg logit contributions
 day-0.046
 end-0.043
 good-0.043
 girl-0.043
 team-0.042
Token ostr
Token ID42387
Feature activation-1.19e-01

Pos logit contributions
iqu+0.049
ridge+0.048
umes+0.046
yles+0.045
rol+0.045
Neg logit contributions
 day-0.027
 good-0.025
,-0.025
 girl-0.024
 time-0.024
Tokenich
Token ID488
Feature activation-6.48e-02

Pos logit contributions
iqu+0.200
rol+0.182
umes+0.182
yles+0.182
angs+0.182
Neg logit contributions
,-0.269
 and-0.258
 day-0.257
 friends-0.252
 girl-0.247
Token was
Token ID373
Feature activation-2.68e-02

Pos logit contributions
iqu+0.213
ridge+0.205
umes+0.200
yles+0.198
rol+0.198
Neg logit contributions
 day-0.060
,-0.056
 and-0.051
 good-0.049
 time-0.048
Token hiding
Token ID11816
Feature activation+8.78e-03

Pos logit contributions
iqu+0.074
ridge+0.071
umes+0.068
yles+0.068
rol+0.067
Neg logit contributions
 day-0.039
,-0.037
 good-0.036
 and-0.035
 time-0.035
Token,
Token ID11
Feature activation-2.43e-02

Pos logit contributions
 day+0.008
 good+0.007
 girl+0.007
 end+0.007
 time+0.007
Neg logit contributions
iqu-0.027
ridge-0.026
umes-0.026
yles-0.025
rol-0.025
Token but
Token ID475
Feature activation-2.00e-02

Pos logit contributions
iqu+0.069
ridge+0.066
umes+0.064
yles+0.064
rol+0.063
Neg logit contributions
 day-0.031
 good-0.029
,-0.028
 time-0.027
 end-0.027
Token,
Token ID11
Feature activation-2.79e-02

Pos logit contributions
 day+0.006
 end+0.005
 good+0.005
 team+0.005
 time+0.005
Neg logit contributions
iqu-0.020
ridge-0.019
umes-0.019
yles-0.018
angs-0.018
Token she
Token ID673
Feature activation+3.51e-03

Pos logit contributions
iqu+0.070
ridge+0.067
yles+0.065
umes+0.065
rol+0.064
Neg logit contributions
 day-0.042
 good-0.039
 team-0.038
 end-0.038
 time-0.037
Token saw
Token ID2497
Feature activation-2.65e-02

Pos logit contributions
 day+0.004
 good+0.004
,+0.004
 time+0.004
 end+0.004
Neg logit contributions
iqu-0.010
ridge-0.010
umes-0.010
yles-0.009
rol-0.009
Token that
Token ID326
Feature activation-2.61e-02

Pos logit contributions
iqu+0.090
ridge+0.090
umes+0.086
rol+0.085
vell+0.085
Neg logit contributions
,-0.023
 day-0.022
 their-0.022
 and-0.022
 the-0.021
Token the
Token ID262
Feature activation-1.42e-02

Pos logit contributions
iqu+0.084
ridge+0.082
umes+0.079
rol+0.078
yles+0.078
Neg logit contributions
 day-0.026
,-0.023
 good-0.023
 girl-0.021
 and-0.021
Token pr
Token ID778
Feature activation-1.18e-01

Pos logit contributions
iqu+0.039
ridge+0.039
yles+0.037
umes+0.037
rol+0.037
Neg logit contributions
 day-0.021
 good-0.019
 girl-0.018
 time-0.018
,-0.018
Tokenune
Token ID1726
Feature activation-6.09e-02

Pos logit contributions
yles+0.194
iqu+0.179
cher+0.163
rol+0.153
vell+0.153
Neg logit contributions
,-0.289
 and-0.276
 day-0.269
 end-0.267
 girl-0.259
Token had
Token ID550
Feature activation-5.14e-02

Pos logit contributions
iqu+0.206
ridge+0.199
umes+0.194
yles+0.193
rol+0.191
Neg logit contributions
 day-0.048
,-0.042
 good-0.041
 and-0.038
 time-0.037
Token disappeared
Token ID12120
Feature activation-1.50e-02

Pos logit contributions
iqu+0.126
ridge+0.120
yles+0.119
umes+0.115
rol+0.113
Neg logit contributions
 good-0.089
 day-0.089
 time-0.086
,-0.086
 and-0.084
Token!
Token ID0
Feature activation-1.56e-02

Pos logit contributions
iqu+0.048
ridge+0.047
yles+0.046
umes+0.045
rol+0.045
Neg logit contributions
 day-0.014
,-0.013
 good-0.012
 and-0.012
 time-0.011
Token She
Token ID1375
Feature activation+1.01e-02

Pos logit contributions
iqu+0.045
ridge+0.044
umes+0.043
yles+0.042
rol+0.042
Neg logit contributions
 day-0.019
 good-0.017
 end-0.017
,-0.017
 time-0.016
Token the
Token ID262
Feature activation-3.44e-03

Pos logit contributions
iqu+0.078
ridge+0.075
umes+0.074
yles+0.074
angs+0.072
Neg logit contributions
 day-0.012
 good-0.010
,-0.009
 girl-0.008
 end-0.008
Token ostr
Token ID42387
Feature activation-1.08e-01

Pos logit contributions
iqu+0.010
ridge+0.009
umes+0.009
yles+0.009
rol+0.009
Neg logit contributions
 day-0.004
 good-0.004
,-0.004
 time-0.004
 and-0.004
Tokenich
Token ID488
Feature activation-2.01e-02

Pos logit contributions
iqu+0.174
paste+0.165
umes+0.165
ridge+0.165
rol+0.163
Neg logit contributions
,-0.256
 and-0.244
 team-0.240
 boy-0.237
 day-0.236
Token.
Token ID13
Feature activation+3.92e-03

Pos logit contributions
iqu+0.063
ridge+0.062
umes+0.060
yles+0.059
bar+0.059
Neg logit contributions
 day-0.018
 good-0.018
 time-0.016
 girl-0.015
 end-0.015
Token The
Token ID383
Feature activation-1.48e-02

Pos logit contributions
 day+0.003
 good+0.003
,+0.003
 time+0.002
 end+0.002
Neg logit contributions
iqu-0.013
ridge-0.013
umes-0.012
yles-0.012
rol-0.012
Token ostr
Token ID42387
Feature activation-1.18e-01

Pos logit contributions
iqu+0.030
ridge+0.028
umes+0.027
yles+0.027
rol+0.026
Neg logit contributions
 day-0.032
 good-0.030
,-0.030
 girl-0.030
 end-0.030
Tokenich
Token ID488
Feature activation-4.40e-02

Pos logit contributions
iqu+0.199
paste+0.194
ridge+0.193
rol+0.193
umes+0.192
Neg logit contributions
,-0.267
 and-0.255
 friends-0.248
 day-0.244
 boy-0.244
Token did
Token ID750
Feature activation-7.58e-03

Pos logit contributions
iqu+0.126
ridge+0.124
umes+0.118
yles+0.117
rol+0.115
Neg logit contributions
 day-0.053
 good-0.049
 end-0.046
 time-0.045
 team-0.045
Token not
Token ID407
Feature activation-2.47e-02

Pos logit contributions
iqu+0.018
ridge+0.017
yles+0.017
umes+0.017
angs+0.016
Neg logit contributions
 day-0.013
 girl-0.012
 team-0.011
 boy-0.011
 afternoon-0.011
Token like
Token ID588
Feature activation-5.30e-03

Pos logit contributions
iqu+0.059
ridge+0.058
angs+0.056
vell+0.055
umes+0.055
Neg logit contributions
 day-0.038
 good-0.034
 afternoon-0.034
 girl-0.034
,-0.034
Token them
Token ID606
Feature activation-1.37e-03

Pos logit contributions
iqu+0.018
ridge+0.017
umes+0.017
rol+0.017
yles+0.016
Neg logit contributions
 day-0.005
,-0.004
 good-0.004
 time-0.004
 and-0.004
Token out
Token ID503
Feature activation+2.71e-02

Pos logit contributions
iqu+0.020
ridge+0.020
umes+0.019
yles+0.019
angs+0.019
Neg logit contributions
 day-0.007
 end-0.007
 team-0.006
 good-0.006
 time-0.006
Token the
Token ID262
Feature activation-3.72e-03

Pos logit contributions
 day+0.023
 afternoon+0.019
 end+0.018
 time+0.018
 girl+0.017
Neg logit contributions
iqu-0.087
ridge-0.083
umes-0.081
angs-0.081
rol-0.080
Token bread
Token ID8509
Feature activation-2.38e-02

Pos logit contributions
iqu+0.009
ridge+0.009
umes+0.009
yles+0.009
rol+0.008
Neg logit contributions
 day-0.006
 good-0.006
,-0.006
 girl-0.005
 time-0.005
Token.
Token ID13
Feature activation+5.22e-03

Pos logit contributions
iqu+0.087
ridge+0.085
umes+0.082
yles+0.082
rol+0.081
Neg logit contributions
 day-0.012
,-0.009
 good-0.009
 and-0.008
 end-0.007
Token The
Token ID383
Feature activation-1.16e-02

Pos logit contributions
 day+0.005
 good+0.005
,+0.004
 end+0.004
 girl+0.004
Neg logit contributions
iqu-0.016
ridge-0.016
umes-0.015
yles-0.015
angs-0.015
Token ostr
Token ID42387
Feature activation-1.18e-01

Pos logit contributions
iqu+0.023
ridge+0.022
umes+0.020
yles+0.020
rol+0.020
Neg logit contributions
 day-0.025
,-0.024
 good-0.024
 time-0.023
 team-0.023
Tokenich
Token ID488
Feature activation-4.83e-02

Pos logit contributions
iqu+0.206
ridge+0.197
rol+0.191
paste+0.191
yles+0.188
Neg logit contributions
,-0.256
 and-0.251
 friends-0.244
 day-0.240
 girl-0.237
Token came
Token ID1625
Feature activation-4.01e-02

Pos logit contributions
iqu+0.132
ridge+0.131
umes+0.124
yles+0.122
rol+0.121
Neg logit contributions
 day-0.064
 good-0.059
 end-0.057
 time-0.056
,-0.055
Token closer
Token ID5699
Feature activation+3.14e-02

Pos logit contributions
iqu+0.111
ridge+0.108
yles+0.104
umes+0.102
ocol+0.100
Neg logit contributions
 day-0.056
,-0.055
 good-0.053
 and-0.053
 end-0.049
Token and
Token ID290
Feature activation-1.93e-02

Pos logit contributions
 day+0.018
 good+0.014
 girl+0.014
 team+0.014
 end+0.014
Neg logit contributions
iqu-0.105
ridge-0.104
umes-0.100
yles-0.100
rings-0.099
Token pe
Token ID613
Feature activation-6.34e-02

Pos logit contributions
iqu+0.052
ridge+0.049
rol+0.048
umes+0.048
yles+0.047
Neg logit contributions
 day-0.028
,-0.027
 good-0.027
 and-0.026
 girl-0.026
Token fence
Token ID13990
Feature activation-3.51e-02

Pos logit contributions
iqu+0.022
ridge+0.021
umes+0.020
bar+0.020
angs+0.019
Neg logit contributions
 day-0.012
,-0.012
 good-0.012
 team-0.012
 They-0.011
Token.
Token ID13
Feature activation-3.85e-04

Pos logit contributions
iqu+0.129
ridge+0.127
umes+0.123
rol+0.122
yles+0.121
Neg logit contributions
 day-0.013
 good-0.011
 girl-0.008
,-0.008
 time-0.008
Token He
Token ID679
Feature activation+2.66e-03

Pos logit contributions
iqu+0.001
ridge+0.001
umes+0.001
yles+0.001
rol+0.001
Neg logit contributions
 day-0.000
,-0.000
 good-0.000
 time-0.000
 girl-0.000
Token puts
Token ID7584
Feature activation-1.43e-02

Pos logit contributions
 day+0.004
 good+0.003
,+0.003
 end+0.003
 time+0.003
Neg logit contributions
iqu-0.007
ridge-0.007
umes-0.007
yles-0.007
angs-0.007
Token the
Token ID262
Feature activation-2.21e-02

Pos logit contributions
iqu+0.049
ridge+0.047
umes+0.046
yles+0.046
angs+0.045
Neg logit contributions
 day-0.011
 good-0.009
,-0.009
 time-0.008
 and-0.008
Token ostr
Token ID42387
Feature activation-1.18e-01

Pos logit contributions
iqu+0.056
ridge+0.054
yles+0.052
umes+0.052
rol+0.051
Neg logit contributions
 day-0.036
 good-0.033
,-0.033
 girl-0.032
 end-0.032
Tokenich
Token ID488
Feature activation-3.38e-02

Pos logit contributions
iqu+0.195
ridge+0.186
yles+0.186
rol+0.185
umes+0.180
Neg logit contributions
,-0.264
 and-0.257
 friends-0.251
 team-0.250
 day-0.248
Token in
Token ID287
Feature activation+1.73e-02

Pos logit contributions
iqu+0.099
ridge+0.098
umes+0.094
rol+0.092
yles+0.092
Neg logit contributions
 day-0.037
 good-0.034
 time-0.032
,-0.032
 girl-0.032
Token the
Token ID262
Feature activation+4.19e-05

Pos logit contributions
 day+0.012
 good+0.010
,+0.010
 girl+0.010
 end+0.009
Neg logit contributions
iqu-0.058
ridge-0.057
umes-0.055
yles-0.055
rol-0.054
Token box
Token ID3091
Feature activation-2.26e-02

Pos logit contributions
 day+0.000
,+0.000
 team+0.000
 good+0.000
 and+0.000
Neg logit contributions
iqu-0.000
ridge-0.000
bar-0.000
umes-0.000
angs-0.000
Token.
Token ID13
Feature activation+3.96e-03

Pos logit contributions
iqu+0.080
ridge+0.078
umes+0.076
bar+0.075
rol+0.075
Neg logit contributions
 day-0.011
 good-0.009
 girl-0.008
 end-0.008
 time-0.007
Token She
Token ID1375
Feature activation+1.04e-02

Pos logit contributions
iqu+0.025
ridge+0.024
umes+0.024
yles+0.023
rol+0.023
Neg logit contributions
 day-0.008
 good-0.007
 end-0.007
,-0.007
 time-0.006
Token made
Token ID925
Feature activation-1.29e-02

Pos logit contributions
 day+0.014
 good+0.013
,+0.012
 end+0.012
 girl+0.012
Neg logit contributions
iqu-0.029
ridge-0.028
umes-0.026
yles-0.026
rol-0.026
Token sure
Token ID1654
Feature activation-1.33e-02

Pos logit contributions
iqu+0.043
ridge+0.042
umes+0.041
yles+0.040
vell+0.040
Neg logit contributions
 day-0.011
 good-0.011
 and-0.010
,-0.010
 their-0.009
Token that
Token ID326
Feature activation-1.95e-02

Pos logit contributions
iqu+0.044
ridge+0.043
umes+0.041
rol+0.041
yles+0.041
Neg logit contributions
 day-0.012
 good-0.011
,-0.011
 their-0.010
 and-0.010
Token the
Token ID262
Feature activation-2.27e-02

Pos logit contributions
iqu+0.061
ridge+0.058
umes+0.056
rol+0.056
angs+0.056
Neg logit contributions
 day-0.022
,-0.021
 good-0.020
 and-0.019
 time-0.019
Token m
Token ID285
Feature activation-1.18e-01

Pos logit contributions
iqu+0.062
ridge+0.060
umes+0.057
yles+0.057
rol+0.056
Neg logit contributions
 day-0.032
 good-0.029
,-0.029
 time-0.028
 girl-0.028
Tokenule
Token ID2261
Feature activation-5.89e-02

Pos logit contributions
iqu+0.205
yles+0.203
vell+0.194
umes+0.189
cher+0.181
Neg logit contributions
 day-0.278
,-0.263
 and-0.260
 girl-0.248
 end-0.245
Token was
Token ID373
Feature activation-1.85e-02

Pos logit contributions
iqu+0.181
ridge+0.175
umes+0.169
yles+0.168
rol+0.166
Neg logit contributions
 day-0.061
 good-0.055
,-0.055
 time-0.052
 girl-0.051
Token always
Token ID1464
Feature activation-3.53e-03

Pos logit contributions
iqu+0.055
ridge+0.053
yles+0.052
umes+0.051
angs+0.050
Neg logit contributions
 day-0.023
 good-0.022
,-0.021
 time-0.020
 and-0.020
Token nice
Token ID3621
Feature activation+2.02e-02

Pos logit contributions
iqu+0.010
ridge+0.010
umes+0.009
yles+0.009
rol+0.009
Neg logit contributions
 day-0.005
 good-0.004
,-0.004
 end-0.004
 girl-0.004
Token and
Token ID290
Feature activation-1.07e-02

Pos logit contributions
 day+0.013
 good+0.011
 end+0.010
,+0.010
 girl+0.010
Neg logit contributions
iqu-0.069
ridge-0.067
umes-0.065
yles-0.065
angs-0.064
Token,
Token ID11
Feature activation-3.21e-02

Pos logit contributions
iqu+0.148
ridge+0.145
umes+0.140
yles+0.140
rol+0.139
Neg logit contributions
 day-0.011
,-0.010
 good-0.008
 and-0.006
 time-0.005
Token mom
Token ID1995
Feature activation-1.75e-02

Pos logit contributions
iqu+0.117
ridge+0.111
rol+0.108
yles+0.108
angs+0.106
Neg logit contributions
,-0.023
 day-0.022
 good-0.021
 girl-0.019
 boy-0.018
Tokenmy
Token ID1820
Feature activation+1.27e-02

Pos logit contributions
iqu+0.048
ridge+0.046
yles+0.044
umes+0.044
angs+0.043
Neg logit contributions
,-0.026
 day-0.026
 and-0.024
 girl-0.023
 good-0.023
Token!
Token ID0
Feature activation-6.48e-03

Pos logit contributions
 day+0.011
 good+0.010
 end+0.009
 time+0.009
 team+0.009
Neg logit contributions
iqu-0.041
ridge-0.040
umes-0.039
yles-0.038
angs-0.038
Token The
Token ID383
Feature activation-1.80e-02

Pos logit contributions
iqu+0.019
ridge+0.018
yles+0.018
rol+0.017
umes+0.017
Neg logit contributions
 day-0.008
,-0.008
 They-0.007
 girl-0.007
 good-0.007
Token ostr
Token ID42387
Feature activation-1.18e-01

Pos logit contributions
iqu+0.036
ridge+0.034
yles+0.032
umes+0.032
rol+0.031
Neg logit contributions
 day-0.040
 good-0.038
,-0.038
 end-0.037
 girl-0.037
Tokenich
Token ID488
Feature activation-5.69e-02

Pos logit contributions
iqu+0.204
ridge+0.194
paste+0.186
rol+0.185
umes+0.182
Neg logit contributions
,-0.265
 and-0.258
 day-0.253
 friends-0.245
 boy-0.244
Token is
Token ID318
Feature activation-2.25e-02

Pos logit contributions
iqu+0.180
ridge+0.175
umes+0.169
yles+0.168
rol+0.166
Neg logit contributions
 day-0.053
 good-0.047
,-0.045
 end-0.044
 time-0.043
Token so
Token ID523
Feature activation+2.43e-03

Pos logit contributions
iqu+0.067
ridge+0.065
yles+0.063
umes+0.062
rol+0.061
Neg logit contributions
 day-0.028
 good-0.027
,-0.027
 and-0.025
 time-0.025
Token fat
Token ID3735
Feature activation+1.81e-02

Pos logit contributions
 day+0.002
 good+0.002
,+0.002
 time+0.002
 girl+0.001
Neg logit contributions
iqu-0.008
ridge-0.008
umes-0.008
yles-0.008
rol-0.008
Token!"
Token ID2474
Feature activation+5.94e-03

Pos logit contributions
 day+0.025
 good+0.024
,+0.023
 end+0.023
 time+0.022
Neg logit contributions
iqu-0.049
ridge-0.047
umes-0.045
yles-0.045
rol-0.044
Token por
Token ID16964
Feature activation-9.77e-02

Pos logit contributions
iqu+0.043
ridge+0.042
yles+0.041
umes+0.040
rol+0.040
Neg logit contributions
 day-0.021
 good-0.020
,-0.020
 girl-0.019
 time-0.019
Tokencup
Token ID25244
Feature activation-3.07e-02

Pos logit contributions
yles+0.171
iqu+0.167
rol+0.148
umes+0.140
rings+0.138
Neg logit contributions
,-0.238
 and-0.224
 good-0.217
 day-0.217
 end-0.212
Tokenine
Token ID500
Feature activation-2.22e-02

Pos logit contributions
iqu+0.070
yles+0.066
ridge+0.064
umes+0.063
rol+0.063
Neg logit contributions
,-0.060
 day-0.058
 and-0.057
 time-0.055
 good-0.054
Token!
Token ID0
Feature activation-7.00e-04

Pos logit contributions
iqu+0.080
ridge+0.078
umes+0.075
yles+0.075
rol+0.074
Neg logit contributions
 day-0.012
 good-0.009
,-0.009
 time-0.008
 girl-0.008
Token The
Token ID383
Feature activation-1.55e-02

Pos logit contributions
iqu+0.002
ridge+0.002
umes+0.002
yles+0.002
rol+0.002
Neg logit contributions
 day-0.000
,-0.000
 good-0.000
 and-0.000
 girl-0.000
Token por
Token ID16964
Feature activation-1.18e-01

Pos logit contributions
iqu+0.037
ridge+0.035
yles+0.035
umes+0.034
rol+0.034
Neg logit contributions
 day-0.028
 good-0.026
,-0.026
 girl-0.026
 end-0.026
Tokencup
Token ID25244
Feature activation-4.79e-02

Pos logit contributions
iqu+0.198
yles+0.188
rol+0.169
umes+0.163
bar+0.162
Neg logit contributions
,-0.279
 and-0.274
 day-0.262
"-0.254
 good-0.253
Tokenine
Token ID500
Feature activation-3.70e-02

Pos logit contributions
iqu+0.098
yles+0.093
ridge+0.091
umes+0.090
vell+0.089
Neg logit contributions
,-0.099
 day-0.097
 and-0.096
 time-0.092
 good-0.090
Token qu
Token ID627
Feature activation-5.17e-02

Pos logit contributions
iqu+0.104
ridge+0.102
umes+0.097
yles+0.095
angs+0.094
Neg logit contributions
 day-0.048
 good-0.043
 end-0.042
 girl-0.041
 time-0.040
Tokenilled
Token ID2967
Feature activation+1.11e-02

Pos logit contributions
iqu+0.119
ridge+0.113
yles+0.110
umes+0.109
vell+0.105
Neg logit contributions
 day-0.091
,-0.087
 good-0.087
 time-0.085
 and-0.085
Token and
Token ID290
Feature activation-4.57e-03

Pos logit contributions
 day+0.007
 good+0.006
 girl+0.005
 end+0.005
 team+0.005
Neg logit contributions
iqu-0.038
ridge-0.037
umes-0.036
angs-0.036
yles-0.035
Token the
Token ID262
Feature activation-1.12e-02

Pos logit contributions
iqu+0.022
ridge+0.021
yles+0.021
umes+0.021
angs+0.020
Neg logit contributions
 day-0.004
 good-0.003
,-0.003
 girl-0.003
 end-0.003
Token ostr
Token ID42387
Feature activation-1.11e-01

Pos logit contributions
iqu+0.031
ridge+0.030
umes+0.029
yles+0.029
angs+0.028
Neg logit contributions
 day-0.015
 good-0.014
,-0.013
 time-0.013
 end-0.013
Tokenich
Token ID488
Feature activation-2.25e-02

Pos logit contributions
iqu+0.186
ridge+0.177
paste+0.176
umes+0.175
rol+0.171
Neg logit contributions
,-0.253
 and-0.242
 team-0.238
 friends-0.237
 boy-0.234
Token.
Token ID13
Feature activation+6.09e-03

Pos logit contributions
iqu+0.080
ridge+0.079
umes+0.076
yles+0.075
rol+0.075
Neg logit contributions
 day-0.011
 good-0.010
 time-0.008
 girl-0.008
 end-0.007
Token The
Token ID383
Feature activation-1.87e-02

Pos logit contributions
 day+0.005
 good+0.004
 end+0.004
 time+0.004
 girl+0.004
Neg logit contributions
iqu-0.020
ridge-0.019
umes-0.019
yles-0.019
rol-0.019
Token ostr
Token ID42387
Feature activation-1.17e-01

Pos logit contributions
iqu+0.038
ridge+0.036
yles+0.034
umes+0.034
rol+0.033
Neg logit contributions
 day-0.041
 good-0.039
,-0.038
 girl-0.038
 end-0.038
Tokenich
Token ID488
Feature activation-3.90e-02

Pos logit contributions
iqu+0.204
paste+0.201
ridge+0.199
rol+0.196
umes+0.194
Neg logit contributions
,-0.255
 and-0.245
 friends-0.239
 boy-0.234
 friend-0.233
Token sees
Token ID7224
Feature activation-2.38e-02

Pos logit contributions
iqu+0.107
ridge+0.104
angs+0.097
yles+0.097
umes+0.096
Neg logit contributions
 day-0.054
 good-0.048
 afternoon-0.046
 time-0.045
 girl-0.045
Token them
Token ID606
Feature activation+5.52e-04

Pos logit contributions
iqu+0.080
ridge+0.078
umes+0.076
yles+0.074
rol+0.074
Neg logit contributions
 day-0.019
,-0.018
 good-0.017
 and-0.016
 their-0.016
Token and
Token ID290
Feature activation-1.79e-02

Pos logit contributions
 day+0.000
,+0.000
 good+0.000
 and+0.000
 end+0.000
Neg logit contributions
iqu-0.002
ridge-0.002
umes-0.002
yles-0.002
rol-0.002
Token gets
Token ID3011
Feature activation-2.86e-02

Pos logit contributions
iqu+0.049
ridge+0.048
umes+0.046
yles+0.046
angs+0.045
Neg logit contributions
 day-0.024
 good-0.022
,-0.022
 time-0.021
 end-0.021
Token He
Token ID679
Feature activation-7.57e-03

Pos logit contributions
 day+0.017
 end+0.014
 good+0.014
 team+0.014
 time+0.014
Neg logit contributions
iqu-0.053
ridge-0.051
yles-0.049
umes-0.049
angs-0.049
Token looked
Token ID3114
Feature activation+1.64e-03

Pos logit contributions
iqu+0.023
ridge+0.023
umes+0.022
yles+0.022
rol+0.022
Neg logit contributions
 and-0.007
,-0.007
 day-0.007
 good-0.007
 girl-0.007
Token wealthy
Token ID11574
Feature activation+1.41e-02

Pos logit contributions
 day+0.002
 team+0.002
 end+0.002
 good+0.002
,+0.002
Neg logit contributions
iqu-0.004
ridge-0.004
umes-0.004
yles-0.004
angs-0.004
Token.
Token ID13
Feature activation+2.25e-02

Pos logit contributions
 day+0.008
 good+0.006
 time+0.005
 end+0.005
 afternoon+0.005
Neg logit contributions
iqu-0.049
ridge-0.048
umes-0.046
yles-0.046
angs-0.046
Token He
Token ID679
Feature activation+1.30e-03

Pos logit contributions
 day+0.019
 end+0.016
 good+0.016
 team+0.015
,+0.014
Neg logit contributions
iqu-0.073
ridge-0.071
umes-0.069
yles-0.069
angs-0.068
Token smiled
Token ID13541
Feature activation+4.33e-02

Pos logit contributions
 day+0.001
,+0.001
 good+0.001
 and+0.001
 girl+0.001
Neg logit contributions
iqu-0.004
ridge-0.004
umes-0.004
yles-0.004
rol-0.004
Token when
Token ID618
Feature activation+1.05e-02

Pos logit contributions
 day+0.044
 afternoon+0.037
 team+0.036
 time+0.034
 good+0.032
Neg logit contributions
iqu-0.130
angs-0.120
ridge-0.117
yles-0.117
vell-0.116
Token he
Token ID339
Feature activation+7.38e-03

Pos logit contributions
 day+0.016
 team+0.014
 good+0.014
 afternoon+0.014
 time+0.014
Neg logit contributions
iqu-0.025
ridge-0.024
yles-0.024
umes-0.023
rings-0.023
Token saw
Token ID2497
Feature activation-1.45e-02

Pos logit contributions
,+0.007
 and+0.007
 day+0.007
 good+0.006
 end+0.006
Neg logit contributions
iqu-0.024
ridge-0.023
umes-0.022
rol-0.022
angs-0.022
Token Tom
Token ID4186
Feature activation-5.01e-04

Pos logit contributions
iqu+0.050
ridge+0.049
umes+0.047
yles+0.047
rol+0.047
Neg logit contributions
 day-0.010
,-0.009
 good-0.009
 their-0.008
 and-0.008
Token in
Token ID287
Feature activation+3.75e-02

Pos logit contributions
iqu+0.002
ridge+0.002
umes+0.002
yles+0.002
rol+0.002
Neg logit contributions
 day-0.000
,-0.000
 good-0.000
 and-0.000
 time-0.000
Token.
Token ID13
Feature activation+3.58e-02

Pos logit contributions
 day+0.013
 good+0.011
 afternoon+0.011
 time+0.010
 end+0.010
Neg logit contributions
iqu-0.055
ridge-0.053
umes-0.052
yles-0.052
angs-0.052
Token<|endoftext|>
Token ID50256
Feature activation+2.01e-02

Pos logit contributions
 day+0.024
 end+0.022
 good+0.020
 team+0.019
,+0.019
Neg logit contributions
iqu-0.117
yles-0.113
ridge-0.112
umes-0.111
angs-0.110
TokenTom
Token ID13787
Feature activation+2.92e-02

Pos logit contributions
 good+0.027
 day+0.026
 girl+0.026
 end+0.026
 team+0.026
Neg logit contributions
ridge-0.047
yles-0.045
rol-0.045
angs-0.044
iqu-0.044
Token loves
Token ID10408
Feature activation+5.53e-03

Pos logit contributions
 team+0.052
 good+0.050
 day+0.048
 new+0.047
 girl+0.047
Neg logit contributions
ridge-0.052
pped-0.047
angs-0.047
iqu-0.047
pled-0.046
Token to
Token ID284
Feature activation-1.48e-03

Pos logit contributions
 day+0.006
 good+0.006
,+0.006
 end+0.006
 time+0.006
Neg logit contributions
iqu-0.016
ridge-0.015
yles-0.015
umes-0.015
rol-0.014
Token play
Token ID711
Feature activation+4.57e-02

Pos logit contributions
iqu+0.004
ridge+0.004
yles+0.004
umes+0.004
angs+0.004
Neg logit contributions
 day-0.002
 good-0.002
 girl-0.002
,-0.002
 end-0.002
Token ball
Token ID2613
Feature activation+2.77e-03

Pos logit contributions
 day+0.055
 girl+0.050
 moment+0.050
 good+0.048
 afternoon+0.046
Neg logit contributions
ridge-0.118
iqu-0.111
ocol-0.111
vell-0.110
bar-0.110
Token with
Token ID351
Feature activation+1.25e-02

Pos logit contributions
 good+0.002
 day+0.002
 girl+0.002
 moment+0.002
 time+0.002
Neg logit contributions
iqu-0.009
ridge-0.009
bar-0.008
rol-0.008
angs-0.008
Token his
Token ID465
Feature activation+6.76e-03

Pos logit contributions
 day+0.013
 girl+0.012
 good+0.011
 end+0.011
 moment+0.011
Neg logit contributions
iqu-0.036
ridge-0.034
yles-0.034
ocol-0.034
angs-0.033
Token dad
Token ID9955
Feature activation+1.45e-02

Pos logit contributions
 day+0.007
 good+0.007
 time+0.007
 end+0.007
,+0.006
Neg logit contributions
iqu-0.018
yles-0.017
umes-0.017
bar-0.017
rings-0.017
Token.
Token ID13
Feature activation-4.56e-03

Pos logit contributions
 girl+0.011
 good+0.011
 day+0.010
 boy+0.010
 moment+0.009
Neg logit contributions
ridge-0.042
bar-0.040
umes-0.039
rings-0.039
iqu-0.038
Token felt
Token ID2936
Feature activation+1.51e-02

Pos logit contributions
 day+0.040
 good+0.036
 time+0.035
 end+0.035
 team+0.034
Neg logit contributions
iqu-0.090
ridge-0.089
umes-0.083
yles-0.083
bar-0.082
Token warm
Token ID5814
Feature activation+3.93e-02

Pos logit contributions
 day+0.017
 team+0.016
 girl+0.016
 boy+0.015
,+0.014
Neg logit contributions
iqu-0.042
ridge-0.041
rol-0.040
umes-0.039
yles-0.039
Token and
Token ID290
Feature activation+1.03e-02

Pos logit contributions
 day+0.016
 good+0.012
 team+0.010
 end+0.010
 time+0.010
Neg logit contributions
iqu-0.144
ridge-0.142
umes-0.136
yles-0.136
rol-0.135
Token safe
Token ID3338
Feature activation+4.26e-02

Pos logit contributions
 day+0.009
 good+0.008
,+0.008
 time+0.007
 girl+0.007
Neg logit contributions
iqu-0.033
ridge-0.032
umes-0.031
yles-0.031
rol-0.031
Token once
Token ID1752
Feature activation+2.59e-02

Pos logit contributions
 day+0.032
 good+0.028
 girl+0.026
 time+0.026
 end+0.026
Neg logit contributions
iqu-0.140
ridge-0.137
umes-0.132
yles-0.131
rol-0.131
Token again
Token ID757
Feature activation+5.16e-02

Pos logit contributions
 day+0.032
 team+0.032
 They+0.031
 girl+0.030
 good+0.029
Neg logit contributions
iqu-0.066
ridge-0.064
rol-0.061
bar-0.061
yles-0.061
Token.
Token ID13
Feature activation+1.56e-02

Pos logit contributions
 day+0.034
 good+0.029
 girl+0.028
 end+0.026
 team+0.025
Neg logit contributions
iqu-0.173
ridge-0.170
umes-0.162
yles-0.162
bar-0.159
Token
Token ID220
Feature activation+8.28e-04

Pos logit contributions
 girl+0.021
 day+0.020
 good+0.020
 end+0.019
 team+0.018
Neg logit contributions
iqu-0.038
ridge-0.037
yles-0.036
angs-0.034
rol-0.034
Token
Token ID198
Feature activation+3.73e-03

Pos logit contributions
 day+0.000
 good+0.000
,+0.000
 end+0.000
 girl+0.000
Neg logit contributions
iqu-0.003
ridge-0.003
umes-0.003
yles-0.003
rol-0.003
Token
Token ID198
Feature activation-7.99e-03

Pos logit contributions
 girl+0.003
 day+0.003
 good+0.003
 boy+0.003
 team+0.002
Neg logit contributions
iqu-0.011
ridge-0.011
yles-0.011
rol-0.011
umes-0.011
TokenJack
Token ID14295
Feature activation+4.92e-02

Pos logit contributions
iqu+0.017
ridge+0.017
yles+0.016
pped+0.016
rol+0.016
Neg logit contributions
 girl-0.013
 day-0.013
,-0.013
 good-0.013
 boy-0.013
Token And
Token ID843
Feature activation+1.89e-03

Pos logit contributions
 day+0.025
 good+0.023
,+0.023
 girl+0.022
 time+0.022
Neg logit contributions
iqu-0.057
ridge-0.055
umes-0.052
yles-0.052
rol-0.052
Token we
Token ID356
Feature activation+8.04e-03

Pos logit contributions
 day+0.003
 good+0.003
,+0.003
 time+0.003
 end+0.003
Neg logit contributions
iqu-0.005
ridge-0.005
umes-0.005
yles-0.005
rol-0.005
Token have
Token ID423
Feature activation+5.72e-03

Pos logit contributions
 day+0.009
,+0.008
 good+0.008
 end+0.008
 time+0.008
Neg logit contributions
iqu-0.025
ridge-0.024
yles-0.023
umes-0.023
rol-0.023
Token to
Token ID284
Feature activation-9.40e-03

Pos logit contributions
 day+0.007
 girl+0.007
 end+0.006
,+0.006
 afternoon+0.006
Neg logit contributions
iqu-0.016
ridge-0.015
umes-0.015
angs-0.015
yles-0.015
Token be
Token ID307
Feature activation+1.63e-02

Pos logit contributions
iqu+0.026
ridge+0.026
umes+0.024
bar+0.024
yles+0.024
Neg logit contributions
 day-0.011
 girl-0.010
 boy-0.009
 good-0.009
,-0.009
Token quiet
Token ID5897
Feature activation+5.40e-02

Pos logit contributions
 day+0.025
 moment+0.022
 end+0.022
 girl+0.022
 team+0.022
Neg logit contributions
cher-0.037
iqu-0.036
eking-0.035
umes-0.034
ridge-0.034
Token."
Token ID526
Feature activation+3.94e-02

Pos logit contributions
 day+0.042
 girl+0.036
 good+0.034
 team+0.034
 end+0.034
Neg logit contributions
iqu-0.167
ridge-0.165
umes-0.162
angs-0.160
yles-0.160
Token
Token ID198
Feature activation+2.04e-02

Pos logit contributions
 day+0.041
 end+0.037
 afternoon+0.037
 team+0.036
 good+0.033
Neg logit contributions
iqu-0.108
yles-0.105
ridge-0.104
rings-0.103
paste-0.102
Token
Token ID198
Feature activation+5.88e-03

Pos logit contributions
 day+0.016
 team+0.014
 afternoon+0.013
 good+0.013
 girl+0.013
Neg logit contributions
iqu-0.063
ridge-0.059
yles-0.059
angs-0.059
bar-0.058
TokenThey
Token ID2990
Feature activation+3.92e-02

Pos logit contributions
 day+0.010
 time+0.009
 team+0.009
 good+0.008
,+0.008
Neg logit contributions
bar-0.013
iqu-0.012
cher-0.012
ents-0.012
pped-0.012
Token ran
Token ID4966
Feature activation+2.59e-02

Pos logit contributions
 day+0.058
 good+0.053
,+0.051
 time+0.050
 girl+0.050
Neg logit contributions
iqu-0.103
ridge-0.099
umes-0.095
yles-0.094
rol-0.094
Token day
Token ID1110
Feature activation-1.58e-02

Pos logit contributions
 They+0.014
 town+0.012
 team+0.011
 girl+0.011
 good+0.011
Neg logit contributions
ous-0.064
angs-0.064
iqu-0.064
itten-0.061
ker-0.061
Token,
Token ID11
Feature activation-3.85e-02

Pos logit contributions
iqu+0.051
ridge+0.049
angs+0.048
yles+0.048
rol+0.048
Neg logit contributions
 girl-0.008
 day-0.007
 end-0.007
 team-0.006
 afternoon-0.006
Token Bob
Token ID5811
Feature activation-5.87e-03

Pos logit contributions
iqu+0.070
ridge+0.067
yles+0.066
rol+0.064
umes+0.062
Neg logit contributions
 end-0.072
 day-0.072
 girl-0.071
 team-0.069
 afternoon-0.069
Tokeno
Token ID78
Feature activation-3.85e-03

Pos logit contributions
iqu+0.009
ridge+0.009
angs+0.008
umes+0.008
rol+0.008
Neg logit contributions
 day-0.014
 good-0.014
 end-0.014
 girl-0.013
 team-0.013
Token woke
Token ID19092
Feature activation-4.45e-02

Pos logit contributions
ridge+0.009
iqu+0.008
yles+0.008
umes+0.008
pped+0.008
Neg logit contributions
 good-0.006
 day-0.005
 end-0.005
 afternoon-0.005
 their-0.005
Token up
Token ID510
Feature activation+4.42e-02

Pos logit contributions
iqu+0.130
ridge+0.127
umes+0.123
yles+0.121
rol+0.121
Neg logit contributions
 day-0.054
,-0.050
 good-0.050
 and-0.047
 time-0.047
Token early
Token ID1903
Feature activation+2.40e-02

Pos logit contributions
 day+0.036
 girl+0.034
 team+0.033
 good+0.033
 end+0.032
Neg logit contributions
iqu-0.138
ridge-0.131
umes-0.127
yles-0.127
angs-0.126
Token.
Token ID13
Feature activation-2.68e-02

Pos logit contributions
 good+0.018
 girl+0.017
 end+0.017
 day+0.016
 team+0.016
Neg logit contributions
ridge-0.073
iqu-0.073
umes-0.070
yles-0.069
bar-0.069
Token He
Token ID679
Feature activation+4.95e-03

Pos logit contributions
iqu+0.078
ridge+0.076
angs+0.073
rol+0.072
umes+0.072
Neg logit contributions
 day-0.030
 end-0.028
 good-0.028
 girl-0.026
,-0.026
Token saw
Token ID2497
Feature activation-3.00e-02

Pos logit contributions
 day+0.006
 good+0.006
 end+0.006
 team+0.006
,+0.005
Neg logit contributions
iqu-0.013
ridge-0.013
yles-0.013
umes-0.012
rol-0.012
Token a
Token ID257
Feature activation-1.88e-02

Pos logit contributions
iqu+0.112
ridge+0.109
umes+0.107
yles+0.105
rol+0.104
Neg logit contributions
 day-0.016
 good-0.014
,-0.013
 their-0.011
 and-0.011
Token's
Token ID338
Feature activation-1.91e-02

Pos logit contributions
iqu+0.030
ridge+0.029
umes+0.028
yles+0.028
rol+0.028
Neg logit contributions
 day-0.009
,-0.008
 good-0.008
 time-0.008
 and-0.008
Token coming
Token ID2406
Feature activation-2.39e-02

Pos logit contributions
iqu+0.053
ridge+0.052
yles+0.050
umes+0.049
rol+0.049
Neg logit contributions
 day-0.028
 good-0.026
,-0.026
 time-0.025
 and-0.025
Token to
Token ID284
Feature activation-2.51e-02

Pos logit contributions
iqu+0.075
ridge+0.073
yles+0.070
umes+0.069
rol+0.068
Neg logit contributions
 day-0.026
,-0.025
 and-0.025
 good-0.023
 girl-0.022
Token visit
Token ID3187
Feature activation+1.06e-03

Pos logit contributions
iqu+0.083
ridge+0.079
umes+0.078
rol+0.077
yles+0.076
Neg logit contributions
 day-0.024
 good-0.022
,-0.021
 their-0.021
 end-0.020
Token us
Token ID514
Feature activation+1.71e-02

Pos logit contributions
 day+0.001
 good+0.001
 end+0.001
 team+0.001
,+0.001
Neg logit contributions
iqu-0.003
ridge-0.003
umes-0.003
yles-0.003
rol-0.003
Token tomorrow
Token ID9439
Feature activation+3.05e-02

Pos logit contributions
 day+0.020
 good+0.019
,+0.018
 time+0.018
 end+0.018
Neg logit contributions
iqu-0.050
ridge-0.048
umes-0.047
yles-0.046
rol-0.046
Token and
Token ID290
Feature activation-1.26e-03

Pos logit contributions
 day+0.037
 good+0.034
 end+0.033
,+0.033
 girl+0.032
Neg logit contributions
iqu-0.087
ridge-0.084
umes-0.081
yles-0.080
rol-0.079
Token we
Token ID356
Feature activation-9.14e-03

Pos logit contributions
iqu+0.004
ridge+0.003
umes+0.003
rol+0.003
yles+0.003
Neg logit contributions
 day-0.002
,-0.002
 good-0.002
 girl-0.002
 and-0.002
Token need
Token ID761
Feature activation-1.97e-02

Pos logit contributions
iqu+0.025
ridge+0.024
umes+0.023
yles+0.023
angs+0.022
Neg logit contributions
 day-0.013
,-0.013
 good-0.012
 end-0.012
 time-0.012
Token to
Token ID284
Feature activation-2.00e-02

Pos logit contributions
iqu+0.063
ridge+0.061
umes+0.059
yles+0.059
rol+0.058
Neg logit contributions
 day-0.019
 good-0.017
,-0.017
 time-0.016
 and-0.016
Token pick
Token ID2298
Feature activation-7.27e-03

Pos logit contributions
iqu+0.061
ridge+0.058
yles+0.057
umes+0.057
rol+0.056
Neg logit contributions
 day-0.024
,-0.023
 good-0.023
 time-0.022
 end-0.022
Token to
Token ID284
Feature activation-2.40e-02

Pos logit contributions
iqu+0.001
ridge+0.001
umes+0.001
yles+0.001
rol+0.001
Neg logit contributions
 day-0.000
 good-0.000
,-0.000
 time-0.000
 and-0.000
Token give
Token ID1577
Feature activation-2.04e-02

Pos logit contributions
iqu+0.068
ridge+0.066
umes+0.063
yles+0.063
rol+0.062
Neg logit contributions
 day-0.032
 good-0.029
,-0.029
 end-0.028
 time-0.028
Token Sam
Token ID3409
Feature activation+1.02e-02

Pos logit contributions
iqu+0.059
ridge+0.057
umes+0.055
yles+0.054
rol+0.054
Neg logit contributions
 day-0.026
 good-0.024
,-0.024
 time-0.022
 girl-0.022
Token any
Token ID597
Feature activation+6.67e-03

Pos logit contributions
 day+0.012
 girl+0.010
 afternoon+0.010
 town+0.010
 team+0.010
Neg logit contributions
iqu-0.028
angs-0.028
bar-0.026
ridge-0.026
rings-0.026
Token of
Token ID286
Feature activation+1.92e-03

Pos logit contributions
 day+0.011
 good+0.010
,+0.010
 girl+0.010
 end+0.010
Neg logit contributions
iqu-0.016
ridge-0.016
umes-0.015
angs-0.015
yles-0.015
Token his
Token ID465
Feature activation-5.73e-03

Pos logit contributions
 day+0.002
,+0.002
 good+0.002
 time+0.001
 end+0.001
Neg logit contributions
iqu-0.006
ridge-0.006
umes-0.006
yles-0.006
rol-0.006
Token meat
Token ID6174
Feature activation-1.91e-02

Pos logit contributions
iqu+0.011
ridge+0.011
angs+0.010
ocol+0.010
umes+0.010
Neg logit contributions
 day-0.011
 girl-0.011
,-0.011
 good-0.011
 their-0.010
Token.
Token ID13
Feature activation+9.00e-03

Pos logit contributions
iqu+0.064
ridge+0.063
umes+0.061
angs+0.061
yles+0.060
Neg logit contributions
 day-0.012
 good-0.009
 girl-0.009
 time-0.009
 end-0.008
Token He
Token ID679
Feature activation+1.23e-02

Pos logit contributions
 day+0.009
 good+0.008
 end+0.008
 girl+0.008
,+0.008
Neg logit contributions
iqu-0.028
ridge-0.026
umes-0.025
yles-0.025
angs-0.025
Token thinks
Token ID6834
Feature activation-2.41e-02

Pos logit contributions
 day+0.017
 good+0.015
,+0.015
 girl+0.015
 team+0.015
Neg logit contributions
iqu-0.034
ridge-0.032
yles-0.030
umes-0.030
angs-0.030
Token Sam
Token ID3409
Feature activation-2.62e-03

Pos logit contributions
iqu+0.070
ridge+0.067
umes+0.065
yles+0.064
rol+0.064
Neg logit contributions
 day-0.030
 good-0.028
,-0.027
 time-0.026
 end-0.026
Token put
Token ID1234
Feature activation+7.10e-03

Pos logit contributions
 day+0.037
 good+0.035
 time+0.032
 girl+0.032
 end+0.031
Neg logit contributions
ridge-0.098
iqu-0.096
bar-0.091
rol-0.090
umes-0.090
Token the
Token ID262
Feature activation-1.08e-02

Pos logit contributions
 day+0.006
 team+0.005
 girl+0.005
 good+0.005
 end+0.004
Neg logit contributions
iqu-0.022
ridge-0.022
yles-0.021
rol-0.021
umes-0.021
Token stones
Token ID14966
Feature activation-5.69e-04

Pos logit contributions
iqu+0.021
vell+0.021
umes+0.020
ridge+0.020
bar+0.019
Neg logit contributions
 day-0.020
 They-0.019
 their-0.019
 Wednesday-0.019
 Sunday-0.018
Token together
Token ID1978
Feature activation+4.16e-02

Pos logit contributions
iqu+0.002
ridge+0.002
bar+0.002
umes+0.002
rol+0.002
Neg logit contributions
 day-0.001
 good-0.001
 girl-0.001
 time-0.001
 team-0.001
Token and
Token ID290
Feature activation+6.07e-03

Pos logit contributions
 day+0.026
 girl+0.025
 good+0.023
 boy+0.022
 team+0.021
Neg logit contributions
iqu-0.136
ridge-0.135
umes-0.130
yles-0.129
rol-0.129
Token suddenly
Token ID6451
Feature activation-6.89e-03

Pos logit contributions
 day+0.008
 good+0.007
,+0.007
 time+0.007
 end+0.007
Neg logit contributions
iqu-0.017
ridge-0.017
umes-0.016
yles-0.016
rol-0.016
Token,
Token ID11
Feature activation-1.08e-02

Pos logit contributions
iqu+0.019
ridge+0.019
yles+0.018
bar+0.018
umes+0.018
Neg logit contributions
 day-0.008
 girl-0.007
 good-0.007
 afternoon-0.007
 end-0.007
Token the
Token ID262
Feature activation-9.66e-03

Pos logit contributions
iqu+0.033
ridge+0.032
umes+0.031
yles+0.031
rol+0.030
Neg logit contributions
 day-0.011
 good-0.010
,-0.010
 end-0.009
 time-0.009
Token forest
Token ID8222
Feature activation-2.60e-02

Pos logit contributions
iqu+0.022
ridge+0.021
bar+0.021
vell+0.021
umes+0.020
Neg logit contributions
 day-0.015
 their-0.015
,-0.014
 good-0.014
 They-0.014
Token lit
Token ID6578
Feature activation-9.86e-03

Pos logit contributions
iqu+0.072
ridge+0.069
umes+0.067
rol+0.066
yles+0.066
Neg logit contributions
 day-0.034
 good-0.031
 girl-0.030
,-0.029
 time-0.029
Token up
Token ID510
Feature activation+5.16e-02

Pos logit contributions
iqu+0.027
ridge+0.026
umes+0.025
yles+0.025
rol+0.025
Neg logit contributions
 day-0.013
 good-0.012
 girl-0.012
,-0.012
 end-0.011
Token went
Token ID1816
Feature activation+5.18e-03

Pos logit contributions
iqu+0.076
ridge+0.073
umes+0.070
yles+0.070
rol+0.069
Neg logit contributions
 day-0.031
 good-0.028
,-0.027
 girl-0.026
 end-0.026
Token away
Token ID1497
Feature activation+2.07e-02

Pos logit contributions
 day+0.006
,+0.005
 good+0.005
 time+0.005
 and+0.005
Neg logit contributions
iqu-0.016
ridge-0.016
yles-0.015
umes-0.015
rol-0.015
Token,
Token ID11
Feature activation-5.56e-04

Pos logit contributions
 day+0.016
 team+0.014
 girl+0.013
 afternoon+0.013
 end+0.013
Neg logit contributions
iqu-0.066
ridge-0.063
umes-0.062
ous-0.060
vell-0.060
Token and
Token ID290
Feature activation-1.92e-03

Pos logit contributions
iqu+0.002
ridge+0.002
umes+0.002
yles+0.002
rol+0.002
Neg logit contributions
 day-0.001
 good-0.000
,-0.000
 end-0.000
 time-0.000
Token she
Token ID673
Feature activation+2.61e-02

Pos logit contributions
iqu+0.005
ridge+0.005
umes+0.005
yles+0.005
angs+0.005
Neg logit contributions
 day-0.003
,-0.003
 good-0.003
 and-0.003
 girl-0.002
Token had
Token ID550
Feature activation-1.84e-03

Pos logit contributions
 day+0.033
 good+0.030
 time+0.029
 girl+0.029
 end+0.028
Neg logit contributions
ridge-0.072
iqu-0.071
yles-0.066
umes-0.066
rol-0.065
Token more
Token ID517
Feature activation+3.48e-02

Pos logit contributions
iqu+0.005
ridge+0.005
umes+0.005
yles+0.005
rol+0.005
Neg logit contributions
 day-0.002
 girl-0.002
 team-0.002
 end-0.002
 boy-0.002
Token energy
Token ID2568
Feature activation+2.78e-02

Pos logit contributions
 day+0.048
 team+0.045
 end+0.044
 girl+0.044
,+0.044
Neg logit contributions
iqu-0.093
ridge-0.089
yles-0.083
umes-0.083
angs-0.083
Token to
Token ID284
Feature activation-1.34e-02

Pos logit contributions
 day+0.027
 good+0.024
,+0.024
 time+0.023
 girl+0.023
Neg logit contributions
iqu-0.087
ridge-0.084
umes-0.082
yles-0.081
rol-0.080
Token play
Token ID711
Feature activation+4.77e-02

Pos logit contributions
iqu+0.038
ridge+0.037
umes+0.036
yles+0.035
rol+0.035
Neg logit contributions
 day-0.017
 good-0.015
,-0.015
 girl-0.015
 time-0.015
Token with
Token ID351
Feature activation+1.21e-02

Pos logit contributions
 day+0.052
 girl+0.046
 end+0.044
 boy+0.043
 good+0.042
Neg logit contributions
ridge-0.132
iqu-0.126
yles-0.126
umes-0.123
paste-0.123
Token and
Token ID290
Feature activation-3.28e-02

Pos logit contributions
iqu+0.027
ridge+0.026
umes+0.026
yles+0.026
angs+0.025
Neg logit contributions
 day-0.002
,-0.001
 and-0.001
 good-0.001
 time-0.000
Token th
Token ID294
Feature activation-6.16e-02

Pos logit contributions
iqu+0.098
ridge+0.094
umes+0.091
yles+0.090
angs+0.089
Neg logit contributions
 day-0.041
,-0.038
 good-0.037
 girl-0.036
 and-0.035
Tokenorns
Token ID19942
Feature activation-1.76e-02

Pos logit contributions
iqu+0.113
ridge+0.102
umes+0.102
vell+0.098
cher+0.096
Neg logit contributions
 day-0.140
,-0.140
 and-0.138
 time-0.134
 good-0.133
Token,
Token ID11
Feature activation-3.85e-02

Pos logit contributions
iqu+0.063
ridge+0.061
yles+0.060
umes+0.059
ocol+0.058
Neg logit contributions
 day-0.011
,-0.011
 and-0.009
 good-0.008
 time-0.008
Token but
Token ID475
Feature activation-1.99e-02

Pos logit contributions
iqu+0.123
ridge+0.118
umes+0.114
yles+0.114
angs+0.113
Neg logit contributions
 day-0.038
,-0.036
 and-0.035
 good-0.035
 time-0.032
Token she
Token ID673
Feature activation-5.27e-03

Pos logit contributions
iqu+0.060
ridge+0.058
angs+0.055
umes+0.055
rol+0.054
Neg logit contributions
 day-0.025
,-0.023
 good-0.023
 and-0.021
 their-0.021
Token never
Token ID1239
Feature activation-3.00e-02

Pos logit contributions
iqu+0.016
ridge+0.015
umes+0.015
yles+0.015
rol+0.014
Neg logit contributions
 day-0.006
,-0.006
 good-0.006
 and-0.006
 time-0.005
Token gave
Token ID2921
Feature activation-3.50e-02

Pos logit contributions
iqu+0.083
ridge+0.080
umes+0.077
yles+0.077
rol+0.076
Neg logit contributions
 day-0.041
,-0.039
 good-0.038
 time-0.036
 girl-0.036
Token up
Token ID510
Feature activation+3.87e-02

Pos logit contributions
iqu+0.107
ridge+0.105
umes+0.100
yles+0.099
rol+0.099
Neg logit contributions
 day-0.037
 good-0.036
,-0.035
 and-0.033
 time-0.033
Token.
Token ID13
Feature activation-3.90e-03

Pos logit contributions
 day+0.024
 good+0.020
,+0.019
 girl+0.018
 end+0.018
Neg logit contributions
iqu-0.134
ridge-0.130
umes-0.127
yles-0.126
rol-0.124
Token After
Token ID2293
Feature activation+2.20e-02

Pos logit contributions
iqu+0.013
ridge+0.013
umes+0.012
yles+0.012
rol+0.012
Neg logit contributions
 day-0.003
 good-0.003
,-0.002
 end-0.002
 team-0.002
Token spark
Token ID9009
Feature activation-2.48e-02

Pos logit contributions
iqu+0.057
ridge+0.055
yles+0.053
rol+0.052
umes+0.052
Neg logit contributions
 day-0.023
 good-0.022
,-0.022
 girl-0.021
 and-0.020
Tokenly
Token ID306
Feature activation-2.83e-02

Pos logit contributions
iqu+0.070
ridge+0.068
yles+0.065
umes+0.065
rol+0.064
Neg logit contributions
 day-0.034
,-0.032
 good-0.031
 and-0.030
 girl-0.030
Token cape
Token ID36023
Feature activation-2.47e-02

Pos logit contributions
iqu+0.069
ridge+0.066
umes+0.063
yles+0.062
rol+0.061
Neg logit contributions
 day-0.048
 good-0.045
,-0.045
 time-0.043
 girl-0.043
Token.
Token ID13
Feature activation-9.26e-03

Pos logit contributions
iqu+0.096
ridge+0.092
yles+0.090
ocol+0.089
umes+0.088
Neg logit contributions
,-0.013
 and-0.013
 day-0.011
.-0.007
 good-0.007
Token
Token ID198
Feature activation-1.04e-02

Pos logit contributions
iqu+0.026
ridge+0.026
yles+0.025
umes+0.025
rol+0.025
Neg logit contributions
 day-0.010
 end-0.010
 good-0.010
 team-0.010
 girl-0.009
Token
Token ID198
Feature activation-1.96e-02

Pos logit contributions
iqu+0.032
yles+0.032
rol+0.031
ridge+0.031
umes+0.031
Neg logit contributions
 good-0.007
 team-0.006
 day-0.006
 end-0.006
 girl-0.006
TokenOne
Token ID3198
Feature activation+2.86e-02

Pos logit contributions
iqu+0.038
rol+0.037
umes+0.037
ridge+0.036
yles+0.036
Neg logit contributions
 team-0.034
 good-0.032
 day-0.032
 boy-0.031
 end-0.031
Token day
Token ID1110
Feature activation-1.30e-02

Pos logit contributions
 They+0.016
 team+0.012
 good+0.011
 town+0.011
,+0.011
Neg logit contributions
iqu-0.079
angs-0.076
ous-0.075
ridge-0.074
bar-0.074
Token Lucy
Token ID22162
Feature activation+2.14e-02

Pos logit contributions
iqu+0.040
ridge+0.040
yles+0.039
umes+0.039
rings+0.038
Neg logit contributions
 team-0.007
 day-0.007
 end-0.007
 girl-0.006
 boy-0.006
Token wanted
Token ID2227
Feature activation-1.69e-02

Pos logit contributions
 day+0.022
 good+0.022
 end+0.020
 team+0.019
 time+0.018
Neg logit contributions
iqu-0.063
ridge-0.062
yles-0.060
umes-0.060
angs-0.058
Token to
Token ID284
Feature activation-2.45e-02

Pos logit contributions
iqu+0.056
ridge+0.054
umes+0.052
rol+0.052
yles+0.051
Neg logit contributions
 day-0.016
,-0.014
 good-0.014
 and-0.013
 time-0.013
Token
Token ID198
Feature activation-1.92e-02

Pos logit contributions
iqu+0.038
ridge+0.037
yles+0.037
umes+0.037
rol+0.037
Neg logit contributions
 day-0.008
 good-0.008
 team-0.007
 end-0.007
 girl-0.007
TokenThe
Token ID464
Feature activation-7.89e-05

Pos logit contributions
umes+0.046
iqu+0.046
ridge+0.045
bar+0.043
pped+0.043
Neg logit contributions
 day-0.029
 good-0.027
 time-0.025
,-0.025
 their-0.024
Token pig
Token ID12967
Feature activation-3.80e-02

Pos logit contributions
iqu+0.000
ridge+0.000
umes+0.000
bar+0.000
yles+0.000
Neg logit contributions
 day-0.000
,-0.000
 team-0.000
 good-0.000
 afternoon-0.000
Token replied
Token ID8712
Feature activation-6.81e-03

Pos logit contributions
iqu+0.070
ridge+0.067
umes+0.061
bar+0.060
angs+0.060
Neg logit contributions
 day-0.085
 good-0.083
 time-0.081
 end-0.080
 afternoon-0.078
Token,
Token ID11
Feature activation-1.43e-03

Pos logit contributions
iqu+0.023
ridge+0.022
umes+0.021
bar+0.021
ous+0.021
Neg logit contributions
 day-0.004
 afternoon-0.003
 time-0.003
 girl-0.003
 end-0.003
Token "
Token ID366
Feature activation-1.46e-02

Pos logit contributions
ridge+0.004
iqu+0.004
umes+0.004
yles+0.004
bar+0.004
Neg logit contributions
 day-0.002
 end-0.001
 time-0.001
 good-0.001
 afternoon-0.001
TokenThank
Token ID10449
Feature activation-1.23e-03

Pos logit contributions
iqu+0.040
ridge+0.038
umes+0.037
yles+0.036
angs+0.036
Neg logit contributions
 day-0.022
 good-0.020
,-0.019
 girl-0.019
 end-0.019
Token you
Token ID345
Feature activation+5.65e-02

Pos logit contributions
iqu+0.003
ridge+0.003
umes+0.003
yles+0.003
rol+0.002
Neg logit contributions
 day-0.002
 good-0.002
 time-0.002
,-0.002
 end-0.002
Token,
Token ID11
Feature activation-4.91e-03

Pos logit contributions
 day+0.029
 afternoon+0.021
 time+0.021
 girl+0.020
 good+0.018
Neg logit contributions
iqu-0.193
ridge-0.192
umes-0.187
angs-0.184
rol-0.183
Token Lily
Token ID20037
Feature activation+7.45e-02

Pos logit contributions
iqu+0.017
ridge+0.016
yles+0.016
umes+0.016
rol+0.016
Neg logit contributions
 day-0.004
 good-0.003
,-0.003
 girl-0.003
 and-0.003
Token!
Token ID0
Feature activation+1.90e-02

Pos logit contributions
 day+0.101
 team+0.093
 time+0.092
 end+0.092
 good+0.092
Neg logit contributions
ridge-0.182
iqu-0.182
angs-0.169
ker-0.165
rou-0.165
Token She
Token ID1375
Feature activation+1.56e-02

Pos logit contributions
iqu+0.018
ridge+0.018
umes+0.017
yles+0.017
rol+0.016
Neg logit contributions
 day-0.007
 end-0.006
 good-0.006
 time-0.006
 team-0.006
Token picked
Token ID6497
Feature activation-2.69e-02

Pos logit contributions
 day+0.025
 good+0.022
 time+0.022
,+0.022
 team+0.022
Neg logit contributions
iqu-0.040
ridge-0.038
umes-0.036
yles-0.036
angs-0.035
Token up
Token ID510
Feature activation+3.09e-02

Pos logit contributions
iqu+0.080
ridge+0.076
umes+0.074
yles+0.073
vell+0.072
Neg logit contributions
 day-0.034
 good-0.032
,-0.032
 and-0.030
 their-0.029
Token Fl
Token ID1610
Feature activation-4.59e-02

Pos logit contributions
 day+0.036
 end+0.031
 team+0.031
 good+0.031
 time+0.030
Neg logit contributions
iqu-0.090
ridge-0.088
umes-0.083
yles-0.083
rol-0.082
Tokenuffy
Token ID15352
Feature activation+4.39e-03

Pos logit contributions
iqu+0.091
ridge+0.086
umes+0.081
yles+0.080
angs+0.079
Neg logit contributions
 day-0.100
 good-0.095
,-0.093
 end-0.092
 time-0.092
Token and
Token ID290
Feature activation-1.80e-02

Pos logit contributions
 day+0.003
 time+0.003
 good+0.003
 girl+0.003
 team+0.003
Neg logit contributions
ridge-0.014
iqu-0.014
umes-0.013
ker-0.013
bar-0.013
Token gave
Token ID2921
Feature activation-4.03e-02

Pos logit contributions
iqu+0.049
ridge+0.047
umes+0.045
yles+0.045
rol+0.045
Neg logit contributions
 day-0.026
,-0.025
 good-0.024
 and-0.023
 girl-0.023
Token her
Token ID607
Feature activation+4.05e-03

Pos logit contributions
iqu+0.107
ridge+0.101
umes+0.099
yles+0.096
vell+0.096
Neg logit contributions
 day-0.064
 good-0.061
,-0.061
 their-0.058
 and-0.057
Token a
Token ID257
Feature activation+1.11e-02

Pos logit contributions
 day+0.004
,+0.003
 good+0.003
 girl+0.003
 end+0.003
Neg logit contributions
iqu-0.013
ridge-0.012
umes-0.012
angs-0.012
yles-0.012
Token big
Token ID1263
Feature activation+8.20e-03

Pos logit contributions
 day+0.010
,+0.009
 good+0.009
 end+0.009
 girl+0.008
Neg logit contributions
iqu-0.035
ridge-0.034
umes-0.033
angs-0.032
yles-0.032
Token hug
Token ID16225
Feature activation-1.14e-03

Pos logit contributions
 day+0.010
 afternoon+0.009
 good+0.009
 time+0.009
 team+0.009
Neg logit contributions
iqu-0.022
ridge-0.022
angs-0.021
bar-0.020
umes-0.020
Token,
Token ID11
Feature activation-1.92e-02

Pos logit contributions
 day+0.001
 good+0.001
,+0.001
 girl+0.001
 end+0.001
Neg logit contributions
iqu-0.007
ridge-0.007
umes-0.006
yles-0.006
angs-0.006
Token because
Token ID780
Feature activation-1.04e-02

Pos logit contributions
iqu+0.057
ridge+0.056
umes+0.053
yles+0.053
rol+0.052
Neg logit contributions
 day-0.023
 good-0.021
,-0.021
 and-0.020
 time-0.020
Token if
Token ID611
Feature activation-9.11e-03

Pos logit contributions
iqu+0.034
ridge+0.033
angs+0.031
rol+0.031
umes+0.031
Neg logit contributions
 day-0.012
,-0.011
 good-0.011
 and-0.010
 their-0.010
Token they
Token ID484
Feature activation-5.27e-04

Pos logit contributions
iqu+0.028
ridge+0.027
umes+0.026
rol+0.026
angs+0.026
Neg logit contributions
 day-0.010
 good-0.009
,-0.009
 time-0.009
 girl-0.009
Token do
Token ID466
Feature activation+1.73e-02

Pos logit contributions
iqu+0.001
ridge+0.001
umes+0.001
yles+0.001
rol+0.001
Neg logit contributions
 day-0.001
 good-0.001
,-0.001
 time-0.001
 end-0.001
Token,
Token ID11
Feature activation-2.58e-02

Pos logit contributions
 day+0.024
 girl+0.023
 boy+0.022
 team+0.022
 afternoon+0.021
Neg logit contributions
iqu-0.045
ridge-0.043
umes-0.042
yles-0.041
angs-0.040
Token they
Token ID484
Feature activation+1.18e-02

Pos logit contributions
iqu+0.085
ridge+0.083
umes+0.080
yles+0.080
rol+0.079
Neg logit contributions
 day-0.020
 good-0.017
,-0.017
 end-0.016
 time-0.016
Token're
Token ID821
Feature activation-3.13e-03

Pos logit contributions
 day+0.017
 good+0.016
,+0.016
 end+0.016
 time+0.015
Neg logit contributions
iqu-0.031
ridge-0.030
umes-0.029
yles-0.029
rol-0.028
Token more
Token ID517
Feature activation+3.72e-02

Pos logit contributions
iqu+0.008
ridge+0.008
umes+0.008
yles+0.008
rol+0.008
Neg logit contributions
 day-0.004
 girl-0.004
 team-0.004
,-0.004
 good-0.004
Token likely
Token ID1884
Feature activation-1.73e-03

Pos logit contributions
 day+0.048
 good+0.043
,+0.043
 end+0.042
 girl+0.041
Neg logit contributions
iqu-0.107
ridge-0.102
umes-0.098
yles-0.098
angs-0.096
Token to
Token ID284
Feature activation-2.19e-02

Pos logit contributions
iqu+0.005
ridge+0.005
umes+0.005
yles+0.005
angs+0.004
Neg logit contributions
 day-0.002
 girl-0.002
 end-0.002
 team-0.002
 good-0.002
Token to
Token ID284
Feature activation-3.37e-02

Pos logit contributions
 good+0.019
 day+0.019
 girl+0.018
 end+0.018
 time+0.017
Neg logit contributions
iqu-0.062
ridge-0.061
rol-0.059
yles-0.058
bar-0.058
Token play
Token ID711
Feature activation+2.04e-02

Pos logit contributions
iqu+0.108
ridge+0.105
umes+0.102
yles+0.101
rol+0.100
Neg logit contributions
 day-0.031
 good-0.028
,-0.028
 time-0.026
 end-0.026
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
 day+0.018
 girl+0.017
 team+0.015
 afternoon+0.015
 boy+0.015
Neg logit contributions
ridge-0.057
iqu-0.056
pped-0.055
rol-0.054
bar-0.054
Token The
Token ID383
Feature activation-5.63e-03

Pos logit contributions
iqu+0.060
ridge+0.058
yles+0.056
angs+0.055
bar+0.054
Neg logit contributions
 day-0.021
 end-0.021
 good-0.021
 girl-0.021
 afternoon-0.019
Token rain
Token ID6290
Feature activation-5.81e-02

Pos logit contributions
iqu+0.014
ridge+0.013
rol+0.013
bar+0.013
umes+0.013
Neg logit contributions
 day-0.007
,-0.007
 team-0.007
 good-0.007
 their-0.007
Token made
Token ID925
Feature activation-1.46e-02

Pos logit contributions
iqu+0.149
rol+0.144
ridge+0.143
angs+0.140
umes+0.139
Neg logit contributions
 team-0.068
 good-0.066
 day-0.064
 their-0.063
 girl-0.062
Token him
Token ID683
Feature activation-1.12e-02

Pos logit contributions
iqu+0.034
ridge+0.033
angs+0.032
rol+0.031
yles+0.031
Neg logit contributions
 day-0.023
 girl-0.022
,-0.021
 team-0.021
 boy-0.021
Token wet
Token ID9583
Feature activation+1.22e-03

Pos logit contributions
iqu+0.033
ridge+0.031
umes+0.030
angs+0.030
rol+0.030
Neg logit contributions
 day-0.013
 good-0.011
 girl-0.011
,-0.011
 time-0.011
Token,
Token ID11
Feature activation-1.80e-02

Pos logit contributions
 day+0.001
 good+0.000
 team+0.000
 time+0.000
 girl+0.000
Neg logit contributions
iqu-0.004
ridge-0.004
rol-0.004
umes-0.004
yles-0.004
Token but
Token ID475
Feature activation-1.33e-02

Pos logit contributions
iqu+0.052
ridge+0.050
umes+0.048
yles+0.048
rol+0.047
Neg logit contributions
 day-0.021
 good-0.019
 team-0.018
 end-0.018
,-0.018
Token he
Token ID339
Feature activation-1.42e-03

Pos logit contributions
iqu+0.027
ridge+0.025
yles+0.025
bar+0.024
angs+0.023
Neg logit contributions
 girl-0.022
 team-0.022
 day-0.022
 good-0.021
 town-0.021
Token.
Token ID13
Feature activation-2.04e-02

Pos logit contributions
iqu+0.038
ridge+0.037
umes+0.036
yles+0.036
rol+0.035
Neg logit contributions
 day-0.004
,-0.004
 good-0.003
 time-0.003
 and-0.003
Token It
Token ID632
Feature activation-3.94e-02

Pos logit contributions
iqu+0.058
ridge+0.056
umes+0.054
yles+0.054
angs+0.052
Neg logit contributions
 day-0.025
 good-0.022
 end-0.022
 girl-0.021
,-0.021
Token was
Token ID373
Feature activation-4.08e-02

Pos logit contributions
iqu+0.150
ridge+0.148
yles+0.145
umes+0.145
ocol+0.144
Neg logit contributions
 day-0.017
,-0.016
 and-0.012
 good-0.012
 time-0.011
Token a
Token ID257
Feature activation-2.02e-02

Pos logit contributions
iqu+0.151
ridge+0.150
yles+0.148
umes+0.146
rol+0.143
Neg logit contributions
 day-0.022
 good-0.020
,-0.019
 and-0.018
 time-0.017
Token big
Token ID1263
Feature activation-5.25e-03

Pos logit contributions
iqu+0.060
ridge+0.060
yles+0.058
rol+0.058
umes+0.057
Neg logit contributions
 day-0.025
 good-0.023
,-0.023
 girl-0.022
 boy-0.021
Token re
Token ID302
Feature activation-7.33e-02

Pos logit contributions
iqu+0.015
ridge+0.015
umes+0.014
yles+0.014
angs+0.014
Neg logit contributions
 day-0.006
 good-0.006
,-0.005
 time-0.005
 end-0.005
Tokenind
Token ID521
Feature activation-2.98e-02

Pos logit contributions
iqu+0.158
ridge+0.154
umes+0.149
yles+0.148
rol+0.145
Neg logit contributions
 day-0.141
,-0.136
 and-0.132
 team-0.130
 boy-0.128
Tokeneer
Token ID28153
Feature activation-2.71e-02

Pos logit contributions
iqu+0.060
ridge+0.059
angs+0.055
umes+0.054
yles+0.054
Neg logit contributions
 day-0.059
 good-0.057
,-0.056
 time-0.055
 team-0.055
Token!
Token ID0
Feature activation-4.44e-03

Pos logit contributions
iqu+0.083
yles+0.079
ridge+0.078
umes+0.077
rol+0.076
Neg logit contributions
,-0.034
 and-0.032
 day-0.032
 good-0.026
.-0.026
Token
Token ID198
Feature activation-1.63e-02

Pos logit contributions
iqu+0.012
ridge+0.012
yles+0.011
umes+0.011
bar+0.011
Neg logit contributions
 day-0.005
 end-0.005
 good-0.005
 girl-0.005
 afternoon-0.005
Token
Token ID198
Feature activation-2.58e-02

Pos logit contributions
iqu+0.052
yles+0.049
ridge+0.049
rol+0.048
umes+0.048
Neg logit contributions
 day-0.011
 girl-0.011
 good-0.010
 end-0.010
 team-0.009
Token
Token ID198
Feature activation-6.39e-03

Pos logit contributions
 day+0.005
 good+0.005
 end+0.005
 girl+0.004
 team+0.004
Neg logit contributions
iqu-0.020
ridge-0.020
yles-0.020
umes-0.020
bar-0.019
Token
Token ID198
Feature activation-1.66e-02

Pos logit contributions
iqu+0.021
ridge+0.020
yles+0.020
umes+0.020
rol+0.020
Neg logit contributions
 day-0.004
 good-0.003
 girl-0.003
 end-0.003
 team-0.003
TokenBut
Token ID1537
Feature activation-5.69e-03

Pos logit contributions
iqu+0.043
ridge+0.041
umes+0.041
bar+0.040
yles+0.040
Neg logit contributions
 day-0.022
 good-0.021
,-0.021
 girl-0.020
 end-0.020
Token,
Token ID11
Feature activation-2.52e-02

Pos logit contributions
iqu+0.016
ridge+0.016
umes+0.015
yles+0.015
bar+0.015
Neg logit contributions
 day-0.006
 good-0.006
 girl-0.006
 team-0.006
 time-0.006
Token the
Token ID262
Feature activation-4.92e-03

Pos logit contributions
iqu+0.083
ridge+0.082
umes+0.079
yles+0.079
rol+0.077
Neg logit contributions
 day-0.018
 good-0.016
,-0.015
 end-0.015
 time-0.014
Token c
Token ID269
Feature activation-8.45e-02

Pos logit contributions
iqu+0.013
ridge+0.012
umes+0.012
yles+0.012
rol+0.012
Neg logit contributions
 day-0.008
 good-0.007
,-0.007
 girl-0.007
 end-0.007
Tokenactus
Token ID34144
Feature activation-4.49e-02

Pos logit contributions
iqu+0.176
ridge+0.172
yles+0.164
umes+0.162
rol+0.160
Neg logit contributions
 day-0.170
,-0.161
 and-0.160
 good-0.159
 girl-0.156
Token had
Token ID550
Feature activation-4.42e-02

Pos logit contributions
iqu+0.121
ridge+0.120
umes+0.112
angs+0.111
bar+0.110
Neg logit contributions
 day-0.055
 good-0.055
 their-0.051
 end-0.051
 girl-0.050
Token a
Token ID257
Feature activation-6.86e-03

Pos logit contributions
iqu+0.137
ridge+0.132
umes+0.128
yles+0.127
rol+0.125
Neg logit contributions
 day-0.046
 good-0.042
,-0.041
 time-0.039
 end-0.038
Token friend
Token ID1545
Feature activation-2.26e-02

Pos logit contributions
iqu+0.020
ridge+0.020
umes+0.019
angs+0.018
yles+0.018
Neg logit contributions
 day-0.007
,-0.007
 good-0.007
 girl-0.006
 end-0.006
Token -
Token ID532
Feature activation-3.40e-02

Pos logit contributions
iqu+0.073
ridge+0.071
umes+0.068
angs+0.068
rol+0.067
Neg logit contributions
 day-0.019
 good-0.017
 end-0.015
 girl-0.015
 team-0.015
Token ate
Token ID15063
Feature activation-3.94e-03

Pos logit contributions
 day+0.006
 good+0.005
,+0.005
 time+0.005
 end+0.005
Neg logit contributions
iqu-0.014
ridge-0.014
umes-0.013
yles-0.013
rol-0.013
Token them
Token ID606
Feature activation+2.96e-02

Pos logit contributions
iqu+0.011
ridge+0.010
yles+0.010
umes+0.010
rol+0.010
Neg logit contributions
 day-0.005
 team-0.004
 girl-0.004
 end-0.004
 boy-0.004
Token together
Token ID1978
Feature activation+4.02e-02

Pos logit contributions
 girl+0.038
 boy+0.036
 day+0.035
 end+0.033
 team+0.032
Neg logit contributions
ridge-0.074
rol-0.072
iqu-0.072
umes-0.071
vell-0.070
Token.
Token ID13
Feature activation+3.39e-03

Pos logit contributions
 girl+0.032
 day+0.030
 boy+0.029
 end+0.026
 team+0.025
Neg logit contributions
ridge-0.123
iqu-0.120
umes-0.119
yles-0.116
rol-0.115
Token The
Token ID383
Feature activation+7.61e-03

Pos logit contributions
 day+0.003
 good+0.003
 end+0.003
 girl+0.002
 team+0.002
Neg logit contributions
iqu-0.011
ridge-0.010
rol-0.010
umes-0.010
yles-0.010
Token p
Token ID279
Feature activation-7.36e-02

Pos logit contributions
 their+0.005
 They+0.004
 day+0.004
,+0.004
 and+0.004
Neg logit contributions
iqu-0.025
ridge-0.023
umes-0.023
rol-0.023
ocol-0.022
Tokenears
Token ID4127
Feature activation-2.36e-02

Pos logit contributions
iqu+0.116
ridge+0.109
umes+0.101
yles+0.100
rol+0.097
Neg logit contributions
 day-0.188
 good-0.181
,-0.180
 girl-0.176
 time-0.176
Token were
Token ID547
Feature activation-7.22e-03

Pos logit contributions
iqu+0.058
ridge+0.058
rol+0.055
umes+0.054
yles+0.053
Neg logit contributions
 day-0.035
 good-0.033
 time-0.032
 girl-0.032
 team-0.032
Token sweet
Token ID6029
Feature activation+4.09e-02

Pos logit contributions
iqu+0.022
ridge+0.021
umes+0.020
rol+0.020
yles+0.020
Neg logit contributions
 day-0.007
 girl-0.007
 team-0.006
,-0.006
 good-0.006
Token and
Token ID290
Feature activation-4.32e-03

Pos logit contributions
 day+0.022
 end+0.017
 good+0.017
 team+0.015
 their+0.015
Neg logit contributions
ridge-0.133
iqu-0.132
umes-0.125
rol-0.124
ous-0.124
Token juicy
Token ID35075
Feature activation+8.23e-03

Pos logit contributions
iqu+0.012
ridge+0.011
umes+0.011
yles+0.011
rol+0.011
Neg logit contributions
 day-0.006
 good-0.006
,-0.005
 time-0.005
 end-0.005
Token day
Token ID1110
Feature activation-1.72e-02

Pos logit contributions
 good+0.009
 They+0.009
 team+0.007
 girl+0.007
 their+0.006
Neg logit contributions
bar-0.056
iqu-0.054
ker-0.053
ous-0.053
cream-0.052
Token,
Token ID11
Feature activation-3.77e-02

Pos logit contributions
iqu+0.061
ridge+0.059
bar+0.058
yles+0.057
angs+0.057
Neg logit contributions
 day-0.006
 girl-0.005
 team-0.004
 end-0.004
 good-0.004
Token he
Token ID339
Feature activation+7.81e-04

Pos logit contributions
iqu+0.100
ridge+0.096
umes+0.092
yles+0.091
rol+0.090
Neg logit contributions
 day-0.058
,-0.054
 good-0.054
 time-0.052
 girl-0.051
Token lost
Token ID2626
Feature activation-2.44e-02

Pos logit contributions
 day+0.001
 good+0.001
 end+0.001
,+0.001
 girl+0.001
Neg logit contributions
iqu-0.002
ridge-0.002
yles-0.002
umes-0.002
bar-0.002
Token his
Token ID465
Feature activation-1.64e-02

Pos logit contributions
iqu+0.079
ridge+0.077
umes+0.074
yles+0.074
angs+0.073
Neg logit contributions
 day-0.020
 good-0.017
,-0.017
 team-0.016
 end-0.016
Token er
Token ID1931
Feature activation-8.49e-02

Pos logit contributions
iqu+0.050
ridge+0.048
umes+0.047
yles+0.046
angs+0.046
Neg logit contributions
 day-0.016
 good-0.015
,-0.015
 girl-0.014
 end-0.014
Tokenaser
Token ID6005
Feature activation-3.51e-02

Pos logit contributions
iqu+0.122
ridge+0.119
yles+0.104
umes+0.098
rol+0.096
Neg logit contributions
,-0.226
 and-0.223
 day-0.223
 girl-0.218
 team-0.217
Token while
Token ID981
Feature activation+3.44e-03

Pos logit contributions
iqu+0.119
ridge+0.115
umes+0.112
yles+0.111
rol+0.110
Neg logit contributions
 day-0.026
 good-0.022
,-0.021
 girl-0.020
 time-0.020
Token he
Token ID339
Feature activation-1.51e-02

Pos logit contributions
 day+0.005
,+0.005
 good+0.005
 and+0.005
 time+0.005
Neg logit contributions
iqu-0.009
ridge-0.009
umes-0.009
rol-0.008
angs-0.008
Token was
Token ID373
Feature activation-2.38e-02

Pos logit contributions
iqu+0.044
ridge+0.042
umes+0.041
yles+0.041
rol+0.040
Neg logit contributions
 day-0.019
,-0.018
 good-0.017
 and-0.017
 time-0.017
Token drawing
Token ID8263
Feature activation-1.38e-02

Pos logit contributions
iqu+0.066
ridge+0.064
umes+0.061
yles+0.061
rol+0.060
Neg logit contributions
 day-0.034
,-0.033
 good-0.033
 time-0.031
 and-0.031
Token But
Token ID887
Feature activation-1.86e-02

Pos logit contributions
iqu+0.043
ridge+0.042
umes+0.041
yles+0.041
rings+0.041
Neg logit contributions
 day-0.016
 good-0.015
 team-0.014
 end-0.013
 girl-0.013
Token one
Token ID530
Feature activation+1.01e-02

Pos logit contributions
iqu+0.062
ridge+0.060
umes+0.058
angs+0.058
rol+0.057
Neg logit contributions
 day-0.016
,-0.014
 good-0.014
 end-0.013
 time-0.013
Token day
Token ID1110
Feature activation-2.56e-02

Pos logit contributions
 They+0.004
 good+0.003
 their+0.002
 team+0.002
 town+0.002
Neg logit contributions
iqu-0.032
ridge-0.032
yles-0.031
angs-0.031
paste-0.031
Token,
Token ID11
Feature activation-4.38e-02

Pos logit contributions
iqu+0.092
ridge+0.090
umes+0.088
yles+0.088
rings+0.087
Neg logit contributions
 day-0.008
 team-0.006
 girl-0.006
 end-0.005
 good-0.005
Token she
Token ID673
Feature activation-9.60e-03

Pos logit contributions
iqu+0.130
ridge+0.125
umes+0.121
yles+0.120
rol+0.119
Neg logit contributions
 day-0.053
 good-0.049
,-0.049
 girl-0.046
 time-0.046
Token accidentally
Token ID14716
Feature activation-7.13e-02

Pos logit contributions
iqu+0.024
ridge+0.024
yles+0.022
umes+0.022
angs+0.022
Neg logit contributions
 day-0.015
 good-0.014
,-0.014
 end-0.013
 girl-0.013
Token knocked
Token ID13642
Feature activation-5.44e-02

Pos logit contributions
iqu+0.174
rol+0.158
ocol+0.156
yles+0.156
umes+0.156
Neg logit contributions
,-0.130
 day-0.126
 and-0.125
 good-0.123
 their-0.120
Token over
Token ID625
Feature activation+3.64e-03

Pos logit contributions
iqu+0.177
ridge+0.168
umes+0.161
vell+0.160
yles+0.159
Neg logit contributions
 day-0.062
,-0.055
 and-0.054
 good-0.054
 time-0.052
Token some
Token ID617
Feature activation-1.31e-02

Pos logit contributions
 day+0.003
 good+0.002
,+0.002
 end+0.002
 team+0.002
Neg logit contributions
iqu-0.012
ridge-0.012
umes-0.011
yles-0.011
angs-0.011
Token furniture
Token ID13091
Feature activation-2.60e-02

Pos logit contributions
iqu+0.036
ridge+0.034
umes+0.033
yles+0.033
rol+0.032
Neg logit contributions
 day-0.018
 good-0.017
,-0.017
 time-0.016
 end-0.016
Token.
Token ID13
Feature activation-3.34e-02

Pos logit contributions
iqu+0.098
ridge+0.094
yles+0.092
umes+0.092
rol+0.091
Neg logit contributions
 day-0.011
,-0.010
 and-0.009
 good-0.008
 end-0.007