TOP ACTIVATIONS
MAX = 0.111

This block is six inches long ." Ben does
treasure from a long time ago ." The
famous artist a long time ago ." The
This block is eight inches long . This block is six
it ! I have the highest tower ! I win !"
bigger . " Ha , ha , I did
had breakfast a little while ago , so I 'm not
foot . " Ha ha , you can 't
ground . " Ha ha , that 's fun
is from a long time ago , when people lived here
away . " Ha ha !" said Max .
, six , five , four , three , two ,
, seven , six , five , four , three ,
is from a long time ago , when people used different
queen from long , long ago . Mia listened and her
more . " Ha , ha . We are
cry . " Ha , ha , ha !
away . " Ha ha !" Tom laughs .
air . " Ha , I got your ball
sad . " Ha ha , I am better

MIN ACTIVATIONS
MIN = -0.100

They did not want to dress . " Ben
to play with their puppy , Spot . Spot is a
the woods when he heard someone shouting . He stopped and
good care of them and feed them and love them .
He always brought his bin ocular s , his sketch book
learned that sometimes things can look scary , but it 's
udd ling with their big , soft blanket . Every night
He did not want to cough and hurt anymore . He
to play with their dog , Spot . Spot is a
play outside with their dog , Spot . Spot is a
Mitt ens learned not to lie on the couch and play
also nervous about having to lie . He never used the
to breathe deeply and to cough when they needed to .
her that sometimes things can look scary but they 're not
if he could check her ears , her eyes , and
morning they would rush to the park together and have lots
. She also liked to dress him up with her old
because it could make him cough and hurt his lungs .
the animals with his bin ocular s and seeing how they
play outside with their dog , Max . One day ,

INTERVAL 0.058 - 0.111
CONTAINS 0.193%

, three , four , five , six , seven ,
and scared her . " Oh no , the bird scared
mug was missing . " Oh no ! Where is my
a thin bug on a l ily pad . Ho ppy
says . He puts his red block on top of Sam

INTERVAL 0.005 - 0.058
CONTAINS 39.274%

. Now you can practice and have more fun on your
. Everyone waved goodbye , and the yellow boat sailed off
day , the egg began to crack . The mom my
. " No , it is mine . You
the poor caul iflower and knew he had to rescue him

INTERVAL -0.048 - 0.005
CONTAINS 59.813%

water . She was surprised when the folder started to get
castle and saw a sad little girl .
for her help . She knew that her prayers were answered
What are you doing ?" A voice answers ,
and sprinkled some glitter . They looked at each

INTERVAL -0.100 - -0.048
CONTAINS 0.720%

me adow , making all the grass spark le .
on , Lily remembered to listen to her breathing and use
flowers . One day , Lily 's mom told her they
it 's going to rain or if there 's a big
't want to get ice cream there . The
Token This
Token ID770
Feature activation+7.21e-02

Pos logit contributions
pie+0.050
anish+0.050
 accomplishment+0.047
 seconds+0.046
 arrest+0.045
Neg logit contributions
s-0.049
 jog-0.043
ashing-0.041
aa-0.041
 especially-0.041
Token block
Token ID2512
Feature activation+4.98e-02

Pos logit contributions
anish+0.129
 arrest+0.129
 accomplishment+0.125
pie+0.118
 seconds+0.117
Neg logit contributions
s-0.123
 jog-0.089
 would-0.088
 especially-0.085
 hot-0.084
Token is
Token ID318
Feature activation+4.67e-02

Pos logit contributions
anish+0.096
 arrest+0.091
 seconds+0.090
 accomplishment+0.090
pie+0.090
Neg logit contributions
s-0.083
act-0.058
aa-0.057
udd-0.056
 jog-0.056
Token six
Token ID2237
Feature activation+7.62e-02

Pos logit contributions
anish+0.111
 arrest+0.099
pie+0.098
rum+0.092
 course+0.091
Neg logit contributions
s-0.054
 especially-0.048
 warm-0.044
 jog-0.043
 hot-0.042
Token inches
Token ID8331
Feature activation+4.57e-02

Pos logit contributions
anish+0.133
 accomplishment+0.112
pie+0.109
 arrest+0.109
ses+0.105
Neg logit contributions
s-0.135
 especially-0.120
act-0.108
 jog-0.103
 jackets-0.103
Token long
Token ID890
Feature activation+1.11e-01

Pos logit contributions
anish+0.103
pie+0.092
 course+0.088
 arrest+0.087
rum+0.084
Neg logit contributions
s-0.066
ashing-0.051
aa-0.050
act-0.049
 jog-0.049
Token."
Token ID526
Feature activation-3.01e-02

Pos logit contributions
anish+0.277
 accomplishment+0.259
pie+0.259
 course+0.254
nexpected+0.252
Neg logit contributions
s-0.090
 and-0.053
 jackets-0.048
 early-0.047
 especially-0.046
Token
Token ID198
Feature activation-4.14e-02

Pos logit contributions
ashing+0.052
s+0.051
ind+0.050
 sax+0.049
ling+0.049
Neg logit contributions
anish-0.041
 seconds-0.040
 eight-0.039
 nine-0.037
pie-0.037
Token
Token ID198
Feature activation-3.70e-02

Pos logit contributions
ashing+0.082
cases+0.079
ory+0.077
 trends+0.075
itionally+0.075
Neg logit contributions
anish-0.045
 seconds-0.044
 eight-0.044
 nine-0.042
 course-0.041
TokenBen
Token ID11696
Feature activation+2.09e-02

Pos logit contributions
ashing+0.064
cases+0.063
s+0.061
rig+0.060
udd+0.059
Neg logit contributions
 nine-0.051
anish-0.049
 eight-0.049
 seconds-0.047
 six-0.047
Token does
Token ID857
Feature activation+2.87e-02

Pos logit contributions
anish+0.039
pie+0.037
 seconds+0.035
 arrest+0.034
 accomplishment+0.034
Neg logit contributions
s-0.034
 jog-0.029
ashing-0.028
act-0.028
aa-0.027
Token treasure
Token ID14068
Feature activation+1.81e-02

Pos logit contributions
anish+0.025
pie+0.024
 seconds+0.022
 arrest+0.021
ps+0.021
Neg logit contributions
s-0.018
 jog-0.016
ashing-0.015
aa-0.015
act-0.015
Token from
Token ID422
Feature activation+1.76e-02

Pos logit contributions
anish+0.025
pie+0.024
 seconds+0.022
 eight+0.021
 arrest+0.020
Neg logit contributions
s-0.037
 jog-0.033
ashing-0.032
ory-0.032
aa-0.032
Token a
Token ID257
Feature activation+7.66e-04

Pos logit contributions
anish+0.031
pie+0.028
 seconds+0.027
rum+0.025
 arrest+0.025
Neg logit contributions
s-0.030
 jog-0.027
ashing-0.026
aa-0.025
ory-0.025
Token long
Token ID890
Feature activation+5.31e-02

Pos logit contributions
anish+0.001
pie+0.001
 seconds+0.001
 arrest+0.001
 eight+0.001
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token time
Token ID640
Feature activation+4.86e-02

Pos logit contributions
anish+0.109
pie+0.103
ps+0.096
rum+0.095
ocolate+0.092
Neg logit contributions
 jog-0.075
s-0.074
ory-0.065
aa-0.064
ashing-0.063
Token ago
Token ID2084
Feature activation+1.08e-01

Pos logit contributions
anish+0.073
pie+0.069
 seconds+0.061
 course+0.061
 accomplishment+0.059
Neg logit contributions
s-0.094
ashing-0.084
 jog-0.084
 especially-0.081
ory-0.080
Token."
Token ID526
Feature activation+1.16e-02

Pos logit contributions
anish+0.207
pie+0.204
 seconds+0.197
 accomplishment+0.191
 course+0.188
Neg logit contributions
s-0.151
 jog-0.115
 early-0.113
ory-0.112
ashing-0.112
Token
Token ID198
Feature activation-1.37e-02

Pos logit contributions
anish+0.025
pie+0.024
 seconds+0.023
 arrest+0.022
ra+0.022
Neg logit contributions
s-0.015
ashing-0.013
 jog-0.012
aa-0.012
ory-0.012
Token
Token ID220
Feature activation+5.17e-03

Pos logit contributions
s+0.021
ashing+0.020
act+0.019
 jog+0.019
aa+0.019
Neg logit contributions
anish-0.024
pie-0.022
 seconds-0.022
 course-0.021
 eight-0.021
Token
Token ID198
Feature activation-9.85e-03

Pos logit contributions
anish+0.011
pie+0.010
 seconds+0.010
 arrest+0.009
 course+0.009
Neg logit contributions
s-0.007
ashing-0.006
 jog-0.006
aa-0.006
act-0.006
TokenThe
Token ID464
Feature activation-1.19e-02

Pos logit contributions
s+0.015
ashing+0.014
aa+0.014
act+0.013
ory+0.013
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.017
 eight-0.016
 course-0.016
Token famous
Token ID5863
Feature activation+4.19e-02

Pos logit contributions
anish+0.002
pie+0.002
 seconds+0.002
 arrest+0.002
 eight+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.001
aa-0.001
act-0.001
Token artist
Token ID6802
Feature activation+6.42e-02

Pos logit contributions
anish+0.074
pie+0.067
rum+0.065
ps+0.065
 seconds+0.065
Neg logit contributions
s-0.067
 jog-0.064
 jazz-0.060
 sax-0.059
ory-0.059
Token a
Token ID257
Feature activation+1.31e-02

Pos logit contributions
anish+0.125
 seconds+0.115
pie+0.113
 accomplishment+0.111
 eight+0.109
Neg logit contributions
s-0.101
 jog-0.076
 especially-0.075
aa-0.075
ashing-0.073
Token long
Token ID890
Feature activation+7.01e-02

Pos logit contributions
anish+0.026
pie+0.025
 seconds+0.024
 arrest+0.023
rum+0.023
Neg logit contributions
s-0.019
 jog-0.016
ashing-0.016
aa-0.015
act-0.015
Token time
Token ID640
Feature activation+4.96e-02

Pos logit contributions
anish+0.134
pie+0.131
ps+0.126
 seconds+0.125
icked+0.122
Neg logit contributions
 jog-0.093
s-0.087
 jazz-0.081
aa-0.078
 sax-0.077
Token ago
Token ID2084
Feature activation+1.08e-01

Pos logit contributions
anish+0.087
pie+0.079
 course+0.074
 seconds+0.073
 accomplishment+0.071
Neg logit contributions
s-0.086
ashing-0.075
 jog-0.075
 especially-0.072
aa-0.071
Token."
Token ID526
Feature activation+2.30e-02

Pos logit contributions
anish+0.243
pie+0.237
 accomplishment+0.234
 course+0.232
 seconds+0.229
Neg logit contributions
s-0.113
ashing-0.079
 early-0.077
 jackets-0.075
 jog-0.075
Token
Token ID220
Feature activation+1.40e-02

Pos logit contributions
anish+0.045
pie+0.045
 seconds+0.042
 arrest+0.040
ra+0.040
Neg logit contributions
s-0.037
ashing-0.029
 jog-0.029
aa-0.028
udd-0.027
Token
Token ID198
Feature activation-1.55e-03

Pos logit contributions
anish+0.031
pie+0.030
 seconds+0.029
 arrest+0.027
 course+0.027
Neg logit contributions
s-0.018
ashing-0.014
 jog-0.014
aa-0.014
act-0.013
Token
Token ID198
Feature activation-5.57e-03

Pos logit contributions
s+0.002
ashing+0.002
 jog+0.002
aa+0.002
act+0.002
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.003
 course-0.003
 eight-0.003
TokenThe
Token ID464
Feature activation+2.94e-02

Pos logit contributions
s+0.008
ashing+0.007
aa+0.007
 jog+0.007
ory+0.007
Neg logit contributions
anish-0.011
pie-0.010
 seconds-0.010
 course-0.009
 eight-0.009
Token This
Token ID770
Feature activation+7.15e-02

Pos logit contributions
anish+0.058
pie+0.058
 accomplishment+0.054
 seconds+0.053
 arrest+0.053
Neg logit contributions
s-0.053
 jog-0.046
 especially-0.044
ashing-0.044
aa-0.044
Token block
Token ID2512
Feature activation+4.94e-02

Pos logit contributions
 arrest+0.128
anish+0.125
 accomplishment+0.124
pie+0.118
 seconds+0.116
Neg logit contributions
s-0.124
 jog-0.090
 would-0.088
 especially-0.086
 hot-0.083
Token is
Token ID318
Feature activation+4.51e-02

Pos logit contributions
anish+0.095
 arrest+0.091
pie+0.090
 seconds+0.090
 accomplishment+0.090
Neg logit contributions
s-0.082
act-0.056
aa-0.055
 jog-0.055
udd-0.055
Token eight
Token ID3624
Feature activation+8.45e-02

Pos logit contributions
anish+0.107
pie+0.096
 arrest+0.096
rum+0.090
 course+0.088
Neg logit contributions
s-0.052
 especially-0.045
 warm-0.042
 jog-0.041
 hot-0.039
Token inches
Token ID8331
Feature activation+3.46e-02

Pos logit contributions
anish+0.172
pie+0.147
 accomplishment+0.138
ceed+0.137
 course+0.137
Neg logit contributions
s-0.132
 especially-0.103
 jog-0.098
act-0.097
 hot-0.097
Token long
Token ID890
Feature activation+1.07e-01

Pos logit contributions
anish+0.078
pie+0.072
 arrest+0.067
 course+0.067
 seconds+0.066
Neg logit contributions
s-0.049
ashing-0.038
aa-0.037
act-0.036
 jog-0.036
Token.
Token ID13
Feature activation+2.97e-02

Pos logit contributions
anish+0.275
pie+0.263
 accomplishment+0.255
 course+0.245
nexpected+0.242
Neg logit contributions
s-0.095
 and-0.052
 early-0.047
 jackets-0.046
ashing-0.044
Token This
Token ID770
Feature activation+7.21e-02

Pos logit contributions
pie+0.050
anish+0.050
 accomplishment+0.047
 seconds+0.046
 arrest+0.045
Neg logit contributions
s-0.049
 jog-0.043
ashing-0.041
aa-0.041
 especially-0.041
Token block
Token ID2512
Feature activation+4.98e-02

Pos logit contributions
anish+0.129
 arrest+0.129
 accomplishment+0.125
pie+0.118
 seconds+0.117
Neg logit contributions
s-0.123
 jog-0.089
 would-0.088
 especially-0.085
 hot-0.084
Token is
Token ID318
Feature activation+4.67e-02

Pos logit contributions
anish+0.096
 arrest+0.091
 seconds+0.090
 accomplishment+0.090
pie+0.090
Neg logit contributions
s-0.083
act-0.058
aa-0.057
udd-0.056
 jog-0.056
Token six
Token ID2237
Feature activation+7.62e-02

Pos logit contributions
anish+0.111
 arrest+0.099
pie+0.098
rum+0.092
 course+0.091
Neg logit contributions
s-0.054
 especially-0.048
 warm-0.044
 jog-0.043
 hot-0.042
Token it
Token ID340
Feature activation+1.49e-02

Pos logit contributions
anish+0.085
pie+0.076
 seconds+0.071
rum+0.069
ASH+0.067
Neg logit contributions
s-0.062
 jog-0.053
 especially-0.051
 jazz-0.049
 early-0.046
Token!
Token ID0
Feature activation+2.53e-02

Pos logit contributions
anish+0.019
pie+0.017
 seconds+0.016
 arrest+0.015
 course+0.015
Neg logit contributions
s-0.033
 jog-0.029
ashing-0.029
aa-0.028
act-0.028
Token I
Token ID314
Feature activation+2.50e-02

Pos logit contributions
pie+0.042
anish+0.042
 seconds+0.038
 accomplishment+0.037
 course+0.037
Neg logit contributions
s-0.043
 jog-0.039
aa-0.038
ashing-0.037
 jazz-0.036
Token have
Token ID423
Feature activation+5.85e-02

Pos logit contributions
anish+0.045
pie+0.043
 seconds+0.040
 eight+0.039
 accomplishment+0.039
Neg logit contributions
s-0.041
 jog-0.036
ashing-0.034
aa-0.034
 especially-0.034
Token the
Token ID262
Feature activation+7.85e-02

Pos logit contributions
anish+0.119
pie+0.107
rum+0.105
 accomplishment+0.101
eros+0.096
Neg logit contributions
s-0.086
 especially-0.073
 jog-0.071
 jazz-0.066
 jackets-0.065
Token highest
Token ID4511
Feature activation+1.06e-01

Pos logit contributions
rum+0.161
anish+0.161
 arrest+0.150
 seconds+0.145
nexpected+0.144
Neg logit contributions
 especially-0.088
 jog-0.083
s-0.083
 warm-0.081
 hot-0.080
Token tower
Token ID10580
Feature activation+5.47e-02

Pos logit contributions
anish+0.176
rum+0.160
ps+0.156
ped+0.155
pie+0.152
Neg logit contributions
s-0.161
 especially-0.153
 jog-0.151
 early-0.136
 hot-0.136
Token!
Token ID0
Feature activation+2.97e-02

Pos logit contributions
anish+0.083
pie+0.073
 seconds+0.070
 accomplishment+0.067
 eight+0.066
Neg logit contributions
s-0.114
ashing-0.091
 jog-0.091
ding-0.089
 especially-0.089
Token I
Token ID314
Feature activation+2.88e-02

Pos logit contributions
pie+0.050
anish+0.050
 seconds+0.045
 accomplishment+0.045
 course+0.044
Neg logit contributions
s-0.050
 jog-0.044
aa-0.043
ashing-0.042
 jazz-0.041
Token win
Token ID1592
Feature activation+5.85e-02

Pos logit contributions
anish+0.053
pie+0.051
 seconds+0.047
 eight+0.046
 accomplishment+0.046
Neg logit contributions
s-0.047
 jog-0.041
ashing-0.039
aa-0.038
 especially-0.038
Token!"
Token ID2474
Feature activation+2.93e-03

Pos logit contributions
anish+0.097
pie+0.088
 arrest+0.081
 accomplishment+0.080
rum+0.079
Neg logit contributions
s-0.106
 especially-0.087
 jog-0.085
 early-0.082
 jazz-0.082
Token bigger
Token ID5749
Feature activation+3.37e-02

Pos logit contributions
anish+0.008
pie+0.007
 seconds+0.007
 arrest+0.007
 eight+0.007
Neg logit contributions
s-0.006
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
Token.
Token ID13
Feature activation+9.12e-03

Pos logit contributions
anish+0.071
pie+0.065
 arrest+0.060
 seconds+0.059
 course+0.059
Neg logit contributions
s-0.048
 jog-0.038
ashing-0.037
aa-0.037
act-0.037
Token
Token ID198
Feature activation+8.90e-03

Pos logit contributions
anish+0.019
pie+0.018
 seconds+0.017
 arrest+0.016
 course+0.016
Neg logit contributions
s-0.013
ashing-0.011
 jog-0.011
aa-0.011
act-0.010
Token
Token ID198
Feature activation+1.10e-02

Pos logit contributions
anish+0.019
pie+0.018
 seconds+0.017
 arrest+0.017
 course+0.016
Neg logit contributions
s-0.012
 jog-0.010
ashing-0.010
aa-0.009
act-0.009
Token"
Token ID1
Feature activation+2.21e-02

Pos logit contributions
anish+0.020
pie+0.019
 seconds+0.018
 arrest+0.017
 accomplishment+0.017
Neg logit contributions
s-0.018
 jog-0.015
ashing-0.015
aa-0.015
act-0.014
TokenHa
Token ID23303
Feature activation+1.06e-01

Pos logit contributions
anish+0.044
pie+0.042
 arrest+0.041
 accomplishment+0.041
rum+0.040
Neg logit contributions
s-0.029
 jog-0.026
 especially-0.024
aa-0.024
ashing-0.024
Token,
Token ID11
Feature activation+3.86e-02

Pos logit contributions
aces+0.185
alks+0.183
pie+0.169
 Glass+0.167
alk+0.167
Neg logit contributions
 jog-0.140
aa-0.124
s-0.122
 sheep-0.112
 j-0.107
Token ha
Token ID387
Feature activation+6.99e-02

Pos logit contributions
anish+0.070
pie+0.066
 seconds+0.061
 accomplishment+0.060
rum+0.059
Neg logit contributions
s-0.058
 jog-0.057
aa-0.052
ory-0.049
 especially-0.048
Token,
Token ID11
Feature activation+3.45e-02

Pos logit contributions
anish+0.132
 accomplishment+0.125
pie+0.124
 itself+0.122
 amulet+0.119
Neg logit contributions
s-0.116
aa-0.100
udd-0.090
ory-0.089
 jog-0.087
Token I
Token ID314
Feature activation+2.35e-02

Pos logit contributions
anish+0.063
pie+0.060
 seconds+0.057
 accomplishment+0.055
rum+0.055
Neg logit contributions
s-0.050
 jog-0.050
aa-0.045
ory-0.042
 especially-0.042
Token did
Token ID750
Feature activation+2.83e-02

Pos logit contributions
anish+0.047
pie+0.045
 seconds+0.042
 accomplishment+0.041
 arrest+0.040
Neg logit contributions
s-0.034
 jog-0.030
aa-0.029
ashing-0.028
 especially-0.028
Token had
Token ID550
Feature activation+3.12e-02

Pos logit contributions
anish+0.004
pie+0.003
 seconds+0.003
 arrest+0.003
 eight+0.003
Neg logit contributions
s-0.003
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token breakfast
Token ID12607
Feature activation+1.29e-02

Pos logit contributions
anish+0.059
pie+0.056
 seconds+0.050
rum+0.050
 accomplishment+0.050
Neg logit contributions
s-0.050
 jog-0.042
 especially-0.041
aa-0.040
ory-0.040
Token a
Token ID257
Feature activation+5.89e-03

Pos logit contributions
anish+0.020
pie+0.019
 seconds+0.018
 arrest+0.017
 eight+0.016
Neg logit contributions
s-0.024
ashing-0.021
 jog-0.021
aa-0.021
ory-0.021
Token little
Token ID1310
Feature activation+2.26e-02

Pos logit contributions
anish+0.011
pie+0.011
 seconds+0.010
 arrest+0.010
 course+0.010
Neg logit contributions
s-0.009
 jog-0.008
ashing-0.008
aa-0.008
act-0.008
Token while
Token ID981
Feature activation+3.06e-02

Pos logit contributions
anish+0.041
pie+0.040
 seconds+0.036
 arrest+0.035
ps+0.035
Neg logit contributions
s-0.037
 jog-0.032
ashing-0.031
aa-0.031
ory-0.030
Token ago
Token ID2084
Feature activation+1.05e-01

Pos logit contributions
anish+0.054
pie+0.051
 seconds+0.046
 course+0.045
 arrest+0.045
Neg logit contributions
s-0.052
 jog-0.046
ashing-0.044
aa-0.044
ory-0.044
Token,
Token ID11
Feature activation-1.78e-02

Pos logit contributions
anish+0.209
pie+0.205
 seconds+0.197
 fireball+0.191
 accomplishment+0.186
Neg logit contributions
s-0.143
 jog-0.110
ory-0.106
 especially-0.104
 early-0.104
Token so
Token ID523
Feature activation+1.53e-02

Pos logit contributions
s+0.032
ashing+0.029
act+0.028
aa+0.028
udd+0.027
Neg logit contributions
anish-0.029
 seconds-0.027
pie-0.027
 eight-0.026
 course-0.025
Token I
Token ID314
Feature activation+1.39e-03

Pos logit contributions
anish+0.026
pie+0.025
 seconds+0.023
 arrest+0.022
 course+0.022
Neg logit contributions
s-0.027
 jog-0.024
aa-0.023
ashing-0.023
ory-0.022
Token'm
Token ID1101
Feature activation+2.82e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.002
 arrest+0.002
 course+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token not
Token ID407
Feature activation+3.67e-02

Pos logit contributions
anish+0.062
pie+0.055
 seconds+0.052
 course+0.051
 arrest+0.051
Neg logit contributions
s-0.035
 especially-0.033
 jog-0.033
ashing-0.033
aa-0.031
Token foot
Token ID2366
Feature activation+2.52e-02

Pos logit contributions
anish+0.043
pie+0.040
 seconds+0.038
 course+0.037
 arrest+0.036
Neg logit contributions
s-0.046
 jog-0.039
ashing-0.036
aa-0.036
 sax-0.035
Token.
Token ID13
Feature activation-2.47e-03

Pos logit contributions
anish+0.052
pie+0.049
 seconds+0.047
 arrest+0.046
 accomplishment+0.045
Neg logit contributions
s-0.037
ashing-0.030
 jog-0.030
aa-0.028
 especially-0.027
Token
Token ID198
Feature activation-4.24e-03

Pos logit contributions
s+0.004
ashing+0.003
 jog+0.003
act+0.003
aa+0.003
Neg logit contributions
anish-0.005
pie-0.004
 seconds-0.004
 eight-0.004
 arrest-0.004
Token
Token ID198
Feature activation-7.43e-04

Pos logit contributions
s+0.006
ashing+0.005
 jog+0.005
act+0.005
aa+0.005
Neg logit contributions
anish-0.008
pie-0.008
 seconds-0.008
 eight-0.007
 arrest-0.007
Token"
Token ID1
Feature activation-1.44e-03

Pos logit contributions
s+0.001
ashing+0.001
 jog+0.001
act+0.001
aa+0.001
Neg logit contributions
anish-0.001
pie-0.001
 seconds-0.001
 arrest-0.001
 eight-0.001
TokenHa
Token ID23303
Feature activation+1.05e-01

Pos logit contributions
s+0.002
ashing+0.002
aa+0.002
 jog+0.002
act+0.002
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.002
 course-0.002
 arrest-0.002
Token ha
Token ID387
Feature activation+8.36e-02

Pos logit contributions
aces+0.185
alks+0.181
stack+0.171
pie+0.171
alk+0.168
Neg logit contributions
 jog-0.153
aa-0.134
s-0.134
 j-0.118
 march-0.116
Token,
Token ID11
Feature activation+3.59e-02

Pos logit contributions
 itself+0.163
pie+0.163
anish+0.162
 accomplishment+0.157
ps+0.157
Neg logit contributions
s-0.136
aa-0.111
ory-0.101
udd-0.095
 jog-0.095
Token you
Token ID345
Feature activation+4.80e-02

Pos logit contributions
anish+0.065
pie+0.064
 seconds+0.060
rum+0.058
 accomplishment+0.058
Neg logit contributions
 jog-0.054
s-0.053
aa-0.045
 jazz-0.045
 sax-0.045
Token can
Token ID460
Feature activation+2.46e-02

Pos logit contributions
pie+0.095
anish+0.092
 seconds+0.091
 accomplishment+0.091
rum+0.089
Neg logit contributions
s-0.071
 jog-0.062
udd-0.053
ory-0.053
 especially-0.051
Token't
Token ID470
Feature activation+3.29e-02

Pos logit contributions
anish+0.041
 seconds+0.038
pie+0.038
rum+0.034
 accomplishment+0.034
Neg logit contributions
s-0.046
 jog-0.042
 especially-0.037
ashing-0.037
aa-0.036
Token ground
Token ID2323
Feature activation+2.72e-02

Pos logit contributions
anish+0.006
pie+0.006
 seconds+0.006
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.006
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
Token.
Token ID13
Feature activation+1.59e-02

Pos logit contributions
anish+0.058
pie+0.055
 seconds+0.052
 arrest+0.051
 course+0.051
Neg logit contributions
s-0.039
ashing-0.031
 jog-0.030
aa-0.028
act-0.027
Token
Token ID198
Feature activation+1.56e-03

Pos logit contributions
anish+0.032
pie+0.031
 seconds+0.029
 arrest+0.028
 course+0.028
Neg logit contributions
s-0.023
ashing-0.019
 jog-0.019
aa-0.019
act-0.018
Token
Token ID198
Feature activation+1.53e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 arrest+0.003
 eight+0.003
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token"
Token ID1
Feature activation+1.84e-03

Pos logit contributions
anish+0.024
pie+0.024
 seconds+0.022
 accomplishment+0.021
 arrest+0.021
Neg logit contributions
s-0.029
 jog-0.025
ashing-0.025
aa-0.024
act-0.023
TokenHa
Token ID23303
Feature activation+1.04e-01

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 course+0.003
 eight+0.003
Neg logit contributions
s-0.003
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token ha
Token ID387
Feature activation+7.97e-02

Pos logit contributions
aces+0.195
alks+0.187
pie+0.183
 itself+0.178
alk+0.176
Neg logit contributions
 jog-0.139
aa-0.127
s-0.126
 j-0.110
 hot-0.107
Token,
Token ID11
Feature activation+4.10e-02

Pos logit contributions
anish+0.145
pie+0.145
 itself+0.142
 accomplishment+0.139
ps+0.136
Neg logit contributions
s-0.143
aa-0.125
ory-0.111
 jog-0.106
ashing-0.105
Token that
Token ID326
Feature activation+1.50e-02

Pos logit contributions
anish+0.083
pie+0.082
rum+0.079
aces+0.079
 accomplishment+0.077
Neg logit contributions
 jog-0.051
s-0.050
aa-0.043
 jazz-0.041
 sheep-0.041
Token's
Token ID338
Feature activation+1.87e-02

Pos logit contributions
anish+0.030
pie+0.028
 seconds+0.027
 accomplishment+0.026
 arrest+0.026
Neg logit contributions
s-0.023
 jog-0.019
ashing-0.018
aa-0.018
act-0.018
Token fun
Token ID1257
Feature activation+2.47e-02

Pos logit contributions
anish+0.035
pie+0.033
 arrest+0.030
 seconds+0.030
rum+0.029
Neg logit contributions
s-0.029
 jog-0.026
ashing-0.025
aa-0.025
 especially-0.025
Token is
Token ID318
Feature activation+2.01e-02

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 arrest+0.000
 course+0.000
Neg logit contributions
s-0.000
ashing-0.000
 jog-0.000
aa-0.000
act-0.000
Token from
Token ID422
Feature activation+2.54e-02

Pos logit contributions
anish+0.038
pie+0.035
 seconds+0.032
 arrest+0.032
 course+0.032
Neg logit contributions
s-0.032
 jog-0.028
ashing-0.027
act-0.026
aa-0.026
Token a
Token ID257
Feature activation+1.47e-02

Pos logit contributions
anish+0.046
pie+0.042
 seconds+0.039
rum+0.038
 course+0.037
Neg logit contributions
s-0.040
 jog-0.037
ashing-0.035
aa-0.035
 especially-0.035
Token long
Token ID890
Feature activation+7.31e-02

Pos logit contributions
anish+0.025
pie+0.023
 seconds+0.022
 arrest+0.021
ps+0.021
Neg logit contributions
s-0.026
 jog-0.023
ashing-0.022
aa-0.022
ory-0.022
Token time
Token ID640
Feature activation+4.11e-02

Pos logit contributions
anish+0.145
ps+0.135
pie+0.133
rum+0.133
oor+0.131
Neg logit contributions
 jog-0.100
s-0.086
 sheep-0.080
aa-0.080
 jacket-0.078
Token ago
Token ID2084
Feature activation+1.04e-01

Pos logit contributions
anish+0.057
pie+0.051
 course+0.046
 accomplishment+0.044
 seconds+0.044
Neg logit contributions
s-0.086
ashing-0.077
 jog-0.076
 especially-0.074
aa-0.073
Token,
Token ID11
Feature activation-5.92e-03

Pos logit contributions
anish+0.227
pie+0.223
 accomplishment+0.219
 course+0.214
nexpected+0.214
Neg logit contributions
s-0.116
 jackets-0.086
 early-0.079
 jog-0.078
ashing-0.073
Token when
Token ID618
Feature activation+1.88e-02

Pos logit contributions
s+0.009
ashing+0.008
aa+0.008
act+0.008
 jog+0.008
Neg logit contributions
anish-0.011
pie-0.011
 seconds-0.011
 eight-0.010
 arrest-0.010
Token people
Token ID661
Feature activation-2.72e-03

Pos logit contributions
anish+0.035
pie+0.033
 seconds+0.030
rum+0.030
 course+0.029
Neg logit contributions
s-0.030
 jog-0.027
aa-0.025
ashing-0.025
act-0.024
Token lived
Token ID5615
Feature activation+6.37e-03

Pos logit contributions
s+0.004
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.005
 arrest-0.004
 course-0.004
Token here
Token ID994
Feature activation+7.80e-03

Pos logit contributions
anish+0.012
pie+0.011
 seconds+0.011
 arrest+0.010
 course+0.010
Neg logit contributions
s-0.010
ashing-0.009
 jog-0.009
aa-0.009
act-0.008
Token away
Token ID1497
Feature activation+3.25e-02

Pos logit contributions
anish+0.033
pie+0.031
 seconds+0.029
 course+0.028
 eight+0.028
Neg logit contributions
s-0.017
 jog-0.014
ashing-0.013
aa-0.013
ory-0.012
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
anish+0.076
pie+0.069
 course+0.067
 seconds+0.065
ra+0.064
Neg logit contributions
s-0.044
 jog-0.033
ashing-0.033
 early-0.032
aa-0.031
Token
Token ID198
Feature activation+2.99e-03

Pos logit contributions
anish+0.028
pie+0.027
 seconds+0.025
 course+0.024
 arrest+0.024
Neg logit contributions
s-0.019
ashing-0.016
 jog-0.016
aa-0.015
ory-0.015
Token
Token ID198
Feature activation+3.61e-03

Pos logit contributions
anish+0.006
pie+0.006
 seconds+0.006
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.004
ashing-0.004
 jog-0.003
aa-0.003
act-0.003
Token"
Token ID1
Feature activation+1.28e-02

Pos logit contributions
anish+0.007
pie+0.007
 seconds+0.006
 arrest+0.006
 eight+0.006
Neg logit contributions
s-0.006
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
TokenHa
Token ID23303
Feature activation+1.03e-01

Pos logit contributions
anish+0.024
pie+0.023
 seconds+0.022
 accomplishment+0.022
 eight+0.021
Neg logit contributions
s-0.019
 jog-0.017
ashing-0.016
aa-0.016
act-0.016
Token ha
Token ID387
Feature activation+8.18e-02

Pos logit contributions
aces+0.188
alks+0.186
alk+0.171
bons+0.170
pie+0.169
Neg logit contributions
 jog-0.136
s-0.127
aa-0.119
 sheep-0.107
 j-0.105
Token!"
Token ID2474
Feature activation+2.01e-02

Pos logit contributions
anish+0.154
 itself+0.150
pie+0.149
 accomplishment+0.147
ps+0.144
Neg logit contributions
s-0.139
aa-0.118
ory-0.106
udd-0.100
ashing-0.099
Token said
Token ID531
Feature activation+2.31e-02

Pos logit contributions
anish+0.041
pie+0.039
 seconds+0.036
 accomplishment+0.035
rum+0.035
Neg logit contributions
s-0.030
 jog-0.025
aa-0.024
ashing-0.023
ory-0.023
Token Max
Token ID5436
Feature activation+2.35e-02

Pos logit contributions
anish+0.044
pie+0.041
 seconds+0.039
 course+0.038
 arrest+0.038
Neg logit contributions
s-0.035
 jog-0.030
aa-0.030
ashing-0.029
ory-0.029
Token.
Token ID13
Feature activation+2.80e-02

Pos logit contributions
anish+0.048
pie+0.044
 seconds+0.043
 course+0.042
 arrest+0.041
Neg logit contributions
s-0.036
ashing-0.029
 jog-0.029
aa-0.028
ory-0.027
Token,
Token ID11
Feature activation+6.43e-02

Pos logit contributions
anish+0.220
pie+0.205
stack+0.199
rum+0.195
ceed+0.188
Neg logit contributions
s-0.100
 jog-0.068
 especially-0.060
 jackets-0.058
 early-0.058
Token six
Token ID2237
Feature activation+9.36e-02

Pos logit contributions
anish+0.071
rum+0.059
pie+0.059
 accomplishment+0.057
 powers+0.045
Neg logit contributions
 jog-0.146
s-0.144
aa-0.130
 sheep-0.129
 ranc-0.124
Token,
Token ID11
Feature activation+6.85e-02

Pos logit contributions
anish+0.201
pie+0.187
stack+0.177
ocolate+0.176
ses+0.176
Neg logit contributions
s-0.131
 jog-0.085
 especially-0.084
 jackets-0.082
 early-0.078
Token five
Token ID1936
Feature activation+1.03e-01

Pos logit contributions
anish+0.107
rum+0.093
pie+0.086
 accomplishment+0.082
ocolate+0.076
Neg logit contributions
 jog-0.125
s-0.125
 warm-0.111
 sheep-0.111
 especially-0.109
Token,
Token ID11
Feature activation+7.47e-02

Pos logit contributions
anish+0.220
pie+0.190
ocolate+0.189
stack+0.189
ASH+0.182
Neg logit contributions
s-0.163
 especially-0.106
 jog-0.100
 jackets-0.099
ling-0.099
Token four
Token ID1440
Feature activation+1.03e-01

Pos logit contributions
anish+0.113
rum+0.101
pie+0.090
 accomplishment+0.089
ocolate+0.079
Neg logit contributions
 jog-0.139
s-0.136
 especially-0.127
 warm-0.121
 sheep-0.121
Token,
Token ID11
Feature activation+7.34e-02

Pos logit contributions
anish+0.211
pie+0.188
ocolate+0.185
ses+0.184
stack+0.184
Neg logit contributions
s-0.170
 especially-0.108
 jackets-0.104
 jog-0.102
cases-0.095
Token three
Token ID1115
Feature activation+9.15e-02

Pos logit contributions
anish+0.111
rum+0.099
pie+0.089
 accomplishment+0.085
nexpected+0.080
Neg logit contributions
 jog-0.138
s-0.133
 especially-0.127
 warm-0.120
 sheep-0.120
Token,
Token ID11
Feature activation+7.46e-02

Pos logit contributions
anish+0.183
pie+0.164
ses+0.160
stack+0.159
ACK+0.158
Neg logit contributions
s-0.159
 especially-0.106
 jackets-0.101
ling-0.099
 jog-0.096
Token two
Token ID734
Feature activation+8.18e-02

Pos logit contributions
anish+0.109
rum+0.097
pie+0.091
 accomplishment+0.080
 arrest+0.079
Neg logit contributions
 jog-0.142
s-0.138
 especially-0.131
 warm-0.126
 sheep-0.123
Token,
Token ID11
Feature activation+7.52e-02

Pos logit contributions
anish+0.156
pie+0.143
ocolate+0.130
 arrest+0.128
 accomplishment+0.128
Neg logit contributions
s-0.147
 jog-0.104
 especially-0.102
cases-0.095
 warm-0.093
Token,
Token ID11
Feature activation+5.43e-02

Pos logit contributions
anish+0.229
pie+0.205
ocolate+0.193
ceed+0.192
stack+0.188
Neg logit contributions
s-0.126
 jog-0.089
 jackets-0.076
 sheep-0.073
 early-0.072
Token seven
Token ID3598
Feature activation+9.02e-02

Pos logit contributions
anish+0.064
rum+0.058
pie+0.057
 accomplishment+0.049
 tal+0.045
Neg logit contributions
 jog-0.120
s-0.115
 especially-0.108
 warm-0.108
 sheep-0.104
Token,
Token ID11
Feature activation+6.43e-02

Pos logit contributions
anish+0.220
pie+0.205
stack+0.199
rum+0.195
ceed+0.188
Neg logit contributions
s-0.100
 jog-0.068
 especially-0.060
 jackets-0.058
 early-0.058
Token six
Token ID2237
Feature activation+9.36e-02

Pos logit contributions
anish+0.071
rum+0.059
pie+0.059
 accomplishment+0.057
 powers+0.045
Neg logit contributions
 jog-0.146
s-0.144
aa-0.130
 sheep-0.129
 ranc-0.124
Token,
Token ID11
Feature activation+6.85e-02

Pos logit contributions
anish+0.201
pie+0.187
stack+0.177
ocolate+0.176
ses+0.176
Neg logit contributions
s-0.131
 jog-0.085
 especially-0.084
 jackets-0.082
 early-0.078
Token five
Token ID1936
Feature activation+1.03e-01

Pos logit contributions
anish+0.107
rum+0.093
pie+0.086
 accomplishment+0.082
ocolate+0.076
Neg logit contributions
 jog-0.125
s-0.125
 warm-0.111
 sheep-0.111
 especially-0.109
Token,
Token ID11
Feature activation+7.47e-02

Pos logit contributions
anish+0.220
pie+0.190
ocolate+0.189
stack+0.189
ASH+0.182
Neg logit contributions
s-0.163
 especially-0.106
 jog-0.100
 jackets-0.099
ling-0.099
Token four
Token ID1440
Feature activation+1.03e-01

Pos logit contributions
anish+0.113
rum+0.101
pie+0.090
 accomplishment+0.089
ocolate+0.079
Neg logit contributions
 jog-0.139
s-0.136
 especially-0.127
 warm-0.121
 sheep-0.121
Token,
Token ID11
Feature activation+7.34e-02

Pos logit contributions
anish+0.211
pie+0.188
ocolate+0.185
ses+0.184
stack+0.184
Neg logit contributions
s-0.170
 especially-0.108
 jackets-0.104
 jog-0.102
cases-0.095
Token three
Token ID1115
Feature activation+9.15e-02

Pos logit contributions
anish+0.111
rum+0.099
pie+0.089
 accomplishment+0.085
nexpected+0.080
Neg logit contributions
 jog-0.138
s-0.133
 especially-0.127
 warm-0.120
 sheep-0.120
Token,
Token ID11
Feature activation+7.46e-02

Pos logit contributions
anish+0.183
pie+0.164
ses+0.160
stack+0.159
ACK+0.158
Neg logit contributions
s-0.159
 especially-0.106
 jackets-0.101
ling-0.099
 jog-0.096
Token is
Token ID318
Feature activation+1.76e-02

Pos logit contributions
anish+0.009
pie+0.009
 seconds+0.008
 arrest+0.008
 eight+0.008
Neg logit contributions
s-0.009
ashing-0.008
 jog-0.008
aa-0.008
act-0.008
Token from
Token ID422
Feature activation+2.97e-02

Pos logit contributions
anish+0.037
pie+0.034
 arrest+0.032
 seconds+0.032
 course+0.031
Neg logit contributions
s-0.024
 jog-0.021
ashing-0.020
 especially-0.020
ory-0.019
Token a
Token ID257
Feature activation+1.25e-02

Pos logit contributions
anish+0.055
pie+0.050
 seconds+0.046
rum+0.046
 arrest+0.044
Neg logit contributions
s-0.045
 jog-0.042
 especially-0.040
ashing-0.040
ory-0.040
Token long
Token ID890
Feature activation+7.50e-02

Pos logit contributions
anish+0.023
pie+0.021
 seconds+0.020
 arrest+0.019
 course+0.019
Neg logit contributions
s-0.020
 jog-0.018
ashing-0.017
aa-0.017
ory-0.017
Token time
Token ID640
Feature activation+5.19e-02

Pos logit contributions
anish+0.146
pie+0.141
ps+0.137
rum+0.131
oor+0.129
Neg logit contributions
 jog-0.104
s-0.092
 jacket-0.089
ory-0.088
 sheep-0.084
Token ago
Token ID2084
Feature activation+1.03e-01

Pos logit contributions
anish+0.080
pie+0.073
 course+0.064
 seconds+0.061
rum+0.061
Neg logit contributions
s-0.102
ashing-0.093
 jog-0.089
 especially-0.087
 early-0.085
Token,
Token ID11
Feature activation-6.03e-03

Pos logit contributions
anish+0.249
pie+0.246
 course+0.235
 accomplishment+0.233
 fireball+0.232
Neg logit contributions
s-0.094
 early-0.065
ashing-0.059
 jackets-0.058
 jog-0.056
Token when
Token ID618
Feature activation+1.31e-02

Pos logit contributions
s+0.008
ashing+0.007
aa+0.007
 jog+0.007
act+0.007
Neg logit contributions
anish-0.012
pie-0.012
 seconds-0.011
 eight-0.011
 course-0.011
Token people
Token ID661
Feature activation-4.85e-04

Pos logit contributions
anish+0.023
pie+0.022
 seconds+0.020
 arrest+0.019
 course+0.019
Neg logit contributions
s-0.022
 jog-0.019
ashing-0.019
aa-0.019
act-0.018
Token used
Token ID973
Feature activation+4.11e-02

Pos logit contributions
s+0.001
ashing+0.001
 jog+0.001
aa+0.001
act+0.001
Neg logit contributions
anish-0.001
pie-0.001
 seconds-0.001
 arrest-0.001
 course-0.001
Token different
Token ID1180
Feature activation+3.41e-02

Pos logit contributions
anish+0.084
pie+0.079
 seconds+0.071
 arrest+0.069
rum+0.069
Neg logit contributions
s-0.063
 jog-0.048
 jackets-0.047
aa-0.046
act-0.045
Token queen
Token ID16599
Feature activation+4.63e-02

Pos logit contributions
anish+0.062
pie+0.057
 course+0.052
rum+0.051
ps+0.051
Neg logit contributions
s-0.056
 jog-0.053
ory-0.052
aa-0.050
ashing-0.050
Token from
Token ID422
Feature activation+2.02e-02

Pos logit contributions
anish+0.105
 seconds+0.094
pie+0.094
 accomplishment+0.093
 arrest+0.091
Neg logit contributions
s-0.061
 jog-0.045
 especially-0.041
ashing-0.041
ory-0.040
Token long
Token ID890
Feature activation+6.20e-02

Pos logit contributions
anish+0.039
pie+0.035
 seconds+0.033
rum+0.032
 arrest+0.032
Neg logit contributions
s-0.030
 jog-0.028
ashing-0.027
aa-0.026
ory-0.026
Token,
Token ID11
Feature activation+1.37e-03

Pos logit contributions
anish+0.121
pie+0.107
oor+0.106
ps+0.105
rum+0.103
Neg logit contributions
s-0.093
 jog-0.091
aa-0.076
 especially-0.074
 café-0.074
Token long
Token ID890
Feature activation+6.32e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 arrest+0.002
 eight+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token ago
Token ID2084
Feature activation+1.02e-01

Pos logit contributions
anish+0.112
pie+0.097
oor+0.096
ocolate+0.093
rum+0.093
Neg logit contributions
s-0.102
 jog-0.098
 early-0.085
 especially-0.084
aa-0.084
Token.
Token ID13
Feature activation+2.67e-02

Pos logit contributions
anish+0.248
pie+0.239
 course+0.233
 fireball+0.231
elve+0.229
Neg logit contributions
s-0.091
 early-0.063
 jackets-0.059
 jog-0.058
 especially-0.055
Token Mia
Token ID32189
Feature activation+2.43e-02

Pos logit contributions
anish+0.054
pie+0.053
 arrest+0.049
ra+0.048
 course+0.047
Neg logit contributions
s-0.039
ashing-0.034
 jog-0.032
aa-0.031
ory-0.030
Token listened
Token ID16399
Feature activation-1.82e-02

Pos logit contributions
anish+0.050
pie+0.047
 seconds+0.045
 arrest+0.044
rum+0.044
Neg logit contributions
s-0.035
 jog-0.029
 especially-0.027
ory-0.026
act-0.026
Token and
Token ID290
Feature activation+1.59e-02

Pos logit contributions
act+0.023
s+0.023
 jog+0.022
aa+0.022
ashing+0.022
Neg logit contributions
anish-0.034
 seconds-0.033
pie-0.033
 eight-0.032
 accomplishment-0.032
Token her
Token ID607
Feature activation+1.87e-02

Pos logit contributions
anish+0.030
pie+0.029
 seconds+0.027
rum+0.026
 arrest+0.026
Neg logit contributions
s-0.025
 jog-0.022
ashing-0.021
aa-0.020
ory-0.020
Token more
Token ID517
Feature activation+3.55e-02

Pos logit contributions
anish+0.054
pie+0.050
ra+0.047
rum+0.046
 arrest+0.045
Neg logit contributions
s-0.043
 jog-0.039
ashing-0.035
aa-0.035
act-0.034
Token.
Token ID13
Feature activation+5.28e-03

Pos logit contributions
anish+0.086
pie+0.075
 course+0.072
ra+0.071
ocolate+0.070
Neg logit contributions
s-0.045
 jog-0.035
ashing-0.034
aa-0.032
 early-0.032
Token
Token ID198
Feature activation-6.13e-04

Pos logit contributions
anish+0.010
pie+0.010
 seconds+0.009
 arrest+0.009
 course+0.009
Neg logit contributions
s-0.008
ashing-0.007
 jog-0.007
aa-0.007
act-0.007
Token
Token ID198
Feature activation+6.27e-03

Pos logit contributions
s+0.001
ashing+0.001
 jog+0.001
aa+0.001
act+0.001
Neg logit contributions
anish-0.001
pie-0.001
 seconds-0.001
 eight-0.001
 arrest-0.001
Token"
Token ID1
Feature activation+1.07e-02

Pos logit contributions
anish+0.012
pie+0.012
 seconds+0.011
 arrest+0.011
 course+0.010
Neg logit contributions
s-0.009
 jog-0.008
ashing-0.008
aa-0.008
act-0.008
TokenHa
Token ID23303
Feature activation+1.02e-01

Pos logit contributions
anish+0.020
pie+0.019
 seconds+0.018
 accomplishment+0.017
 arrest+0.017
Neg logit contributions
s-0.017
 jog-0.015
ashing-0.014
aa-0.014
act-0.014
Token,
Token ID11
Feature activation+2.98e-02

Pos logit contributions
aces+0.176
alks+0.175
bons+0.160
pie+0.159
alk+0.158
Neg logit contributions
 jog-0.150
s-0.140
aa-0.133
 j-0.116
 sheep-0.116
Token ha
Token ID387
Feature activation+7.09e-02

Pos logit contributions
anish+0.050
pie+0.047
 seconds+0.044
rum+0.043
 accomplishment+0.042
Neg logit contributions
 jog-0.050
s-0.050
aa-0.046
ory-0.043
ashing-0.042
Token.
Token ID13
Feature activation+2.08e-02

Pos logit contributions
anish+0.131
 accomplishment+0.124
pie+0.123
 itself+0.120
 rectangle+0.118
Neg logit contributions
s-0.127
aa-0.107
ory-0.096
 jog-0.094
udd-0.092
Token We
Token ID775
Feature activation+3.07e-02

Pos logit contributions
anish+0.032
pie+0.032
 seconds+0.031
 accomplishment+0.030
 course+0.028
Neg logit contributions
s-0.038
 jog-0.034
aa-0.034
ashing-0.033
ory-0.032
Token are
Token ID389
Feature activation+2.89e-02

Pos logit contributions
anish+0.063
 seconds+0.059
pie+0.058
 accomplishment+0.058
rum+0.056
Neg logit contributions
s-0.042
 jog-0.039
 especially-0.034
aa-0.034
act-0.034
Token cry
Token ID3960
Feature activation-2.60e-02

Pos logit contributions
s+0.005
ashing+0.005
aa+0.005
ory+0.005
act+0.005
Neg logit contributions
anish-0.006
pie-0.006
 seconds-0.005
 arrest-0.005
 eight-0.005
Token.
Token ID13
Feature activation-8.79e-03

Pos logit contributions
act+0.051
s+0.049
ory+0.049
 jog+0.049
ashing+0.048
Neg logit contributions
anish-0.032
 seconds-0.029
pie-0.028
rum-0.028
ps-0.027
Token
Token ID198
Feature activation-6.77e-03

Pos logit contributions
s+0.013
ashing+0.012
 jog+0.012
ory+0.011
act+0.011
Neg logit contributions
anish-0.017
pie-0.016
 seconds-0.015
 eight-0.015
 arrest-0.014
Token
Token ID198
Feature activation-4.26e-03

Pos logit contributions
s+0.010
ashing+0.009
 jog+0.009
act+0.009
ory+0.009
Neg logit contributions
anish-0.013
pie-0.012
 seconds-0.012
 eight-0.011
 arrest-0.011
Token"
Token ID1
Feature activation-2.06e-03

Pos logit contributions
s+0.008
ashing+0.007
act+0.007
 jog+0.007
ory+0.007
Neg logit contributions
anish-0.007
pie-0.006
 seconds-0.006
 eight-0.006
 arrest-0.006
TokenHa
Token ID23303
Feature activation+1.02e-01

Pos logit contributions
s+0.003
ashing+0.003
 jog+0.003
aa+0.003
ory+0.003
Neg logit contributions
anish-0.004
pie-0.004
 seconds-0.003
 arrest-0.003
 course-0.003
Token,
Token ID11
Feature activation+3.73e-02

Pos logit contributions
aces+0.192
alks+0.185
stack+0.179
pie+0.174
 itself+0.173
Neg logit contributions
 jog-0.138
s-0.135
aa-0.125
 warm-0.109
 j-0.108
Token ha
Token ID387
Feature activation+7.77e-02

Pos logit contributions
anish+0.071
pie+0.070
 seconds+0.066
rum+0.064
aces+0.064
Neg logit contributions
s-0.052
 jog-0.051
aa-0.044
 jazz-0.044
 coffee-0.043
Token,
Token ID11
Feature activation+3.59e-02

Pos logit contributions
pie+0.156
 itself+0.155
anish+0.154
 accomplishment+0.153
 amulet+0.152
Neg logit contributions
s-0.123
aa-0.097
udd-0.091
ory-0.083
ashing-0.083
Token ha
Token ID387
Feature activation+6.93e-02

Pos logit contributions
anish+0.071
pie+0.069
 seconds+0.066
rum+0.065
 accomplishment+0.064
Neg logit contributions
s-0.047
 jog-0.047
aa-0.040
 jazz-0.039
 sax-0.037
Token!
Token ID0
Feature activation+2.05e-02

Pos logit contributions
anish+0.130
pie+0.129
 accomplishment+0.127
 itself+0.125
 amulet+0.122
Neg logit contributions
s-0.119
aa-0.094
udd-0.092
ashing-0.084
 jog-0.084
Token away
Token ID1497
Feature activation+3.14e-02

Pos logit contributions
anish+0.015
pie+0.014
 seconds+0.014
 arrest+0.013
 course+0.013
Neg logit contributions
s-0.008
ashing-0.007
 jog-0.007
aa-0.007
act-0.006
Token.
Token ID13
Feature activation+2.17e-02

Pos logit contributions
anish+0.075
pie+0.071
 course+0.069
 arrest+0.064
 seconds+0.064
Neg logit contributions
s-0.040
 early-0.029
 jog-0.028
ashing-0.028
aa-0.026
Token
Token ID198
Feature activation+6.41e-04

Pos logit contributions
anish+0.048
pie+0.047
 seconds+0.044
 accomplishment+0.043
ra+0.043
Neg logit contributions
s-0.027
 jog-0.021
aa-0.021
ashing-0.020
act-0.019
Token
Token ID198
Feature activation+2.00e-02

Pos logit contributions
anish+0.001
pie+0.001
 seconds+0.001
 arrest+0.001
 eight+0.001
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token"
Token ID1
Feature activation+1.92e-02

Pos logit contributions
anish+0.039
pie+0.038
 accomplishment+0.036
 seconds+0.035
ps+0.034
Neg logit contributions
s-0.032
 jog-0.025
aa-0.025
ashing-0.025
act-0.023
TokenHa
Token ID23303
Feature activation+1.02e-01

Pos logit contributions
anish+0.035
pie+0.033
 accomplishment+0.031
rum+0.031
 seconds+0.030
Neg logit contributions
s-0.030
 jog-0.027
aa-0.026
ashing-0.025
 especially-0.025
Token ha
Token ID387
Feature activation+8.12e-02

Pos logit contributions
aces+0.180
alks+0.173
pie+0.163
alk+0.161
 itself+0.160
Neg logit contributions
 jog-0.143
aa-0.136
s-0.135
 sheep-0.123
 j-0.116
Token!"
Token ID2474
Feature activation+2.85e-02

Pos logit contributions
pie+0.148
anish+0.147
 itself+0.147
 accomplishment+0.145
 amulet+0.138
Neg logit contributions
s-0.142
aa-0.123
udd-0.106
ory-0.105
 jog-0.102
Token Tom
Token ID4186
Feature activation+1.20e-03

Pos logit contributions
anish+0.057
pie+0.053
rum+0.049
 seconds+0.048
ra+0.047
Neg logit contributions
s-0.046
 jog-0.038
aa-0.035
 early-0.034
 jazz-0.033
Token laughs
Token ID22051
Feature activation-4.07e-03

Pos logit contributions
anish+0.003
pie+0.002
 seconds+0.002
 arrest+0.002
 eight+0.002
Neg logit contributions
s-0.002
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token.
Token ID13
Feature activation+3.23e-02

Pos logit contributions
s+0.007
ashing+0.006
act+0.006
 jog+0.006
aa+0.006
Neg logit contributions
anish-0.007
pie-0.006
 seconds-0.006
 eight-0.006
 arrest-0.006
Token air
Token ID1633
Feature activation-4.49e-03

Pos logit contributions
anish+0.055
pie+0.052
 seconds+0.050
rum+0.049
 accomplishment+0.047
Neg logit contributions
s-0.049
 jog-0.044
ashing-0.041
act-0.040
udd-0.039
Token.
Token ID13
Feature activation-7.62e-03

Pos logit contributions
s+0.007
 jog+0.006
ashing+0.006
aa+0.006
act+0.006
Neg logit contributions
anish-0.008
pie-0.008
 seconds-0.007
 course-0.007
 arrest-0.007
Token
Token ID198
Feature activation+4.62e-03

Pos logit contributions
s+0.010
 jog+0.009
ashing+0.009
aa+0.009
act+0.009
Neg logit contributions
anish-0.016
pie-0.015
 seconds-0.014
 eight-0.014
 course-0.014
Token
Token ID198
Feature activation+8.17e-03

Pos logit contributions
anish+0.010
pie+0.009
 seconds+0.009
 arrest+0.008
 eight+0.008
Neg logit contributions
s-0.006
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
Token"
Token ID1
Feature activation+4.71e-03

Pos logit contributions
anish+0.016
pie+0.016
 seconds+0.015
 arrest+0.014
 accomplishment+0.014
Neg logit contributions
s-0.012
 jog-0.010
ashing-0.010
aa-0.010
act-0.009
TokenHa
Token ID23303
Feature activation+1.02e-01

Pos logit contributions
anish+0.009
pie+0.009
 seconds+0.008
 arrest+0.008
 eight+0.008
Neg logit contributions
s-0.007
ashing-0.006
 jog-0.006
aa-0.006
act-0.006
Token,
Token ID11
Feature activation+3.30e-02

Pos logit contributions
aces+0.184
alks+0.172
stack+0.165
alk+0.162
 ago+0.162
Neg logit contributions
 jog-0.138
s-0.131
aa-0.124
 j-0.110
 hot-0.109
Token I
Token ID314
Feature activation+2.43e-02

Pos logit contributions
anish+0.059
pie+0.058
 seconds+0.054
rum+0.053
aces+0.051
Neg logit contributions
s-0.050
 jog-0.050
aa-0.043
 jazz-0.043
ory-0.042
Token got
Token ID1392
Feature activation+2.28e-02

Pos logit contributions
anish+0.044
pie+0.043
 seconds+0.040
 accomplishment+0.039
rum+0.038
Neg logit contributions
s-0.040
 jog-0.036
aa-0.034
ashing-0.034
udd-0.033
Token your
Token ID534
Feature activation+1.73e-02

Pos logit contributions
anish+0.044
pie+0.039
 seconds+0.037
rum+0.036
 accomplishment+0.035
Neg logit contributions
s-0.039
 jog-0.032
aa-0.031
ashing-0.031
ory-0.030
Token ball
Token ID2613
Feature activation+3.84e-02

Pos logit contributions
anish+0.033
pie+0.031
 seconds+0.030
rum+0.028
 arrest+0.027
Neg logit contributions
s-0.028
 jog-0.024
ashing-0.023
aa-0.022
ory-0.022
Token sad
Token ID6507
Feature activation-1.95e-02

Pos logit contributions
anish+0.077
pie+0.067
 course+0.067
rum+0.066
 arrest+0.066
Neg logit contributions
s-0.033
 especially-0.031
 jog-0.030
ashing-0.029
 jazz-0.028
Token.
Token ID13
Feature activation-2.26e-02

Pos logit contributions
s+0.035
act+0.035
aa+0.034
ory+0.034
ashing+0.034
Neg logit contributions
anish-0.027
 seconds-0.026
pie-0.024
 nine-0.023
 eight-0.022
Token
Token ID198
Feature activation-1.75e-02

Pos logit contributions
s+0.034
ashing+0.033
act+0.032
 jog+0.032
ling+0.032
Neg logit contributions
anish-0.040
 eight-0.035
 seconds-0.035
pie-0.035
 nine-0.034
Token
Token ID198
Feature activation-1.83e-02

Pos logit contributions
s+0.027
ashing+0.026
act+0.025
ory+0.025
 jog+0.025
Neg logit contributions
anish-0.030
pie-0.028
 seconds-0.028
 eight-0.026
 course-0.026
Token"
Token ID1
Feature activation-8.29e-03

Pos logit contributions
s+0.038
act+0.037
ashing+0.037
ory+0.036
aa+0.036
Neg logit contributions
anish-0.021
 seconds-0.019
pie-0.019
 course-0.018
 nine-0.018
TokenHa
Token ID23303
Feature activation+1.02e-01

Pos logit contributions
s+0.013
ashing+0.012
aa+0.011
act+0.011
udd+0.011
Neg logit contributions
anish-0.016
pie-0.015
 seconds-0.015
 course-0.014
ra-0.014
Token ha
Token ID387
Feature activation+8.24e-02

Pos logit contributions
alks+0.181
aces+0.178
alk+0.168
pie+0.168
stack+0.168
Neg logit contributions
 jog-0.145
s-0.133
aa-0.131
 j-0.116
 warm-0.110
Token,
Token ID11
Feature activation+3.57e-02

Pos logit contributions
pie+0.163
 itself+0.162
anish+0.160
 accomplishment+0.157
ps+0.154
Neg logit contributions
s-0.134
aa-0.107
ory-0.099
udd-0.094
ashing-0.092
Token I
Token ID314
Feature activation+2.61e-02

Pos logit contributions
anish+0.067
pie+0.067
rum+0.063
 seconds+0.062
 accomplishment+0.062
Neg logit contributions
 jog-0.049
s-0.049
 jazz-0.042
 especially-0.041
ory-0.041
Token am
Token ID716
Feature activation+5.59e-02

Pos logit contributions
anish+0.047
pie+0.046
 seconds+0.043
rum+0.042
 accomplishment+0.041
Neg logit contributions
s-0.043
 jog-0.039
ashing-0.036
aa-0.035
 especially-0.035
Token better
Token ID1365
Feature activation+3.27e-02

Pos logit contributions
anish+0.134
 course+0.112
rum+0.112
pie+0.112
 arrest+0.109
Neg logit contributions
 especially-0.062
 tired-0.061
 sleepy-0.060
 early-0.057
s-0.057
Token They
Token ID1119
Feature activation-3.46e-02

Pos logit contributions
ling+0.056
s+0.055
actus+0.054
act+0.054
 jog+0.054
Neg logit contributions
anish-0.064
 eight-0.060
 seconds-0.059
 rectangle-0.056
 nine-0.056
Token did
Token ID750
Feature activation-8.31e-03

Pos logit contributions
s+0.070
ashing+0.066
act+0.065
aa+0.064
udd+0.063
Neg logit contributions
 seconds-0.043
anish-0.041
pie-0.041
 eight-0.041
ra-0.040
Token not
Token ID407
Feature activation-2.62e-02

Pos logit contributions
s+0.018
aa+0.016
ashing+0.016
act+0.016
ory+0.016
Neg logit contributions
anish-0.010
 seconds-0.010
pie-0.010
 accomplishment-0.009
 arrest-0.009
Token want
Token ID765
Feature activation-5.35e-02

Pos logit contributions
s+0.047
act+0.039
actus+0.038
udd+0.038
ashing+0.037
Neg logit contributions
pie-0.043
 seconds-0.042
anish-0.042
 eight-0.038
ps-0.038
Token to
Token ID284
Feature activation-4.91e-02

Pos logit contributions
ashing+0.094
actus+0.091
act+0.088
ling+0.088
udd+0.088
Neg logit contributions
 seconds-0.085
 eight-0.076
 zero-0.073
 example-0.073
 nine-0.072
Token dress
Token ID6576
Feature activation-1.00e-01

Pos logit contributions
s+0.082
ashing+0.080
act+0.078
actus+0.075
udd+0.075
Neg logit contributions
ra-0.074
 seconds-0.072
 arrest-0.072
pie-0.072
anish-0.071
Token.
Token ID13
Feature activation-4.10e-02

Pos logit contributions
ashing+0.180
rig+0.171
Ice+0.165
 café+0.165
ory+0.163
Neg logit contributions
ps-0.140
ra-0.109
ered-0.105
pie-0.102
 itself-0.101
Token
Token ID198
Feature activation-1.98e-02

Pos logit contributions
ling+0.068
act+0.065
actus+0.065
ashing+0.065
 jog+0.063
Neg logit contributions
anish-0.066
 eight-0.061
 seconds-0.059
 nine-0.056
 rectangle-0.056
Token
Token ID198
Feature activation-2.23e-02

Pos logit contributions
s+0.031
ashing+0.031
act+0.030
ory+0.030
 jog+0.030
Neg logit contributions
anish-0.032
 seconds-0.030
pie-0.030
 eight-0.029
 course-0.028
Token"
Token ID1
Feature activation-3.51e-02

Pos logit contributions
ashing+0.036
s+0.035
act+0.035
ory+0.034
actus+0.034
Neg logit contributions
anish-0.035
 seconds-0.034
 eight-0.033
ra-0.032
 nine-0.032
TokenBen
Token ID11696
Feature activation+4.03e-02

Pos logit contributions
ashing+0.064
s+0.063
udd+0.062
act+0.061
cases+0.059
Neg logit contributions
anish-0.056
 seconds-0.055
ra-0.048
 half-0.046
ASH-0.044
Token to
Token ID284
Feature activation-1.82e-02

Pos logit contributions
s+0.067
udd+0.065
act+0.064
ory+0.063
ling+0.063
Neg logit contributions
anish-0.061
 eight-0.060
pie-0.058
 solving-0.057
 arrest-0.056
Token play
Token ID711
Feature activation-7.51e-03

Pos logit contributions
s+0.023
ashing+0.019
act+0.019
aa+0.018
udd+0.018
Neg logit contributions
anish-0.040
pie-0.038
 seconds-0.036
 course-0.036
 arrest-0.036
Token with
Token ID351
Feature activation-3.66e-02

Pos logit contributions
s+0.012
aa+0.011
ashing+0.011
ory+0.010
act+0.010
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.012
 course-0.012
 arrest-0.012
Token their
Token ID511
Feature activation-3.31e-02

Pos logit contributions
ashing+0.067
s+0.062
aa+0.062
ory+0.061
itating+0.060
Neg logit contributions
pie-0.052
anish-0.051
 seconds-0.049
 eight-0.049
 arrest-0.049
Token puppy
Token ID26188
Feature activation-2.78e-02

Pos logit contributions
s+0.064
ashing+0.062
aa+0.061
ory+0.060
udd+0.058
Neg logit contributions
anish-0.042
pie-0.041
 accomplishment-0.038
 arrest-0.038
 seconds-0.038
Token,
Token ID11
Feature activation-9.67e-02

Pos logit contributions
s+0.046
aa+0.045
ashing+0.045
act+0.043
ory+0.043
Neg logit contributions
anish-0.042
pie-0.041
 course-0.039
 seconds-0.037
 arrest-0.036
Token Spot
Token ID15899
Feature activation-2.06e-02

Pos logit contributions
ashing+0.203
act+0.202
udd+0.195
ling+0.186
s+0.184
Neg logit contributions
 arrest-0.102
 no-0.101
 course-0.099
 eight-0.098
pie-0.096
Token.
Token ID13
Feature activation-4.68e-02

Pos logit contributions
act+0.035
ashing+0.035
s+0.035
aa+0.034
ory+0.034
Neg logit contributions
anish-0.029
pie-0.028
 course-0.027
 seconds-0.027
 eight-0.027
Token Spot
Token ID15899
Feature activation-5.49e-02

Pos logit contributions
act+0.091
udd+0.088
ashing+0.085
actus+0.081
ling+0.081
Neg logit contributions
 eight-0.060
 seconds-0.059
 course-0.057
 nine-0.057
anish-0.056
Token is
Token ID318
Feature activation-3.60e-02

Pos logit contributions
ashing+0.116
ory+0.112
aa+0.111
act+0.111
udd+0.108
Neg logit contributions
 course-0.051
pie-0.050
 eight-0.050
 seconds-0.047
 tries-0.045
Token a
Token ID257
Feature activation-5.67e-02

Pos logit contributions
s+0.079
act+0.074
ashing+0.071
udd+0.070
ory+0.070
Neg logit contributions
 eight-0.043
 six-0.040
pie-0.040
 nine-0.039
 seconds-0.039
Token the
Token ID262
Feature activation-5.38e-02

Pos logit contributions
anish+0.027
pie+0.025
 seconds+0.023
ra+0.022
rum+0.022
Neg logit contributions
s-0.024
 jog-0.022
ashing-0.020
 especially-0.020
aa-0.020
Token woods
Token ID16479
Feature activation+2.57e-02

Pos logit contributions
s+0.112
ashing+0.105
aa+0.104
cases+0.104
ory+0.104
Neg logit contributions
 course-0.067
 nine-0.060
 arrest-0.059
 eight-0.058
pie-0.058
Token when
Token ID618
Feature activation-9.25e-03

Pos logit contributions
anish+0.056
pie+0.052
 seconds+0.050
ra+0.048
 eight+0.048
Neg logit contributions
s-0.035
ashing-0.030
 jog-0.030
 especially-0.028
 early-0.027
Token he
Token ID339
Feature activation-2.23e-02

Pos logit contributions
s+0.013
ashing+0.012
aa+0.012
act+0.012
ory+0.011
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.017
 arrest-0.016
 course-0.016
Token heard
Token ID2982
Feature activation-3.13e-02

Pos logit contributions
s+0.037
ashing+0.035
act+0.034
aa+0.034
ory+0.033
Neg logit contributions
pie-0.034
anish-0.034
 seconds-0.033
 eight-0.032
 arrest-0.031
Token someone
Token ID2130
Feature activation-9.38e-02

Pos logit contributions
s+0.072
ashing+0.070
act+0.067
ory+0.067
aa+0.066
Neg logit contributions
anish-0.033
 seconds-0.032
pie-0.031
 eight-0.028
 course-0.027
Token shouting
Token ID19642
Feature activation-6.02e-02

Pos logit contributions
act+0.206
ory+0.195
actus+0.188
udd+0.187
ashing+0.187
Neg logit contributions
 counting-0.086
 solving-0.084
pie-0.084
 figuring-0.080
 finding-0.075
Token.
Token ID13
Feature activation-1.27e-02

Pos logit contributions
act+0.140
s+0.128
actus+0.127
udd+0.127
ashing+0.125
Neg logit contributions
anish-0.061
 seconds-0.058
pie-0.053
 itself-0.053
ps-0.052
Token He
Token ID679
Feature activation-3.58e-02

Pos logit contributions
s+0.022
ashing+0.021
act+0.021
aa+0.020
ory+0.020
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.018
 eight-0.017
 arrest-0.017
Token stopped
Token ID5025
Feature activation-5.87e-02

Pos logit contributions
ashing+0.065
act+0.063
s+0.061
ory+0.059
udd+0.059
Neg logit contributions
 seconds-0.051
 eight-0.049
anish-0.048
pie-0.047
ra-0.045
Token and
Token ID290
Feature activation-2.99e-02

Pos logit contributions
act+0.107
actus+0.106
udd+0.103
cases+0.102
ory+0.101
Neg logit contributions
pie-0.083
 solving-0.076
 seconds-0.076
 eight-0.075
anish-0.074
Token good
Token ID922
Feature activation+2.08e-02

Pos logit contributions
ashing+0.042
s+0.041
udd+0.039
act+0.038
ory+0.038
Neg logit contributions
anish-0.043
 seconds-0.042
 arrest-0.041
 eight-0.041
pie-0.040
Token care
Token ID1337
Feature activation-1.90e-02

Pos logit contributions
pie+0.041
anish+0.040
 seconds+0.036
 eight+0.035
rum+0.035
Neg logit contributions
s-0.031
 jog-0.028
aa-0.027
ashing-0.026
ory-0.024
Token of
Token ID286
Feature activation-5.34e-02

Pos logit contributions
s+0.033
 jog+0.033
act+0.032
ashing+0.032
aa+0.032
Neg logit contributions
anish-0.029
pie-0.026
 seconds-0.026
 eight-0.024
 course-0.024
Token them
Token ID606
Feature activation-2.57e-02

Pos logit contributions
ashing+0.095
s+0.086
udd+0.086
ding+0.082
ory+0.081
Neg logit contributions
 course-0.083
 seconds-0.080
 itself-0.078
pie-0.077
ps-0.077
Token and
Token ID290
Feature activation-2.51e-02

Pos logit contributions
aa+0.041
s+0.041
ashing+0.041
act+0.040
 jog+0.039
Neg logit contributions
anish-0.041
pie-0.039
 seconds-0.038
 course-0.036
ps-0.036
Token feed
Token ID3745
Feature activation-9.38e-02

Pos logit contributions
s+0.041
ashing+0.038
act+0.038
aa+0.037
ory+0.036
Neg logit contributions
anish-0.042
 seconds-0.040
pie-0.040
 arrest-0.039
 eight-0.038
Token them
Token ID606
Feature activation-2.28e-02

Pos logit contributions
ashing+0.200
udd+0.181
rig+0.177
 jog+0.164
act+0.164
Neg logit contributions
 course-0.107
 itself-0.103
ps-0.101
anish-0.092
 half-0.088
Token and
Token ID290
Feature activation-1.75e-02

Pos logit contributions
s+0.039
ashing+0.037
aa+0.037
act+0.036
 jog+0.036
Neg logit contributions
anish-0.035
pie-0.034
 seconds-0.031
ps-0.031
 eight-0.030
Token love
Token ID1842
Feature activation-6.33e-02

Pos logit contributions
s+0.027
ashing+0.024
act+0.023
aa+0.023
ory+0.023
Neg logit contributions
anish-0.032
pie-0.031
 seconds-0.031
 arrest-0.030
 eight-0.029
Token them
Token ID606
Feature activation-1.99e-02

Pos logit contributions
ashing+0.113
udd+0.111
cases+0.107
rig+0.107
ding+0.103
Neg logit contributions
 seconds-0.086
 itself-0.083
 course-0.079
anish-0.078
ps-0.076
Token.
Token ID13
Feature activation-3.28e-02

Pos logit contributions
s+0.035
ashing+0.034
aa+0.034
act+0.033
 jog+0.033
Neg logit contributions
anish-0.030
pie-0.028
 seconds-0.027
 course-0.026
 eight-0.025
Token He
Token ID679
Feature activation-3.23e-02

Pos logit contributions
s+0.012
aa+0.011
ashing+0.011
 jog+0.011
act+0.011
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.013
 course-0.013
 eight-0.013
Token always
Token ID1464
Feature activation-1.52e-02

Pos logit contributions
s+0.051
ashing+0.047
act+0.045
aa+0.044
ory+0.044
Neg logit contributions
pie-0.053
 seconds-0.053
anish-0.051
 eight-0.051
ra-0.049
Token brought
Token ID3181
Feature activation-1.75e-02

Pos logit contributions
s+0.026
act+0.023
ashing+0.023
 jog+0.022
ory+0.022
Neg logit contributions
anish-0.024
pie-0.024
 seconds-0.023
 eight-0.022
 course-0.021
Token his
Token ID465
Feature activation-1.52e-02

Pos logit contributions
s+0.029
ashing+0.028
 jog+0.026
aa+0.026
udd+0.025
Neg logit contributions
anish-0.029
pie-0.028
 seconds-0.028
 eight-0.026
 course-0.026
Token bin
Token ID9874
Feature activation-5.55e-02

Pos logit contributions
s+0.024
ashing+0.023
aa+0.021
ory+0.021
act+0.021
Neg logit contributions
anish-0.026
pie-0.025
 seconds-0.024
 eight-0.024
 course-0.023
Tokenocular
Token ID37320
Feature activation-9.36e-02

Pos logit contributions
Kids+0.102
 jog+0.091
aa+0.090
 jazz+0.090
ashing+0.089
Neg logit contributions
pie-0.074
anish-0.072
 eight-0.071
ra-0.071
 nine-0.068
Tokens
Token ID82
Feature activation-3.14e-02

Pos logit contributions
Kids+0.096
ashing+0.087
 shepherd+0.087
 spa+0.077
 jog+0.074
Neg logit contributions
 course-0.195
pie-0.189
 eight-0.183
on-0.183
 zero-0.181
Token,
Token ID11
Feature activation-3.66e-02

Pos logit contributions
s+0.046
ashing+0.044
 jog+0.043
aa+0.043
act+0.043
Neg logit contributions
anish-0.055
pie-0.052
 seconds-0.049
 eight-0.049
ra-0.048
Token his
Token ID465
Feature activation-2.35e-03

Pos logit contributions
s+0.062
ashing+0.059
act+0.057
aa+0.056
ory+0.056
Neg logit contributions
pie-0.057
 seconds-0.056
 course-0.055
 arrest-0.055
anish-0.055
Token sketch
Token ID17548
Feature activation-2.14e-03

Pos logit contributions
s+0.003
ashing+0.003
 jog+0.003
aa+0.003
act+0.003
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.004
 eight-0.004
 arrest-0.004
Tokenbook
Token ID2070
Feature activation-2.10e-02

Pos logit contributions
s+0.005
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.002
 course-0.002
 eight-0.002
Token learned
Token ID4499
Feature activation-2.87e-02

Pos logit contributions
s+0.015
ashing+0.014
 jog+0.013
aa+0.013
ory+0.013
Neg logit contributions
anish-0.015
pie-0.014
 seconds-0.014
 arrest-0.013
 course-0.013
Token that
Token ID326
Feature activation-7.02e-03

Pos logit contributions
ashing+0.056
s+0.055
ory+0.055
aa+0.052
ding+0.052
Neg logit contributions
anish-0.037
 seconds-0.035
pie-0.035
 course-0.034
 itself-0.034
Token sometimes
Token ID3360
Feature activation-5.03e-03

Pos logit contributions
s+0.012
ashing+0.011
aa+0.011
ory+0.011
act+0.010
Neg logit contributions
anish-0.012
pie-0.011
 seconds-0.011
 course-0.010
 eight-0.010
Token things
Token ID1243
Feature activation-5.28e-02

Pos logit contributions
s+0.009
ashing+0.009
aa+0.008
ory+0.008
act+0.008
Neg logit contributions
anish-0.008
pie-0.007
 seconds-0.007
 course-0.007
 arrest-0.007
Token can
Token ID460
Feature activation-3.42e-02

Pos logit contributions
ory+0.092
ashing+0.091
 café+0.087
aa+0.087
s+0.087
Neg logit contributions
anish-0.076
pie-0.075
 course-0.071
 seconds-0.070
ps-0.068
Token look
Token ID804
Feature activation-9.32e-02

Pos logit contributions
ashing+0.067
s+0.065
ory+0.064
act+0.062
aa+0.062
Neg logit contributions
pie-0.046
anish-0.045
 seconds-0.040
 course-0.039
 eight-0.039
Token scary
Token ID14343
Feature activation-2.72e-02

Pos logit contributions
Kids+0.175
aa+0.164
actus+0.164
ido+0.160
 café+0.153
Neg logit contributions
 six-0.122
 seconds-0.117
 itself-0.114
 accomplishment-0.114
 granted-0.113
Token,
Token ID11
Feature activation-5.13e-02

Pos logit contributions
s+0.043
aa+0.041
ashing+0.040
ory+0.039
 jog+0.039
Neg logit contributions
anish-0.047
 seconds-0.044
pie-0.044
 course-0.041
ps-0.040
Token but
Token ID475
Feature activation-1.80e-02

Pos logit contributions
s+0.083
ory+0.079
actus+0.078
aa+0.077
ashing+0.077
Neg logit contributions
anish-0.085
 seconds-0.084
pie-0.083
 course-0.078
 arrest-0.076
Token it
Token ID340
Feature activation-6.60e-03

Pos logit contributions
s+0.032
ashing+0.030
aa+0.029
ory+0.029
act+0.029
Neg logit contributions
anish-0.029
pie-0.027
 seconds-0.027
 course-0.025
 accomplishment-0.025
Token's
Token ID338
Feature activation+5.97e-04

Pos logit contributions
s+0.008
ashing+0.008
aa+0.007
 jog+0.007
ory+0.007
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.013
 course-0.012
 eight-0.012
Tokenudd
Token ID4185
Feature activation-6.26e-02

Pos logit contributions
 jog+0.075
Kids+0.073
 shepherd+0.072
 sax+0.070
Say+0.068
Neg logit contributions
ra-0.118
 arrest-0.113
anish-0.113
oor-0.111
 accomplishment-0.110
Tokenling
Token ID1359
Feature activation-3.69e-02

Pos logit contributions
 jog+0.083
ashing+0.082
 especially+0.081
Ice+0.080
 café+0.080
Neg logit contributions
 course-0.099
ps-0.098
 accomplishment-0.094
anish-0.094
pie-0.090
Token with
Token ID351
Feature activation-5.35e-02

Pos logit contributions
ashing+0.062
ory+0.061
aa+0.060
ido+0.059
s+0.058
Neg logit contributions
anish-0.056
 course-0.054
 seconds-0.052
ps-0.049
pie-0.049
Token their
Token ID511
Feature activation-5.72e-02

Pos logit contributions
ashing+0.090
rig+0.080
itating+0.080
act+0.075
abs+0.074
Neg logit contributions
 eight-0.084
 course-0.084
 arrest-0.081
 seconds-0.079
ra-0.078
Token big
Token ID1263
Feature activation-2.07e-02

Pos logit contributions
ashing+0.090
itating+0.080
ido+0.080
ding+0.078
s+0.076
Neg logit contributions
 eight-0.088
ra-0.086
 course-0.086
 accomplishment-0.084
 seconds-0.084
Token,
Token ID11
Feature activation-9.29e-02

Pos logit contributions
s+0.034
ashing+0.032
act+0.031
aa+0.030
 jog+0.029
Neg logit contributions
anish-0.034
pie-0.032
 eight-0.031
 seconds-0.031
 course-0.030
Token soft
Token ID2705
Feature activation-2.72e-02

Pos logit contributions
s+0.179
ashing+0.171
itating+0.170
act+0.170
abs+0.167
Neg logit contributions
 eight-0.108
 no-0.101
 claws-0.100
 six-0.098
 spe-0.097
Token blanket
Token ID18447
Feature activation-5.54e-04

Pos logit contributions
s+0.043
ashing+0.041
act+0.038
aa+0.037
ory+0.037
Neg logit contributions
anish-0.046
pie-0.042
 seconds-0.042
 eight-0.041
 course-0.040
Token.
Token ID13
Feature activation-1.35e-02

Pos logit contributions
s+0.001
ashing+0.001
 jog+0.001
aa+0.001
act+0.001
Neg logit contributions
anish-0.001
pie-0.001
 seconds-0.001
 arrest-0.001
 course-0.001
Token Every
Token ID3887
Feature activation-1.11e-02

Pos logit contributions
s+0.024
ashing+0.022
act+0.022
aa+0.022
ory+0.021
Neg logit contributions
anish-0.022
pie-0.019
 seconds-0.019
 course-0.019
 eight-0.019
Token night
Token ID1755
Feature activation-2.04e-02

Pos logit contributions
s+0.019
act+0.017
ashing+0.017
ory+0.017
aa+0.016
Neg logit contributions
anish-0.018
 seconds-0.017
pie-0.017
 course-0.016
 eight-0.016
Token He
Token ID679
Feature activation-1.20e-02

Pos logit contributions
anish+0.002
pie+0.002
 seconds+0.002
 eight+0.002
 arrest+0.002
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token did
Token ID750
Feature activation+1.19e-02

Pos logit contributions
s+0.020
ashing+0.018
 jog+0.017
act+0.017
aa+0.017
Neg logit contributions
anish-0.020
pie-0.020
 seconds-0.020
ra-0.018
 eight-0.018
Token not
Token ID407
Feature activation+1.18e-04

Pos logit contributions
anish+0.021
pie+0.019
 seconds+0.018
 arrest+0.017
 course+0.017
Neg logit contributions
s-0.021
 jog-0.019
ashing-0.018
aa-0.017
act-0.017
Token want
Token ID765
Feature activation-8.85e-03

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 arrest+0.000
 eight+0.000
Neg logit contributions
s-0.000
ashing-0.000
 jog-0.000
aa-0.000
act-0.000
Token to
Token ID284
Feature activation-2.08e-02

Pos logit contributions
s+0.012
ashing+0.011
 jog+0.010
aa+0.010
udd+0.010
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.017
 eight-0.016
 course-0.016
Token cough
Token ID22094
Feature activation-9.26e-02

Pos logit contributions
s+0.035
ashing+0.032
udd+0.030
act+0.030
aa+0.029
Neg logit contributions
pie-0.035
anish-0.035
 arrest-0.033
 seconds-0.032
ra-0.032
Token and
Token ID290
Feature activation-2.96e-03

Pos logit contributions
act+0.195
rig+0.193
 jog+0.187
Kids+0.187
 shepherd+0.187
Neg logit contributions
anish-0.091
ps-0.071
 no-0.070
rum-0.070
 six-0.068
Token hurt
Token ID5938
Feature activation-1.41e-02

Pos logit contributions
s+0.004
ashing+0.004
aa+0.004
act+0.004
 jog+0.004
Neg logit contributions
anish-0.006
pie-0.006
 seconds-0.005
 arrest-0.005
 eight-0.005
Token anymore
Token ID7471
Feature activation+9.03e-03

Pos logit contributions
s+0.025
ashing+0.024
 jog+0.023
aa+0.023
act+0.023
Neg logit contributions
anish-0.023
pie-0.021
 seconds-0.021
 eight-0.019
 course-0.019
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
anish+0.017
pie+0.016
 seconds+0.016
 arrest+0.015
 eight+0.015
Neg logit contributions
s-0.014
 jog-0.012
ashing-0.012
aa-0.011
act-0.011
Token He
Token ID679
Feature activation-1.19e-02

Pos logit contributions
anish+0.022
pie+0.022
 seconds+0.021
 arrest+0.020
 accomplishment+0.020
Neg logit contributions
s-0.013
 jog-0.011
ashing-0.011
aa-0.011
act-0.010
Token to
Token ID284
Feature activation-2.78e-02

Pos logit contributions
udd+0.100
ling+0.099
s+0.099
ory+0.098
act+0.098
Neg logit contributions
 eight-0.072
anish-0.071
 solving-0.066
pie-0.066
 seconds-0.065
Token play
Token ID711
Feature activation-5.48e-03

Pos logit contributions
s+0.039
act+0.034
ashing+0.034
udd+0.032
actus+0.031
Neg logit contributions
anish-0.057
pie-0.053
 seconds-0.052
 arrest-0.051
 course-0.051
Token with
Token ID351
Feature activation-3.55e-02

Pos logit contributions
s+0.008
aa+0.007
ashing+0.007
ory+0.007
act+0.007
Neg logit contributions
anish-0.010
pie-0.010
 seconds-0.009
 course-0.009
 arrest-0.009
Token their
Token ID511
Feature activation-3.97e-02

Pos logit contributions
ashing+0.066
ory+0.063
aa+0.062
s+0.061
ido+0.060
Neg logit contributions
pie-0.048
anish-0.048
 seconds-0.046
 arrest-0.045
 course-0.045
Token dog
Token ID3290
Feature activation-3.52e-02

Pos logit contributions
ashing+0.078
s+0.077
aa+0.076
ory+0.074
act+0.073
Neg logit contributions
anish-0.047
 accomplishment-0.043
pie-0.043
 arrest-0.042
 eight-0.042
Token,
Token ID11
Feature activation-9.23e-02

Pos logit contributions
s+0.063
act+0.063
aa+0.063
ashing+0.063
ory+0.062
Neg logit contributions
anish-0.046
pie-0.044
 course-0.042
 seconds-0.040
 eight-0.039
Token Spot
Token ID15899
Feature activation-2.47e-02

Pos logit contributions
act+0.199
ashing+0.195
udd+0.190
ling+0.183
s+0.181
Neg logit contributions
 eight-0.101
 arrest-0.098
pie-0.094
 course-0.090
 no-0.089
Token.
Token ID13
Feature activation-4.52e-02

Pos logit contributions
act+0.044
ashing+0.043
s+0.042
aa+0.042
udd+0.042
Neg logit contributions
anish-0.033
pie-0.032
 course-0.031
 eight-0.030
 seconds-0.030
Token Spot
Token ID15899
Feature activation-4.97e-02

Pos logit contributions
act+0.088
udd+0.083
ashing+0.081
ling+0.080
actus+0.080
Neg logit contributions
 seconds-0.058
 eight-0.056
 course-0.056
 arrest-0.056
anish-0.056
Token is
Token ID318
Feature activation-3.64e-02

Pos logit contributions
ashing+0.106
ory+0.104
act+0.103
aa+0.102
udd+0.100
Neg logit contributions
 course-0.044
pie-0.043
 eight-0.043
 seconds-0.042
anish-0.040
Token a
Token ID257
Feature activation-5.70e-02

Pos logit contributions
s+0.082
act+0.078
ory+0.075
actus+0.074
udd+0.073
Neg logit contributions
 eight-0.041
 six-0.039
pie-0.037
 nine-0.037
 seconds-0.037
Token play
Token ID711
Feature activation-4.62e-03

Pos logit contributions
s+0.027
act+0.023
ashing+0.023
udd+0.022
aa+0.022
Neg logit contributions
anish-0.043
pie-0.041
 seconds-0.040
 course-0.039
 arrest-0.039
Token outside
Token ID2354
Feature activation+5.98e-03

Pos logit contributions
s+0.007
aa+0.006
ashing+0.006
ory+0.006
act+0.006
Neg logit contributions
anish-0.009
pie-0.008
 seconds-0.008
 course-0.008
 arrest-0.007
Token with
Token ID351
Feature activation-4.07e-02

Pos logit contributions
anish+0.012
pie+0.011
 seconds+0.011
 arrest+0.010
 eight+0.010
Neg logit contributions
s-0.009
 jog-0.008
ashing-0.008
aa-0.007
act-0.007
Token their
Token ID511
Feature activation-3.80e-02

Pos logit contributions
ashing+0.072
s+0.068
ding+0.066
actus+0.066
itating+0.065
Neg logit contributions
 seconds-0.060
 eight-0.060
 half-0.059
anish-0.058
 nine-0.056
Token dog
Token ID3290
Feature activation-3.34e-02

Pos logit contributions
ashing+0.069
s+0.069
ory+0.067
aa+0.065
act+0.065
Neg logit contributions
anish-0.049
 eight-0.047
 accomplishment-0.046
pie-0.046
 seconds-0.045
Token,
Token ID11
Feature activation-9.21e-02

Pos logit contributions
s+0.067
ashing+0.066
act+0.066
aa+0.065
ory+0.065
Neg logit contributions
anish-0.037
pie-0.034
 course-0.033
 eight-0.032
 seconds-0.032
Token Spot
Token ID15899
Feature activation-2.18e-02

Pos logit contributions
act+0.195
ashing+0.187
udd+0.184
s+0.183
ind+0.183
Neg logit contributions
 eight-0.100
pie-0.093
 arrest-0.093
 course-0.087
 no-0.086
Token.
Token ID13
Feature activation-6.19e-02

Pos logit contributions
act+0.039
s+0.038
ashing+0.038
udd+0.037
ory+0.037
Neg logit contributions
anish-0.028
pie-0.027
 eight-0.026
 course-0.026
 seconds-0.026
Token Spot
Token ID15899
Feature activation-5.30e-02

Pos logit contributions
act+0.126
udd+0.122
ling+0.120
ooky+0.117
ashing+0.116
Neg logit contributions
 eight-0.070
 nine-0.067
 six-0.065
 seconds-0.063
 course-0.060
Token is
Token ID318
Feature activation-3.76e-02

Pos logit contributions
ashing+0.106
ory+0.106
act+0.103
aa+0.103
udd+0.101
Neg logit contributions
 course-0.053
 eight-0.052
pie-0.051
 seconds-0.050
 tries-0.047
Token a
Token ID257
Feature activation-6.19e-02

Pos logit contributions
s+0.084
act+0.080
ory+0.077
actus+0.076
cases+0.076
Neg logit contributions
 eight-0.043
 six-0.041
 nine-0.039
pie-0.039
 seconds-0.039
Token Mitt
Token ID16627
Feature activation-3.22e-02

Pos logit contributions
s+0.024
ashing+0.023
act+0.022
ory+0.021
aa+0.021
Neg logit contributions
anish-0.028
pie-0.027
 seconds-0.026
 course-0.025
 eight-0.025
Tokenens
Token ID641
Feature activation-4.31e-03

Pos logit contributions
s+0.030
 jog+0.026
 sax+0.025
ashing+0.022
 especially+0.022
Neg logit contributions
anish-0.077
 course-0.072
pie-0.071
 seconds-0.068
 arrest-0.066
Token learned
Token ID4499
Feature activation-3.58e-02

Pos logit contributions
s+0.007
ashing+0.006
 jog+0.006
aa+0.006
ory+0.006
Neg logit contributions
anish-0.008
pie-0.008
 seconds-0.007
 arrest-0.007
 course-0.007
Token not
Token ID407
Feature activation-3.01e-02

Pos logit contributions
ashing+0.070
ory+0.068
s+0.067
ding+0.065
aa+0.065
Neg logit contributions
anish-0.045
 course-0.043
 seconds-0.042
 itself-0.041
 example-0.040
Token to
Token ID284
Feature activation-3.25e-02

Pos logit contributions
s+0.044
ory+0.040
actus+0.040
act+0.039
udd+0.037
Neg logit contributions
anish-0.054
pie-0.053
 eight-0.052
 seconds-0.050
 solving-0.049
Token lie
Token ID6486
Feature activation-9.11e-02

Pos logit contributions
s+0.061
cases+0.056
actus+0.055
act+0.053
ding+0.052
Neg logit contributions
anish-0.050
pie-0.044
 seconds-0.044
 arrest-0.043
ra-0.042
Token on
Token ID319
Feature activation-3.96e-02

Pos logit contributions
Kids+0.165
Say+0.160
rig+0.156
 jog+0.153
itionally+0.152
Neg logit contributions
ps-0.099
pie-0.095
 claws-0.095
 inches-0.093
 itself-0.093
Token the
Token ID262
Feature activation-2.24e-02

Pos logit contributions
s+0.070
ashing+0.069
ory+0.064
aa+0.063
udd+0.062
Neg logit contributions
 seconds-0.058
anish-0.058
pie-0.057
 course-0.057
 arrest-0.056
Token couch
Token ID18507
Feature activation-2.52e-02

Pos logit contributions
s+0.045
ory+0.043
ashing+0.042
aa+0.040
 jog+0.038
Neg logit contributions
anish-0.031
pie-0.030
 seconds-0.028
 course-0.028
 eight-0.027
Token and
Token ID290
Feature activation-2.07e-02

Pos logit contributions
s+0.038
aa+0.035
act+0.034
ashing+0.033
 jog+0.033
Neg logit contributions
anish-0.046
pie-0.044
 seconds-0.042
 course-0.041
 accomplishment-0.039
Token play
Token ID711
Feature activation-4.48e-03

Pos logit contributions
s+0.033
ashing+0.030
ory+0.029
aa+0.029
act+0.029
Neg logit contributions
anish-0.037
pie-0.035
 seconds-0.034
 arrest-0.033
 course-0.033
Token also
Token ID635
Feature activation+2.07e-02

Pos logit contributions
ashing+0.051
s+0.050
act+0.049
aa+0.048
 jog+0.048
Neg logit contributions
anish-0.041
 seconds-0.039
pie-0.037
 course-0.037
 eight-0.037
Token nervous
Token ID10927
Feature activation-2.35e-02

Pos logit contributions
anish+0.040
pie+0.037
 seconds+0.035
 arrest+0.034
 course+0.034
Neg logit contributions
s-0.031
 jog-0.027
ashing-0.027
aa-0.026
ory-0.025
Token about
Token ID546
Feature activation-9.59e-03

Pos logit contributions
s+0.038
act+0.036
ashing+0.035
actus+0.034
 jog+0.034
Neg logit contributions
anish-0.039
 seconds-0.037
pie-0.037
 course-0.034
 eight-0.034
Token having
Token ID1719
Feature activation+5.90e-03

Pos logit contributions
s+0.018
ashing+0.016
aa+0.016
act+0.016
 jog+0.015
Neg logit contributions
anish-0.015
pie-0.014
 seconds-0.014
 course-0.013
 eight-0.013
Token to
Token ID284
Feature activation-2.21e-03

Pos logit contributions
anish+0.010
pie+0.010
 seconds+0.009
 arrest+0.009
 eight+0.009
Neg logit contributions
s-0.010
 jog-0.009
ashing-0.009
aa-0.009
act-0.008
Token lie
Token ID6486
Feature activation-9.11e-02

Pos logit contributions
s+0.003
ashing+0.003
aa+0.003
act+0.003
 jog+0.003
Neg logit contributions
anish-0.004
pie-0.004
 seconds-0.004
 arrest-0.004
 course-0.004
Token.
Token ID13
Feature activation-2.88e-03

Pos logit contributions
Kids+0.157
aa+0.148
ido+0.147
 shepherd+0.144
 jog+0.144
Neg logit contributions
anish-0.105
 seconds-0.102
ps-0.102
pie-0.100
 itself-0.100
Token He
Token ID679
Feature activation-2.97e-02

Pos logit contributions
s+0.004
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.005
 eight-0.005
 arrest-0.005
Token never
Token ID1239
Feature activation-1.75e-02

Pos logit contributions
s+0.049
ashing+0.047
aa+0.044
act+0.043
ory+0.043
Neg logit contributions
 seconds-0.047
anish-0.047
pie-0.047
 eight-0.043
 course-0.043
Token used
Token ID973
Feature activation+9.10e-03

Pos logit contributions
s+0.027
act+0.024
ashing+0.024
aa+0.024
ory+0.023
Neg logit contributions
anish-0.031
pie-0.030
 seconds-0.029
 eight-0.027
 arrest-0.027
Token the
Token ID262
Feature activation-1.41e-02

Pos logit contributions
anish+0.017
pie+0.016
 seconds+0.015
 arrest+0.014
ra+0.014
Neg logit contributions
s-0.015
 jog-0.013
ashing-0.013
aa-0.012
act-0.012
Token to
Token ID284
Feature activation+9.68e-03

Pos logit contributions
s+0.047
ashing+0.045
act+0.042
aa+0.042
actus+0.042
Neg logit contributions
anish-0.037
 seconds-0.036
pie-0.035
 arrest-0.035
 eight-0.034
Token breathe
Token ID18044
Feature activation-3.86e-02

Pos logit contributions
anish+0.018
pie+0.017
 seconds+0.016
 arrest+0.015
 eight+0.015
Neg logit contributions
s-0.015
 jog-0.014
ashing-0.013
aa-0.013
act-0.013
Token deeply
Token ID7744
Feature activation-5.47e-02

Pos logit contributions
s+0.060
ashing+0.058
ory+0.056
aa+0.056
act+0.054
Neg logit contributions
anish-0.068
 seconds-0.064
pie-0.062
ps-0.062
 accomplishment-0.060
Token and
Token ID290
Feature activation-1.20e-02

Pos logit contributions
ashing+0.089
act+0.088
Kids+0.085
aa+0.085
actus+0.085
Neg logit contributions
anish-0.080
 seconds-0.078
 accomplishment-0.074
pie-0.073
 itself-0.073
Token to
Token ID284
Feature activation+3.06e-03

Pos logit contributions
s+0.019
ashing+0.017
act+0.017
aa+0.017
ory+0.016
Neg logit contributions
anish-0.022
pie-0.021
 seconds-0.020
 arrest-0.019
 eight-0.019
Token cough
Token ID22094
Feature activation-9.09e-02

Pos logit contributions
anish+0.006
pie+0.005
 seconds+0.005
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.005
ashing-0.004
 jog-0.004
aa-0.004
act-0.004
Token when
Token ID618
Feature activation-1.75e-02

Pos logit contributions
 jog+0.160
act+0.159
actus+0.154
ashing+0.153
rig+0.152
Neg logit contributions
anish-0.118
ps-0.107
 itself-0.103
 seconds-0.093
rum-0.093
Token they
Token ID484
Feature activation-2.18e-02

Pos logit contributions
s+0.020
ashing+0.018
act+0.018
aa+0.017
udd+0.017
Neg logit contributions
anish-0.038
 seconds-0.037
pie-0.037
 arrest-0.035
 eight-0.035
Token needed
Token ID2622
Feature activation+2.46e-03

Pos logit contributions
s+0.034
ashing+0.033
act+0.031
aa+0.030
ory+0.030
Neg logit contributions
anish-0.038
pie-0.037
 seconds-0.036
 course-0.034
 arrest-0.034
Token to
Token ID284
Feature activation-2.01e-02

Pos logit contributions
anish+0.005
pie+0.004
 seconds+0.004
 arrest+0.004
 eight+0.004
Neg logit contributions
s-0.004
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token.
Token ID13
Feature activation-1.46e-02

Pos logit contributions
s+0.035
ashing+0.033
act+0.031
aa+0.031
ory+0.030
Neg logit contributions
anish-0.033
pie-0.032
 seconds-0.031
 arrest-0.030
 course-0.029
Token her
Token ID607
Feature activation+6.36e-03

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 course+0.000
 arrest+0.000
Neg logit contributions
s-0.000
ashing-0.000
 jog-0.000
aa-0.000
act-0.000
Token that
Token ID326
Feature activation-1.98e-02

Pos logit contributions
anish+0.012
pie+0.011
 seconds+0.011
 arrest+0.010
 course+0.010
Neg logit contributions
s-0.010
 jog-0.009
ashing-0.009
aa-0.009
act-0.008
Token sometimes
Token ID3360
Feature activation-2.08e-02

Pos logit contributions
s+0.032
ashing+0.030
ory+0.029
act+0.029
aa+0.029
Neg logit contributions
anish-0.035
pie-0.034
 seconds-0.032
 course-0.031
 arrest-0.030
Token things
Token ID1243
Feature activation-4.50e-02

Pos logit contributions
s+0.039
ory+0.038
ashing+0.038
act+0.037
aa+0.036
Neg logit contributions
anish-0.030
 seconds-0.029
pie-0.028
 arrest-0.026
 course-0.026
Token can
Token ID460
Feature activation-4.39e-02

Pos logit contributions
ory+0.074
ashing+0.073
s+0.071
act+0.070
aa+0.069
Neg logit contributions
anish-0.069
pie-0.069
 course-0.065
ps-0.062
 seconds-0.062
Token look
Token ID804
Feature activation-9.06e-02

Pos logit contributions
ashing+0.097
ory+0.093
act+0.093
udd+0.091
s+0.089
Neg logit contributions
pie-0.048
anish-0.045
ps-0.040
 arrest-0.040
 eight-0.039
Token scary
Token ID14343
Feature activation-3.65e-02

Pos logit contributions
Kids+0.168
actus+0.162
aa+0.158
eddy+0.151
act+0.151
Neg logit contributions
 six-0.116
 seconds-0.110
 claws-0.106
 itself-0.106
 eight-0.106
Token but
Token ID475
Feature activation-2.51e-02

Pos logit contributions
s+0.059
aa+0.057
ashing+0.057
ory+0.056
act+0.055
Neg logit contributions
anish-0.060
 seconds-0.056
pie-0.055
 course-0.052
ps-0.051
Token they
Token ID484
Feature activation-1.37e-02

Pos logit contributions
s+0.038
ashing+0.036
act+0.036
aa+0.035
ory+0.035
Neg logit contributions
anish-0.045
pie-0.042
 seconds-0.042
 accomplishment-0.040
 eight-0.040
Token're
Token ID821
Feature activation-2.57e-04

Pos logit contributions
s+0.021
ashing+0.020
aa+0.020
act+0.019
ory+0.019
Neg logit contributions
anish-0.024
pie-0.023
 seconds-0.022
 course-0.021
 arrest-0.021
Token not
Token ID407
Feature activation+1.04e-02

Pos logit contributions
s+0.000
ashing+0.000
 jog+0.000
aa+0.000
act+0.000
Neg logit contributions
anish-0.000
pie-0.000
 seconds-0.000
 eight-0.000
 arrest-0.000
Token if
Token ID611
Feature activation-3.10e-02

Pos logit contributions
act+0.047
ashing+0.046
s+0.046
aa+0.045
 sax+0.045
Neg logit contributions
anish-0.056
pie-0.053
 eight-0.052
 nine-0.051
 seconds-0.050
Token he
Token ID339
Feature activation-1.17e-02

Pos logit contributions
ashing+0.046
s+0.046
udd+0.043
act+0.041
aa+0.040
Neg logit contributions
anish-0.054
 seconds-0.054
pie-0.053
 eight-0.051
 course-0.050
Token could
Token ID714
Feature activation-2.84e-02

Pos logit contributions
s+0.017
ashing+0.016
 jog+0.015
act+0.015
aa+0.015
Neg logit contributions
anish-0.022
pie-0.022
 seconds-0.021
ra-0.020
 eight-0.020
Token check
Token ID2198
Feature activation-3.23e-02

Pos logit contributions
s+0.049
ashing+0.048
act+0.046
udd+0.045
actus+0.044
Neg logit contributions
anish-0.044
pie-0.043
 arrest-0.042
ra-0.041
 seconds-0.041
Token her
Token ID607
Feature activation-4.01e-02

Pos logit contributions
ashing+0.057
s+0.056
act+0.054
udd+0.053
ory+0.052
Neg logit contributions
anish-0.049
pie-0.047
 seconds-0.047
 nine-0.044
 eight-0.044
Token ears
Token ID11368
Feature activation-9.01e-02

Pos logit contributions
ashing+0.068
aa+0.066
act+0.066
Kids+0.063
s+0.063
Neg logit contributions
anish-0.061
ra-0.058
 example-0.057
 accomplishment-0.056
 seconds-0.056
Token,
Token ID11
Feature activation-3.77e-02

Pos logit contributions
act+0.162
ashing+0.156
Kids+0.151
actus+0.148
aa+0.146
Neg logit contributions
anish-0.129
pie-0.125
 nine-0.115
 course-0.113
on-0.113
Token her
Token ID607
Feature activation-9.54e-03

Pos logit contributions
s+0.058
act+0.054
ashing+0.054
actus+0.051
aa+0.051
Neg logit contributions
anish-0.067
 seconds-0.066
pie-0.065
 eight-0.062
 course-0.062
Token eyes
Token ID2951
Feature activation-5.05e-02

Pos logit contributions
s+0.016
ashing+0.014
act+0.014
 jog+0.014
aa+0.014
Neg logit contributions
anish-0.016
pie-0.015
 seconds-0.015
ra-0.014
 eight-0.014
Token,
Token ID11
Feature activation-2.89e-02

Pos logit contributions
act+0.087
ashing+0.086
s+0.083
ory+0.082
aa+0.082
Neg logit contributions
anish-0.072
pie-0.070
 seconds-0.069
ps-0.065
 eight-0.065
Token and
Token ID290
Feature activation-6.14e-03

Pos logit contributions
s+0.042
act+0.040
ashing+0.040
aa+0.038
ory+0.038
Neg logit contributions
anish-0.054
pie-0.052
 seconds-0.051
 course-0.049
 eight-0.048
Token morning
Token ID3329
Feature activation+1.31e-02

Pos logit contributions
s+0.014
act+0.012
ashing+0.012
ory+0.012
 jog+0.012
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.013
 course-0.012
 eight-0.012
Token they
Token ID484
Feature activation-5.74e-02

Pos logit contributions
anish+0.021
pie+0.021
 seconds+0.019
 arrest+0.018
ra+0.018
Neg logit contributions
s-0.024
 jog-0.021
ashing-0.021
aa-0.020
ory-0.020
Token would
Token ID561
Feature activation-4.31e-02

Pos logit contributions
s+0.071
 sax+0.068
act+0.068
ashing+0.067
actus+0.066
Neg logit contributions
 eight-0.102
pie-0.102
 seconds-0.100
anish-0.100
 course-0.098
Token rush
Token ID10484
Feature activation-3.13e-02

Pos logit contributions
s+0.064
act+0.060
ashing+0.056
ory+0.055
actus+0.054
Neg logit contributions
anish-0.079
pie-0.074
 arrest-0.074
 eight-0.074
 seconds-0.074
Token to
Token ID284
Feature activation-3.24e-02

Pos logit contributions
act+0.049
actus+0.049
ory+0.047
itionally+0.047
 shepherd+0.047
Neg logit contributions
anish-0.047
 seconds-0.046
 nine-0.046
pie-0.045
 eight-0.044
Token the
Token ID262
Feature activation-8.98e-02

Pos logit contributions
s+0.068
ashing+0.063
act+0.062
ory+0.061
aa+0.059
Neg logit contributions
 arrest-0.041
pie-0.039
 eight-0.039
anish-0.038
 course-0.037
Token park
Token ID3952
Feature activation-1.60e-02

Pos logit contributions
s+0.194
ashing+0.182
ory+0.181
aa+0.177
bs+0.174
Neg logit contributions
 course-0.093
 nine-0.091
 six-0.091
 eight-0.088
 five-0.084
Token together
Token ID1978
Feature activation-7.88e-03

Pos logit contributions
s+0.022
act+0.021
ory+0.020
aa+0.020
actus+0.019
Neg logit contributions
anish-0.030
pie-0.029
 seconds-0.028
 course-0.027
 eight-0.027
Token and
Token ID290
Feature activation-3.97e-02

Pos logit contributions
s+0.012
act+0.011
ashing+0.011
aa+0.011
 jog+0.010
Neg logit contributions
anish-0.014
pie-0.014
 seconds-0.013
 eight-0.013
 course-0.012
Token have
Token ID423
Feature activation-5.11e-03

Pos logit contributions
s+0.067
act+0.064
aa+0.061
udd+0.061
actus+0.060
Neg logit contributions
pie-0.064
anish-0.063
 seconds-0.060
 arrest-0.060
 eight-0.060
Token lots
Token ID6041
Feature activation-3.76e-02

Pos logit contributions
s+0.009
ashing+0.008
aa+0.008
 jog+0.008
act+0.008
Neg logit contributions
anish-0.008
pie-0.008
 seconds-0.008
 course-0.007
 eight-0.007
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
anish+0.012
pie+0.011
 seconds+0.011
 arrest+0.010
 eight+0.010
Neg logit contributions
s-0.010
ashing-0.008
 jog-0.008
aa-0.008
act-0.008
Token She
Token ID1375
Feature activation-1.06e-02

Pos logit contributions
s+0.032
ashing+0.030
aa+0.029
act+0.029
 jog+0.029
Neg logit contributions
anish-0.036
pie-0.033
 eight-0.033
 course-0.032
 seconds-0.031
Token also
Token ID635
Feature activation-3.80e-03

Pos logit contributions
s+0.017
ashing+0.015
aa+0.015
 jog+0.015
act+0.014
Neg logit contributions
anish-0.019
pie-0.018
 seconds-0.018
 course-0.017
 eight-0.017
Token liked
Token ID8288
Feature activation-3.58e-02

Pos logit contributions
s+0.006
ashing+0.006
 jog+0.005
aa+0.005
act+0.005
Neg logit contributions
anish-0.007
pie-0.006
 seconds-0.006
 course-0.006
 arrest-0.006
Token to
Token ID284
Feature activation+5.17e-03

Pos logit contributions
s+0.056
ashing+0.055
udd+0.052
act+0.051
cases+0.050
Neg logit contributions
 eight-0.058
 seconds-0.058
anish-0.055
 arrest-0.055
pie-0.055
Token dress
Token ID6576
Feature activation-8.98e-02

Pos logit contributions
anish+0.010
pie+0.010
 seconds+0.010
 arrest+0.009
 eight+0.009
Neg logit contributions
s-0.007
 jog-0.006
ashing-0.006
aa-0.006
act-0.006
Token him
Token ID683
Feature activation-2.83e-02

Pos logit contributions
ashing+0.161
Kids+0.143
Ice+0.141
ido+0.136
rig+0.135
Neg logit contributions
ra-0.109
ps-0.108
 course-0.102
eros-0.100
 itself-0.099
Token up
Token ID510
Feature activation-3.13e-03

Pos logit contributions
s+0.046
ashing+0.045
aa+0.045
act+0.043
ory+0.042
Neg logit contributions
anish-0.047
pie-0.044
ra-0.041
 seconds-0.041
ps-0.040
Token with
Token ID351
Feature activation-2.59e-02

Pos logit contributions
s+0.004
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.006
pie-0.006
 seconds-0.006
 eight-0.006
 arrest-0.006
Token her
Token ID607
Feature activation-1.39e-02

Pos logit contributions
ashing+0.042
s+0.041
act+0.037
udd+0.036
aa+0.036
Neg logit contributions
anish-0.043
 course-0.042
 seconds-0.042
 eight-0.041
pie-0.041
Token old
Token ID1468
Feature activation-1.84e-02

Pos logit contributions
s+0.024
ashing+0.023
act+0.022
aa+0.022
 jog+0.022
Neg logit contributions
anish-0.021
pie-0.020
 seconds-0.020
 eight-0.019
ra-0.019
Token because
Token ID780
Feature activation-1.32e-02

Pos logit contributions
s+0.009
 jog+0.008
ashing+0.008
aa+0.008
act+0.008
Neg logit contributions
anish-0.011
pie-0.010
 seconds-0.010
 arrest-0.009
 accomplishment-0.009
Token it
Token ID340
Feature activation-2.54e-02

Pos logit contributions
s+0.024
ashing+0.022
aa+0.021
 jog+0.021
ory+0.021
Neg logit contributions
anish-0.021
pie-0.020
 seconds-0.019
 course-0.018
 arrest-0.018
Token could
Token ID714
Feature activation-2.54e-02

Pos logit contributions
ashing+0.042
ory+0.041
s+0.041
aa+0.040
 jog+0.040
Neg logit contributions
anish-0.040
pie-0.040
 seconds-0.037
 arrest-0.035
 course-0.035
Token make
Token ID787
Feature activation-1.01e-02

Pos logit contributions
s+0.048
ashing+0.046
act+0.045
udd+0.044
ory+0.043
Neg logit contributions
pie-0.037
anish-0.036
 arrest-0.035
 seconds-0.035
 eight-0.034
Token him
Token ID683
Feature activation-5.62e-02

Pos logit contributions
s+0.016
ashing+0.015
 jog+0.014
ory+0.014
aa+0.014
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.016
 arrest-0.016
 course-0.015
Token cough
Token ID22094
Feature activation-8.96e-02

Pos logit contributions
Kids+0.080
act+0.079
s+0.077
actus+0.073
ory+0.072
Neg logit contributions
pie-0.095
ps-0.090
 accomplishment-0.090
anish-0.090
 arrest-0.090
Token and
Token ID290
Feature activation-1.16e-02

Pos logit contributions
act+0.171
Kids+0.170
ido+0.168
ory+0.165
actus+0.164
Neg logit contributions
anish-0.102
ps-0.101
 itself-0.092
 inches-0.088
 no-0.088
Token hurt
Token ID5938
Feature activation-1.26e-02

Pos logit contributions
s+0.018
ashing+0.017
act+0.016
ory+0.016
aa+0.016
Neg logit contributions
anish-0.021
pie-0.020
 seconds-0.019
 arrest-0.018
 eight-0.018
Token his
Token ID465
Feature activation-6.44e-03

Pos logit contributions
s+0.022
ashing+0.020
 jog+0.020
aa+0.019
act+0.019
Neg logit contributions
anish-0.021
pie-0.020
 seconds-0.019
 eight-0.018
 course-0.018
Token lungs
Token ID21726
Feature activation-2.99e-02

Pos logit contributions
s+0.011
ashing+0.010
 jog+0.010
act+0.010
aa+0.009
Neg logit contributions
anish-0.011
pie-0.010
 seconds-0.010
 accomplishment-0.009
 arrest-0.009
Token.
Token ID13
Feature activation+8.56e-03

Pos logit contributions
s+0.052
aa+0.049
act+0.049
ashing+0.048
ory+0.048
Neg logit contributions
anish-0.046
pie-0.044
 seconds-0.042
rum-0.040
ps-0.039
Token the
Token ID262
Feature activation-4.10e-02

Pos logit contributions
s+0.004
ashing+0.004
 jog+0.003
aa+0.003
act+0.003
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.003
 course-0.003
 arrest-0.003
Token animals
Token ID4695
Feature activation-5.02e-02

Pos logit contributions
ashing+0.072
s+0.071
aa+0.068
act+0.063
 jog+0.063
Neg logit contributions
 course-0.058
 eight-0.055
anish-0.054
 nine-0.053
pie-0.052
Token with
Token ID351
Feature activation-1.53e-02

Pos logit contributions
aa+0.082
act+0.079
Kids+0.078
ashing+0.078
udd+0.076
Neg logit contributions
pie-0.077
 course-0.075
anish-0.070
 solving-0.069
 seconds-0.068
Token his
Token ID465
Feature activation-2.97e-02

Pos logit contributions
s+0.026
ashing+0.024
aa+0.023
act+0.023
udd+0.022
Neg logit contributions
anish-0.025
pie-0.025
 seconds-0.024
 course-0.023
 arrest-0.023
Token bin
Token ID9874
Feature activation-5.07e-02

Pos logit contributions
ashing+0.043
s+0.042
act+0.040
udd+0.039
aa+0.039
Neg logit contributions
pie-0.050
anish-0.050
 eight-0.049
 seconds-0.048
 accomplishment-0.048
Tokenocular
Token ID37320
Feature activation-8.93e-02

Pos logit contributions
 jog+0.078
Kids+0.076
aa+0.075
ashing+0.074
 shepherd+0.071
Neg logit contributions
pie-0.077
anish-0.075
 course-0.071
 eight-0.070
ra-0.069
Tokens
Token ID82
Feature activation-2.75e-02

Pos logit contributions
ashing+0.073
Kids+0.071
 shepherd+0.067
 jog+0.063
 especially+0.060
Neg logit contributions
 course-0.200
pie-0.191
on-0.188
 arrest-0.183
 eight-0.182
Token and
Token ID290
Feature activation-2.10e-02

Pos logit contributions
s+0.047
ashing+0.045
aa+0.045
 jog+0.044
act+0.044
Neg logit contributions
anish-0.041
pie-0.039
 seconds-0.036
 course-0.036
 eight-0.035
Token seeing
Token ID4379
Feature activation-4.13e-02

Pos logit contributions
s+0.037
ashing+0.036
aa+0.035
act+0.034
ory+0.033
Neg logit contributions
anish-0.031
pie-0.030
 seconds-0.030
 arrest-0.029
 eight-0.029
Token how
Token ID703
Feature activation-4.98e-02

Pos logit contributions
ashing+0.078
s+0.076
aa+0.075
ory+0.070
ding+0.069
Neg logit contributions
 course-0.058
pie-0.055
anish-0.055
 seconds-0.054
 accomplishment-0.050
Token they
Token ID484
Feature activation-4.00e-02

Pos logit contributions
ashing+0.070
act+0.066
 sax+0.066
aa+0.065
s+0.065
Neg logit contributions
 eight-0.079
pie-0.078
 seconds-0.075
anish-0.074
 accomplishment-0.073
Token play
Token ID711
Feature activation-2.68e-03

Pos logit contributions
s+0.032
act+0.028
ashing+0.028
udd+0.027
aa+0.026
Neg logit contributions
anish-0.049
pie-0.046
 seconds-0.045
 arrest-0.044
 eight-0.044
Token outside
Token ID2354
Feature activation+7.69e-03

Pos logit contributions
s+0.004
ashing+0.004
aa+0.004
 jog+0.003
ory+0.003
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.005
 course-0.005
 arrest-0.004
Token with
Token ID351
Feature activation-3.20e-02

Pos logit contributions
anish+0.015
pie+0.014
 seconds+0.014
 arrest+0.013
 accomplishment+0.013
Neg logit contributions
s-0.012
 jog-0.010
ashing-0.010
aa-0.009
act-0.009
Token their
Token ID511
Feature activation-3.19e-02

Pos logit contributions
ashing+0.052
s+0.051
ory+0.048
actus+0.048
ding+0.047
Neg logit contributions
 seconds-0.051
anish-0.050
 eight-0.050
 half-0.049
pie-0.048
Token dog
Token ID3290
Feature activation-3.21e-02

Pos logit contributions
s+0.057
ashing+0.056
ory+0.054
aa+0.053
act+0.053
Neg logit contributions
anish-0.045
 eight-0.042
pie-0.042
 seconds-0.041
 accomplishment-0.040
Token,
Token ID11
Feature activation-8.88e-02

Pos logit contributions
s+0.064
ashing+0.064
act+0.063
aa+0.062
ory+0.061
Neg logit contributions
anish-0.036
pie-0.033
 course-0.032
 seconds-0.031
 eight-0.030
Token Max
Token ID5436
Feature activation-2.90e-03

Pos logit contributions
act+0.190
ashing+0.185
udd+0.182
ind+0.182
s+0.176
Neg logit contributions
 eight-0.091
pie-0.087
 arrest-0.082
 course-0.082
 no-0.081
Token.
Token ID13
Feature activation-3.74e-02

Pos logit contributions
s+0.005
ashing+0.004
act+0.004
 jog+0.004
aa+0.004
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.005
 course-0.004
 eight-0.004
Token One
Token ID1881
Feature activation-4.20e-02

Pos logit contributions
act+0.078
actus+0.073
ashing+0.072
udd+0.071
ling+0.071
Neg logit contributions
 eight-0.045
anish-0.044
 rectangle-0.042
 course-0.042
 seconds-0.042
Token day
Token ID1110
Feature activation-3.00e-03

Pos logit contributions
act+0.094
ory+0.092
itionally+0.089
cases+0.089
ind+0.087
Neg logit contributions
anish-0.039
 course-0.037
 eight-0.033
 seconds-0.032
 half-0.031
Token,
Token ID11
Feature activation-2.89e-02

Pos logit contributions
s+0.006
ashing+0.005
act+0.005
 jog+0.005
aa+0.005
Neg logit contributions
anish-0.004
pie-0.004
 seconds-0.004
 eight-0.004
 arrest-0.004
Token,
Token ID11
Feature activation+4.73e-02

Pos logit contributions
anish+0.055
pie+0.052
 seconds+0.047
 arrest+0.046
 accomplishment+0.046
Neg logit contributions
s-0.055
 jog-0.046
aa-0.043
 especially-0.043
ashing-0.043
Token three
Token ID1115
Feature activation+5.77e-02

Pos logit contributions
anish+0.069
pie+0.066
rum+0.064
 accomplishment+0.060
 arrest+0.057
Neg logit contributions
s-0.089
 jog-0.086
 especially-0.081
 jackets-0.077
 jazz-0.076
Token,
Token ID11
Feature activation+5.00e-02

Pos logit contributions
anish+0.117
pie+0.109
 arrest+0.098
 accomplishment+0.096
ses+0.095
Neg logit contributions
s-0.097
 especially-0.071
 jog-0.069
 jackets-0.068
aa-0.066
Token four
Token ID1440
Feature activation+6.14e-02

Pos logit contributions
anish+0.067
pie+0.065
rum+0.062
 accomplishment+0.056
 arrest+0.054
Neg logit contributions
s-0.099
 jog-0.096
 especially-0.090
 jackets-0.086
 jazz-0.085
Token,
Token ID11
Feature activation+5.63e-02

Pos logit contributions
anish+0.129
pie+0.119
 arrest+0.110
 accomplishment+0.105
ocolate+0.103
Neg logit contributions
s-0.100
 especially-0.073
 jog-0.071
 jackets-0.069
aa-0.064
Token five
Token ID1936
Feature activation+6.60e-02

Pos logit contributions
anish+0.071
pie+0.067
rum+0.065
 accomplishment+0.060
 arrest+0.055
Neg logit contributions
s-0.116
 jog-0.112
 especially-0.106
 jackets-0.100
 sheep-0.100
Token,
Token ID11
Feature activation+5.68e-02

Pos logit contributions
anish+0.137
pie+0.124
 arrest+0.113
stack+0.111
ocolate+0.111
Neg logit contributions
s-0.108
 especially-0.080
 jog-0.079
 jackets-0.075
cases-0.072
Token six
Token ID2237
Feature activation+6.31e-02

Pos logit contributions
anish+0.070
pie+0.064
rum+0.064
 accomplishment+0.060
 tal+0.052
Neg logit contributions
s-0.119
 jog-0.114
 especially-0.109
 jackets-0.102
 sheep-0.102
Token,
Token ID11
Feature activation+5.78e-02

Pos logit contributions
anish+0.132
pie+0.124
 arrest+0.114
 accomplishment+0.110
stack+0.108
Neg logit contributions
s-0.097
 jog-0.071
 especially-0.071
 jackets-0.067
act-0.063
Token seven
Token ID3598
Feature activation+6.72e-02

Pos logit contributions
anish+0.078
rum+0.067
pie+0.067
 accomplishment+0.062
 arrest+0.056
Neg logit contributions
s-0.121
 jog-0.115
 especially-0.107
 jackets-0.103
aa-0.102
Token,
Token ID11
Feature activation+6.78e-02

Pos logit contributions
anish+0.147
pie+0.136
stack+0.126
rum+0.125
 accomplishment+0.124
Neg logit contributions
s-0.097
 jog-0.072
 especially-0.072
 jackets-0.065
 movies-0.062
Token and
Token ID290
Feature activation+1.36e-02

Pos logit contributions
anish+0.029
pie+0.028
 seconds+0.026
 arrest+0.026
 course+0.025
Neg logit contributions
s-0.018
 jog-0.014
ashing-0.014
aa-0.014
act-0.013
Token scared
Token ID12008
Feature activation+6.80e-04

Pos logit contributions
anish+0.024
pie+0.023
 seconds+0.021
 arrest+0.020
 accomplishment+0.020
Neg logit contributions
s-0.023
 jog-0.020
aa-0.020
ashing-0.020
act-0.019
Token her
Token ID607
Feature activation+3.98e-03

Pos logit contributions
anish+0.001
pie+0.001
 seconds+0.001
 eight+0.001
 arrest+0.001
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token.
Token ID13
Feature activation-2.91e-04

Pos logit contributions
anish+0.007
pie+0.007
 seconds+0.007
 arrest+0.006
 course+0.006
Neg logit contributions
s-0.006
ashing-0.006
 jog-0.006
aa-0.005
act-0.005
Token "
Token ID366
Feature activation-9.81e-03

Pos logit contributions
s+0.001
ashing+0.000
 jog+0.000
aa+0.000
act+0.000
Neg logit contributions
anish-0.000
pie-0.000
 seconds-0.000
 eight-0.000
 arrest-0.000
TokenOh
Token ID5812
Feature activation+7.40e-02

Pos logit contributions
s+0.015
ashing+0.014
act+0.013
 jog+0.013
ory+0.013
Neg logit contributions
anish-0.019
pie-0.018
 seconds-0.017
 course-0.017
 arrest-0.016
Token no
Token ID645
Feature activation+4.19e-02

Pos logit contributions
pie+0.136
alks+0.121
ps+0.121
ates+0.119
anish+0.118
Neg logit contributions
s-0.119
 warm-0.111
aa-0.102
 cold-0.101
 chilly-0.099
Token,
Token ID11
Feature activation+4.19e-02

Pos logit contributions
anish+0.079
pie+0.078
bridge+0.070
ps+0.069
aces+0.069
Neg logit contributions
s-0.069
aa-0.066
 jog-0.060
ashing-0.057
ory-0.054
Token the
Token ID262
Feature activation+6.77e-02

Pos logit contributions
anish+0.085
pie+0.081
rum+0.076
aces+0.074
 seconds+0.074
Neg logit contributions
 jog-0.052
s-0.052
 warm-0.049
 sheep-0.048
aa-0.047
Token bird
Token ID6512
Feature activation+4.50e-02

Pos logit contributions
rum+0.148
anish+0.141
ates+0.134
 seconds+0.134
icians+0.131
Neg logit contributions
 warm-0.074
s-0.074
 hot-0.073
 sheep-0.069
 cold-0.068
Token scared
Token ID12008
Feature activation+2.33e-04

Pos logit contributions
anish+0.082
 seconds+0.076
 accomplishment+0.073
pie+0.073
 example+0.070
Neg logit contributions
s-0.083
aa-0.067
 jog-0.063
ashing-0.062
ling-0.062
Token mug
Token ID25152
Feature activation-4.25e-03

Pos logit contributions
s+0.039
ashing+0.037
udd+0.035
act+0.035
aa+0.035
Neg logit contributions
anish-0.032
pie-0.030
 seconds-0.028
 course-0.028
 arrest-0.027
Token was
Token ID373
Feature activation-1.34e-02

Pos logit contributions
s+0.007
ashing+0.006
 jog+0.006
aa+0.006
act+0.006
Neg logit contributions
anish-0.008
pie-0.008
 seconds-0.007
 course-0.007
 arrest-0.007
Token missing
Token ID4814
Feature activation+1.16e-02

Pos logit contributions
s+0.023
ashing+0.021
aa+0.021
act+0.020
 jog+0.020
Neg logit contributions
anish-0.022
pie-0.022
 seconds-0.021
 eight-0.020
 course-0.019
Token.
Token ID13
Feature activation-2.06e-02

Pos logit contributions
anish+0.022
pie+0.021
 seconds+0.020
 arrest+0.019
 course+0.019
Neg logit contributions
s-0.018
 jog-0.015
ashing-0.015
aa-0.015
act-0.015
Token "
Token ID366
Feature activation-1.94e-02

Pos logit contributions
act+0.036
ashing+0.035
ory+0.035
s+0.033
aa+0.033
Neg logit contributions
anish-0.032
pie-0.028
 eight-0.028
 seconds-0.027
 arrest-0.027
TokenOh
Token ID5812
Feature activation+6.25e-02

Pos logit contributions
s+0.031
ashing+0.029
act+0.028
ory+0.027
aa+0.027
Neg logit contributions
anish-0.037
pie-0.034
 seconds-0.033
 arrest-0.032
 course-0.032
Token no
Token ID645
Feature activation+3.89e-02

Pos logit contributions
pie+0.108
anish+0.098
alks+0.096
ps+0.094
ates+0.088
Neg logit contributions
s-0.108
 warm-0.097
 jog-0.093
 especially-0.088
 handsome-0.088
Token!
Token ID0
Feature activation+2.22e-02

Pos logit contributions
pie+0.071
anish+0.071
ps+0.063
bridge+0.061
 seconds+0.061
Neg logit contributions
s-0.067
aa-0.063
 jog-0.059
ory-0.057
ashing-0.056
Token Where
Token ID6350
Feature activation+1.99e-02

Pos logit contributions
anish+0.037
pie+0.036
 seconds+0.035
 accomplishment+0.034
 course+0.033
Neg logit contributions
s-0.038
 jog-0.033
aa-0.033
ashing-0.032
ory-0.032
Token is
Token ID318
Feature activation+2.96e-02

Pos logit contributions
anish+0.038
pie+0.037
 seconds+0.036
 accomplishment+0.035
 arrest+0.034
Neg logit contributions
s-0.031
 jog-0.026
aa-0.024
ashing-0.024
udd-0.023
Token my
Token ID616
Feature activation-2.27e-02

Pos logit contributions
anish+0.055
pie+0.050
 seconds+0.046
 accomplishment+0.044
ra+0.044
Neg logit contributions
s-0.050
 jog-0.043
aa-0.040
ory-0.039
 jackets-0.039
Token a
Token ID257
Feature activation-3.63e-02

Pos logit contributions
s+0.005
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.003
 eight-0.003
 course-0.003
Token thin
Token ID7888
Feature activation+8.54e-03

Pos logit contributions
s+0.059
ashing+0.056
cases+0.053
act+0.052
udd+0.051
Neg logit contributions
anish-0.059
pie-0.057
 eight-0.057
 seconds-0.053
 rectangle-0.052
Token bug
Token ID5434
Feature activation+1.91e-02

Pos logit contributions
anish+0.014
pie+0.013
 seconds+0.012
 arrest+0.012
 course+0.012
Neg logit contributions
s-0.015
 jog-0.014
ashing-0.014
aa-0.013
act-0.013
Token on
Token ID319
Feature activation+2.71e-02

Pos logit contributions
anish+0.037
pie+0.034
 seconds+0.034
 accomplishment+0.033
 eight+0.032
Neg logit contributions
s-0.030
 jog-0.026
ashing-0.025
aa-0.024
act-0.024
Token a
Token ID257
Feature activation-1.63e-02

Pos logit contributions
anish+0.052
pie+0.047
 seconds+0.046
 accomplishment+0.044
rum+0.044
Neg logit contributions
s-0.041
 jog-0.038
aa-0.034
 jazz-0.034
act-0.033
Token l
Token ID300
Feature activation+6.45e-02

Pos logit contributions
s+0.035
ashing+0.034
aa+0.032
act+0.032
udd+0.032
Neg logit contributions
pie-0.019
anish-0.019
 eight-0.017
 seconds-0.016
 course-0.016
Tokenily
Token ID813
Feature activation+1.56e-02

Pos logit contributions
anish+0.104
 accomplishment+0.102
 seconds+0.095
ASH+0.093
 Majesty+0.090
Neg logit contributions
s-0.104
ashing-0.099
act-0.099
actus-0.098
aa-0.094
Token pad
Token ID14841
Feature activation+7.56e-02

Pos logit contributions
anish+0.028
pie+0.026
 seconds+0.025
 accomplishment+0.024
 arrest+0.023
Neg logit contributions
s-0.027
ashing-0.023
 jog-0.023
aa-0.022
act-0.022
Token.
Token ID13
Feature activation-1.34e-02

Pos logit contributions
 accomplishment+0.166
anish+0.164
 seconds+0.164
 Majesty+0.161
 course+0.160
Neg logit contributions
s-0.100
ashing-0.073
aa-0.065
 jog-0.060
ding-0.058
Token Ho
Token ID9544
Feature activation+1.91e-02

Pos logit contributions
s+0.022
ashing+0.021
act+0.020
aa+0.020
 jog+0.020
Neg logit contributions
anish-0.022
pie-0.020
 eight-0.019
 seconds-0.019
 arrest-0.018
Tokenppy
Token ID14097
Feature activation+5.84e-03

Pos logit contributions
anish+0.043
pie+0.043
 seconds+0.042
 accomplishment+0.041
 arrest+0.041
Neg logit contributions
s-0.022
ashing-0.018
ory-0.017
aa-0.017
act-0.017
Token says
Token ID1139
Feature activation+1.53e-02

Pos logit contributions
anish+0.019
pie+0.018
 seconds+0.018
 arrest+0.017
 accomplishment+0.017
Neg logit contributions
s-0.013
 jog-0.011
ashing-0.011
aa-0.011
act-0.010
Token.
Token ID13
Feature activation-2.52e-03

Pos logit contributions
anish+0.031
pie+0.029
 course+0.028
 seconds+0.028
 arrest+0.028
Neg logit contributions
s-0.023
ashing-0.019
 jog-0.019
aa-0.018
ory-0.017
Token He
Token ID679
Feature activation+4.74e-02

Pos logit contributions
s+0.004
ashing+0.003
 jog+0.003
act+0.003
ory+0.003
Neg logit contributions
anish-0.005
pie-0.004
 seconds-0.004
 eight-0.004
 arrest-0.004
Token puts
Token ID7584
Feature activation+3.02e-02

Pos logit contributions
anish+0.093
pie+0.089
rum+0.081
 arrest+0.081
 seconds+0.079
Neg logit contributions
s-0.080
 jog-0.064
ashing-0.058
act-0.058
aa-0.057
Token his
Token ID465
Feature activation+8.56e-02

Pos logit contributions
anish+0.064
pie+0.058
 arrest+0.057
 seconds+0.055
rum+0.055
Neg logit contributions
s-0.044
 jog-0.035
aa-0.033
 jackets-0.033
 especially-0.032
Token red
Token ID2266
Feature activation+7.61e-02

Pos logit contributions
anish+0.183
 arrest+0.164
rum+0.161
ross+0.161
oor+0.157
Neg logit contributions
s-0.118
 favorite-0.108
 warm-0.106
 hot-0.101
 jog-0.101
Token block
Token ID2512
Feature activation+3.40e-02

Pos logit contributions
 seconds+0.112
anish+0.112
 figuring+0.111
 accomplishment+0.110
 arrest+0.107
Neg logit contributions
s-0.156
 jog-0.122
 jacket-0.114
 jackets-0.113
 favorite-0.109
Token on
Token ID319
Feature activation+3.42e-02

Pos logit contributions
anish+0.075
 arrest+0.070
pie+0.069
 seconds+0.069
rum+0.067
Neg logit contributions
s-0.047
 jog-0.034
ashing-0.033
aa-0.031
 early-0.031
Token top
Token ID1353
Feature activation+2.39e-02

Pos logit contributions
anish+0.072
pie+0.067
 seconds+0.062
rum+0.062
 Majesty+0.062
Neg logit contributions
s-0.045
 jog-0.038
 especially-0.035
 jazz-0.035
 early-0.034
Token of
Token ID286
Feature activation+4.20e-02

Pos logit contributions
anish+0.051
pie+0.047
 seconds+0.045
 arrest+0.045
 accomplishment+0.043
Neg logit contributions
s-0.034
 jog-0.026
ashing-0.026
aa-0.025
act-0.024
Token Sam
Token ID3409
Feature activation+1.40e-02

Pos logit contributions
anish+0.091
pie+0.081
 seconds+0.078
rum+0.078
ra+0.077
Neg logit contributions
s-0.052
 jog-0.045
 especially-0.043
 jazz-0.042
 warm-0.042
Token.
Token ID13
Feature activation-4.56e-03

Pos logit contributions
s+0.023
ashing+0.022
aa+0.022
 jog+0.022
act+0.022
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.019
 course-0.017
 eight-0.017
Token Now
Token ID2735
Feature activation+3.26e-03

Pos logit contributions
s+0.009
ashing+0.008
act+0.008
 jog+0.008
aa+0.008
Neg logit contributions
anish-0.007
pie-0.006
 seconds-0.006
 eight-0.006
 arrest-0.006
Token you
Token ID345
Feature activation+9.36e-03

Pos logit contributions
anish+0.005
pie+0.005
 seconds+0.005
 arrest+0.005
 eight+0.004
Neg logit contributions
s-0.006
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
Token can
Token ID460
Feature activation-1.35e-02

Pos logit contributions
anish+0.018
pie+0.017
 seconds+0.017
 arrest+0.016
 course+0.016
Neg logit contributions
s-0.014
 jog-0.012
ashing-0.012
aa-0.012
ory-0.011
Token practice
Token ID3357
Feature activation+2.00e-02

Pos logit contributions
s+0.019
ashing+0.018
aa+0.017
act+0.017
udd+0.016
Neg logit contributions
anish-0.026
pie-0.025
 course-0.024
 arrest-0.024
 seconds-0.024
Token and
Token ID290
Feature activation+1.93e-02

Pos logit contributions
anish+0.038
pie+0.036
 seconds+0.034
 arrest+0.033
ra+0.032
Neg logit contributions
s-0.032
 jog-0.027
ashing-0.025
 especially-0.025
aa-0.025
Token have
Token ID423
Feature activation+3.07e-02

Pos logit contributions
anish+0.037
pie+0.036
 seconds+0.033
rum+0.032
 arrest+0.032
Neg logit contributions
s-0.029
 jog-0.027
ashing-0.024
 especially-0.023
aa-0.023
Token more
Token ID517
Feature activation+2.72e-02

Pos logit contributions
anish+0.074
pie+0.073
ra+0.067
rum+0.066
 arrest+0.064
Neg logit contributions
s-0.034
 jog-0.029
 especially-0.026
 jackets-0.025
 jazz-0.024
Token fun
Token ID1257
Feature activation+3.25e-02

Pos logit contributions
anish+0.058
pie+0.057
ra+0.051
 arrest+0.051
 seconds+0.050
Neg logit contributions
s-0.038
 jog-0.033
ashing-0.029
 jackets-0.029
 especially-0.028
Token on
Token ID319
Feature activation+3.36e-02

Pos logit contributions
anish+0.061
pie+0.058
 arrest+0.053
 eight+0.053
ra+0.052
Neg logit contributions
s-0.052
 jog-0.045
ashing-0.043
 especially-0.043
 jackets-0.041
Token your
Token ID534
Feature activation+1.53e-02

Pos logit contributions
anish+0.066
pie+0.061
 seconds+0.056
ra+0.055
rum+0.055
Neg logit contributions
s-0.051
 jog-0.047
 jazz-0.042
 especially-0.041
 early-0.041
Token.
Token ID13
Feature activation+6.80e-03

Pos logit contributions
anish+0.044
pie+0.041
 seconds+0.040
 arrest+0.038
 eight+0.038
Neg logit contributions
s-0.034
ashing-0.028
 jog-0.027
aa-0.026
act-0.026
Token Everyone
Token ID11075
Feature activation+2.20e-03

Pos logit contributions
anish+0.012
pie+0.012
 seconds+0.011
 arrest+0.010
 course+0.010
Neg logit contributions
s-0.011
ashing-0.010
 jog-0.010
aa-0.010
act-0.009
Token waved
Token ID26834
Feature activation-2.07e-02

Pos logit contributions
anish+0.004
pie+0.004
 seconds+0.004
 arrest+0.003
 course+0.003
Neg logit contributions
s-0.003
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token goodbye
Token ID24829
Feature activation+1.13e-02

Pos logit contributions
s+0.034
act+0.033
ashing+0.033
actus+0.031
udd+0.031
Neg logit contributions
anish-0.032
 seconds-0.032
pie-0.031
 eight-0.030
 arrest-0.029
Token,
Token ID11
Feature activation+1.14e-02

Pos logit contributions
anish+0.026
pie+0.025
 seconds+0.024
 course+0.023
 arrest+0.023
Neg logit contributions
s-0.013
ashing-0.011
 jog-0.010
aa-0.010
ory-0.010
Token and
Token ID290
Feature activation+9.27e-03

Pos logit contributions
anish+0.025
pie+0.023
 seconds+0.022
ra+0.021
 arrest+0.021
Neg logit contributions
s-0.016
ashing-0.013
 jog-0.013
aa-0.013
act-0.012
Token the
Token ID262
Feature activation+1.32e-02

Pos logit contributions
anish+0.016
pie+0.015
 seconds+0.014
 arrest+0.014
 eight+0.014
Neg logit contributions
s-0.016
 jog-0.014
ashing-0.014
aa-0.014
act-0.013
Token yellow
Token ID7872
Feature activation+6.66e-04

Pos logit contributions
anish+0.030
pie+0.028
 seconds+0.026
rum+0.026
 arrest+0.026
Neg logit contributions
s-0.016
 jog-0.015
aa-0.013
ashing-0.013
act-0.013
Token boat
Token ID8848
Feature activation+3.24e-02

Pos logit contributions
anish+0.001
pie+0.001
 seconds+0.001
 arrest+0.001
 eight+0.001
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token sailed
Token ID33835
Feature activation+9.85e-03

Pos logit contributions
anish+0.061
pie+0.055
 seconds+0.053
rum+0.051
 arrest+0.051
Neg logit contributions
s-0.056
 jog-0.047
aa-0.044
ashing-0.043
 especially-0.042
Token off
Token ID572
Feature activation+2.63e-02

Pos logit contributions
anish+0.021
pie+0.020
 seconds+0.019
 arrest+0.019
 course+0.018
Neg logit contributions
s-0.013
 jog-0.011
ashing-0.011
aa-0.011
ory-0.010
Token day
Token ID1110
Feature activation+3.21e-04

Pos logit contributions
anish+0.070
pie+0.068
 arrest+0.064
 seconds+0.064
ps+0.063
Neg logit contributions
s-0.038
 jog-0.033
aa-0.032
ashing-0.030
act-0.028
Token,
Token ID11
Feature activation-5.96e-03

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 arrest+0.000
 eight+0.000
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token the
Token ID262
Feature activation+1.88e-02

Pos logit contributions
s+0.011
ashing+0.010
act+0.009
aa+0.009
 jog+0.009
Neg logit contributions
anish-0.010
pie-0.009
 seconds-0.009
 course-0.008
 eight-0.008
Token egg
Token ID5935
Feature activation+2.57e-02

Pos logit contributions
anish+0.040
pie+0.037
rum+0.035
 seconds+0.034
 arrest+0.034
Neg logit contributions
s-0.026
 jog-0.023
aa-0.022
ashing-0.020
act-0.020
Token began
Token ID2540
Feature activation+1.83e-02

Pos logit contributions
anish+0.053
pie+0.049
 seconds+0.048
rum+0.046
 course+0.046
Neg logit contributions
s-0.040
 jog-0.030
aa-0.030
ashing-0.029
act-0.028
Token to
Token ID284
Feature activation+1.19e-02

Pos logit contributions
anish+0.047
pie+0.044
 course+0.042
 seconds+0.042
 arrest+0.041
Neg logit contributions
s-0.018
 jog-0.013
ashing-0.013
aa-0.012
 especially-0.011
Token crack
Token ID8469
Feature activation-2.56e-03

Pos logit contributions
anish+0.024
pie+0.023
 seconds+0.022
 arrest+0.021
 course+0.021
Neg logit contributions
s-0.017
 jog-0.015
ashing-0.014
aa-0.014
act-0.013
Token.
Token ID13
Feature activation+3.24e-03

Pos logit contributions
s+0.004
ashing+0.003
 jog+0.003
aa+0.003
ory+0.003
Neg logit contributions
anish-0.005
pie-0.005
 seconds-0.004
 eight-0.004
ra-0.004
Token The
Token ID383
Feature activation+2.53e-02

Pos logit contributions
anish+0.006
pie+0.006
 seconds+0.005
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.005
ashing-0.004
 jog-0.004
aa-0.004
act-0.004
Token mom
Token ID1995
Feature activation+1.33e-02

Pos logit contributions
anish+0.051
pie+0.047
rum+0.045
 arrest+0.043
 seconds+0.043
Neg logit contributions
s-0.037
 jog-0.032
aa-0.031
actus-0.030
 especially-0.030
Tokenmy
Token ID1820
Feature activation+3.61e-02

Pos logit contributions
anish+0.035
pie+0.033
 seconds+0.032
 accomplishment+0.032
 arrest+0.031
Neg logit contributions
s-0.012
aa-0.008
ashing-0.008
 jog-0.008
udd-0.007
Token.
Token ID13
Feature activation+1.82e-02

Pos logit contributions
anish+0.002
pie+0.002
 seconds+0.002
 eight+0.002
 arrest+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token
Token ID198
Feature activation+2.83e-03

Pos logit contributions
anish+0.041
pie+0.040
 arrest+0.038
 course+0.037
 seconds+0.037
Neg logit contributions
s-0.023
 jog-0.018
ashing-0.017
aa-0.017
ory-0.016
Token
Token ID198
Feature activation+4.48e-03

Pos logit contributions
anish+0.006
pie+0.006
 seconds+0.005
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.004
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token"
Token ID1
Feature activation-1.74e-02

Pos logit contributions
anish+0.006
pie+0.006
 seconds+0.005
 arrest+0.005
 course+0.005
Neg logit contributions
s-0.009
ashing-0.008
 jog-0.008
aa-0.008
act-0.008
TokenNo
Token ID2949
Feature activation+2.83e-02

Pos logit contributions
s+0.024
ashing+0.023
act+0.022
ory+0.021
udd+0.021
Neg logit contributions
anish-0.035
 seconds-0.032
pie-0.032
 course-0.031
 eight-0.030
Token,
Token ID11
Feature activation+2.00e-02

Pos logit contributions
pie+0.058
anish+0.057
 seconds+0.052
ra+0.051
 arrest+0.050
Neg logit contributions
s-0.044
 jog-0.036
aa-0.034
ory-0.032
 jackets-0.031
Token it
Token ID340
Feature activation+2.25e-02

Pos logit contributions
anish+0.040
pie+0.039
 seconds+0.036
rum+0.035
 arrest+0.034
Neg logit contributions
s-0.028
 jog-0.025
aa-0.023
ory-0.023
 especially-0.022
Token is
Token ID318
Feature activation+3.84e-02

Pos logit contributions
anish+0.051
pie+0.049
 seconds+0.048
 accomplishment+0.047
 arrest+0.047
Neg logit contributions
s-0.029
 jog-0.021
aa-0.021
act-0.020
udd-0.020
Token mine
Token ID6164
Feature activation+1.38e-02

Pos logit contributions
anish+0.075
pie+0.068
 arrest+0.064
 seconds+0.061
 course+0.060
Neg logit contributions
s-0.061
 especially-0.053
 swe-0.052
 jog-0.051
 warm-0.051
Token.
Token ID13
Feature activation+8.53e-03

Pos logit contributions
anish+0.024
pie+0.023
 seconds+0.022
 arrest+0.021
 course+0.021
Neg logit contributions
s-0.024
ashing-0.020
 jog-0.020
aa-0.020
act-0.019
Token You
Token ID921
Feature activation+2.47e-02

Pos logit contributions
anish+0.013
pie+0.012
 seconds+0.011
 arrest+0.010
 course+0.010
Neg logit contributions
s-0.017
 jog-0.015
ashing-0.015
aa-0.015
act-0.014
Token the
Token ID262
Feature activation+2.59e-02

Pos logit contributions
anish+0.008
pie+0.008
 seconds+0.007
 arrest+0.007
 eight+0.007
Neg logit contributions
s-0.009
ashing-0.008
 jog-0.008
aa-0.008
act-0.007
Token poor
Token ID3595
Feature activation+7.59e-03

Pos logit contributions
anish+0.050
pie+0.044
rum+0.043
 seconds+0.043
 course+0.041
Neg logit contributions
s-0.039
 jog-0.038
ashing-0.033
 especially-0.033
aa-0.033
Token caul
Token ID46087
Feature activation+3.30e-02

Pos logit contributions
anish+0.014
pie+0.013
 seconds+0.013
 course+0.012
 arrest+0.012
Neg logit contributions
s-0.012
 jog-0.011
ashing-0.010
aa-0.010
act-0.010
Tokeniflower
Token ID42642
Feature activation+2.17e-02

Pos logit contributions
anish+0.059
 seconds+0.053
pie+0.051
 accomplishment+0.050
 eight+0.049
Neg logit contributions
s-0.060
act-0.050
actus-0.047
ling-0.047
ind-0.045
Token and
Token ID290
Feature activation+2.58e-02

Pos logit contributions
anish+0.046
pie+0.043
 seconds+0.041
 arrest+0.040
 eight+0.039
Neg logit contributions
s-0.030
 jog-0.025
ashing-0.024
aa-0.022
 especially-0.022
Token knew
Token ID2993
Feature activation+2.32e-02

Pos logit contributions
anish+0.056
pie+0.051
 seconds+0.048
ra+0.047
 arrest+0.047
Neg logit contributions
s-0.036
 jog-0.029
 especially-0.028
ashing-0.027
aa-0.027
Token he
Token ID339
Feature activation+2.29e-03

Pos logit contributions
anish+0.048
pie+0.044
 seconds+0.041
 arrest+0.041
ra+0.041
Neg logit contributions
s-0.032
 jog-0.027
 especially-0.026
aa-0.025
ashing-0.025
Token had
Token ID550
Feature activation+4.95e-02

Pos logit contributions
anish+0.004
pie+0.004
 seconds+0.004
 arrest+0.004
 eight+0.004
Neg logit contributions
s-0.003
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token to
Token ID284
Feature activation+4.08e-02

Pos logit contributions
anish+0.130
pie+0.122
rum+0.115
 rectangle+0.114
eros+0.113
Neg logit contributions
s-0.041
 jog-0.034
 especially-0.033
 would-0.028
ashing-0.027
Token rescue
Token ID9992
Feature activation+7.39e-03

Pos logit contributions
anish+0.082
pie+0.079
 seconds+0.074
 eight+0.072
 rectangle+0.071
Neg logit contributions
 jog-0.057
s-0.047
 especially-0.046
aa-0.043
 warm-0.042
Token him
Token ID683
Feature activation+1.06e-02

Pos logit contributions
anish+0.013
pie+0.012
 seconds+0.011
 arrest+0.011
 eight+0.010
Neg logit contributions
s-0.013
 jog-0.011
ashing-0.011
aa-0.011
act-0.011
Token water
Token ID1660
Feature activation-6.13e-03

Pos logit contributions
s+0.000
ashing+0.000
 jog+0.000
aa+0.000
act+0.000
Neg logit contributions
anish-0.000
pie-0.000
 seconds-0.000
 arrest-0.000
 eight-0.000
Token.
Token ID13
Feature activation+6.18e-03

Pos logit contributions
s+0.010
 jog+0.009
ashing+0.009
aa+0.008
act+0.008
Neg logit contributions
anish-0.011
pie-0.010
 seconds-0.010
 course-0.009
 arrest-0.009
Token She
Token ID1375
Feature activation-1.08e-02

Pos logit contributions
anish+0.013
pie+0.012
 seconds+0.011
 arrest+0.011
 course+0.011
Neg logit contributions
s-0.009
ashing-0.007
 jog-0.007
aa-0.007
act-0.007
Token was
Token ID373
Feature activation+1.76e-02

Pos logit contributions
s+0.018
ashing+0.016
 jog+0.016
aa+0.016
act+0.016
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.017
 eight-0.016
ra-0.016
Token surprised
Token ID6655
Feature activation-1.44e-02

Pos logit contributions
anish+0.041
pie+0.038
 course+0.036
 arrest+0.036
 seconds+0.035
Neg logit contributions
s-0.020
ashing-0.018
 jog-0.017
 especially-0.017
aa-0.016
Token when
Token ID618
Feature activation-1.40e-03

Pos logit contributions
s+0.021
 jog+0.020
ashing+0.019
aa+0.019
act+0.019
Neg logit contributions
anish-0.026
 seconds-0.025
pie-0.025
 course-0.024
 arrest-0.023
Token the
Token ID262
Feature activation-1.84e-03

Pos logit contributions
s+0.002
ashing+0.002
 jog+0.002
aa+0.002
act+0.002
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.003
 eight-0.003
 arrest-0.003
Token folder
Token ID9483
Feature activation+2.09e-02

Pos logit contributions
s+0.003
ashing+0.003
 jog+0.003
aa+0.003
act+0.002
Neg logit contributions
anish-0.003
pie-0.003
 seconds-0.003
 arrest-0.003
 eight-0.003
Token started
Token ID2067
Feature activation-1.95e-03

Pos logit contributions
anish+0.040
pie+0.038
 seconds+0.037
rum+0.035
 arrest+0.035
Neg logit contributions
s-0.032
 jog-0.027
ashing-0.026
act-0.024
aa-0.024
Token to
Token ID284
Feature activation+2.82e-03

Pos logit contributions
s+0.002
ashing+0.002
 jog+0.002
aa+0.002
act+0.002
Neg logit contributions
anish-0.004
pie-0.004
 seconds-0.004
 arrest-0.004
 eight-0.004
Token get
Token ID651
Feature activation-4.42e-03

Pos logit contributions
anish+0.006
pie+0.005
 seconds+0.005
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.004
ashing-0.004
 jog-0.003
aa-0.003
act-0.003
Token castle
Token ID16669
Feature activation+1.89e-02

Pos logit contributions
anish+0.010
pie+0.009
 seconds+0.008
 arrest+0.008
 course+0.008
Neg logit contributions
s-0.013
 jog-0.012
ashing-0.012
act-0.011
aa-0.011
Token and
Token ID290
Feature activation+1.31e-03

Pos logit contributions
anish+0.042
pie+0.039
 seconds+0.038
rum+0.036
ra+0.036
Neg logit contributions
s-0.025
 jog-0.021
ashing-0.021
aa-0.019
udd-0.018
Token saw
Token ID2497
Feature activation+1.32e-02

Pos logit contributions
anish+0.002
pie+0.002
 seconds+0.002
 arrest+0.002
 eight+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token a
Token ID257
Feature activation-1.14e-02

Pos logit contributions
anish+0.023
pie+0.021
 seconds+0.020
ra+0.019
 arrest+0.019
Neg logit contributions
s-0.023
 jog-0.020
ashing-0.019
aa-0.019
act-0.019
Token sad
Token ID6507
Feature activation-4.00e-03

Pos logit contributions
s+0.016
ashing+0.014
aa+0.013
act+0.013
 jog+0.013
Neg logit contributions
anish-0.023
pie-0.022
 seconds-0.021
 eight-0.021
 arrest-0.020
Token little
Token ID1310
Feature activation-5.83e-03

Pos logit contributions
s+0.007
ashing+0.006
 jog+0.006
act+0.006
aa+0.006
Neg logit contributions
anish-0.007
pie-0.007
 seconds-0.006
 arrest-0.006
 eight-0.006
Token girl
Token ID2576
Feature activation-1.01e-02

Pos logit contributions
s+0.009
ashing+0.008
 jog+0.008
act+0.008
aa+0.008
Neg logit contributions
anish-0.011
pie-0.010
 seconds-0.010
 arrest-0.009
 eight-0.009
Token.
Token ID13
Feature activation-1.11e-02

Pos logit contributions
s+0.017
act+0.015
ashing+0.015
aa+0.015
 jog+0.015
Neg logit contributions
anish-0.016
pie-0.016
 seconds-0.015
 arrest-0.015
 course-0.014
Token
Token ID220
Feature activation-3.12e-03

Pos logit contributions
s+0.016
ashing+0.015
act+0.015
 jog+0.015
aa+0.015
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.018
 eight-0.018
 arrest-0.018
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
s+0.004
 jog+0.004
ashing+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.006
pie-0.006
 seconds-0.006
 eight-0.005
 arrest-0.005
Token
Token ID198
Feature activation-1.24e-02

Pos logit contributions
s+0.027
ashing+0.027
act+0.026
ory+0.025
 jog+0.025
Neg logit contributions
anish-0.030
pie-0.028
 seconds-0.027
 course-0.027
 eight-0.027
Token for
Token ID329
Feature activation+4.99e-02

Pos logit contributions
s+0.026
act+0.025
ashing+0.024
 jog+0.024
aa+0.024
Neg logit contributions
anish-0.024
pie-0.023
 seconds-0.022
 eight-0.021
 nine-0.021
Token her
Token ID607
Feature activation+4.51e-03

Pos logit contributions
anish+0.095
pie+0.087
ra+0.085
aces+0.080
 Majesty+0.078
Neg logit contributions
 jog-0.073
 jazz-0.070
 warm-0.065
 especially-0.064
 early-0.062
Token help
Token ID1037
Feature activation-3.86e-03

Pos logit contributions
anish+0.007
pie+0.006
 seconds+0.006
 arrest+0.005
 eight+0.005
Neg logit contributions
s-0.009
 jog-0.008
ashing-0.008
aa-0.008
act-0.008
Token.
Token ID13
Feature activation+1.65e-02

Pos logit contributions
s+0.006
ashing+0.006
act+0.005
 jog+0.005
aa+0.005
Neg logit contributions
anish-0.007
pie-0.007
 seconds-0.006
 course-0.006
 arrest-0.006
Token She
Token ID1375
Feature activation+5.39e-04

Pos logit contributions
anish+0.039
pie+0.038
 seconds+0.036
 course+0.034
 arrest+0.034
Neg logit contributions
s-0.018
 jog-0.015
ashing-0.014
aa-0.014
act-0.013
Token knew
Token ID2993
Feature activation+4.44e-03

Pos logit contributions
anish+0.001
pie+0.001
 seconds+0.001
 course+0.001
 arrest+0.001
Neg logit contributions
s-0.001
ashing-0.001
 jog-0.001
aa-0.001
act-0.001
Token that
Token ID326
Feature activation-5.51e-03

Pos logit contributions
anish+0.008
pie+0.008
 seconds+0.008
 arrest+0.007
 eight+0.007
Neg logit contributions
s-0.007
ashing-0.006
 jog-0.006
aa-0.006
act-0.006
Token her
Token ID607
Feature activation+1.68e-03

Pos logit contributions
s+0.009
ashing+0.008
aa+0.008
ory+0.008
act+0.008
Neg logit contributions
anish-0.010
pie-0.009
 seconds-0.009
 course-0.009
 accomplishment-0.009
Token prayers
Token ID15707
Feature activation-2.25e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 arrest+0.003
 eight+0.003
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token were
Token ID547
Feature activation+5.37e-03

Pos logit contributions
s+0.031
ashing+0.029
act+0.028
 jog+0.028
aa+0.028
Neg logit contributions
anish-0.043
pie-0.041
 seconds-0.040
 course-0.039
ra-0.038
Token answered
Token ID9373
Feature activation-8.08e-03

Pos logit contributions
anish+0.010
pie+0.010
 seconds+0.009
 arrest+0.009
 eight+0.009
Neg logit contributions
s-0.008
ashing-0.007
 jog-0.007
aa-0.007
act-0.007
Token What
Token ID1867
Feature activation-8.93e-03

Pos logit contributions
s+0.025
act+0.023
ashing+0.023
aa+0.022
 jog+0.022
Neg logit contributions
anish-0.019
pie-0.018
 seconds-0.017
 eight-0.017
ra-0.016
Token are
Token ID389
Feature activation+4.70e-03

Pos logit contributions
s+0.015
ashing+0.015
aa+0.014
ory+0.014
act+0.014
Neg logit contributions
anish-0.015
pie-0.014
 seconds-0.013
 eight-0.013
 course-0.013
Token you
Token ID345
Feature activation+1.35e-02

Pos logit contributions
anish+0.008
pie+0.008
 seconds+0.007
 arrest+0.007
 course+0.007
Neg logit contributions
s-0.008
ashing-0.007
 jog-0.007
aa-0.007
act-0.007
Token doing
Token ID1804
Feature activation+6.76e-03

Pos logit contributions
anish+0.022
pie+0.021
 seconds+0.020
 arrest+0.019
 course+0.019
Neg logit contributions
s-0.024
 jog-0.022
ashing-0.022
aa-0.021
act-0.021
Token?"
Token ID1701
Feature activation-1.56e-02

Pos logit contributions
anish+0.013
pie+0.012
 seconds+0.011
 arrest+0.011
 eight+0.011
Neg logit contributions
s-0.011
 jog-0.009
ashing-0.009
aa-0.009
act-0.009
Token
Token ID198
Feature activation-1.77e-02

Pos logit contributions
s+0.026
ashing+0.025
act+0.025
udd+0.023
 jog+0.023
Neg logit contributions
anish-0.026
 seconds-0.024
pie-0.024
 eight-0.023
 course-0.022
Token
Token ID198
Feature activation-6.23e-03

Pos logit contributions
ashing+0.028
s+0.028
act+0.027
ory+0.026
 jog+0.026
Neg logit contributions
anish-0.030
 seconds-0.028
pie-0.027
 eight-0.027
 course-0.026
TokenA
Token ID32
Feature activation+1.54e-02

Pos logit contributions
s+0.010
ashing+0.009
act+0.009
 jog+0.008
aa+0.008
Neg logit contributions
anish-0.012
pie-0.011
 seconds-0.011
 eight-0.010
 course-0.010
Token voice
Token ID3809
Feature activation+1.37e-03

Pos logit contributions
anish+0.029
pie+0.028
 arrest+0.026
 seconds+0.026
ps+0.025
Neg logit contributions
s-0.024
 jog-0.021
aa-0.020
ashing-0.019
ory-0.019
Token answers
Token ID7429
Feature activation+1.02e-02

Pos logit contributions
anish+0.002
pie+0.002
 seconds+0.002
 course+0.002
 arrest+0.002
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token,
Token ID11
Feature activation+2.50e-02

Pos logit contributions
anish+0.018
pie+0.017
 seconds+0.016
 arrest+0.016
 accomplishment+0.015
Neg logit contributions
s-0.017
 jog-0.015
ashing-0.014
aa-0.014
act-0.014
Token and
Token ID290
Feature activation+7.52e-03

Pos logit contributions
anish+0.020
pie+0.019
 seconds+0.019
 arrest+0.018
 course+0.018
Neg logit contributions
s-0.016
ashing-0.013
 jog-0.013
aa-0.013
act-0.013
Token sprinkled
Token ID49692
Feature activation+2.25e-02

Pos logit contributions
anish+0.014
pie+0.013
 seconds+0.012
 arrest+0.012
 course+0.012
Neg logit contributions
s-0.012
 jog-0.011
ashing-0.011
aa-0.010
act-0.010
Token some
Token ID617
Feature activation-2.81e-03

Pos logit contributions
anish+0.041
pie+0.038
 arrest+0.035
 seconds+0.035
rum+0.035
Neg logit contributions
s-0.038
 jog-0.032
aa-0.031
act-0.031
ashing-0.030
Token glitter
Token ID31133
Feature activation-2.59e-02

Pos logit contributions
s+0.005
ashing+0.004
 jog+0.004
aa+0.004
act+0.004
Neg logit contributions
anish-0.005
pie-0.004
 seconds-0.004
 eight-0.004
 course-0.004
Token.
Token ID13
Feature activation-8.67e-03

Pos logit contributions
 jog+0.039
ashing+0.038
s+0.036
act+0.035
 shepherd+0.035
Neg logit contributions
anish-0.044
pie-0.043
ra-0.041
 seconds-0.040
ps-0.040
Token
Token ID198
Feature activation+2.05e-04

Pos logit contributions
s+0.013
ashing+0.011
 jog+0.011
aa+0.011
act+0.011
Neg logit contributions
anish-0.016
pie-0.015
 seconds-0.015
 eight-0.015
 nine-0.014
Token
Token ID198
Feature activation+2.77e-03

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 eight+0.000
 arrest+0.000
Neg logit contributions
s-0.000
ashing-0.000
 jog-0.000
aa-0.000
act-0.000
TokenThey
Token ID2990
Feature activation+8.00e-03

Pos logit contributions
anish+0.004
pie+0.004
 seconds+0.004
 arrest+0.003
 eight+0.003
Neg logit contributions
s-0.005
ashing-0.005
 jog-0.005
aa-0.005
act-0.005
Token looked
Token ID3114
Feature activation-2.70e-02

Pos logit contributions
anish+0.015
pie+0.014
 seconds+0.014
 arrest+0.013
 course+0.013
Neg logit contributions
s-0.012
 jog-0.011
ashing-0.011
aa-0.010
act-0.010
Token at
Token ID379
Feature activation-1.62e-02

Pos logit contributions
s+0.042
aa+0.041
 jog+0.039
Kids+0.038
act+0.038
Neg logit contributions
 seconds-0.043
pie-0.041
 nine-0.041
 six-0.041
 eight-0.041
Token each
Token ID1123
Feature activation+1.56e-02

Pos logit contributions
s+0.028
ashing+0.027
aa+0.026
 jog+0.025
act+0.025
Neg logit contributions
anish-0.025
pie-0.025
 seconds-0.025
 course-0.024
 eight-0.023
Token me
Token ID502
Feature activation-1.84e-02

Pos logit contributions
s+0.084
cases+0.080
ashing+0.079
ido+0.076
aa+0.075
Neg logit contributions
 eight-0.057
pie-0.053
 six-0.050
anish-0.050
 course-0.050
Tokenadow
Token ID4584
Feature activation+2.37e-03

Pos logit contributions
s+0.031
 jog+0.028
 sax+0.027
ory+0.026
ashing+0.026
Neg logit contributions
pie-0.030
anish-0.029
 seconds-0.028
 eight-0.027
ps-0.026
Token,
Token ID11
Feature activation-1.38e-02

Pos logit contributions
anish+0.005
pie+0.005
 seconds+0.004
 arrest+0.004
 eight+0.004
Neg logit contributions
s-0.003
ashing-0.003
 jog-0.003
aa-0.003
act-0.003
Token making
Token ID1642
Feature activation-4.56e-03

Pos logit contributions
s+0.020
act+0.017
aa+0.017
 jog+0.017
ashing+0.017
Neg logit contributions
anish-0.026
pie-0.026
 seconds-0.025
 eight-0.024
 course-0.024
Token all
Token ID477
Feature activation-7.21e-03

Pos logit contributions
s+0.008
ashing+0.007
 jog+0.007
aa+0.007
ory+0.007
Neg logit contributions
anish-0.007
pie-0.007
 seconds-0.007
 course-0.006
 eight-0.006
Token the
Token ID262
Feature activation-5.01e-02

Pos logit contributions
s+0.014
ashing+0.012
 jog+0.012
aa+0.012
ory+0.012
Neg logit contributions
anish-0.010
pie-0.010
 seconds-0.009
 course-0.009
 eight-0.009
Token grass
Token ID8701
Feature activation-4.88e-02

Pos logit contributions
s+0.093
cases+0.089
ding+0.087
ashing+0.087
aa+0.085
Neg logit contributions
 course-0.061
 eight-0.060
anish-0.060
 seconds-0.059
 inches-0.058
Token spark
Token ID9009
Feature activation-4.57e-02

Pos logit contributions
Kids+0.082
ido+0.081
ory+0.079
 jog+0.079
act+0.078
Neg logit contributions
pie-0.069
anish-0.067
ps-0.064
 itself-0.064
 eight-0.062
Tokenle
Token ID293
Feature activation-2.62e-02

Pos logit contributions
ido+0.075
 jog+0.072
ashing+0.070
Kids+0.068
 café+0.067
Neg logit contributions
ps-0.069
anish-0.069
pie-0.069
 seconds-0.068
 itself-0.065
Token.
Token ID13
Feature activation+8.28e-03

Pos logit contributions
 jog+0.043
ashing+0.042
s+0.041
act+0.041
aa+0.040
Neg logit contributions
anish-0.040
pie-0.039
 seconds-0.038
ps-0.036
 eight-0.036
Token
Token ID198
Feature activation+5.61e-03

Pos logit contributions
anish+0.015
pie+0.014
 seconds+0.013
 arrest+0.013
ra+0.013
Neg logit contributions
s-0.014
ashing-0.012
 jog-0.012
aa-0.012
act-0.011
Token on
Token ID319
Feature activation+2.06e-03

Pos logit contributions
anish+0.013
pie+0.012
 seconds+0.011
 arrest+0.011
 eight+0.011
Neg logit contributions
s-0.010
 jog-0.009
ashing-0.009
aa-0.008
act-0.008
Token,
Token ID11
Feature activation-1.01e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 arrest+0.002
 course+0.002
Neg logit contributions
s-0.004
ashing-0.004
 jog-0.004
aa-0.004
act-0.004
Token Lily
Token ID20037
Feature activation-1.13e-02

Pos logit contributions
s+0.015
ashing+0.014
aa+0.014
act+0.014
 jog+0.013
Neg logit contributions
anish-0.019
pie-0.017
 seconds-0.017
 course-0.017
 accomplishment-0.016
Token remembered
Token ID12086
Feature activation-2.07e-02

Pos logit contributions
s+0.018
ashing+0.017
ory+0.016
aa+0.016
 jog+0.016
Neg logit contributions
anish-0.019
pie-0.019
 seconds-0.018
 eight-0.017
 arrest-0.017
Token to
Token ID284
Feature activation+1.67e-03

Pos logit contributions
s+0.037
ashing+0.033
aa+0.032
act+0.031
ory+0.031
Neg logit contributions
anish-0.033
 seconds-0.032
pie-0.031
 eight-0.030
 arrest-0.030
Token listen
Token ID6004
Feature activation-4.85e-02

Pos logit contributions
anish+0.003
pie+0.003
 seconds+0.003
 arrest+0.003
 course+0.003
Neg logit contributions
s-0.002
ashing-0.002
 jog-0.002
aa-0.002
act-0.002
Token to
Token ID284
Feature activation-7.20e-03

Pos logit contributions
act+0.080
ory+0.080
glers+0.080
actus+0.076
aa+0.076
Neg logit contributions
 seconds-0.074
anish-0.068
 eight-0.067
 example-0.066
 accomplishment-0.064
Token her
Token ID607
Feature activation-2.22e-02

Pos logit contributions
s+0.011
ashing+0.010
ory+0.010
act+0.009
aa+0.009
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.013
 accomplishment-0.012
 eight-0.012
Token breathing
Token ID12704
Feature activation-1.80e-02

Pos logit contributions
s+0.039
ashing+0.037
ory+0.036
act+0.036
aa+0.035
Neg logit contributions
 seconds-0.034
anish-0.034
 accomplishment-0.034
pie-0.031
 eight-0.031
Token and
Token ID290
Feature activation-1.54e-02

Pos logit contributions
s+0.028
ory+0.026
ashing+0.026
aa+0.025
act+0.025
Neg logit contributions
anish-0.033
 seconds-0.031
pie-0.031
 accomplishment-0.029
ps-0.029
Token use
Token ID779
Feature activation-5.76e-03

Pos logit contributions
s+0.027
ashing+0.025
ory+0.025
act+0.025
aa+0.024
Neg logit contributions
anish-0.025
pie-0.023
 seconds-0.023
 accomplishment-0.022
 arrest-0.021
Token flowers
Token ID12734
Feature activation+1.54e-02

Pos logit contributions
s+0.045
ashing+0.042
aa+0.042
ory+0.041
act+0.040
Neg logit contributions
pie-0.039
anish-0.039
 eight-0.038
 seconds-0.038
 arrest-0.037
Token.
Token ID13
Feature activation-1.92e-02

Pos logit contributions
anish+0.032
pie+0.030
 seconds+0.029
ra+0.028
 arrest+0.028
Neg logit contributions
s-0.023
 jog-0.018
ashing-0.018
act-0.017
aa-0.017
Token One
Token ID1881
Feature activation-4.50e-02

Pos logit contributions
s+0.035
aa+0.034
act+0.033
ashing+0.032
ory+0.032
Neg logit contributions
anish-0.027
pie-0.025
 course-0.024
 seconds-0.023
 nine-0.023
Token day
Token ID1110
Feature activation+1.07e-02

Pos logit contributions
cases+0.098
act+0.098
s+0.094
ory+0.092
itionally+0.091
Neg logit contributions
 course-0.041
anish-0.040
 eight-0.038
 seconds-0.037
 half-0.037
Token,
Token ID11
Feature activation+1.92e-04

Pos logit contributions
anish+0.017
pie+0.017
 seconds+0.016
 arrest+0.015
 accomplishment+0.014
Neg logit contributions
s-0.020
 jog-0.017
ashing-0.017
aa-0.017
act-0.016
Token Lily
Token ID20037
Feature activation-4.80e-02

Pos logit contributions
anish+0.000
pie+0.000
 seconds+0.000
 eight+0.000
 course+0.000
Neg logit contributions
s-0.000
ashing-0.000
 jog-0.000
aa-0.000
act-0.000
Token's
Token ID338
Feature activation-2.57e-02

Pos logit contributions
 numb+0.082
ido+0.080
 sax+0.080
ashing+0.079
aa+0.078
Neg logit contributions
 arrest-0.061
anish-0.060
pie-0.058
ra-0.057
ASH-0.054
Token mom
Token ID1995
Feature activation-1.05e-02

Pos logit contributions
s+0.047
ashing+0.041
act+0.041
ory+0.041
aa+0.040
Neg logit contributions
anish-0.040
 eight-0.036
pie-0.035
 seconds-0.034
ra-0.034
Token told
Token ID1297
Feature activation-2.63e-02

Pos logit contributions
s+0.015
 jog+0.014
ashing+0.013
act+0.013
aa+0.013
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.018
 course-0.018
 eight-0.017
Token her
Token ID607
Feature activation+7.33e-03

Pos logit contributions
s+0.047
act+0.045
ashing+0.044
actus+0.042
ory+0.041
Neg logit contributions
anish-0.040
 seconds-0.038
 course-0.037
pie-0.036
 arrest-0.035
Token they
Token ID484
Feature activation-1.73e-02

Pos logit contributions
anish+0.015
pie+0.014
 seconds+0.014
 arrest+0.013
 course+0.013
Neg logit contributions
s-0.010
 jog-0.009
ashing-0.009
aa-0.008
act-0.008
Token it
Token ID340
Feature activation-1.21e-02

Pos logit contributions
s+0.019
ashing+0.018
aa+0.017
 jog+0.017
ory+0.017
Neg logit contributions
anish-0.018
pie-0.017
 seconds-0.016
 arrest-0.015
 eight-0.015
Token's
Token ID338
Feature activation-2.66e-02

Pos logit contributions
s+0.019
ashing+0.017
ory+0.017
 jog+0.017
act+0.017
Neg logit contributions
anish-0.022
pie-0.021
 seconds-0.020
 course-0.019
 eight-0.019
Token going
Token ID1016
Feature activation-4.27e-02

Pos logit contributions
s+0.045
ory+0.039
act+0.039
aa+0.038
ashing+0.038
Neg logit contributions
anish-0.044
 seconds-0.044
pie-0.043
ps-0.041
 eight-0.041
Token to
Token ID284
Feature activation-1.04e-02

Pos logit contributions
s+0.064
act+0.061
ory+0.060
aa+0.059
ashing+0.059
Neg logit contributions
pie-0.068
anish-0.067
 eight-0.064
 nine-0.064
 six-0.063
Token rain
Token ID6290
Feature activation-5.07e-02

Pos logit contributions
s+0.019
ashing+0.017
aa+0.016
act+0.016
ory+0.016
Neg logit contributions
anish-0.017
pie-0.016
 seconds-0.015
 arrest-0.015
 course-0.015
Token or
Token ID393
Feature activation-5.30e-02

Pos logit contributions
s+0.081
ory+0.079
act+0.074
actus+0.074
aa+0.074
Neg logit contributions
anish-0.080
pie-0.078
ps-0.070
 seconds-0.070
 course-0.068
Token if
Token ID611
Feature activation-8.22e-03

Pos logit contributions
s+0.087
act+0.082
ory+0.082
ashing+0.081
aa+0.080
Neg logit contributions
pie-0.085
anish-0.083
 eight-0.079
 seconds-0.079
 arrest-0.078
Token there
Token ID612
Feature activation-2.10e-02

Pos logit contributions
s+0.014
ashing+0.013
aa+0.012
 jog+0.012
act+0.012
Neg logit contributions
anish-0.014
pie-0.013
 seconds-0.013
 arrest-0.012
 eight-0.012
Token's
Token ID338
Feature activation-1.40e-02

Pos logit contributions
s+0.034
ashing+0.034
ory+0.032
aa+0.032
act+0.031
Neg logit contributions
anish-0.036
pie-0.034
 seconds-0.032
 course-0.031
 eight-0.030
Token a
Token ID257
Feature activation-2.44e-02

Pos logit contributions
s+0.028
ashing+0.026
aa+0.025
act+0.025
ory+0.025
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.018
 eight-0.017
 nine-0.017
Token big
Token ID1263
Feature activation+1.85e-02

Pos logit contributions
s+0.041
ashing+0.036
act+0.035
aa+0.035
ory+0.035
Neg logit contributions
anish-0.042
 seconds-0.039
pie-0.039
 course-0.037
 eight-0.036
Token't
Token ID470
Feature activation-2.55e-02

Pos logit contributions
ashing+0.066
s+0.065
act+0.064
ory+0.063
udd+0.062
Neg logit contributions
anish-0.022
pie-0.020
 seconds-0.019
 arrest-0.018
 eight-0.018
Token want
Token ID765
Feature activation-2.59e-02

Pos logit contributions
s+0.042
act+0.036
actus+0.035
udd+0.035
ashing+0.033
Neg logit contributions
anish-0.045
pie-0.044
 seconds-0.044
rum-0.040
 eight-0.040
Token to
Token ID284
Feature activation-3.62e-02

Pos logit contributions
s+0.040
ashing+0.038
udd+0.035
aa+0.034
act+0.034
Neg logit contributions
 seconds-0.047
anish-0.046
pie-0.045
 course-0.044
 eight-0.043
Token get
Token ID651
Feature activation-3.34e-02

Pos logit contributions
s+0.066
ashing+0.060
udd+0.059
act+0.058
actus+0.058
Neg logit contributions
anish-0.055
pie-0.053
 arrest-0.051
 seconds-0.051
ra-0.049
Token ice
Token ID4771
Feature activation-4.22e-02

Pos logit contributions
s+0.054
udd+0.050
act+0.049
ashing+0.049
actus+0.048
Neg logit contributions
pie-0.055
 seconds-0.055
 eight-0.052
ps-0.051
 nine-0.050
Token cream
Token ID8566
Feature activation-5.19e-02

Pos logit contributions
 shepherd+0.080
 jog+0.080
itionally+0.077
s+0.077
 ranc+0.076
Neg logit contributions
anish-0.047
pie-0.047
 seconds-0.047
 inches-0.044
ps-0.044
Token there
Token ID612
Feature activation-1.78e-02

Pos logit contributions
act+0.096
ashing+0.095
s+0.094
 jog+0.093
aa+0.092
Neg logit contributions
anish-0.063
pie-0.061
 seconds-0.059
 course-0.059
ra-0.056
Token.
Token ID13
Feature activation-1.11e-02

Pos logit contributions
s+0.031
ashing+0.028
act+0.028
aa+0.028
 jog+0.028
Neg logit contributions
anish-0.029
pie-0.027
 seconds-0.026
 course-0.025
 eight-0.024
Token
Token ID198
Feature activation-1.10e-02

Pos logit contributions
s+0.017
ashing+0.015
 jog+0.015
aa+0.015
act+0.015
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.018
 eight-0.018
 course-0.017
Token
Token ID198
Feature activation-1.78e-02

Pos logit contributions
s+0.016
ashing+0.015
 jog+0.015
act+0.014
ory+0.014
Neg logit contributions
anish-0.020
pie-0.019
 seconds-0.019
 eight-0.018
 course-0.018
TokenThe
Token ID464
Feature activation-1.49e-03

Pos logit contributions
s+0.030
ashing+0.027
act+0.026
 jog+0.025
aa+0.025
Neg logit contributions
anish-0.030
pie-0.029
 seconds-0.028
 arrest-0.027
 eight-0.026