TOP ACTIVATIONS
MAX = 0.178

back on the zip - line , z ipping all around
robe back onto the clothes line . He used all his
try to loosen the b ram bles . As he was
I fell off the clothes line and I can 't get
dizz y when you â re having fun .
the edge of the shore line and just stood there .
post so you wouldn â t forget it when
when it hugged the shore line again , it noticed a
blanket before it was torn S ue decided
of laundry on the clothes line . " He wants to
it would hug the shore line and admire the wrecked boats
a walk along the shore line . But something
so the cat wouldn â t catch him trying
the edge of the shore line and just stood there .
as he explored the shore line . One day
jumped onto the zip - line and z ipped around the
magnet so it wouldn â t fold . The
quick or we won â t be able to
when you open it â . Molly opened the box
away so she wouldn â t have to m

MIN ACTIVATIONS
MIN = -0.168

different tastes . But most importantly , she learned to love
But most importantly , they discussed what they
need it . And most importantly , she learned that a
before . But , most importantly , she could feel the
of things . But most importantly of all , he had
of things . But most importantly of all , he had
, treats , and most importantly , Lily 's love .
beautiful things and , most importantly , they had a lot
. It had her most precious memories and gifts . You
dance . This playful team high - f ived
something difficult . But more importantly , she proved to herself
see . The playful puppy learned that sometimes sne
up . The clever girl took her mother 's
! â . The playful baby tiger â
!" The playful bee was startled . He
grateful . The clever person thanked his friend for
more . The playful puppy was so happy that
see who was the most intelligent chess player . They hugged
distance . The intelligent kid thought of a plan
help . The clever fish smiled and said ,

INTERVAL 0.092 - 0.178
CONTAINS 1.479%

down and try to catch it !" They
see Lily having fun . The two friends
the play . So Anna started to learn all the lines
. Every time he pressed it , the animals became bigger
parent â s home , the little dog in

INTERVAL 0.005 - 0.092
CONTAINS 45.491%

and said , " Hi , do ggy ! Do you
have it back , please ?" Lily asked .
But the boy was safe . He watched as the boat
" Wow , you are so clever !" But
calm and kind . She feels a little bra ver .

INTERVAL -0.081 - 0.005
CONTAINS 49.549%

aps his hands . " More , more ,
, please ask me before you paint toast . OK ?"
sky was turning dark and gray . Suddenly , a big
small bites and taste a little bit first to make sure
anger . It was perfect for his red shirt .

INTERVAL -0.168 - -0.081
CONTAINS 3.481%

heard a loud boom . They looked up and saw dark
, but she was very brave and adventurous . One day
big storm came with lots of wind and rain . Tim
. " What do you mean ?" " We
play and the birds sang beautiful songs .
Token back
Token ID736
Feature activation+7.45e-02

Pos logit contributions
angers+0.103
 salon+0.073
ooter+0.072
ulk+0.070
anger+0.070
Neg logit contributions
oc-0.130
 course-0.120
 deer-0.114
 unity-0.110
 thanks-0.109
Token on
Token ID319
Feature activation-2.08e-02

Pos logit contributions
angers+0.139
 salon+0.104
ooter+0.100
ulk+0.099
aj+0.099
Neg logit contributions
oc-0.157
 course-0.145
 deer-0.138
 thanks-0.134
 unity-0.131
Token the
Token ID262
Feature activation+6.39e-02

Pos logit contributions
oc+0.044
affe+0.036
 deer+0.036
 unity+0.035
 course+0.034
Neg logit contributions
angers-0.038
 salon-0.030
 bathroom-0.029
anger-0.028
 door-0.028
Token zip
Token ID19974
Feature activation+2.07e-02

Pos logit contributions
angers+0.108
 salon+0.081
ooter+0.079
amas+0.075
aj+0.074
Neg logit contributions
oc-0.146
 course-0.132
 deer-0.124
 unity-0.123
 sorts-0.121
Token-
Token ID12
Feature activation+5.10e-02

Pos logit contributions
angers+0.043
 salon+0.033
ooter+0.032
anger+0.032
belt+0.032
Neg logit contributions
oc-0.040
 course-0.036
 deer-0.034
 unity-0.033
 thanks-0.033
Tokenline
Token ID1370
Feature activation+1.64e-01

Pos logit contributions
angers+0.088
 salon+0.064
ooter+0.063
ulk+0.061
belt+0.060
Neg logit contributions
oc-0.111
 course-0.104
 deer-0.102
 thanks-0.098
 unity-0.097
Token,
Token ID11
Feature activation+6.47e-02

Pos logit contributions
angers+0.349
 salon+0.269
ooter+0.261
aj+0.260
ulk+0.260
Neg logit contributions
oc-0.303
 course-0.278
 deer-0.261
 thanks-0.253
 unity-0.247
Token z
Token ID1976
Feature activation-4.87e-05

Pos logit contributions
angers+0.116
 salon+0.093
line+0.085
belt+0.084
 bathroom+0.083
Neg logit contributions
oc-0.143
 course-0.124
 deer-0.118
affe-0.117
 unity-0.114
Tokenipping
Token ID4501
Feature activation-1.70e-02

Pos logit contributions
oc+0.000
 course+0.000
 thanks+0.000
 unity+0.000
 deer+0.000
Neg logit contributions
angers-0.000
ooter-0.000
 salon-0.000
outed-0.000
roller-0.000
Token all
Token ID477
Feature activation-3.14e-03

Pos logit contributions
oc+0.034
 course+0.030
 deer+0.028
 unity+0.028
 matter+0.027
Neg logit contributions
angers-0.031
anger-0.025
ulk-0.023
 salon-0.022
belt-0.022
Token around
Token ID1088
Feature activation+6.27e-03

Pos logit contributions
oc+0.006
 deer+0.005
affe+0.005
 turkey+0.005
 unity+0.005
Neg logit contributions
angers-0.006
 salon-0.005
ulk-0.005
ips-0.004
outed-0.004
Tokenrobe
Token ID25481
Feature activation+6.86e-02

Pos logit contributions
angers+0.083
 salon+0.060
ews+0.059
azes+0.058
ulk+0.056
Neg logit contributions
oc-0.172
 course-0.169
 thanks-0.159
 deer-0.158
 harmony-0.152
Token back
Token ID736
Feature activation+7.00e-02

Pos logit contributions
angers+0.138
 salon+0.107
ooter+0.105
ulk+0.101
anger+0.100
Neg logit contributions
oc-0.135
 course-0.120
 deer-0.116
 unity-0.111
 thanks-0.109
Token onto
Token ID4291
Feature activation-2.09e-03

Pos logit contributions
angers+0.133
 salon+0.099
ooter+0.097
belt+0.093
anger+0.093
Neg logit contributions
oc-0.148
 course-0.134
 deer-0.127
 unity-0.124
 thanks-0.121
Token the
Token ID262
Feature activation+6.08e-02

Pos logit contributions
oc+0.004
affe+0.003
 course+0.003
 bond+0.003
 unity+0.003
Neg logit contributions
angers-0.004
 bathroom-0.003
ulk-0.003
 salon-0.003
anger-0.003
Token clothes
Token ID8242
Feature activation+4.78e-02

Pos logit contributions
angers+0.138
 salon+0.105
outed+0.105
azes+0.105
ulk+0.105
Neg logit contributions
oc-0.103
 deer-0.097
 course-0.096
 thanks-0.088
 unity-0.085
Tokenline
Token ID1370
Feature activation+1.63e-01

Pos logit contributions
angers+0.097
 salon+0.075
ooter+0.072
ulk+0.070
belt+0.070
Neg logit contributions
oc-0.094
 course-0.085
 deer-0.081
 unity-0.078
 thanks-0.077
Token.
Token ID13
Feature activation+1.91e-02

Pos logit contributions
angers+0.338
 salon+0.260
ooter+0.251
aj+0.250
ulk+0.250
Neg logit contributions
oc-0.308
 course-0.284
 deer-0.268
 thanks-0.259
 unity-0.253
Token He
Token ID679
Feature activation-5.94e-02

Pos logit contributions
angers+0.041
 salon+0.031
ooter+0.030
ulk+0.030
aj+0.030
Neg logit contributions
oc-0.035
 course-0.033
 deer-0.031
 unity-0.030
 thanks-0.029
Token used
Token ID973
Feature activation-4.66e-02

Pos logit contributions
oc+0.128
 course+0.115
 unity+0.109
 deer+0.109
 harmony+0.106
Neg logit contributions
angers-0.100
 salon-0.080
outed-0.073
ooter-0.072
aj-0.071
Token all
Token ID477
Feature activation-1.92e-02

Pos logit contributions
oc+0.107
affe+0.082
 deer+0.080
 course+0.079
 bond+0.078
Neg logit contributions
angers-0.080
ulk-0.062
 bathroom-0.059
elt-0.059
ooter-0.059
Token his
Token ID465
Feature activation+2.85e-02

Pos logit contributions
oc+0.039
 deer+0.031
affe+0.029
 unity+0.028
 hunter+0.027
Neg logit contributions
angers-0.036
ulk-0.030
elt-0.030
outed-0.029
ips-0.029
Token try
Token ID1949
Feature activation+3.94e-02

Pos logit contributions
oc+0.022
affe+0.020
 course+0.019
 unity+0.018
 deer+0.017
Neg logit contributions
angers-0.018
 bathroom-0.017
 salon-0.016
ulk-0.015
ooter-0.015
Token to
Token ID284
Feature activation-1.03e-02

Pos logit contributions
angers+0.081
 salon+0.063
ooter+0.061
aj+0.060
ulk+0.060
Neg logit contributions
oc-0.076
 course-0.070
 deer-0.066
 thanks-0.063
 unity-0.062
Token loosen
Token ID40953
Feature activation-2.17e-02

Pos logit contributions
oc+0.021
 course+0.018
 thanks+0.017
affe+0.017
 unity+0.017
Neg logit contributions
angers-0.016
ooter-0.014
 bathroom-0.014
 salon-0.013
our-0.013
Token the
Token ID262
Feature activation+5.73e-02

Pos logit contributions
oc+0.041
 bond+0.036
affe+0.035
 course+0.035
 unity+0.033
Neg logit contributions
angers-0.040
unes-0.031
ulk-0.030
anger-0.029
 salon-0.029
Token b
Token ID275
Feature activation+6.38e-03

Pos logit contributions
angers+0.119
 salon+0.110
ooter+0.106
anger+0.101
alo+0.099
Neg logit contributions
oc-0.091
 sorts-0.077
 course-0.075
 unity-0.073
 harmony-0.070
Tokenram
Token ID859
Feature activation+1.61e-01

Pos logit contributions
angers+0.013
 salon+0.010
aj+0.009
ulk+0.009
ooter+0.009
Neg logit contributions
oc-0.013
 course-0.012
 deer-0.011
 thanks-0.011
 unity-0.010
Tokenbles
Token ID7689
Feature activation+1.17e-02

Pos logit contributions
angers+0.252
 salon+0.209
ulk+0.194
outed+0.188
aj+0.186
Neg logit contributions
oc-0.333
 deer-0.310
 course-0.307
 thanks-0.295
 kindness-0.279
Token.
Token ID13
Feature activation+1.35e-02

Pos logit contributions
angers+0.022
 salon+0.016
ooter+0.016
anger+0.016
ulk+0.015
Neg logit contributions
oc-0.024
 course-0.022
 deer-0.021
 unity-0.021
 harmony-0.020
Token As
Token ID1081
Feature activation-5.67e-03

Pos logit contributions
angers+0.032
 salon+0.026
aj+0.025
ulk+0.024
ooter+0.024
Neg logit contributions
oc-0.022
 course-0.019
 thanks-0.018
 deer-0.018
 unity-0.017
Token he
Token ID339
Feature activation+1.74e-03

Pos logit contributions
oc+0.012
 course+0.011
affe+0.011
bit+0.010
 bond+0.010
Neg logit contributions
angers-0.009
belt-0.008
 salon-0.007
 upstairs-0.007
ulk-0.007
Token was
Token ID373
Feature activation+6.74e-02

Pos logit contributions
angers+0.003
 salon+0.002
outed+0.002
ulk+0.002
elt+0.002
Neg logit contributions
oc-0.003
 course-0.003
 deer-0.003
 unity-0.003
 thanks-0.003
TokenI
Token ID40
Feature activation-6.65e-02

Pos logit contributions
oc+0.012
 course+0.011
 deer+0.010
 thanks+0.010
 unity+0.010
Neg logit contributions
angers-0.009
 salon-0.007
ooter-0.007
elt-0.007
ulk-0.007
Token fell
Token ID3214
Feature activation+2.64e-02

Pos logit contributions
oc+0.138
 course+0.137
 unity+0.128
 deer+0.127
 harmony+0.121
Neg logit contributions
angers-0.113
 salon-0.083
aj-0.082
ooter-0.081
anger-0.079
Token off
Token ID572
Feature activation+1.26e-02

Pos logit contributions
angers+0.057
 salon+0.045
outed+0.043
azes+0.043
ooter+0.043
Neg logit contributions
oc-0.047
 course-0.043
 deer-0.042
 thanks-0.040
 sorts-0.037
Token the
Token ID262
Feature activation+5.95e-02

Pos logit contributions
angers+0.029
 salon+0.023
ulk+0.022
aj+0.022
belt+0.022
Neg logit contributions
oc-0.021
 course-0.018
 deer-0.017
 unity-0.017
affe-0.016
Token clothes
Token ID8242
Feature activation+4.07e-02

Pos logit contributions
angers+0.133
azes+0.106
outed+0.105
elt+0.101
ulk+0.101
Neg logit contributions
oc-0.098
 deer-0.097
 course-0.095
 thanks-0.087
 tiger-0.085
Tokenline
Token ID1370
Feature activation+1.57e-01

Pos logit contributions
angers+0.091
 salon+0.073
ooter+0.071
aj+0.069
belt+0.069
Neg logit contributions
oc-0.071
 course-0.063
 deer-0.060
 unity-0.057
 thanks-0.056
Token and
Token ID290
Feature activation-2.54e-02

Pos logit contributions
angers+0.345
 salon+0.272
ooter+0.266
aj+0.260
ulk+0.259
Neg logit contributions
oc-0.278
 course-0.252
 deer-0.237
 thanks-0.225
 unity-0.224
Token I
Token ID314
Feature activation-4.67e-02

Pos logit contributions
oc+0.045
 course+0.040
 deer+0.037
 unity+0.035
 thanks+0.035
Neg logit contributions
angers-0.056
 salon-0.046
ulk-0.044
ooter-0.044
aj-0.043
Token can
Token ID460
Feature activation+2.11e-02

Pos logit contributions
oc+0.089
 course+0.084
 deer+0.079
 unity+0.077
 thanks+0.074
Neg logit contributions
angers-0.093
 salon-0.072
ooter-0.071
aj-0.070
outed-0.069
Token't
Token ID470
Feature activation-4.86e-02

Pos logit contributions
angers+0.042
 salon+0.032
ooter+0.031
ulk+0.030
aj+0.030
Neg logit contributions
oc-0.042
 course-0.039
 deer-0.037
 thanks-0.035
 unity-0.035
Token get
Token ID651
Feature activation+8.15e-02

Pos logit contributions
oc+0.081
 course+0.073
 deer+0.068
 thanks+0.065
 unity+0.064
Neg logit contributions
angers-0.113
 salon-0.090
ooter-0.088
ulk-0.087
aj-0.086
Token dizz
Token ID29527
Feature activation-8.40e-03

Pos logit contributions
angers+0.077
 salon+0.061
ooter+0.060
aj+0.059
ulk+0.059
Neg logit contributions
oc-0.052
 course-0.047
 deer-0.043
 thanks-0.041
 unity-0.041
Tokeny
Token ID88
Feature activation+2.40e-02

Pos logit contributions
oc+0.008
 deer+0.006
 thanks+0.006
able+0.006
 course+0.005
Neg logit contributions
angers-0.023
 salon-0.020
ews-0.020
ulk-0.020
omm-0.019
Token when
Token ID618
Feature activation+5.40e-02

Pos logit contributions
angers+0.045
 salon+0.033
ooter+0.033
ulk+0.032
belt+0.032
Neg logit contributions
oc-0.050
 course-0.046
 deer-0.044
 unity-0.042
 thanks-0.041
Token you
Token ID345
Feature activation-7.46e-03

Pos logit contributions
angers+0.135
 salon+0.109
ulk+0.106
ooter+0.106
aj+0.105
Neg logit contributions
oc-0.081
 course-0.071
 deer-0.065
 thanks-0.062
 unity-0.061
Tokenâ
Token ID22940
Feature activation+2.60e-02

Pos logit contributions
oc+0.012
 course+0.011
 deer+0.010
 thanks+0.010
 unity+0.009
Neg logit contributions
angers-0.018
 salon-0.014
ooter-0.014
ulk-0.014
aj-0.014
Token
Token ID26391
Feature activation+1.57e-01

Pos logit contributions
angers+0.051
aj+0.036
 salon+0.035
ews+0.035
outed+0.034
Neg logit contributions
 course-0.057
oc-0.056
 thanks-0.049
 matter-0.048
 unity-0.046
Token
Token ID8151
Feature activation+3.92e-02

Pos logit contributions
angers+0.270
 salon+0.206
outed+0.206
ooter+0.192
ulk+0.189
Neg logit contributions
oc-0.358
 course-0.304
 thanks-0.289
 harmony-0.276
 deer-0.274
Tokenre
Token ID260
Feature activation+5.85e-02

Pos logit contributions
angers+0.095
alo+0.075
belt+0.073
amas+0.073
ews+0.073
Neg logit contributions
 course-0.058
oc-0.058
 thanks-0.046
Fred-0.043
 Freddie-0.042
Token having
Token ID1719
Feature activation-1.36e-02

Pos logit contributions
angers+0.123
 salon+0.096
anger+0.090
ulk+0.090
ooter+0.089
Neg logit contributions
oc-0.107
 course-0.100
 thanks-0.097
 deer-0.096
 unity-0.091
Token fun
Token ID1257
Feature activation-4.41e-02

Pos logit contributions
oc+0.020
 course+0.017
 deer+0.017
 thanks+0.015
 unity+0.015
Neg logit contributions
angers-0.034
 salon-0.027
ooter-0.027
ulk-0.027
aj-0.027
Token.
Token ID13
Feature activation+1.50e-02

Pos logit contributions
oc+0.089
 course+0.081
 deer+0.078
 thanks+0.074
 unity+0.074
Neg logit contributions
angers-0.086
 salon-0.065
ooter-0.063
ulk-0.063
aj-0.062
Token the
Token ID262
Feature activation+6.33e-02

Pos logit contributions
oc+0.020
 course+0.017
 deer+0.016
affe+0.015
 unity+0.015
Neg logit contributions
angers-0.029
 salon-0.024
ulk-0.024
 bathroom-0.023
ooter-0.023
Token edge
Token ID5743
Feature activation+9.14e-02

Pos logit contributions
angers+0.129
 salon+0.099
ooter+0.096
ulk+0.095
aj+0.094
Neg logit contributions
oc-0.123
 course-0.113
 deer-0.108
 thanks-0.103
 unity-0.102
Token of
Token ID286
Feature activation-7.91e-02

Pos logit contributions
angers+0.230
 salon+0.188
ooter+0.184
ulk+0.181
aj+0.180
Neg logit contributions
oc-0.132
 course-0.114
 deer-0.108
 unity-0.103
 thanks-0.099
Token the
Token ID262
Feature activation+5.98e-02

Pos logit contributions
oc+0.144
affe+0.112
 course+0.107
 unity+0.103
 deer+0.102
Neg logit contributions
angers-0.176
 salon-0.144
ulk-0.143
 bathroom-0.142
outed-0.132
Token shore
Token ID15191
Feature activation+8.09e-02

Pos logit contributions
angers+0.131
 salon+0.101
ooter+0.098
aj+0.097
ulk+0.097
Neg logit contributions
oc-0.108
 course-0.099
 deer-0.096
 thanks-0.091
 unity-0.089
Tokenline
Token ID1370
Feature activation+1.57e-01

Pos logit contributions
angers+0.195
 salon+0.155
ulk+0.153
ooter+0.152
aj+0.151
Neg logit contributions
oc-0.127
 course-0.114
 deer-0.106
 thanks-0.105
 sorts-0.097
Token and
Token ID290
Feature activation-2.63e-02

Pos logit contributions
angers+0.402
 salon+0.323
aj+0.320
ulk+0.319
ooter+0.312
Neg logit contributions
oc-0.220
 course-0.199
 deer-0.182
 thanks-0.182
 sorts-0.166
Token just
Token ID655
Feature activation-3.02e-02

Pos logit contributions
oc+0.049
 course+0.045
 deer+0.042
 thanks+0.041
 unity+0.040
Neg logit contributions
angers-0.056
 salon-0.043
ooter-0.042
ulk-0.041
aj-0.041
Token stood
Token ID6204
Feature activation+1.06e-01

Pos logit contributions
oc+0.055
 course+0.051
 deer+0.048
 thanks+0.047
 unity+0.045
Neg logit contributions
angers-0.066
 salon-0.051
ooter-0.049
ulk-0.049
amas-0.048
Token there
Token ID612
Feature activation+1.82e-02

Pos logit contributions
angers+0.245
 salon+0.202
ooter+0.201
anger+0.193
unes+0.187
Neg logit contributions
oc-0.169
 deer-0.163
 course-0.162
 thanks-0.160
 harmony-0.145
Token.
Token ID13
Feature activation+1.43e-02

Pos logit contributions
angers+0.037
 salon+0.028
ooter+0.027
ulk+0.027
aj+0.027
Neg logit contributions
oc-0.036
 course-0.033
 deer-0.031
 thanks-0.030
 unity-0.029
Tokenpost
Token ID7353
Feature activation+4.50e-02

Pos logit contributions
angers+0.187
 salon+0.146
ulk+0.144
ooter+0.144
outed+0.141
Neg logit contributions
oc-0.144
 course-0.132
 deer-0.121
 thanks-0.119
 unity-0.114
Token so
Token ID523
Feature activation+2.65e-02

Pos logit contributions
angers+0.102
 salon+0.080
ooter+0.078
aj+0.077
ulk+0.077
Neg logit contributions
oc-0.077
 course-0.070
 deer-0.066
 thanks-0.063
 unity-0.062
Token you
Token ID345
Feature activation-1.55e-02

Pos logit contributions
angers+0.062
 salon+0.050
belt+0.048
ooter+0.048
aj+0.048
Neg logit contributions
oc-0.043
 course-0.039
 deer-0.037
 unity-0.034
 thanks-0.034
Token wouldn
Token ID3636
Feature activation+8.21e-02

Pos logit contributions
oc+0.026
 course+0.024
 deer+0.022
 unity+0.021
 thanks+0.021
Neg logit contributions
angers-0.035
 salon-0.028
ooter-0.027
aj-0.027
belt-0.027
Tokenâ
Token ID22940
Feature activation+3.12e-02

Pos logit contributions
angers+0.186
outed+0.151
aj+0.148
 salon+0.142
elt+0.141
Neg logit contributions
oc-0.139
 course-0.125
 deer-0.119
 thanks-0.116
 love-0.110
Token
Token ID26391
Feature activation+1.56e-01

Pos logit contributions
angers+0.060
aj+0.042
outed+0.042
ews+0.041
 salon+0.040
Neg logit contributions
oc-0.068
 course-0.068
 thanks-0.059
 matter-0.057
 sorts-0.055
Token
Token ID8151
Feature activation+4.96e-02

Pos logit contributions
angers+0.275
outed+0.212
ulk+0.200
 salon+0.199
elt+0.195
Neg logit contributions
oc-0.344
 course-0.293
 thanks-0.279
 deer-0.267
affe-0.265
Tokent
Token ID83
Feature activation-2.02e-02

Pos logit contributions
angers+0.092
ulk+0.068
azes+0.065
 salon+0.065
aj+0.064
Neg logit contributions
oc-0.105
 course-0.094
 thanks-0.086
affe-0.084
 deer-0.083
Token forget
Token ID6044
Feature activation-5.12e-02

Pos logit contributions
oc+0.030
 course+0.027
 deer+0.025
 thanks+0.024
 unity+0.023
Neg logit contributions
angers-0.050
 salon-0.040
ooter-0.040
ulk-0.039
aj-0.039
Token it
Token ID340
Feature activation+1.03e-01

Pos logit contributions
oc+0.081
 course+0.074
 deer+0.069
 thanks+0.066
 unity+0.064
Neg logit contributions
angers-0.123
 salon-0.097
ooter-0.095
ulk-0.094
outed-0.093
Token when
Token ID618
Feature activation+5.43e-02

Pos logit contributions
angers+0.226
 salon+0.178
ooter+0.174
aj+0.172
ulk+0.170
Neg logit contributions
oc-0.181
 course-0.163
 deer-0.154
 thanks-0.146
 unity-0.146
Token when
Token ID618
Feature activation+3.82e-02

Pos logit contributions
angers+0.099
 salon+0.078
elt+0.077
 bathroom+0.075
amas+0.072
Neg logit contributions
oc-0.110
 course-0.093
affe-0.090
 sorts-0.089
 unity-0.088
Token it
Token ID340
Feature activation+9.73e-02

Pos logit contributions
angers+0.100
 salon+0.082
ulk+0.080
ooter+0.079
belt+0.078
Neg logit contributions
oc-0.052
 course-0.045
 deer-0.041
 thanks-0.039
 unity-0.038
Token hugged
Token ID41130
Feature activation-1.14e-03

Pos logit contributions
angers+0.238
 salon+0.184
ooter+0.181
ulk+0.176
unes+0.175
Neg logit contributions
oc-0.172
 course-0.148
 deer-0.141
 thanks-0.138
 matter-0.128
Token the
Token ID262
Feature activation+6.19e-02

Pos logit contributions
oc+0.002
 course+0.001
 sorts+0.001
 unity+0.001
 thanks+0.001
Neg logit contributions
angers-0.003
 salon-0.002
belt-0.002
ooter-0.002
outed-0.002
Token shore
Token ID15191
Feature activation+7.97e-02

Pos logit contributions
angers+0.120
 salon+0.095
ooter+0.093
 bathroom+0.092
belt+0.090
Neg logit contributions
oc-0.125
 course-0.109
 sorts-0.099
 thanks-0.096
 Gary-0.095
Tokenline
Token ID1370
Feature activation+1.56e-01

Pos logit contributions
angers+0.159
 salon+0.122
ooter+0.119
belt+0.117
aj+0.115
Neg logit contributions
oc-0.157
 course-0.144
 deer-0.135
 unity-0.130
 thanks-0.129
Token again
Token ID757
Feature activation+4.75e-02

Pos logit contributions
angers+0.327
 salon+0.253
ooter+0.247
aj+0.242
ulk+0.242
Neg logit contributions
oc-0.292
 course-0.267
 deer-0.251
 thanks-0.241
 unity-0.238
Token,
Token ID11
Feature activation+7.05e-02

Pos logit contributions
angers+0.099
 salon+0.076
ooter+0.074
aj+0.073
ulk+0.073
Neg logit contributions
oc-0.091
 course-0.083
 deer-0.078
 thanks-0.075
 unity-0.074
Token it
Token ID340
Feature activation+8.67e-02

Pos logit contributions
angers+0.154
 salon+0.125
ulk+0.122
anger+0.117
 bathroom+0.116
Neg logit contributions
oc-0.125
 course-0.102
affe-0.093
 sorts-0.092
 deer-0.091
Token noticed
Token ID6810
Feature activation+3.55e-02

Pos logit contributions
angers+0.171
 salon+0.129
ooter+0.125
ulk+0.123
aj+0.123
Neg logit contributions
oc-0.175
 course-0.161
 deer-0.153
 thanks-0.148
 unity-0.145
Token a
Token ID257
Feature activation-2.55e-02

Pos logit contributions
angers+0.075
 salon+0.060
outed+0.058
ulk+0.058
elt+0.057
Neg logit contributions
oc-0.065
 course-0.054
 deer-0.052
 unity-0.051
 thanks-0.050
Token blanket
Token ID18447
Feature activation+7.88e-03

Pos logit contributions
angers+0.105
aj+0.090
 salon+0.087
belt+0.086
ooter+0.084
Neg logit contributions
oc-0.113
 course-0.098
 sorts-0.092
 deer-0.091
 unity-0.088
Token before
Token ID878
Feature activation+6.94e-02

Pos logit contributions
angers+0.014
 salon+0.011
belt+0.010
ooter+0.010
aj+0.010
Neg logit contributions
oc-0.018
 course-0.016
 deer-0.015
 unity-0.015
 harmony-0.014
Token it
Token ID340
Feature activation+1.13e-01

Pos logit contributions
angers+0.152
ooter+0.118
 salon+0.117
aj+0.114
outed+0.113
Neg logit contributions
oc-0.121
 course-0.116
 deer-0.108
 thanks-0.107
 unity-0.101
Token was
Token ID373
Feature activation+8.81e-02

Pos logit contributions
angers+0.235
 salon+0.180
ooter+0.174
ulk+0.174
aj+0.170
Neg logit contributions
oc-0.217
 course-0.200
 deer-0.189
 thanks-0.185
 unity-0.176
Token torn
Token ID12445
Feature activation-3.29e-02

Pos logit contributions
angers+0.191
 salon+0.149
ooter+0.145
ulk+0.143
aj+0.142
Neg logit contributions
oc-0.160
 course-0.146
 deer-0.137
 thanks-0.132
 unity-0.130
Token
Token ID220
Feature activation+1.55e-01

Pos logit contributions
oc+0.056
 course+0.051
 deer+0.049
 thanks+0.048
 unity+0.044
Neg logit contributions
angers-0.072
 salon-0.058
ulk-0.056
ooter-0.056
azes-0.055
Token
Token ID198
Feature activation-2.52e-02

Pos logit contributions
angers+0.302
 salon+0.246
ulk+0.239
ooter+0.223
 bathroom+0.221
Neg logit contributions
oc-0.284
 thanks-0.240
 course-0.234
affe-0.228
 sorts-0.225
Token
Token ID198
Feature activation-3.41e-02

Pos logit contributions
oc+0.046
 course+0.042
 deer+0.039
 thanks+0.038
 unity+0.037
Neg logit contributions
angers-0.055
 salon-0.043
ooter-0.041
ulk-0.041
aj-0.041
TokenS
Token ID50
Feature activation-2.88e-02

Pos logit contributions
oc+0.055
 deer+0.049
 thanks+0.049
 course+0.048
 unity+0.047
Neg logit contributions
angers-0.074
 salon-0.064
ulk-0.061
ooter-0.059
belt-0.059
Tokenue
Token ID518
Feature activation-1.97e-02

Pos logit contributions
oc+0.052
 course+0.050
 thanks+0.043
 deer+0.041
affe+0.040
Neg logit contributions
angers-0.065
belt-0.049
ulk-0.048
 salon-0.048
ooter-0.047
Token decided
Token ID3066
Feature activation+4.27e-02

Pos logit contributions
oc+0.039
 course+0.036
 deer+0.035
 unity+0.034
 harmony+0.033
Neg logit contributions
angers-0.036
 salon-0.030
outed-0.028
ooter-0.028
aj-0.028
Token of
Token ID286
Feature activation-8.84e-02

Pos logit contributions
angers+0.058
 salon+0.048
ooter+0.047
belt+0.046
anger+0.046
Neg logit contributions
oc-0.036
 course-0.031
 deer-0.029
 unity-0.029
 thanks-0.028
Token laundry
Token ID25724
Feature activation+5.49e-02

Pos logit contributions
oc+0.175
 course+0.150
 unity+0.145
 thanks+0.141
affe+0.140
Neg logit contributions
angers-0.161
 bathroom-0.132
ooter-0.125
 salon-0.122
 office-0.119
Token on
Token ID319
Feature activation-3.94e-02

Pos logit contributions
angers+0.109
 salon+0.082
ooter+0.080
ulk+0.079
aj+0.079
Neg logit contributions
oc-0.109
 course-0.101
 deer-0.096
 thanks-0.092
 unity-0.090
Token the
Token ID262
Feature activation+4.62e-02

Pos logit contributions
oc+0.070
 course+0.060
affe+0.059
 deer+0.059
 unity+0.058
Neg logit contributions
angers-0.081
 bathroom-0.066
 salon-0.066
ulk-0.063
belt-0.062
Token clothes
Token ID8242
Feature activation+3.21e-02

Pos logit contributions
angers+0.080
ulk+0.056
 salon+0.056
outed+0.056
aj+0.056
Neg logit contributions
oc-0.102
 course-0.096
 deer-0.093
 thanks-0.088
 unity-0.087
Tokenline
Token ID1370
Feature activation+1.55e-01

Pos logit contributions
angers+0.066
 salon+0.050
ooter+0.049
ulk+0.048
aj+0.048
Neg logit contributions
oc-0.061
 course-0.057
 deer-0.053
 thanks-0.052
 unity-0.050
Token.
Token ID13
Feature activation+1.78e-02

Pos logit contributions
angers+0.328
 salon+0.251
aj+0.244
ulk+0.244
ooter+0.242
Neg logit contributions
oc-0.286
 course-0.263
 deer-0.247
 thanks-0.240
 unity-0.231
Token "
Token ID366
Feature activation-1.27e-02

Pos logit contributions
angers+0.042
 salon+0.034
ooter+0.033
ulk+0.032
aj+0.032
Neg logit contributions
oc-0.029
 course-0.025
 deer-0.024
 thanks-0.024
 unity-0.022
TokenHe
Token ID1544
Feature activation-6.41e-02

Pos logit contributions
oc+0.023
 course+0.022
 thanks+0.021
 deer+0.021
 unity+0.020
Neg logit contributions
angers-0.026
 salon-0.021
ulk-0.020
elt-0.020
belt-0.019
Token wants
Token ID3382
Feature activation-2.62e-02

Pos logit contributions
oc+0.121
 deer+0.111
 course+0.110
 unity+0.110
 harmony+0.110
Neg logit contributions
angers-0.121
aj-0.096
ooter-0.096
 salon-0.096
outed-0.093
Token to
Token ID284
Feature activation-2.57e-02

Pos logit contributions
oc+0.063
 course+0.055
 deer+0.052
 unity+0.051
 sorts+0.051
Neg logit contributions
angers-0.041
 salon-0.031
ooter-0.030
aj-0.029
 bathroom-0.029
Token it
Token ID340
Feature activation+8.54e-02

Pos logit contributions
angers+0.122
 salon+0.100
 bathroom+0.092
line+0.091
belt+0.087
Neg logit contributions
oc-0.133
affe-0.110
 course-0.106
 sorts-0.105
able-0.102
Token would
Token ID561
Feature activation+3.52e-02

Pos logit contributions
angers+0.193
 salon+0.153
ooter+0.149
aj+0.148
outed+0.147
Neg logit contributions
oc-0.142
 course-0.131
 deer-0.121
 unity-0.115
 thanks-0.114
Token hug
Token ID16225
Feature activation-2.66e-02

Pos logit contributions
angers+0.059
 salon+0.052
ooter+0.049
 bathroom+0.048
belt+0.047
Neg logit contributions
oc-0.071
 course-0.061
affe-0.058
able-0.057
 unity-0.056
Token the
Token ID262
Feature activation+5.97e-02

Pos logit contributions
oc+0.042
 course+0.038
 sorts+0.034
 deer+0.033
 unity+0.033
Neg logit contributions
angers-0.060
amas-0.047
 salon-0.047
belt-0.046
ulk-0.045
Token shore
Token ID15191
Feature activation+7.70e-02

Pos logit contributions
angers+0.114
 bathroom+0.091
 salon+0.091
ooter+0.089
amas+0.087
Neg logit contributions
oc-0.118
 course-0.102
 sorts-0.096
 thanks-0.090
 Fred-0.090
Tokenline
Token ID1370
Feature activation+1.55e-01

Pos logit contributions
angers+0.157
 salon+0.120
ooter+0.118
belt+0.115
line+0.112
Neg logit contributions
oc-0.145
 course-0.139
 deer-0.125
 matter-0.123
 unity-0.122
Token and
Token ID290
Feature activation-2.29e-02

Pos logit contributions
angers+0.311
ooter+0.240
 salon+0.239
belt+0.229
ews+0.224
Neg logit contributions
oc-0.300
 course-0.275
 deer-0.252
 unity-0.243
 matter-0.240
Token admire
Token ID21099
Feature activation-2.12e-02

Pos logit contributions
oc+0.043
 course+0.038
affe+0.036
able+0.033
Fred+0.032
Neg logit contributions
angers-0.039
 bathroom-0.035
 salon-0.035
ulk-0.033
belt-0.033
Token the
Token ID262
Feature activation+5.54e-02

Pos logit contributions
oc+0.039
affe+0.030
 course+0.029
Fred+0.026
 deer+0.026
Neg logit contributions
angers-0.045
ulk-0.036
 salon-0.035
outed-0.034
azes-0.034
Token wrecked
Token ID44770
Feature activation+4.54e-03

Pos logit contributions
angers+0.095
 bathroom+0.080
 salon+0.074
 door+0.074
ews+0.073
Neg logit contributions
oc-0.113
 course-0.098
 Charlie-0.091
affe-0.091
 Gary-0.091
Token boats
Token ID16659
Feature activation-4.31e-03

Pos logit contributions
angers+0.011
 salon+0.009
ooter+0.009
aj+0.009
ulk+0.009
Neg logit contributions
oc-0.007
 course-0.006
 deer-0.006
 thanks-0.005
 unity-0.005
Token a
Token ID257
Feature activation-2.62e-02

Pos logit contributions
oc+0.019
 course+0.019
 thanks+0.017
 deer+0.017
 sorts+0.015
Neg logit contributions
angers-0.041
 salon-0.035
outed-0.034
belt-0.034
aj-0.034
Token walk
Token ID2513
Feature activation+1.27e-02

Pos logit contributions
oc+0.048
 course+0.043
 deer+0.039
 unity+0.039
 thanks+0.038
Neg logit contributions
angers-0.056
 salon-0.045
ooter-0.045
aj-0.043
anger-0.042
Token along
Token ID1863
Feature activation-1.23e-02

Pos logit contributions
angers+0.025
 salon+0.019
ooter+0.018
ulk+0.018
aj+0.018
Neg logit contributions
oc-0.026
 course-0.024
 deer-0.022
 thanks-0.022
 unity-0.021
Token the
Token ID262
Feature activation+5.00e-02

Pos logit contributions
oc+0.024
 course+0.022
 deer+0.021
 unity+0.020
 thanks+0.020
Neg logit contributions
angers-0.024
 salon-0.019
ulk-0.018
belt-0.018
ooter-0.018
Token shore
Token ID15191
Feature activation+7.53e-02

Pos logit contributions
angers+0.092
 salon+0.068
ooter+0.066
ulk+0.065
aj+0.065
Neg logit contributions
oc-0.108
 course-0.099
 deer-0.094
 thanks-0.091
 unity-0.090
Tokenline
Token ID1370
Feature activation+1.55e-01

Pos logit contributions
angers+0.158
 salon+0.122
aj+0.120
ulk+0.119
ooter+0.119
Neg logit contributions
oc-0.141
 course-0.128
 deer-0.121
 thanks-0.117
 sorts-0.113
Token.
Token ID13
Feature activation+6.64e-03

Pos logit contributions
angers+0.339
aj+0.263
 salon+0.263
ulk+0.258
outed+0.254
Neg logit contributions
oc-0.268
 course-0.249
 deer-0.232
 thanks-0.230
 sorts-0.221
Token
Token ID198
Feature activation-4.14e-02

Pos logit contributions
angers+0.017
 salon+0.014
belt+0.013
aj+0.013
ooter+0.013
Neg logit contributions
oc-0.010
 course-0.009
 sorts-0.009
 thanks-0.008
 deer-0.008
Token
Token ID198
Feature activation-4.62e-02

Pos logit contributions
oc+0.075
 course+0.069
 deer+0.065
 thanks+0.062
 unity+0.061
Neg logit contributions
angers-0.089
 salon-0.069
ooter-0.068
aj-0.067
ulk-0.066
TokenBut
Token ID1537
Feature activation-5.90e-03

Pos logit contributions
oc+0.079
 course+0.071
 deer+0.067
 thanks+0.065
 unity+0.064
Neg logit contributions
angers-0.104
 salon-0.083
ulk-0.080
ooter-0.080
aj-0.078
Token something
Token ID1223
Feature activation+1.83e-02

Pos logit contributions
oc+0.010
 course+0.009
 deer+0.008
 thanks+0.008
 unity+0.008
Neg logit contributions
angers-0.013
 salon-0.011
ooter-0.010
belt-0.010
ulk-0.010
Token so
Token ID523
Feature activation+2.14e-02

Pos logit contributions
angers+0.069
 salon+0.053
aj+0.052
ooter+0.052
ulk+0.051
Neg logit contributions
oc-0.063
 course-0.058
 deer-0.054
 thanks-0.053
 unity-0.051
Token the
Token ID262
Feature activation+4.84e-02

Pos logit contributions
angers+0.051
 salon+0.041
belt+0.040
ooter+0.040
ulk+0.040
Neg logit contributions
oc-0.034
 course-0.031
 deer-0.028
 thanks-0.026
 unity-0.026
Token cat
Token ID3797
Feature activation-2.53e-02

Pos logit contributions
angers+0.109
 salon+0.089
ooter+0.086
 bathroom+0.084
aj+0.084
Neg logit contributions
oc-0.085
 course-0.074
 deer-0.067
 unity-0.067
 thanks-0.066
Token wouldn
Token ID3636
Feature activation+7.61e-02

Pos logit contributions
oc+0.050
 course+0.045
 unity+0.042
 deer+0.042
 harmony+0.041
Neg logit contributions
angers-0.049
 salon-0.040
ooter-0.038
belt-0.037
aj-0.036
Tokenâ
Token ID22940
Feature activation+2.32e-02

Pos logit contributions
angers+0.158
aj+0.120
 salon+0.118
outed+0.117
ooter+0.114
Neg logit contributions
oc-0.148
 course-0.136
 deer-0.128
 thanks-0.124
 tiger-0.116
Token
Token ID26391
Feature activation+1.55e-01

Pos logit contributions
angers+0.046
aj+0.034
ews+0.033
outed+0.032
 salon+0.031
Neg logit contributions
oc-0.049
 course-0.049
 thanks-0.042
 matter-0.042
 bond-0.039
Token
Token ID8151
Feature activation+4.65e-02

Pos logit contributions
angers+0.276
outed+0.200
 salon+0.195
anger+0.191
ulk+0.190
Neg logit contributions
oc-0.344
 course-0.294
 thanks-0.286
 deer-0.271
 harmony-0.268
Tokent
Token ID83
Feature activation-2.26e-02

Pos logit contributions
angers+0.087
 salon+0.062
ulk+0.061
aj+0.061
azes+0.059
Neg logit contributions
oc-0.100
 course-0.092
 thanks-0.084
 deer-0.083
affe-0.080
Token catch
Token ID4929
Feature activation+7.42e-03

Pos logit contributions
oc+0.043
 course+0.038
 unity+0.035
 deer+0.034
 sorts+0.034
Neg logit contributions
angers-0.046
 salon-0.037
ooter-0.037
ulk-0.035
anger-0.034
Token him
Token ID683
Feature activation+7.83e-02

Pos logit contributions
angers+0.017
ulk+0.014
 salon+0.014
ooter+0.013
amas+0.013
Neg logit contributions
oc-0.012
 course-0.010
 deer-0.010
 unity-0.010
 sorts-0.009
Token trying
Token ID2111
Feature activation+7.35e-03

Pos logit contributions
angers+0.147
 salon+0.112
ooter+0.108
ulk+0.104
aj+0.104
Neg logit contributions
oc-0.162
 course-0.150
 deer-0.142
 unity-0.138
 thanks-0.136
Token the
Token ID262
Feature activation+6.05e-02

Pos logit contributions
oc+0.023
 course+0.019
 deer+0.017
affe+0.017
 unity+0.017
Neg logit contributions
angers-0.033
 salon-0.027
ulk-0.027
 bathroom-0.026
ooter-0.025
Token edge
Token ID5743
Feature activation+9.11e-02

Pos logit contributions
angers+0.125
 salon+0.096
ooter+0.093
ulk+0.092
aj+0.092
Neg logit contributions
oc-0.115
 course-0.106
 deer-0.102
 thanks-0.097
 unity-0.096
Token of
Token ID286
Feature activation-8.15e-02

Pos logit contributions
angers+0.238
 salon+0.195
ooter+0.190
ulk+0.188
aj+0.188
Neg logit contributions
oc-0.125
 course-0.109
 deer-0.101
 unity-0.094
 thanks-0.094
Token the
Token ID262
Feature activation+5.77e-02

Pos logit contributions
oc+0.147
affe+0.115
 course+0.110
 unity+0.106
 deer+0.105
Neg logit contributions
angers-0.182
 salon-0.149
ulk-0.148
 bathroom-0.147
outed-0.136
Token shore
Token ID15191
Feature activation+7.94e-02

Pos logit contributions
angers+0.129
 salon+0.099
ooter+0.097
aj+0.096
ulk+0.096
Neg logit contributions
oc-0.102
 course-0.094
 deer-0.092
 thanks-0.087
 unity-0.085
Tokenline
Token ID1370
Feature activation+1.53e-01

Pos logit contributions
angers+0.197
 salon+0.158
ulk+0.156
ooter+0.156
aj+0.155
Neg logit contributions
oc-0.119
 course-0.106
 thanks-0.099
 deer-0.097
 sorts-0.089
Token and
Token ID290
Feature activation-2.76e-02

Pos logit contributions
angers+0.403
 salon+0.325
aj+0.324
ulk+0.322
anger+0.314
Neg logit contributions
oc-0.204
 course-0.183
 thanks-0.169
 deer-0.168
 sorts-0.153
Token just
Token ID655
Feature activation-3.26e-02

Pos logit contributions
oc+0.051
 course+0.047
 deer+0.044
 thanks+0.042
 unity+0.042
Neg logit contributions
angers-0.059
 salon-0.045
ooter-0.044
ulk-0.044
aj-0.044
Token stood
Token ID6204
Feature activation+1.05e-01

Pos logit contributions
oc+0.058
 course+0.054
 deer+0.051
 thanks+0.051
 unity+0.048
Neg logit contributions
angers-0.072
 salon-0.056
ooter-0.054
amas-0.054
ulk-0.053
Token there
Token ID612
Feature activation+1.74e-02

Pos logit contributions
angers+0.245
ooter+0.204
 salon+0.203
anger+0.195
unes+0.189
Neg logit contributions
oc-0.161
 deer-0.157
 thanks-0.155
 course-0.154
 harmony-0.138
Token.
Token ID13
Feature activation+1.22e-02

Pos logit contributions
angers+0.035
 salon+0.027
ooter+0.026
aj+0.026
ulk+0.026
Neg logit contributions
oc-0.034
 course-0.031
 deer-0.029
 thanks-0.028
 unity-0.028
Token as
Token ID355
Feature activation+1.62e-02

Pos logit contributions
angers+0.137
 salon+0.102
ooter+0.101
belt+0.099
aj+0.099
Neg logit contributions
oc-0.140
 course-0.133
 deer-0.123
 thanks-0.118
 unity-0.118
Token he
Token ID339
Feature activation+8.84e-03

Pos logit contributions
angers+0.038
belt+0.029
 salon+0.029
ooter+0.029
outed+0.028
Neg logit contributions
oc-0.028
 course-0.025
 deer-0.023
 thanks-0.022
 unity-0.021
Token explored
Token ID18782
Feature activation+1.06e-02

Pos logit contributions
angers+0.019
 salon+0.015
ooter+0.014
ulk+0.014
aj+0.014
Neg logit contributions
oc-0.017
 course-0.015
 deer-0.014
 thanks-0.014
 sorts-0.013
Token the
Token ID262
Feature activation+5.14e-02

Pos logit contributions
angers+0.021
ulk+0.017
 salon+0.016
ooter+0.016
belt+0.015
Neg logit contributions
oc-0.020
 course-0.017
affe-0.017
 deer-0.017
 unity-0.016
Token shore
Token ID15191
Feature activation+7.71e-02

Pos logit contributions
angers+0.105
 salon+0.080
ooter+0.077
ulk+0.077
aj+0.076
Neg logit contributions
oc-0.101
 course-0.094
 deer-0.089
 thanks-0.086
 unity-0.084
Tokenline
Token ID1370
Feature activation+1.53e-01

Pos logit contributions
angers+0.161
 salon+0.126
ooter+0.123
ulk+0.122
aj+0.122
Neg logit contributions
oc-0.145
 course-0.132
 deer-0.124
 thanks-0.122
 sorts-0.116
Token.
Token ID13
Feature activation+9.05e-03

Pos logit contributions
angers+0.329
 salon+0.257
aj+0.253
ulk+0.250
ooter+0.246
Neg logit contributions
oc-0.276
 course-0.255
 deer-0.238
 thanks-0.234
 sorts-0.224
Token
Token ID198
Feature activation-3.83e-02

Pos logit contributions
angers+0.018
 salon+0.014
ooter+0.014
ulk+0.014
aj+0.013
Neg logit contributions
oc-0.018
 course-0.016
 deer-0.016
 unity-0.015
 thanks-0.015
Token
Token ID198
Feature activation-4.41e-02

Pos logit contributions
oc+0.071
 course+0.065
 deer+0.062
 thanks+0.059
 unity+0.059
Neg logit contributions
angers-0.081
 salon-0.062
ooter-0.061
ulk-0.060
aj-0.060
TokenOne
Token ID3198
Feature activation-5.16e-02

Pos logit contributions
oc+0.083
 deer+0.076
 thanks+0.073
 course+0.072
 unity+0.071
Neg logit contributions
angers-0.084
boards-0.065
 salon-0.065
outed-0.064
ulk-0.064
Token day
Token ID1110
Feature activation+3.24e-02

Pos logit contributions
oc+0.084
 course+0.075
 deer+0.070
 unity+0.066
 thanks+0.066
Neg logit contributions
angers-0.121
 salon-0.097
ooter-0.095
aj-0.093
outed-0.092
Token jumped
Token ID11687
Feature activation+5.24e-02

Pos logit contributions
oc+0.017
 course+0.015
 deer+0.015
 unity+0.014
 thanks+0.014
Neg logit contributions
angers-0.017
 salon-0.014
ooter-0.013
aj-0.013
outed-0.013
Token onto
Token ID4291
Feature activation-1.08e-02

Pos logit contributions
angers+0.112
 salon+0.089
aj+0.085
outed+0.084
ulk+0.084
Neg logit contributions
oc-0.095
 course-0.087
 deer-0.086
 thanks-0.083
 tiger-0.076
Token the
Token ID262
Feature activation+5.36e-02

Pos logit contributions
oc+0.018
 course+0.015
 deer+0.014
affe+0.014
 unity+0.014
Neg logit contributions
angers-0.025
 salon-0.021
ulk-0.020
anger-0.020
 bathroom-0.019
Token zip
Token ID19974
Feature activation+1.31e-02

Pos logit contributions
angers+0.120
ulk+0.094
outed+0.094
aj+0.093
ews+0.093
Neg logit contributions
oc-0.090
 course-0.087
 deer-0.084
 thanks-0.077
 tiger-0.075
Token-
Token ID12
Feature activation+4.65e-02

Pos logit contributions
angers+0.031
 salon+0.025
ooter+0.024
anger+0.024
belt+0.024
Neg logit contributions
oc-0.021
 course-0.019
 deer-0.018
 unity-0.017
 thanks-0.017
Tokenline
Token ID1370
Feature activation+1.53e-01

Pos logit contributions
angers+0.093
 salon+0.071
ooter+0.068
aj+0.068
ulk+0.068
Neg logit contributions
oc-0.092
 course-0.085
 deer-0.080
 thanks-0.077
 unity-0.076
Token and
Token ID290
Feature activation-2.41e-02

Pos logit contributions
angers+0.389
 salon+0.312
aj+0.312
ulk+0.310
outed+0.303
Neg logit contributions
oc-0.216
 course-0.195
 deer-0.181
 thanks-0.177
 sorts-0.166
Token z
Token ID1976
Feature activation-2.70e-03

Pos logit contributions
oc+0.048
 course+0.044
 deer+0.041
 unity+0.040
 thanks+0.039
Neg logit contributions
angers-0.048
 salon-0.037
ooter-0.035
ulk-0.035
belt-0.035
Tokenipped
Token ID3949
Feature activation+1.93e-02

Pos logit contributions
 course+0.006
oc+0.005
 thanks+0.005
osaurs+0.004
affe+0.004
Neg logit contributions
angers-0.004
 salon-0.004
 bathroom-0.004
amas-0.004
ews-0.003
Token around
Token ID1088
Feature activation+6.40e-03

Pos logit contributions
angers+0.043
 salon+0.033
ooter+0.033
ulk+0.032
anger+0.032
Neg logit contributions
oc-0.034
 course-0.031
 deer-0.029
 unity-0.028
 thanks-0.027
Token the
Token ID262
Feature activation+5.12e-02

Pos logit contributions
angers+0.013
 salon+0.010
ulk+0.010
anger+0.009
ooter+0.009
Neg logit contributions
oc-0.013
 course-0.011
 deer-0.011
 unity-0.010
 thanks-0.010
Token magnet
Token ID19972
Feature activation-1.37e-05

Pos logit contributions
angers+0.133
ooter+0.121
amas+0.120
 bathroom+0.111
 door+0.110
Neg logit contributions
oc-0.107
 sorts-0.092
 course-0.091
 harmony-0.089
 unity-0.086
Token so
Token ID523
Feature activation+2.77e-02

Pos logit contributions
oc+0.000
 course+0.000
 deer+0.000
 unity+0.000
 matter+0.000
Neg logit contributions
angers-0.000
ooter-0.000
belt-0.000
 salon-0.000
amas-0.000
Token it
Token ID340
Feature activation+9.97e-02

Pos logit contributions
angers+0.059
belt+0.048
 salon+0.047
outed+0.046
ulk+0.045
Neg logit contributions
oc-0.049
 course-0.043
 deer-0.040
 sorts-0.040
 matter-0.039
Token wouldn
Token ID3636
Feature activation+7.86e-02

Pos logit contributions
angers+0.180
 salon+0.138
belt+0.135
ooter+0.135
outed+0.135
Neg logit contributions
oc-0.206
 course-0.191
 deer-0.178
 unity-0.178
 harmony-0.175
Tokenâ
Token ID22940
Feature activation+2.53e-02

Pos logit contributions
angers+0.149
 salon+0.111
ooter+0.108
aj+0.106
ulk+0.106
Neg logit contributions
oc-0.164
 course-0.151
 deer-0.144
 thanks-0.139
 unity-0.136
Token
Token ID26391
Feature activation+1.53e-01

Pos logit contributions
angers+0.046
 salon+0.032
aj+0.032
ews+0.032
outed+0.031
Neg logit contributions
oc-0.059
 course-0.056
 thanks-0.050
 matter-0.048
 deer-0.047
Token
Token ID8151
Feature activation+4.60e-02

Pos logit contributions
angers+0.258
 salon+0.187
outed+0.179
ulk+0.176
ooter+0.176
Neg logit contributions
oc-0.357
 course-0.320
 deer-0.304
 thanks-0.299
 unity-0.289
Tokent
Token ID83
Feature activation-2.35e-02

Pos logit contributions
angers+0.082
ulk+0.059
 salon+0.058
aj+0.057
azes+0.056
Neg logit contributions
oc-0.102
 course-0.092
 thanks-0.084
 deer-0.084
affe-0.081
Token fold
Token ID5591
Feature activation+3.19e-03

Pos logit contributions
oc+0.045
 course+0.039
 unity+0.036
 harmony+0.035
 deer+0.035
Neg logit contributions
angers-0.047
ooter-0.040
 salon-0.038
ulk-0.037
belt-0.036
Token.
Token ID13
Feature activation+2.07e-02

Pos logit contributions
angers+0.006
ulk+0.005
belt+0.005
anger+0.005
amas+0.005
Neg logit contributions
oc-0.006
 deer-0.005
 course-0.005
affe-0.005
 tiger-0.005
Token The
Token ID383
Feature activation-2.13e-02

Pos logit contributions
angers+0.043
 salon+0.033
ooter+0.033
ulk+0.032
aj+0.032
Neg logit contributions
oc-0.039
 course-0.036
 deer-0.034
 unity-0.032
 thanks-0.032
Token quick
Token ID2068
Feature activation-7.65e-02

Pos logit contributions
angers+0.157
 salon+0.125
belt+0.122
ooter+0.121
anger+0.117
Neg logit contributions
oc-0.140
 course-0.126
 deer-0.123
 unity-0.114
 tiger-0.110
Token or
Token ID393
Feature activation-1.95e-02

Pos logit contributions
oc+0.149
 course+0.132
 deer+0.125
 unity+0.123
 bond+0.118
Neg logit contributions
angers-0.153
anger-0.113
 salon-0.112
amas-0.112
aj-0.112
Token we
Token ID356
Feature activation-4.18e-02

Pos logit contributions
oc+0.036
 course+0.033
 deer+0.030
 thanks+0.029
 unity+0.029
Neg logit contributions
angers-0.042
 salon-0.033
ooter-0.032
aj-0.032
belt-0.031
Token won
Token ID1839
Feature activation+4.20e-02

Pos logit contributions
oc+0.066
 course+0.059
 deer+0.053
 unity+0.052
 thanks+0.051
Neg logit contributions
angers-0.096
ooter-0.080
 salon-0.078
belt-0.077
aj-0.075
Tokenâ
Token ID22940
Feature activation+1.82e-02

Pos logit contributions
angers+0.098
 salon+0.078
ooter+0.076
aj+0.076
ulk+0.075
Neg logit contributions
oc-0.069
 course-0.062
 deer-0.058
 thanks-0.056
 unity-0.054
Token
Token ID26391
Feature activation+1.53e-01

Pos logit contributions
angers+0.041
aj+0.032
 salon+0.031
outed+0.031
elt+0.030
Neg logit contributions
oc-0.033
 course-0.032
 thanks-0.027
 matter-0.027
 unity-0.026
Token
Token ID8151
Feature activation+4.21e-02

Pos logit contributions
angers+0.264
 salon+0.198
outed+0.189
ooter+0.187
ulk+0.184
Neg logit contributions
oc-0.348
 course-0.303
 thanks-0.290
 deer-0.290
 harmony-0.280
Tokent
Token ID83
Feature activation-2.32e-02

Pos logit contributions
angers+0.078
 salon+0.057
ulk+0.057
aj+0.056
outed+0.054
Neg logit contributions
oc-0.091
 course-0.081
 thanks-0.075
 deer-0.075
affe-0.073
Token be
Token ID307
Feature activation+6.98e-02

Pos logit contributions
oc+0.040
 course+0.037
 deer+0.033
 thanks+0.033
 unity+0.033
Neg logit contributions
angers-0.051
 salon-0.041
ooter-0.041
ulk-0.039
anger-0.038
Token able
Token ID1498
Feature activation-4.04e-02

Pos logit contributions
angers+0.151
 salon+0.118
ooter+0.116
aj+0.115
ulk+0.115
Neg logit contributions
oc-0.126
 course-0.114
 deer-0.106
 unity-0.101
 thanks-0.100
Token to
Token ID284
Feature activation-1.24e-02

Pos logit contributions
oc+0.099
 course+0.092
 deer+0.086
 unity+0.083
 thanks+0.083
Neg logit contributions
angers-0.062
 salon-0.043
ooter-0.042
ulk-0.041
aj-0.041
Token when
Token ID618
Feature activation+4.25e-02

Pos logit contributions
oc+0.057
 course+0.047
 deer+0.046
 unity+0.044
 sorts+0.044
Neg logit contributions
angers-0.069
 salon-0.056
aj-0.054
ooter-0.054
belt-0.053
Token you
Token ID345
Feature activation-1.45e-02

Pos logit contributions
angers+0.117
ooter+0.096
outed+0.095
 salon+0.095
ulk+0.094
Neg logit contributions
oc-0.047
 course-0.045
 deer-0.041
 thanks-0.040
 unity-0.037
Token open
Token ID1280
Feature activation-2.67e-02

Pos logit contributions
oc+0.025
 course+0.022
 deer+0.021
 unity+0.021
 harmony+0.020
Neg logit contributions
angers-0.031
ooter-0.025
 salon-0.024
outed-0.024
aj-0.024
Token it
Token ID340
Feature activation+1.02e-01

Pos logit contributions
oc+0.039
 course+0.033
 deer+0.031
 unity+0.029
 thanks+0.029
Neg logit contributions
angers-0.069
 salon-0.056
aj-0.054
ooter-0.054
anger-0.054
Tokenâ
Token ID22940
Feature activation+2.11e-02

Pos logit contributions
angers+0.230
 salon+0.184
ulk+0.179
ooter+0.178
aj+0.174
Neg logit contributions
oc-0.173
 course-0.157
 deer-0.146
 thanks-0.142
 unity-0.135
Token
Token ID26391
Feature activation+1.53e-01

Pos logit contributions
angers+0.041
 salon+0.030
outed+0.029
aj+0.029
ews+0.028
Neg logit contributions
oc-0.045
 course-0.045
 thanks-0.040
 sorts-0.037
 matter-0.036
Token.
Token ID13
Feature activation+1.58e-02

Pos logit contributions
angers+0.278
outed+0.245
ooter+0.225
 salon+0.224
ulk+0.224
Neg logit contributions
oc-0.318
 course-0.252
 thanks-0.247
 harmony-0.233
 Charlie-0.231
Token Molly
Token ID30236
Feature activation+1.75e-02

Pos logit contributions
angers+0.033
 salon+0.025
aj+0.025
ooter+0.025
ulk+0.025
Neg logit contributions
oc-0.029
 course-0.027
 deer-0.025
 unity-0.024
 thanks-0.024
Token opened
Token ID4721
Feature activation+4.97e-03

Pos logit contributions
angers+0.034
 salon+0.027
aj+0.026
ooter+0.026
outed+0.025
Neg logit contributions
oc-0.033
 course-0.031
 deer-0.029
 unity-0.029
 harmony-0.028
Token the
Token ID262
Feature activation+5.28e-02

Pos logit contributions
angers+0.012
 salon+0.009
aj+0.009
ulk+0.009
anger+0.009
Neg logit contributions
oc-0.009
 course-0.007
 unity-0.007
 deer-0.007
affe-0.007
Token box
Token ID3091
Feature activation+3.32e-02

Pos logit contributions
angers+0.098
ulk+0.068
elt+0.067
outed+0.067
 salon+0.067
Neg logit contributions
oc-0.114
 course-0.109
 deer-0.103
 thanks-0.103
 matter-0.096
Token away
Token ID1497
Feature activation+3.28e-02

Pos logit contributions
oc+0.057
 course+0.054
 deer+0.050
 thanks+0.050
 unity+0.048
Neg logit contributions
angers-0.062
 salon-0.047
ulk-0.046
ooter-0.045
aj-0.045
Token so
Token ID523
Feature activation+2.54e-02

Pos logit contributions
angers+0.070
 salon+0.054
ooter+0.053
aj+0.052
belt+0.052
Neg logit contributions
oc-0.061
 course-0.055
 deer-0.052
 unity-0.049
 thanks-0.049
Token she
Token ID673
Feature activation-8.06e-05

Pos logit contributions
angers+0.055
 salon+0.043
ooter+0.042
ulk+0.041
aj+0.041
Neg logit contributions
oc-0.046
 course-0.042
 deer-0.040
 thanks-0.037
 unity-0.037
Token wouldn
Token ID3636
Feature activation+7.17e-02

Pos logit contributions
oc+0.000
 course+0.000
 deer+0.000
 unity+0.000
 harmony+0.000
Neg logit contributions
angers-0.000
 salon-0.000
belt-0.000
ooter-0.000
outed-0.000
Tokenâ
Token ID22940
Feature activation+2.02e-02

Pos logit contributions
angers+0.155
aj+0.122
outed+0.121
 salon+0.117
elt+0.115
Neg logit contributions
oc-0.129
 course-0.118
 thanks-0.108
 deer-0.108
 love-0.100
Token
Token ID26391
Feature activation+1.53e-01

Pos logit contributions
angers+0.040
aj+0.029
ews+0.027
 salon+0.026
unes+0.026
Neg logit contributions
 course-0.042
oc-0.042
 matter-0.038
 thanks-0.037
 unity-0.035
Token
Token ID8151
Feature activation+4.31e-02

Pos logit contributions
angers+0.259
outed+0.186
 salon+0.183
ooter+0.181
ulk+0.180
Neg logit contributions
oc-0.350
 course-0.302
 thanks-0.288
 love-0.277
 harmony-0.274
Tokent
Token ID83
Feature activation-2.46e-02

Pos logit contributions
angers+0.082
ulk+0.061
azes+0.059
aj+0.058
 salon+0.058
Neg logit contributions
oc-0.087
 course-0.080
 thanks-0.072
affe-0.070
 deer-0.069
Token have
Token ID423
Feature activation-2.83e-02

Pos logit contributions
oc+0.044
 course+0.039
 deer+0.036
 unity+0.034
 thanks+0.034
Neg logit contributions
angers-0.054
 salon-0.045
ooter-0.044
ulk-0.042
belt-0.041
Token to
Token ID284
Feature activation-1.88e-02

Pos logit contributions
oc+0.057
 course+0.055
 deer+0.051
 thanks+0.050
 unity+0.049
Neg logit contributions
angers-0.055
 salon-0.041
ooter-0.040
aj-0.039
belt-0.039
Token m
Token ID285
Feature activation+7.09e-02

Pos logit contributions
oc+0.035
 course+0.030
 deer+0.029
 unity+0.027
affe+0.027
Neg logit contributions
angers-0.039
 salon-0.033
ooter-0.031
ulk-0.031
 bathroom-0.030
Token different
Token ID1180
Feature activation-7.96e-02

Pos logit contributions
oc+0.080
 deer+0.061
 course+0.060
affe+0.059
 tiger+0.059
Neg logit contributions
angers-0.073
outed-0.059
elt-0.058
 salon-0.058
 bathroom-0.057
Token tastes
Token ID18221
Feature activation+1.72e-02

Pos logit contributions
oc+0.142
 course+0.123
 deer+0.120
 tiger+0.109
 unity+0.109
Neg logit contributions
angers-0.172
outed-0.134
ooter-0.132
aj-0.132
elt-0.132
Token.
Token ID13
Feature activation+1.52e-02

Pos logit contributions
angers+0.031
unes+0.024
boards+0.024
elt+0.024
outed+0.023
Neg logit contributions
oc-0.036
 course-0.029
 deer-0.029
affe-0.028
 tiger-0.028
Token But
Token ID887
Feature activation-1.73e-02

Pos logit contributions
angers+0.027
 salon+0.020
ulk+0.020
omm+0.019
elt+0.019
Neg logit contributions
oc-0.031
 deer-0.030
 course-0.029
 unity-0.028
 matter-0.027
Token most
Token ID749
Feature activation-9.00e-02

Pos logit contributions
oc+0.031
 deer+0.027
 course+0.026
affe+0.025
 unity+0.024
Neg logit contributions
angers-0.036
 salon-0.031
ulk-0.028
outed-0.028
ooter-0.028
Token importantly
Token ID11003
Feature activation-1.48e-01

Pos logit contributions
oc+0.152
 course+0.132
 deer+0.128
osaurs+0.123
bit+0.122
Neg logit contributions
angers-0.190
outed-0.154
elt-0.150
azes-0.150
uddled-0.148
Token,
Token ID11
Feature activation+4.79e-02

Pos logit contributions
oc+0.331
 deer+0.288
 course+0.281
 tiger+0.269
 unity+0.269
Neg logit contributions
angers-0.258
 salon-0.197
aj-0.183
outed-0.181
anger-0.179
Token she
Token ID673
Feature activation-1.78e-02

Pos logit contributions
angers+0.082
 salon+0.074
 bathroom+0.061
ulk+0.060
elt+0.057
Neg logit contributions
oc-0.107
affe-0.089
 deer-0.088
 warrior-0.083
bit-0.082
Token learned
Token ID4499
Feature activation-2.15e-02

Pos logit contributions
oc+0.033
 course+0.030
 deer+0.029
 unity+0.028
 harmony+0.027
Neg logit contributions
angers-0.036
 salon-0.029
outed-0.028
aj-0.027
ooter-0.027
Token to
Token ID284
Feature activation-3.08e-02

Pos logit contributions
oc+0.032
 course+0.029
 deer+0.027
 thanks+0.026
 unity+0.025
Neg logit contributions
angers-0.053
 salon-0.043
ooter-0.042
aj-0.042
ulk-0.042
Token love
Token ID1842
Feature activation-5.47e-02

Pos logit contributions
oc+0.060
 deer+0.053
 turkey+0.049
 course+0.049
affe+0.048
Neg logit contributions
angers-0.057
 salon-0.048
ooter-0.047
 bathroom-0.045
OM-0.044
Token
Token ID220
Feature activation+1.17e-01

Pos logit contributions
angers+0.033
amas+0.024
aj+0.024
omm+0.023
ulk+0.023
Neg logit contributions
oc-0.036
 course-0.034
 matter-0.032
 deer-0.032
 unity-0.031
Token
Token ID198
Feature activation-3.22e-02

Pos logit contributions
angers+0.271
 salon+0.225
ooter+0.216
belt+0.215
ulk+0.211
Neg logit contributions
oc-0.188
 course-0.164
 deer-0.153
 thanks-0.152
 unity-0.144
Token
Token ID198
Feature activation-3.76e-02

Pos logit contributions
oc+0.069
 course+0.065
 deer+0.063
 matter+0.062
 thanks+0.061
Neg logit contributions
angers-0.056
ulk-0.039
aj-0.039
 salon-0.038
ews-0.037
TokenBut
Token ID1537
Feature activation-4.94e-03

Pos logit contributions
oc+0.072
 thanks+0.067
 course+0.063
 tiger+0.063
affe+0.063
Neg logit contributions
angers-0.067
ulk-0.053
 salon-0.052
outed-0.049
uddled-0.049
Token most
Token ID749
Feature activation-8.79e-02

Pos logit contributions
oc+0.010
affe+0.009
 course+0.008
 sorts+0.008
 tiger+0.008
Neg logit contributions
angers-0.009
 salon-0.007
belt-0.007
outed-0.007
 bathroom-0.007
Token importantly
Token ID11003
Feature activation-1.48e-01

Pos logit contributions
oc+0.158
bit+0.144
osaurs+0.143
 course+0.142
 thanks+0.128
Neg logit contributions
angers-0.161
outed-0.138
elt-0.129
uddled-0.128
itter-0.125
Token,
Token ID11
Feature activation+4.70e-02

Pos logit contributions
oc+0.362
 course+0.304
 deer+0.300
 tiger+0.293
affe+0.291
Neg logit contributions
angers-0.221
 salon-0.164
outed-0.155
belt-0.150
aj-0.150
Token they
Token ID484
Feature activation-4.59e-02

Pos logit contributions
angers+0.072
 bathroom+0.058
 salon+0.055
belt+0.052
tub+0.052
Neg logit contributions
oc-0.102
affe-0.100
 course-0.086
bit-0.086
osaurs-0.082
Token discussed
Token ID6693
Feature activation+4.25e-03

Pos logit contributions
oc+0.103
 course+0.084
 tiger+0.080
 harmony+0.080
 unity+0.079
Neg logit contributions
angers-0.074
 salon-0.063
outed-0.060
line-0.055
ews-0.055
Token what
Token ID644
Feature activation+8.13e-03

Pos logit contributions
angers+0.008
 salon+0.007
ews+0.006
 bathroom+0.006
 office+0.006
Neg logit contributions
oc-0.008
affe-0.007
 course-0.007
 deer-0.006
 warrior-0.006
Token they
Token ID484
Feature activation-3.89e-02

Pos logit contributions
angers+0.014
 salon+0.013
 bathroom+0.012
ews+0.012
 office+0.012
Neg logit contributions
oc-0.017
 course-0.013
affe-0.013
 deer-0.012
bit-0.012
Token need
Token ID761
Feature activation-5.97e-02

Pos logit contributions
oc+0.018
 course+0.016
 deer+0.015
 unity+0.014
 thanks+0.014
Neg logit contributions
angers-0.028
 salon-0.023
ooter-0.022
aj-0.022
ulk-0.022
Token it
Token ID340
Feature activation+9.04e-02

Pos logit contributions
oc+0.115
 course+0.094
 deer+0.090
affe+0.088
 sorts+0.086
Neg logit contributions
angers-0.123
ulk-0.096
 salon-0.096
 bathroom-0.095
aj-0.093
Token.
Token ID13
Feature activation+1.41e-02

Pos logit contributions
angers+0.171
 salon+0.126
aj+0.125
ooter+0.124
amas+0.122
Neg logit contributions
oc-0.189
 course-0.172
 deer-0.163
 unity-0.156
 thanks-0.153
Token And
Token ID843
Feature activation-6.83e-02

Pos logit contributions
angers+0.028
 salon+0.021
ulk+0.020
aj+0.020
ooter+0.020
Neg logit contributions
oc-0.028
 course-0.026
 deer-0.025
 unity-0.024
 matter-0.023
Token most
Token ID749
Feature activation-8.79e-02

Pos logit contributions
oc+0.152
 course+0.131
 deer+0.131
affe+0.128
 tiger+0.126
Neg logit contributions
angers-0.117
 salon-0.097
azes-0.082
ulk-0.080
 office-0.080
Token importantly
Token ID11003
Feature activation-1.47e-01

Pos logit contributions
oc+0.155
 course+0.136
bit+0.130
osaurs+0.130
 deer+0.125
Neg logit contributions
angers-0.176
uddled-0.138
elt-0.138
outed-0.134
ider-0.134
Token,
Token ID11
Feature activation+5.03e-02

Pos logit contributions
oc+0.328
 deer+0.284
 course+0.283
 unity+0.271
 tiger+0.266
Neg logit contributions
angers-0.260
 salon-0.193
aj-0.184
anger-0.180
outed-0.177
Token she
Token ID673
Feature activation-1.86e-02

Pos logit contributions
angers+0.086
 salon+0.073
 bathroom+0.063
belt+0.060
 office+0.059
Neg logit contributions
oc-0.114
affe-0.096
 deer-0.095
 course-0.094
bit-0.089
Token learned
Token ID4499
Feature activation-2.00e-02

Pos logit contributions
oc+0.034
 course+0.031
 deer+0.029
 unity+0.029
 harmony+0.028
Neg logit contributions
angers-0.038
 salon-0.029
aj-0.028
outed-0.028
ooter-0.028
Token that
Token ID326
Feature activation-2.34e-02

Pos logit contributions
oc+0.036
 course+0.032
 deer+0.030
 unity+0.028
 thanks+0.028
Neg logit contributions
angers-0.044
 salon-0.035
ooter-0.034
ulk-0.033
aj-0.033
Token a
Token ID257
Feature activation-4.67e-02

Pos logit contributions
oc+0.047
affe+0.039
 course+0.039
 deer+0.036
 tiger+0.034
Neg logit contributions
angers-0.044
 salon-0.038
 bathroom-0.035
 office-0.034
ulk-0.033
Token before
Token ID878
Feature activation+6.40e-02

Pos logit contributions
angers+0.020
aj+0.015
outed+0.015
ooter+0.015
ulk+0.015
Neg logit contributions
oc-0.017
 course-0.016
 thanks-0.015
 deer-0.014
 matter-0.014
Token.
Token ID13
Feature activation+1.90e-02

Pos logit contributions
angers+0.143
 salon+0.112
ooter+0.111
aj+0.110
belt+0.108
Neg logit contributions
oc-0.109
 course-0.104
 deer-0.097
 thanks-0.096
 unity-0.090
Token But
Token ID887
Feature activation-1.18e-02

Pos logit contributions
angers+0.039
 salon+0.030
ooter+0.029
aj+0.029
ulk+0.028
Neg logit contributions
oc-0.037
 course-0.034
 deer-0.032
 thanks-0.031
 unity-0.030
Token,
Token ID11
Feature activation+5.69e-02

Pos logit contributions
oc+0.025
 course+0.021
 deer+0.021
affe+0.020
 tiger+0.019
Neg logit contributions
angers-0.022
 salon-0.018
belt-0.017
outed-0.017
 bathroom-0.016
Token most
Token ID749
Feature activation-8.78e-02

Pos logit contributions
angers+0.096
 salon+0.073
belt+0.068
 office+0.066
 bathroom+0.066
Neg logit contributions
oc-0.132
 deer-0.110
 course-0.110
affe-0.109
 tiger-0.107
Token importantly
Token ID11003
Feature activation-1.45e-01

Pos logit contributions
oc+0.150
 course+0.128
osaurs+0.118
bit+0.117
 deer+0.115
Neg logit contributions
angers-0.192
elt-0.154
outed-0.153
 salon-0.150
azes-0.149
Token,
Token ID11
Feature activation+5.07e-02

Pos logit contributions
oc+0.337
 deer+0.281
 course+0.280
 tiger+0.269
 unity+0.268
Neg logit contributions
angers-0.256
 salon-0.192
anger-0.179
outed-0.178
aj-0.176
Token she
Token ID673
Feature activation-1.52e-02

Pos logit contributions
angers+0.074
 salon+0.054
 bathroom+0.051
 office+0.049
ulk+0.048
Neg logit contributions
oc-0.132
affe-0.109
 deer-0.106
 course-0.104
 tiger-0.104
Token could
Token ID714
Feature activation+2.01e-02

Pos logit contributions
oc+0.030
 course+0.027
 deer+0.026
 unity+0.025
 harmony+0.025
Neg logit contributions
angers-0.029
 salon-0.023
outed-0.022
aj-0.022
belt-0.022
Token feel
Token ID1254
Feature activation+1.23e-02

Pos logit contributions
angers+0.041
 salon+0.034
ooter+0.033
belt+0.031
ulk+0.031
Neg logit contributions
oc-0.038
 course-0.033
 deer-0.031
 unity-0.030
 tiger-0.029
Token the
Token ID262
Feature activation+3.94e-02

Pos logit contributions
angers+0.028
 salon+0.023
ulk+0.022
ews+0.022
outed+0.021
Neg logit contributions
oc-0.021
 course-0.017
 deer-0.016
affe-0.016
 tiger-0.015
Token of
Token ID286
Feature activation-8.92e-02

Pos logit contributions
oc+0.101
 course+0.088
 deer+0.087
 unity+0.080
 harmony+0.079
Neg logit contributions
angers-0.127
ooter-0.099
 salon-0.098
anger-0.097
ulk-0.095
Token things
Token ID1243
Feature activation+2.18e-02

Pos logit contributions
oc+0.164
affe+0.130
 deer+0.129
 course+0.122
Fin+0.113
Neg logit contributions
angers-0.177
ulk-0.150
 bathroom-0.147
 salon-0.147
ooter-0.141
Token.
Token ID13
Feature activation+1.68e-02

Pos logit contributions
angers+0.042
 salon+0.032
ooter+0.032
belt+0.031
anger+0.030
Neg logit contributions
oc-0.045
 course-0.039
 deer-0.038
 unity-0.036
 harmony-0.036
Token But
Token ID887
Feature activation-1.27e-02

Pos logit contributions
angers+0.031
ooter+0.023
ulk+0.023
aj+0.023
 salon+0.023
Neg logit contributions
oc-0.033
 course-0.032
 deer-0.031
 unity-0.030
 matter-0.030
Token most
Token ID749
Feature activation-8.72e-02

Pos logit contributions
oc+0.022
 course+0.020
 deer+0.020
 sorts+0.019
affe+0.019
Neg logit contributions
angers-0.025
 salon-0.021
belt-0.021
ulk-0.019
 office-0.018
Token importantly
Token ID11003
Feature activation-1.45e-01

Pos logit contributions
oc+0.145
 deer+0.127
 course+0.125
bit+0.124
 harmony+0.120
Neg logit contributions
angers-0.177
outed-0.142
azes-0.139
elt-0.137
ooter-0.137
Token of
Token ID286
Feature activation-1.02e-01

Pos logit contributions
oc+0.342
 deer+0.297
 course+0.288
 unity+0.282
 harmony+0.281
Neg logit contributions
angers-0.229
 salon-0.166
belt-0.162
anger-0.156
aj-0.154
Token all
Token ID477
Feature activation-2.01e-02

Pos logit contributions
oc+0.231
affe+0.185
 deer+0.174
Blue+0.158
bit+0.157
Neg logit contributions
angers-0.168
azes-0.137
ulk-0.137
 salon-0.137
 bathroom-0.135
Token,
Token ID11
Feature activation+5.34e-02

Pos logit contributions
oc+0.041
 deer+0.036
Blue+0.034
Mic+0.034
 unity+0.034
Neg logit contributions
angers-0.032
 salon-0.025
azes-0.025
 office-0.025
 bathroom-0.024
Token he
Token ID339
Feature activation-1.71e-02

Pos logit contributions
angers+0.097
 salon+0.076
 bathroom+0.070
belt+0.070
 office+0.068
Neg logit contributions
oc-0.105
 course-0.090
affe-0.089
 deer-0.086
 sorts-0.085
Token had
Token ID550
Feature activation-2.28e-02

Pos logit contributions
oc+0.035
 course+0.031
 deer+0.030
 unity+0.030
 harmony+0.029
Neg logit contributions
angers-0.031
 salon-0.025
outed-0.023
belt-0.023
ooter-0.023
Token of
Token ID286
Feature activation-8.92e-02

Pos logit contributions
oc+0.101
 course+0.088
 deer+0.087
 unity+0.080
 harmony+0.079
Neg logit contributions
angers-0.127
ooter-0.099
 salon-0.098
anger-0.097
ulk-0.095
Token things
Token ID1243
Feature activation+2.18e-02

Pos logit contributions
oc+0.164
affe+0.130
 deer+0.129
 course+0.122
Fin+0.113
Neg logit contributions
angers-0.177
ulk-0.150
 bathroom-0.147
 salon-0.147
ooter-0.141
Token.
Token ID13
Feature activation+1.68e-02

Pos logit contributions
angers+0.042
 salon+0.032
ooter+0.032
belt+0.031
anger+0.030
Neg logit contributions
oc-0.045
 course-0.039
 deer-0.038
 unity-0.036
 harmony-0.036
Token But
Token ID887
Feature activation-1.27e-02

Pos logit contributions
angers+0.031
ooter+0.023
ulk+0.023
aj+0.023
 salon+0.023
Neg logit contributions
oc-0.033
 course-0.032
 deer-0.031
 unity-0.030
 matter-0.030
Token most
Token ID749
Feature activation-8.72e-02

Pos logit contributions
oc+0.022
 course+0.020
 deer+0.020
 sorts+0.019
affe+0.019
Neg logit contributions
angers-0.025
 salon-0.021
belt-0.021
ulk-0.019
 office-0.018
Token importantly
Token ID11003
Feature activation-1.45e-01

Pos logit contributions
oc+0.145
 deer+0.127
 course+0.125
bit+0.124
 harmony+0.120
Neg logit contributions
angers-0.177
outed-0.142
azes-0.139
elt-0.137
ooter-0.137
Token of
Token ID286
Feature activation-1.02e-01

Pos logit contributions
oc+0.342
 deer+0.297
 course+0.288
 unity+0.282
 harmony+0.281
Neg logit contributions
angers-0.229
 salon-0.166
belt-0.162
anger-0.156
aj-0.154
Token all
Token ID477
Feature activation-2.01e-02

Pos logit contributions
oc+0.231
affe+0.185
 deer+0.174
Blue+0.158
bit+0.157
Neg logit contributions
angers-0.168
azes-0.137
ulk-0.137
 salon-0.137
 bathroom-0.135
Token,
Token ID11
Feature activation+5.34e-02

Pos logit contributions
oc+0.041
 deer+0.036
Blue+0.034
Mic+0.034
 unity+0.034
Neg logit contributions
angers-0.032
 salon-0.025
azes-0.025
 office-0.025
 bathroom-0.024
Token he
Token ID339
Feature activation-1.71e-02

Pos logit contributions
angers+0.097
 salon+0.076
 bathroom+0.070
belt+0.070
 office+0.068
Neg logit contributions
oc-0.105
 course-0.090
affe-0.089
 deer-0.086
 sorts-0.085
Token had
Token ID550
Feature activation-2.28e-02

Pos logit contributions
oc+0.035
 course+0.031
 deer+0.030
 unity+0.030
 harmony+0.029
Neg logit contributions
angers-0.031
 salon-0.025
outed-0.023
belt-0.023
ooter-0.023
Token,
Token ID11
Feature activation+5.33e-02

Pos logit contributions
oc+0.089
 course+0.076
 deer+0.075
 unity+0.072
 harmony+0.069
Neg logit contributions
angers-0.092
 salon-0.073
ooter-0.071
belt-0.068
anger-0.068
Token treats
Token ID18432
Feature activation-5.28e-02

Pos logit contributions
angers+0.088
 salon+0.070
aj+0.065
ooter+0.064
belt+0.063
Neg logit contributions
oc-0.121
 unity-0.097
 course-0.097
affe-0.094
 sorts-0.093
Token,
Token ID11
Feature activation+5.40e-02

Pos logit contributions
oc+0.106
 course+0.091
 deer+0.089
 unity+0.088
 harmony+0.085
Neg logit contributions
angers-0.105
 salon-0.081
ooter-0.078
belt-0.077
anger-0.077
Token and
Token ID290
Feature activation-3.39e-02

Pos logit contributions
angers+0.093
 salon+0.074
aj+0.071
ooter+0.068
elt+0.066
Neg logit contributions
oc-0.119
 unity-0.097
 course-0.096
 sorts-0.095
 harmony-0.094
Token most
Token ID749
Feature activation-8.88e-02

Pos logit contributions
oc+0.075
affe+0.059
 course+0.059
 deer+0.054
 unity+0.053
Neg logit contributions
angers-0.056
 salon-0.049
 bathroom-0.046
belt-0.042
 office-0.042
Token importantly
Token ID11003
Feature activation-1.44e-01

Pos logit contributions
oc+0.169
 course+0.152
osaurs+0.138
 Gary+0.136
bit+0.134
Neg logit contributions
angers-0.170
uddled-0.130
aj-0.130
ooter-0.128
elt-0.126
Token,
Token ID11
Feature activation+4.64e-02

Pos logit contributions
oc+0.289
 course+0.259
 deer+0.247
 unity+0.236
 thanks+0.231
Neg logit contributions
angers-0.290
 salon-0.219
aj-0.211
ooter-0.210
anger-0.209
Token Lily
Token ID20037
Feature activation-5.80e-03

Pos logit contributions
angers+0.083
 salon+0.064
 bathroom+0.058
aj+0.058
belt+0.055
Neg logit contributions
oc-0.107
 course-0.084
affe-0.083
 deer-0.082
 unity-0.082
Token's
Token ID338
Feature activation+3.15e-02

Pos logit contributions
oc+0.011
 course+0.010
 deer+0.010
 unity+0.010
 harmony+0.009
Neg logit contributions
angers-0.012
 salon-0.009
ooter-0.009
belt-0.009
aj-0.008
Token love
Token ID1842
Feature activation-5.98e-02

Pos logit contributions
angers+0.058
 salon+0.048
aj+0.047
ooter+0.046
 bathroom+0.046
Neg logit contributions
oc-0.065
 course-0.055
 sorts-0.050
affe-0.050
 deer-0.049
Token.
Token ID13
Feature activation+1.53e-02

Pos logit contributions
oc+0.135
 course+0.115
 unity+0.110
 deer+0.109
 sorts+0.105
Neg logit contributions
angers-0.104
 salon-0.078
belt-0.074
ooter-0.073
elt-0.070
Token beautiful
Token ID4950
Feature activation-9.05e-02

Pos logit contributions
angers+0.041
 salon+0.034
outed+0.033
 office+0.033
 bathroom+0.032
Neg logit contributions
oc-0.052
 matter-0.039
affe-0.038
 course-0.038
Mic-0.037
Token things
Token ID1243
Feature activation+2.09e-02

Pos logit contributions
oc+0.183
 course+0.162
 deer+0.147
affe+0.145
 unity+0.144
Neg logit contributions
angers-0.171
 salon-0.130
aj-0.127
ews-0.126
ulk-0.125
Token and
Token ID290
Feature activation-3.17e-02

Pos logit contributions
angers+0.040
 salon+0.031
ooter+0.031
line+0.030
elt+0.030
Neg logit contributions
oc-0.044
 course-0.035
 deer-0.034
affe-0.033
 unity-0.033
Token,
Token ID11
Feature activation+5.33e-02

Pos logit contributions
oc+0.065
 course+0.057
affe+0.050
 deer+0.050
 tiger+0.048
Neg logit contributions
angers-0.060
 salon-0.047
outed-0.046
ews-0.045
belt-0.045
Token most
Token ID749
Feature activation-8.57e-02

Pos logit contributions
angers+0.096
 salon+0.077
line+0.075
 bathroom+0.068
belt+0.068
Neg logit contributions
oc-0.117
 course-0.095
affe-0.094
 deer-0.088
Mic-0.085
Token importantly
Token ID11003
Feature activation-1.43e-01

Pos logit contributions
oc+0.149
 course+0.133
osaurs+0.118
 Gary+0.117
 Freddie+0.116
Neg logit contributions
angers-0.183
uddled-0.143
elt-0.143
aj-0.142
outed-0.142
Token,
Token ID11
Feature activation+4.74e-02

Pos logit contributions
oc+0.305
 course+0.263
 deer+0.254
 tiger+0.239
 unity+0.238
Neg logit contributions
angers-0.271
 salon-0.204
outed-0.195
anger-0.194
aj-0.194
Token they
Token ID484
Feature activation-4.68e-02

Pos logit contributions
angers+0.086
 salon+0.067
line+0.065
 bathroom+0.063
 office+0.062
Neg logit contributions
oc-0.104
affe-0.084
 course-0.083
Mic-0.076
 deer-0.076
Token had
Token ID550
Feature activation-2.21e-02

Pos logit contributions
oc+0.106
 course+0.088
 harmony+0.082
 tiger+0.081
 unity+0.081
Neg logit contributions
angers-0.078
 salon-0.061
outed-0.060
line-0.059
belt-0.058
Token a
Token ID257
Feature activation-4.38e-02

Pos logit contributions
oc+0.044
 course+0.037
 deer+0.036
affe+0.033
 tiger+0.033
Neg logit contributions
angers-0.045
 salon-0.035
outed-0.034
ulk-0.034
ooter-0.034
Token lot
Token ID1256
Feature activation-6.85e-02

Pos logit contributions
oc+0.077
 course+0.064
affe+0.061
bit+0.055
 Freddie+0.055
Neg logit contributions
angers-0.081
 salon-0.078
 bathroom-0.076
ooter-0.072
 office-0.071
Token.
Token ID13
Feature activation+1.69e-02

Pos logit contributions
angers+0.029
 salon+0.024
ooter+0.023
ulk+0.023
azes+0.023
Neg logit contributions
oc-0.018
 course-0.016
 deer-0.016
 thanks-0.015
 unity-0.014
Token It
Token ID632
Feature activation-1.72e-03

Pos logit contributions
angers+0.034
aj+0.026
 salon+0.026
ooter+0.025
elt+0.025
Neg logit contributions
oc-0.032
 course-0.030
 deer-0.028
 unity-0.027
 sorts-0.026
Token had
Token ID550
Feature activation-2.94e-02

Pos logit contributions
oc+0.004
 course+0.003
 unity+0.003
 deer+0.003
 harmony+0.003
Neg logit contributions
angers-0.003
 salon-0.002
outed-0.002
ooter-0.002
belt-0.002
Token her
Token ID607
Feature activation+4.53e-02

Pos logit contributions
oc+0.060
 course+0.051
 deer+0.047
 unity+0.046
 sorts+0.045
Neg logit contributions
angers-0.059
 salon-0.047
ooter-0.045
aj-0.044
belt-0.043
Token most
Token ID749
Feature activation-8.19e-02

Pos logit contributions
angers+0.090
 salon+0.073
ooter+0.072
aj+0.072
belt+0.070
Neg logit contributions
oc-0.087
 course-0.079
 deer-0.071
 sorts-0.071
 unity-0.070
Token precious
Token ID14186
Feature activation-1.42e-01

Pos logit contributions
oc+0.130
 course+0.120
 deer+0.108
 unity+0.103
 Gary+0.102
Neg logit contributions
angers-0.188
aj-0.154
ooter-0.150
 salon-0.150
belt-0.146
Token memories
Token ID9846
Feature activation-4.74e-02

Pos logit contributions
oc+0.239
 course+0.215
 unity+0.190
 deer+0.186
 harmony+0.184
Neg logit contributions
angers-0.312
ooter-0.250
anger-0.243
 salon-0.243
belt-0.243
Token and
Token ID290
Feature activation-4.25e-02

Pos logit contributions
oc+0.091
 course+0.077
 deer+0.075
 unity+0.071
 harmony+0.068
Neg logit contributions
angers-0.099
 salon-0.079
ooter-0.077
belt-0.075
aj-0.074
Token gifts
Token ID13201
Feature activation-7.79e-02

Pos logit contributions
oc+0.083
 course+0.067
affe+0.064
 deer+0.061
Fred+0.058
Neg logit contributions
angers-0.085
 salon-0.071
 bathroom-0.068
belt-0.067
aj-0.066
Token.
Token ID13
Feature activation+1.57e-02

Pos logit contributions
oc+0.156
 course+0.139
 deer+0.131
 unity+0.127
 harmony+0.123
Neg logit contributions
angers-0.157
 salon-0.121
ooter-0.118
aj-0.115
belt-0.115
Token You
Token ID921
Feature activation-7.76e-02

Pos logit contributions
angers+0.032
aj+0.024
 salon+0.024
ooter+0.024
elt+0.024
Neg logit contributions
oc-0.029
 course-0.028
 deer-0.026
 unity-0.025
 matter-0.024
Token dance
Token ID9280
Feature activation-9.91e-03

Pos logit contributions
oc+0.212
 course+0.193
 deer+0.175
 unity+0.174
 kindness+0.171
Neg logit contributions
angers-0.192
 salon-0.147
amas-0.141
 bathroom-0.140
ews-0.137
Token.
Token ID13
Feature activation+8.61e-03

Pos logit contributions
oc+0.021
 course+0.019
 deer+0.018
 unity+0.017
 thanks+0.017
Neg logit contributions
angers-0.019
 salon-0.014
ooter-0.014
anger-0.014
ulk-0.013
Token
Token ID198
Feature activation-3.96e-02

Pos logit contributions
angers+0.016
 salon+0.012
ooter+0.011
ulk+0.011
elt+0.011
Neg logit contributions
oc-0.018
 course-0.016
 deer-0.016
 unity-0.015
 thanks-0.015
Token
Token ID198
Feature activation-4.51e-02

Pos logit contributions
oc+0.078
 course+0.072
 deer+0.069
 thanks+0.067
 unity+0.067
Neg logit contributions
angers-0.077
 salon-0.058
ulk-0.056
ooter-0.056
aj-0.055
TokenThis
Token ID1212
Feature activation-4.29e-02

Pos logit contributions
oc+0.096
 thanks+0.089
 kindness+0.085
 deer+0.083
 course+0.082
Neg logit contributions
angers-0.074
 salon-0.059
OM-0.052
ulk-0.052
ews-0.051
Token playful
Token ID34264
Feature activation-1.41e-01

Pos logit contributions
oc+0.070
 course+0.062
 deer+0.058
 thanks+0.054
 unity+0.054
Neg logit contributions
angers-0.101
 salon-0.082
ooter-0.079
ulk-0.078
aj-0.078
Token team
Token ID1074
Feature activation-7.26e-02

Pos logit contributions
oc+0.199
 course+0.174
 deer+0.154
 thanks+0.146
 unity+0.144
Neg logit contributions
angers-0.366
 salon-0.299
ooter-0.293
aj-0.286
anger-0.285
Token high
Token ID1029
Feature activation-6.61e-02

Pos logit contributions
oc+0.145
 course+0.127
 deer+0.122
 unity+0.120
 thanks+0.116
Neg logit contributions
angers-0.137
 salon-0.112
ooter-0.105
outed-0.101
belt-0.099
Token-
Token ID12
Feature activation+2.65e-02

Pos logit contributions
oc+0.110
 course+0.101
 deer+0.096
 thanks+0.092
 unity+0.088
Neg logit contributions
angers-0.152
 salon-0.122
ulk-0.119
ooter-0.118
aj-0.115
Tokenf
Token ID69
Feature activation-1.66e-02

Pos logit contributions
angers+0.045
 salon+0.032
aj+0.031
ooter+0.031
ulk+0.031
Neg logit contributions
oc-0.061
 course-0.057
 deer-0.054
 thanks-0.052
 unity-0.052
Tokenived
Token ID1572
Feature activation-1.99e-02

Pos logit contributions
oc+0.034
 course+0.033
 thanks+0.032
 deer+0.031
affe+0.029
Neg logit contributions
angers-0.029
 salon-0.022
belt-0.022
ulk-0.021
ooter-0.021
Token something
Token ID1223
Feature activation+3.54e-02

Pos logit contributions
oc+0.017
 deer+0.015
 course+0.014
 unity+0.013
 tiger+0.013
Neg logit contributions
angers-0.020
 salon-0.016
ulk-0.016
ooter-0.015
belt-0.015
Token difficult
Token ID2408
Feature activation-5.75e-02

Pos logit contributions
angers+0.074
 salon+0.061
ooter+0.058
belt+0.057
aj+0.056
Neg logit contributions
oc-0.066
 deer-0.056
 course-0.055
 tiger-0.051
affe-0.050
Token.
Token ID13
Feature activation+2.54e-02

Pos logit contributions
oc+0.119
 deer+0.106
 course+0.103
 unity+0.097
 tiger+0.096
Neg logit contributions
angers-0.110
 salon-0.083
ooter-0.080
ulk-0.079
anger-0.079
Token But
Token ID887
Feature activation-1.03e-02

Pos logit contributions
angers+0.051
 salon+0.039
ooter+0.038
ulk+0.038
aj+0.037
Neg logit contributions
oc-0.050
 course-0.045
 deer-0.043
 thanks-0.041
 unity-0.041
Token more
Token ID517
Feature activation-2.80e-02

Pos logit contributions
oc+0.020
 deer+0.017
 course+0.017
affe+0.015
 unity+0.015
Neg logit contributions
angers-0.021
 salon-0.018
belt-0.016
ulk-0.016
ooter-0.015
Token importantly
Token ID11003
Feature activation-1.40e-01

Pos logit contributions
oc+0.054
 course+0.045
 deer+0.045
affe+0.042
 tiger+0.040
Neg logit contributions
angers-0.057
 salon-0.045
azes-0.044
ulk-0.043
ews-0.042
Token,
Token ID11
Feature activation+5.77e-02

Pos logit contributions
oc+0.325
 deer+0.284
 course+0.266
 tiger+0.263
 unity+0.260
Neg logit contributions
angers-0.240
 salon-0.187
belt-0.169
anger-0.166
aj-0.166
Token she
Token ID673
Feature activation-9.85e-03

Pos logit contributions
angers+0.098
 salon+0.082
ulk+0.072
 office+0.071
belt+0.071
Neg logit contributions
oc-0.130
 deer-0.113
 course-0.108
affe-0.103
 tiger-0.103
Token proved
Token ID8302
Feature activation+8.78e-03

Pos logit contributions
oc+0.019
 deer+0.017
 course+0.017
 unity+0.016
 harmony+0.016
Neg logit contributions
angers-0.019
 salon-0.015
outed-0.014
ooter-0.014
belt-0.014
Token to
Token ID284
Feature activation-2.71e-02

Pos logit contributions
angers+0.019
 salon+0.015
ooter+0.014
ulk+0.014
aj+0.014
Neg logit contributions
oc-0.017
 course-0.015
 deer-0.014
 unity-0.013
 thanks-0.013
Token herself
Token ID5223
Feature activation+4.48e-02

Pos logit contributions
oc+0.056
 deer+0.046
 course+0.044
 tiger+0.042
affe+0.042
Neg logit contributions
angers-0.052
 salon-0.043
ooter-0.041
ulk-0.040
belt-0.040
Token see
Token ID766
Feature activation+3.05e-02

Pos logit contributions
oc+0.042
 unity+0.036
 course+0.036
 sorts+0.035
 thanks+0.035
Neg logit contributions
angers-0.047
 salon-0.039
ooter-0.039
belt-0.037
anger-0.037
Token.
Token ID13
Feature activation+1.88e-02

Pos logit contributions
angers+0.055
belt+0.040
 salon+0.040
ooter+0.040
outed+0.040
Neg logit contributions
oc-0.059
 course-0.053
 unity-0.053
 bond-0.052
 deer-0.052
Token
Token ID198
Feature activation-3.49e-02

Pos logit contributions
angers+0.045
 salon+0.036
ooter+0.035
ulk+0.034
aj+0.034
Neg logit contributions
oc-0.031
 course-0.027
 deer-0.025
 thanks-0.024
 unity-0.024
Token
Token ID198
Feature activation-4.08e-02

Pos logit contributions
oc+0.065
 course+0.060
 deer+0.057
 thanks+0.055
 unity+0.054
Neg logit contributions
angers-0.073
 salon-0.056
ooter-0.055
ulk-0.054
aj-0.054
TokenThe
Token ID464
Feature activation-2.50e-02

Pos logit contributions
oc+0.071
 thanks+0.069
 course+0.068
 deer+0.067
 unity+0.064
Neg logit contributions
angers-0.082
ulk-0.063
amas-0.063
belt-0.063
ooter-0.062
Token playful
Token ID34264
Feature activation-1.40e-01

Pos logit contributions
oc+0.052
 sorts+0.044
 course+0.041
 unity+0.038
 thanks+0.038
Neg logit contributions
angers-0.045
ooter-0.037
 salon-0.037
 bathroom-0.037
amas-0.035
Token puppy
Token ID26188
Feature activation-6.44e-02

Pos logit contributions
oc+0.211
 course+0.181
 sorts+0.177
 unity+0.160
 thanks+0.155
Neg logit contributions
angers-0.312
ooter-0.270
amas-0.264
 salon-0.263
elt-0.257
Token learned
Token ID4499
Feature activation-2.22e-02

Pos logit contributions
oc+0.135
 course+0.127
 unity+0.122
 sorts+0.120
 harmony+0.120
Neg logit contributions
angers-0.106
 salon-0.084
ooter-0.078
outed-0.077
belt-0.075
Token that
Token ID326
Feature activation-2.45e-02

Pos logit contributions
oc+0.037
 course+0.032
 deer+0.030
 unity+0.028
 thanks+0.028
Neg logit contributions
angers-0.051
ulk-0.041
 salon-0.041
ooter-0.041
aj-0.039
Token sometimes
Token ID3360
Feature activation-2.37e-06

Pos logit contributions
oc+0.043
 course+0.038
affe+0.038
 deer+0.035
bit+0.033
Neg logit contributions
angers-0.046
 salon-0.039
 bathroom-0.039
 office-0.038
ulk-0.037
Token sne
Token ID10505
Feature activation+2.39e-03

Feature not active on prev token;
no predictions were affected.
Token up
Token ID510
Feature activation+3.35e-02

Pos logit contributions
oc+0.008
 course+0.007
 sorts+0.006
 unity+0.006
 harmony+0.006
Neg logit contributions
angers-0.007
ooter-0.005
 salon-0.005
unes-0.005
outed-0.005
Token.
Token ID13
Feature activation+9.90e-03

Pos logit contributions
angers+0.061
 salon+0.047
aj+0.046
ooter+0.046
belt+0.045
Neg logit contributions
oc-0.069
 course-0.061
 sorts-0.057
 deer-0.056
 unity-0.056
Token
Token ID198
Feature activation-3.77e-02

Pos logit contributions
angers+0.020
 salon+0.015
ooter+0.015
aj+0.015
ulk+0.015
Neg logit contributions
oc-0.019
 course-0.018
 deer-0.017
 unity-0.016
 thanks-0.016
Token
Token ID198
Feature activation-4.08e-02

Pos logit contributions
oc+0.074
 course+0.068
 deer+0.065
 thanks+0.063
 unity+0.063
Neg logit contributions
angers-0.074
 salon-0.056
ulk-0.056
ooter-0.055
aj-0.055
TokenThe
Token ID464
Feature activation-3.39e-02

Pos logit contributions
oc+0.070
 thanks+0.064
 deer+0.063
 course+0.062
 unity+0.059
Neg logit contributions
angers-0.085
 salon-0.069
ulk-0.068
ooter-0.066
belt-0.066
Token clever
Token ID14169
Feature activation-1.40e-01

Pos logit contributions
oc+0.061
 course+0.052
 sorts+0.050
 thanks+0.048
 unity+0.046
Neg logit contributions
angers-0.071
 salon-0.060
ooter-0.058
belt-0.056
 bathroom-0.055
Token girl
Token ID2576
Feature activation-3.31e-02

Pos logit contributions
oc+0.231
 course+0.215
 thanks+0.191
 sorts+0.189
 harmony+0.185
Neg logit contributions
angers-0.285
 salon-0.246
ooter-0.239
anger-0.230
belt-0.226
Token took
Token ID1718
Feature activation-1.30e-02

Pos logit contributions
oc+0.067
 course+0.060
 deer+0.059
 unity+0.058
 harmony+0.057
Neg logit contributions
angers-0.059
 salon-0.049
outed-0.047
ooter-0.045
belt-0.044
Token her
Token ID607
Feature activation+6.44e-02

Pos logit contributions
oc+0.027
 course+0.022
 deer+0.022
 unity+0.021
affe+0.021
Neg logit contributions
angers-0.024
ulk-0.019
anger-0.018
ooter-0.018
 salon-0.017
Token mother
Token ID2802
Feature activation+2.17e-02

Pos logit contributions
angers+0.112
ooter+0.102
 salon+0.097
belt+0.093
aj+0.092
Neg logit contributions
oc-0.131
 course-0.109
 sorts-0.109
 harmony-0.104
 unity-0.103
Token's
Token ID338
Feature activation+5.00e-02

Pos logit contributions
angers+0.030
ooter+0.025
 salon+0.025
 office+0.024
itter+0.022
Neg logit contributions
oc-0.049
 course-0.045
 unity-0.045
 deer-0.044
 harmony-0.043
Token!
Token ID0
Feature activation+1.68e-02

Pos logit contributions
oc+0.161
 deer+0.137
 course+0.132
 unity+0.130
 harmony+0.125
Neg logit contributions
angers-0.158
 salon-0.130
ooter-0.125
belt-0.121
aj-0.121
Tokenâ
Token ID22940
Feature activation-1.10e-03

Pos logit contributions
angers+0.037
 salon+0.029
ooter+0.029
aj+0.028
ulk+0.028
Neg logit contributions
oc-0.029
 course-0.027
 deer-0.025
 unity-0.024
 thanks-0.024
Token
Token ID26391
Feature activation+1.23e-01

Pos logit contributions
oc+0.003
 course+0.003
 thanks+0.002
 unity+0.002
 sorts+0.002
Neg logit contributions
angers-0.002
 salon-0.001
aj-0.001
outed-0.001
unes-0.001
Token.
Token ID13
Feature activation+1.26e-02

Pos logit contributions
angers+0.279
 salon+0.217
outed+0.216
ooter+0.211
ulk+0.208
Neg logit contributions
oc-0.224
 course-0.192
 thanks-0.175
 deer-0.168
 sorts-0.161
Token The
Token ID383
Feature activation-2.81e-02

Pos logit contributions
angers+0.028
 salon+0.022
ooter+0.022
aj+0.021
ulk+0.021
Neg logit contributions
oc-0.022
 course-0.020
 deer-0.019
 thanks-0.018
 unity-0.018
Token playful
Token ID34264
Feature activation-1.40e-01

Pos logit contributions
oc+0.054
 deer+0.045
 course+0.044
 unity+0.044
 sorts+0.044
Neg logit contributions
angers-0.055
 salon-0.047
ooter-0.044
belt-0.042
aj-0.041
Token baby
Token ID5156
Feature activation-2.95e-02

Pos logit contributions
oc+0.232
 course+0.201
 unity+0.192
 deer+0.190
 thanks+0.184
Neg logit contributions
angers-0.307
 salon-0.260
ooter-0.250
anger-0.239
aj-0.235
Token tiger
Token ID26241
Feature activation-7.84e-02

Pos logit contributions
oc+0.053
 course+0.047
 unity+0.046
 deer+0.045
 harmony+0.044
Neg logit contributions
angers-0.059
 salon-0.051
ooter-0.047
belt-0.046
aj-0.045
Tokenâ
Token ID22940
Feature activation+5.64e-03

Pos logit contributions
oc+0.148
 course+0.133
 unity+0.131
 deer+0.130
 harmony+0.128
Neg logit contributions
angers-0.145
 salon-0.122
ooter-0.113
outed-0.109
belt-0.109
Token
Token ID26391
Feature activation+1.29e-01

Pos logit contributions
angers+0.010
 salon+0.007
aj+0.007
ews+0.006
outed+0.006
Neg logit contributions
oc-0.013
 course-0.013
 thanks-0.012
 deer-0.011
 unity-0.011
Token
Token ID8151
Feature activation+2.35e-02

Pos logit contributions
angers+0.280
 salon+0.219
outed+0.213
ooter+0.210
ulk+0.208
Neg logit contributions
oc-0.248
 course-0.213
 thanks-0.196
 deer-0.194
 tiger-0.182
Token!"
Token ID2474
Feature activation+4.48e-02

Pos logit contributions
oc+0.002
 course+0.001
 deer+0.001
 unity+0.001
 sorts+0.001
Neg logit contributions
angers-0.002
 salon-0.001
ooter-0.001
amas-0.001
aj-0.001
Token
Token ID220
Feature activation+1.20e-01

Pos logit contributions
angers+0.074
elt+0.056
ulk+0.056
amas+0.055
ooter+0.055
Neg logit contributions
oc-0.091
 matter-0.086
 unity-0.085
 course-0.085
 deer-0.084
Token
Token ID198
Feature activation-2.77e-02

Pos logit contributions
angers+0.289
 salon+0.240
belt+0.229
ooter+0.227
aj+0.223
Neg logit contributions
oc-0.194
 course-0.167
 thanks-0.154
 deer-0.154
 unity-0.141
Token
Token ID198
Feature activation-3.51e-02

Pos logit contributions
oc+0.056
 course+0.052
 deer+0.050
 unity+0.050
 thanks+0.049
Neg logit contributions
angers-0.051
ulk-0.038
 salon-0.038
ooter-0.037
aj-0.037
TokenThe
Token ID464
Feature activation-2.36e-02

Pos logit contributions
oc+0.058
 thanks+0.057
 unity+0.055
 sorts+0.055
 course+0.054
Neg logit contributions
angers-0.063
 salon-0.054
ulk-0.053
ooter-0.050
belt-0.050
Token playful
Token ID34264
Feature activation-1.40e-01

Pos logit contributions
oc+0.036
 course+0.030
 deer+0.027
 thanks+0.027
 unity+0.027
Neg logit contributions
angers-0.059
 salon-0.049
ooter-0.046
aj-0.046
outed-0.046
Token bee
Token ID20697
Feature activation-2.68e-02

Pos logit contributions
oc+0.205
 course+0.177
 unity+0.153
 deer+0.153
 thanks+0.151
Neg logit contributions
angers-0.348
 salon-0.289
anger-0.276
ooter-0.275
outed-0.272
Token was
Token ID373
Feature activation+6.23e-02

Pos logit contributions
oc+0.054
 unity+0.051
 harmony+0.051
 course+0.050
 sorts+0.050
Neg logit contributions
angers-0.039
outed-0.034
 salon-0.032
uddled-0.031
aj-0.030
Token startled
Token ID36249
Feature activation-5.64e-02

Pos logit contributions
angers+0.129
 salon+0.099
ooter+0.097
ulk+0.095
aj+0.095
Neg logit contributions
oc-0.120
 course-0.110
 deer-0.104
 thanks-0.100
 unity-0.098
Token.
Token ID13
Feature activation+1.27e-02

Pos logit contributions
oc+0.117
 unity+0.101
 course+0.101
 sorts+0.098
 bond+0.097
Neg logit contributions
angers-0.101
elt-0.080
ulk-0.076
outed-0.074
itter-0.074
Token He
Token ID679
Feature activation-6.24e-02

Pos logit contributions
angers+0.030
 salon+0.024
aj+0.023
ulk+0.023
ooter+0.023
Neg logit contributions
oc-0.021
 course-0.019
 deer-0.018
 thanks-0.017
 unity-0.016
Token grateful
Token ID14066
Feature activation-5.71e-02

Pos logit contributions
angers+0.035
outed+0.029
belt+0.029
 salon+0.029
aj+0.028
Neg logit contributions
oc-0.034
 course-0.033
 deer-0.028
affe-0.027
 tiger-0.027
Token.
Token ID13
Feature activation+1.54e-02

Pos logit contributions
oc+0.135
 course+0.121
 deer+0.119
 unity+0.113
 harmony+0.111
Neg logit contributions
angers-0.089
belt-0.064
ooter-0.063
 salon-0.061
aj-0.061
Token
Token ID198
Feature activation-3.50e-02

Pos logit contributions
angers+0.027
ulk+0.020
ooter+0.020
azes+0.020
elt+0.019
Neg logit contributions
oc-0.030
 matter-0.030
 course-0.029
 deer-0.029
 unity-0.028
Token
Token ID198
Feature activation-4.08e-02

Pos logit contributions
oc+0.072
 course+0.067
 deer+0.066
 matter+0.065
 unity+0.064
Neg logit contributions
angers-0.062
ulk-0.046
ooter-0.045
 salon-0.044
aj-0.044
TokenThe
Token ID464
Feature activation-2.57e-02

Pos logit contributions
 thanks+0.068
 matter+0.068
oc+0.066
 deer+0.065
 alien+0.064
Neg logit contributions
angers-0.071
ulk-0.065
 salon-0.059
mage-0.055
belt-0.055
Token clever
Token ID14169
Feature activation-1.40e-01

Pos logit contributions
oc+0.047
 course+0.040
 sorts+0.038
 thanks+0.037
 deer+0.035
Neg logit contributions
angers-0.052
 salon-0.044
ooter-0.042
 office-0.041
belt-0.041
Token person
Token ID1048
Feature activation-3.22e-02

Pos logit contributions
oc+0.266
 course+0.255
 sorts+0.222
 thanks+0.222
 harmony+0.215
Neg logit contributions
angers-0.260
 salon-0.217
ooter-0.211
belt-0.199
outed-0.198
Token thanked
Token ID26280
Feature activation-2.43e-02

Pos logit contributions
oc+0.063
 harmony+0.056
 unity+0.056
 matter+0.056
 course+0.055
Neg logit contributions
angers-0.048
 salon-0.045
outed-0.040
 office-0.040
ooter-0.040
Token his
Token ID465
Feature activation+5.30e-02

Pos logit contributions
oc+0.048
 sorts+0.043
 bond+0.040
able+0.040
 sizes+0.039
Neg logit contributions
angers-0.039
 salon-0.031
omm-0.030
ulk-0.030
 office-0.029
Token friend
Token ID1545
Feature activation-6.04e-02

Pos logit contributions
angers+0.090
ooter+0.075
 office+0.075
itter+0.070
omm+0.069
Neg logit contributions
 course-0.100
oc-0.099
 sorts-0.092
 Freddie-0.086
 harmony-0.084
Token for
Token ID329
Feature activation-4.29e-02

Pos logit contributions
oc+0.126
 matter+0.120
 harmony+0.119
 unity+0.117
 course+0.117
Neg logit contributions
angers-0.086
ooter-0.070
 office-0.067
 salon-0.067
itter-0.063
Token more
Token ID517
Feature activation-1.71e-02

Pos logit contributions
oc+0.065
 deer+0.052
affe+0.050
 Mickey+0.050
 unity+0.050
Neg logit contributions
angers-0.068
 salon-0.059
ulk-0.055
aj-0.054
 bathroom-0.054
Token.
Token ID13
Feature activation+1.69e-02

Pos logit contributions
oc+0.036
 course+0.032
 deer+0.031
 unity+0.031
 Mickey+0.031
Neg logit contributions
angers-0.027
 salon-0.022
aj-0.021
ooter-0.021
belt-0.019
Token
Token ID198
Feature activation-3.50e-02

Pos logit contributions
angers+0.033
ooter+0.024
ulk+0.024
aj+0.024
 salon+0.024
Neg logit contributions
oc-0.033
 course-0.031
 deer-0.030
 unity-0.028
 matter-0.028
Token
Token ID198
Feature activation-4.09e-02

Pos logit contributions
oc+0.070
 course+0.065
 deer+0.063
 unity+0.061
 matter+0.060
Neg logit contributions
angers-0.066
ulk-0.049
 salon-0.049
aj-0.048
ooter-0.048
TokenThe
Token ID464
Feature activation-2.50e-02

Pos logit contributions
oc+0.065
 thanks+0.061
 deer+0.060
 matter+0.059
 unity+0.059
Neg logit contributions
angers-0.083
ulk-0.071
 salon-0.069
belt-0.067
OM-0.066
Token playful
Token ID34264
Feature activation-1.39e-01

Pos logit contributions
oc+0.053
 sorts+0.043
 course+0.042
 unity+0.041
 thanks+0.040
Neg logit contributions
angers-0.043
 salon-0.036
ooter-0.034
 bathroom-0.034
belt-0.034
Token puppy
Token ID26188
Feature activation-6.44e-02

Pos logit contributions
oc+0.220
 course+0.179
 sorts+0.164
 unity+0.159
 thanks+0.155
Neg logit contributions
angers-0.319
ooter-0.267
 salon-0.266
amas-0.260
 bathroom-0.256
Token was
Token ID373
Feature activation+6.33e-02

Pos logit contributions
oc+0.133
 course+0.124
 unity+0.120
 matter+0.116
 sorts+0.115
Neg logit contributions
angers-0.103
 salon-0.085
aj-0.079
belt-0.078
ooter-0.077
Token so
Token ID523
Feature activation+2.08e-02

Pos logit contributions
angers+0.129
 salon+0.105
aj+0.100
ulk+0.100
ooter+0.099
Neg logit contributions
oc-0.121
 course-0.106
 deer-0.099
 unity-0.096
affe-0.093
Token happy
Token ID3772
Feature activation-5.67e-02

Pos logit contributions
angers+0.045
aj+0.037
outed+0.037
 salon+0.036
elt+0.036
Neg logit contributions
oc-0.035
 course-0.032
 deer-0.030
 unity-0.028
 matter-0.028
Token that
Token ID326
Feature activation-2.05e-02

Pos logit contributions
oc+0.121
 course+0.109
 unity+0.103
 deer+0.102
 matter+0.099
Neg logit contributions
angers-0.102
elt-0.076
ulk-0.075
 salon-0.073
ooter-0.072
Token see
Token ID766
Feature activation+2.13e-02

Pos logit contributions
angers+0.048
 salon+0.043
ooter+0.039
ulk+0.038
 bathroom+0.037
Neg logit contributions
oc-0.052
 deer-0.043
 course-0.042
able-0.042
affe-0.040
Token who
Token ID508
Feature activation-6.07e-02

Pos logit contributions
angers+0.040
 salon+0.035
 office+0.031
ews+0.031
ulk+0.030
Neg logit contributions
oc-0.041
affe-0.034
Fred-0.031
 deer-0.031
 course-0.031
Token was
Token ID373
Feature activation+6.93e-02

Pos logit contributions
oc+0.123
 deer+0.107
 course+0.103
 unity+0.101
affe+0.099
Neg logit contributions
angers-0.109
 salon-0.094
belt-0.085
outed-0.084
ooter-0.082
Token the
Token ID262
Feature activation+4.00e-02

Pos logit contributions
angers+0.141
 salon+0.110
ooter+0.105
ulk+0.104
aj+0.103
Neg logit contributions
oc-0.137
 course-0.123
 deer-0.116
 unity-0.109
 thanks-0.109
Token most
Token ID749
Feature activation-7.75e-02

Pos logit contributions
angers+0.070
 salon+0.060
 bathroom+0.055
ooter+0.052
 office+0.051
Neg logit contributions
oc-0.090
 deer-0.073
 course-0.071
affe-0.068
 sorts-0.065
Token intelligent
Token ID12661
Feature activation-1.39e-01

Pos logit contributions
oc+0.142
 course+0.125
 deer+0.123
affe+0.109
 Hunter+0.108
Neg logit contributions
angers-0.159
 salon-0.130
ooter-0.120
aj-0.118
 bathroom-0.117
Token chess
Token ID19780
Feature activation-1.47e-02

Pos logit contributions
oc+0.264
 course+0.208
 deer+0.206
 unity+0.192
Fred+0.182
Neg logit contributions
angers-0.282
 salon-0.243
anger-0.221
 bathroom-0.215
belt-0.213
Token player
Token ID2137
Feature activation-5.73e-02

Pos logit contributions
oc+0.028
 deer+0.024
 course+0.023
 unity+0.022
 thanks+0.022
Neg logit contributions
angers-0.030
 salon-0.024
anger-0.023
boards-0.023
belt-0.023
Token.
Token ID13
Feature activation+1.23e-02

Pos logit contributions
oc+0.115
 unity+0.096
 deer+0.095
 course+0.095
 harmony+0.092
Neg logit contributions
angers-0.107
 salon-0.085
belt-0.083
anger-0.080
ooter-0.080
Token They
Token ID1119
Feature activation-9.44e-02

Pos logit contributions
angers+0.026
 salon+0.020
ooter+0.020
ulk+0.020
amas+0.019
Neg logit contributions
oc-0.023
 course-0.021
 deer-0.020
 unity-0.019
 thanks-0.018
Token hugged
Token ID41130
Feature activation-1.27e-02

Pos logit contributions
oc+0.215
 turkey+0.176
 course+0.172
 harmony+0.169
 deer+0.167
Neg logit contributions
angers-0.136
 salon-0.117
outed-0.112
ooter-0.108
belt-0.107
Token distance
Token ID5253
Feature activation+1.82e-02

Pos logit contributions
angers+0.082
 salon+0.062
aj+0.061
ooter+0.061
outed+0.061
Neg logit contributions
oc-0.060
 course-0.055
 deer-0.054
 thanks-0.052
 harmony-0.049
Token.
Token ID13
Feature activation+1.07e-02

Pos logit contributions
angers+0.026
 salon+0.018
ooter+0.018
line+0.017
belt+0.017
Neg logit contributions
oc-0.044
 course-0.041
 deer-0.038
 unity-0.038
 bond-0.037
Token
Token ID198
Feature activation-3.50e-02

Pos logit contributions
angers+0.025
 salon+0.020
ooter+0.019
aj+0.019
ulk+0.019
Neg logit contributions
oc-0.018
 course-0.016
 deer-0.015
 thanks-0.015
 unity-0.014
Token
Token ID198
Feature activation-4.04e-02

Pos logit contributions
oc+0.068
 course+0.063
 deer+0.060
 thanks+0.058
 unity+0.058
Neg logit contributions
angers-0.070
 salon-0.053
ooter-0.052
ulk-0.052
aj-0.051
TokenThe
Token ID464
Feature activation-2.60e-02

Pos logit contributions
oc+0.068
 thanks+0.065
 course+0.064
 deer+0.062
 unity+0.061
Neg logit contributions
angers-0.081
ulk-0.066
 salon-0.064
belt-0.063
ooter-0.062
Token intelligent
Token ID12661
Feature activation-1.39e-01

Pos logit contributions
oc+0.052
 course+0.047
 deer+0.043
 sorts+0.042
 unity+0.041
Neg logit contributions
angers-0.051
 salon-0.040
ooter-0.038
aj-0.038
 bathroom-0.038
Token kid
Token ID5141
Feature activation-2.98e-02

Pos logit contributions
oc+0.233
 course+0.217
 deer+0.183
 unity+0.180
 sorts+0.180
Neg logit contributions
angers-0.309
 salon-0.245
ooter-0.242
anger-0.236
aj-0.233
Token thought
Token ID1807
Feature activation+2.25e-02

Pos logit contributions
oc+0.061
 course+0.056
 unity+0.053
 harmony+0.053
 deer+0.053
Neg logit contributions
angers-0.053
 salon-0.042
outed-0.040
ooter-0.040
belt-0.039
Token of
Token ID286
Feature activation-9.90e-02

Pos logit contributions
angers+0.040
ooter+0.031
anger+0.031
 salon+0.030
belt+0.029
Neg logit contributions
oc-0.050
 course-0.041
 deer-0.039
 harmony-0.039
 unity-0.038
Token a
Token ID257
Feature activation-4.87e-02

Pos logit contributions
oc+0.218
affe+0.193
 deer+0.170
lest+0.166
able+0.164
Neg logit contributions
angers-0.134
 office-0.117
 bathroom-0.110
azes-0.110
ulk-0.106
Token plan
Token ID1410
Feature activation-1.23e-02

Pos logit contributions
oc+0.101
 course+0.090
affe+0.083
ck+0.080
bit+0.078
Neg logit contributions
 bathroom-0.065
 salon-0.064
 office-0.061
angers-0.060
 door-0.059
Token help
Token ID1037
Feature activation-1.62e-02

Pos logit contributions
oc+0.075
 course+0.058
 sorts+0.058
affe+0.057
 deer+0.054
Neg logit contributions
angers-0.084
ulk-0.068
 bathroom-0.065
azes-0.065
aj-0.063
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
oc+0.035
 course+0.033
 deer+0.031
 unity+0.031
 thanks+0.030
Neg logit contributions
angers-0.028
 salon-0.020
ooter-0.020
ulk-0.020
amas-0.019
Token
Token ID198
Feature activation-3.85e-02

Pos logit contributions
angers+0.024
 salon+0.019
ooter+0.019
ulk+0.018
aj+0.018
Neg logit contributions
oc-0.019
 course-0.017
 deer-0.016
 thanks-0.016
 unity-0.015
Token
Token ID198
Feature activation-4.40e-02

Pos logit contributions
oc+0.075
 course+0.071
 deer+0.067
 unity+0.066
 thanks+0.065
Neg logit contributions
angers-0.074
 salon-0.056
ooter-0.055
ulk-0.055
aj-0.054
TokenThe
Token ID464
Feature activation-2.35e-02

Pos logit contributions
 course+0.074
 thanks+0.073
oc+0.071
 unity+0.070
 kindness+0.068
Neg logit contributions
angers-0.086
belt-0.070
ooter-0.069
 salon-0.068
ulk-0.066
Token clever
Token ID14169
Feature activation-1.39e-01

Pos logit contributions
oc+0.034
 course+0.030
 sorts+0.029
 unity+0.025
 matter+0.024
Neg logit contributions
angers-0.054
 bathroom-0.046
tub-0.045
outed-0.044
aj-0.044
Token fish
Token ID5916
Feature activation-5.77e-02

Pos logit contributions
 course+0.160
oc+0.159
 sorts+0.130
 unity+0.122
 harmony+0.120
Neg logit contributions
angers-0.340
tub-0.291
outed-0.285
anger-0.281
ooter-0.276
Token smiled
Token ID13541
Feature activation+1.81e-02

Pos logit contributions
oc+0.107
 course+0.105
 harmony+0.100
 unity+0.100
 matter+0.098
Neg logit contributions
angers-0.095
outed-0.085
elt-0.078
am-0.075
ooter-0.075
Token and
Token ID290
Feature activation-3.04e-02

Pos logit contributions
angers+0.041
 salon+0.032
ooter+0.031
aj+0.031
outed+0.031
Neg logit contributions
oc-0.031
 course-0.029
 deer-0.026
 unity-0.025
 thanks-0.025
Token said
Token ID531
Feature activation+2.72e-02

Pos logit contributions
oc+0.058
 course+0.052
 unity+0.047
 harmony+0.046
 matter+0.046
Neg logit contributions
angers-0.060
elt-0.046
outed-0.046
aj-0.045
ulk-0.045
Token,
Token ID11
Feature activation+6.74e-02

Pos logit contributions
angers+0.060
 salon+0.047
ooter+0.045
aj+0.045
ulk+0.045
Neg logit contributions
oc-0.049
 course-0.045
 deer-0.043
 thanks-0.041
 unity-0.039
Token down
Token ID866
Feature activation+5.88e-02

Pos logit contributions
oc+0.100
 course+0.091
 matter+0.087
 unity+0.086
 deer+0.083
Neg logit contributions
angers-0.072
 salon-0.054
ooter-0.054
belt-0.051
anger-0.051
Token and
Token ID290
Feature activation-2.93e-02

Pos logit contributions
angers+0.112
 salon+0.087
anger+0.082
ulk+0.082
ooter+0.081
Neg logit contributions
oc-0.119
 matter-0.098
 deer-0.096
 unity-0.095
 course-0.093
Token try
Token ID1949
Feature activation+3.30e-02

Pos logit contributions
oc+0.057
 course+0.048
affe+0.044
able+0.043
 sorts+0.041
Neg logit contributions
angers-0.053
 salon-0.047
ooter-0.045
 bathroom-0.045
line-0.042
Token to
Token ID284
Feature activation-1.12e-02

Pos logit contributions
angers+0.068
 salon+0.052
ooter+0.051
aj+0.051
ulk+0.050
Neg logit contributions
oc-0.064
 course-0.058
 deer-0.053
 unity-0.052
 thanks-0.051
Token catch
Token ID4929
Feature activation+7.74e-04

Pos logit contributions
oc+0.023
 course+0.019
 unity+0.018
 sorts+0.018
 teamwork+0.018
Neg logit contributions
angers-0.016
ooter-0.014
 salon-0.014
rolling-0.013
our-0.013
Token it
Token ID340
Feature activation+1.01e-01

Pos logit contributions
angers+0.001
ulk+0.001
uck+0.001
unes+0.001
line+0.001
Neg logit contributions
oc-0.001
 bond-0.001
 matter-0.001
affe-0.001
 unity-0.001
Token!"
Token ID2474
Feature activation+4.87e-02

Pos logit contributions
angers+0.219
 salon+0.174
ooter+0.173
amas+0.171
anger+0.170
Neg logit contributions
oc-0.170
 course-0.156
 unity-0.146
 deer-0.143
 bond-0.139
Token
Token ID220
Feature activation+1.24e-01

Pos logit contributions
angers+0.111
 salon+0.087
ooter+0.084
aj+0.084
belt+0.083
Neg logit contributions
oc-0.085
 course-0.078
 deer-0.072
 thanks-0.071
 unity-0.068
Token
Token ID198
Feature activation-2.74e-02

Pos logit contributions
angers+0.313
 salon+0.263
ooter+0.254
belt+0.254
ulk+0.246
Neg logit contributions
oc-0.181
 course-0.148
 thanks-0.143
 deer-0.136
 Charlie-0.127
Token
Token ID198
Feature activation-3.56e-02

Pos logit contributions
oc+0.050
 course+0.045
 deer+0.043
 thanks+0.041
 unity+0.040
Neg logit contributions
angers-0.060
 salon-0.047
ooter-0.045
aj-0.045
ulk-0.045
TokenThey
Token ID2990
Feature activation-7.33e-02

Pos logit contributions
oc+0.058
 course+0.052
 deer+0.049
 thanks+0.048
 unity+0.047
Neg logit contributions
angers-0.079
 salon-0.067
ews-0.064
ulk-0.063
outed-0.062
Token see
Token ID766
Feature activation+2.52e-02

Pos logit contributions
oc+0.036
 course+0.032
 deer+0.030
 thanks+0.028
 unity+0.028
Neg logit contributions
angers-0.062
 salon-0.051
ooter-0.050
ulk-0.049
belt-0.048
Token Lily
Token ID20037
Feature activation+2.05e-02

Pos logit contributions
angers+0.040
 salon+0.035
 bathroom+0.031
ulk+0.030
 office+0.030
Neg logit contributions
oc-0.058
affe-0.046
 bond-0.046
 unity-0.045
 harmony-0.045
Token having
Token ID1719
Feature activation-2.72e-02

Pos logit contributions
angers+0.040
 salon+0.032
ooter+0.031
belt+0.029
ulk+0.029
Neg logit contributions
oc-0.040
 course-0.038
 deer-0.035
 unity-0.034
 harmony-0.033
Token fun
Token ID1257
Feature activation-5.52e-02

Pos logit contributions
oc+0.042
 course+0.034
 deer+0.034
 unity+0.031
affe+0.030
Neg logit contributions
angers-0.067
 salon-0.055
ulk-0.054
aj-0.053
ooter-0.053
Token.
Token ID13
Feature activation+1.86e-02

Pos logit contributions
oc+0.103
 course+0.095
 deer+0.088
 thanks+0.086
 unity+0.084
Neg logit contributions
angers-0.117
 salon-0.091
ooter-0.089
aj-0.088
outed-0.087
Token
Token ID220
Feature activation+1.16e-01

Pos logit contributions
angers+0.039
 salon+0.030
ooter+0.029
aj+0.029
ulk+0.029
Neg logit contributions
oc-0.035
 course-0.032
 deer-0.030
 thanks-0.029
 unity-0.029
Token
Token ID198
Feature activation-3.25e-02

Pos logit contributions
angers+0.292
 salon+0.256
belt+0.240
ooter+0.236
ulk+0.230
Neg logit contributions
oc-0.157
 course-0.133
 thanks-0.127
 deer-0.119
 Charlie-0.115
Token
Token ID198
Feature activation-3.84e-02

Pos logit contributions
oc+0.059
 course+0.054
 deer+0.051
 thanks+0.049
 unity+0.048
Neg logit contributions
angers-0.070
 salon-0.055
ooter-0.053
aj-0.053
ulk-0.053
TokenThe
Token ID464
Feature activation-2.51e-02

Pos logit contributions
oc+0.068
 course+0.064
 thanks+0.063
 deer+0.062
 unity+0.061
Neg logit contributions
angers-0.078
 salon-0.062
ulk-0.060
belt-0.059
amas-0.057
Token two
Token ID734
Feature activation-7.88e-02

Pos logit contributions
oc+0.056
 course+0.046
 sorts+0.045
 deer+0.044
 thanks+0.044
Neg logit contributions
angers-0.042
 salon-0.034
ooter-0.034
belt-0.033
 bathroom-0.033
Token friends
Token ID2460
Feature activation-9.10e-02

Pos logit contributions
oc+0.163
 course+0.130
 deer+0.129
 unity+0.129
 harmony+0.127
Neg logit contributions
angers-0.151
 salon-0.121
outed-0.116
ooter-0.113
uddled-0.112
Token the
Token ID262
Feature activation+3.40e-02

Pos logit contributions
oc+0.026
affe+0.022
 deer+0.021
 course+0.020
 tiger+0.019
Neg logit contributions
angers-0.025
 salon-0.021
ulk-0.020
 bathroom-0.020
 office-0.019
Token play
Token ID711
Feature activation+2.26e-03

Pos logit contributions
angers+0.069
 salon+0.052
ooter+0.051
aj+0.050
ulk+0.050
Neg logit contributions
oc-0.066
 course-0.062
 deer-0.058
 thanks-0.057
 unity-0.056
Token.
Token ID13
Feature activation+1.72e-02

Pos logit contributions
angers+0.004
 salon+0.003
ooter+0.003
ulk+0.003
belt+0.003
Neg logit contributions
oc-0.005
 course-0.004
 deer-0.004
 thanks-0.004
 unity-0.004
Token So
Token ID1406
Feature activation-4.60e-02

Pos logit contributions
angers+0.036
 salon+0.029
ulk+0.027
ooter+0.027
aj+0.027
Neg logit contributions
oc-0.032
 course-0.029
 deer-0.028
 unity-0.026
 thanks-0.026
Token Anna
Token ID11735
Feature activation+8.89e-03

Pos logit contributions
oc+0.104
 deer+0.092
 course+0.089
 unity+0.085
 sorts+0.085
Neg logit contributions
angers-0.078
 salon-0.068
ulk-0.056
belt-0.055
 bathroom-0.054
Token started
Token ID2067
Feature activation+1.01e-01

Pos logit contributions
angers+0.018
 salon+0.014
ooter+0.013
aj+0.013
outed+0.013
Neg logit contributions
oc-0.017
 course-0.016
 deer-0.015
 unity-0.015
 thanks-0.014
Token to
Token ID284
Feature activation-1.02e-02

Pos logit contributions
angers+0.195
ooter+0.145
 salon+0.145
aj+0.145
outed+0.144
Neg logit contributions
oc-0.198
 course-0.184
 deer-0.179
 thanks-0.178
 sorts-0.172
Token learn
Token ID2193
Feature activation-7.59e-03

Pos logit contributions
oc+0.019
 course+0.017
affe+0.016
 deer+0.015
 turkey+0.015
Neg logit contributions
angers-0.021
 salon-0.018
 bathroom-0.017
ooter-0.016
ulk-0.016
Token all
Token ID477
Feature activation-5.05e-03

Pos logit contributions
oc+0.014
 course+0.012
 deer+0.011
 unity+0.011
 thanks+0.010
Neg logit contributions
angers-0.017
 salon-0.014
ulk-0.013
ooter-0.013
aj-0.013
Token the
Token ID262
Feature activation+4.37e-02

Pos logit contributions
oc+0.008
 deer+0.006
 course+0.006
affe+0.005
 unity+0.005
Neg logit contributions
angers-0.012
 salon-0.010
ulk-0.010
ooter-0.010
 office-0.010
Token lines
Token ID3951
Feature activation+4.39e-03

Pos logit contributions
angers+0.094
 salon+0.075
 bathroom+0.073
ooter+0.073
ulk+0.071
Neg logit contributions
oc-0.082
 course-0.065
affe-0.065
 deer-0.064
Fred-0.059
Token.
Token ID13
Feature activation+1.67e-02

Pos logit contributions
oc+0.178
 course+0.159
 deer+0.151
 unity+0.147
affe+0.142
Neg logit contributions
angers-0.145
 salon-0.108
belt-0.105
ooter-0.104
anger-0.103
Token Every
Token ID3887
Feature activation-7.78e-02

Pos logit contributions
angers+0.036
 salon+0.028
ooter+0.027
ulk+0.027
aj+0.027
Neg logit contributions
oc-0.030
 course-0.028
 deer-0.027
 thanks-0.025
 unity-0.025
Token time
Token ID640
Feature activation+2.61e-02

Pos logit contributions
oc+0.159
Mic+0.133
 deer+0.131
Blue+0.130
affe+0.130
Neg logit contributions
angers-0.112
 office-0.103
 salon-0.103
 bathroom-0.097
belt-0.092
Token he
Token ID339
Feature activation-4.65e-03

Pos logit contributions
angers+0.060
 salon+0.048
ulk+0.046
belt+0.046
ooter+0.046
Neg logit contributions
oc-0.043
 course-0.037
 deer-0.034
 unity-0.033
 thanks-0.032
Token pressed
Token ID12070
Feature activation-5.23e-03

Pos logit contributions
oc+0.009
 course+0.009
 unity+0.008
 deer+0.008
 harmony+0.008
Neg logit contributions
angers-0.008
 salon-0.007
outed-0.006
ooter-0.006
belt-0.006
Token it
Token ID340
Feature activation+9.25e-02

Pos logit contributions
oc+0.009
 course+0.008
 deer+0.008
 bond+0.007
affe+0.007
Neg logit contributions
angers-0.011
anger-0.008
ooter-0.008
belt-0.008
ulk-0.008
Token,
Token ID11
Feature activation+6.79e-02

Pos logit contributions
angers+0.179
ooter+0.133
 salon+0.132
anger+0.132
belt+0.127
Neg logit contributions
oc-0.187
 course-0.169
 deer-0.164
 unity-0.157
 thanks-0.156
Token the
Token ID262
Feature activation+4.47e-02

Pos logit contributions
angers+0.138
 salon+0.110
ulk+0.106
belt+0.102
line+0.101
Neg logit contributions
oc-0.127
 course-0.108
 bond-0.102
affe-0.100
 deer-0.097
Token animals
Token ID4695
Feature activation-5.01e-02

Pos logit contributions
angers+0.108
 salon+0.092
ooter+0.092
 bathroom+0.090
anger+0.090
Neg logit contributions
oc-0.061
 course-0.053
 thanks-0.047
 sorts-0.046
 unity-0.044
Token became
Token ID2627
Feature activation+7.51e-02

Pos logit contributions
oc+0.106
 course+0.092
 unity+0.086
 deer+0.085
 thanks+0.085
Neg logit contributions
angers-0.091
 salon-0.072
ooter-0.069
belt-0.066
outed-0.066
Token bigger
Token ID5749
Feature activation-3.92e-02

Pos logit contributions
angers+0.136
outed+0.103
 salon+0.102
ulk+0.100
ews+0.099
Neg logit contributions
oc-0.165
 course-0.144
affe-0.133
 deer-0.123
 bond-0.119
Tokenparent
Token ID8000
Feature activation-1.47e-02

Pos logit contributions
oc+0.083
 course+0.072
 deer+0.071
 unity+0.066
 thanks+0.065
Neg logit contributions
angers-0.106
 salon-0.085
ooter-0.083
aj-0.083
ulk-0.081
Tokenâ
Token ID22940
Feature activation+1.68e-02

Pos logit contributions
oc+0.028
 course+0.026
 deer+0.025
 unity+0.024
 thanks+0.023
Neg logit contributions
angers-0.030
 salon-0.024
ooter-0.023
aj-0.022
ulk-0.022
Token
Token ID26391
Feature activation+1.44e-01

Pos logit contributions
angers+0.033
 salon+0.024
aj+0.023
outed+0.023
anger+0.023
Neg logit contributions
oc-0.035
 course-0.035
 thanks-0.030
 sorts-0.030
 unity-0.029
Token
Token ID8151
Feature activation+3.63e-02

Pos logit contributions
angers+0.285
 salon+0.220
outed+0.217
ooter+0.216
ulk+0.215
Neg logit contributions
oc-0.296
 course-0.249
 thanks-0.235
 Charlie-0.224
 harmony-0.221
Tokens
Token ID82
Feature activation+2.32e-02

Pos logit contributions
angers+0.073
ulk+0.056
 salon+0.055
aj+0.053
amas+0.053
Neg logit contributions
oc-0.070
 course-0.066
 thanks-0.058
 deer-0.057
 sorts-0.055
Token home
Token ID1363
Feature activation+1.09e-01

Pos logit contributions
angers+0.037
 salon+0.027
ooter+0.026
aj+0.025
ulk+0.025
Neg logit contributions
oc-0.055
 course-0.051
 deer-0.049
 thanks-0.047
 unity-0.047
Token,
Token ID11
Feature activation+6.05e-02

Pos logit contributions
angers+0.247
 salon+0.192
ooter+0.190
ulk+0.190
aj+0.186
Neg logit contributions
oc-0.181
 course-0.168
 deer-0.154
 thanks-0.154
 unity-0.144
Token the
Token ID262
Feature activation+4.44e-02

Pos logit contributions
angers+0.109
 salon+0.090
 bathroom+0.085
belt+0.084
aj+0.080
Neg logit contributions
oc-0.131
affe-0.107
 course-0.107
 deer-0.104
 unity-0.100
Token little
Token ID1310
Feature activation-8.20e-02

Pos logit contributions
angers+0.089
 salon+0.070
ooter+0.068
belt+0.066
aj+0.066
Neg logit contributions
oc-0.088
 course-0.078
 deer-0.073
 thanks-0.072
 unity-0.070
Token dog
Token ID3290
Feature activation-4.21e-02

Pos logit contributions
oc+0.144
 course+0.126
 deer+0.114
 unity+0.113
 thanks+0.111
Neg logit contributions
angers-0.185
 salon-0.150
ooter-0.144
aj-0.141
belt-0.140
Token in
Token ID287
Feature activation-1.13e-02

Pos logit contributions
oc+0.082
 course+0.072
 unity+0.069
 deer+0.068
 harmony+0.066
Neg logit contributions
angers-0.083
 salon-0.066
ooter-0.063
aj-0.062
belt-0.061
Token and
Token ID290
Feature activation-3.13e-02

Pos logit contributions
oc+0.056
 course+0.052
 unity+0.051
 deer+0.049
 harmony+0.048
Neg logit contributions
angers-0.050
 salon-0.040
ooter-0.037
aj-0.035
ulk-0.035
Token said
Token ID531
Feature activation+3.53e-02

Pos logit contributions
oc+0.059
 course+0.054
 unity+0.050
 sorts+0.048
 deer+0.048
Neg logit contributions
angers-0.063
 salon-0.051
ooter-0.048
outed-0.048
ulk-0.047
Token,
Token ID11
Feature activation+7.08e-02

Pos logit contributions
angers+0.088
 salon+0.070
amas+0.070
ulk+0.068
ooter+0.067
Neg logit contributions
oc-0.053
 course-0.049
 deer-0.047
 thanks-0.046
 kindness-0.043
Token "
Token ID366
Feature activation+2.49e-03

Pos logit contributions
angers+0.156
 salon+0.123
ulk+0.122
aj+0.119
ooter+0.118
Neg logit contributions
oc-0.125
 deer-0.113
 course-0.113
 thanks-0.106
 tiger-0.103
TokenHi
Token ID17250
Feature activation-2.69e-02

Pos logit contributions
angers+0.005
 salon+0.004
elt+0.004
ulk+0.004
outed+0.004
Neg logit contributions
oc-0.005
 course-0.004
 thanks-0.004
 sorts-0.004
 unity-0.004
Token,
Token ID11
Feature activation+4.68e-02

Pos logit contributions
oc+0.049
 course+0.044
 deer+0.042
 thanks+0.040
 unity+0.039
Neg logit contributions
angers-0.060
 salon-0.047
ooter-0.046
ulk-0.045
belt-0.045
Token do
Token ID466
Feature activation-2.68e-02

Pos logit contributions
angers+0.076
 salon+0.071
 bathroom+0.065
ooter+0.063
anger+0.060
Neg logit contributions
oc-0.103
 sorts-0.089
 harmony-0.085
 unity-0.085
 course-0.080
Tokenggy
Token ID19970
Feature activation-2.60e-02

Pos logit contributions
oc+0.038
 course+0.034
 deer+0.031
 thanks+0.029
 unity+0.028
Neg logit contributions
angers-0.070
 salon-0.057
ooter-0.056
aj-0.055
ulk-0.055
Token!
Token ID0
Feature activation+8.72e-03

Pos logit contributions
oc+0.053
 course+0.049
 deer+0.047
 unity+0.045
 thanks+0.045
Neg logit contributions
angers-0.050
 salon-0.038
ooter-0.036
aj-0.036
ulk-0.036
Token Do
Token ID2141
Feature activation-3.78e-02

Pos logit contributions
angers+0.017
 salon+0.013
ooter+0.013
elt+0.013
aj+0.013
Neg logit contributions
oc-0.016
 course-0.016
 unity-0.015
 deer-0.015
 sorts-0.015
Token you
Token ID345
Feature activation-3.59e-02

Pos logit contributions
oc+0.053
 course+0.046
 deer+0.043
 thanks+0.041
 unity+0.040
Neg logit contributions
angers-0.097
 salon-0.079
ooter-0.078
ulk-0.077
aj-0.077
Token have
Token ID423
Feature activation-4.42e-02

Pos logit contributions
oc+0.093
 course+0.080
 unity+0.078
 sorts+0.078
 harmony+0.077
Neg logit contributions
angers-0.097
 salon-0.083
ooter-0.082
 bathroom-0.074
belt-0.074
Token it
Token ID340
Feature activation+6.91e-02

Pos logit contributions
oc+0.077
 sorts+0.059
affe+0.059
 deer+0.058
 course+0.058
Neg logit contributions
angers-0.092
 salon-0.078
ulk-0.077
 bathroom-0.074
aj-0.074
Token back
Token ID736
Feature activation+5.12e-02

Pos logit contributions
angers+0.149
ulk+0.112
 salon+0.111
azes+0.111
ews+0.109
Neg logit contributions
oc-0.127
 course-0.121
 deer-0.112
 thanks-0.111
 love-0.102
Token,
Token ID11
Feature activation+5.52e-02

Pos logit contributions
angers+0.104
 salon+0.077
ooter+0.075
ulk+0.075
amas+0.074
Neg logit contributions
oc-0.102
 course-0.095
 deer-0.090
 thanks-0.088
 unity-0.083
Token please
Token ID3387
Feature activation-5.89e-03

Pos logit contributions
angers+0.077
 salon+0.068
 bathroom+0.061
ooter+0.059
line+0.059
Neg logit contributions
oc-0.129
 sorts-0.118
 course-0.118
 unity-0.109
 harmony-0.109
Token?"
Token ID1701
Feature activation+4.35e-02

Pos logit contributions
oc+0.012
 course+0.011
 thanks+0.010
 deer+0.010
 unity+0.010
Neg logit contributions
angers-0.012
 salon-0.009
ooter-0.009
ulk-0.009
outed-0.009
Token Lily
Token ID20037
Feature activation+2.60e-02

Pos logit contributions
angers+0.082
 salon+0.064
ulk+0.062
 bathroom+0.060
outed+0.058
Neg logit contributions
oc-0.087
 course-0.080
 unity-0.075
 bond-0.074
 sorts-0.074
Token asked
Token ID1965
Feature activation+1.51e-02

Pos logit contributions
angers+0.071
anger+0.056
belt+0.056
 salon+0.056
ooter+0.056
Neg logit contributions
oc-0.035
 course-0.030
 deer-0.029
 thanks-0.027
 tiger-0.024
Token.
Token ID13
Feature activation+2.62e-02

Pos logit contributions
angers+0.034
 salon+0.027
ooter+0.026
aj+0.026
belt+0.026
Neg logit contributions
oc-0.026
 course-0.024
 deer-0.023
 thanks-0.022
 unity-0.021
Token
Token ID198
Feature activation-2.71e-02

Pos logit contributions
angers+0.058
 salon+0.046
ooter+0.045
aj+0.045
ulk+0.044
Neg logit contributions
oc-0.046
 course-0.042
 deer-0.040
 thanks-0.038
 unity-0.037
Token
Token ID198
Feature activation-3.38e-02

Pos logit contributions
oc+0.053
 course+0.049
 deer+0.047
 unity+0.045
 thanks+0.045
Neg logit contributions
angers-0.054
 salon-0.041
ulk-0.040
ooter-0.040
aj-0.039
Token But
Token ID887
Feature activation-1.26e-02

Pos logit contributions
angers+0.041
 salon+0.031
ooter+0.030
aj+0.030
ulk+0.030
Neg logit contributions
oc-0.037
 course-0.034
 deer-0.032
 unity-0.031
 thanks-0.030
Token the
Token ID262
Feature activation+3.54e-02

Pos logit contributions
oc+0.022
 course+0.019
 sorts+0.018
 deer+0.018
 unity+0.018
Neg logit contributions
angers-0.026
 salon-0.022
belt-0.022
ulk-0.020
outed-0.020
Token boy
Token ID2933
Feature activation-2.16e-02

Pos logit contributions
angers+0.074
 salon+0.061
ooter+0.059
belt+0.059
 bathroom+0.058
Neg logit contributions
oc-0.065
 course-0.056
 sorts-0.053
 thanks-0.051
 unity-0.048
Token was
Token ID373
Feature activation+6.57e-02

Pos logit contributions
oc+0.044
 course+0.042
 deer+0.040
 unity+0.040
 harmony+0.039
Neg logit contributions
angers-0.038
 salon-0.030
outed-0.029
ooter-0.029
belt-0.028
Token safe
Token ID3338
Feature activation-4.25e-02

Pos logit contributions
angers+0.148
 salon+0.114
ooter+0.111
anger+0.111
aj+0.110
Neg logit contributions
oc-0.113
 course-0.105
 deer-0.101
 thanks-0.098
 unity-0.094
Token.
Token ID13
Feature activation+2.00e-02

Pos logit contributions
oc+0.083
 course+0.075
 deer+0.070
 unity+0.069
 sorts+0.066
Neg logit contributions
angers-0.085
 salon-0.067
belt-0.064
ooter-0.063
ulk-0.063
Token He
Token ID679
Feature activation-5.84e-02

Pos logit contributions
angers+0.048
 salon+0.039
ooter+0.037
ulk+0.037
belt+0.037
Neg logit contributions
oc-0.032
 course-0.028
 thanks-0.026
 deer-0.026
 unity-0.024
Token watched
Token ID7342
Feature activation+3.15e-02

Pos logit contributions
oc+0.121
 course+0.114
 unity+0.109
 harmony+0.107
 sorts+0.105
Neg logit contributions
angers-0.096
 salon-0.078
outed-0.077
belt-0.074
aj-0.072
Token as
Token ID355
Feature activation+1.12e-02

Pos logit contributions
angers+0.069
 salon+0.052
amas+0.051
ulk+0.051
anger+0.051
Neg logit contributions
oc-0.058
 course-0.049
 sorts-0.047
 deer-0.046
 unity-0.045
Token the
Token ID262
Feature activation+5.76e-02

Pos logit contributions
angers+0.025
 salon+0.021
belt+0.020
ulk+0.020
outed+0.019
Neg logit contributions
oc-0.019
 course-0.017
 bond-0.015
 deer-0.015
 sorts-0.015
Token boat
Token ID8848
Feature activation+2.50e-02

Pos logit contributions
angers+0.121
 salon+0.099
ooter+0.097
belt+0.095
 bathroom+0.094
Neg logit contributions
oc-0.105
 course-0.091
 thanks-0.081
 sorts-0.080
 unity-0.076
Token "
Token ID366
Feature activation-6.18e-04

Pos logit contributions
angers+0.147
 salon+0.113
aj+0.111
ulk+0.110
ooter+0.110
Neg logit contributions
oc-0.144
 course-0.133
 deer-0.132
 thanks-0.125
 tiger-0.119
TokenWow
Token ID22017
Feature activation-2.03e-02

Pos logit contributions
oc+0.001
 course+0.001
 thanks+0.001
 unity+0.001
 deer+0.001
Neg logit contributions
angers-0.001
elt-0.001
 salon-0.001
ooter-0.001
ulk-0.001
Token,
Token ID11
Feature activation+4.43e-02

Pos logit contributions
oc+0.037
 course+0.034
 deer+0.032
 unity+0.031
 thanks+0.030
Neg logit contributions
angers-0.044
 salon-0.034
ooter-0.033
aj-0.033
ulk-0.032
Token you
Token ID345
Feature activation-4.46e-02

Pos logit contributions
 salon+0.070
angers+0.066
ooter+0.058
 door+0.057
 bathroom+0.056
Neg logit contributions
oc-0.098
 sorts-0.081
 harmony-0.079
 unity-0.079
able-0.078
Token are
Token ID389
Feature activation+2.43e-02

Pos logit contributions
oc+0.069
 course+0.061
 deer+0.059
 unity+0.056
 thanks+0.054
Neg logit contributions
angers-0.105
 salon-0.084
ooter-0.084
ulk-0.081
aj-0.081
Token so
Token ID523
Feature activation+2.00e-02

Pos logit contributions
angers+0.053
 salon+0.041
ooter+0.040
ulk+0.040
aj+0.040
Neg logit contributions
oc-0.044
 course-0.040
 deer-0.037
 thanks-0.036
 unity-0.035
Token clever
Token ID14169
Feature activation-1.19e-01

Pos logit contributions
angers+0.049
 salon+0.039
ooter+0.039
ulk+0.038
aj+0.038
Neg logit contributions
oc-0.031
 course-0.027
 deer-0.025
 thanks-0.024
 unity-0.024
Token!"
Token ID2474
Feature activation+4.20e-02

Pos logit contributions
oc+0.237
 course+0.212
 unity+0.208
 bond+0.198
 harmony+0.194
Neg logit contributions
angers-0.234
aj-0.169
elt-0.169
anger-0.167
ooter-0.167
Token
Token ID198
Feature activation-3.10e-02

Pos logit contributions
angers+0.098
 salon+0.079
belt+0.079
ooter+0.076
aj+0.076
Neg logit contributions
oc-0.069
 course-0.064
 deer-0.059
 thanks-0.058
 sorts-0.052
Token
Token ID198
Feature activation-3.81e-02

Pos logit contributions
oc+0.052
 course+0.046
 deer+0.043
 thanks+0.041
 sorts+0.039
Neg logit contributions
angers-0.073
 salon-0.059
ooter-0.057
belt-0.056
aj-0.055
TokenBut
Token ID1537
Feature activation-2.65e-03

Pos logit contributions
oc+0.064
 course+0.057
 deer+0.054
 thanks+0.053
 unity+0.051
Neg logit contributions
angers-0.088
 salon-0.071
ooter-0.068
ulk-0.068
aj-0.067
Token calm
Token ID9480
Feature activation-4.40e-02

Pos logit contributions
angers+0.078
 salon+0.060
ooter+0.058
ulk+0.057
aj+0.057
Neg logit contributions
oc-0.079
 course-0.073
 deer-0.070
 thanks-0.067
 unity-0.066
Token and
Token ID290
Feature activation-2.33e-02

Pos logit contributions
oc+0.056
 course+0.049
 thanks+0.048
 deer+0.048
 harmony+0.041
Neg logit contributions
angers-0.119
 salon-0.098
ooter-0.097
belt-0.095
anger-0.095
Token kind
Token ID1611
Feature activation-9.15e-02

Pos logit contributions
oc+0.037
 course+0.033
 deer+0.029
affe+0.027
 sorts+0.027
Neg logit contributions
angers-0.055
 salon-0.045
aj-0.045
outed-0.044
belt-0.044
Token.
Token ID13
Feature activation+1.90e-02

Pos logit contributions
oc+0.189
 course+0.175
 deer+0.165
 thanks+0.158
 unity+0.158
Neg logit contributions
angers-0.176
 salon-0.132
ooter-0.127
ulk-0.126
aj-0.125
Token She
Token ID1375
Feature activation-6.42e-02

Pos logit contributions
angers+0.033
 salon+0.023
ulk+0.022
ooter+0.022
aj+0.022
Neg logit contributions
oc-0.043
 course-0.041
 unity-0.038
 deer-0.038
 thanks-0.037
Token feels
Token ID5300
Feature activation+1.75e-02

Pos logit contributions
oc+0.115
 course+0.101
 deer+0.093
 unity+0.092
 harmony+0.090
Neg logit contributions
angers-0.139
 salon-0.113
aj-0.109
outed-0.108
ulk-0.107
Token a
Token ID257
Feature activation-3.43e-02

Pos logit contributions
angers+0.028
outed+0.020
ulk+0.020
 salon+0.020
 bathroom+0.019
Neg logit contributions
oc-0.041
 course-0.037
affe-0.034
 matter-0.033
 deer-0.033
Token little
Token ID1310
Feature activation-7.96e-02

Pos logit contributions
oc+0.054
 course+0.050
 sorts+0.044
 unity+0.044
 deer+0.043
Neg logit contributions
angers-0.078
 salon-0.064
amas-0.061
aj-0.061
ooter-0.061
Token bra
Token ID8290
Feature activation-1.94e-02

Pos logit contributions
oc+0.169
 course+0.166
 sorts+0.146
 unity+0.145
 Charlie+0.144
Neg logit contributions
angers-0.118
 salon-0.101
 bathroom-0.098
outed-0.093
ider-0.092
Tokenver
Token ID332
Feature activation-5.27e-02

Pos logit contributions
oc+0.036
 course+0.031
 deer+0.029
 thanks+0.028
affe+0.028
Neg logit contributions
angers-0.040
ulk-0.032
amas-0.032
aj-0.031
 salon-0.031
Token.
Token ID13
Feature activation+2.38e-02

Pos logit contributions
oc+0.116
 course+0.111
 unity+0.105
 deer+0.102
 thanks+0.101
Neg logit contributions
angers-0.084
 salon-0.065
ulk-0.060
belt-0.059
ooter-0.058
Tokenaps
Token ID1686
Feature activation-3.37e-02

Pos logit contributions
angers+0.046
 salon+0.036
belt+0.035
ooter+0.034
anger+0.032
Neg logit contributions
oc-0.063
 thanks-0.055
 course-0.055
 harmony-0.052
 sorts-0.052
Token his
Token ID465
Feature activation+5.93e-02

Pos logit contributions
oc+0.071
 course+0.062
 deer+0.060
 unity+0.058
 sorts+0.056
Neg logit contributions
angers-0.063
outed-0.047
ulk-0.047
 salon-0.046
aj-0.046
Token hands
Token ID2832
Feature activation+6.88e-03

Pos logit contributions
angers+0.106
aj+0.079
outed+0.077
ooter+0.076
 salon+0.076
Neg logit contributions
oc-0.130
 course-0.121
 deer-0.113
 sorts-0.108
 unity-0.107
Token.
Token ID13
Feature activation+2.82e-02

Pos logit contributions
angers+0.011
 salon+0.008
ooter+0.007
ulk+0.007
outed+0.007
Neg logit contributions
oc-0.016
 course-0.015
 deer-0.014
 unity-0.014
 sorts-0.013
Token
Token ID198
Feature activation-2.87e-02

Pos logit contributions
angers+0.072
 salon+0.062
ooter+0.060
belt+0.059
anger+0.057
Neg logit contributions
oc-0.039
 course-0.032
 thanks-0.032
 deer-0.032
 tiger-0.029
Token
Token ID198
Feature activation-3.44e-02

Pos logit contributions
oc+0.047
 course+0.041
 deer+0.038
 thanks+0.038
 sorts+0.035
Neg logit contributions
angers-0.069
 salon-0.056
ooter-0.054
belt-0.054
aj-0.052
Token"
Token ID1
Feature activation+2.71e-02

Pos logit contributions
oc+0.057
 course+0.051
 deer+0.048
 thanks+0.046
 unity+0.045
Neg logit contributions
angers-0.081
 salon-0.064
ooter-0.063
ulk-0.062
aj-0.062
TokenMore
Token ID5167
Feature activation-4.19e-02

Pos logit contributions
angers+0.057
 salon+0.046
ulk+0.045
elt+0.044
OM+0.043
Neg logit contributions
oc-0.050
 course-0.041
 deer-0.041
 thanks-0.040
 sorts-0.039
Token,
Token ID11
Feature activation+5.62e-02

Pos logit contributions
oc+0.082
 course+0.071
 deer+0.070
 unity+0.066
 thanks+0.065
Neg logit contributions
angers-0.086
 salon-0.067
aj-0.064
ooter-0.063
ulk-0.063
Token more
Token ID517
Feature activation-2.09e-02

Pos logit contributions
angers+0.096
 salon+0.081
 bathroom+0.078
ider+0.078
outed+0.068
Neg logit contributions
oc-0.122
 sorts-0.101
 deer-0.098
affe-0.094
 harmony-0.092
Token,
Token ID11
Feature activation+5.71e-02

Pos logit contributions
oc+0.039
 course+0.033
 deer+0.032
affe+0.029
 unity+0.029
Neg logit contributions
angers-0.045
aj-0.035
 salon-0.035
 bathroom-0.034
ulk-0.033
Token,
Token ID11
Feature activation+6.03e-02

Pos logit contributions
angers+0.067
belt+0.053
ooter+0.053
outed+0.052
ulk+0.052
Neg logit contributions
oc-0.038
 course-0.036
 thanks-0.034
 deer-0.031
 love-0.031
Token please
Token ID3387
Feature activation-3.42e-03

Pos logit contributions
angers+0.125
 salon+0.110
aj+0.099
 bathroom+0.098
ooter+0.098
Neg logit contributions
oc-0.111
 course-0.092
 deer-0.090
 sorts-0.088
 unity-0.085
Token ask
Token ID1265
Feature activation+1.14e-02

Pos logit contributions
oc+0.006
 course+0.005
 deer+0.005
 unity+0.004
 sorts+0.004
Neg logit contributions
angers-0.008
 salon-0.006
aj-0.006
ooter-0.006
ulk-0.006
Token me
Token ID502
Feature activation+6.49e-02

Pos logit contributions
angers+0.025
 salon+0.019
ooter+0.019
ulk+0.019
aj+0.019
Neg logit contributions
oc-0.020
 course-0.019
 deer-0.017
 thanks-0.017
 unity-0.017
Token before
Token ID878
Feature activation+5.70e-02

Pos logit contributions
angers+0.129
 salon+0.099
ooter+0.095
aj+0.095
ulk+0.092
Neg logit contributions
oc-0.130
 course-0.118
 deer-0.113
 thanks-0.107
 unity-0.107
Token you
Token ID345
Feature activation-1.32e-02

Pos logit contributions
angers+0.150
ooter+0.123
 salon+0.121
anger+0.119
belt+0.119
Neg logit contributions
oc-0.074
 course-0.070
 deer-0.063
 thanks-0.061
 unity-0.058
Token paint
Token ID7521
Feature activation+1.67e-02

Pos logit contributions
oc+0.020
 course+0.018
 deer+0.017
 unity+0.016
 thanks+0.016
Neg logit contributions
angers-0.031
 salon-0.025
ooter-0.025
aj-0.024
outed-0.024
Token toast
Token ID27805
Feature activation+3.32e-02

Pos logit contributions
angers+0.041
 salon+0.033
ooter+0.033
ulk+0.032
aj+0.032
Neg logit contributions
oc-0.025
 course-0.023
 deer-0.021
 thanks-0.020
 unity-0.019
Token.
Token ID13
Feature activation+1.62e-02

Pos logit contributions
angers+0.073
 salon+0.057
ooter+0.057
ulk+0.056
outed+0.055
Neg logit contributions
oc-0.058
 course-0.053
 deer-0.049
 thanks-0.048
 unity-0.046
Token OK
Token ID7477
Feature activation-3.94e-02

Pos logit contributions
angers+0.034
 salon+0.026
aj+0.026
ooter+0.026
elt+0.026
Neg logit contributions
oc-0.029
 course-0.027
 deer-0.026
 unity-0.024
 sorts-0.024
Token?"
Token ID1701
Feature activation+6.69e-02

Pos logit contributions
oc+0.067
 course+0.062
 thanks+0.058
 matter+0.058
 deer+0.058
Neg logit contributions
angers-0.094
ulk-0.071
ooter-0.070
boards-0.069
anger-0.069
Token sky
Token ID6766
Feature activation+3.49e-02

Pos logit contributions
angers+0.081
 salon+0.062
ooter+0.060
aj+0.059
ulk+0.058
Neg logit contributions
oc-0.085
 course-0.079
 deer-0.073
 thanks-0.071
 unity-0.071
Token was
Token ID373
Feature activation+7.43e-02

Pos logit contributions
angers+0.085
 salon+0.065
ooter+0.065
aj+0.065
ulk+0.064
Neg logit contributions
oc-0.058
 course-0.052
 thanks-0.049
 deer-0.049
 matter-0.046
Token turning
Token ID6225
Feature activation+2.21e-02

Pos logit contributions
angers+0.179
ooter+0.140
 salon+0.139
anger+0.138
ulk+0.136
Neg logit contributions
oc-0.111
 course-0.108
 deer-0.108
 thanks-0.105
 matter-0.099
Token dark
Token ID3223
Feature activation-3.17e-02

Pos logit contributions
angers+0.054
 salon+0.044
ulk+0.043
ooter+0.043
azes+0.042
Neg logit contributions
oc-0.033
 deer-0.030
 course-0.030
 thanks-0.029
 sorts-0.027
Token and
Token ID290
Feature activation-2.09e-02

Pos logit contributions
oc+0.050
 course+0.045
 deer+0.044
 thanks+0.043
 sorts+0.040
Neg logit contributions
angers-0.076
 salon-0.063
ooter-0.061
aj-0.060
outed-0.060
Token gray
Token ID12768
Feature activation-4.27e-02

Pos logit contributions
oc+0.032
 course+0.029
 deer+0.027
 thanks+0.026
 unity+0.025
Neg logit contributions
angers-0.051
 salon-0.041
ooter-0.040
aj-0.039
ulk-0.039
Token.
Token ID13
Feature activation+9.76e-03

Pos logit contributions
oc+0.069
 course+0.064
 deer+0.063
 thanks+0.060
 tiger+0.056
Neg logit contributions
angers-0.100
 salon-0.083
aj-0.081
ooter-0.079
ulk-0.078
Token Suddenly
Token ID24975
Feature activation+3.41e-03

Pos logit contributions
angers+0.029
 salon+0.026
ider+0.024
ulk+0.023
ips+0.023
Neg logit contributions
oc-0.011
 Charlie-0.009
 Fred-0.008
 Buddy-0.008
 Mickey-0.008
Token,
Token ID11
Feature activation+6.33e-02

Pos logit contributions
angers+0.007
 salon+0.006
ooter+0.005
aj+0.005
ulk+0.005
Neg logit contributions
oc-0.006
 course-0.006
 deer-0.005
 thanks-0.005
 unity-0.005
Token a
Token ID257
Feature activation-3.45e-02

Pos logit contributions
angers+0.144
 salon+0.115
ulk+0.114
belt+0.111
elt+0.110
Neg logit contributions
oc-0.106
 course-0.090
 thanks-0.083
 unity-0.083
 deer-0.082
Token big
Token ID1263
Feature activation-9.05e-02

Pos logit contributions
oc+0.056
 course+0.050
 deer+0.045
 thanks+0.045
 unity+0.044
Neg logit contributions
angers-0.080
 salon-0.065
ooter-0.063
ulk-0.061
aj-0.061
Token small
Token ID1402
Feature activation-6.15e-02

Pos logit contributions
oc+0.026
 deer+0.024
 course+0.022
 unity+0.022
affe+0.022
Neg logit contributions
angers-0.040
ulk-0.032
amas-0.032
uck-0.031
unes-0.031
Token bites
Token ID26081
Feature activation-2.70e-02

Pos logit contributions
oc+0.106
 course+0.099
 deer+0.097
 unity+0.093
affe+0.089
Neg logit contributions
angers-0.125
 salon-0.098
elt-0.098
aj-0.097
 bathroom-0.096
Token and
Token ID290
Feature activation-3.01e-02

Pos logit contributions
oc+0.047
 course+0.043
 deer+0.043
 unity+0.042
 thanks+0.039
Neg logit contributions
angers-0.057
 salon-0.043
unes-0.043
ooter-0.042
amas-0.042
Token taste
Token ID6938
Feature activation-1.21e-03

Pos logit contributions
oc+0.050
 course+0.044
 deer+0.042
affe+0.039
 sorts+0.038
Neg logit contributions
angers-0.063
 salon-0.054
ulk-0.052
 bathroom-0.052
aj-0.052
Token a
Token ID257
Feature activation-3.08e-02

Pos logit contributions
oc+0.002
 deer+0.002
 course+0.002
affe+0.002
 tiger+0.002
Neg logit contributions
angers-0.002
unes-0.002
ulk-0.002
aj-0.002
 bathroom-0.002
Token little
Token ID1310
Feature activation-7.68e-02

Pos logit contributions
 course+0.050
oc+0.049
 deer+0.044
 unity+0.042
 sorts+0.041
Neg logit contributions
angers-0.059
 salon-0.053
aj-0.052
 bathroom-0.051
amas-0.050
Token bit
Token ID1643
Feature activation-1.81e-02

Pos logit contributions
 course+0.126
oc+0.123
 deer+0.111
 unity+0.109
Charlie+0.106
Neg logit contributions
angers-0.143
 salon-0.127
 bathroom-0.125
elt-0.122
 door-0.116
Token first
Token ID717
Feature activation+3.26e-02

Pos logit contributions
oc+0.034
 course+0.031
 deer+0.030
 unity+0.029
 tiger+0.027
Neg logit contributions
angers-0.035
outed-0.028
elt-0.027
 salon-0.027
aj-0.027
Token to
Token ID284
Feature activation-1.79e-02

Pos logit contributions
angers+0.068
 salon+0.052
ooter+0.051
aj+0.051
belt+0.050
Neg logit contributions
oc-0.062
 course-0.056
 deer-0.053
 unity-0.051
 sorts-0.049
Token make
Token ID787
Feature activation+1.08e-02

Pos logit contributions
oc+0.032
 sorts+0.026
 course+0.026
 deer+0.025
 unity+0.025
Neg logit contributions
angers-0.034
 salon-0.030
 bathroom-0.028
ulk-0.027
ooter-0.027
Token sure
Token ID1654
Feature activation-2.44e-02

Pos logit contributions
angers+0.024
amas+0.019
ulk+0.018
azes+0.018
ews+0.018
Neg logit contributions
oc-0.018
 course-0.017
 deer-0.016
 unity-0.015
 bond-0.014
Tokenanger
Token ID2564
Feature activation+5.05e-02

Pos logit contributions
outed+0.038
ulk+0.038
 bathroom+0.032
 pant+0.032
elt+0.030
Neg logit contributions
oc-0.131
 course-0.122
affe-0.117
 thanks-0.111
osaurs-0.106
Token.
Token ID13
Feature activation+1.45e-02

Pos logit contributions
angers+0.108
 salon+0.084
ooter+0.083
belt+0.081
aj+0.080
Neg logit contributions
oc-0.093
 course-0.084
 deer-0.080
 unity-0.076
 thanks-0.075
Token It
Token ID632
Feature activation-2.17e-03

Pos logit contributions
angers+0.037
 salon+0.031
aj+0.029
ulk+0.029
belt+0.029
Neg logit contributions
oc-0.021
 course-0.018
 thanks-0.016
 deer-0.016
 tiger-0.015
Token was
Token ID373
Feature activation+6.27e-02

Pos logit contributions
oc+0.004
 course+0.004
 deer+0.004
 thanks+0.004
 unity+0.004
Neg logit contributions
angers-0.004
 salon-0.003
ooter-0.003
aj-0.003
ulk-0.003
Token perfect
Token ID2818
Feature activation-8.37e-02

Pos logit contributions
angers+0.139
 salon+0.110
ooter+0.106
aj+0.105
belt+0.104
Neg logit contributions
oc-0.112
 course-0.100
 deer-0.094
 thanks-0.089
 unity-0.089
Token for
Token ID329
Feature activation-4.97e-02

Pos logit contributions
oc+0.155
 course+0.140
 deer+0.132
 thanks+0.125
 unity+0.125
Neg logit contributions
angers-0.181
 salon-0.141
ooter-0.137
aj-0.135
ulk-0.134
Token his
Token ID465
Feature activation+4.48e-02

Pos logit contributions
oc+0.078
 course+0.069
 deer+0.064
 thanks+0.061
 unity+0.060
Neg logit contributions
angers-0.121
 salon-0.097
ooter-0.094
ulk-0.094
aj-0.093
Token red
Token ID2266
Feature activation-4.68e-02

Pos logit contributions
angers+0.096
ulk+0.076
amas+0.072
 salon+0.072
outed+0.071
Neg logit contributions
oc-0.080
 course-0.074
 deer-0.070
 thanks-0.069
 unity-0.066
Token shirt
Token ID10147
Feature activation+6.32e-02

Pos logit contributions
oc+0.090
 course+0.087
 deer+0.080
 love+0.076
 matter+0.074
Neg logit contributions
angers-0.089
ulk-0.072
azes-0.071
outed-0.070
ews-0.070
Token.
Token ID13
Feature activation+1.52e-02

Pos logit contributions
angers+0.135
 salon+0.104
ulk+0.103
aj+0.101
outed+0.101
Neg logit contributions
oc-0.114
 course-0.105
 deer-0.098
 thanks-0.096
 unity-0.091
Token
Token ID220
Feature activation+1.19e-01

Pos logit contributions
angers+0.033
 salon+0.026
ooter+0.025
ulk+0.025
aj+0.025
Neg logit contributions
oc-0.027
 course-0.025
 deer-0.023
 thanks-0.022
 unity-0.022
Token heard
Token ID2982
Feature activation+2.51e-02

Pos logit contributions
oc+0.041
 deer+0.040
 course+0.038
 thanks+0.035
 lion+0.033
Neg logit contributions
angers-0.075
ooter-0.060
belt-0.059
 salon-0.059
anger-0.058
Token a
Token ID257
Feature activation-2.11e-02

Pos logit contributions
angers+0.054
 salon+0.042
ulk+0.041
outed+0.040
ooter+0.040
Neg logit contributions
oc-0.046
 course-0.041
 deer-0.037
 unity-0.036
 thanks-0.036
Token loud
Token ID7812
Feature activation-1.10e-01

Pos logit contributions
oc+0.031
 course+0.028
 deer+0.025
 thanks+0.025
 unity+0.025
Neg logit contributions
angers-0.052
 salon-0.042
ooter-0.041
ulk-0.041
aj-0.040
Token boom
Token ID14166
Feature activation-4.78e-03

Pos logit contributions
oc+0.193
 course+0.176
 deer+0.163
 unity+0.161
 thanks+0.160
Neg logit contributions
angers-0.234
 salon-0.184
ooter-0.180
ulk-0.179
amas-0.176
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
oc+0.009
 course+0.008
 deer+0.008
 thanks+0.007
 unity+0.007
Neg logit contributions
angers-0.010
 salon-0.008
ooter-0.008
aj-0.008
belt-0.008
Token They
Token ID1119
Feature activation-9.36e-02

Pos logit contributions
angers+0.028
 salon+0.025
ooter+0.023
aj+0.023
ulk+0.023
Neg logit contributions
oc-0.014
 deer-0.012
 course-0.012
 thanks-0.011
 Freddie-0.010
Token looked
Token ID3114
Feature activation+6.97e-02

Pos logit contributions
oc+0.177
 course+0.162
 deer+0.153
 thanks+0.147
 unity+0.145
Neg logit contributions
angers-0.196
 salon-0.151
ooter-0.147
aj-0.144
ulk-0.144
Token up
Token ID510
Feature activation+2.60e-02

Pos logit contributions
angers+0.107
ooter+0.073
outed+0.072
 salon+0.072
ulk+0.071
Neg logit contributions
oc-0.168
 course-0.156
 unity-0.142
 deer-0.138
 harmony-0.137
Token and
Token ID290
Feature activation-2.53e-02

Pos logit contributions
angers+0.062
 salon+0.050
ooter+0.048
ulk+0.048
aj+0.048
Neg logit contributions
oc-0.042
 course-0.037
 deer-0.034
 unity-0.032
 thanks-0.032
Token saw
Token ID2497
Feature activation+3.00e-02

Pos logit contributions
oc+0.045
 course+0.041
 deer+0.040
 thanks+0.038
 unity+0.037
Neg logit contributions
angers-0.055
 salon-0.043
ooter-0.042
aj-0.042
ulk-0.041
Token dark
Token ID3223
Feature activation-3.32e-02

Pos logit contributions
angers+0.056
 bathroom+0.044
 salon+0.044
ulk+0.043
outed+0.042
Neg logit contributions
oc-0.063
 course-0.052
affe-0.048
 matter-0.048
 bond-0.047
Token,
Token ID11
Feature activation+5.33e-02

Pos logit contributions
oc+0.060
 course+0.055
 deer+0.052
 thanks+0.050
 unity+0.048
Neg logit contributions
angers-0.081
 salon-0.064
ooter-0.063
ulk-0.062
aj-0.062
Token but
Token ID475
Feature activation-3.03e-02

Pos logit contributions
angers+0.104
 salon+0.083
belt+0.075
elt+0.075
 bathroom+0.075
Neg logit contributions
oc-0.113
 unity-0.092
 course-0.092
affe-0.088
 deer-0.088
Token she
Token ID673
Feature activation-1.17e-02

Pos logit contributions
oc+0.063
 course+0.055
 deer+0.052
 unity+0.050
 thanks+0.050
Neg logit contributions
angers-0.058
 salon-0.046
outed-0.043
belt-0.043
aj-0.042
Token was
Token ID373
Feature activation+6.56e-02

Pos logit contributions
oc+0.019
 course+0.017
 deer+0.016
 thanks+0.015
 unity+0.015
Neg logit contributions
angers-0.028
 salon-0.022
ooter-0.022
ulk-0.022
aj-0.022
Token very
Token ID845
Feature activation+1.45e-02

Pos logit contributions
angers+0.165
anger+0.136
amas+0.131
ooter+0.131
ulk+0.129
Neg logit contributions
oc-0.092
 deer-0.087
 thanks-0.086
 course-0.085
 love-0.079
Token brave
Token ID14802
Feature activation-8.53e-02

Pos logit contributions
angers+0.037
ulk+0.030
 salon+0.030
amas+0.030
anger+0.030
Neg logit contributions
oc-0.020
 deer-0.017
 course-0.017
 thanks-0.017
 love-0.016
Token and
Token ID290
Feature activation-1.78e-02

Pos logit contributions
oc+0.193
 course+0.181
 unity+0.169
 deer+0.163
 harmony+0.158
Neg logit contributions
angers-0.145
 salon-0.103
aj-0.100
elt-0.099
ooter-0.097
Token adventurous
Token ID39770
Feature activation-9.45e-02

Pos logit contributions
oc+0.030
 course+0.027
 deer+0.025
 thanks+0.023
 unity+0.023
Neg logit contributions
angers-0.041
 salon-0.033
ooter-0.032
belt-0.031
outed-0.031
Token.
Token ID13
Feature activation+8.39e-03

Pos logit contributions
oc+0.204
 course+0.188
 deer+0.175
 unity+0.171
 thanks+0.167
Neg logit contributions
angers-0.175
 salon-0.130
ooter-0.126
ulk-0.123
aj-0.122
Token One
Token ID1881
Feature activation-4.01e-02

Pos logit contributions
angers+0.020
 salon+0.016
ooter+0.016
belt+0.016
outed+0.015
Neg logit contributions
oc-0.013
 course-0.012
 thanks-0.011
 deer-0.011
 sorts-0.010
Token day
Token ID1110
Feature activation+3.48e-02

Pos logit contributions
oc+0.064
 course+0.058
 deer+0.053
 thanks+0.051
 unity+0.051
Neg logit contributions
angers-0.095
 salon-0.077
ooter-0.075
aj-0.073
belt-0.073
Token big
Token ID1263
Feature activation-9.49e-02

Pos logit contributions
oc+0.077
 course+0.068
 deer+0.063
 thanks+0.060
 unity+0.059
Neg logit contributions
angers-0.135
 salon-0.109
ooter-0.107
ulk-0.106
aj-0.106
Token storm
Token ID6388
Feature activation-2.16e-03

Pos logit contributions
oc+0.168
 course+0.142
 unity+0.132
 sorts+0.128
 thanks+0.127
Neg logit contributions
angers-0.200
 salon-0.172
ooter-0.163
 bathroom-0.156
 office-0.153
Token came
Token ID1625
Feature activation+8.81e-02

Pos logit contributions
oc+0.004
 course+0.004
 deer+0.004
 thanks+0.004
 matter+0.003
Neg logit contributions
angers-0.004
aj-0.003
 salon-0.003
ooter-0.003
belt-0.003
Token with
Token ID351
Feature activation-3.63e-02

Pos logit contributions
angers+0.247
 salon+0.213
belt+0.204
aj+0.201
ulk+0.201
Neg logit contributions
oc-0.109
 course-0.091
 deer-0.090
 thanks-0.088
 sorts-0.076
Token lots
Token ID6041
Feature activation-7.76e-02

Pos logit contributions
oc+0.070
 course+0.058
 deer+0.055
 unity+0.055
 thanks+0.053
Neg logit contributions
angers-0.073
outed-0.058
elt-0.057
ulk-0.056
 salon-0.056
Token of
Token ID286
Feature activation-9.11e-02

Pos logit contributions
oc+0.129
 course+0.108
 unity+0.105
 deer+0.103
 bond+0.099
Neg logit contributions
angers-0.176
 salon-0.140
ooter-0.138
anger-0.136
elt-0.135
Token wind
Token ID2344
Feature activation+3.79e-02

Pos logit contributions
oc+0.175
 course+0.147
 thanks+0.136
 deer+0.135
 unity+0.133
Neg logit contributions
angers-0.187
 salon-0.144
outed-0.143
ews-0.142
aj-0.141
Token and
Token ID290
Feature activation-3.07e-02

Pos logit contributions
angers+0.091
 salon+0.075
omm+0.073
ooter+0.073
belt+0.072
Neg logit contributions
oc-0.055
 course-0.051
 deer-0.050
 thanks-0.049
 tiger-0.043
Token rain
Token ID6290
Feature activation+3.58e-02

Pos logit contributions
oc+0.057
 course+0.051
 deer+0.046
 unity+0.045
 thanks+0.045
Neg logit contributions
angers-0.064
 salon-0.050
ulk-0.049
outed-0.048
elt-0.048
Token.
Token ID13
Feature activation+3.84e-03

Pos logit contributions
angers+0.086
omm+0.072
 salon+0.071
addy+0.070
azes+0.068
Neg logit contributions
oc-0.048
 course-0.047
 thanks-0.046
 love-0.045
 deer-0.042
Token Tim
Token ID5045
Feature activation-1.17e-02

Pos logit contributions
angers+0.010
 salon+0.009
aj+0.008
ulk+0.008
belt+0.008
Neg logit contributions
oc-0.005
 course-0.005
 deer-0.004
 thanks-0.004
 tiger-0.004
Token.
Token ID13
Feature activation+1.49e-02

Pos logit contributions
angers+0.015
 salon+0.012
ooter+0.012
aj+0.011
ulk+0.011
Neg logit contributions
oc-0.014
 course-0.013
 deer-0.012
 thanks-0.012
 unity-0.012
Token "
Token ID366
Feature activation-1.04e-02

Pos logit contributions
angers+0.032
 salon+0.026
aj+0.025
ulk+0.025
belt+0.025
Neg logit contributions
oc-0.025
 course-0.022
 thanks-0.021
 deer-0.021
 Charlie-0.021
TokenWhat
Token ID2061
Feature activation-1.33e-02

Pos logit contributions
oc+0.022
 deer+0.020
 thanks+0.019
 course+0.019
 shields+0.019
Neg logit contributions
angers-0.017
elt-0.014
 salon-0.014
outed-0.013
belt-0.013
Token do
Token ID466
Feature activation-3.19e-02

Pos logit contributions
oc+0.025
 course+0.022
 deer+0.021
 unity+0.019
 harmony+0.019
Neg logit contributions
angers-0.026
 salon-0.022
ooter-0.021
azes-0.020
outed-0.020
Token you
Token ID345
Feature activation-3.89e-02

Pos logit contributions
oc+0.055
 course+0.046
 deer+0.043
 thanks+0.042
 sorts+0.041
Neg logit contributions
angers-0.070
 salon-0.057
ooter-0.056
belt-0.055
azes-0.055
Token mean
Token ID1612
Feature activation-8.38e-02

Pos logit contributions
oc+0.058
 course+0.049
 deer+0.048
 unity+0.043
 tiger+0.042
Neg logit contributions
angers-0.093
ooter-0.077
 salon-0.077
outed-0.074
azes-0.074
Token?"
Token ID1701
Feature activation+5.92e-02

Pos logit contributions
oc+0.181
 course+0.168
 deer+0.160
 thanks+0.155
 unity+0.152
Neg logit contributions
angers-0.153
 salon-0.113
ooter-0.109
ulk-0.107
aj-0.107
Token
Token ID198
Feature activation-2.18e-02

Pos logit contributions
angers+0.120
 salon+0.092
ooter+0.089
ulk+0.088
aj+0.087
Neg logit contributions
oc-0.116
 course-0.106
 deer-0.101
 thanks-0.097
 unity-0.096
Token
Token ID198
Feature activation-2.84e-02

Pos logit contributions
oc+0.033
 course+0.029
 deer+0.026
 thanks+0.024
 Charlie+0.023
Neg logit contributions
angers-0.056
 salon-0.045
unes-0.044
outed-0.043
ooter-0.043
Token"
Token ID1
Feature activation+3.12e-02

Pos logit contributions
oc+0.042
 course+0.037
 deer+0.034
 unity+0.031
 thanks+0.030
Neg logit contributions
angers-0.074
 salon-0.059
unes-0.058
ooter-0.057
aj-0.057
TokenWe
Token ID1135
Feature activation-9.07e-02

Pos logit contributions
angers+0.061
 salon+0.048
outed+0.045
elt+0.045
ooter+0.045
Neg logit contributions
oc-0.065
 course-0.056
 thanks-0.055
 deer-0.055
 unity-0.051
Token play
Token ID711
Feature activation+5.48e-03

Pos logit contributions
oc+0.019
affe+0.016
 course+0.015
 tiger+0.014
able+0.014
Neg logit contributions
angers-0.015
 bathroom-0.013
 salon-0.013
ulk-0.012
line-0.012
Token and
Token ID290
Feature activation-2.34e-02

Pos logit contributions
angers+0.011
 salon+0.008
ulk+0.008
anger+0.008
amas+0.008
Neg logit contributions
oc-0.010
 course-0.009
 deer-0.009
 unity-0.009
 thanks-0.009
Token the
Token ID262
Feature activation+4.89e-02

Pos logit contributions
oc+0.046
 course+0.041
affe+0.037
 deer+0.035
 unity+0.034
Neg logit contributions
angers-0.043
 salon-0.036
ulk-0.034
 bathroom-0.033
ooter-0.033
Token birds
Token ID10087
Feature activation-1.89e-02

Pos logit contributions
angers+0.110
 salon+0.088
aj+0.083
ooter+0.083
amas+0.083
Neg logit contributions
oc-0.087
 course-0.075
 deer-0.068
 sorts-0.067
 thanks-0.067
Token sang
Token ID25889
Feature activation-2.17e-02

Pos logit contributions
oc+0.038
 course+0.033
 unity+0.033
 matter+0.031
 harmony+0.031
Neg logit contributions
angers-0.034
 salon-0.027
outed-0.025
elt-0.025
uddled-0.025
Token beautiful
Token ID4950
Feature activation-9.36e-02

Pos logit contributions
oc+0.041
 course+0.037
 deer+0.032
 matter+0.032
 unity+0.031
Neg logit contributions
angers-0.043
ulk-0.032
itter-0.032
amas-0.031
uddled-0.031
Token songs
Token ID7259
Feature activation-7.10e-02

Pos logit contributions
oc+0.128
 course+0.115
 deer+0.101
 unity+0.097
 Gary+0.093
Neg logit contributions
angers-0.231
 salon-0.185
aj-0.182
ulk-0.182
amas-0.181
Token.
Token ID13
Feature activation+1.60e-02

Pos logit contributions
oc+0.181
 course+0.160
 matter+0.151
 deer+0.149
 unity+0.149
Neg logit contributions
angers-0.102
 salon-0.071
anger-0.069
ooter-0.065
uddled-0.062
Token
Token ID220
Feature activation+1.16e-01

Pos logit contributions
angers+0.033
ooter+0.025
outed+0.025
 salon+0.025
ulk+0.024
Neg logit contributions
oc-0.030
 course-0.027
 deer-0.026
 matter-0.025
 unity-0.024
Token
Token ID198
Feature activation-3.66e-02

Pos logit contributions
angers+0.280
 salon+0.237
belt+0.223
ooter+0.223
ulk+0.216
Neg logit contributions
oc-0.181
 course-0.157
 thanks-0.145
 deer-0.142
 unity-0.135
Token
Token ID198
Feature activation-4.26e-02

Pos logit contributions
oc+0.073
 course+0.069
 deer+0.065
 thanks+0.064
 unity+0.063
Neg logit contributions
angers-0.070
 salon-0.051
ulk-0.051
ooter-0.050
aj-0.049