TOP ACTIVATIONS
MAX = 0.130

happily started to go down it . Suddenly ,
when she was walking outside it started to snow , and
almost every day sliding down it . One day it was
When it reached the bottom it went rolling off into the
he climbed on top of it , he realized that it
hill and wanted to climb it , but it looked too
that day on , whenever it r ained , the gir
could appear and disappear whenever it wanted . The people in
tree and decided to climb it . Tim went first and
the afternoon bouncing until soon it was time to go home
tree with many mushrooms under it . Tim wanted to look
tree and decided to climb it . As Tom
and seek all afternoon until it was time for Sam to
Finally the day came when it was time for Anna to
that day on , whenever it r ained , Lily would
One day , when it was very warm and sunny
that day on , whenever it r ained , Lily would
rolling through the corridors when it ran into a tiny mouse
. One day , when it was very cold outside ,
that day on , whenever it r ained , Lily would

MIN ACTIVATIONS
MIN = -0.137

. He also has a sle igh and re ind eer
burn you !" They wear apr ons and hats and act
I also have a magic sle igh and re ind eer
Tom said , holding a stuffed dog as his bird .
outside and showed them his sle igh and re ind eer
put a shiny saddle and brid le on him . Finally
. She grabbed the fire extingu isher and put out the
put a saddle and a brid le on the horse .
. He gave her a stuffed bear and a night light
trial . They used a stuffed dog and a stuffed cat
come in ," said his grand pa . They all turned
so he gave her a stuffed bear . The
girl and she had a stuffed bear called Bear y .
quickly checked for a fire extingu isher , but there wasn
Tom and Lily thanked their grand pa and ran to the
and thanked her grandma and grand pa . She said they
They thanked their grandma and grand pa in their hearts and
. She had a blue stuffed animal named Paul .
a stuffed dog and a stuffed cat as the witnesses .
The kid had a grand old time . He laughed

INTERVAL 0.063 - 0.130
CONTAINS 3.248%

<|endoftext|> Once upon a time there was a yellow bear who
was very careful to do it just like her mummy and
day , the boy walked down the street and stopped in
her pocket . She put it in a special box and
y ummy , she ate it all up .

INTERVAL -0.004 - 0.063
CONTAINS 52.871%

She does not know what she did wrong . She pulls
every hot day and had the best time . <|endoftext|> Once
happy to hear from him . Lily smiled , knowing that
like to play in the park . One day , they
Tom and Mia get the je ep back .

INTERVAL -0.071 - -0.004
CONTAINS 42.518%

took out two past els from the box . One was
big claws and picked up a big fish .
it must have blown in from the picnic they were planning
, and then he started to play on the stair .
carry Anna and he can 't leave her alone . He

INTERVAL -0.137 - -0.071
CONTAINS 1.364%

to his mom 's warning and kept his lungs safe .
waved their fins at Ben and Mia . They thanked them
M ia hugged her mom and dad . She said
, Lily reminded her dad and he thanked her for being
to their room . They gave him food and water .
Token happily
Token ID18177
Feature activation-9.47e-03

Pos logit contributions
 envelop+0.092
Offic+0.087
ons+0.082
 Amen+0.078
 read+0.077
Neg logit contributions
tub-0.123
gan-0.113
ider-0.113
ses-0.104
 Moe-0.103
Token started
Token ID2067
Feature activation+5.92e-02

Pos logit contributions
 envelop+0.017
Offic+0.015
 Amen+0.015
ons+0.014
oc+0.013
Neg logit contributions
tub-0.019
gan-0.019
 Moe-0.017
ider-0.017
ahl-0.016
Token to
Token ID284
Feature activation-2.16e-02

Pos logit contributions
tub+0.106
gan+0.100
 Moe+0.090
ider+0.087
 Mil+0.085
Neg logit contributions
 envelop-0.123
Offic-0.114
 Amen-0.113
ons-0.106
Mic-0.104
Token go
Token ID467
Feature activation+1.34e-02

Pos logit contributions
 envelop+0.042
Offic+0.039
 Amen+0.038
ons+0.036
Mic+0.035
Neg logit contributions
tub-0.041
gan-0.039
 Moe-0.035
ider-0.034
 Mil-0.034
Token down
Token ID866
Feature activation+8.19e-02

Pos logit contributions
tub+0.024
gan+0.023
ider+0.020
 Moe+0.020
ses+0.020
Neg logit contributions
 envelop-0.027
Offic-0.025
 Amen-0.025
ons-0.024
Mic-0.022
Token it
Token ID340
Feature activation+1.28e-01

Pos logit contributions
tub+0.159
gan+0.151
ider+0.136
ses+0.133
 Moe+0.132
Neg logit contributions
 envelop-0.156
Offic-0.144
 Amen-0.140
ons-0.137
oc-0.123
Token.
Token ID13
Feature activation+1.32e-02

Pos logit contributions
tub+0.253
gan+0.248
ses+0.230
sed+0.223
 Moe+0.222
Neg logit contributions
 envelop-0.217
ons-0.192
Offic-0.190
 read-0.187
 Amen-0.182
Token
Token ID198
Feature activation-1.35e-02

Pos logit contributions
tub+0.023
gan+0.021
ider+0.020
 Moe+0.019
ahl+0.018
Neg logit contributions
 envelop-0.028
Offic-0.026
 Amen-0.026
ons-0.025
Mic-0.023
Token
Token ID198
Feature activation-9.88e-03

Pos logit contributions
 envelop+0.035
Offic+0.032
 Amen+0.032
ons+0.030
itivity+0.030
Neg logit contributions
tub-0.017
gan-0.015
 Moe-0.013
ahl-0.012
 Mil-0.012
TokenSuddenly
Token ID38582
Feature activation+5.34e-02

Pos logit contributions
 envelop+0.019
 Amen+0.017
ons+0.016
Offic+0.016
 parcels+0.015
Neg logit contributions
tub-0.019
gan-0.018
ider-0.017
 Moe-0.016
ahl-0.016
Token,
Token ID11
Feature activation+5.03e-02

Pos logit contributions
tub+0.131
gan+0.115
ider+0.115
ipples+0.113
ses+0.108
Neg logit contributions
 envelop-0.078
ons-0.076
 read-0.064
Offic-0.061
 Amen-0.059
Token when
Token ID618
Feature activation+5.55e-02

Pos logit contributions
tub+0.124
gan+0.113
ider+0.104
ses+0.103
ahl+0.101
Neg logit contributions
 envelop-0.139
ons-0.128
Offic-0.124
 Amen-0.118
 read-0.114
Token she
Token ID673
Feature activation+5.19e-02

Pos logit contributions
tub+0.093
gan+0.088
ider+0.078
 Moe+0.076
ahl+0.074
Neg logit contributions
 envelop-0.122
Offic-0.114
 Amen-0.111
ons-0.108
Mic-0.102
Token was
Token ID373
Feature activation+3.51e-02

Pos logit contributions
tub+0.085
gan+0.083
ses+0.073
ider+0.073
 Moe+0.071
Neg logit contributions
 envelop-0.110
Offic-0.104
 Amen-0.103
ons-0.102
oc-0.098
Token walking
Token ID6155
Feature activation+2.09e-02

Pos logit contributions
tub+0.076
gan+0.073
 Moe+0.067
ider+0.066
ahl+0.065
Neg logit contributions
 envelop-0.058
Offic-0.053
 Amen-0.052
ons-0.050
Mic-0.045
Token outside
Token ID2354
Feature activation+8.09e-02

Pos logit contributions
gan+0.046
tub+0.045
ses+0.044
 fault+0.043
sed+0.042
Neg logit contributions
 envelop-0.029
 Amen-0.027
ons-0.026
 read-0.026
 saddle-0.025
Token it
Token ID340
Feature activation+1.26e-01

Pos logit contributions
gan+0.168
tub+0.161
ider+0.154
ses+0.153
ahl+0.147
Neg logit contributions
 envelop-0.134
ons-0.132
 read-0.120
 Amen-0.118
Offic-0.114
Token started
Token ID2067
Feature activation+8.42e-02

Pos logit contributions
gan+0.310
 fault+0.294
ses+0.292
sed+0.274
ider+0.273
Neg logit contributions
 read-0.173
ons-0.172
 envelop-0.143
Offic-0.131
 chat-0.131
Token to
Token ID284
Feature activation-8.38e-03

Pos logit contributions
gan+0.184
tub+0.180
ahl+0.167
ses+0.167
 Moe+0.165
Neg logit contributions
 envelop-0.127
ons-0.115
Offic-0.108
 read-0.104
 Amen-0.102
Token snow
Token ID6729
Feature activation+3.99e-02

Pos logit contributions
 envelop+0.016
Offic+0.014
 Amen+0.014
 read+0.013
ons+0.013
Neg logit contributions
gan-0.016
tub-0.016
ider-0.015
 Moe-0.014
ahl-0.014
Token,
Token ID11
Feature activation+3.22e-02

Pos logit contributions
gan+0.076
ses+0.071
ahl+0.070
tub+0.070
 Moe+0.070
Neg logit contributions
 envelop-0.062
ons-0.060
Offic-0.056
 Amen-0.056
oc-0.055
Token and
Token ID290
Feature activation-3.91e-02

Pos logit contributions
tub+0.065
gan+0.061
ider+0.060
ahl+0.057
ses+0.057
Neg logit contributions
 envelop-0.057
ons-0.051
Offic-0.050
 read-0.047
 Amen-0.047
Token almost
Token ID2048
Feature activation+3.47e-02

Pos logit contributions
 envelop+0.015
Offic+0.015
 Amen+0.014
Mic+0.013
oc+0.012
Neg logit contributions
tub-0.012
gan-0.011
 Moe-0.010
 Mil-0.009
ider-0.009
Token every
Token ID790
Feature activation-7.87e-04

Pos logit contributions
tub+0.073
gan+0.067
ahl+0.061
ses+0.061
 Moe+0.061
Neg logit contributions
 envelop-0.055
ons-0.053
Offic-0.052
 read-0.049
 rode-0.048
Token day
Token ID1110
Feature activation+4.70e-02

Pos logit contributions
 envelop+0.002
Offic+0.002
 Amen+0.002
ons+0.001
Mic+0.001
Neg logit contributions
tub-0.001
gan-0.001
 Moe-0.001
ider-0.001
ahl-0.001
Token sliding
Token ID22292
Feature activation+2.82e-02

Pos logit contributions
tub+0.079
gan+0.074
ider+0.067
 Moe+0.066
 Mil+0.062
Neg logit contributions
 envelop-0.102
Offic-0.096
 Amen-0.094
ons-0.089
Mic-0.087
Token down
Token ID866
Feature activation+8.54e-02

Pos logit contributions
tub+0.060
gan+0.059
ses+0.057
ider+0.055
sed+0.054
Neg logit contributions
 envelop-0.043
ons-0.039
Offic-0.037
 saddle-0.035
 Amen-0.035
Token it
Token ID340
Feature activation+1.25e-01

Pos logit contributions
tub+0.176
gan+0.174
ses+0.161
ider+0.159
sed+0.158
Neg logit contributions
 envelop-0.146
ons-0.130
Offic-0.129
 Amen-0.120
oc-0.111
Token.
Token ID13
Feature activation+1.54e-02

Pos logit contributions
gan+0.231
tub+0.230
ses+0.204
 Moe+0.203
ider+0.200
Neg logit contributions
 envelop-0.238
ons-0.214
Offic-0.209
 Amen-0.201
oc-0.192
Token One
Token ID1881
Feature activation+1.12e-02

Pos logit contributions
tub+0.031
ider+0.030
gan+0.030
ses+0.029
sed+0.029
Neg logit contributions
 envelop-0.025
 Amen-0.023
oc-0.023
Offic-0.023
ons-0.020
Token day
Token ID1110
Feature activation+6.62e-02

Pos logit contributions
tub+0.025
ider+0.024
ses+0.024
gan+0.024
sed+0.024
Neg logit contributions
 envelop-0.015
ons-0.014
Offic-0.014
oc-0.014
 read-0.013
Token it
Token ID340
Feature activation+1.06e-01

Pos logit contributions
tub+0.141
ses+0.138
gan+0.130
ider+0.127
sed+0.125
Neg logit contributions
 envelop-0.104
ons-0.098
oc-0.092
Offic-0.087
 rode-0.087
Token was
Token ID373
Feature activation+3.56e-02

Pos logit contributions
ses+0.241
ider+0.227
sed+0.218
tub+0.213
gan+0.210
Neg logit contributions
 rode-0.138
 read-0.131
Offic-0.130
 envelop-0.128
ons-0.125
Token When
Token ID1649
Feature activation+4.50e-02

Pos logit contributions
tub+0.027
gan+0.025
ider+0.024
ahl+0.023
ses+0.023
Neg logit contributions
 envelop-0.031
Offic-0.030
 Amen-0.028
ons-0.027
Mic-0.027
Token it
Token ID340
Feature activation+1.05e-01

Pos logit contributions
tub+0.105
gan+0.100
ider+0.096
ses+0.091
ahl+0.090
Neg logit contributions
 envelop-0.068
Offic-0.061
ons-0.059
 Amen-0.058
 read-0.050
Token reached
Token ID4251
Feature activation+4.04e-02

Pos logit contributions
ses+0.226
tub+0.221
 fault+0.216
gan+0.214
ider+0.206
Neg logit contributions
 rode-0.144
ons-0.140
Offic-0.140
 read-0.139
oc-0.130
Token the
Token ID262
Feature activation+5.99e-02

Pos logit contributions
ider+0.118
tub+0.114
inter+0.110
gan+0.108
auc+0.105
Neg logit contributions
 envelop-0.035
ons-0.024
Offic-0.018
 exchange-0.018
 helicopters-0.017
Token bottom
Token ID4220
Feature activation+9.96e-02

Pos logit contributions
gan+0.105
tub+0.105
ider+0.093
ahl+0.092
 Mil+0.090
Neg logit contributions
 envelop-0.119
Offic-0.110
 Amen-0.102
 saddle-0.098
oc-0.097
Token it
Token ID340
Feature activation+1.24e-01

Pos logit contributions
ses+0.213
gan+0.205
tub+0.204
ahl+0.201
sed+0.199
Neg logit contributions
ons-0.149
 envelop-0.137
Offic-0.109
oc-0.108
 read-0.106
Token went
Token ID1816
Feature activation+3.39e-02

Pos logit contributions
gan+0.266
ses+0.262
tub+0.252
ider+0.244
 fault+0.244
Neg logit contributions
ons-0.163
 rode-0.162
 read-0.152
 envelop-0.134
 chat-0.132
Token rolling
Token ID10708
Feature activation+2.42e-02

Pos logit contributions
gan+0.068
tub+0.067
ses+0.065
sed+0.060
 Mil+0.059
Neg logit contributions
 envelop-0.055
 Amen-0.050
Offic-0.049
ons-0.047
 saddle-0.045
Token off
Token ID572
Feature activation+3.27e-02

Pos logit contributions
tub+0.053
gan+0.052
ider+0.051
ses+0.049
sed+0.047
Neg logit contributions
 envelop-0.034
ons-0.029
 saddle-0.027
 Amen-0.026
Offic-0.026
Token into
Token ID656
Feature activation+2.34e-02

Pos logit contributions
tub+0.071
gan+0.070
ider+0.065
ses+0.063
ahl+0.062
Neg logit contributions
 envelop-0.052
ons-0.045
Offic-0.043
 Amen-0.041
 arrest-0.038
Token the
Token ID262
Feature activation+5.62e-02

Pos logit contributions
tub+0.049
gan+0.047
ider+0.044
 Moe+0.041
ahl+0.041
Neg logit contributions
 envelop-0.042
Offic-0.038
 Amen-0.036
ons-0.035
Mic-0.032
Token he
Token ID339
Feature activation+5.44e-02

Pos logit contributions
tub+0.078
gan+0.072
 Moe+0.063
ider+0.063
ahl+0.060
Neg logit contributions
 envelop-0.118
Offic-0.111
 Amen-0.108
ons-0.105
Mic-0.101
Token climbed
Token ID19952
Feature activation+1.52e-02

Pos logit contributions
tub+0.101
gan+0.096
 Moe+0.085
ider+0.085
ahl+0.082
Neg logit contributions
 envelop-0.108
Offic-0.101
 Amen-0.098
ons-0.095
Mic-0.090
Token on
Token ID319
Feature activation+1.06e-02

Pos logit contributions
tub+0.037
gan+0.036
ider+0.035
ses+0.034
sed+0.034
Neg logit contributions
ons-0.016
 envelop-0.016
 saddle-0.015
Offic-0.015
oc-0.014
Token top
Token ID1353
Feature activation+7.48e-02

Pos logit contributions
tub+0.022
gan+0.021
ider+0.019
 Moe+0.018
ahl+0.018
Neg logit contributions
 envelop-0.019
Offic-0.017
 Amen-0.017
ons-0.017
oc-0.015
Token of
Token ID286
Feature activation+8.67e-04

Pos logit contributions
ses+0.210
gan+0.208
sed+0.206
tub+0.201
ider+0.196
Neg logit contributions
ons-0.071
 envelop-0.057
oc-0.052
Offic-0.041
 arrest-0.039
Token it
Token ID340
Feature activation+1.24e-01

Pos logit contributions
tub+0.002
gan+0.001
 Moe+0.001
ider+0.001
ahl+0.001
Neg logit contributions
 envelop-0.002
Offic-0.002
 Amen-0.002
ons-0.002
Mic-0.001
Token,
Token ID11
Feature activation+3.34e-02

Pos logit contributions
ses+0.336
gan+0.334
sed+0.317
tub+0.302
 fault+0.300
Neg logit contributions
ons-0.123
 read-0.102
oc-0.088
 course-0.085
 rode-0.075
Token he
Token ID339
Feature activation+2.27e-02

Pos logit contributions
tub+0.056
gan+0.054
ider+0.047
 Moe+0.046
ahl+0.045
Neg logit contributions
 envelop-0.073
Offic-0.067
 Amen-0.066
ons-0.064
Mic-0.061
Token realized
Token ID6939
Feature activation+5.23e-03

Pos logit contributions
tub+0.045
gan+0.043
 Moe+0.039
ider+0.039
ses+0.038
Neg logit contributions
 envelop-0.040
Offic-0.038
ons-0.037
oc-0.034
 read-0.034
Token that
Token ID326
Feature activation+8.89e-03

Pos logit contributions
tub+0.011
gan+0.010
ider+0.009
ses+0.009
ahl+0.009
Neg logit contributions
 envelop-0.010
Offic-0.009
ons-0.009
 Amen-0.008
oc-0.008
Token it
Token ID340
Feature activation+9.00e-02

Pos logit contributions
tub+0.016
gan+0.015
 Moe+0.013
ider+0.013
ahl+0.013
Neg logit contributions
 envelop-0.018
Offic-0.017
 Amen-0.017
ons-0.016
Mic-0.015
Token hill
Token ID12788
Feature activation+5.58e-02

Pos logit contributions
 envelop+0.075
Offic+0.069
 Amen+0.068
ons+0.065
Mic+0.062
Neg logit contributions
tub-0.067
gan-0.064
ider-0.057
 Moe-0.057
ahl-0.055
Token and
Token ID290
Feature activation-4.62e-02

Pos logit contributions
gan+0.107
tub+0.103
ses+0.098
 Moe+0.097
ahl+0.097
Neg logit contributions
 envelop-0.092
ons-0.092
Offic-0.082
 Amen-0.080
oc-0.077
Token wanted
Token ID2227
Feature activation-8.54e-03

Pos logit contributions
 envelop+0.083
Offic+0.076
 Amen+0.073
ons+0.071
oc+0.066
Neg logit contributions
tub-0.093
gan-0.090
ider-0.084
 Moe-0.080
ahl-0.080
Token to
Token ID284
Feature activation-3.23e-02

Pos logit contributions
 envelop+0.015
ons+0.014
Offic+0.014
 Amen+0.013
 parcels+0.012
Neg logit contributions
tub-0.018
gan-0.016
ider-0.016
ahl-0.015
 Moe-0.015
Token climb
Token ID12080
Feature activation+1.41e-02

Pos logit contributions
 envelop+0.064
Offic+0.060
 Amen+0.059
Mic+0.055
ons+0.055
Neg logit contributions
tub-0.059
gan-0.057
 Moe-0.052
 Mil-0.050
ahl-0.048
Token it
Token ID340
Feature activation+1.24e-01

Pos logit contributions
tub+0.035
gan+0.033
ider+0.032
ses+0.031
sed+0.030
Neg logit contributions
 envelop-0.020
ons-0.017
Offic-0.016
 Amen-0.016
 saddle-0.014
Token,
Token ID11
Feature activation+2.25e-02

Pos logit contributions
tub+0.242
gan+0.240
ses+0.231
sed+0.227
ider+0.217
Neg logit contributions
 envelop-0.209
ons-0.190
Offic-0.176
 Amen-0.176
oc-0.169
Token but
Token ID475
Feature activation+3.32e-03

Pos logit contributions
tub+0.034
gan+0.032
ider+0.028
ahl+0.027
 Moe+0.027
Neg logit contributions
 envelop-0.052
Offic-0.049
 Amen-0.048
ons-0.046
Mic-0.044
Token it
Token ID340
Feature activation+7.32e-02

Pos logit contributions
tub+0.007
ider+0.006
gan+0.006
ahl+0.006
ses+0.006
Neg logit contributions
 envelop-0.005
Offic-0.005
ons-0.005
 Amen-0.005
 read-0.004
Token looked
Token ID3114
Feature activation+3.51e-02

Pos logit contributions
tub+0.133
ider+0.117
gan+0.117
 Moe+0.111
ses+0.109
Neg logit contributions
ons-0.131
 envelop-0.130
Offic-0.127
oc-0.124
 read-0.118
Token too
Token ID1165
Feature activation+2.88e-02

Pos logit contributions
tub+0.071
gan+0.067
ider+0.063
ahl+0.061
ses+0.060
Neg logit contributions
 envelop-0.063
 Amen-0.052
ons-0.051
Offic-0.050
 parcels-0.049
Token that
Token ID326
Feature activation+2.29e-02

Pos logit contributions
 envelop+0.084
Offic+0.079
 Amen+0.077
ons+0.074
Mic+0.069
Neg logit contributions
tub-0.089
gan-0.082
ider-0.076
ahl-0.072
ses-0.072
Token day
Token ID1110
Feature activation+4.05e-02

Pos logit contributions
tub+0.040
gan+0.038
ider+0.033
 Moe+0.033
ahl+0.032
Neg logit contributions
 envelop-0.048
Offic-0.045
 Amen-0.044
ons-0.043
oc-0.041
Token on
Token ID319
Feature activation+1.27e-02

Pos logit contributions
tub+0.086
gan+0.083
 Moe+0.075
ider+0.075
ahl+0.072
Neg logit contributions
 envelop-0.069
Offic-0.063
 Amen-0.062
ons-0.059
Mic-0.055
Token,
Token ID11
Feature activation+2.79e-02

Pos logit contributions
tub+0.023
gan+0.023
ider+0.020
ses+0.020
ahl+0.020
Neg logit contributions
 envelop-0.024
ons-0.023
 Amen-0.021
Offic-0.020
oc-0.020
Token whenever
Token ID8797
Feature activation+5.28e-02

Pos logit contributions
tub+0.051
gan+0.047
ider+0.044
ses+0.042
ahl+0.042
Neg logit contributions
 envelop-0.056
Offic-0.052
 Amen-0.050
ons-0.049
oc-0.047
Token it
Token ID340
Feature activation+1.24e-01

Pos logit contributions
tub+0.106
gan+0.098
ider+0.093
ses+0.091
sed+0.089
Neg logit contributions
 envelop-0.096
Offic-0.089
 Amen-0.087
ons-0.082
 read-0.079
Token r
Token ID374
Feature activation+5.93e-02

Pos logit contributions
gan+0.236
tub+0.223
ses+0.214
 Moe+0.210
ider+0.205
Neg logit contributions
ons-0.215
Offic-0.203
oc-0.203
 envelop-0.203
 Amen-0.199
Tokenained
Token ID1328
Feature activation+6.47e-02

Pos logit contributions
tub+0.090
gan+0.080
 Moe+0.079
 Mil+0.075
ahl+0.075
Neg logit contributions
 envelop-0.125
ons-0.123
Offic-0.119
 Amen-0.111
Mic-0.110
Token,
Token ID11
Feature activation+3.32e-02

Pos logit contributions
gan+0.134
ses+0.119
tub+0.119
ahl+0.119
ider+0.116
Neg logit contributions
ons-0.102
 envelop-0.096
 Amen-0.089
 lending-0.088
 course-0.085
Token the
Token ID262
Feature activation+4.57e-02

Pos logit contributions
tub+0.065
gan+0.061
ider+0.059
ses+0.057
ahl+0.056
Neg logit contributions
 envelop-0.062
Offic-0.056
ons-0.053
 Amen-0.052
Mic-0.049
Token gir
Token ID37370
Feature activation-6.38e-02

Pos logit contributions
tub+0.096
gan+0.093
ider+0.084
 Moe+0.084
ahl+0.082
Neg logit contributions
 envelop-0.078
 Amen-0.072
Offic-0.070
ons-0.067
Mic-0.062
Token could
Token ID714
Feature activation+1.51e-02

Pos logit contributions
gan+0.029
tub+0.029
 Moe+0.026
ider+0.025
ahl+0.025
Neg logit contributions
 envelop-0.036
ons-0.035
 Amen-0.032
Offic-0.031
oc-0.031
Token appear
Token ID1656
Feature activation+4.75e-02

Pos logit contributions
tub+0.029
gan+0.027
 Moe+0.025
ider+0.024
ahl+0.024
Neg logit contributions
 envelop-0.029
Offic-0.027
 Amen-0.026
ons-0.025
Mic-0.024
Token and
Token ID290
Feature activation-4.32e-02

Pos logit contributions
tub+0.091
gan+0.091
ses+0.081
 Moe+0.081
ider+0.081
Neg logit contributions
 envelop-0.088
ons-0.076
Offic-0.076
 Amen-0.074
oc-0.070
Token disappear
Token ID10921
Feature activation+3.93e-02

Pos logit contributions
 envelop+0.082
Offic+0.076
 Amen+0.074
ons+0.071
Mic+0.068
Neg logit contributions
tub-0.084
gan-0.079
 Moe-0.071
ider-0.071
ahl-0.069
Token whenever
Token ID8797
Feature activation+5.09e-02

Pos logit contributions
gan+0.078
tub+0.075
 Moe+0.072
ses+0.072
ider+0.072
Neg logit contributions
 envelop-0.067
ons-0.054
 armor-0.050
 parcels-0.050
 airplanes-0.049
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.121
gan+0.116
ider+0.109
ses+0.106
inter+0.105
Neg logit contributions
 envelop-0.074
 Amen-0.067
Offic-0.066
ons-0.064
 read-0.056
Token wanted
Token ID2227
Feature activation+1.33e-02

Pos logit contributions
gan+0.231
tub+0.227
 Moe+0.209
ider+0.201
 Mil+0.191
Neg logit contributions
 envelop-0.231
ons-0.219
Offic-0.209
 Amen-0.200
oc-0.198
Token.
Token ID13
Feature activation+1.66e-02

Pos logit contributions
tub+0.025
gan+0.024
ider+0.021
 Moe+0.021
ahl+0.020
Neg logit contributions
 envelop-0.027
Offic-0.024
 Amen-0.023
ons-0.023
oc-0.021
Token The
Token ID383
Feature activation+2.00e-02

Pos logit contributions
tub+0.029
gan+0.028
ider+0.026
 Moe+0.024
ahl+0.024
Neg logit contributions
 envelop-0.034
 Amen-0.032
Offic-0.032
ons-0.030
Mic-0.029
Token people
Token ID661
Feature activation-8.24e-03

Pos logit contributions
tub+0.038
gan+0.037
 Moe+0.033
ider+0.033
ahl+0.032
Neg logit contributions
 envelop-0.039
Offic-0.035
 Amen-0.033
ons-0.033
Mic-0.032
Token in
Token ID287
Feature activation+9.63e-03

Pos logit contributions
 envelop+0.018
Offic+0.017
 Amen+0.017
ons+0.016
Mic+0.015
Neg logit contributions
tub-0.013
gan-0.013
 Moe-0.011
ider-0.011
ahl-0.011
Token tree
Token ID5509
Feature activation+1.64e-02

Pos logit contributions
 envelop+0.114
Offic+0.106
 Amen+0.104
ons+0.100
Mic+0.096
Neg logit contributions
tub-0.095
gan-0.089
 Moe-0.079
ider-0.078
ahl-0.076
Token and
Token ID290
Feature activation-5.19e-02

Pos logit contributions
tub+0.030
gan+0.029
 Moe+0.026
ider+0.025
ahl+0.025
Neg logit contributions
 envelop-0.032
Offic-0.030
 Amen-0.029
ons-0.028
oc-0.026
Token decided
Token ID3066
Feature activation+2.29e-03

Pos logit contributions
 envelop+0.104
Offic+0.096
 Amen+0.094
ons+0.091
Mic+0.085
Neg logit contributions
tub-0.096
gan-0.091
ider-0.082
 Moe-0.081
ahl-0.079
Token to
Token ID284
Feature activation-2.36e-02

Pos logit contributions
tub+0.005
gan+0.005
ses+0.005
ider+0.005
ahl+0.004
Neg logit contributions
 envelop-0.004
Offic-0.003
 Amen-0.003
ons-0.003
 read-0.003
Token climb
Token ID12080
Feature activation+1.78e-02

Pos logit contributions
 envelop+0.046
Offic+0.044
 Amen+0.042
Mic+0.039
ons+0.039
Neg logit contributions
tub-0.044
gan-0.042
 Moe-0.038
ahl-0.036
 Mil-0.035
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.044
gan+0.043
ider+0.041
ses+0.038
sed+0.038
Neg logit contributions
 envelop-0.024
ons-0.021
Offic-0.020
 Amen-0.020
oc-0.017
Token.
Token ID13
Feature activation+7.97e-03

Pos logit contributions
gan+0.247
tub+0.243
ses+0.233
sed+0.228
ider+0.219
Neg logit contributions
 envelop-0.208
ons-0.187
Offic-0.177
oc-0.169
 read-0.169
Token Tim
Token ID5045
Feature activation-7.79e-03

Pos logit contributions
tub+0.013
ider+0.013
gan+0.012
ses+0.012
ahl+0.012
Neg logit contributions
 envelop-0.016
Offic-0.015
 Amen-0.014
 parcels-0.013
ons-0.013
Token went
Token ID1816
Feature activation+8.36e-03

Pos logit contributions
 envelop+0.014
Offic+0.013
 Amen+0.013
ons+0.012
Mic+0.011
Neg logit contributions
tub-0.016
gan-0.015
 Moe-0.014
ider-0.014
ahl-0.013
Token first
Token ID717
Feature activation+2.32e-02

Pos logit contributions
tub+0.014
gan+0.014
ses+0.012
ider+0.012
 Moe+0.012
Neg logit contributions
 envelop-0.017
Offic-0.016
 Amen-0.016
ons-0.015
 saddle-0.014
Token and
Token ID290
Feature activation-5.35e-02

Pos logit contributions
tub+0.047
gan+0.044
ider+0.041
ses+0.041
ahl+0.039
Neg logit contributions
 envelop-0.041
ons-0.037
Offic-0.037
 read-0.034
oc-0.032
Token the
Token ID262
Feature activation+4.25e-02

Pos logit contributions
 envelop+0.020
 Amen+0.019
Offic+0.019
ons+0.017
Mic+0.017
Neg logit contributions
tub-0.018
gan-0.017
 Moe-0.016
 Mil-0.015
ahl-0.014
Token afternoon
Token ID6672
Feature activation+4.26e-02

Pos logit contributions
tub+0.081
gan+0.077
 Moe+0.068
ider+0.068
ahl+0.066
Neg logit contributions
 envelop-0.082
Offic-0.076
 Amen-0.074
ons-0.073
Mic-0.068
Token bouncing
Token ID30713
Feature activation+1.71e-02

Pos logit contributions
tub+0.089
gan+0.085
ider+0.076
 Moe+0.076
ahl+0.075
Neg logit contributions
 envelop-0.075
Offic-0.068
 Amen-0.066
ons-0.065
Mic-0.060
Token until
Token ID1566
Feature activation+4.74e-02

Pos logit contributions
gan+0.035
tub+0.034
ider+0.033
ses+0.030
ahl+0.029
Neg logit contributions
 envelop-0.029
ons-0.027
Offic-0.024
oc-0.023
 Amen-0.023
Token soon
Token ID2582
Feature activation+5.38e-02

Pos logit contributions
tub+0.104
gan+0.098
ider+0.097
ses+0.091
ahl+0.091
Neg logit contributions
 envelop-0.077
ons-0.068
Offic-0.067
 Amen-0.066
 lending-0.058
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
ider+0.120
tub+0.120
ses+0.114
gan+0.113
sed+0.106
Neg logit contributions
 envelop-0.085
ons-0.076
Offic-0.068
 arrest-0.064
itivity-0.062
Token was
Token ID373
Feature activation+4.55e-02

Pos logit contributions
ses+0.268
ider+0.258
gan+0.252
 spine+0.250
sing+0.246
Neg logit contributions
ons-0.161
 envelop-0.137
 rode-0.136
 read-0.130
oc-0.129
Token time
Token ID640
Feature activation+5.27e-02

Pos logit contributions
tub+0.097
gan+0.093
ider+0.089
ahl+0.084
ses+0.084
Neg logit contributions
 envelop-0.076
 Amen-0.065
Offic-0.064
ons-0.063
Mic-0.057
Token to
Token ID284
Feature activation-6.13e-03

Pos logit contributions
tub+0.111
gan+0.105
 Moe+0.097
ahl+0.096
ider+0.095
Neg logit contributions
 envelop-0.090
ons-0.076
 Amen-0.076
Offic-0.073
 permitting-0.067
Token go
Token ID467
Feature activation+2.63e-02

Pos logit contributions
 envelop+0.010
Offic+0.009
 Amen+0.009
ons+0.008
Mic+0.008
Neg logit contributions
tub-0.013
gan-0.013
ider-0.012
 Moe-0.012
ahl-0.011
Token home
Token ID1363
Feature activation+7.38e-02

Pos logit contributions
gan+0.054
tub+0.054
ider+0.051
ses+0.049
sed+0.048
Neg logit contributions
 envelop-0.044
ons-0.038
 Amen-0.038
Offic-0.037
 saddle-0.034
Token tree
Token ID5509
Feature activation+1.20e-02

Pos logit contributions
 envelop+0.110
Offic+0.102
 Amen+0.100
ons+0.096
Mic+0.092
Neg logit contributions
tub-0.090
gan-0.085
 Moe-0.075
ider-0.075
ahl-0.072
Token with
Token ID351
Feature activation-2.29e-02

Pos logit contributions
tub+0.024
gan+0.023
 Moe+0.021
ider+0.020
ahl+0.020
Neg logit contributions
 envelop-0.022
Offic-0.020
 Amen-0.020
ons-0.019
oc-0.018
Token many
Token ID867
Feature activation-4.04e-02

Pos logit contributions
 envelop+0.045
Offic+0.043
 Amen+0.043
ons+0.040
Mic+0.038
Neg logit contributions
tub-0.042
gan-0.040
 Moe-0.036
ider-0.034
ahl-0.034
Token mushrooms
Token ID23452
Feature activation-8.50e-03

Pos logit contributions
 envelop+0.086
 Amen+0.085
Offic+0.082
Mic+0.075
ons+0.073
Neg logit contributions
tub-0.064
gan-0.063
 Moe-0.055
ider-0.054
ahl-0.053
Token under
Token ID739
Feature activation+1.84e-02

Pos logit contributions
 envelop+0.017
Offic+0.016
ons+0.016
 Amen+0.015
Mic+0.014
Neg logit contributions
tub-0.015
gan-0.015
 Moe-0.013
ahl-0.013
ider-0.012
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.036
gan+0.034
 Moe+0.031
 Mil+0.030
ider+0.030
Neg logit contributions
 envelop-0.034
Offic-0.032
 Amen-0.032
ons-0.030
Mic-0.029
Token.
Token ID13
Feature activation+1.11e-02

Pos logit contributions
gan+0.236
tub+0.236
 Moe+0.211
ider+0.206
ahl+0.204
Neg logit contributions
 envelop-0.229
ons-0.211
Offic-0.197
 Amen-0.183
oc-0.179
Token Tim
Token ID5045
Feature activation-5.27e-03

Pos logit contributions
tub+0.021
ider+0.020
gan+0.019
ahl+0.018
ses+0.018
Neg logit contributions
 envelop-0.019
Offic-0.019
 Amen-0.018
ons-0.017
oc-0.017
Token wanted
Token ID2227
Feature activation-1.46e-02

Pos logit contributions
 envelop+0.010
Offic+0.009
 Amen+0.009
ons+0.009
Mic+0.008
Neg logit contributions
tub-0.010
gan-0.010
ider-0.009
 Moe-0.009
ahl-0.008
Token to
Token ID284
Feature activation-3.47e-02

Pos logit contributions
 envelop+0.031
Offic+0.029
 Amen+0.029
ons+0.027
Mic+0.026
Neg logit contributions
tub-0.025
gan-0.023
 Moe-0.021
ider-0.020
ahl-0.020
Token look
Token ID804
Feature activation+2.41e-02

Pos logit contributions
 envelop+0.073
 Amen+0.070
Offic+0.069
Mic+0.064
ons+0.063
Neg logit contributions
gan-0.058
tub-0.055
 Moe-0.052
 Mil-0.050
ahl-0.048
Token tree
Token ID5509
Feature activation+2.02e-02

Pos logit contributions
 envelop+0.102
Offic+0.097
 Amen+0.095
ons+0.091
Mic+0.089
Neg logit contributions
tub-0.080
gan-0.072
 Moe-0.066
ses-0.064
ider-0.063
Token and
Token ID290
Feature activation-4.73e-02

Pos logit contributions
tub+0.037
gan+0.035
 Moe+0.032
ider+0.031
ahl+0.031
Neg logit contributions
 envelop-0.040
Offic-0.037
 Amen-0.037
ons-0.035
Mic-0.033
Token decided
Token ID3066
Feature activation+6.37e-03

Pos logit contributions
 envelop+0.095
Offic+0.089
 Amen+0.087
ons+0.083
Mic+0.078
Neg logit contributions
tub-0.087
gan-0.082
ider-0.073
 Moe-0.073
ahl-0.071
Token to
Token ID284
Feature activation-1.89e-02

Pos logit contributions
tub+0.014
gan+0.012
ider+0.012
ses+0.012
 Moe+0.011
Neg logit contributions
 envelop-0.011
Offic-0.010
 Amen-0.010
ons-0.009
 read-0.009
Token climb
Token ID12080
Feature activation+1.89e-02

Pos logit contributions
 envelop+0.038
Offic+0.035
 Amen+0.034
ons+0.032
Mic+0.032
Neg logit contributions
tub-0.034
gan-0.033
 Moe-0.030
 Mil-0.028
ahl-0.028
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.043
gan+0.041
ider+0.039
ses+0.037
sed+0.036
Neg logit contributions
 envelop-0.030
ons-0.026
Offic-0.026
 Amen-0.026
oc-0.022
Token.
Token ID13
Feature activation+1.16e-02

Pos logit contributions
tub+0.245
gan+0.245
ses+0.234
sed+0.228
 Moe+0.219
Neg logit contributions
 envelop-0.201
ons-0.183
Offic-0.174
 Amen-0.169
oc-0.168
Token
Token ID198
Feature activation-1.28e-02

Pos logit contributions
tub+0.019
gan+0.017
ider+0.017
ses+0.016
ahl+0.015
Neg logit contributions
 envelop-0.025
Offic-0.024
 Amen-0.023
ons-0.022
Mic-0.022
Token
Token ID198
Feature activation-1.08e-02

Pos logit contributions
 envelop+0.033
 Amen+0.030
Offic+0.030
ons+0.028
itivity+0.028
Neg logit contributions
tub-0.017
gan-0.016
 Moe-0.014
 Mil-0.013
ahl-0.013
TokenAs
Token ID1722
Feature activation+2.32e-02

Pos logit contributions
 envelop+0.021
 Amen+0.019
ons+0.018
Offic+0.017
 parcels+0.016
Neg logit contributions
tub-0.021
gan-0.020
 Moe-0.018
ider-0.018
 Mil-0.017
Token Tom
Token ID4186
Feature activation+3.07e-02

Pos logit contributions
tub+0.041
gan+0.037
ider+0.035
ses+0.034
ahl+0.033
Neg logit contributions
 envelop-0.048
Offic-0.045
ons-0.043
 Amen-0.043
Mic-0.041
Token and
Token ID290
Feature activation-5.79e-02

Pos logit contributions
 envelop+0.021
Offic+0.019
 Amen+0.018
ons+0.018
oc+0.017
Neg logit contributions
tub-0.025
gan-0.024
ider-0.021
 Moe-0.021
ahl-0.021
Token seek
Token ID5380
Feature activation-2.39e-02

Pos logit contributions
 envelop+0.108
Offic+0.099
 Amen+0.096
ons+0.093
Mic+0.088
Neg logit contributions
tub-0.114
gan-0.109
ider-0.098
 Moe-0.096
ahl-0.094
Token all
Token ID477
Feature activation-9.18e-03

Pos logit contributions
 envelop+0.046
ons+0.040
Offic+0.040
 Amen+0.039
oc+0.037
Neg logit contributions
tub-0.044
gan-0.043
ses-0.038
ahl-0.038
ider-0.037
Token afternoon
Token ID6672
Feature activation+5.16e-02

Pos logit contributions
 envelop+0.016
ons+0.014
 Amen+0.014
Offic+0.014
oc+0.013
Neg logit contributions
tub-0.019
gan-0.018
ider-0.018
ses-0.017
sed-0.016
Token until
Token ID1566
Feature activation+5.66e-02

Pos logit contributions
tub+0.087
gan+0.084
ahl+0.073
 Moe+0.072
ider+0.071
Neg logit contributions
 envelop-0.111
Offic-0.100
 Amen-0.098
ons-0.098
Mic-0.091
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.115
gan+0.109
ider+0.102
ses+0.096
ahl+0.096
Neg logit contributions
 envelop-0.103
Offic-0.093
 Amen-0.090
ons-0.090
Mic-0.079
Token was
Token ID373
Feature activation+4.46e-02

Pos logit contributions
ses+0.241
ider+0.229
gan+0.227
tub+0.223
 fault+0.218
Neg logit contributions
 envelop-0.181
Offic-0.178
ons-0.175
 read-0.169
oc-0.167
Token time
Token ID640
Feature activation+5.00e-02

Pos logit contributions
tub+0.086
gan+0.082
 Moe+0.073
ider+0.073
ahl+0.071
Neg logit contributions
 envelop-0.086
Offic-0.079
 Amen-0.077
ons-0.074
Mic-0.070
Token for
Token ID329
Feature activation-6.75e-03

Pos logit contributions
tub+0.106
gan+0.100
 Moe+0.094
ahl+0.092
ider+0.090
Neg logit contributions
 envelop-0.083
 Amen-0.071
Offic-0.068
ons-0.068
 lending-0.063
Token Sam
Token ID3409
Feature activation+1.14e-02

Pos logit contributions
 envelop+0.015
Offic+0.014
 Amen+0.014
oc+0.013
ons+0.013
Neg logit contributions
tub-0.011
gan-0.010
 Moe-0.010
 Mil-0.009
ahl-0.009
Token to
Token ID284
Feature activation-2.17e-02

Pos logit contributions
tub+0.021
gan+0.020
ider+0.018
 Moe+0.018
ahl+0.017
Neg logit contributions
 envelop-0.023
Offic-0.021
 Amen-0.020
ons-0.020
Mic-0.019
TokenFinally
Token ID11158
Feature activation+5.03e-02

Pos logit contributions
 envelop+0.016
Offic+0.014
 Amen+0.014
ons+0.013
oc+0.012
Neg logit contributions
tub-0.016
gan-0.016
ider-0.014
 Moe-0.014
ahl-0.014
Token the
Token ID262
Feature activation+4.64e-02

Pos logit contributions
tub+0.097
gan+0.086
ses+0.082
inter+0.079
ider+0.079
Neg logit contributions
 envelop-0.102
ons-0.091
Offic-0.088
 Amen-0.085
 read-0.085
Token day
Token ID1110
Feature activation+5.77e-02

Pos logit contributions
tub+0.093
gan+0.090
 Moe+0.079
ider+0.079
ahl+0.078
Neg logit contributions
 envelop-0.086
Offic-0.078
 Amen-0.077
ons-0.074
Mic-0.069
Token came
Token ID1625
Feature activation+5.18e-02

Pos logit contributions
tub+0.118
gan+0.111
ses+0.108
ider+0.105
sed+0.103
Neg logit contributions
 envelop-0.103
ons-0.093
Offic-0.091
 read-0.082
oc-0.081
Token when
Token ID618
Feature activation+5.22e-02

Pos logit contributions
tub+0.114
ses+0.105
ider+0.101
gan+0.100
auc+0.100
Neg logit contributions
 envelop-0.080
ons-0.063
Offic-0.063
 read-0.062
 lending-0.062
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.096
gan+0.087
ider+0.083
inter+0.081
ses+0.080
Neg logit contributions
 envelop-0.103
Offic-0.092
 Amen-0.092
ons-0.091
 read-0.086
Token was
Token ID373
Feature activation+4.12e-02

Pos logit contributions
ses+0.254
gan+0.247
tub+0.238
 fault+0.233
ider+0.227
Neg logit contributions
 envelop-0.184
Offic-0.180
ons-0.179
 read-0.174
oc-0.165
Token time
Token ID640
Feature activation+5.51e-02

Pos logit contributions
tub+0.072
gan+0.069
 Moe+0.060
ider+0.059
ahl+0.057
Neg logit contributions
 envelop-0.086
Offic-0.082
 Amen-0.079
ons-0.076
oc-0.076
Token for
Token ID329
Feature activation-5.85e-03

Pos logit contributions
tub+0.121
gan+0.115
 Moe+0.107
ider+0.104
ahl+0.103
Neg logit contributions
 envelop-0.090
ons-0.074
Offic-0.071
 Amen-0.071
 read-0.065
Token Anna
Token ID11735
Feature activation+2.53e-02

Pos logit contributions
 envelop+0.012
Offic+0.011
 Amen+0.011
ons+0.011
Mic+0.010
Neg logit contributions
tub-0.010
gan-0.010
 Moe-0.009
ider-0.009
ahl-0.008
Token to
Token ID284
Feature activation-2.24e-02

Pos logit contributions
tub+0.045
gan+0.043
 Moe+0.038
ider+0.038
ahl+0.037
Neg logit contributions
 envelop-0.052
Offic-0.047
 Amen-0.047
ons-0.046
Mic-0.043
Token that
Token ID326
Feature activation+2.41e-02

Pos logit contributions
 envelop+0.067
Offic+0.064
 Amen+0.062
ons+0.060
 read+0.056
Neg logit contributions
tub-0.079
gan-0.074
ider-0.068
 Moe-0.065
ses-0.064
Token day
Token ID1110
Feature activation+4.52e-02

Pos logit contributions
tub+0.029
gan+0.023
 Moe+0.021
ahl+0.020
 rains+0.019
Neg logit contributions
 envelop-0.065
Offic-0.061
 Amen-0.058
ons-0.057
 parcels-0.057
Token on
Token ID319
Feature activation+1.40e-02

Pos logit contributions
tub+0.076
gan+0.074
ider+0.067
 Moe+0.063
ahl+0.063
Neg logit contributions
 envelop-0.096
Offic-0.089
 Amen-0.086
 parcels-0.080
Mic-0.079
Token,
Token ID11
Feature activation+3.38e-02

Pos logit contributions
tub+0.028
gan+0.026
ider+0.024
ses+0.024
ipples+0.023
Neg logit contributions
 envelop-0.026
ons-0.025
 Amen-0.022
Offic-0.022
 read-0.021
Token whenever
Token ID8797
Feature activation+5.41e-02

Pos logit contributions
tub+0.041
gan+0.038
ider+0.032
 Moe+0.032
ahl+0.030
Neg logit contributions
 envelop-0.089
Offic-0.084
 Amen-0.083
ons-0.081
Mic-0.077
Token it
Token ID340
Feature activation+1.23e-01

Pos logit contributions
tub+0.097
gan+0.093
ider+0.083
ses+0.080
ahl+0.077
Neg logit contributions
 envelop-0.112
Offic-0.103
 Amen-0.102
ons-0.099
 read-0.095
Token r
Token ID374
Feature activation+5.83e-02

Pos logit contributions
gan+0.212
tub+0.192
ses+0.191
 Moe+0.189
ider+0.187
Neg logit contributions
ons-0.223
 read-0.217
Offic-0.211
 envelop-0.210
 Amen-0.207
Tokenained
Token ID1328
Feature activation+6.29e-02

Pos logit contributions
tub+0.086
 Moe+0.081
gan+0.080
 Mil+0.076
ahl+0.075
Neg logit contributions
 envelop-0.122
ons-0.120
Offic-0.116
 Amen-0.110
Mic-0.105
Token,
Token ID11
Feature activation+4.03e-02

Pos logit contributions
gan+0.143
tub+0.127
ses+0.126
ider+0.125
ahl+0.123
Neg logit contributions
ons-0.086
 envelop-0.075
 course-0.073
 lending-0.071
 read-0.071
Token Lily
Token ID20037
Feature activation+3.01e-02

Pos logit contributions
tub+0.074
gan+0.073
ider+0.072
ses+0.066
sed+0.064
Neg logit contributions
 envelop-0.078
ons-0.070
Offic-0.067
 read-0.064
 Amen-0.063
Token would
Token ID561
Feature activation-8.84e-03

Pos logit contributions
tub+0.059
gan+0.058
ider+0.054
 Moe+0.054
ahl+0.051
Neg logit contributions
 envelop-0.053
Offic-0.049
ons-0.047
 Amen-0.047
 read-0.044
Token
Token ID198
Feature activation-9.16e-03

Pos logit contributions
 envelop+0.030
Offic+0.028
 Amen+0.027
ons+0.026
Mic+0.026
Neg logit contributions
tub-0.016
gan-0.014
 Moe-0.013
ahl-0.012
 Mil-0.012
TokenOne
Token ID3198
Feature activation+1.04e-02

Pos logit contributions
 envelop+0.017
 Amen+0.015
Offic+0.014
ons+0.014
 parcels+0.013
Neg logit contributions
tub-0.019
gan-0.016
 Moe-0.015
ider-0.015
ahl-0.014
Token day
Token ID1110
Feature activation+6.67e-02

Pos logit contributions
tub+0.021
gan+0.021
ider+0.020
ses+0.019
sed+0.019
Neg logit contributions
 Amen-0.017
 envelop-0.017
Offic-0.016
oc-0.015
ons-0.015
Token,
Token ID11
Feature activation+4.91e-02

Pos logit contributions
tub+0.099
gan+0.095
ider+0.081
 Moe+0.080
ses+0.079
Neg logit contributions
 envelop-0.158
Offic-0.147
 Amen-0.142
ons-0.142
oc-0.133
Token when
Token ID618
Feature activation+6.08e-02

Pos logit contributions
tub+0.081
gan+0.075
 Moe+0.067
ider+0.065
 Mil+0.062
Neg logit contributions
 envelop-0.108
Offic-0.101
 Amen-0.101
ons-0.095
Mic-0.092
Token it
Token ID340
Feature activation+1.22e-01

Pos logit contributions
tub+0.105
gan+0.099
 Moe+0.087
ider+0.086
ahl+0.083
Neg logit contributions
 envelop-0.130
Offic-0.121
 Amen-0.118
ons-0.114
Mic-0.109
Token was
Token ID373
Feature activation+4.54e-02

Pos logit contributions
gan+0.233
ses+0.225
 fault+0.213
tub+0.206
ider+0.205
Neg logit contributions
 Amen-0.205
oc-0.200
 envelop-0.194
ons-0.194
Offic-0.191
Token very
Token ID845
Feature activation+4.17e-03

Pos logit contributions
tub+0.083
gan+0.079
 Moe+0.070
ider+0.070
ahl+0.067
Neg logit contributions
 envelop-0.092
Offic-0.085
 Amen-0.084
ons-0.080
Mic-0.076
Token warm
Token ID5814
Feature activation+1.35e-03

Pos logit contributions
tub+0.008
gan+0.008
ses+0.007
 Moe+0.007
ahl+0.007
Neg logit contributions
 envelop-0.007
 Amen-0.007
Offic-0.007
ons-0.007
Mic-0.006
Token and
Token ID290
Feature activation-3.78e-02

Pos logit contributions
gan+0.003
ahl+0.003
tub+0.003
ses+0.003
ider+0.003
Neg logit contributions
 envelop-0.002
ons-0.002
 Amen-0.002
Offic-0.002
pa-0.002
Token sunny
Token ID27737
Feature activation+3.60e-02

Pos logit contributions
 envelop+0.083
Offic+0.077
 Amen+0.076
ons+0.073
Mic+0.070
Neg logit contributions
tub-0.064
gan-0.059
 Moe-0.052
ider-0.051
ahl-0.049
Token that
Token ID326
Feature activation+2.53e-02

Pos logit contributions
 envelop+0.067
Offic+0.063
ons+0.061
 Amen+0.058
 read+0.055
Neg logit contributions
tub-0.092
gan-0.081
ider-0.078
inter-0.075
ses-0.074
Token day
Token ID1110
Feature activation+4.45e-02

Pos logit contributions
tub+0.035
gan+0.030
 Moe+0.027
ider+0.025
ahl+0.025
Neg logit contributions
 envelop-0.064
Offic-0.060
 Amen-0.057
ons-0.056
 parcels-0.054
Token on
Token ID319
Feature activation+1.43e-02

Pos logit contributions
tub+0.073
gan+0.071
ider+0.065
 Moe+0.062
ahl+0.061
Neg logit contributions
 envelop-0.095
Offic-0.089
 Amen-0.085
 parcels-0.080
Mic-0.079
Token,
Token ID11
Feature activation+3.22e-02

Pos logit contributions
tub+0.028
gan+0.027
ider+0.024
ses+0.024
ipples+0.023
Neg logit contributions
 envelop-0.027
ons-0.026
 Amen-0.023
Offic-0.023
 lending-0.022
Token whenever
Token ID8797
Feature activation+5.47e-02

Pos logit contributions
tub+0.046
gan+0.043
ider+0.038
ses+0.037
 Moe+0.035
Neg logit contributions
 envelop-0.079
Offic-0.072
ons-0.071
 Amen-0.070
Mic-0.067
Token it
Token ID340
Feature activation+1.22e-01

Pos logit contributions
tub+0.099
gan+0.093
ider+0.087
ses+0.085
inter+0.083
Neg logit contributions
 envelop-0.114
ons-0.101
Offic-0.100
 Amen-0.099
 read-0.098
Token r
Token ID374
Feature activation+5.96e-02

Pos logit contributions
gan+0.234
tub+0.221
ses+0.221
 Moe+0.215
ider+0.208
Neg logit contributions
ons-0.206
 read-0.195
 rode-0.184
 envelop-0.183
Offic-0.182
Tokenained
Token ID1328
Feature activation+6.36e-02

Pos logit contributions
tub+0.093
 Moe+0.087
gan+0.084
ahl+0.082
 Mil+0.081
Neg logit contributions
ons-0.118
 envelop-0.118
Offic-0.110
 Amen-0.102
Mic-0.100
Token,
Token ID11
Feature activation+3.74e-02

Pos logit contributions
gan+0.138
ses+0.125
tub+0.125
ahl+0.123
ider+0.121
Neg logit contributions
ons-0.093
 envelop-0.084
 course-0.078
 lending-0.074
 read-0.073
Token Lily
Token ID20037
Feature activation+2.67e-02

Pos logit contributions
tub+0.066
gan+0.065
ider+0.063
ses+0.060
sed+0.058
Neg logit contributions
 envelop-0.076
ons-0.068
Offic-0.065
 read-0.061
 Amen-0.060
Token would
Token ID561
Feature activation-1.03e-02

Pos logit contributions
tub+0.052
gan+0.051
ider+0.047
 Moe+0.047
ahl+0.045
Neg logit contributions
 envelop-0.049
Offic-0.045
ons-0.043
 Amen-0.043
oc-0.040
Tokenrolling
Token ID18886
Feature activation+6.89e-03

Pos logit contributions
 envelop+0.005
Offic+0.005
ons+0.004
 Amen+0.004
Mic+0.004
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.004
ses-0.004
 Mil-0.004
Token through
Token ID832
Feature activation+2.72e-02

Pos logit contributions
gan+0.013
tub+0.013
ses+0.013
sed+0.012
ider+0.012
Neg logit contributions
 envelop-0.012
Offic-0.011
ons-0.010
oc-0.010
 Amen-0.010
Token the
Token ID262
Feature activation+4.93e-02

Pos logit contributions
tub+0.057
gan+0.056
ider+0.051
ahl+0.049
 Moe+0.049
Neg logit contributions
 envelop-0.049
Offic-0.044
ons-0.041
 Amen-0.041
oc-0.038
Token corridors
Token ID32807
Feature activation+6.84e-02

Pos logit contributions
tub+0.090
gan+0.088
ider+0.078
 Moe+0.077
ses+0.075
Neg logit contributions
 envelop-0.100
Offic-0.092
 Amen-0.089
ons-0.087
Mic-0.081
Token when
Token ID618
Feature activation+5.05e-02

Pos logit contributions
tub+0.128
gan+0.127
ahl+0.115
 Moe+0.115
ses+0.114
Neg logit contributions
 envelop-0.127
ons-0.119
Offic-0.115
oc-0.104
 Amen-0.103
Token it
Token ID340
Feature activation+1.22e-01

Pos logit contributions
tub+0.110
gan+0.107
ahl+0.101
ider+0.099
inter+0.097
Neg logit contributions
 envelop-0.077
ons-0.071
 Amen-0.071
Offic-0.069
 read-0.065
Token ran
Token ID4966
Feature activation+1.76e-02

Pos logit contributions
gan+0.290
 fault+0.287
ses+0.286
tub+0.276
 Moe+0.273
Neg logit contributions
 envelop-0.144
 read-0.141
Offic-0.140
ons-0.133
 rode-0.113
Token into
Token ID656
Feature activation+1.48e-02

Pos logit contributions
gan+0.035
tub+0.034
ses+0.031
ider+0.031
 Moe+0.030
Neg logit contributions
 envelop-0.033
Offic-0.028
 Amen-0.026
ons-0.025
Mic-0.025
Token a
Token ID257
Feature activation-2.08e-02

Pos logit contributions
tub+0.024
gan+0.022
 Moe+0.020
 Mil+0.019
ahl+0.018
Neg logit contributions
 envelop-0.032
 Amen-0.031
Offic-0.031
ons-0.029
Mic-0.029
Token tiny
Token ID7009
Feature activation-2.16e-02

Pos logit contributions
 envelop+0.045
Offic+0.042
 Amen+0.041
ons+0.040
Mic+0.038
Neg logit contributions
tub-0.035
gan-0.032
 Moe-0.029
ider-0.029
ahl-0.028
Token mouse
Token ID10211
Feature activation-1.80e-02

Pos logit contributions
 envelop+0.046
Offic+0.043
 Amen+0.042
ons+0.041
Mic+0.039
Neg logit contributions
tub-0.037
gan-0.035
ider-0.031
 Moe-0.031
ahl-0.029
Token.
Token ID13
Feature activation+1.80e-02

Pos logit contributions
tub+0.091
gan+0.083
 Moe+0.073
ider+0.072
 Mil+0.067
Neg logit contributions
 envelop-0.153
Offic-0.148
 Amen-0.138
ons-0.135
Mic-0.133
Token One
Token ID1881
Feature activation+1.81e-02

Pos logit contributions
tub+0.036
gan+0.036
ider+0.034
 Moe+0.031
ses+0.030
Neg logit contributions
 envelop-0.031
 Amen-0.030
Offic-0.029
ons-0.028
Mic-0.026
Token day
Token ID1110
Feature activation+7.23e-02

Pos logit contributions
tub+0.036
gan+0.035
ider+0.034
 Moe+0.032
ses+0.031
Neg logit contributions
 envelop-0.031
Offic-0.029
 Amen-0.029
ons-0.028
oc-0.027
Token,
Token ID11
Feature activation+5.18e-02

Pos logit contributions
tub+0.139
gan+0.131
ider+0.116
 Moe+0.112
 fault+0.111
Neg logit contributions
 envelop-0.134
ons-0.133
 Amen-0.120
Offic-0.119
oc-0.114
Token when
Token ID618
Feature activation+6.30e-02

Pos logit contributions
tub+0.086
gan+0.082
ider+0.073
 Moe+0.070
ses+0.068
Neg logit contributions
 envelop-0.109
Offic-0.102
 Amen-0.102
ons-0.101
Mic-0.095
Token it
Token ID340
Feature activation+1.22e-01

Pos logit contributions
tub+0.110
gan+0.107
ider+0.095
 Moe+0.090
ses+0.088
Neg logit contributions
 envelop-0.127
 Amen-0.122
Offic-0.120
ons-0.118
Mic-0.108
Token was
Token ID373
Feature activation+4.22e-02

Pos logit contributions
ses+0.243
gan+0.232
 fault+0.230
sed+0.223
tub+0.220
Neg logit contributions
ons-0.185
 read-0.180
 Amen-0.176
oc-0.170
Offic-0.168
Token very
Token ID845
Feature activation+5.30e-03

Pos logit contributions
tub+0.076
gan+0.073
 Moe+0.065
ider+0.065
ahl+0.063
Neg logit contributions
 envelop-0.085
Offic-0.078
 Amen-0.077
ons-0.075
Mic-0.071
Token cold
Token ID4692
Feature activation+4.51e-02

Pos logit contributions
tub+0.010
gan+0.010
ses+0.009
 Moe+0.009
ahl+0.009
Neg logit contributions
 envelop-0.008
 Amen-0.008
ons-0.008
Offic-0.007
 read-0.007
Token outside
Token ID2354
Feature activation+8.45e-02

Pos logit contributions
gan+0.094
tub+0.089
ider+0.086
ahl+0.084
ses+0.084
Neg logit contributions
 envelop-0.071
ons-0.070
 Amen-0.067
Offic-0.062
Mic-0.059
Token,
Token ID11
Feature activation+4.87e-02

Pos logit contributions
gan+0.170
ider+0.162
tub+0.160
ses+0.158
ahl+0.153
Neg logit contributions
 envelop-0.147
ons-0.144
 Amen-0.129
 read-0.120
pa-0.118
Token that
Token ID326
Feature activation+2.42e-02

Pos logit contributions
 envelop+0.083
Offic+0.078
 Amen+0.077
ons+0.074
Mic+0.069
Neg logit contributions
tub-0.090
gan-0.081
ider-0.076
 Moe-0.072
ses-0.071
Token day
Token ID1110
Feature activation+4.31e-02

Pos logit contributions
tub+0.029
gan+0.023
 Moe+0.019
ahl+0.019
ider+0.018
Neg logit contributions
 envelop-0.065
Offic-0.061
ons-0.057
itivity-0.057
 Amen-0.057
Token on
Token ID319
Feature activation+1.40e-02

Pos logit contributions
tub+0.071
gan+0.069
ider+0.063
 Moe+0.059
ahl+0.059
Neg logit contributions
 envelop-0.093
Offic-0.086
 Amen-0.082
 parcels-0.077
Mic-0.077
Token,
Token ID11
Feature activation+2.91e-02

Pos logit contributions
tub+0.026
gan+0.025
ider+0.023
ses+0.022
ahl+0.021
Neg logit contributions
 envelop-0.028
ons-0.027
 Amen-0.024
Offic-0.024
Mic-0.023
Token whenever
Token ID8797
Feature activation+5.40e-02

Pos logit contributions
tub+0.037
gan+0.034
ider+0.029
 Moe+0.028
ahl+0.027
Neg logit contributions
 envelop-0.076
Offic-0.071
 Amen-0.070
ons-0.068
Mic-0.065
Token it
Token ID340
Feature activation+1.22e-01

Pos logit contributions
tub+0.096
gan+0.091
ider+0.083
ses+0.080
inter+0.078
Neg logit contributions
 envelop-0.114
Offic-0.102
ons-0.101
 Amen-0.101
 read-0.097
Token r
Token ID374
Feature activation+5.94e-02

Pos logit contributions
gan+0.222
tub+0.216
 Moe+0.208
ses+0.192
 Mil+0.192
Neg logit contributions
 envelop-0.222
ons-0.218
 Amen-0.215
Offic-0.215
oc-0.199
Tokenained
Token ID1328
Feature activation+6.32e-02

Pos logit contributions
tub+0.083
gan+0.077
 Moe+0.075
 Mil+0.071
ahl+0.070
Neg logit contributions
 envelop-0.136
ons-0.129
Offic-0.128
 Amen-0.124
Mic-0.118
Token,
Token ID11
Feature activation+3.57e-02

Pos logit contributions
gan+0.136
tub+0.123
ses+0.120
ahl+0.120
ider+0.116
Neg logit contributions
ons-0.095
 envelop-0.090
 Amen-0.085
 lending-0.082
 course-0.080
Token Lily
Token ID20037
Feature activation+2.54e-02

Pos logit contributions
tub+0.061
gan+0.060
ider+0.057
ses+0.054
sed+0.053
Neg logit contributions
 envelop-0.075
ons-0.067
Offic-0.065
 read-0.060
 Amen-0.060
Token would
Token ID561
Feature activation-9.25e-03

Pos logit contributions
tub+0.049
gan+0.047
 Moe+0.043
ider+0.042
ahl+0.041
Neg logit contributions
 envelop-0.048
Offic-0.045
 Amen-0.043
ons-0.042
Mic-0.039
Token.
Token ID13
Feature activation+1.33e-02

Pos logit contributions
 envelop+0.104
Offic+0.097
 Amen+0.095
ons+0.092
Mic+0.087
Neg logit contributions
tub-0.075
gan-0.071
 Moe-0.062
ider-0.062
ahl-0.060
Token He
Token ID679
Feature activation-3.85e-03

Pos logit contributions
tub+0.021
gan+0.020
ider+0.018
ahl+0.018
ses+0.018
Neg logit contributions
 envelop-0.030
Offic-0.028
 Amen-0.028
Mic-0.025
ons-0.025
Token also
Token ID635
Feature activation-1.37e-02

Pos logit contributions
 envelop+0.008
Offic+0.007
 Amen+0.007
ons+0.007
oc+0.007
Neg logit contributions
tub-0.007
gan-0.007
ider-0.006
 Moe-0.006
ahl-0.006
Token has
Token ID468
Feature activation-1.60e-02

Pos logit contributions
 envelop+0.027
Offic+0.025
 Amen+0.024
ons+0.023
Mic+0.022
Neg logit contributions
tub-0.026
gan-0.024
 Moe-0.022
ider-0.022
ahl-0.021
Token a
Token ID257
Feature activation-4.86e-02

Pos logit contributions
 envelop+0.034
Offic+0.033
 Amen+0.033
ons+0.031
oc+0.030
Neg logit contributions
tub-0.026
gan-0.025
 Moe-0.022
ider-0.021
 Mil-0.020
Token sle
Token ID3133
Feature activation-1.37e-01

Pos logit contributions
 envelop+0.097
Offic+0.094
ons+0.090
 Amen+0.090
Mic+0.083
Neg logit contributions
tub-0.089
gan-0.078
 Moe-0.071
ider-0.070
ses-0.069
Tokenigh
Token ID394
Feature activation-5.09e-02

Pos logit contributions
Offic+0.160
 envelop+0.157
Mic+0.151
 pic+0.129
 inviting+0.127
Neg logit contributions
tub-0.314
ider-0.314
gan-0.307
ses-0.299
sed-0.294
Token and
Token ID290
Feature activation-7.63e-02

Pos logit contributions
 envelop+0.103
Offic+0.097
 Amen+0.095
ons+0.089
Mic+0.087
Neg logit contributions
tub-0.092
gan-0.086
ider-0.077
 Moe-0.076
ahl-0.072
Token re
Token ID302
Feature activation-1.01e-01

Pos logit contributions
 envelop+0.163
Offic+0.155
 Amen+0.151
ons+0.148
Mic+0.144
Neg logit contributions
tub-0.119
gan-0.108
 Moe-0.098
ses-0.097
 Mil-0.095
Tokenind
Token ID521
Feature activation-5.97e-02

Pos logit contributions
 envelop+0.147
Offic+0.141
Mic+0.139
 lending+0.125
 permitting+0.122
Neg logit contributions
gan-0.212
tub-0.210
flies-0.193
ider-0.189
uffy-0.187
Tokeneer
Token ID28153
Feature activation-4.79e-02

Pos logit contributions
 envelop+0.110
Offic+0.101
 Amen+0.099
ons+0.095
Mic+0.089
Neg logit contributions
tub-0.119
gan-0.114
 Moe-0.103
ider-0.102
ahl-0.099
Token burn
Token ID4245
Feature activation-5.24e-03

Pos logit contributions
tub+0.020
gan+0.019
ider+0.017
 Moe+0.017
ahl+0.017
Neg logit contributions
 envelop-0.019
 Amen-0.017
Offic-0.017
ons-0.016
Mic-0.015
Token you
Token ID345
Feature activation+2.40e-02

Pos logit contributions
 envelop+0.009
ons+0.008
 Amen+0.008
Offic+0.007
oc+0.007
Neg logit contributions
tub-0.011
gan-0.011
ider-0.010
 Moe-0.010
ahl-0.010
Token!"
Token ID2474
Feature activation+4.38e-03

Pos logit contributions
tub+0.045
gan+0.043
 Moe+0.040
ider+0.039
ahl+0.039
Neg logit contributions
 envelop-0.046
 Amen-0.041
ons-0.041
Offic-0.040
oc-0.038
Token They
Token ID1119
Feature activation-1.70e-02

Pos logit contributions
tub+0.007
gan+0.006
ider+0.006
 Moe+0.006
ahl+0.005
Neg logit contributions
 envelop-0.010
Offic-0.009
 Amen-0.009
ons-0.009
Mic-0.008
Token wear
Token ID5806
Feature activation-3.98e-02

Pos logit contributions
 envelop+0.030
Offic+0.027
 Amen+0.026
ons+0.025
Mic+0.024
Neg logit contributions
tub-0.036
gan-0.034
ider-0.031
 Moe-0.031
ahl-0.030
Token apr
Token ID46593
Feature activation-1.34e-01

Pos logit contributions
 envelop+0.079
Offic+0.076
 Amen+0.074
ons+0.070
Mic+0.068
Neg logit contributions
tub-0.073
gan-0.069
 Moe-0.061
ider-0.059
ahl-0.058
Tokenons
Token ID684
Feature activation-6.58e-02

Pos logit contributions
Offic+0.129
 Amen+0.122
 envelop+0.112
Mic+0.089
 saddle+0.088
Neg logit contributions
tub-0.351
gan-0.347
 Moe-0.330
ia-0.320
icky-0.311
Token and
Token ID290
Feature activation-8.29e-02

Pos logit contributions
 envelop+0.123
Offic+0.108
 Amen+0.106
ons+0.106
Mic+0.096
Neg logit contributions
tub-0.130
gan-0.125
ider-0.114
 Moe-0.113
ahl-0.111
Token hats
Token ID23910
Feature activation-7.19e-02

Pos logit contributions
Offic+0.154
 Amen+0.154
 envelop+0.154
Mic+0.141
ons+0.137
Neg logit contributions
tub-0.147
gan-0.142
ses-0.124
 Mil-0.122
 Moe-0.122
Token and
Token ID290
Feature activation-7.69e-02

Pos logit contributions
 envelop+0.139
Offic+0.122
 Amen+0.120
ons+0.119
oc+0.107
Neg logit contributions
tub-0.136
gan-0.130
 Moe-0.119
ahl-0.117
ider-0.117
Token act
Token ID719
Feature activation-1.82e-02

Pos logit contributions
Offic+0.140
 Amen+0.140
 envelop+0.139
ons+0.126
Mic+0.126
Neg logit contributions
tub-0.139
gan-0.136
ses-0.120
sed-0.117
 Mil-0.116
Token I
Token ID314
Feature activation-1.82e-03

Pos logit contributions
tub+0.006
gan+0.006
 Moe+0.005
ider+0.005
ahl+0.005
Neg logit contributions
 envelop-0.010
 Amen-0.009
Offic-0.009
ons-0.009
oc-0.009
Token also
Token ID635
Feature activation-1.30e-02

Pos logit contributions
 envelop+0.003
Offic+0.003
 Amen+0.003
ons+0.003
oc+0.003
Neg logit contributions
tub-0.004
gan-0.003
 Moe-0.003
ider-0.003
ahl-0.003
Token have
Token ID423
Feature activation-2.47e-02

Pos logit contributions
 envelop+0.024
Offic+0.022
 Amen+0.021
Mic+0.021
ons+0.020
Neg logit contributions
tub-0.025
gan-0.024
ider-0.022
ses-0.021
 Moe-0.021
Token a
Token ID257
Feature activation-5.00e-02

Pos logit contributions
 envelop+0.054
Offic+0.053
 Amen+0.053
Mic+0.049
ons+0.049
Neg logit contributions
tub-0.037
gan-0.036
 Moe-0.030
ider-0.030
 Mil-0.029
Token magic
Token ID5536
Feature activation-1.34e-02

Pos logit contributions
Offic+0.091
ons+0.089
 envelop+0.088
Mic+0.083
 Amen+0.081
Neg logit contributions
tub-0.092
ses-0.078
gan-0.078
sed-0.077
 fault-0.075
Token sle
Token ID3133
Feature activation-1.34e-01

Pos logit contributions
 envelop+0.023
Offic+0.022
 Amen+0.020
ons+0.020
Mic+0.019
Neg logit contributions
tub-0.028
gan-0.026
ider-0.024
 Moe-0.023
ses-0.023
Tokenigh
Token ID394
Feature activation-5.31e-02

Pos logit contributions
Offic+0.157
 envelop+0.155
Mic+0.147
 pic+0.123
 inviting+0.121
Neg logit contributions
ider-0.311
tub-0.303
gan-0.300
ses-0.290
sed-0.282
Token and
Token ID290
Feature activation-7.26e-02

Pos logit contributions
 envelop+0.107
Offic+0.099
 Amen+0.097
ons+0.093
Mic+0.088
Neg logit contributions
tub-0.097
gan-0.092
 Moe-0.082
ider-0.081
ahl-0.079
Token re
Token ID302
Feature activation-9.99e-02

Pos logit contributions
 envelop+0.140
Offic+0.135
 Amen+0.130
ons+0.125
Mic+0.125
Neg logit contributions
tub-0.126
gan-0.118
ses-0.107
ider-0.107
 Moe-0.105
Tokenind
Token ID521
Feature activation-6.01e-02

Pos logit contributions
Mic+0.152
 envelop+0.152
Offic+0.148
 lending+0.134
 permitting+0.134
Neg logit contributions
gan-0.187
tub-0.176
flies-0.171
ider-0.171
uffy-0.167
Tokeneer
Token ID28153
Feature activation-4.83e-02

Pos logit contributions
 envelop+0.116
Offic+0.108
 Amen+0.104
ons+0.100
Mic+0.097
Neg logit contributions
tub-0.114
gan-0.109
ider-0.097
 Moe-0.096
ahl-0.093
Token Tom
Token ID4186
Feature activation-6.15e-03

Pos logit contributions
tub+0.006
gan+0.005
ider+0.005
 Moe+0.005
ahl+0.005
Neg logit contributions
 envelop-0.009
Offic-0.008
 Amen-0.008
ons-0.007
Mic-0.007
Token said
Token ID531
Feature activation-2.57e-02

Pos logit contributions
 envelop+0.011
Offic+0.010
 Amen+0.009
ons+0.009
oc+0.009
Neg logit contributions
tub-0.013
gan-0.012
ider-0.011
 Moe-0.011
ahl-0.011
Token,
Token ID11
Feature activation-8.86e-03

Pos logit contributions
 envelop+0.047
 Amen+0.042
Offic+0.039
ons+0.038
oc+0.038
Neg logit contributions
tub-0.050
ses-0.044
gan-0.044
ahl-0.043
ider-0.042
Token holding
Token ID4769
Feature activation-2.40e-02

Pos logit contributions
 envelop+0.017
Offic+0.015
ons+0.015
 Amen+0.015
Mic+0.014
Neg logit contributions
tub-0.017
gan-0.017
ider-0.015
 Moe-0.015
 Mil-0.014
Token a
Token ID257
Feature activation-5.11e-02

Pos logit contributions
 envelop+0.047
Offic+0.044
 Amen+0.043
ons+0.041
Mic+0.039
Neg logit contributions
tub-0.045
gan-0.043
 Moe-0.039
ider-0.038
ahl-0.037
Token stuffed
Token ID22259
Feature activation-1.33e-01

Pos logit contributions
Offic+0.103
 envelop+0.101
 Amen+0.100
ons+0.097
itivity+0.093
Neg logit contributions
tub-0.088
gan-0.075
ses-0.070
sed-0.069
ahl-0.067
Token dog
Token ID3290
Feature activation-5.70e-02

Pos logit contributions
 envelop+0.289
Offic+0.281
 Amen+0.275
ons+0.273
itivity+0.264
Neg logit contributions
tub-0.194
gan-0.173
 Moe-0.164
ahl-0.162
ter-0.161
Token as
Token ID355
Feature activation-3.93e-03

Pos logit contributions
 envelop+0.127
Offic+0.124
 Amen+0.121
ons+0.116
Mic+0.113
Neg logit contributions
tub-0.086
gan-0.081
ider-0.075
 Moe-0.073
ses-0.066
Token his
Token ID465
Feature activation+1.15e-02

Pos logit contributions
 envelop+0.008
Offic+0.007
ons+0.007
 Amen+0.007
 parcels+0.006
Neg logit contributions
tub-0.007
gan-0.007
ahl-0.006
ider-0.006
 Moe-0.006
Token bird
Token ID6512
Feature activation-1.31e-02

Pos logit contributions
tub+0.024
gan+0.023
 Moe+0.020
ider+0.020
ahl+0.020
Neg logit contributions
 envelop-0.020
Offic-0.019
 Amen-0.018
ons-0.017
Mic-0.016
Token.
Token ID13
Feature activation+1.37e-03

Pos logit contributions
 envelop+0.026
Offic+0.024
 Amen+0.024
ons+0.023
Mic+0.022
Neg logit contributions
tub-0.024
gan-0.023
ider-0.020
 Moe-0.020
ahl-0.020
Token outside
Token ID2354
Feature activation+3.94e-02

Pos logit contributions
tub+0.044
gan+0.042
 Moe+0.037
ider+0.035
ahl+0.034
Neg logit contributions
 envelop-0.064
 Amen-0.063
Offic-0.063
ons-0.059
Mic-0.058
Token and
Token ID290
Feature activation-7.23e-02

Pos logit contributions
tub+0.073
gan+0.068
 Moe+0.061
ider+0.060
 Mil+0.058
Neg logit contributions
 envelop-0.079
Offic-0.074
 Amen-0.072
ons-0.069
Mic-0.068
Token showed
Token ID3751
Feature activation-4.95e-02

Pos logit contributions
 envelop+0.129
Offic+0.119
 Amen+0.117
ons+0.111
Mic+0.105
Neg logit contributions
tub-0.148
gan-0.142
 Moe-0.128
ider-0.127
ahl-0.123
Token them
Token ID606
Feature activation+1.95e-02

Pos logit contributions
 envelop+0.068
Offic+0.061
 Amen+0.059
ons+0.055
Mic+0.051
Neg logit contributions
tub-0.122
gan-0.118
ider-0.108
 Moe-0.108
ahl-0.105
Token his
Token ID465
Feature activation+6.50e-03

Pos logit contributions
tub+0.045
gan+0.042
ider+0.041
ses+0.039
sed+0.038
Neg logit contributions
 envelop-0.032
Offic-0.027
ons-0.026
 Amen-0.025
 parcels-0.023
Token sle
Token ID3133
Feature activation-1.33e-01

Pos logit contributions
tub+0.013
gan+0.012
ses+0.011
sed+0.011
ider+0.011
Neg logit contributions
Offic-0.011
 envelop-0.011
Mic-0.011
ons-0.010
 Amen-0.010
Tokenigh
Token ID394
Feature activation-4.88e-02

Pos logit contributions
Offic+0.150
 envelop+0.143
Mic+0.138
 permitting+0.114
 inviting+0.112
Neg logit contributions
tub-0.310
gan-0.308
ider-0.299
uffy-0.293
ses-0.292
Token and
Token ID290
Feature activation-8.15e-02

Pos logit contributions
 envelop+0.107
Offic+0.100
 Amen+0.098
ons+0.094
Mic+0.091
Neg logit contributions
tub-0.080
gan-0.075
ider-0.066
 Moe-0.066
ahl-0.063
Token re
Token ID302
Feature activation-1.01e-01

Pos logit contributions
 Amen+0.136
Offic+0.134
 envelop+0.129
Mic+0.125
ons+0.124
Neg logit contributions
tub-0.156
gan-0.143
ses-0.136
sed-0.135
 Moe-0.134
Tokenind
Token ID521
Feature activation-6.64e-02

Pos logit contributions
 envelop+0.157
Offic+0.151
Mic+0.149
 permitting+0.137
 lending+0.135
Neg logit contributions
gan-0.197
tub-0.187
uffy-0.180
flies-0.170
anda-0.170
Tokeneer
Token ID28153
Feature activation-4.86e-02

Pos logit contributions
 envelop+0.126
Offic+0.117
 Amen+0.113
ons+0.109
Mic+0.103
Neg logit contributions
tub-0.129
gan-0.123
 Moe-0.110
ider-0.110
ahl-0.106
Token put
Token ID1234
Feature activation-4.22e-02

Pos logit contributions
 envelop+0.006
Offic+0.006
 Amen+0.005
ons+0.005
Mic+0.005
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.004
ider-0.004
ahl-0.004
Token a
Token ID257
Feature activation-4.48e-02

Pos logit contributions
 envelop+0.072
Offic+0.066
 Amen+0.065
ons+0.061
Mic+0.058
Neg logit contributions
tub-0.090
gan-0.086
 Moe-0.078
ider-0.078
ahl-0.076
Token shiny
Token ID22441
Feature activation-6.27e-02

Pos logit contributions
 envelop+0.083
Offic+0.083
 Amen+0.081
ons+0.076
Mic+0.072
Neg logit contributions
tub-0.085
gan-0.077
ider-0.069
 Moe-0.068
ahl-0.068
Token saddle
Token ID33083
Feature activation-9.04e-02

Pos logit contributions
 envelop+0.108
Offic+0.097
 Amen+0.095
ons+0.091
Mic+0.085
Neg logit contributions
tub-0.133
gan-0.128
 Moe-0.115
ider-0.115
ahl-0.112
Token and
Token ID290
Feature activation-7.42e-02

Pos logit contributions
 envelop+0.165
Offic+0.150
 Amen+0.148
ons+0.142
Mic+0.132
Neg logit contributions
tub-0.183
gan-0.174
 Moe-0.157
ider-0.156
ahl-0.152
Token brid
Token ID38265
Feature activation-1.33e-01

Pos logit contributions
 envelop+0.146
 Amen+0.141
Offic+0.137
ons+0.127
Mic+0.125
Neg logit contributions
tub-0.132
gan-0.128
 Moe-0.113
ider-0.110
 Mil-0.110
Tokenle
Token ID293
Feature activation-1.19e-02

Pos logit contributions
 envelop+0.252
Offic+0.235
 Amen+0.228
Mic+0.211
ons+0.210
Neg logit contributions
tub-0.249
gan-0.247
 Moe-0.221
ider-0.220
ahl-0.214
Token on
Token ID319
Feature activation-2.38e-02

Pos logit contributions
 envelop+0.023
ons+0.020
Offic+0.020
 Amen+0.019
oc+0.018
Neg logit contributions
tub-0.023
gan-0.022
ider-0.020
 Moe-0.020
ahl-0.020
Token him
Token ID683
Feature activation+2.65e-02

Pos logit contributions
 envelop+0.046
 Amen+0.046
Offic+0.045
ons+0.041
itivity+0.040
Neg logit contributions
tub-0.041
gan-0.040
 Moe-0.037
 Mil-0.036
ahl-0.034
Token.
Token ID13
Feature activation+1.44e-02

Pos logit contributions
tub+0.046
gan+0.043
ider+0.038
 Moe+0.038
ahl+0.036
Neg logit contributions
 envelop-0.056
Offic-0.053
 Amen-0.051
ons-0.049
Mic-0.047
Token Finally
Token ID9461
Feature activation+4.95e-02

Pos logit contributions
tub+0.024
gan+0.022
ider+0.021
ahl+0.019
 Moe+0.019
Neg logit contributions
 envelop-0.032
Offic-0.030
 Amen-0.028
Mic-0.027
ons-0.027
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
tub+0.046
gan+0.043
ider+0.038
 Moe+0.037
ahl+0.036
Neg logit contributions
 envelop-0.061
Offic-0.056
 Amen-0.055
ons-0.053
Mic-0.051
Token She
Token ID1375
Feature activation-1.53e-03

Pos logit contributions
tub+0.020
ider+0.017
gan+0.016
ulu+0.016
auc+0.016
Neg logit contributions
 envelop-0.020
 Amen-0.020
Offic-0.018
 parcels-0.017
 read-0.016
Token grabbed
Token ID13176
Feature activation-2.63e-02

Pos logit contributions
 envelop+0.003
Offic+0.003
 Amen+0.003
ons+0.003
Mic+0.003
Neg logit contributions
tub-0.003
gan-0.003
 Moe-0.002
ider-0.002
ahl-0.002
Token the
Token ID262
Feature activation+2.30e-02

Pos logit contributions
 envelop+0.053
Offic+0.052
 Amen+0.051
ons+0.047
oc+0.047
Neg logit contributions
tub-0.045
gan-0.044
 Moe-0.040
 Mil-0.038
ahl-0.037
Token fire
Token ID2046
Feature activation-3.49e-03

Pos logit contributions
tub+0.047
gan+0.046
 Moe+0.041
ider+0.041
ahl+0.041
Neg logit contributions
 envelop-0.042
 Amen-0.038
Offic-0.038
ons-0.035
oc-0.033
Token extingu
Token ID24995
Feature activation-1.33e-01

Pos logit contributions
 envelop+0.006
 Amen+0.005
Offic+0.005
oc+0.005
ons+0.005
Neg logit contributions
tub-0.007
gan-0.007
ahl-0.006
 Moe-0.006
ider-0.006
Tokenisher
Token ID4828
Feature activation+3.21e-03

Pos logit contributions
 envelop+0.307
Offic+0.298
 Amen+0.282
Mic+0.279
ons+0.274
Neg logit contributions
gan-0.190
tub-0.186
ider-0.162
 Moe-0.161
ahl-0.150
Token and
Token ID290
Feature activation-7.00e-02

Pos logit contributions
tub+0.006
gan+0.006
 Moe+0.006
ider+0.006
ahl+0.005
Neg logit contributions
 envelop-0.006
Offic-0.005
 Amen-0.005
ons-0.005
oc-0.005
Token put
Token ID1234
Feature activation-3.81e-02

Pos logit contributions
 envelop+0.132
 Amen+0.123
Offic+0.123
ons+0.114
Mic+0.112
Neg logit contributions
tub-0.135
gan-0.130
 Moe-0.116
ider-0.112
 Mil-0.111
Token out
Token ID503
Feature activation+3.70e-03

Pos logit contributions
 envelop+0.067
Offic+0.064
 Amen+0.061
ons+0.057
Mic+0.056
Neg logit contributions
tub-0.078
gan-0.075
 Moe-0.068
ider-0.065
ahl-0.065
Token the
Token ID262
Feature activation+3.19e-02

Pos logit contributions
tub+0.008
gan+0.007
 Moe+0.007
ider+0.007
ahl+0.007
Neg logit contributions
 envelop-0.006
Offic-0.006
 Amen-0.006
ons-0.005
Mic-0.005
Token put
Token ID1234
Feature activation-3.88e-02

Pos logit contributions
 envelop+0.006
 Amen+0.005
Offic+0.005
ons+0.005
Mic+0.005
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.004
ider-0.004
 Mil-0.004
Token a
Token ID257
Feature activation-4.61e-02

Pos logit contributions
 envelop+0.074
Offic+0.071
 Amen+0.068
ons+0.064
Mic+0.063
Neg logit contributions
tub-0.072
gan-0.070
 Moe-0.065
ses-0.061
ahl-0.060
Token saddle
Token ID33083
Feature activation-9.27e-02

Pos logit contributions
ons+0.089
Offic+0.089
 envelop+0.089
 Amen+0.088
itivity+0.085
Neg logit contributions
tub-0.079
gan-0.067
ider-0.062
sed-0.062
ses-0.062
Token and
Token ID290
Feature activation-7.66e-02

Pos logit contributions
 envelop+0.179
Offic+0.167
 Amen+0.162
ons+0.156
Mic+0.151
Neg logit contributions
tub-0.177
gan-0.169
ider-0.152
 Moe-0.149
ahl-0.144
Token a
Token ID257
Feature activation-5.02e-02

Pos logit contributions
 envelop+0.153
 Amen+0.149
Offic+0.142
ons+0.136
Mic+0.132
Neg logit contributions
tub-0.131
gan-0.128
 Moe-0.113
ses-0.111
 Mil-0.110
Token brid
Token ID38265
Feature activation-1.33e-01

Pos logit contributions
 envelop+0.086
Offic+0.084
ons+0.084
 Amen+0.082
itivity+0.076
Neg logit contributions
tub-0.101
gan-0.090
ider-0.086
ses-0.082
sed-0.081
Tokenle
Token ID293
Feature activation-1.66e-02

Pos logit contributions
 envelop+0.270
Offic+0.248
 Amen+0.240
Mic+0.236
 parcels+0.226
Neg logit contributions
tub-0.222
gan-0.220
ider-0.204
 Moe-0.198
ahl-0.192
Token on
Token ID319
Feature activation-1.92e-02

Pos logit contributions
 envelop+0.031
Offic+0.028
 Amen+0.027
ons+0.026
oc+0.024
Neg logit contributions
tub-0.033
gan-0.032
 Moe-0.029
ider-0.028
ahl-0.028
Token the
Token ID262
Feature activation+3.62e-02

Pos logit contributions
 Amen+0.040
 envelop+0.039
Offic+0.038
ons+0.036
itivity+0.036
Neg logit contributions
tub-0.029
gan-0.029
 Moe-0.027
 Mil-0.026
ses-0.025
Token horse
Token ID8223
Feature activation-2.91e-02

Pos logit contributions
tub+0.076
gan+0.073
ses+0.067
sed+0.066
ider+0.066
Neg logit contributions
 envelop-0.056
ons-0.055
Offic-0.054
 Amen-0.052
Mic-0.048
Token.
Token ID13
Feature activation+1.54e-02

Pos logit contributions
 envelop+0.071
Offic+0.069
 Amen+0.068
Mic+0.064
ons+0.063
Neg logit contributions
tub-0.039
gan-0.035
ider-0.031
 Moe-0.029
ses-0.028
Token.
Token ID13
Feature activation+1.02e-02

Pos logit contributions
tub+0.006
gan+0.004
 Moe+0.004
 Mil+0.004
ider+0.004
Neg logit contributions
Offic-0.013
Mic-0.013
 envelop-0.013
itivity-0.012
ons-0.012
Token He
Token ID679
Feature activation-6.93e-03

Pos logit contributions
tub+0.017
gan+0.016
ider+0.014
ter+0.014
ses+0.014
Neg logit contributions
 envelop-0.022
 Amen-0.021
Offic-0.020
 read-0.018
oc-0.018
Token gave
Token ID2921
Feature activation-6.92e-02

Pos logit contributions
 envelop+0.013
Offic+0.012
 Amen+0.012
ons+0.012
Mic+0.011
Neg logit contributions
tub-0.013
gan-0.013
 Moe-0.011
ider-0.011
ahl-0.011
Token her
Token ID607
Feature activation-2.86e-03

Pos logit contributions
 envelop+0.115
 Amen+0.102
Offic+0.098
ons+0.097
Mic+0.085
Neg logit contributions
tub-0.154
gan-0.143
ider-0.137
ses-0.131
 Moe-0.129
Token a
Token ID257
Feature activation-5.81e-02

Pos logit contributions
 envelop+0.006
Offic+0.006
 Amen+0.006
ons+0.006
Mic+0.006
Neg logit contributions
tub-0.004
gan-0.004
 Moe-0.004
ider-0.004
ahl-0.003
Token stuffed
Token ID22259
Feature activation-1.31e-01

Pos logit contributions
 envelop+0.100
Offic+0.094
 Amen+0.090
ons+0.086
Mic+0.081
Neg logit contributions
tub-0.123
gan-0.115
 Moe-0.105
ider-0.104
ahl-0.103
Token bear
Token ID6842
Feature activation-5.24e-02

Pos logit contributions
 envelop+0.281
Offic+0.267
 Amen+0.259
ons+0.249
Mic+0.242
Neg logit contributions
tub-0.217
gan-0.201
 Moe-0.179
ahl-0.177
 Mil-0.171
Token and
Token ID290
Feature activation-8.09e-02

Pos logit contributions
 envelop+0.108
Offic+0.104
 Amen+0.101
ons+0.097
Mic+0.093
Neg logit contributions
tub-0.090
gan-0.086
ider-0.075
 Moe-0.075
ahl-0.071
Token a
Token ID257
Feature activation-5.77e-02

Pos logit contributions
 envelop+0.177
Offic+0.166
 Amen+0.163
ons+0.157
Mic+0.150
Neg logit contributions
tub-0.133
gan-0.125
 Moe-0.110
ider-0.109
ahl-0.105
Token night
Token ID1755
Feature activation+3.86e-02

Pos logit contributions
 envelop+0.099
Offic+0.092
 Amen+0.089
ons+0.085
Mic+0.080
Neg logit contributions
tub-0.122
gan-0.115
ider-0.105
 Moe-0.104
ahl-0.102
Token light
Token ID1657
Feature activation-1.17e-02

Pos logit contributions
gan+0.074
tub+0.070
 Moe+0.065
ahl+0.065
ses+0.063
Neg logit contributions
 envelop-0.074
 Amen-0.064
ons-0.064
Offic-0.063
oc-0.061
Token trial
Token ID4473
Feature activation+1.01e-02

Pos logit contributions
tub+0.096
ider+0.083
gan+0.081
ses+0.079
sed+0.075
Neg logit contributions
 Amen-0.122
 envelop-0.115
Offic-0.112
itivity-0.106
ons-0.104
Token.
Token ID13
Feature activation+1.20e-02

Pos logit contributions
tub+0.019
gan+0.018
ider+0.016
 Moe+0.016
ses+0.015
Neg logit contributions
 envelop-0.020
Offic-0.019
 Amen-0.019
ons-0.017
Mic-0.017
Token They
Token ID1119
Feature activation-9.96e-03

Pos logit contributions
tub+0.020
gan+0.019
ider+0.018
ahl+0.017
ble+0.017
Neg logit contributions
 envelop-0.027
Offic-0.023
 Amen-0.022
 parcels-0.021
 read-0.020
Token used
Token ID973
Feature activation-4.99e-02

Pos logit contributions
 Amen+0.019
 envelop+0.019
Offic+0.018
ons+0.017
Mic+0.016
Neg logit contributions
gan-0.018
tub-0.018
 Moe-0.016
ahl-0.015
 Mil-0.015
Token a
Token ID257
Feature activation-5.05e-02

Pos logit contributions
 Amen+0.103
Offic+0.103
 envelop+0.100
Mic+0.092
oc+0.092
Neg logit contributions
tub-0.083
gan-0.078
 Moe-0.072
ses-0.069
 Mil-0.067
Token stuffed
Token ID22259
Feature activation-1.31e-01

Pos logit contributions
 envelop+0.093
 Amen+0.092
Offic+0.091
ons+0.085
Mic+0.080
Neg logit contributions
tub-0.095
gan-0.084
ses-0.078
ider-0.077
 Moe-0.077
Token dog
Token ID3290
Feature activation-5.88e-02

Pos logit contributions
 Amen+0.269
 envelop+0.262
Offic+0.258
ons+0.244
Mic+0.243
Neg logit contributions
tub-0.201
gan-0.179
ses-0.178
sed-0.168
 Moe-0.168
Token and
Token ID290
Feature activation-7.85e-02

Pos logit contributions
 envelop+0.121
Offic+0.113
 Amen+0.110
ons+0.106
Mic+0.101
Neg logit contributions
tub-0.105
gan-0.099
 Moe-0.088
ider-0.088
ahl-0.085
Token a
Token ID257
Feature activation-4.90e-02

Pos logit contributions
 envelop+0.162
 Amen+0.155
Offic+0.154
ons+0.145
Mic+0.139
Neg logit contributions
tub-0.133
gan-0.127
 Moe-0.112
ider-0.109
ses-0.108
Token stuffed
Token ID22259
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.091
Offic+0.086
 Amen+0.085
ons+0.081
Mic+0.076
Neg logit contributions
tub-0.095
gan-0.088
ider-0.080
 Moe-0.079
ses-0.078
Token cat
Token ID3797
Feature activation-5.05e-02

Pos logit contributions
 envelop+0.260
 Amen+0.251
Offic+0.246
ons+0.242
Mic+0.229
Neg logit contributions
tub-0.209
gan-0.192
ses-0.178
 Moe-0.177
sed-0.170
Token come
Token ID1282
Feature activation+3.35e-02

Pos logit contributions
 envelop+0.030
Offic+0.028
 Amen+0.027
ons+0.027
Mic+0.025
Neg logit contributions
tub-0.022
gan-0.020
ider-0.018
 Moe-0.018
ahl-0.017
Token in
Token ID287
Feature activation+2.10e-02

Pos logit contributions
tub+0.056
gan+0.053
 Moe+0.046
ider+0.046
ahl+0.044
Neg logit contributions
 envelop-0.073
Offic-0.068
 Amen-0.066
ons-0.064
Mic-0.061
Token,"
Token ID553
Feature activation-2.50e-02

Pos logit contributions
tub+0.035
gan+0.033
 Moe+0.029
ider+0.029
ahl+0.027
Neg logit contributions
 envelop-0.045
Offic-0.043
 Amen-0.041
ons-0.040
Mic-0.039
Token said
Token ID531
Feature activation-6.80e-03

Pos logit contributions
 envelop+0.054
Offic+0.049
 Amen+0.049
ons+0.047
Mic+0.045
Neg logit contributions
tub-0.043
gan-0.039
ider-0.036
ahl-0.035
 Moe-0.034
Token his
Token ID465
Feature activation-6.59e-03

Pos logit contributions
 envelop+0.013
 Amen+0.012
Offic+0.012
ons+0.011
Mic+0.010
Neg logit contributions
tub-0.013
gan-0.012
ider-0.011
ahl-0.011
ses-0.011
Token grand
Token ID4490
Feature activation-1.29e-01

Pos logit contributions
 envelop+0.012
 Amen+0.011
Offic+0.010
ons+0.009
Mic+0.009
Neg logit contributions
tub-0.014
gan-0.012
ahl-0.011
 Moe-0.011
ider-0.011
Tokenpa
Token ID8957
Feature activation-7.82e-02

Pos logit contributions
 envelop+0.105
Offic+0.102
ons+0.081
Mic+0.080
 Amen+0.079
Neg logit contributions
gan-0.360
tub-0.359
 Moe-0.331
ider-0.327
ses-0.324
Token.
Token ID13
Feature activation+4.93e-04

Pos logit contributions
 envelop+0.164
 Amen+0.151
Offic+0.150
ons+0.145
oc+0.136
Neg logit contributions
tub-0.138
gan-0.127
ider-0.114
 Moe-0.114
ahl-0.112
Token They
Token ID1119
Feature activation-2.52e-02

Pos logit contributions
tub+0.001
gan+0.001
ider+0.001
 Moe+0.001
ahl+0.001
Neg logit contributions
 envelop-0.001
Offic-0.001
 Amen-0.001
ons-0.001
Mic-0.001
Token all
Token ID477
Feature activation-1.53e-02

Pos logit contributions
 envelop+0.050
 Amen+0.047
Offic+0.046
ons+0.044
Mic+0.043
Neg logit contributions
tub-0.045
gan-0.044
 Moe-0.039
 Mil-0.036
ahl-0.036
Token turned
Token ID2900
Feature activation+8.76e-03

Pos logit contributions
 envelop+0.029
Offic+0.027
 Amen+0.026
ons+0.025
Mic+0.024
Neg logit contributions
tub-0.030
gan-0.029
 Moe-0.026
ider-0.025
ahl-0.024
Token so
Token ID523
Feature activation-2.31e-02

Pos logit contributions
tub+0.021
gan+0.020
 Moe+0.017
ider+0.016
 Mil+0.016
Neg logit contributions
 envelop-0.033
Offic-0.031
 Amen-0.031
ons-0.029
Mic-0.028
Token he
Token ID339
Feature activation+1.78e-02

Pos logit contributions
 envelop+0.054
Offic+0.051
 Amen+0.050
ons+0.049
Mic+0.046
Neg logit contributions
tub-0.035
gan-0.032
ider-0.028
 Moe-0.028
ahl-0.027
Token gave
Token ID2921
Feature activation-6.42e-02

Pos logit contributions
tub+0.036
gan+0.035
 Moe+0.031
ider+0.031
ahl+0.030
Neg logit contributions
 envelop-0.032
Offic-0.029
 Amen-0.029
ons-0.027
Mic-0.026
Token her
Token ID607
Feature activation-4.80e-04

Pos logit contributions
 envelop+0.110
Offic+0.108
 Amen+0.104
Mic+0.094
oc+0.093
Neg logit contributions
gan-0.128
tub-0.126
 Moe-0.120
 Mil-0.112
ahl-0.110
Token a
Token ID257
Feature activation-5.71e-02

Pos logit contributions
 envelop+0.001
Offic+0.001
 Amen+0.001
ons+0.001
Mic+0.001
Neg logit contributions
tub-0.001
gan-0.001
ider-0.001
 Moe-0.001
ahl-0.001
Token stuffed
Token ID22259
Feature activation-1.29e-01

Pos logit contributions
 envelop+0.103
Offic+0.097
 Amen+0.094
ons+0.090
Mic+0.085
Neg logit contributions
tub-0.116
gan-0.108
 Moe-0.098
ider-0.096
ahl-0.095
Token bear
Token ID6842
Feature activation-5.03e-02

Pos logit contributions
 envelop+0.274
Offic+0.260
 Amen+0.257
ons+0.247
Mic+0.239
Neg logit contributions
tub-0.214
gan-0.199
 Moe-0.181
ahl-0.172
 Mil-0.167
Token.
Token ID13
Feature activation+8.26e-03

Pos logit contributions
 envelop+0.108
Offic+0.102
 Amen+0.100
ons+0.096
Mic+0.092
Neg logit contributions
tub-0.084
gan-0.079
 Moe-0.069
ider-0.069
ahl-0.066
Token
Token ID198
Feature activation-2.08e-02

Pos logit contributions
tub+0.014
gan+0.013
ider+0.012
ses+0.011
ahl+0.011
Neg logit contributions
 envelop-0.018
Offic-0.017
 Amen-0.016
ons-0.016
Mic-0.015
Token
Token ID198
Feature activation-1.71e-02

Pos logit contributions
 envelop+0.055
 Amen+0.050
Offic+0.050
 parcels+0.047
itivity+0.047
Neg logit contributions
tub-0.025
gan-0.022
 Moe-0.020
 Mil-0.018
ahl-0.018
TokenThe
Token ID464
Feature activation+1.99e-02

Pos logit contributions
 envelop+0.039
 Amen+0.035
ons+0.033
 parcels+0.032
pa+0.032
Neg logit contributions
tub-0.024
gan-0.023
 Moe-0.022
ulu-0.020
ider-0.020
Token girl
Token ID2576
Feature activation+1.10e-03

Pos logit contributions
 envelop+0.045
 Amen+0.042
Offic+0.042
ons+0.039
oc+0.039
Neg logit contributions
tub-0.029
gan-0.029
ider-0.026
 Moe-0.025
ahl-0.025
Token and
Token ID290
Feature activation-5.88e-02

Pos logit contributions
tub+0.002
gan+0.002
 Mil+0.002
ider+0.002
 Moe+0.002
Neg logit contributions
 envelop-0.002
Offic-0.002
 parcels-0.002
 Amen-0.002
itivity-0.002
Token she
Token ID673
Feature activation+6.95e-03

Pos logit contributions
 envelop+0.112
 Amen+0.107
Offic+0.104
ons+0.099
Mic+0.093
Neg logit contributions
tub-0.115
gan-0.107
ider-0.099
 Moe-0.096
ahl-0.093
Token had
Token ID550
Feature activation-1.13e-02

Pos logit contributions
tub+0.014
gan+0.014
 Moe+0.013
ider+0.013
ahl+0.012
Neg logit contributions
 envelop-0.012
 Amen-0.011
Offic-0.011
ons-0.011
oc-0.010
Token a
Token ID257
Feature activation-4.85e-02

Pos logit contributions
 envelop+0.025
Offic+0.023
 Amen+0.023
ons+0.022
Mic+0.021
Neg logit contributions
tub-0.019
gan-0.018
ider-0.016
 Moe-0.015
ahl-0.015
Token stuffed
Token ID22259
Feature activation-1.29e-01

Pos logit contributions
 envelop+0.096
Offic+0.089
 Amen+0.088
ons+0.083
Mic+0.080
Neg logit contributions
tub-0.090
gan-0.086
 Moe-0.077
ider-0.077
ahl-0.074
Token bear
Token ID6842
Feature activation-5.03e-02

Pos logit contributions
 envelop+0.264
Offic+0.262
ons+0.253
 Amen+0.253
itivity+0.239
Neg logit contributions
tub-0.207
gan-0.185
ahl-0.166
 Moe-0.166
ter-0.165
Token called
Token ID1444
Feature activation-4.07e-02

Pos logit contributions
 envelop+0.090
Offic+0.087
 Amen+0.083
ons+0.079
Mic+0.077
Neg logit contributions
tub-0.102
gan-0.094
ider-0.088
 Moe-0.084
ses-0.081
Token Bear
Token ID14732
Feature activation-4.34e-02

Pos logit contributions
 envelop+0.087
Offic+0.079
 Amen+0.078
ons+0.076
Mic+0.072
Neg logit contributions
tub-0.070
gan-0.066
ider-0.061
 Moe-0.057
ahl-0.057
Tokeny
Token ID88
Feature activation-2.43e-02

Pos logit contributions
 envelop+0.093
 Amen+0.084
Offic+0.084
ons+0.080
oc+0.077
Neg logit contributions
tub-0.073
gan-0.070
ider-0.063
 Moe-0.062
ahl-0.060
Token.
Token ID13
Feature activation+1.27e-02

Pos logit contributions
 envelop+0.051
 Amen+0.047
Offic+0.046
ons+0.044
oc+0.042
Neg logit contributions
tub-0.042
gan-0.040
ider-0.036
 Moe-0.035
ahl-0.035
Token quickly
Token ID2952
Feature activation-6.55e-04

Pos logit contributions
tub+0.019
gan+0.019
 Moe+0.016
ider+0.016
ahl+0.016
Neg logit contributions
 envelop-0.022
Offic-0.020
 Amen-0.020
ons-0.019
Mic-0.018
Token checked
Token ID10667
Feature activation-1.87e-02

Pos logit contributions
 envelop+0.001
 Amen+0.001
Offic+0.001
ons+0.001
Mic+0.001
Neg logit contributions
tub-0.001
gan-0.001
 Moe-0.001
ider-0.001
ahl-0.001
Token for
Token ID329
Feature activation-2.88e-02

Pos logit contributions
 envelop+0.035
Offic+0.032
 Amen+0.032
ons+0.030
Mic+0.029
Neg logit contributions
tub-0.037
gan-0.035
 Moe-0.032
ider-0.031
ahl-0.031
Token a
Token ID257
Feature activation-3.64e-02

Pos logit contributions
 envelop+0.063
 Amen+0.062
Offic+0.060
Mic+0.057
oc+0.057
Neg logit contributions
tub-0.043
gan-0.041
 Moe-0.038
 Mil-0.035
ider-0.034
Token fire
Token ID2046
Feature activation-5.62e-03

Pos logit contributions
 envelop+0.076
Offic+0.071
 Amen+0.070
ons+0.068
Mic+0.064
Neg logit contributions
tub-0.064
gan-0.059
ider-0.053
 Moe-0.053
ahl-0.050
Token extingu
Token ID24995
Feature activation-1.29e-01

Pos logit contributions
 envelop+0.011
Offic+0.009
 Amen+0.009
oc+0.009
ons+0.009
Neg logit contributions
gan-0.011
tub-0.010
 Moe-0.010
ahl-0.010
 Mil-0.009
Tokenisher
Token ID4828
Feature activation-2.74e-03

Pos logit contributions
 envelop+0.296
Offic+0.285
 Amen+0.273
Mic+0.264
ons+0.264
Neg logit contributions
tub-0.186
gan-0.183
ider-0.158
 Moe-0.156
ses-0.147
Token,
Token ID11
Feature activation+1.87e-02

Pos logit contributions
 envelop+0.006
Offic+0.005
 Amen+0.005
ons+0.005
oc+0.005
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.004
ahl-0.004
ider-0.004
Token but
Token ID475
Feature activation+6.25e-03

Pos logit contributions
tub+0.026
gan+0.024
 Moe+0.021
ider+0.020
ahl+0.019
Neg logit contributions
 envelop-0.046
Offic-0.043
 Amen-0.043
ons-0.041
Mic-0.040
Token there
Token ID612
Feature activation+5.10e-02

Pos logit contributions
tub+0.013
gan+0.012
ider+0.011
ahl+0.011
 Moe+0.011
Neg logit contributions
 envelop-0.011
Offic-0.010
 Amen-0.010
ons-0.010
 read-0.009
Token wasn
Token ID2492
Feature activation+9.35e-03

Pos logit contributions
tub+0.081
gan+0.079
ider+0.077
ahl+0.068
ses+0.068
Neg logit contributions
 envelop-0.112
ons-0.104
Offic-0.102
 Amen-0.097
oc-0.093
Token Tom
Token ID4186
Feature activation-1.05e-02

Pos logit contributions
tub+0.015
ider+0.015
ses+0.014
gan+0.014
ter+0.014
Neg logit contributions
 envelop-0.020
Offic-0.018
 Amen-0.017
ons-0.017
Mic-0.016
Token and
Token ID290
Feature activation-7.05e-02

Pos logit contributions
 envelop+0.020
Offic+0.019
 Amen+0.019
ons+0.017
Mic+0.017
Neg logit contributions
tub-0.020
gan-0.019
 Moe-0.017
ider-0.017
ahl-0.016
Token Lily
Token ID20037
Feature activation-3.34e-03

Pos logit contributions
 envelop+0.195
Offic+0.186
 Amen+0.183
ons+0.179
Mic+0.172
Neg logit contributions
tub-0.076
gan-0.069
 Moe-0.056
ider-0.053
ahl-0.050
Token thanked
Token ID26280
Feature activation-5.98e-02

Pos logit contributions
 envelop+0.007
 Amen+0.006
Offic+0.006
ons+0.006
Mic+0.006
Neg logit contributions
tub-0.006
gan-0.006
 Moe-0.005
 Mil-0.005
ahl-0.005
Token their
Token ID511
Feature activation-2.44e-02

Pos logit contributions
 envelop+0.118
Offic+0.117
oc+0.108
 Amen+0.108
Mic+0.107
Neg logit contributions
gan-0.098
tub-0.096
 Moe-0.095
 Mil-0.090
 Mia-0.080
Token grand
Token ID4490
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.045
ons+0.043
Offic+0.042
 Amen+0.041
Mic+0.041
Neg logit contributions
gan-0.044
tub-0.044
 Moe-0.039
sed-0.039
ses-0.039
Tokenpa
Token ID8957
Feature activation-8.34e-02

Pos logit contributions
 envelop+0.123
Offic+0.115
Mic+0.098
ons+0.094
 Amen+0.093
Neg logit contributions
gan-0.341
tub-0.335
 Moe-0.315
ses-0.314
ia-0.310
Token and
Token ID290
Feature activation-9.78e-02

Pos logit contributions
 envelop+0.175
Offic+0.162
 Amen+0.159
ons+0.153
Mic+0.145
Neg logit contributions
tub-0.146
gan-0.138
 Moe-0.122
ider-0.122
ahl-0.117
Token ran
Token ID4966
Feature activation-1.53e-02

Pos logit contributions
 envelop+0.188
Offic+0.181
 Amen+0.179
ons+0.175
Mic+0.171
Neg logit contributions
gan-0.166
tub-0.164
 Moe-0.144
 Mil-0.139
ses-0.136
Token to
Token ID284
Feature activation-4.00e-02

Pos logit contributions
 envelop+0.034
Offic+0.033
 Amen+0.032
ons+0.031
Mic+0.030
Neg logit contributions
tub-0.023
gan-0.023
 Moe-0.021
ider-0.019
ahl-0.019
Token the
Token ID262
Feature activation+4.02e-02

Pos logit contributions
 envelop+0.092
Offic+0.090
ons+0.089
oc+0.088
Mic+0.088
Neg logit contributions
 Moe-0.051
tub-0.048
gan-0.048
 Mil-0.046
 Mia-0.042
Token and
Token ID290
Feature activation-8.67e-02

Pos logit contributions
gan+0.021
tub+0.021
 Moe+0.018
ider+0.017
ahl+0.016
Neg logit contributions
 envelop-0.027
Offic-0.027
 Amen-0.025
ons-0.024
Mic-0.023
Token thanked
Token ID26280
Feature activation-5.63e-02

Pos logit contributions
 envelop+0.171
Offic+0.162
 Amen+0.161
ons+0.153
Mic+0.151
Neg logit contributions
tub-0.153
gan-0.148
 Moe-0.132
ses-0.130
 Mil-0.128
Token her
Token ID607
Feature activation-7.45e-03

Pos logit contributions
 envelop+0.098
Offic+0.096
 Amen+0.094
Mic+0.091
oc+0.091
Neg logit contributions
gan-0.105
tub-0.104
 Mil-0.097
 Moe-0.095
ahl-0.087
Token grandma
Token ID49890
Feature activation-9.10e-02

Pos logit contributions
 envelop+0.014
Offic+0.013
ons+0.013
 Amen+0.013
Mic+0.012
Neg logit contributions
tub-0.014
gan-0.013
ider-0.012
ses-0.011
 Moe-0.011
Token and
Token ID290
Feature activation-1.06e-01

Pos logit contributions
 envelop+0.213
Offic+0.204
 Amen+0.198
ons+0.192
Mic+0.186
Neg logit contributions
tub-0.133
gan-0.127
 Moe-0.109
ider-0.107
ahl-0.102
Token grand
Token ID4490
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.194
 Amen+0.188
ons+0.188
Offic+0.186
oc+0.184
Neg logit contributions
gan-0.172
tub-0.167
ses-0.162
sed-0.157
 Mil-0.146
Tokenpa
Token ID8957
Feature activation-8.26e-02

Pos logit contributions
Offic+0.114
 envelop+0.112
Mic+0.103
 Amen+0.089
ons+0.089
Neg logit contributions
gan-0.335
tub-0.326
ses-0.315
 fault-0.307
 Moe-0.306
Token.
Token ID13
Feature activation+1.00e-02

Pos logit contributions
 envelop+0.185
Offic+0.171
 Amen+0.169
ons+0.163
Mic+0.154
Neg logit contributions
tub-0.133
gan-0.125
ider-0.109
 Moe-0.109
ahl-0.105
Token She
Token ID1375
Feature activation-1.65e-02

Pos logit contributions
tub+0.015
gan+0.014
ider+0.013
 Moe+0.012
ahl+0.012
Neg logit contributions
 envelop-0.023
Offic-0.022
 Amen-0.022
ons-0.020
Mic-0.020
Token said
Token ID531
Feature activation-3.30e-02

Pos logit contributions
 envelop+0.033
Offic+0.031
 Amen+0.030
ons+0.029
Mic+0.028
Neg logit contributions
tub-0.030
gan-0.029
 Moe-0.026
ider-0.025
ahl-0.024
Token they
Token ID484
Feature activation-2.86e-02

Pos logit contributions
 envelop+0.075
 Amen+0.070
Offic+0.070
ons+0.067
Mic+0.063
Neg logit contributions
tub-0.053
gan-0.049
ider-0.044
 Moe-0.042
ahl-0.042
Token They
Token ID1119
Feature activation-1.59e-02

Pos logit contributions
tub+0.025
gan+0.025
ider+0.024
ahl+0.022
sed+0.021
Neg logit contributions
 envelop-0.030
Offic-0.028
 Amen-0.027
 parcels-0.024
oc-0.024
Token thanked
Token ID26280
Feature activation-5.93e-02

Pos logit contributions
 envelop+0.032
 Amen+0.031
Offic+0.030
ons+0.029
Mic+0.028
Neg logit contributions
tub-0.028
gan-0.027
 Moe-0.024
ider-0.023
 Mil-0.023
Token their
Token ID511
Feature activation-2.23e-02

Pos logit contributions
Offic+0.122
 envelop+0.120
Mic+0.117
 Amen+0.116
ons+0.116
Neg logit contributions
gan-0.093
tub-0.091
 Mil-0.084
 Moe-0.084
ahl-0.070
Token grandma
Token ID49890
Feature activation-8.65e-02

Pos logit contributions
 envelop+0.043
Offic+0.041
 Amen+0.041
ons+0.040
Mic+0.039
Neg logit contributions
tub-0.040
gan-0.039
 Moe-0.035
ses-0.034
ider-0.034
Token and
Token ID290
Feature activation-1.06e-01

Pos logit contributions
 envelop+0.184
Offic+0.180
 Amen+0.174
ons+0.167
Mic+0.167
Neg logit contributions
tub-0.139
gan-0.135
 Moe-0.116
ider-0.111
 Mil-0.109
Token grand
Token ID4490
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.194
 Amen+0.189
ons+0.187
Offic+0.184
Mic+0.179
Neg logit contributions
gan-0.188
tub-0.188
ses-0.167
 Moe-0.164
sed-0.162
Tokenpa
Token ID8957
Feature activation-8.14e-02

Pos logit contributions
Offic+0.117
 envelop+0.111
Mic+0.107
 Amen+0.092
ons+0.088
Neg logit contributions
gan-0.335
tub-0.326
ses-0.313
 Moe-0.306
 fault-0.305
Token in
Token ID287
Feature activation-1.63e-02

Pos logit contributions
 envelop+0.174
Offic+0.165
 Amen+0.160
ons+0.154
Mic+0.149
Neg logit contributions
tub-0.137
gan-0.130
 Moe-0.114
ider-0.113
ahl-0.109
Token their
Token ID511
Feature activation-1.69e-02

Pos logit contributions
ons+0.037
Offic+0.037
Mic+0.037
itivity+0.034
 envelop+0.034
Neg logit contributions
tub-0.019
 Moe-0.019
 Mil-0.018
gan-0.017
ses-0.016
Token hearts
Token ID11954
Feature activation+1.41e-03

Pos logit contributions
Offic+0.034
 Amen+0.033
ons+0.033
Mic+0.032
 envelop+0.031
Neg logit contributions
tub-0.027
gan-0.025
ses-0.024
sed-0.023
 Moe-0.023
Token and
Token ID290
Feature activation-8.78e-02

Pos logit contributions
tub+0.002
gan+0.002
 Moe+0.002
ider+0.002
ahl+0.002
Neg logit contributions
 envelop-0.003
Offic-0.003
 Amen-0.003
ons-0.003
Mic-0.002
Token.
Token ID13
Feature activation+2.10e-02

Pos logit contributions
 envelop+0.051
Offic+0.050
 Amen+0.046
ons+0.046
 parcels+0.046
Neg logit contributions
gan-0.021
tub-0.021
 Mil-0.016
 Moe-0.015
ahl-0.015
Token She
Token ID1375
Feature activation-3.02e-03

Pos logit contributions
tub+0.026
gan+0.026
 Moe+0.020
ses+0.020
ahl+0.020
Neg logit contributions
 envelop-0.053
Offic-0.050
 Amen-0.048
ons-0.047
Mic-0.046
Token had
Token ID550
Feature activation-1.65e-02

Pos logit contributions
 envelop+0.005
 Amen+0.005
Offic+0.005
ons+0.005
Mic+0.004
Neg logit contributions
tub-0.006
gan-0.006
 Moe-0.005
ider-0.005
ahl-0.005
Token a
Token ID257
Feature activation-4.74e-02

Pos logit contributions
 envelop+0.037
Offic+0.035
 Amen+0.035
ons+0.033
Mic+0.032
Neg logit contributions
tub-0.026
gan-0.025
 Moe-0.021
ider-0.021
ahl-0.020
Token blue
Token ID4171
Feature activation-4.71e-02

Pos logit contributions
 envelop+0.097
Offic+0.090
 Amen+0.088
ons+0.085
Mic+0.081
Neg logit contributions
tub-0.085
gan-0.081
 Moe-0.072
ider-0.071
ahl-0.069
Token stuffed
Token ID22259
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.080
 Amen+0.072
Offic+0.071
ons+0.067
Mic+0.063
Neg logit contributions
tub-0.100
gan-0.097
 Moe-0.088
ider-0.087
ahl-0.086
Token animal
Token ID5044
Feature activation-3.32e-02

Pos logit contributions
 envelop+0.269
Offic+0.267
 Amen+0.256
ons+0.252
itivity+0.244
Neg logit contributions
tub-0.204
gan-0.182
 Moe-0.163
ahl-0.161
ter-0.159
Token named
Token ID3706
Feature activation-4.55e-02

Pos logit contributions
 envelop+0.048
Offic+0.042
 Amen+0.041
ons+0.039
Mic+0.036
Neg logit contributions
tub-0.080
gan-0.077
 Moe-0.071
ider-0.070
ahl-0.069
Token Paul
Token ID3362
Feature activation-3.71e-02

Pos logit contributions
 envelop+0.095
Offic+0.088
 Amen+0.085
ons+0.082
Mic+0.080
Neg logit contributions
tub-0.081
gan-0.075
ider-0.069
 Moe-0.066
ahl-0.065
Token.
Token ID13
Feature activation+1.24e-02

Pos logit contributions
 envelop+0.077
 Amen+0.069
Offic+0.069
ons+0.067
oc+0.064
Neg logit contributions
tub-0.067
gan-0.063
ider-0.057
 Moe-0.056
ahl-0.055
Token
Token ID198
Feature activation-1.26e-02

Pos logit contributions
tub+0.015
gan+0.013
 Moe+0.012
ahl+0.010
 Mil+0.010
Neg logit contributions
 envelop-0.032
Offic-0.029
 Amen-0.028
ons-0.028
Mic-0.028
Token a
Token ID257
Feature activation-5.05e-02

Pos logit contributions
 Amen+0.103
Offic+0.103
 envelop+0.100
Mic+0.092
oc+0.092
Neg logit contributions
tub-0.083
gan-0.078
 Moe-0.072
ses-0.069
 Mil-0.067
Token stuffed
Token ID22259
Feature activation-1.31e-01

Pos logit contributions
 envelop+0.093
 Amen+0.092
Offic+0.091
ons+0.085
Mic+0.080
Neg logit contributions
tub-0.095
gan-0.084
ses-0.078
ider-0.077
 Moe-0.077
Token dog
Token ID3290
Feature activation-5.88e-02

Pos logit contributions
 Amen+0.269
 envelop+0.262
Offic+0.258
ons+0.244
Mic+0.243
Neg logit contributions
tub-0.201
gan-0.179
ses-0.178
sed-0.168
 Moe-0.168
Token and
Token ID290
Feature activation-7.85e-02

Pos logit contributions
 envelop+0.121
Offic+0.113
 Amen+0.110
ons+0.106
Mic+0.101
Neg logit contributions
tub-0.105
gan-0.099
 Moe-0.088
ider-0.088
ahl-0.085
Token a
Token ID257
Feature activation-4.90e-02

Pos logit contributions
 envelop+0.162
 Amen+0.155
Offic+0.154
ons+0.145
Mic+0.139
Neg logit contributions
tub-0.133
gan-0.127
 Moe-0.112
ider-0.109
ses-0.108
Token stuffed
Token ID22259
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.091
Offic+0.086
 Amen+0.085
ons+0.081
Mic+0.076
Neg logit contributions
tub-0.095
gan-0.088
ider-0.080
 Moe-0.079
ses-0.078
Token cat
Token ID3797
Feature activation-5.05e-02

Pos logit contributions
 envelop+0.260
 Amen+0.251
Offic+0.246
ons+0.242
Mic+0.229
Neg logit contributions
tub-0.209
gan-0.192
ses-0.178
 Moe-0.177
sed-0.170
Token as
Token ID355
Feature activation+4.03e-03

Pos logit contributions
 envelop+0.111
Offic+0.105
 Amen+0.103
ons+0.099
Mic+0.095
Neg logit contributions
tub-0.082
gan-0.077
 Moe-0.067
ider-0.067
ahl-0.064
Token the
Token ID262
Feature activation+4.63e-02

Pos logit contributions
tub+0.007
gan+0.007
ider+0.006
 Moe+0.006
ahl+0.006
Neg logit contributions
 envelop-0.008
Offic-0.008
 Amen-0.008
ons-0.007
Mic-0.007
Token witnesses
Token ID11432
Feature activation-4.26e-02

Pos logit contributions
tub+0.082
gan+0.077
ider+0.070
 Moe+0.069
ses+0.067
Neg logit contributions
 envelop-0.094
Offic-0.088
 Amen-0.087
ons-0.084
Mic-0.079
Token.
Token ID13
Feature activation+8.84e-03

Pos logit contributions
 envelop+0.091
Offic+0.085
 Amen+0.083
ons+0.080
Mic+0.076
Neg logit contributions
tub-0.073
gan-0.069
 Moe-0.060
ider-0.060
ahl-0.058
Token
Token ID198
Feature activation-1.80e-02

Pos logit contributions
 envelop+0.045
Offic+0.042
 Amen+0.042
ons+0.040
Mic+0.038
Neg logit contributions
tub-0.035
gan-0.033
 Moe-0.029
ider-0.029
ahl-0.028
TokenThe
Token ID464
Feature activation+1.54e-02

Pos logit contributions
 envelop+0.033
 Amen+0.029
Offic+0.029
ons+0.028
Mic+0.025
Neg logit contributions
tub-0.036
gan-0.035
 Moe-0.032
ider-0.032
 Mil-0.030
Token kid
Token ID5141
Feature activation-4.98e-03

Pos logit contributions
tub+0.030
gan+0.029
 Moe+0.026
ider+0.026
ahl+0.025
Neg logit contributions
 envelop-0.029
Offic-0.027
 Amen-0.026
ons-0.024
Mic-0.024
Token had
Token ID550
Feature activation-1.76e-02

Pos logit contributions
 envelop+0.010
Offic+0.009
 Amen+0.009
ons+0.009
oc+0.008
Neg logit contributions
tub-0.009
gan-0.009
 Moe-0.008
ider-0.008
ahl-0.008
Token a
Token ID257
Feature activation-5.16e-02

Pos logit contributions
 envelop+0.038
 Amen+0.038
Offic+0.037
ons+0.035
Mic+0.035
Neg logit contributions
tub-0.027
gan-0.026
 Moe-0.023
 Mil-0.022
ahl-0.022
Token grand
Token ID4490
Feature activation-1.27e-01

Pos logit contributions
 envelop+0.099
Offic+0.093
ons+0.090
 Amen+0.090
Mic+0.082
Neg logit contributions
tub-0.098
gan-0.089
 Moe-0.082
ses-0.081
ider-0.080
Token old
Token ID1468
Feature activation-3.86e-02

Pos logit contributions
 envelop+0.189
Offic+0.184
Mic+0.172
 Amen+0.169
ons+0.157
Neg logit contributions
tub-0.270
gan-0.260
ses-0.251
sed-0.247
 Moe-0.244
Token time
Token ID640
Feature activation+2.92e-02

Pos logit contributions
 envelop+0.067
Offic+0.060
 Amen+0.059
ons+0.056
oc+0.052
Neg logit contributions
tub-0.082
gan-0.080
 Moe-0.072
ider-0.071
ahl-0.070
Token.
Token ID13
Feature activation+8.08e-03

Pos logit contributions
tub+0.055
gan+0.052
ider+0.047
 Moe+0.047
ahl+0.044
Neg logit contributions
 envelop-0.057
Offic-0.053
 Amen-0.052
ons-0.049
Mic-0.048
Token He
Token ID679
Feature activation-3.48e-03

Pos logit contributions
tub+0.013
gan+0.013
ider+0.013
ahl+0.011
arling+0.011
Neg logit contributions
 envelop-0.017
Offic-0.016
 Amen-0.015
oc-0.015
 parcels-0.014
Token laughed
Token ID13818
Feature activation-1.85e-02

Pos logit contributions
 envelop+0.007
Offic+0.007
 Amen+0.007
ons+0.006
Mic+0.006
Neg logit contributions
tub-0.006
gan-0.006
 Moe-0.005
ider-0.005
ahl-0.005
Token<|endoftext|>
Token ID50256
Feature activation+2.78e-02

Pos logit contributions
tub+0.031
gan+0.029
 Moe+0.025
ider+0.025
ahl+0.024
Neg logit contributions
 envelop-0.047
Offic-0.044
 Amen-0.043
ons-0.042
Mic-0.040
TokenOnce
Token ID7454
Feature activation+6.83e-02

Pos logit contributions
gan+0.050
tub+0.045
ider+0.045
ses+0.045
sed+0.044
Neg logit contributions
 envelop-0.055
ons-0.053
 Amen-0.052
Offic-0.052
oc-0.048
Token upon
Token ID2402
Feature activation+5.35e-02

Pos logit contributions
ider+0.105
tub+0.102
gan+0.097
ses+0.094
sed+0.088
Neg logit contributions
ons-0.145
 envelop-0.143
 Amen-0.130
pa-0.127
Offic-0.124
Token a
Token ID257
Feature activation-1.39e-02

Pos logit contributions
tub+0.082
gan+0.077
 Moe+0.067
ider+0.066
ahl+0.063
Neg logit contributions
 envelop-0.124
Offic-0.116
 Amen-0.113
ons-0.110
Mic-0.105
Token time
Token ID640
Feature activation+8.15e-02

Pos logit contributions
 envelop+0.029
Offic+0.027
 Amen+0.026
ons+0.025
Mic+0.023
Neg logit contributions
tub-0.025
gan-0.024
 Moe-0.021
ider-0.021
ahl-0.020
Token there
Token ID612
Feature activation+8.93e-02

Pos logit contributions
tub+0.185
gan+0.173
ipples+0.171
ider+0.168
ses+0.166
Neg logit contributions
ons-0.107
 envelop-0.095
 Amen-0.093
oc-0.087
 lending-0.084
Token was
Token ID373
Feature activation+3.03e-02

Pos logit contributions
ider+0.213
tub+0.187
gan+0.184
ses+0.181
arling+0.175
Neg logit contributions
ons-0.130
oc-0.124
 envelop-0.121
 read-0.113
 Amen-0.110
Token a
Token ID257
Feature activation-3.42e-02

Pos logit contributions
tub+0.051
gan+0.048
ider+0.042
 Moe+0.042
ahl+0.040
Neg logit contributions
 envelop-0.066
Offic-0.061
 Amen-0.060
ons-0.058
Mic-0.056
Token yellow
Token ID7872
Feature activation-3.42e-02

Pos logit contributions
 envelop+0.079
Offic+0.074
 Amen+0.072
ons+0.070
Mic+0.067
Neg logit contributions
tub-0.053
gan-0.049
 Moe-0.043
ider-0.043
ahl-0.041
Token bear
Token ID6842
Feature activation-2.58e-02

Pos logit contributions
 envelop+0.065
 Amen+0.059
Offic+0.058
ons+0.055
oc+0.053
Neg logit contributions
tub-0.065
gan-0.063
ider-0.057
 Moe-0.057
ahl-0.056
Token who
Token ID508
Feature activation-8.26e-03

Pos logit contributions
 envelop+0.038
Offic+0.034
 Amen+0.032
ons+0.030
Mic+0.029
Neg logit contributions
tub-0.062
gan-0.059
ider-0.054
 Moe-0.054
 Mil-0.052
Token was
Token ID373
Feature activation+1.50e-02

Pos logit contributions
 envelop+0.114
Offic+0.105
 Amen+0.103
ons+0.098
Mic+0.094
Neg logit contributions
tub-0.116
gan-0.111
 Moe-0.099
ider-0.098
ahl-0.095
Token very
Token ID845
Feature activation-2.24e-02

Pos logit contributions
tub+0.027
gan+0.026
 Moe+0.023
ider+0.023
 Mil+0.022
Neg logit contributions
 envelop-0.031
Offic-0.029
 Amen-0.028
ons-0.027
Mic-0.026
Token careful
Token ID8161
Feature activation+1.64e-02

Pos logit contributions
 envelop+0.041
Offic+0.037
 Amen+0.037
ons+0.035
Mic+0.033
Neg logit contributions
tub-0.045
gan-0.043
 Moe-0.039
ider-0.039
ahl-0.038
Token to
Token ID284
Feature activation-3.68e-02

Pos logit contributions
tub+0.030
gan+0.028
ahl+0.025
 Moe+0.025
ider+0.025
Neg logit contributions
 envelop-0.033
 Amen-0.030
ons-0.029
Offic-0.029
 lending-0.026
Token do
Token ID466
Feature activation+5.07e-03

Pos logit contributions
 envelop+0.065
Offic+0.059
 Amen+0.058
ons+0.055
Mic+0.052
Neg logit contributions
tub-0.077
gan-0.073
ider-0.067
 Moe-0.066
ahl-0.065
Token it
Token ID340
Feature activation+8.26e-02

Pos logit contributions
tub+0.011
gan+0.010
ider+0.009
 Moe+0.009
ses+0.009
Neg logit contributions
 envelop-0.009
ons-0.008
Offic-0.008
 Amen-0.008
oc-0.007
Token just
Token ID655
Feature activation+3.71e-02

Pos logit contributions
tub+0.166
gan+0.164
ses+0.155
sed+0.150
ahl+0.144
Neg logit contributions
 envelop-0.142
ons-0.129
Offic-0.118
 read-0.118
 lending-0.115
Token like
Token ID588
Feature activation+3.85e-03

Pos logit contributions
tub+0.072
gan+0.064
ider+0.058
ahl+0.057
 Moe+0.057
Neg logit contributions
 envelop-0.073
Offic-0.063
ons-0.063
 read-0.062
 Amen-0.060
Token her
Token ID607
Feature activation+3.25e-02

Pos logit contributions
tub+0.007
gan+0.007
 Moe+0.006
ider+0.006
ahl+0.006
Neg logit contributions
 envelop-0.007
Offic-0.007
 Amen-0.007
ons-0.006
Mic-0.006
Token mummy
Token ID38567
Feature activation-4.22e-02

Pos logit contributions
tub+0.060
gan+0.058
ahl+0.052
ider+0.051
 Moe+0.051
Neg logit contributions
 envelop-0.064
 Amen-0.059
ons-0.055
Offic-0.054
oc-0.052
Token and
Token ID290
Feature activation-7.15e-02

Pos logit contributions
 envelop+0.085
Offic+0.078
 Amen+0.077
ons+0.074
Mic+0.070
Neg logit contributions
tub-0.077
gan-0.073
 Moe-0.065
ider-0.065
ahl-0.063
Token day
Token ID1110
Feature activation+6.09e-02

Pos logit contributions
 envelop+0.004
 Amen+0.003
ons+0.003
Offic+0.003
oc+0.003
Neg logit contributions
tub-0.005
gan-0.005
ses-0.005
ider-0.005
sed-0.005
Token,
Token ID11
Feature activation+4.01e-02

Pos logit contributions
tub+0.116
gan+0.107
ses+0.106
ider+0.101
 fault+0.099
Neg logit contributions
 envelop-0.120
ons-0.106
Offic-0.103
 Amen-0.095
oc-0.094
Token the
Token ID262
Feature activation+4.31e-02

Pos logit contributions
tub+0.075
gan+0.068
ider+0.066
ses+0.065
ahl+0.063
Neg logit contributions
 envelop-0.079
Offic-0.072
ons-0.067
 Amen-0.066
Mic-0.064
Token boy
Token ID2933
Feature activation+1.43e-02

Pos logit contributions
tub+0.079
gan+0.079
ahl+0.070
 Moe+0.069
ider+0.068
Neg logit contributions
 envelop-0.083
Offic-0.075
 Amen-0.074
ons-0.070
Mic-0.069
Token walked
Token ID6807
Feature activation+2.67e-03

Pos logit contributions
tub+0.029
gan+0.028
ider+0.026
 Moe+0.025
ahl+0.025
Neg logit contributions
 envelop-0.026
Offic-0.024
ons-0.022
 Amen-0.022
oc-0.022
Token down
Token ID866
Feature activation+6.91e-02

Pos logit contributions
tub+0.005
gan+0.005
ses+0.005
ider+0.004
sed+0.004
Neg logit contributions
 envelop-0.005
Offic-0.004
 Amen-0.004
ons-0.004
 saddle-0.004
Token the
Token ID262
Feature activation+7.09e-02

Pos logit contributions
gan+0.148
tub+0.147
ider+0.135
ses+0.128
ahl+0.126
Neg logit contributions
 envelop-0.118
Offic-0.107
 Amen-0.098
ons-0.096
 read-0.089
Token street
Token ID4675
Feature activation+4.37e-02

Pos logit contributions
gan+0.154
tub+0.147
ider+0.138
 Moe+0.135
ahl+0.135
Neg logit contributions
 envelop-0.115
Offic-0.107
Mic-0.097
 Amen-0.097
 read-0.093
Token and
Token ID290
Feature activation-5.20e-02

Pos logit contributions
tub+0.085
gan+0.085
ahl+0.079
ses+0.077
ider+0.076
Neg logit contributions
 envelop-0.078
Offic-0.070
ons-0.065
 Amen-0.063
 read-0.062
Token stopped
Token ID5025
Feature activation+4.59e-02

Pos logit contributions
 envelop+0.082
Offic+0.077
 read+0.069
ons+0.066
 Amen+0.065
Neg logit contributions
tub-0.113
gan-0.107
ider-0.106
 fault-0.102
ses-0.101
Token in
Token ID287
Feature activation+2.37e-02

Pos logit contributions
tub+0.095
gan+0.093
ider+0.084
ahl+0.082
 Moe+0.081
Neg logit contributions
 envelop-0.081
Offic-0.071
 Amen-0.071
ons-0.068
 lending-0.064
Token her
Token ID607
Feature activation+3.36e-02

Pos logit contributions
 Amen+0.040
Offic+0.039
oc+0.038
 envelop+0.038
Mic+0.037
Neg logit contributions
tub-0.028
 Moe-0.025
gan-0.025
 Mil-0.023
ahl-0.021
Token pocket
Token ID10000
Feature activation+2.29e-03

Pos logit contributions
tub+0.061
gan+0.057
 Moe+0.051
ider+0.051
ahl+0.049
Neg logit contributions
 envelop-0.069
Offic-0.065
 Amen-0.064
ons-0.060
Mic-0.058
Token.
Token ID13
Feature activation+7.92e-03

Pos logit contributions
tub+0.004
gan+0.004
ider+0.003
 Moe+0.003
ahl+0.003
Neg logit contributions
 envelop-0.005
Offic-0.005
 Amen-0.005
ons-0.004
Mic-0.004
Token She
Token ID1375
Feature activation-4.74e-03

Pos logit contributions
tub+0.012
gan+0.012
ider+0.011
ses+0.011
ahl+0.011
Neg logit contributions
 envelop-0.017
 Amen-0.016
Offic-0.015
ons-0.015
 parcels-0.014
Token put
Token ID1234
Feature activation-4.32e-02

Pos logit contributions
 envelop+0.010
 Amen+0.010
Offic+0.010
Mic+0.009
ons+0.009
Neg logit contributions
tub-0.008
gan-0.007
 Moe-0.006
 Mil-0.006
ider-0.006
Token it
Token ID340
Feature activation+6.46e-02

Pos logit contributions
Offic+0.081
 envelop+0.081
 Amen+0.079
Mic+0.072
oc+0.070
Neg logit contributions
tub-0.081
gan-0.078
 Moe-0.072
 Mil-0.068
ahl-0.068
Token in
Token ID287
Feature activation-1.04e-03

Pos logit contributions
tub+0.118
gan+0.112
 Moe+0.102
ider+0.100
ses+0.100
Neg logit contributions
 envelop-0.129
Offic-0.114
ons-0.113
 Amen-0.111
Mic-0.104
Token a
Token ID257
Feature activation-3.65e-02

Pos logit contributions
Offic+0.002
 Amen+0.002
oc+0.002
ons+0.002
Mic+0.002
Neg logit contributions
tub-0.002
 Moe-0.001
gan-0.001
 Mil-0.001
ahl-0.001
Token special
Token ID2041
Feature activation-3.63e-02

Pos logit contributions
Offic+0.080
 Amen+0.080
 envelop+0.078
ons+0.076
Mic+0.072
Neg logit contributions
tub-0.056
gan-0.045
 Moe-0.044
ahl-0.041
ider-0.040
Token box
Token ID3091
Feature activation-3.44e-03

Pos logit contributions
Offic+0.077
 envelop+0.077
 Amen+0.074
ons+0.072
Mic+0.072
Neg logit contributions
tub-0.060
gan-0.051
 Moe-0.048
ider-0.045
ahl-0.045
Token and
Token ID290
Feature activation-6.63e-02

Pos logit contributions
 envelop+0.007
Offic+0.006
 Amen+0.006
ons+0.006
Mic+0.006
Neg logit contributions
tub-0.006
gan-0.006
 Moe-0.005
ider-0.005
ahl-0.005
Token y
Token ID331
Feature activation-4.72e-02

Pos logit contributions
 envelop+0.028
Offic+0.025
 Amen+0.024
ons+0.023
Mic+0.022
Neg logit contributions
tub-0.036
gan-0.035
 Moe-0.032
ider-0.031
ahl-0.031
Tokenummy
Token ID13513
Feature activation-2.57e-02

Pos logit contributions
 envelop+0.102
ons+0.097
Offic+0.093
 Amen+0.092
Mic+0.091
Neg logit contributions
tub-0.074
gan-0.067
ses-0.061
ider-0.058
ahl-0.058
Token,
Token ID11
Feature activation+1.24e-02

Pos logit contributions
 envelop+0.044
Offic+0.040
 Amen+0.038
ons+0.038
oc+0.035
Neg logit contributions
tub-0.055
gan-0.052
ider-0.047
ahl-0.046
 Moe-0.046
Token she
Token ID673
Feature activation+1.34e-02

Pos logit contributions
tub+0.021
gan+0.020
 Moe+0.017
ider+0.016
 Mil+0.016
Neg logit contributions
 envelop-0.027
Offic-0.026
 Amen-0.025
ons-0.023
Mic-0.023
Token ate
Token ID15063
Feature activation-2.23e-02

Pos logit contributions
tub+0.024
gan+0.023
 Moe+0.020
ider+0.020
 Mil+0.019
Neg logit contributions
 envelop-0.028
Offic-0.026
 Amen-0.025
ons-0.024
Mic-0.023
Token it
Token ID340
Feature activation+7.57e-02

Pos logit contributions
 envelop+0.040
Offic+0.038
 Amen+0.037
ons+0.034
Mic+0.033
Neg logit contributions
tub-0.045
gan-0.044
 Moe-0.040
ider-0.038
ahl-0.038
Token all
Token ID477
Feature activation-9.73e-03

Pos logit contributions
tub+0.156
gan+0.148
ses+0.136
ider+0.136
 Moe+0.133
Neg logit contributions
 envelop-0.132
ons-0.116
Offic-0.115
 Amen-0.113
oc-0.104
Token up
Token ID510
Feature activation+4.96e-02

Pos logit contributions
 envelop+0.016
ons+0.016
 Amen+0.015
 read+0.014
Offic+0.013
Neg logit contributions
tub-0.020
ses-0.019
ider-0.019
sed-0.018
gan-0.018
Token.
Token ID13
Feature activation+1.79e-02

Pos logit contributions
tub+0.092
gan+0.086
ider+0.078
 Moe+0.077
ahl+0.075
Neg logit contributions
 envelop-0.100
Offic-0.091
 Amen-0.089
ons-0.087
Mic-0.081
Token
Token ID198
Feature activation-1.67e-02

Pos logit contributions
tub+0.033
gan+0.030
ider+0.029
 Moe+0.027
ses+0.026
Neg logit contributions
 envelop-0.036
Offic-0.034
 Amen-0.034
ons-0.032
Mic-0.031
Token
Token ID198
Feature activation-1.38e-02

Pos logit contributions
 envelop+0.042
Offic+0.038
 Amen+0.037
ons+0.035
 parcels+0.035
Neg logit contributions
tub-0.023
gan-0.021
 Moe-0.019
ahl-0.018
 Mil-0.017
Token She
Token ID1375
Feature activation-9.33e-03

Pos logit contributions
tub+0.025
ider+0.023
gan+0.022
ahl+0.020
flies+0.020
Neg logit contributions
 envelop-0.022
Offic-0.021
 Amen-0.020
Mic-0.018
ons-0.018
Token does
Token ID857
Feature activation+4.62e-03

Pos logit contributions
 envelop+0.019
Offic+0.018
 Amen+0.017
ons+0.017
Mic+0.016
Neg logit contributions
tub-0.017
gan-0.016
 Moe-0.014
ider-0.014
ahl-0.014
Token not
Token ID407
Feature activation-1.20e-02

Pos logit contributions
tub+0.005
gan+0.005
 Moe+0.004
ider+0.004
 Mil+0.004
Neg logit contributions
 envelop-0.012
Offic-0.011
 Amen-0.011
ons-0.011
Mic-0.011
Token know
Token ID760
Feature activation-2.84e-02

Pos logit contributions
 envelop+0.020
Offic+0.019
 Amen+0.018
ons+0.018
Mic+0.016
Neg logit contributions
tub-0.026
gan-0.025
 Moe-0.022
ider-0.022
ahl-0.022
Token what
Token ID644
Feature activation+1.83e-02

Pos logit contributions
 envelop+0.042
ons+0.037
Offic+0.036
 Amen+0.033
 read+0.030
Neg logit contributions
tub-0.070
gan-0.061
ider-0.060
ahl-0.059
ses-0.058
Token she
Token ID673
Feature activation+4.18e-02

Pos logit contributions
tub+0.036
gan+0.034
ider+0.031
 Moe+0.030
ahl+0.030
Neg logit contributions
 envelop-0.036
Offic-0.032
ons-0.031
 Amen-0.031
 parcels-0.028
Token did
Token ID750
Feature activation+3.47e-02

Pos logit contributions
tub+0.062
gan+0.058
 Moe+0.050
ider+0.050
ahl+0.047
Neg logit contributions
 envelop-0.099
Offic-0.094
 Amen-0.092
ons-0.089
oc-0.085
Token wrong
Token ID2642
Feature activation+8.14e-02

Pos logit contributions
tub+0.055
gan+0.052
 Moe+0.045
ider+0.045
ahl+0.043
Neg logit contributions
 envelop-0.078
Offic-0.073
 Amen-0.072
ons-0.069
Mic-0.066
Token.
Token ID13
Feature activation+1.63e-02

Pos logit contributions
tub+0.155
gan+0.148
ahl+0.133
 Moe+0.132
ider+0.130
Neg logit contributions
 envelop-0.159
 Amen-0.137
Offic-0.137
ons-0.135
oc-0.127
Token She
Token ID1375
Feature activation-8.23e-03

Pos logit contributions
tub+0.034
ider+0.029
gan+0.028
flies+0.027
ahl+0.026
Neg logit contributions
 envelop-0.028
Offic-0.028
 Amen-0.027
Mic-0.023
ons-0.023
Token pulls
Token ID16194
Feature activation-2.30e-02

Pos logit contributions
 envelop+0.017
Offic+0.016
 Amen+0.015
ons+0.015
oc+0.014
Neg logit contributions
tub-0.015
gan-0.014
ider-0.013
 Moe-0.013
ahl-0.012
Token every
Token ID790
Feature activation-3.92e-03

Pos logit contributions
 envelop+0.014
Offic+0.013
 Amen+0.013
ons+0.012
Mic+0.012
Neg logit contributions
tub-0.013
gan-0.013
 Moe-0.011
ider-0.011
ahl-0.011
Token hot
Token ID3024
Feature activation-4.95e-03

Pos logit contributions
 envelop+0.009
Offic+0.009
 Amen+0.009
Mic+0.008
ons+0.008
Neg logit contributions
tub-0.006
gan-0.005
ider-0.004
 Moe-0.004
ses-0.004
Token day
Token ID1110
Feature activation+4.70e-02

Pos logit contributions
 envelop+0.010
Offic+0.009
 Amen+0.009
ons+0.008
Mic+0.008
Neg logit contributions
tub-0.009
gan-0.009
 Moe-0.008
ider-0.008
ahl-0.008
Token and
Token ID290
Feature activation-6.01e-02

Pos logit contributions
tub+0.076
gan+0.072
 Moe+0.063
ider+0.063
ahl+0.060
Neg logit contributions
 envelop-0.104
Offic-0.098
 Amen-0.096
ons-0.092
Mic-0.088
Token had
Token ID550
Feature activation-1.35e-02

Pos logit contributions
 envelop+0.122
Offic+0.115
 Amen+0.115
ons+0.107
Mic+0.105
Neg logit contributions
tub-0.107
gan-0.101
 Moe-0.091
 Mil-0.086
ahl-0.085
Token the
Token ID262
Feature activation+2.47e-02

Pos logit contributions
 envelop+0.025
 Amen+0.025
Offic+0.025
oc+0.022
ons+0.022
Neg logit contributions
tub-0.026
gan-0.025
 Moe-0.022
 Mil-0.021
ahl-0.021
Token best
Token ID1266
Feature activation-7.24e-03

Pos logit contributions
tub+0.030
gan+0.028
ider+0.024
 Moe+0.024
ahl+0.022
Neg logit contributions
 envelop-0.066
Offic-0.060
 Amen-0.060
ons-0.059
Mic-0.056
Token time
Token ID640
Feature activation+4.03e-02

Pos logit contributions
 envelop+0.015
Offic+0.015
 Amen+0.014
Mic+0.013
ons+0.012
Neg logit contributions
tub-0.013
gan-0.012
 Moe-0.011
ider-0.011
ses-0.010
Token.
Token ID13
Feature activation+9.50e-03

Pos logit contributions
tub+0.064
gan+0.061
 Moe+0.053
ider+0.053
 Mil+0.050
Neg logit contributions
 envelop-0.089
 Amen-0.085
Offic-0.085
Mic-0.078
ons-0.077
Token<|endoftext|>
Token ID50256
Feature activation+2.46e-02

Pos logit contributions
tub+0.015
gan+0.014
ider+0.013
 Moe+0.012
ahl+0.012
Neg logit contributions
 envelop-0.022
Offic-0.020
 Amen-0.020
ons-0.019
Mic-0.018
TokenOnce
Token ID7454
Feature activation+6.92e-02

Pos logit contributions
tub+0.044
gan+0.044
ider+0.040
ses+0.038
sed+0.037
Neg logit contributions
 envelop-0.049
Offic-0.045
 Amen-0.045
ons-0.044
oc-0.041
Token happy
Token ID3772
Feature activation+4.22e-04

Pos logit contributions
 envelop+0.117
Offic+0.105
 Amen+0.104
ons+0.104
Mic+0.096
Neg logit contributions
tub-0.112
gan-0.110
 Moe-0.098
 Mil-0.093
ider-0.093
Token to
Token ID284
Feature activation-4.55e-02

Pos logit contributions
tub+0.001
gan+0.001
ahl+0.001
ider+0.001
ses+0.001
Neg logit contributions
 envelop-0.001
ons-0.001
Offic-0.001
 Amen-0.001
 lending-0.001
Token hear
Token ID3285
Feature activation-1.11e-03

Pos logit contributions
 envelop+0.086
Offic+0.078
ons+0.077
 Amen+0.075
Mic+0.070
Neg logit contributions
tub-0.084
gan-0.083
 Moe-0.075
ahl-0.072
 Mil-0.072
Token from
Token ID422
Feature activation-2.74e-02

Pos logit contributions
 envelop+0.002
Offic+0.002
 Amen+0.002
Mic+0.002
ons+0.002
Neg logit contributions
tub-0.002
gan-0.002
 Moe-0.001
 Mil-0.001
der-0.001
Token him
Token ID683
Feature activation+2.84e-02

Pos logit contributions
 envelop+0.052
ons+0.050
Offic+0.049
 Amen+0.048
oc+0.046
Neg logit contributions
gan-0.046
tub-0.046
 Moe-0.044
ses-0.041
 Mil-0.039
Token.
Token ID13
Feature activation+1.02e-02

Pos logit contributions
tub+0.041
gan+0.038
ider+0.033
 Moe+0.033
ahl+0.031
Neg logit contributions
 envelop-0.067
Offic-0.064
 Amen-0.063
ons-0.061
Mic-0.059
Token Lily
Token ID20037
Feature activation+1.42e-02

Pos logit contributions
tub+0.015
gan+0.014
ider+0.013
 Moe+0.012
ses+0.012
Neg logit contributions
 envelop-0.024
Offic-0.022
 Amen-0.022
ons-0.020
Mic-0.020
Token smiled
Token ID13541
Feature activation-1.98e-02

Pos logit contributions
tub+0.027
gan+0.026
 Moe+0.023
ider+0.023
ahl+0.022
Neg logit contributions
 envelop-0.028
Offic-0.025
 Amen-0.025
ons-0.024
Mic-0.023
Token,
Token ID11
Feature activation-5.86e-03

Pos logit contributions
 envelop+0.035
Offic+0.032
 Amen+0.032
ons+0.029
Mic+0.028
Neg logit contributions
tub-0.041
gan-0.039
ider-0.035
 Moe-0.035
ahl-0.035
Token knowing
Token ID6970
Feature activation-1.95e-02

Pos logit contributions
 envelop+0.013
ons+0.012
 Amen+0.012
Offic+0.011
igh+0.011
Neg logit contributions
gan-0.009
tub-0.009
der-0.007
sed-0.007
 Moe-0.007
Token that
Token ID326
Feature activation+3.25e-03

Pos logit contributions
 envelop+0.049
 Amen+0.046
Offic+0.046
ons+0.044
oc+0.043
Neg logit contributions
tub-0.023
gan-0.023
 Moe-0.021
 Mil-0.018
der-0.018
Token like
Token ID588
Feature activation-2.41e-02

Pos logit contributions
 envelop+0.036
Offic+0.033
 Amen+0.033
ons+0.031
Mic+0.029
Neg logit contributions
tub-0.033
gan-0.031
ider-0.028
 Moe-0.028
ahl-0.027
Token to
Token ID284
Feature activation-3.12e-02

Pos logit contributions
 envelop+0.058
Offic+0.056
 Amen+0.056
oc+0.053
ons+0.052
Neg logit contributions
tub-0.034
gan-0.030
 Moe-0.026
 Mil-0.025
ses-0.025
Token play
Token ID711
Feature activation+6.46e-03

Pos logit contributions
 envelop+0.052
Offic+0.049
 Amen+0.046
ons+0.043
Mic+0.042
Neg logit contributions
tub-0.068
gan-0.064
 Moe-0.059
ider-0.057
 Mil-0.056
Token in
Token ID287
Feature activation+2.21e-02

Pos logit contributions
tub+0.012
gan+0.011
 Moe+0.010
ider+0.010
ahl+0.010
Neg logit contributions
 envelop-0.013
Offic-0.012
 Amen-0.012
ons-0.011
Mic-0.011
Token the
Token ID262
Feature activation+4.13e-02

Pos logit contributions
tub+0.037
gan+0.032
 Moe+0.031
 Mil+0.029
ahl+0.028
Neg logit contributions
 envelop-0.047
Offic-0.046
 Amen-0.045
Mic-0.042
ons-0.042
Token park
Token ID3952
Feature activation+4.80e-02

Pos logit contributions
gan+0.077
tub+0.076
ider+0.070
 Moe+0.066
ses+0.065
Neg logit contributions
 envelop-0.082
Offic-0.074
 Amen-0.071
ons-0.070
Mic-0.067
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
tub+0.085
gan+0.081
 Moe+0.073
ahl+0.073
ider+0.071
Neg logit contributions
 envelop-0.097
Offic-0.088
 Amen-0.086
ons-0.085
oc-0.081
Token One
Token ID1881
Feature activation+1.55e-02

Pos logit contributions
tub+0.036
ider+0.033
gan+0.033
ahl+0.031
ter+0.030
Neg logit contributions
 envelop-0.029
 Amen-0.026
Offic-0.025
oc-0.023
ons-0.022
Token day
Token ID1110
Feature activation+6.81e-02

Pos logit contributions
ider+0.036
tub+0.035
gan+0.034
der+0.033
sed+0.033
Neg logit contributions
 envelop-0.019
 read-0.018
ons-0.018
Offic-0.017
oc-0.017
Token,
Token ID11
Feature activation+4.81e-02

Pos logit contributions
tub+0.144
ses+0.131
ahl+0.123
ider+0.121
sing+0.120
Neg logit contributions
 envelop-0.109
ons-0.102
 read-0.087
 Amen-0.087
 exchange-0.086
Token they
Token ID484
Feature activation+1.23e-02

Pos logit contributions
tub+0.093
ider+0.085
ses+0.083
ahl+0.081
sed+0.076
Neg logit contributions
 envelop-0.086
ons-0.072
Offic-0.070
 Amen-0.070
 read-0.068
Token
Token ID198
Feature activation-2.19e-02

Pos logit contributions
tub+0.020
gan+0.019
ider+0.017
ahl+0.016
 Moe+0.016
Neg logit contributions
 envelop-0.024
Offic-0.023
ons-0.022
 Amen-0.021
oc-0.020
Token
Token ID198
Feature activation-1.81e-02

Pos logit contributions
 envelop+0.046
Offic+0.044
 Amen+0.042
ons+0.041
Mic+0.039
Neg logit contributions
tub-0.038
gan-0.036
ider-0.032
 Moe-0.031
ses-0.030
TokenTom
Token ID13787
Feature activation-3.56e-03

Pos logit contributions
 envelop+0.034
 Amen+0.031
Offic+0.029
ons+0.028
 parcels+0.027
Neg logit contributions
tub-0.034
gan-0.034
 Moe-0.031
ider-0.030
ahl-0.030
Token and
Token ID290
Feature activation-7.02e-02

Pos logit contributions
 envelop+0.007
Offic+0.007
ons+0.007
 Amen+0.007
oc+0.006
Neg logit contributions
tub-0.006
gan-0.006
ider-0.005
 Moe-0.005
ahl-0.005
Token Mia
Token ID32189
Feature activation-4.62e-03

Pos logit contributions
 envelop+0.192
Offic+0.184
 Amen+0.176
ons+0.175
oc+0.166
Neg logit contributions
tub-0.084
gan-0.069
ider-0.063
 Moe-0.055
ahl-0.051
Token get
Token ID651
Feature activation+3.72e-03

Pos logit contributions
 envelop+0.009
Offic+0.008
 Amen+0.008
ons+0.008
oc+0.008
Neg logit contributions
tub-0.009
gan-0.008
ider-0.007
 Moe-0.007
ahl-0.007
Token the
Token ID262
Feature activation+4.24e-02

Pos logit contributions
tub+0.006
gan+0.006
 Moe+0.005
 Mil+0.005
ahl+0.004
Neg logit contributions
Offic-0.008
 envelop-0.008
 Amen-0.008
Mic-0.008
ons-0.007
Token je
Token ID11223
Feature activation-3.94e-02

Pos logit contributions
tub+0.069
gan+0.063
ses+0.056
ider+0.055
ahl+0.055
Neg logit contributions
 envelop-0.089
 Amen-0.087
Offic-0.087
ons-0.081
Mic-0.079
Tokenep
Token ID538
Feature activation-1.51e-02

Pos logit contributions
 Amen+0.048
 envelop+0.046
Offic+0.041
Mic+0.038
 lending+0.038
Neg logit contributions
tub-0.095
gan-0.090
ider-0.089
anger-0.086
ahl-0.085
Token back
Token ID736
Feature activation+5.08e-02

Pos logit contributions
 envelop+0.032
Offic+0.030
 Amen+0.029
ons+0.028
oc+0.027
Neg logit contributions
tub-0.026
gan-0.025
 Moe-0.022
ider-0.022
ahl-0.021
Token.
Token ID13
Feature activation+6.94e-03

Pos logit contributions
tub+0.085
gan+0.081
 Moe+0.072
ider+0.071
ahl+0.069
Neg logit contributions
 envelop-0.111
Offic-0.101
 Amen-0.098
ons-0.098
oc-0.091
Token took
Token ID1718
Feature activation-5.60e-02

Pos logit contributions
 envelop+0.094
Offic+0.086
 Amen+0.086
ons+0.081
Mic+0.078
Neg logit contributions
tub-0.088
gan-0.085
 Moe-0.075
ider-0.074
ahl-0.072
Token out
Token ID503
Feature activation+7.79e-03

Pos logit contributions
Offic+0.122
 envelop+0.122
 Amen+0.122
ons+0.112
oc+0.110
Neg logit contributions
tub-0.083
gan-0.081
 Moe-0.074
 Mil-0.068
ahl-0.068
Token two
Token ID734
Feature activation-5.08e-02

Pos logit contributions
tub+0.012
gan+0.012
 Moe+0.011
 Mil+0.010
ahl+0.010
Neg logit contributions
 Amen-0.017
 envelop-0.017
Offic-0.017
oc-0.016
Mic-0.015
Token past
Token ID1613
Feature activation-3.44e-02

Pos logit contributions
 Amen+0.110
Offic+0.106
Mic+0.096
 envelop+0.092
oc+0.091
Neg logit contributions
tub-0.081
gan-0.073
 Moe-0.071
ider-0.066
ses-0.066
Tokenels
Token ID1424
Feature activation-4.13e-02

Pos logit contributions
 envelop+0.049
Offic+0.046
 Amen+0.045
ons+0.040
Mic+0.039
Neg logit contributions
tub-0.082
gan-0.079
 Moe-0.073
ider-0.071
ahl-0.070
Token from
Token ID422
Feature activation-3.69e-02

Pos logit contributions
 envelop+0.080
Offic+0.073
 Amen+0.071
ons+0.070
Mic+0.064
Neg logit contributions
tub-0.079
gan-0.075
ider-0.067
 Moe-0.067
ahl-0.065
Token the
Token ID262
Feature activation+3.37e-02

Pos logit contributions
 envelop+0.073
Offic+0.068
 Amen+0.066
ons+0.063
Mic+0.060
Neg logit contributions
tub-0.070
gan-0.066
ider-0.059
 Moe-0.059
ahl-0.057
Token box
Token ID3091
Feature activation+1.95e-02

Pos logit contributions
tub+0.066
gan+0.063
 Moe+0.057
ider+0.057
ahl+0.054
Neg logit contributions
 envelop-0.066
Offic-0.059
 Amen-0.056
ons-0.055
Mic-0.052
Token.
Token ID13
Feature activation+4.89e-03

Pos logit contributions
tub+0.034
gan+0.032
 Moe+0.028
ider+0.028
ahl+0.027
Neg logit contributions
 envelop-0.041
Offic-0.038
 Amen-0.037
ons-0.036
Mic-0.034
Token One
Token ID1881
Feature activation+3.74e-03

Pos logit contributions
tub+0.008
gan+0.007
ider+0.007
ahl+0.006
 Moe+0.006
Neg logit contributions
 envelop-0.011
Offic-0.010
 Amen-0.010
ons-0.009
Mic-0.009
Token was
Token ID373
Feature activation+1.79e-02

Pos logit contributions
tub+0.007
gan+0.007
 Moe+0.007
ider+0.007
ahl+0.006
Neg logit contributions
 envelop-0.007
ons-0.006
Offic-0.006
 Amen-0.006
oc-0.006
Token big
Token ID1263
Feature activation-6.47e-02

Pos logit contributions
tub+0.060
gan+0.059
ider+0.053
 Moe+0.052
ahl+0.052
Neg logit contributions
 envelop-0.061
Offic-0.053
 Amen-0.052
ons-0.052
oc-0.049
Token claws
Token ID28421
Feature activation-3.93e-03

Pos logit contributions
 envelop+0.132
Offic+0.121
 Amen+0.119
ons+0.114
Mic+0.108
Neg logit contributions
tub-0.117
gan-0.112
ider-0.099
 Moe-0.099
ahl-0.096
Token and
Token ID290
Feature activation-6.19e-02

Pos logit contributions
 envelop+0.007
ons+0.006
Offic+0.005
 parcels+0.005
 saddle+0.005
Neg logit contributions
tub-0.008
gan-0.008
 Moe-0.007
ahl-0.007
 Mil-0.007
Token picked
Token ID6497
Feature activation-1.45e-02

Pos logit contributions
 envelop+0.121
Offic+0.112
 Amen+0.109
ons+0.105
Mic+0.099
Neg logit contributions
tub-0.117
gan-0.111
ider-0.100
 Moe-0.099
ahl-0.096
Token up
Token ID510
Feature activation+2.37e-02

Pos logit contributions
 envelop+0.026
ons+0.022
Offic+0.021
 Amen+0.021
 parcels+0.019
Neg logit contributions
tub-0.031
gan-0.030
ider-0.029
ses-0.026
ahl-0.026
Token a
Token ID257
Feature activation-3.07e-02

Pos logit contributions
tub+0.047
gan+0.045
ider+0.041
 Moe+0.040
ahl+0.039
Neg logit contributions
 envelop-0.045
Offic-0.040
 Amen-0.039
ons-0.038
oc-0.035
Token big
Token ID1263
Feature activation-5.83e-02

Pos logit contributions
 envelop+0.063
Offic+0.059
 Amen+0.057
ons+0.055
Mic+0.052
Neg logit contributions
tub-0.055
gan-0.052
 Moe-0.046
ider-0.046
ahl-0.044
Token fish
Token ID5916
Feature activation-2.51e-02

Pos logit contributions
 envelop+0.126
Offic+0.118
 Amen+0.115
ons+0.111
Mic+0.107
Neg logit contributions
tub-0.098
gan-0.092
 Moe-0.082
ider-0.080
ahl-0.077
Token.
Token ID13
Feature activation+4.01e-03

Pos logit contributions
 envelop+0.053
Offic+0.049
 Amen+0.048
ons+0.047
oc+0.044
Neg logit contributions
tub-0.043
gan-0.041
 Moe-0.036
ider-0.036
ahl-0.035
Token
Token ID220
Feature activation+3.59e-02

Pos logit contributions
tub+0.007
gan+0.006
ider+0.006
ahl+0.006
ses+0.006
Neg logit contributions
 envelop-0.008
Offic-0.008
 Amen-0.008
ons-0.007
oc-0.007
Token
Token ID198
Feature activation-2.15e-02

Pos logit contributions
tub+0.073
ider+0.073
gan+0.071
ses+0.068
sed+0.067
Neg logit contributions
 envelop-0.058
ons-0.055
Offic-0.051
oc-0.048
 Amen-0.048
Token it
Token ID340
Feature activation+7.71e-02

Pos logit contributions
 envelop+0.006
Offic+0.005
 Amen+0.005
ons+0.005
Mic+0.005
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.004
ider-0.004
ahl-0.004
Token must
Token ID1276
Feature activation+5.31e-03

Pos logit contributions
gan+0.148
tub+0.146
ider+0.135
 Moe+0.131
ahl+0.129
Neg logit contributions
 envelop-0.143
ons-0.135
Offic-0.125
 Amen-0.125
oc-0.122
Token have
Token ID423
Feature activation-7.06e-03

Pos logit contributions
tub+0.012
gan+0.011
ider+0.010
 Moe+0.010
ses+0.010
Neg logit contributions
 envelop-0.009
 Amen-0.008
Offic-0.008
ons-0.007
oc-0.007
Token blown
Token ID16318
Feature activation-3.99e-02

Pos logit contributions
 envelop+0.013
Offic+0.011
ons+0.011
 Amen+0.011
 parcels+0.010
Neg logit contributions
tub-0.014
gan-0.014
ider-0.013
 Moe-0.012
ahl-0.012
Token in
Token ID287
Feature activation+2.05e-03

Pos logit contributions
 envelop+0.071
Offic+0.064
 Amen+0.063
ons+0.061
Mic+0.057
Neg logit contributions
tub-0.083
gan-0.079
ider-0.072
 Moe-0.071
ahl-0.069
Token from
Token ID422
Feature activation-3.40e-02

Pos logit contributions
tub+0.003
gan+0.003
 Moe+0.003
 Mil+0.003
ses+0.002
Neg logit contributions
Offic-0.004
 envelop-0.004
 Amen-0.004
ons-0.004
Mic-0.004
Token the
Token ID262
Feature activation+2.26e-02

Pos logit contributions
 envelop+0.068
Offic+0.064
 Amen+0.062
ons+0.059
Mic+0.057
Neg logit contributions
tub-0.062
gan-0.059
 Moe-0.053
ider-0.051
ses-0.050
Token picnic
Token ID35715
Feature activation-4.42e-03

Pos logit contributions
gan+0.040
tub+0.040
ahl+0.035
ider+0.035
 Moe+0.034
Neg logit contributions
 envelop-0.047
Offic-0.041
 Amen-0.041
ons-0.038
 parcels-0.037
Token they
Token ID484
Feature activation+5.82e-03

Pos logit contributions
 envelop+0.009
Offic+0.008
 Amen+0.008
ons+0.008
oc+0.007
Neg logit contributions
tub-0.008
gan-0.008
ider-0.007
 Moe-0.007
ahl-0.007
Token were
Token ID547
Feature activation+5.56e-03

Pos logit contributions
tub+0.009
gan+0.008
 Moe+0.007
ider+0.007
ses+0.007
Neg logit contributions
 envelop-0.013
Offic-0.013
 Amen-0.013
ons-0.012
Mic-0.012
Token planning
Token ID5410
Feature activation+5.42e-03

Pos logit contributions
tub+0.011
gan+0.010
 Moe+0.009
ider+0.009
ses+0.008
Neg logit contributions
 envelop-0.011
Offic-0.010
 Amen-0.010
ons-0.009
Mic-0.009
Token,
Token ID11
Feature activation+1.04e-02

Pos logit contributions
tub+0.083
gan+0.080
 Moe+0.071
ider+0.069
ahl+0.069
Neg logit contributions
 envelop-0.104
Offic-0.095
 Amen-0.093
ons-0.091
oc-0.085
Token and
Token ID290
Feature activation-4.90e-02

Pos logit contributions
tub+0.016
gan+0.015
ider+0.013
 Moe+0.013
ahl+0.013
Neg logit contributions
 envelop-0.024
Offic-0.023
 Amen-0.022
ons-0.021
Mic-0.020
Token then
Token ID788
Feature activation+2.03e-02

Pos logit contributions
 envelop+0.093
Offic+0.085
 Amen+0.082
ons+0.079
 read+0.075
Neg logit contributions
tub-0.095
gan-0.088
ider-0.084
ahl-0.081
 Moe-0.080
Token he
Token ID339
Feature activation+1.94e-02

Pos logit contributions
tub+0.041
gan+0.038
ider+0.036
ahl+0.035
ses+0.034
Neg logit contributions
 envelop-0.038
Offic-0.033
ons-0.033
 read-0.032
 Amen-0.031
Token started
Token ID2067
Feature activation+6.15e-02

Pos logit contributions
tub+0.037
gan+0.035
 Moe+0.032
ider+0.031
ahl+0.030
Neg logit contributions
 envelop-0.038
Offic-0.035
 Amen-0.033
ons-0.033
Mic-0.031
Token to
Token ID284
Feature activation-1.99e-02

Pos logit contributions
tub+0.110
gan+0.104
ider+0.093
 Moe+0.092
ahl+0.089
Neg logit contributions
 envelop-0.127
Offic-0.118
 Amen-0.115
ons-0.111
Mic-0.105
Token play
Token ID711
Feature activation+7.78e-03

Pos logit contributions
 envelop+0.036
Offic+0.033
 Amen+0.032
ons+0.030
Mic+0.029
Neg logit contributions
tub-0.041
gan-0.039
ider-0.035
 Moe-0.035
ahl-0.034
Token on
Token ID319
Feature activation+6.51e-03

Pos logit contributions
gan+0.015
tub+0.015
ses+0.013
ider+0.013
 Moe+0.013
Neg logit contributions
 envelop-0.015
Offic-0.013
 Amen-0.013
ons-0.013
oc-0.012
Token the
Token ID262
Feature activation+5.55e-02

Pos logit contributions
tub+0.013
gan+0.013
ider+0.012
 Moe+0.011
ahl+0.011
Neg logit contributions
 envelop-0.012
Offic-0.011
 Amen-0.010
ons-0.010
Mic-0.009
Token stair
Token ID18761
Feature activation+6.77e-02

Pos logit contributions
tub+0.091
gan+0.087
ider+0.077
 Moe+0.076
ses+0.073
Neg logit contributions
 envelop-0.119
Offic-0.113
ons-0.108
 Amen-0.105
Mic-0.100
Token.
Token ID13
Feature activation+1.57e-02

Pos logit contributions
tub+0.118
gan+0.114
 Moe+0.102
ses+0.101
ahl+0.100
Neg logit contributions
 envelop-0.138
Offic-0.126
ons-0.122
 Amen-0.122
oc-0.114
Token carry
Token ID3283
Feature activation-3.12e-02

Pos logit contributions
 envelop+0.005
Offic+0.004
 Amen+0.004
ons+0.004
Mic+0.004
Neg logit contributions
tub-0.005
gan-0.005
 Moe-0.005
ider-0.005
ahl-0.004
Token Anna
Token ID11735
Feature activation+4.63e-03

Pos logit contributions
 envelop+0.056
Offic+0.049
ons+0.048
 Amen+0.048
Mic+0.044
Neg logit contributions
tub-0.066
gan-0.062
ider-0.057
 Moe-0.055
ahl-0.055
Token and
Token ID290
Feature activation-8.45e-02

Pos logit contributions
tub+0.009
gan+0.008
ider+0.007
 Moe+0.007
ahl+0.007
Neg logit contributions
 envelop-0.009
Offic-0.008
 Amen-0.008
ons-0.008
oc-0.007
Token he
Token ID339
Feature activation-6.94e-04

Pos logit contributions
 envelop+0.191
Offic+0.177
 Amen+0.174
ons+0.167
Mic+0.159
Neg logit contributions
tub-0.139
gan-0.127
ider-0.112
 Moe-0.110
ahl-0.107
Token can
Token ID460
Feature activation+1.03e-02

Pos logit contributions
 envelop+0.002
Offic+0.001
 Amen+0.001
ons+0.001
oc+0.001
Neg logit contributions
tub-0.001
gan-0.001
 Moe-0.001
ider-0.001
ahl-0.001
Token't
Token ID470
Feature activation-5.05e-03

Pos logit contributions
tub+0.020
gan+0.019
 Moe+0.017
ider+0.017
ahl+0.017
Neg logit contributions
 envelop-0.020
Offic-0.018
 Amen-0.018
ons-0.017
Mic-0.016
Token leave
Token ID2666
Feature activation+9.96e-03

Pos logit contributions
 envelop+0.009
Offic+0.008
 Amen+0.008
ons+0.008
Mic+0.007
Neg logit contributions
tub-0.010
gan-0.010
 Moe-0.009
ider-0.009
ahl-0.008
Token her
Token ID607
Feature activation+3.68e-02

Pos logit contributions
tub+0.019
gan+0.018
 Moe+0.016
ider+0.016
ahl+0.016
Neg logit contributions
 envelop-0.019
Offic-0.018
 Amen-0.017
ons-0.017
Mic-0.016
Token alone
Token ID3436
Feature activation+6.91e-02

Pos logit contributions
tub+0.065
gan+0.062
 Moe+0.056
ider+0.055
ahl+0.054
Neg logit contributions
 envelop-0.077
 Amen-0.067
Offic-0.067
ons-0.065
oc-0.061
Token.
Token ID13
Feature activation+1.25e-02

Pos logit contributions
tub+0.131
gan+0.127
ahl+0.117
 Moe+0.116
ses+0.113
Neg logit contributions
 envelop-0.130
 Amen-0.116
Offic-0.111
ons-0.110
oc-0.105
Token He
Token ID679
Feature activation-5.71e-03

Pos logit contributions
tub+0.024
gan+0.021
anger+0.020
ider+0.020
sed+0.020
Neg logit contributions
 Amen-0.023
 envelop-0.023
Offic-0.022
oc-0.021
Mic-0.019
Token to
Token ID284
Feature activation-4.05e-02

Pos logit contributions
tub+0.002
gan+0.002
ses+0.002
 Moe+0.002
ahl+0.002
Neg logit contributions
 envelop-0.002
Offic-0.002
 Amen-0.002
ons-0.002
oc-0.002
Token his
Token ID465
Feature activation+9.08e-03

Pos logit contributions
 envelop+0.086
Offic+0.081
 Amen+0.077
ons+0.076
Mic+0.074
Neg logit contributions
tub-0.069
gan-0.063
 Moe-0.059
 Mil-0.057
ahl-0.052
Token mom
Token ID1995
Feature activation-6.87e-02

Pos logit contributions
tub+0.015
gan+0.015
 Moe+0.013
ider+0.013
ahl+0.012
Neg logit contributions
 envelop-0.020
Offic-0.019
 Amen-0.018
ons-0.017
Mic-0.017
Token's
Token ID338
Feature activation-1.38e-02

Pos logit contributions
 envelop+0.139
Offic+0.132
 Amen+0.128
ons+0.122
Mic+0.119
Neg logit contributions
tub-0.123
gan-0.116
 Moe-0.103
ider-0.102
ahl-0.097
Token warning
Token ID6509
Feature activation-2.38e-02

Pos logit contributions
 envelop+0.027
Offic+0.025
ons+0.024
 Amen+0.024
Mic+0.023
Neg logit contributions
tub-0.026
gan-0.024
ider-0.022
 Moe-0.022
ses-0.022
Token and
Token ID290
Feature activation-7.67e-02

Pos logit contributions
 envelop+0.041
Offic+0.036
ons+0.036
 Amen+0.035
 lending+0.032
Neg logit contributions
tub-0.050
gan-0.049
ider-0.045
ahl-0.044
 Moe-0.043
Token kept
Token ID4030
Feature activation+9.73e-03

Pos logit contributions
 envelop+0.145
Offic+0.134
 Amen+0.131
ons+0.125
Mic+0.119
Neg logit contributions
tub-0.150
gan-0.143
 Moe-0.128
ider-0.127
ahl-0.123
Token his
Token ID465
Feature activation+3.67e-03

Pos logit contributions
tub+0.016
gan+0.015
 Moe+0.014
 Mil+0.013
ter+0.013
Neg logit contributions
 envelop-0.020
Offic-0.019
oc-0.019
 Amen-0.018
Mic-0.018
Token lungs
Token ID21726
Feature activation-3.72e-02

Pos logit contributions
tub+0.007
gan+0.006
ider+0.006
 Moe+0.006
ses+0.005
Neg logit contributions
 envelop-0.007
Offic-0.007
 Amen-0.007
ons-0.007
Mic-0.006
Token safe
Token ID3338
Feature activation+1.59e-02

Pos logit contributions
 envelop+0.070
Offic+0.063
 Amen+0.063
ons+0.060
Mic+0.056
Neg logit contributions
tub-0.073
gan-0.070
 Moe-0.063
ider-0.062
ahl-0.061
Token.
Token ID13
Feature activation+1.21e-02

Pos logit contributions
gan+0.029
tub+0.029
ahl+0.027
ider+0.026
 Moe+0.026
Neg logit contributions
 envelop-0.030
 Amen-0.027
Offic-0.025
ons-0.025
oc-0.025
Token waved
Token ID26834
Feature activation-6.39e-02

Pos logit contributions
 envelop+0.115
Offic+0.106
 Amen+0.104
ons+0.099
Mic+0.094
Neg logit contributions
tub-0.118
gan-0.112
 Moe-0.100
ider-0.100
ahl-0.097
Token their
Token ID511
Feature activation-3.95e-03

Pos logit contributions
 envelop+0.136
Offic+0.128
 Amen+0.125
ons+0.122
Mic+0.117
Neg logit contributions
tub-0.108
gan-0.103
 Moe-0.091
ider-0.089
ahl-0.087
Token fins
Token ID42080
Feature activation-3.52e-02

Pos logit contributions
 envelop+0.007
Offic+0.007
 Amen+0.007
ons+0.007
Mic+0.006
Neg logit contributions
tub-0.008
gan-0.007
 Moe-0.007
ider-0.007
ses-0.006
Token at
Token ID379
Feature activation-3.49e-02

Pos logit contributions
 envelop+0.071
Offic+0.064
 Amen+0.063
ons+0.061
oc+0.057
Neg logit contributions
tub-0.064
gan-0.061
 Moe-0.054
ider-0.054
ahl-0.053
Token Ben
Token ID3932
Feature activation+1.62e-02

Pos logit contributions
 envelop+0.063
Offic+0.059
 Amen+0.058
ons+0.055
Mic+0.052
Neg logit contributions
tub-0.071
gan-0.067
 Moe-0.061
ider-0.060
ahl-0.059
Token and
Token ID290
Feature activation-8.33e-02

Pos logit contributions
tub+0.026
gan+0.025
 Moe+0.022
ider+0.022
ahl+0.021
Neg logit contributions
 envelop-0.036
Offic-0.033
 Amen-0.033
ons-0.032
Mic-0.030
Token Mia
Token ID32189
Feature activation+2.88e-03

Pos logit contributions
 envelop+0.177
Offic+0.165
 Amen+0.158
ons+0.146
oc+0.145
Neg logit contributions
tub-0.145
gan-0.130
ider-0.127
anger-0.113
ahl-0.112
Token.
Token ID13
Feature activation+5.42e-03

Pos logit contributions
tub+0.005
gan+0.004
 Moe+0.004
ider+0.004
ahl+0.004
Neg logit contributions
 envelop-0.006
Offic-0.006
 Amen-0.006
ons-0.006
oc-0.005
Token They
Token ID1119
Feature activation-1.99e-02

Pos logit contributions
gan+0.010
tub+0.009
ider+0.009
ahl+0.008
ter+0.008
Neg logit contributions
 envelop-0.010
Offic-0.010
 Amen-0.009
oc-0.008
 exchange-0.008
Token thanked
Token ID26280
Feature activation-6.02e-02

Pos logit contributions
 envelop+0.039
 Amen+0.036
Offic+0.035
ons+0.034
Mic+0.032
Neg logit contributions
tub-0.038
gan-0.036
 Moe-0.033
ider-0.031
ahl-0.031
Token them
Token ID606
Feature activation+1.12e-02

Pos logit contributions
 envelop+0.103
Offic+0.095
 Amen+0.092
ons+0.088
Mic+0.084
Neg logit contributions
tub-0.127
gan-0.123
 Moe-0.115
 Mil-0.110
ahl-0.107
Token
Token ID198
Feature activation-2.76e-02

Pos logit contributions
 envelop+0.063
Offic+0.060
 Amen+0.058
ons+0.056
oc+0.053
Neg logit contributions
tub-0.059
gan-0.056
ider-0.052
 Moe-0.049
ses-0.048
TokenM
Token ID44
Feature activation-3.56e-02

Pos logit contributions
 envelop+0.045
Offic+0.042
 Amen+0.040
ons+0.038
Mic+0.036
Neg logit contributions
tub-0.062
gan-0.058
ider-0.053
 Moe-0.053
ses-0.052
Tokenia
Token ID544
Feature activation-1.25e-02

Pos logit contributions
 envelop+0.084
ons+0.083
Offic+0.078
pa+0.076
 Amen+0.073
Neg logit contributions
tub-0.050
 Moe-0.035
robe-0.034
gan-0.034
ter-0.033
Token hugged
Token ID41130
Feature activation-6.77e-02

Pos logit contributions
 envelop+0.022
Offic+0.020
ons+0.019
 Amen+0.019
 read+0.018
Neg logit contributions
tub-0.026
gan-0.023
ider-0.022
ahl-0.022
 Moe-0.022
Token her
Token ID607
Feature activation-4.71e-03

Pos logit contributions
 envelop+0.117
Offic+0.107
 Amen+0.105
ons+0.099
Mic+0.093
Neg logit contributions
tub-0.144
gan-0.139
 Moe-0.127
ider-0.124
ahl-0.122
Token mom
Token ID1995
Feature activation-7.73e-02

Pos logit contributions
 envelop+0.011
Offic+0.010
 Amen+0.010
ons+0.010
Mic+0.009
Neg logit contributions
tub-0.007
gan-0.007
 Moe-0.006
ider-0.006
ses-0.006
Token and
Token ID290
Feature activation-9.99e-02

Pos logit contributions
 envelop+0.172
Offic+0.171
 Amen+0.162
ons+0.156
Mic+0.153
Neg logit contributions
gan-0.119
tub-0.113
 Moe-0.099
ider-0.091
ahl-0.090
Token dad
Token ID9955
Feature activation-8.76e-02

Pos logit contributions
 envelop+0.198
 Amen+0.187
Offic+0.186
ons+0.177
oc+0.173
Neg logit contributions
gan-0.176
tub-0.175
 Moe-0.154
 Mil-0.149
ses-0.146
Token.
Token ID13
Feature activation+4.23e-03

Pos logit contributions
 envelop+0.190
Offic+0.174
 Amen+0.172
ons+0.167
oc+0.157
Neg logit contributions
tub-0.150
gan-0.139
ider-0.124
 Moe-0.122
ahl-0.119
Token She
Token ID1375
Feature activation-1.55e-02

Pos logit contributions
tub+0.007
gan+0.007
ider+0.007
ahl+0.006
anger+0.006
Neg logit contributions
 envelop-0.009
Offic-0.008
 Amen-0.008
Mic-0.007
ons-0.007
Token said
Token ID531
Feature activation-3.33e-02

Pos logit contributions
 envelop+0.035
Offic+0.033
 Amen+0.032
ons+0.031
Mic+0.030
Neg logit contributions
tub-0.024
gan-0.023
 Moe-0.020
ider-0.020
ahl-0.019
Token,
Token ID11
Feature activation+1.77e-02

Pos logit contributions
gan+0.048
tub+0.046
ider+0.043
ses+0.042
ahl+0.041
Neg logit contributions
 envelop-0.056
Offic-0.050
ons-0.050
 Amen-0.046
oc-0.045
Token Lily
Token ID20037
Feature activation+1.88e-02

Pos logit contributions
tub+0.028
gan+0.026
ider+0.026
ses+0.025
ahl+0.023
Neg logit contributions
 envelop-0.039
Offic-0.035
ons-0.034
 Amen-0.034
 read-0.032
Token reminded
Token ID14516
Feature activation-3.09e-02

Pos logit contributions
tub+0.037
gan+0.035
 Moe+0.031
ider+0.031
ahl+0.030
Neg logit contributions
 envelop-0.035
Offic-0.033
 Amen-0.033
ons-0.031
Mic-0.029
Token her
Token ID607
Feature activation+3.31e-03

Pos logit contributions
 envelop+0.054
Offic+0.051
 Amen+0.049
oc+0.048
ons+0.044
Neg logit contributions
gan-0.063
tub-0.062
 Moe-0.058
 Mil-0.054
ahl-0.054
Token dad
Token ID9955
Feature activation-8.79e-02

Pos logit contributions
tub+0.005
gan+0.005
 Moe+0.004
 Mil+0.003
ider+0.003
Neg logit contributions
 envelop-0.008
Offic-0.008
 Amen-0.008
Mic-0.007
oc-0.007
Token and
Token ID290
Feature activation-7.63e-02

Pos logit contributions
 envelop+0.178
Offic+0.163
 Amen+0.160
ons+0.157
Mic+0.144
Neg logit contributions
tub-0.161
gan-0.152
ider-0.139
 Moe-0.134
ahl-0.131
Token he
Token ID339
Feature activation-9.49e-03

Pos logit contributions
 envelop+0.151
Offic+0.146
 Amen+0.140
Mic+0.135
oc+0.134
Neg logit contributions
tub-0.137
gan-0.132
 Moe-0.117
 Mil-0.111
ahl-0.105
Token thanked
Token ID26280
Feature activation-6.17e-02

Pos logit contributions
 envelop+0.019
Offic+0.017
 Amen+0.017
ons+0.016
Mic+0.015
Neg logit contributions
tub-0.018
gan-0.017
ider-0.015
 Moe-0.015
ahl-0.014
Token her
Token ID607
Feature activation-4.60e-03

Pos logit contributions
Offic+0.126
 envelop+0.122
oc+0.118
 saddle+0.117
ons+0.114
Neg logit contributions
 Moe-0.091
gan-0.088
tub-0.086
 Mil-0.086
ter-0.076
Token for
Token ID329
Feature activation-3.90e-02

Pos logit contributions
 envelop+0.011
Offic+0.010
 Amen+0.010
ons+0.010
Mic+0.009
Neg logit contributions
tub-0.007
gan-0.007
 Moe-0.006
ider-0.006
ahl-0.005
Token being
Token ID852
Feature activation+2.46e-03

Pos logit contributions
oc+0.084
 envelop+0.075
Offic+0.072
 Amen+0.071
Mic+0.071
Neg logit contributions
tub-0.061
gan-0.059
 Moe-0.057
ter-0.057
 Mil-0.052
Token to
Token ID284
Feature activation-3.88e-02

Pos logit contributions
 envelop+0.109
Offic+0.104
 Amen+0.102
ons+0.097
Mic+0.094
Neg logit contributions
tub-0.081
gan-0.076
 Moe-0.066
ider-0.066
ahl-0.064
Token their
Token ID511
Feature activation-6.17e-03

Pos logit contributions
 envelop+0.082
ons+0.080
 Amen+0.079
Offic+0.079
Mic+0.074
Neg logit contributions
tub-0.062
gan-0.059
 Moe-0.054
 Mil-0.051
ses-0.049
Token room
Token ID2119
Feature activation+1.77e-02

Pos logit contributions
 envelop+0.014
Offic+0.013
 Amen+0.013
ons+0.012
Mic+0.012
Neg logit contributions
tub-0.010
gan-0.009
 Moe-0.008
ider-0.008
ses-0.008
Token.
Token ID13
Feature activation+1.63e-02

Pos logit contributions
tub+0.028
gan+0.026
ider+0.022
 Moe+0.022
ahl+0.021
Neg logit contributions
 envelop-0.040
Offic-0.038
 Amen-0.037
ons-0.036
Mic-0.035
Token They
Token ID1119
Feature activation-2.22e-02

Pos logit contributions
tub+0.027
gan+0.027
ider+0.024
ahl+0.023
 Moe+0.023
Neg logit contributions
 envelop-0.035
 Amen-0.032
Offic-0.032
ons-0.029
oc-0.028
Token gave
Token ID2921
Feature activation-7.26e-02

Pos logit contributions
 envelop+0.043
 Amen+0.041
Offic+0.041
ons+0.039
Mic+0.038
Neg logit contributions
tub-0.041
gan-0.040
 Moe-0.035
 Mil-0.033
ahl-0.033
Token him
Token ID683
Feature activation-1.14e-03

Pos logit contributions
Offic+0.145
 envelop+0.139
 Amen+0.134
Mic+0.129
ons+0.125
Neg logit contributions
tub-0.125
gan-0.124
 Moe-0.113
ses-0.104
 Mil-0.103
Token food
Token ID2057
Feature activation-4.16e-02

Pos logit contributions
 envelop+0.002
Offic+0.002
 Amen+0.002
ons+0.002
Mic+0.002
Neg logit contributions
tub-0.002
gan-0.002
 Moe-0.002
ider-0.002
ahl-0.002
Token and
Token ID290
Feature activation-8.83e-02

Pos logit contributions
 envelop+0.081
Offic+0.071
 Amen+0.069
ons+0.069
oc+0.064
Neg logit contributions
tub-0.079
gan-0.077
ider-0.070
 Moe-0.069
ahl-0.068
Token water
Token ID1660
Feature activation-1.36e-02

Pos logit contributions
Offic+0.159
 envelop+0.159
 Amen+0.158
Mic+0.145
ons+0.145
Neg logit contributions
tub-0.167
gan-0.156
 Moe-0.141
ses-0.141
sed-0.135
Token.
Token ID13
Feature activation+1.86e-02

Pos logit contributions
 envelop+0.028
Offic+0.024
ons+0.024
 Amen+0.024
oc+0.022
Neg logit contributions
tub-0.024
gan-0.024
ider-0.022
 Moe-0.020
ahl-0.020