TOP ACTIVATIONS
MAX = 0.110

could be and never mis beh aved . <|endoftext|> Once upon
" ch oo cho o cho o ". They
be happy and well - beh aved . The end .
a little bit of mis beh aving . The
the sky before they walked fore shadow ed the down pour
told Max to stop mis beh aving . But he didn
to go to work . Tears started flowing down his cheeks
became friendly and well - beh aved , just like Buddy
night , the wild cat lur ked through the woods ,
needle or a st eth oscope . They did not want
very naughty and always mis beh aving . One day she
listen . He always mis beh aved and caused trouble wherever
knee and it started to bleed . Tim
saw some other kids mis beh aving and running around .
play . THE <|endoftext|> Once upon a time
went to his room and sul ked . Then
for being so well - beh aved and Jack was delighted
girl didn 't understand . Tears started rolling down her face
told him to stop mis beh aving and to be kind
. THE <|endoftext|> Once upon a time

MIN ACTIVATIONS
MIN = -0.094

, fourteen , fifteen , sixteen , seventeen , eighteen ,
, seven . There were seven ants in the line .
. " They can help each other fight the bad guys
think . They look at each other . They have an
tomato . They looked at each other . They felt bad
to climb . They help each other when the ground is
. Mom said they could each choose one fruit to take
was . They looked at each other and felt sorry .
hoop . They look at each other . "
stick . They look at each other . They
" Tom , what are you doing with that worm
say " sorry " to each other . They hug and
, sweet ie . What are you expecting ?" Dad says
looks . She sees the ele venth pengu in . She
They look at each other . They hug the
and grateful . They loved each other very much . <|endoftext|>
, sixteen , seventeen , eighteen , nineteen , twenty ,
, fourteen , fifteen , sixteen , seventeen , eighteen ,
They learned to recommend to each other what was best for
Then they try to find each other 's fruit . They

INTERVAL 0.059 - 0.110
CONTAINS 0.024%

for treasures . When it was time to go
broken and it didn 't work anymore . Lily was sad
history . Every day and night , a spray of water
their medicine and stuck a therm ometer in their mouth .
have to pay one dollar per cup . Or you can

INTERVAL 0.008 - 0.059
CONTAINS 29.917%

, " L ily , you have to eat your broccoli
it and saw the pepper resting in a bush . He
said sorry and helped Lily print her drawings .
a time , there was a mighty land . In this
The silly refrigerator and the little girl became the

INTERVAL -0.043 - 0.008
CONTAINS 69.740%

with their new things . They are very happy . They
happy and he thanked his mom my for getting him a
bear looked up and saw a friendly ot ter . "
is hurt . Tom is scared . He runs
a fancy cabin in the woods . She had a pet

INTERVAL -0.094 - -0.043
CONTAINS 0.319%

Lily said . " Can we take some ?"
He asked Annie what she was doing under the post .
, " Of course ! Thank you for offering to help
but he cannot grab the o at . Tom
a big board at the edge of the pool . She
Token could
Token ID714
Feature activation+7.18e-03

Pos logit contributions
azes+0.047
 writing+0.041
isdom+0.041
 patiently+0.041
pit+0.040
Neg logit contributions
sc-0.026
ch-0.026
Br-0.025
B-0.025
 subs-0.025
Token be
Token ID307
Feature activation-1.47e-02

Pos logit contributions
B+0.009
 subs+0.008
ch+0.008
ched+0.008
sc+0.008
Neg logit contributions
azes-0.015
ires-0.013
pit-0.013
 fridge-0.013
 oven-0.013
Token and
Token ID290
Feature activation+1.05e-02

Pos logit contributions
azes+0.028
pit+0.024
mmm+0.023
ires+0.023
isdom+0.023
Neg logit contributions
ch-0.021
B-0.021
sc-0.021
 subs-0.020
ched-0.019
Token never
Token ID1239
Feature activation+2.48e-02

Pos logit contributions
ch+0.014
sc+0.014
B+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.021
pit-0.018
mmm-0.017
ement-0.017
ires-0.017
Token mis
Token ID2984
Feature activation+6.79e-02

Pos logit contributions
sc+0.036
B+0.035
ch+0.034
text+0.032
Fl+0.031
Neg logit contributions
azes-0.048
 writing-0.041
isdom-0.039
pit-0.039
castle-0.039
Tokenbeh
Token ID20709
Feature activation+9.44e-02

Pos logit contributions
sc+0.079
ch+0.078
 subs+0.075
B+0.074
fl+0.072
Neg logit contributions
azes-0.143
isdom-0.121
ires-0.118
eria-0.117
pit-0.117
Tokenaved
Token ID9586
Feature activation+8.32e-03

Pos logit contributions
ch+0.134
B+0.133
sc+0.133
 subs+0.128
bo+0.120
Neg logit contributions
azes-0.179
pit-0.151
isdom-0.148
mmm-0.147
ires-0.147
Token.
Token ID13
Feature activation+1.74e-03

Pos logit contributions
ch+0.011
sc+0.011
B+0.011
 subs+0.011
bo+0.010
Neg logit contributions
azes-0.016
pit-0.014
ement-0.013
isdom-0.013
ires-0.013
Token<|endoftext|>
Token ID50256
Feature activation+1.71e-02

Pos logit contributions
sc+0.002
ch+0.002
B+0.002
 subs+0.002
ched+0.002
Neg logit contributions
azes-0.004
pit-0.003
isdom-0.003
mmm-0.003
ires-0.003
TokenOnce
Token ID7454
Feature activation+1.48e-02

Pos logit contributions
 subs+0.024
ch+0.024
sc+0.024
B+0.023
 scrap+0.023
Neg logit contributions
azes-0.031
pit-0.026
ement-0.026
 fridge-0.025
 refrigerator-0.025
Token upon
Token ID2402
Feature activation+2.83e-02

Pos logit contributions
ch+0.021
B+0.021
sc+0.021
 subs+0.020
bo+0.019
Neg logit contributions
azes-0.028
pit-0.023
mmm-0.023
isdom-0.023
 oven-0.023
Token "
Token ID366
Feature activation+1.13e-02

Pos logit contributions
B+0.038
ch+0.038
sc+0.038
 subs+0.036
ched+0.034
Neg logit contributions
azes-0.062
pit-0.053
 oven-0.052
isdom-0.052
mmm-0.052
Tokench
Token ID354
Feature activation+5.21e-02

Pos logit contributions
sc+0.012
ch+0.012
 subs+0.012
B+0.011
text+0.011
Neg logit contributions
azes-0.023
mmm-0.021
pit-0.021
ocolate-0.021
ement-0.020
Tokenoo
Token ID2238
Feature activation+5.07e-02

Pos logit contributions
 subs+0.088
B+0.087
sc+0.084
Free+0.082
 scrap+0.082
Neg logit contributions
azes-0.082
ocolate-0.068
mmm-0.067
pit-0.065
 oven-0.065
Token cho
Token ID1727
Feature activation+8.76e-02

Pos logit contributions
B+0.075
sc+0.074
 subs+0.073
Free+0.068
 scrap+0.067
Neg logit contributions
azes-0.084
mmm-0.075
 oven-0.074
 patiently-0.072
ocolate-0.072
Tokeno
Token ID78
Feature activation+3.55e-02

Pos logit contributions
 subs+0.155
 scrap+0.148
B+0.147
Free+0.140
sc+0.140
Neg logit contributions
azes-0.129
ocolate-0.113
ruits-0.105
mmm-0.103
sing-0.101
Token cho
Token ID1727
Feature activation+9.33e-02

Pos logit contributions
B+0.048
 subs+0.046
sc+0.044
 scrap+0.043
Free+0.043
Neg logit contributions
azes-0.061
 oven-0.056
mmm-0.055
ocolate-0.055
ummy-0.053
Tokeno
Token ID78
Feature activation+3.87e-02

Pos logit contributions
 subs+0.162
 scrap+0.157
B+0.154
Free+0.150
sc+0.149
Neg logit contributions
azes-0.135
ocolate-0.120
sing-0.109
ruits-0.109
mmm-0.105
Token".
Token ID1911
Feature activation-7.27e-03

Pos logit contributions
B+0.054
 subs+0.052
Free+0.049
 scrap+0.049
sc+0.049
Neg logit contributions
azes-0.063
 oven-0.058
ocolate-0.056
sing-0.056
mmm-0.056
Token
Token ID198
Feature activation+2.96e-03

Pos logit contributions
azes+0.015
pit+0.013
isdom+0.013
ires+0.013
ement+0.012
Neg logit contributions
ch-0.009
B-0.009
sc-0.009
 subs-0.009
bo-0.008
Token
Token ID198
Feature activation-2.74e-03

Pos logit contributions
ch+0.003
sc+0.003
B+0.003
 subs+0.003
ched+0.003
Neg logit contributions
azes-0.006
pit-0.005
mmm-0.005
isdom-0.005
ires-0.005
TokenThey
Token ID2990
Feature activation-4.40e-03

Pos logit contributions
azes+0.006
pit+0.005
isdom+0.005
ires+0.005
mmm+0.005
Neg logit contributions
ch-0.003
B-0.003
sc-0.003
 subs-0.003
bo-0.003
Token be
Token ID307
Feature activation-1.24e-02

Pos logit contributions
 subs+0.003
sc+0.003
bo+0.003
ch+0.003
B+0.003
Neg logit contributions
azes-0.005
ires-0.005
pit-0.005
ummy-0.004
ement-0.004
Token happy
Token ID3772
Feature activation-7.11e-03

Pos logit contributions
azes+0.021
pit+0.018
mmm+0.017
isdom+0.017
ires+0.017
Neg logit contributions
ch-0.020
sc-0.020
B-0.020
 subs-0.019
bo-0.018
Token and
Token ID290
Feature activation+1.22e-02

Pos logit contributions
azes+0.017
mmm+0.015
ocolate+0.015
ruits+0.015
isdom+0.015
Neg logit contributions
B-0.006
 subs-0.006
ch-0.006
fl-0.006
sc-0.006
Token well
Token ID880
Feature activation+3.09e-02

Pos logit contributions
ch+0.020
B+0.020
sc+0.020
 subs+0.019
bo+0.018
Neg logit contributions
azes-0.021
pit-0.017
isdom-0.017
mmm-0.017
 writing-0.017
Token-
Token ID12
Feature activation+3.96e-02

Pos logit contributions
sc+0.038
ch+0.038
B+0.038
 subs+0.036
ched+0.034
Neg logit contributions
azes-0.065
pit-0.055
mmm-0.054
isdom-0.054
ires-0.054
Tokenbeh
Token ID20709
Feature activation+8.59e-02

Pos logit contributions
 subs+0.047
sc+0.046
ch+0.046
text+0.046
B+0.045
Neg logit contributions
azes-0.082
isdom-0.072
mmm-0.070
ires-0.069
castle-0.069
Tokenaved
Token ID9586
Feature activation+8.28e-03

Pos logit contributions
ch+0.109
sc+0.109
B+0.108
 subs+0.104
bo+0.097
Neg logit contributions
azes-0.176
pit-0.150
isdom-0.147
mmm-0.147
ires-0.147
Token.
Token ID13
Feature activation-2.24e-04

Pos logit contributions
ch+0.011
B+0.011
sc+0.011
 subs+0.011
bo+0.010
Neg logit contributions
azes-0.016
pit-0.014
ement-0.014
isdom-0.014
 oven-0.014
Token The
Token ID383
Feature activation+2.24e-02

Pos logit contributions
azes+0.001
pit+0.000
ires+0.000
ement+0.000
isdom+0.000
Neg logit contributions
B-0.000
ch-0.000
sc-0.000
 subs-0.000
bo-0.000
Token end
Token ID886
Feature activation+1.92e-02

Pos logit contributions
ch+0.028
B+0.028
sc+0.027
 subs+0.026
 scrap+0.025
Neg logit contributions
azes-0.047
mmm-0.041
isdom-0.040
pit-0.039
ires-0.039
Token.
Token ID13
Feature activation+6.53e-03

Pos logit contributions
sc+0.026
B+0.026
ch+0.025
 subs+0.025
ched+0.023
Neg logit contributions
azes-0.037
pit-0.031
 oven-0.031
isdom-0.031
ement-0.031
Token a
Token ID257
Feature activation-1.90e-03

Pos logit contributions
ch+0.011
 subs+0.010
sc+0.010
ched+0.010
B+0.010
Neg logit contributions
azes-0.013
pit-0.011
ement-0.011
eria-0.011
mmm-0.011
Token little
Token ID1310
Feature activation-5.69e-04

Pos logit contributions
azes+0.004
pit+0.003
 fridge+0.003
 oven+0.003
isdom+0.003
Neg logit contributions
sc-0.002
ch-0.002
B-0.002
 subs-0.002
ched-0.002
Token bit
Token ID1643
Feature activation+1.36e-02

Pos logit contributions
azes+0.001
ires+0.001
pit+0.001
mmm+0.001
isdom+0.001
Neg logit contributions
ch-0.001
sc-0.001
B-0.001
 subs-0.001
bo-0.001
Token of
Token ID286
Feature activation+5.73e-03

Pos logit contributions
ch+0.017
sc+0.017
B+0.017
 subs+0.016
ched+0.015
Neg logit contributions
azes-0.028
pit-0.024
isdom-0.024
mmm-0.024
ires-0.023
Token mis
Token ID2984
Feature activation+4.54e-02

Pos logit contributions
ch+0.008
B+0.007
sc+0.007
 scrap+0.007
 subs+0.007
Neg logit contributions
azes-0.012
pit-0.010
eria-0.010
castle-0.010
ires-0.010
Tokenbeh
Token ID20709
Feature activation+8.52e-02

Pos logit contributions
ch+0.051
sc+0.051
B+0.051
 subs+0.049
ched+0.045
Neg logit contributions
azes-0.099
pit-0.086
isdom-0.084
mmm-0.084
ires-0.084
Tokenaving
Token ID2703
Feature activation+2.04e-02

Pos logit contributions
sc+0.119
ch+0.117
B+0.114
 subs+0.109
ched+0.108
Neg logit contributions
azes-0.166
pit-0.144
ires-0.143
ement-0.142
mmm-0.141
Token.
Token ID13
Feature activation-7.33e-03

Pos logit contributions
ch+0.029
B+0.029
sc+0.029
 subs+0.028
Br+0.026
Neg logit contributions
azes-0.037
pit-0.032
 oven-0.032
 fridge-0.032
ement-0.031
Token
Token ID198
Feature activation+8.57e-04

Pos logit contributions
azes+0.017
castle+0.017
pit+0.016
 balcony+0.015
 throne+0.015
Neg logit contributions
B-0.007
sc-0.006
ch-0.006
bo-0.006
 subs-0.006
Token
Token ID198
Feature activation-3.42e-03

Pos logit contributions
B+0.001
ch+0.001
sc+0.001
 subs+0.001
ched+0.001
Neg logit contributions
azes-0.002
pit-0.002
castle-0.002
 oven-0.002
 fridge-0.002
TokenThe
Token ID464
Feature activation+4.16e-03

Pos logit contributions
azes+0.007
pit+0.006
mmm+0.006
isdom+0.006
ires+0.006
Neg logit contributions
ch-0.004
sc-0.004
B-0.004
 subs-0.004
ched-0.004
Token the
Token ID262
Feature activation+7.43e-03

Pos logit contributions
sc+0.006
ch+0.006
B+0.005
 subs+0.005
 scrap+0.005
Neg logit contributions
azes-0.015
mmm-0.013
isdom-0.013
ires-0.013
ruits-0.013
Token sky
Token ID6766
Feature activation+8.08e-03

Pos logit contributions
ch+0.014
sc+0.013
B+0.013
 subs+0.013
bo+0.013
Neg logit contributions
azes-0.012
mmm-0.009
isdom-0.009
ires-0.009
ement-0.009
Token before
Token ID878
Feature activation+2.41e-02

Pos logit contributions
ch+0.009
sc+0.009
fl+0.008
 subs+0.008
bo+0.008
Neg logit contributions
azes-0.018
isdom-0.016
ocolate-0.016
mmm-0.016
ires-0.016
Token they
Token ID484
Feature activation-1.30e-02

Pos logit contributions
ch+0.027
B+0.026
 subs+0.024
sc+0.023
fl+0.022
Neg logit contributions
azes-0.055
pit-0.049
ires-0.048
ruits-0.048
isdom-0.048
Token walked
Token ID6807
Feature activation+1.87e-02

Pos logit contributions
azes+0.031
pit+0.027
mmm+0.027
ement+0.027
ires+0.027
Neg logit contributions
ch-0.013
sc-0.013
 subs-0.012
B-0.012
bo-0.012
Token fore
Token ID1674
Feature activation+8.36e-02

Pos logit contributions
ch+0.023
sc+0.022
B+0.022
 subs+0.021
ched+0.020
Neg logit contributions
azes-0.040
pit-0.034
isdom-0.034
mmm-0.033
ires-0.033
Tokenshadow
Token ID19106
Feature activation+1.83e-02

Pos logit contributions
 subs+0.125
text+0.125
B+0.123
ch+0.119
fl+0.117
Neg logit contributions
azes-0.145
 patiently-0.129
eria-0.120
pit-0.120
isdom-0.119
Tokened
Token ID276
Feature activation-8.98e-03

Pos logit contributions
sc+0.016
ch+0.016
B+0.016
 subs+0.015
 scrap+0.014
Neg logit contributions
azes-0.043
ement-0.039
pit-0.038
ires-0.038
ummy-0.037
Token the
Token ID262
Feature activation+2.04e-02

Pos logit contributions
azes+0.019
pit+0.016
mmm+0.016
isdom+0.016
ires+0.016
Neg logit contributions
sc-0.011
ch-0.011
B-0.011
 subs-0.011
ched-0.010
Token down
Token ID866
Feature activation-1.35e-02

Pos logit contributions
ch+0.031
sc+0.030
B+0.030
 subs+0.030
bo+0.028
Neg logit contributions
azes-0.037
pit-0.031
isdom-0.030
ires-0.030
mmm-0.030
Tokenpour
Token ID48681
Feature activation+5.17e-02

Pos logit contributions
azes+0.029
isdom+0.024
mmm+0.024
pit+0.024
ires+0.024
Neg logit contributions
sc-0.016
ch-0.016
 subs-0.016
B-0.015
ched-0.014
Token told
Token ID1297
Feature activation+6.71e-03

Pos logit contributions
sc+0.017
ch+0.017
B+0.017
 subs+0.015
text+0.015
Neg logit contributions
azes-0.029
 fridge-0.025
 oven-0.025
 writing-0.025
 patiently-0.025
Token Max
Token ID5436
Feature activation+7.03e-03

Pos logit contributions
ch+0.010
B+0.010
sc+0.009
 subs+0.009
bo+0.009
Neg logit contributions
azes-0.013
pit-0.011
mmm-0.011
castle-0.011
ires-0.011
Token to
Token ID284
Feature activation+9.39e-03

Pos logit contributions
B+0.011
ch+0.011
sc+0.011
 subs+0.010
ched+0.010
Neg logit contributions
azes-0.013
pit-0.011
ires-0.010
mmm-0.010
ement-0.010
Token stop
Token ID2245
Feature activation-1.41e-02

Pos logit contributions
B+0.011
ch+0.010
 subs+0.010
bo+0.010
sc+0.010
Neg logit contributions
azes-0.021
ires-0.018
ement-0.018
pit-0.018
mmm-0.017
Token mis
Token ID2984
Feature activation+5.07e-02

Pos logit contributions
azes+0.025
eria+0.022
ae+0.022
mmm+0.022
pit+0.022
Neg logit contributions
ch-0.021
B-0.021
 scrap-0.020
 subs-0.020
bo-0.019
Tokenbeh
Token ID20709
Feature activation+8.35e-02

Pos logit contributions
 subs+0.055
fl+0.053
ch+0.053
sc+0.051
 scrap+0.050
Neg logit contributions
azes-0.109
eria-0.094
isdom-0.093
mmm-0.091
idge-0.090
Tokenaving
Token ID2703
Feature activation+1.51e-02

Pos logit contributions
sc+0.119
ch+0.118
B+0.117
 subs+0.111
ched+0.108
Neg logit contributions
azes-0.159
pit-0.136
ires-0.134
mmm-0.133
ement-0.133
Token.
Token ID13
Feature activation+5.06e-03

Pos logit contributions
ch+0.020
sc+0.020
B+0.020
 subs+0.019
bo+0.018
Neg logit contributions
azes-0.030
pit-0.025
ement-0.025
 oven-0.025
isdom-0.025
Token But
Token ID887
Feature activation+3.58e-04

Pos logit contributions
B+0.007
ch+0.007
sc+0.006
 subs+0.006
 scrap+0.006
Neg logit contributions
azes-0.010
castle-0.009
pit-0.009
ement-0.009
ires-0.009
Token he
Token ID339
Feature activation-8.93e-03

Pos logit contributions
ch+0.001
sc+0.001
B+0.001
 subs+0.001
ched+0.000
Neg logit contributions
azes-0.001
pit-0.001
isdom-0.001
mmm-0.001
ires-0.001
Token didn
Token ID1422
Feature activation+3.71e-02

Pos logit contributions
azes+0.021
mmm+0.018
ires+0.018
pit+0.018
 balcony+0.018
Neg logit contributions
B-0.009
ch-0.009
 subs-0.009
ched-0.009
bo-0.009
Token to
Token ID284
Feature activation+2.73e-02

Pos logit contributions
ch+0.018
sc+0.018
B+0.018
 subs+0.017
ched+0.016
Neg logit contributions
azes-0.025
pit-0.021
isdom-0.021
 oven-0.021
mmm-0.021
Token go
Token ID467
Feature activation+2.25e-03

Pos logit contributions
B+0.034
ch+0.034
sc+0.033
 subs+0.032
bo+0.031
Neg logit contributions
azes-0.056
pit-0.049
ement-0.048
ires-0.048
isdom-0.047
Token to
Token ID284
Feature activation+1.17e-02

Pos logit contributions
ch+0.003
sc+0.003
B+0.003
 subs+0.003
bo+0.003
Neg logit contributions
azes-0.005
pit-0.004
mmm-0.004
ires-0.004
isdom-0.004
Token work
Token ID670
Feature activation+4.96e-02

Pos logit contributions
ch+0.013
B+0.013
sc+0.012
 subs+0.012
 scrap+0.012
Neg logit contributions
azes-0.026
pit-0.023
ement-0.023
ires-0.022
mmm-0.022
Token.
Token ID13
Feature activation+8.55e-03

Pos logit contributions
sc+0.059
B+0.056
ch+0.056
 subs+0.052
ched+0.050
Neg logit contributions
azes-0.109
pit-0.094
mmm-0.093
isdom-0.093
ires-0.093
Token Tears
Token ID40328
Feature activation+8.25e-02

Pos logit contributions
B+0.012
ch+0.012
sc+0.012
 subs+0.012
 scrap+0.011
Neg logit contributions
azes-0.015
castle-0.014
mmm-0.014
isdom-0.014
pit-0.013
Token started
Token ID2067
Feature activation+3.66e-02

Pos logit contributions
fl+0.091
B+0.089
sc+0.089
Fl+0.087
ch+0.086
Neg logit contributions
azes-0.158
ement-0.150
 patiently-0.146
 oven-0.144
 writing-0.143
Token flowing
Token ID17609
Feature activation+4.55e-02

Pos logit contributions
sc+0.053
ch+0.051
B+0.051
 subs+0.048
bo+0.047
Neg logit contributions
azes-0.069
pit-0.058
 writing-0.057
 patiently-0.057
 oven-0.057
Token down
Token ID866
Feature activation-1.18e-02

Pos logit contributions
ch+0.056
B+0.055
sc+0.055
 subs+0.053
text+0.051
Neg logit contributions
azes-0.091
pit-0.080
ement-0.079
 fridge-0.078
 patiently-0.078
Token his
Token ID465
Feature activation+2.28e-02

Pos logit contributions
azes+0.024
mmm+0.020
isdom+0.020
pit+0.020
ires+0.020
Neg logit contributions
ch-0.015
B-0.015
sc-0.015
 subs-0.015
ched-0.014
Token cheeks
Token ID25839
Feature activation+2.88e-02

Pos logit contributions
ch+0.040
sc+0.040
B+0.040
 subs+0.039
bo+0.038
Neg logit contributions
azes-0.035
pit-0.029
mmm-0.029
isdom-0.029
ires-0.028
Token became
Token ID2627
Feature activation-8.19e-03

Pos logit contributions
sc+0.010
B+0.009
ch+0.009
 subs+0.009
Free+0.008
Neg logit contributions
azes-0.011
 patiently-0.010
isdom-0.010
 oven-0.010
 Jamie-0.010
Token friendly
Token ID8030
Feature activation-2.67e-02

Pos logit contributions
azes+0.015
pit+0.013
 writing+0.013
 oven+0.013
 patiently+0.013
Neg logit contributions
sc-0.012
ch-0.012
B-0.011
 subs-0.011
bo-0.011
Token and
Token ID290
Feature activation+4.53e-03

Pos logit contributions
azes+0.052
mmm+0.045
isdom+0.044
 oven+0.044
ires+0.043
Neg logit contributions
B-0.037
sc-0.036
ch-0.036
 subs-0.035
fl-0.034
Token well
Token ID880
Feature activation+2.50e-02

Pos logit contributions
sc+0.007
ch+0.007
B+0.007
 subs+0.006
text+0.006
Neg logit contributions
azes-0.008
 writing-0.007
 patiently-0.007
isdom-0.006
mmm-0.006
Token-
Token ID12
Feature activation+2.85e-02

Pos logit contributions
sc+0.034
ch+0.034
B+0.033
 subs+0.032
ched+0.031
Neg logit contributions
azes-0.049
pit-0.042
mmm-0.041
isdom-0.041
 fridge-0.041
Tokenbeh
Token ID20709
Feature activation+8.24e-02

Pos logit contributions
sc+0.035
 subs+0.035
ch+0.034
B+0.034
text+0.034
Neg logit contributions
azes-0.058
isdom-0.050
mmm-0.049
ires-0.048
castle-0.048
Tokenaved
Token ID9586
Feature activation+9.84e-04

Pos logit contributions
ch+0.105
B+0.105
sc+0.104
 subs+0.100
bo+0.093
Neg logit contributions
azes-0.167
pit-0.143
isdom-0.140
mmm-0.139
ement-0.139
Token,
Token ID11
Feature activation-1.72e-02

Pos logit contributions
ch+0.001
sc+0.001
B+0.001
 subs+0.001
bo+0.001
Neg logit contributions
azes-0.002
pit-0.002
ement-0.002
isdom-0.002
ires-0.002
Token just
Token ID655
Feature activation+3.75e-03

Pos logit contributions
azes+0.032
ires+0.027
mmm+0.027
pit+0.026
ement+0.026
Neg logit contributions
ch-0.025
sc-0.025
B-0.025
 subs-0.024
ched-0.023
Token like
Token ID588
Feature activation-1.38e-02

Pos logit contributions
ch+0.006
sc+0.006
B+0.006
 subs+0.006
ched+0.006
Neg logit contributions
azes-0.006
pit-0.005
ement-0.005
mmm-0.005
ires-0.005
Token Buddy
Token ID36896
Feature activation-1.53e-02

Pos logit contributions
azes+0.028
pit+0.024
mmm+0.024
ement+0.024
ires+0.023
Neg logit contributions
ch-0.018
B-0.018
sc-0.017
 subs-0.017
ched-0.016
Token night
Token ID1755
Feature activation+5.26e-02

Pos logit contributions
ch+0.031
B+0.030
sc+0.030
 subs+0.029
bo+0.026
Neg logit contributions
azes-0.067
pit-0.058
mmm-0.057
ires-0.057
isdom-0.057
Token,
Token ID11
Feature activation-6.12e-03

Pos logit contributions
sc+0.098
B+0.097
ch+0.095
 subs+0.094
ched+0.091
Neg logit contributions
azes-0.077
pit-0.061
 oven-0.060
 fridge-0.060
 writing-0.060
Token the
Token ID262
Feature activation+1.59e-02

Pos logit contributions
azes+0.017
pit+0.015
isdom+0.015
ement+0.015
mmm+0.015
Neg logit contributions
ch-0.004
B-0.004
sc-0.004
 subs-0.003
Free-0.003
Token wild
Token ID4295
Feature activation+1.16e-02

Pos logit contributions
ch+0.016
B+0.016
sc+0.014
 subs+0.014
Spot+0.013
Neg logit contributions
azes-0.038
mmm-0.035
pit-0.034
isdom-0.034
ement-0.033
Token cat
Token ID3797
Feature activation+7.97e-03

Pos logit contributions
ch+0.014
B+0.013
sc+0.013
 subs+0.012
fl+0.012
Neg logit contributions
azes-0.025
mmm-0.023
pit-0.022
ement-0.022
castle-0.022
Token lur
Token ID20605
Feature activation+8.09e-02

Pos logit contributions
ch+0.007
B+0.007
sc+0.007
 subs+0.006
bo+0.006
Neg logit contributions
azes-0.020
mmm-0.017
ires-0.017
ement-0.017
pit-0.017
Tokenked
Token ID9091
Feature activation+1.72e-02

Pos logit contributions
 subs+0.108
B+0.103
fl+0.101
Br+0.097
Spot+0.097
Neg logit contributions
azes-0.142
isdom-0.124
 patiently-0.122
ues-0.122
sing-0.122
Token through
Token ID832
Feature activation-2.58e-03

Pos logit contributions
ch+0.022
sc+0.022
B+0.021
 subs+0.020
ched+0.020
Neg logit contributions
azes-0.036
mmm-0.031
pit-0.031
ement-0.030
isdom-0.030
Token the
Token ID262
Feature activation-1.76e-03

Pos logit contributions
azes+0.007
pit+0.006
mmm+0.006
castle+0.006
ement+0.006
Neg logit contributions
B-0.002
ch-0.002
sc-0.002
 subs-0.002
 scrap-0.002
Token woods
Token ID16479
Feature activation+4.92e-03

Pos logit contributions
azes+0.003
pit+0.002
mmm+0.002
isdom+0.002
ement+0.002
Neg logit contributions
ch-0.003
B-0.003
sc-0.003
 subs-0.003
bo-0.003
Token,
Token ID11
Feature activation-8.53e-03

Pos logit contributions
sc+0.005
ch+0.005
B+0.005
ny+0.005
fl+0.005
Neg logit contributions
azes-0.011
mmm-0.011
mmmm-0.010
isdom-0.010
ocolate-0.010
Token needle
Token ID17598
Feature activation+4.85e-02

Pos logit contributions
sc+0.035
ch+0.035
B+0.035
 subs+0.033
ched+0.032
Neg logit contributions
azes-0.041
pit-0.035
 oven-0.034
 fridge-0.034
isdom-0.034
Token or
Token ID393
Feature activation+2.41e-02

Pos logit contributions
B+0.071
ch+0.070
sc+0.070
 subs+0.068
text+0.064
Neg logit contributions
azes-0.087
pit-0.076
 oven-0.073
 fridge-0.073
castle-0.073
Token a
Token ID257
Feature activation+1.99e-02

Pos logit contributions
sc+0.038
B+0.038
ch+0.037
 subs+0.037
ched+0.035
Neg logit contributions
azes-0.040
 fridge-0.035
 oven-0.035
 writing-0.034
 pasta-0.034
Token st
Token ID336
Feature activation+6.98e-02

Pos logit contributions
ch+0.027
B+0.027
sc+0.027
 subs+0.026
 scrap+0.025
Neg logit contributions
azes-0.040
ires-0.034
mmm-0.034
isdom-0.033
ement-0.033
Tokeneth
Token ID2788
Feature activation+4.21e-02

Pos logit contributions
B+0.088
sc+0.086
ch+0.081
fl+0.080
 subs+0.079
Neg logit contributions
azes-0.150
 fridge-0.126
pit-0.126
 oven-0.125
ires-0.125
Tokenoscope
Token ID40326
Feature activation+8.06e-02

Pos logit contributions
B+0.071
 subs+0.068
sc+0.067
ch+0.066
W+0.063
Neg logit contributions
azes-0.065
 oven-0.056
 fridge-0.056
 writing-0.056
pit-0.056
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
B+0.122
sc+0.119
ch+0.117
 subs+0.115
Br+0.110
Neg logit contributions
azes-0.140
 oven-0.120
pit-0.119
 fridge-0.118
 writing-0.117
Token They
Token ID1119
Feature activation+2.08e-03

Pos logit contributions
ch+0.017
sc+0.017
B+0.017
 subs+0.016
ched+0.015
Neg logit contributions
azes-0.023
pit-0.019
ires-0.019
mmm-0.019
isdom-0.019
Token did
Token ID750
Feature activation+2.39e-02

Pos logit contributions
ch+0.003
sc+0.003
B+0.003
 subs+0.003
fl+0.002
Neg logit contributions
azes-0.004
pit-0.004
 fridge-0.004
isdom-0.004
 oven-0.004
Token not
Token ID407
Feature activation+2.48e-02

Pos logit contributions
sc+0.049
ch+0.049
B+0.049
 subs+0.048
bo+0.045
Neg logit contributions
azes-0.031
isdom-0.023
ires-0.023
mmm-0.023
pit-0.023
Token want
Token ID765
Feature activation-4.43e-04

Pos logit contributions
sc+0.040
ch+0.040
B+0.040
 subs+0.039
ched+0.037
Neg logit contributions
azes-0.042
pit-0.034
isdom-0.034
mmm-0.034
 oven-0.033
Token very
Token ID845
Feature activation-8.86e-03

Pos logit contributions
azes+0.028
pit+0.024
mmm+0.024
ement+0.024
isdom+0.024
Neg logit contributions
ch-0.021
B-0.020
sc-0.020
 subs-0.020
bo-0.019
Token naughty
Token ID45128
Feature activation+5.03e-03

Pos logit contributions
azes+0.018
pit+0.015
ement+0.015
isdom+0.015
mmm+0.015
Neg logit contributions
ch-0.012
B-0.012
sc-0.011
 subs-0.011
bo-0.010
Token and
Token ID290
Feature activation+1.66e-02

Pos logit contributions
ch+0.006
B+0.006
sc+0.005
fl+0.005
 subs+0.005
Neg logit contributions
azes-0.011
isdom-0.010
mmm-0.010
ocolate-0.009
castle-0.009
Token always
Token ID1464
Feature activation+1.57e-02

Pos logit contributions
sc+0.024
B+0.024
ch+0.024
 subs+0.023
text+0.022
Neg logit contributions
azes-0.031
 oven-0.026
 fridge-0.026
pit-0.026
isdom-0.026
Token mis
Token ID2984
Feature activation+5.04e-02

Pos logit contributions
ch+0.018
sc+0.018
B+0.018
 subs+0.017
ched+0.016
Neg logit contributions
azes-0.034
pit-0.029
mmm-0.029
isdom-0.029
ires-0.029
Tokenbeh
Token ID20709
Feature activation+8.04e-02

Pos logit contributions
ch+0.049
sc+0.049
 subs+0.048
B+0.045
text+0.044
Neg logit contributions
azes-0.117
isdom-0.099
pit-0.098
eria-0.098
mmm-0.097
Tokenaving
Token ID2703
Feature activation+2.17e-02

Pos logit contributions
sc+0.103
ch+0.102
B+0.101
 subs+0.095
ched+0.093
Neg logit contributions
azes-0.165
pit-0.143
ires-0.141
ement-0.141
mmm-0.141
Token.
Token ID13
Feature activation+2.15e-03

Pos logit contributions
ch+0.030
sc+0.030
B+0.030
 subs+0.029
bo+0.027
Neg logit contributions
azes-0.042
pit-0.035
isdom-0.034
 oven-0.034
ires-0.034
Token One
Token ID1881
Feature activation-1.30e-02

Pos logit contributions
B+0.002
Fl+0.002
ch+0.002
|+0.002
W+0.002
Neg logit contributions
castle-0.005
azes-0.005
 patiently-0.005
ues-0.005
pit-0.005
Token day
Token ID1110
Feature activation+1.81e-02

Pos logit contributions
azes+0.026
 fridge+0.022
 oven+0.022
pit+0.022
 refrigerator+0.022
Neg logit contributions
sc-0.017
 subs-0.016
B-0.016
ch-0.016
ched-0.016
Token she
Token ID673
Feature activation-1.45e-03

Pos logit contributions
sc+0.033
ch+0.033
B+0.032
 subs+0.032
bo+0.031
Neg logit contributions
azes-0.027
pit-0.022
mmm-0.021
 oven-0.021
 writing-0.021
Token listen
Token ID6004
Feature activation+2.24e-03

Pos logit contributions
sc+0.022
B+0.022
ch+0.021
 subs+0.020
text+0.020
Neg logit contributions
azes-0.026
pit-0.021
 writing-0.021
isdom-0.021
 fridge-0.021
Token.
Token ID13
Feature activation-4.41e-03

Pos logit contributions
ch+0.002
sc+0.002
B+0.002
 subs+0.002
ched+0.002
Neg logit contributions
azes-0.005
pit-0.005
mmm-0.005
ement-0.005
isdom-0.005
Token He
Token ID679
Feature activation-6.31e-03

Pos logit contributions
azes+0.009
 studio+0.008
castle+0.008
 writing+0.008
 canvas+0.008
Neg logit contributions
ch-0.005
B-0.005
ny-0.005
sc-0.005
Spot-0.005
Token always
Token ID1464
Feature activation+1.64e-02

Pos logit contributions
azes+0.013
pit+0.011
mmm+0.011
ires+0.011
isdom+0.011
Neg logit contributions
ch-0.008
B-0.008
sc-0.008
 subs-0.008
bo-0.007
Token mis
Token ID2984
Feature activation+4.73e-02

Pos logit contributions
ch+0.018
B+0.018
sc+0.018
 subs+0.017
bo+0.016
Neg logit contributions
azes-0.037
ement-0.032
pit-0.032
ires-0.032
mmm-0.032
Tokenbeh
Token ID20709
Feature activation+7.98e-02

Pos logit contributions
sc+0.041
 subs+0.041
ch+0.040
B+0.037
text+0.036
Neg logit contributions
azes-0.114
isdom-0.099
pit-0.096
mmm-0.096
eria-0.096
Tokenaved
Token ID9586
Feature activation-6.41e-03

Pos logit contributions
sc+0.109
ch+0.109
B+0.107
 subs+0.102
ched+0.099
Neg logit contributions
azes-0.157
pit-0.134
ires-0.132
ement-0.132
mmm-0.131
Token and
Token ID290
Feature activation+1.81e-02

Pos logit contributions
azes+0.013
mmm+0.011
pit+0.011
isdom+0.011
ement+0.011
Neg logit contributions
ch-0.008
sc-0.008
B-0.008
 subs-0.008
ched-0.008
Token caused
Token ID4073
Feature activation+9.85e-04

Pos logit contributions
ch+0.024
B+0.024
sc+0.024
 subs+0.023
ched+0.022
Neg logit contributions
azes-0.036
pit-0.031
ement-0.031
mmm-0.030
ires-0.030
Token trouble
Token ID5876
Feature activation+2.64e-02

Pos logit contributions
sc+0.001
ch+0.001
B+0.001
 subs+0.001
ched+0.001
Neg logit contributions
azes-0.002
pit-0.002
isdom-0.002
ires-0.002
mmm-0.002
Token wherever
Token ID14530
Feature activation+2.14e-02

Pos logit contributions
ch+0.039
sc+0.038
B+0.038
 subs+0.037
bo+0.035
Neg logit contributions
azes-0.048
pit-0.040
 writing-0.040
isdom-0.039
 oven-0.039
Token knee
Token ID10329
Feature activation+2.33e-02

Pos logit contributions
sc+0.035
ch+0.034
B+0.034
 subs+0.032
ched+0.032
Neg logit contributions
azes-0.019
pit-0.014
 fridge-0.014
 writing-0.014
ires-0.014
Token and
Token ID290
Feature activation+3.49e-02

Pos logit contributions
B+0.045
sc+0.045
Ar+0.045
text+0.044
Free+0.044
Neg logit contributions
azes-0.027
 fridge-0.024
 patiently-0.023
 writing-0.023
 oven-0.022
Token it
Token ID340
Feature activation+2.85e-02

Pos logit contributions
sc+0.051
ch+0.045
B+0.045
 subs+0.043
Ar+0.043
Neg logit contributions
azes-0.066
 fridge-0.060
 writing-0.060
 pasta-0.057
 desk-0.057
Token started
Token ID2067
Feature activation+3.97e-02

Pos logit contributions
Ar+0.040
text+0.038
|+0.038
sc+0.038
fl+0.034
Neg logit contributions
 fridge-0.042
 patiently-0.041
 writing-0.038
azes-0.038
 desk-0.037
Token to
Token ID284
Feature activation+2.27e-02

Pos logit contributions
sc+0.071
text+0.070
Ar+0.065
Fl+0.064
B+0.063
Neg logit contributions
azes-0.057
 writing-0.054
 counting-0.050
 reading-0.049
 patiently-0.049
Token bleed
Token ID30182
Feature activation+7.95e-02

Pos logit contributions
sc+0.030
text+0.028
B+0.026
 subs+0.026
fl+0.026
Neg logit contributions
azes-0.041
 fridge-0.040
 oven-0.038
 balcony-0.038
 writing-0.037
Token.
Token ID13
Feature activation+1.80e-02

Pos logit contributions
B+0.129
sc+0.128
ch+0.125
 subs+0.125
Br+0.123
Neg logit contributions
azes-0.117
 patiently-0.104
 fridge-0.102
 writing-0.102
 oven-0.099
Token
Token ID220
Feature activation+2.20e-02

Pos logit contributions
sc+0.025
ch+0.025
B+0.024
 subs+0.024
ched+0.022
Neg logit contributions
azes-0.035
pit-0.030
mmm-0.030
 fridge-0.029
isdom-0.029
Token
Token ID198
Feature activation+2.76e-02

Pos logit contributions
ch+0.027
sc+0.027
B+0.026
 subs+0.026
bo+0.024
Neg logit contributions
azes-0.046
pit-0.039
ires-0.039
mmm-0.038
isdom-0.038
Token
Token ID198
Feature activation+2.22e-02

Pos logit contributions
 subs+0.036
sc+0.036
ch+0.036
B+0.035
fl+0.033
Neg logit contributions
azes-0.054
mmm-0.045
ires-0.045
ement-0.044
pit-0.044
TokenTim
Token ID14967
Feature activation+1.43e-02

Pos logit contributions
sc+0.035
 subs+0.034
fl+0.032
text+0.032
ch+0.031
Neg logit contributions
azes-0.038
mmm-0.033
isdom-0.032
 fridge-0.032
ocolate-0.031
Token saw
Token ID2497
Feature activation-3.91e-02

Pos logit contributions
azes+0.005
pit+0.004
 fridge+0.004
castle+0.004
 writing+0.004
Neg logit contributions
sc-0.003
B-0.003
ch-0.003
 subs-0.002
Br-0.002
Token some
Token ID617
Feature activation-9.84e-03

Pos logit contributions
azes+0.088
mmm+0.078
pit+0.075
isdom+0.075
ement+0.074
Neg logit contributions
B-0.046
ch-0.046
 scrap-0.041
ched-0.041
sc-0.041
Token other
Token ID584
Feature activation-1.61e-02

Pos logit contributions
azes+0.020
mmm+0.018
isdom+0.017
pit+0.017
ocolate+0.017
Neg logit contributions
ch-0.013
B-0.012
sc-0.012
 subs-0.012
 scrap-0.011
Token kids
Token ID3988
Feature activation-1.13e-02

Pos logit contributions
azes+0.030
mmm+0.026
isdom+0.026
ement+0.025
ocolate+0.025
Neg logit contributions
sc-0.023
ch-0.023
B-0.023
 subs-0.022
text-0.021
Token mis
Token ID2984
Feature activation+4.27e-02

Pos logit contributions
azes+0.018
mmm+0.017
isdom+0.016
ires+0.016
ae+0.016
Neg logit contributions
sc-0.018
ch-0.018
B-0.018
 subs-0.018
ched-0.017
Tokenbeh
Token ID20709
Feature activation+7.88e-02

Pos logit contributions
ch+0.052
sc+0.050
 subs+0.050
B+0.049
fl+0.047
Neg logit contributions
azes-0.089
pit-0.076
isdom-0.074
mmm-0.073
eria-0.073
Tokenaving
Token ID2703
Feature activation+1.21e-02

Pos logit contributions
ch+0.102
sc+0.102
B+0.102
 subs+0.097
bo+0.091
Neg logit contributions
azes-0.159
pit-0.135
isdom-0.133
mmm-0.132
ires-0.132
Token and
Token ID290
Feature activation+1.11e-02

Pos logit contributions
ch+0.017
B+0.017
sc+0.017
 subs+0.016
bo+0.015
Neg logit contributions
azes-0.023
pit-0.020
isdom-0.019
ement-0.019
 oven-0.019
Token running
Token ID2491
Feature activation+1.89e-02

Pos logit contributions
ch+0.014
sc+0.014
B+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.023
mmm-0.019
ement-0.019
pit-0.019
ires-0.019
Token around
Token ID1088
Feature activation-2.20e-02

Pos logit contributions
 subs+0.025
ch+0.025
B+0.024
text+0.024
Br+0.024
Neg logit contributions
azes-0.033
pit-0.030
 patiently-0.029
ement-0.028
castle-0.028
Token.
Token ID13
Feature activation-2.61e-03

Pos logit contributions
azes+0.050
mmm+0.043
isdom+0.043
pit+0.042
ires+0.042
Neg logit contributions
sc-0.025
B-0.025
ch-0.024
 subs-0.022
ched-0.021
Token play
Token ID711
Feature activation+7.85e-03

Pos logit contributions
ch+0.009
B+0.009
 subs+0.009
 scrap+0.009
sc+0.008
Neg logit contributions
azes-0.022
ires-0.019
mmm-0.019
pit-0.019
ement-0.019
Token.
Token ID13
Feature activation+7.15e-04

Pos logit contributions
B+0.009
sc+0.008
ch+0.008
ched+0.007
 subs+0.007
Neg logit contributions
azes-0.018
isdom-0.016
mmm-0.016
ruits-0.016
ires-0.016
Token
Token ID220
Feature activation+8.77e-03

Pos logit contributions
ch+0.001
sc+0.001
B+0.001
 subs+0.001
bo+0.001
Neg logit contributions
azes-0.001
pit-0.001
isdom-0.001
ires-0.001
mmm-0.001
Token
Token ID198
Feature activation-5.77e-04

Pos logit contributions
ch+0.009
B+0.009
sc+0.009
 subs+0.009
ched+0.008
Neg logit contributions
azes-0.019
pit-0.017
isdom-0.017
 oven-0.017
ement-0.017
Token
Token ID198
Feature activation-7.01e-03

Pos logit contributions
azes+0.001
mmm+0.001
ocolate+0.001
ires+0.001
 writing+0.001
Neg logit contributions
sc-0.001
 subs-0.001
ch-0.001
B-0.001
fl-0.001
TokenTHE
Token ID10970
Feature activation+7.84e-02

Pos logit contributions
azes+0.012
mmm+0.011
ocolate+0.011
eria+0.011
ae+0.011
Neg logit contributions
fl-0.010
sc-0.010
 subs-0.009
ch-0.009
bo-0.009
Token<|endoftext|>
Token ID50256
Feature activation+1.14e-02

Pos logit contributions
ch+0.095
sc+0.095
B+0.094
 subs+0.091
bo+0.085
Neg logit contributions
azes-0.165
pit-0.143
ires-0.141
isdom-0.140
ement-0.140
TokenOnce
Token ID7454
Feature activation+1.79e-02

Pos logit contributions
 subs+0.018
sc+0.018
beh+0.017
burn+0.017
ch+0.017
Neg logit contributions
azes-0.018
 fridge-0.015
ement-0.015
 refrigerator-0.015
pit-0.015
Token upon
Token ID2402
Feature activation+3.38e-02

Pos logit contributions
sc+0.027
ch+0.027
B+0.027
 subs+0.026
ched+0.025
Neg logit contributions
azes-0.032
pit-0.027
ires-0.026
isdom-0.026
ement-0.026
Token a
Token ID257
Feature activation+3.56e-02

Pos logit contributions
sc+0.054
 subs+0.053
ch+0.053
B+0.051
fl+0.049
Neg logit contributions
azes-0.058
 writing-0.047
 fridge-0.046
mmm-0.046
pit-0.046
Token time
Token ID640
Feature activation-4.22e-03

Pos logit contributions
sc+0.035
ch+0.034
B+0.034
 subs+0.033
ched+0.030
Neg logit contributions
azes-0.082
pit-0.072
mmm-0.071
isdom-0.071
 oven-0.071
Token went
Token ID1816
Feature activation-2.48e-03

Pos logit contributions
sc+0.007
ch+0.007
B+0.007
 subs+0.006
fl+0.006
Neg logit contributions
azes-0.013
 oven-0.011
pit-0.011
 fridge-0.011
 patiently-0.011
Token to
Token ID284
Feature activation+4.49e-03

Pos logit contributions
azes+0.005
pit+0.004
 oven+0.004
 fridge+0.004
 patiently+0.004
Neg logit contributions
B-0.003
ch-0.003
sc-0.003
 subs-0.003
text-0.003
Token his
Token ID465
Feature activation-3.94e-03

Pos logit contributions
ch+0.004
B+0.004
 scrap+0.004
sc+0.004
 subs+0.004
Neg logit contributions
azes-0.011
ement-0.010
pit-0.010
mmm-0.010
ires-0.010
Token room
Token ID2119
Feature activation+1.25e-02

Pos logit contributions
azes+0.007
mmm+0.006
otion+0.006
isdom+0.006
ocolate+0.006
Neg logit contributions
B-0.006
sc-0.006
ch-0.006
 subs-0.006
 scrap-0.006
Token and
Token ID290
Feature activation+5.41e-03

Pos logit contributions
sc+0.012
 subs+0.010
B+0.010
 went+0.010
ch+0.010
Neg logit contributions
azes-0.030
mmm-0.028
isdom-0.026
ires-0.026
mmmm-0.026
Token sul
Token ID33154
Feature activation+7.83e-02

Pos logit contributions
ch+0.006
sc+0.005
B+0.005
 subs+0.005
 scrap+0.005
Neg logit contributions
azes-0.013
ement-0.011
mmm-0.011
ires-0.011
pit-0.011
Tokenked
Token ID9091
Feature activation+1.97e-02

Pos logit contributions
B+0.079
ch+0.079
sc+0.078
 subs+0.074
text+0.068
Neg logit contributions
azes-0.181
pit-0.158
ires-0.154
isdom-0.154
 fridge-0.154
Token.
Token ID13
Feature activation+8.05e-03

Pos logit contributions
ch+0.024
sc+0.024
B+0.024
 subs+0.023
ched+0.022
Neg logit contributions
azes-0.041
pit-0.035
mmm-0.035
isdom-0.035
ires-0.034
Token
Token ID198
Feature activation+2.17e-02

Pos logit contributions
B+0.006
ch+0.006
sc+0.006
Br+0.005
 subs+0.005
Neg logit contributions
azes-0.020
castle-0.019
pit-0.018
isdom-0.018
ement-0.018
Token
Token ID198
Feature activation+1.92e-02

Pos logit contributions
ch+0.025
sc+0.025
B+0.024
 subs+0.023
ched+0.022
Neg logit contributions
azes-0.047
pit-0.041
isdom-0.040
mmm-0.040
ires-0.040
TokenThen
Token ID6423
Feature activation+3.00e-02

Pos logit contributions
ch+0.024
sc+0.024
B+0.024
 subs+0.023
ched+0.022
Neg logit contributions
azes-0.040
pit-0.034
isdom-0.033
mmm-0.033
ires-0.033
Token for
Token ID329
Feature activation-1.83e-02

Pos logit contributions
azes+0.035
ires+0.029
isdom+0.028
 Majesty+0.028
pit+0.028
Neg logit contributions
sc-0.021
ch-0.021
B-0.020
 subs-0.018
ched-0.018
Token being
Token ID852
Feature activation+2.25e-03

Pos logit contributions
azes+0.038
pit+0.035
ires+0.033
 oven+0.032
 fridge+0.032
Neg logit contributions
B-0.023
ch-0.023
sc-0.021
 subs-0.021
 scrap-0.020
Token so
Token ID523
Feature activation+3.42e-03

Pos logit contributions
ch+0.004
sc+0.004
B+0.004
 subs+0.003
ched+0.003
Neg logit contributions
azes-0.004
pit-0.003
isdom-0.003
mmm-0.003
ires-0.003
Token well
Token ID880
Feature activation+3.61e-02

Pos logit contributions
sc+0.005
ch+0.005
B+0.005
 subs+0.005
ched+0.005
Neg logit contributions
azes-0.006
pit-0.005
ires-0.005
isdom-0.005
mmm-0.005
Token-
Token ID12
Feature activation+3.68e-02

Pos logit contributions
sc+0.049
ch+0.048
B+0.048
 subs+0.046
ched+0.044
Neg logit contributions
azes-0.072
pit-0.060
isdom-0.059
mmm-0.059
ires-0.059
Tokenbeh
Token ID20709
Feature activation+7.77e-02

Pos logit contributions
text+0.047
sc+0.047
 subs+0.046
ch+0.044
B+0.043
Neg logit contributions
azes-0.071
isdom-0.063
mmm-0.061
 patiently-0.060
ae-0.060
Tokenaved
Token ID9586
Feature activation+5.36e-03

Pos logit contributions
ch+0.099
sc+0.099
B+0.099
 subs+0.094
bo+0.088
Neg logit contributions
azes-0.158
pit-0.135
isdom-0.133
mmm-0.132
ires-0.132
Token and
Token ID290
Feature activation+1.04e-02

Pos logit contributions
ch+0.008
B+0.008
sc+0.008
 subs+0.007
bo+0.007
Neg logit contributions
azes-0.010
pit-0.009
ement-0.008
 oven-0.008
isdom-0.008
Token Jack
Token ID3619
Feature activation+4.12e-03

Pos logit contributions
ch+0.017
sc+0.017
B+0.017
 subs+0.016
ched+0.015
Neg logit contributions
azes-0.018
pit-0.014
ires-0.014
isdom-0.014
mmm-0.014
Token was
Token ID373
Feature activation-2.06e-02

Pos logit contributions
ch+0.004
sc+0.004
B+0.004
 subs+0.004
bo+0.003
Neg logit contributions
azes-0.010
pit-0.009
ires-0.008
mmm-0.008
isdom-0.008
Token delighted
Token ID20707
Feature activation-3.23e-02

Pos logit contributions
azes+0.035
pit+0.029
isdom+0.028
ires+0.028
mmm+0.028
Neg logit contributions
ch-0.034
B-0.033
sc-0.033
 subs-0.032
bo-0.031
Token girl
Token ID2576
Feature activation+1.41e-02

Pos logit contributions
ch+0.026
sc+0.025
B+0.024
text+0.024
ched+0.024
Neg logit contributions
azes-0.029
pit-0.028
 fridge-0.028
 oven-0.027
 refrigerator-0.027
Token didn
Token ID1422
Feature activation+3.93e-02

Pos logit contributions
B+0.014
ch+0.014
 subs+0.014
sc+0.014
bo+0.014
Neg logit contributions
azes-0.033
pit-0.028
ires-0.028
mmm-0.028
ocolate-0.027
Token't
Token ID470
Feature activation+1.57e-02

Pos logit contributions
sc+0.066
ch+0.064
B+0.064
 subs+0.063
fl+0.062
Neg logit contributions
azes-0.063
mmm-0.054
ocolate-0.052
ement-0.051
ires-0.051
Token understand
Token ID1833
Feature activation-7.24e-03

Pos logit contributions
sc+0.028
B+0.027
ch+0.027
text+0.026
 subs+0.026
Neg logit contributions
azes-0.024
isdom-0.020
 writing-0.019
mmm-0.019
 patiently-0.019
Token.
Token ID13
Feature activation+2.96e-03

Pos logit contributions
azes+0.015
pit+0.013
castle+0.013
 balcony+0.013
sing+0.013
Neg logit contributions
ch-0.009
B-0.009
sc-0.008
ched-0.008
 scrap-0.008
Token Tears
Token ID40328
Feature activation+7.77e-02

Pos logit contributions
ny+0.003
bo+0.003
s+0.003
 subs+0.003
B+0.003
Neg logit contributions
azes-0.006
castle-0.006
ocket-0.005
mmm-0.005
otion-0.005
Token started
Token ID2067
Feature activation+3.07e-02

Pos logit contributions
fl+0.096
sc+0.089
Fl+0.089
ch+0.088
Rex+0.087
Neg logit contributions
ement-0.138
azes-0.137
 patiently-0.130
 writing-0.126
 oven-0.125
Token rolling
Token ID10708
Feature activation+1.81e-02

Pos logit contributions
sc+0.046
ch+0.043
B+0.042
text+0.041
bo+0.040
Neg logit contributions
azes-0.057
 writing-0.049
 patiently-0.049
 fridge-0.047
pit-0.047
Token down
Token ID866
Feature activation-1.26e-02

Pos logit contributions
ch+0.023
text+0.022
sc+0.022
 subs+0.022
B+0.022
Neg logit contributions
azes-0.032
ement-0.030
pit-0.030
 patiently-0.029
 refrigerator-0.029
Token her
Token ID607
Feature activation+1.80e-02

Pos logit contributions
azes+0.026
pit+0.022
ement+0.021
isdom+0.021
 oven+0.021
Neg logit contributions
ch-0.016
sc-0.016
B-0.016
 subs-0.016
bo-0.015
Token face
Token ID1986
Feature activation+1.28e-02

Pos logit contributions
ch+0.034
sc+0.034
B+0.033
 subs+0.032
ched+0.031
Neg logit contributions
azes-0.027
pit-0.021
ement-0.021
 fridge-0.021
isdom-0.020
Token told
Token ID1297
Feature activation+4.07e-03

Pos logit contributions
azes+0.027
pit+0.023
ires+0.023
mmm+0.023
isdom+0.023
Neg logit contributions
ch-0.012
B-0.012
sc-0.012
 subs-0.011
ched-0.010
Token him
Token ID683
Feature activation-2.18e-02

Pos logit contributions
ch+0.007
B+0.007
sc+0.007
 subs+0.007
bo+0.006
Neg logit contributions
azes-0.007
pit-0.005
ires-0.005
castle-0.005
ement-0.005
Token to
Token ID284
Feature activation+2.42e-03

Pos logit contributions
azes+0.043
pit+0.039
ement+0.037
ires+0.037
mmm+0.037
Neg logit contributions
ch-0.028
B-0.028
sc-0.027
 subs-0.027
ched-0.026
Token stop
Token ID2245
Feature activation-1.96e-02

Pos logit contributions
B+0.003
ch+0.003
sc+0.003
 subs+0.003
bo+0.003
Neg logit contributions
azes-0.005
ires-0.005
ement-0.005
pit-0.005
 oven-0.004
Token mis
Token ID2984
Feature activation+4.80e-02

Pos logit contributions
azes+0.035
eria+0.032
pit+0.031
isdom+0.031
ae+0.031
Neg logit contributions
ch-0.029
B-0.028
 subs-0.027
sc-0.026
 scrap-0.026
Tokenbeh
Token ID20709
Feature activation+7.73e-02

Pos logit contributions
 subs+0.054
fl+0.052
ch+0.050
 scrap+0.049
text+0.048
Neg logit contributions
azes-0.101
eria-0.085
mmm-0.085
isdom-0.084
pit-0.083
Tokenaving
Token ID2703
Feature activation+7.72e-03

Pos logit contributions
sc+0.104
ch+0.104
B+0.103
 subs+0.099
ched+0.094
Neg logit contributions
azes-0.153
pit-0.130
mmm-0.127
ires-0.127
isdom-0.127
Token and
Token ID290
Feature activation+8.50e-03

Pos logit contributions
ch+0.012
sc+0.012
B+0.012
 subs+0.011
ched+0.011
Neg logit contributions
azes-0.014
pit-0.011
isdom-0.011
mmm-0.011
ires-0.011
Token to
Token ID284
Feature activation+1.84e-02

Pos logit contributions
ch+0.012
sc+0.012
B+0.012
 subs+0.012
ched+0.011
Neg logit contributions
azes-0.016
pit-0.014
mmm-0.013
isdom-0.013
ement-0.013
Token be
Token ID307
Feature activation-1.52e-02

Pos logit contributions
ch+0.026
sc+0.026
B+0.026
 subs+0.025
ched+0.023
Neg logit contributions
azes-0.035
pit-0.030
isdom-0.029
mmm-0.029
ires-0.029
Token kind
Token ID1611
Feature activation+1.70e-02

Pos logit contributions
azes+0.023
pit+0.019
 patiently+0.019
 writing+0.018
castle+0.018
Neg logit contributions
sc-0.027
ch-0.026
B-0.026
 subs-0.026
text-0.025
Token.
Token ID13
Feature activation-4.08e-03

Pos logit contributions
azes+0.033
ires+0.027
mmm+0.027
pit+0.027
isdom+0.026
Neg logit contributions
ch-0.018
sc-0.018
B-0.017
 subs-0.017
fl-0.017
Token
Token ID220
Feature activation-6.13e-04

Pos logit contributions
azes+0.009
pit+0.008
ement+0.008
isdom+0.008
castle+0.008
Neg logit contributions
B-0.004
ch-0.004
sc-0.004
 subs-0.004
bo-0.004
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
azes+0.001
ement+0.001
pit+0.001
 oven+0.001
 fridge+0.001
Neg logit contributions
ch-0.001
B-0.001
sc-0.001
 subs-0.001
bo-0.001
Token
Token ID220
Feature activation-5.39e-03

Pos logit contributions
azes+0.022
pit+0.019
ires+0.018
mmm+0.018
isdom+0.018
Neg logit contributions
sc-0.013
ch-0.013
 subs-0.013
B-0.013
bo-0.011
Token
Token ID198
Feature activation-1.54e-02

Pos logit contributions
azes+0.012
ement+0.010
pit+0.010
isdom+0.010
 oven+0.010
Neg logit contributions
ch-0.006
B-0.006
sc-0.006
 subs-0.006
bo-0.005
TokenTHE
Token ID10970
Feature activation+7.71e-02

Pos logit contributions
azes+0.034
mmm+0.030
pit+0.030
ires+0.030
ocolate+0.029
Neg logit contributions
sc-0.016
 subs-0.015
ch-0.015
B-0.015
 scrap-0.014
Token<|endoftext|>
Token ID50256
Feature activation+8.32e-03

Pos logit contributions
B+0.080
ch+0.079
sc+0.078
 subs+0.077
fl+0.074
Neg logit contributions
azes-0.174
pit-0.155
ires-0.155
ement-0.153
isdom-0.151
TokenOnce
Token ID7454
Feature activation+1.72e-02

Pos logit contributions
 subs+0.012
sc+0.012
ch+0.011
 scrap+0.011
burn+0.011
Neg logit contributions
azes-0.015
pit-0.013
ement-0.012
 fridge-0.012
 refrigerator-0.012
Token upon
Token ID2402
Feature activation+3.29e-02

Pos logit contributions
ch+0.026
sc+0.025
B+0.025
 subs+0.024
ched+0.023
Neg logit contributions
azes-0.031
pit-0.026
isdom-0.026
mmm-0.026
ires-0.025
Token a
Token ID257
Feature activation+3.70e-02

Pos logit contributions
sc+0.053
 subs+0.053
ch+0.052
B+0.050
fl+0.048
Neg logit contributions
azes-0.055
 writing-0.045
mmm-0.044
 fridge-0.044
pit-0.044
Token time
Token ID640
Feature activation-3.50e-03

Pos logit contributions
sc+0.036
ch+0.035
B+0.035
 subs+0.034
ched+0.031
Neg logit contributions
azes-0.086
pit-0.075
mmm-0.074
isdom-0.074
 oven-0.074
Token,
Token ID11
Feature activation-1.58e-02

Pos logit contributions
azes+0.043
isdom+0.037
ires+0.036
eria+0.035
 fridge+0.035
Neg logit contributions
ch-0.034
sc-0.033
 subs-0.032
B-0.031
bo-0.030
Token fourteen
Token ID29167
Feature activation-6.24e-02

Pos logit contributions
azes+0.025
isdom+0.020
pit+0.019
ement+0.019
ires+0.019
Neg logit contributions
ch-0.028
sc-0.027
B-0.027
 subs-0.026
bo-0.026
Token,
Token ID11
Feature activation-2.26e-02

Pos logit contributions
azes+0.118
isdom+0.101
eria+0.100
ires+0.099
sing+0.098
Neg logit contributions
ch-0.086
 subs-0.084
sc-0.082
B-0.082
fl-0.080
Token fifteen
Token ID17280
Feature activation-7.23e-02

Pos logit contributions
azes+0.034
isdom+0.027
ement+0.026
pit+0.026
ires+0.026
Neg logit contributions
ch-0.043
sc-0.041
B-0.041
 subs-0.041
bo-0.040
Token,
Token ID11
Feature activation-3.09e-02

Pos logit contributions
azes+0.142
isdom+0.123
eria+0.121
sing+0.121
 workshop+0.119
Neg logit contributions
ch-0.096
sc-0.091
l-0.088
 subs-0.088
B-0.087
Token sixteen
Token ID27913
Feature activation-9.45e-02

Pos logit contributions
azes+0.046
isdom+0.036
pit+0.035
ement+0.035
ires+0.035
Neg logit contributions
ch-0.059
sc-0.057
B-0.055
 subs-0.055
bo-0.054
Token,
Token ID11
Feature activation-3.62e-02

Pos logit contributions
azes+0.183
isdom+0.159
eria+0.158
sing+0.157
 workshop+0.154
Neg logit contributions
ch-0.125
sc-0.122
B-0.119
fl-0.119
 subs-0.118
Token seventeen
Token ID38741
Feature activation-6.57e-02

Pos logit contributions
azes+0.058
isdom+0.046
pit+0.045
ires+0.044
ement+0.044
Neg logit contributions
ch-0.066
sc-0.064
B-0.062
 subs-0.061
bo-0.060
Token,
Token ID11
Feature activation-2.90e-02

Pos logit contributions
azes+0.126
eria+0.113
sing+0.111
 balcony+0.111
isdom+0.111
Neg logit contributions
ch-0.084
fl-0.079
l-0.078
 subs-0.077
bo-0.077
Token eighteen
Token ID29095
Feature activation-7.73e-02

Pos logit contributions
azes+0.045
isdom+0.035
pit+0.034
ement+0.034
ires+0.034
Neg logit contributions
ch-0.054
sc-0.052
B-0.050
 subs-0.050
bo-0.050
Token,
Token ID11
Feature activation-2.65e-02

Pos logit contributions
azes+0.158
isdom+0.144
eria+0.144
 workshop+0.143
sing+0.141
Neg logit contributions
l-0.085
ch-0.085
s-0.082
fl-0.081
 subs-0.080
Token,
Token ID11
Feature activation-3.02e-02

Pos logit contributions
azes+0.101
isdom+0.086
pit+0.084
 fridge+0.084
mmm+0.084
Neg logit contributions
ch-0.087
B-0.085
sc-0.085
 subs-0.082
bo-0.077
Token seven
Token ID3598
Feature activation-7.51e-02

Pos logit contributions
azes+0.054
isdom+0.044
pit+0.043
ires+0.043
ement+0.043
Neg logit contributions
ch-0.048
sc-0.048
B-0.047
 subs-0.045
bo-0.043
Token.
Token ID13
Feature activation-2.88e-03

Pos logit contributions
azes+0.135
isdom+0.118
 fridge+0.116
pit+0.115
ocolate+0.115
Neg logit contributions
ch-0.109
sc-0.105
B-0.104
 subs-0.102
text-0.097
Token There
Token ID1318
Feature activation+1.32e-02

Pos logit contributions
azes+0.007
isdom+0.006
pit+0.006
castle+0.006
ement+0.006
Neg logit contributions
sc-0.003
ch-0.003
B-0.003
ny-0.003
bo-0.002
Token were
Token ID547
Feature activation-5.46e-02

Pos logit contributions
s+0.008
 subs+0.007
sc+0.007
l+0.007
fl+0.007
Neg logit contributions
azes-0.035
isdom-0.033
mmm-0.032
sing-0.031
ires-0.031
Token seven
Token ID3598
Feature activation-8.62e-02

Pos logit contributions
azes+0.098
isdom+0.092
ement+0.088
mmm+0.086
ires+0.082
Neg logit contributions
ch-0.081
B-0.079
sc-0.079
 subs-0.074
fl-0.073
Token ants
Token ID27842
Feature activation-3.41e-02

Pos logit contributions
azes+0.151
isdom+0.131
ires+0.126
 fridge+0.126
mmm+0.125
Neg logit contributions
ch-0.132
sc-0.129
B-0.127
 subs-0.123
text-0.117
Token in
Token ID287
Feature activation-4.78e-03

Pos logit contributions
azes+0.078
isdom+0.075
mmm+0.073
ae+0.070
ires+0.070
Neg logit contributions
ch-0.032
sc-0.031
l-0.030
B-0.029
s-0.028
Token the
Token ID262
Feature activation-2.86e-02

Pos logit contributions
azes+0.011
pit+0.009
 oven+0.009
ires+0.009
mmm+0.009
Neg logit contributions
sc-0.005
ch-0.005
B-0.005
 subs-0.005
ched-0.005
Token line
Token ID1627
Feature activation-1.83e-02

Pos logit contributions
azes+0.050
isdom+0.041
mmm+0.041
ires+0.040
ement+0.040
Neg logit contributions
ch-0.047
sc-0.046
B-0.046
 subs-0.044
bo-0.043
Token.
Token ID13
Feature activation+1.75e-03

Pos logit contributions
azes+0.048
isdom+0.046
mmm+0.045
 Majesty+0.044
sing+0.043
Neg logit contributions
sc-0.013
l-0.010
ny-0.010
ch-0.010
ched-0.010
Token.
Token ID13
Feature activation+2.11e-03

Pos logit contributions
ch+0.008
sc+0.008
B+0.008
 subs+0.007
bo+0.007
Neg logit contributions
azes-0.012
pit-0.010
ement-0.010
 oven-0.010
isdom-0.010
Token "
Token ID366
Feature activation+7.02e-04

Pos logit contributions
B+0.003
ch+0.003
sc+0.003
 subs+0.003
bo+0.003
Neg logit contributions
azes-0.004
isdom-0.003
pit-0.003
mmm-0.003
ement-0.003
TokenThey
Token ID2990
Feature activation-1.28e-02

Pos logit contributions
B+0.001
sc+0.001
ch+0.001
 subs+0.001
bo+0.001
Neg logit contributions
azes-0.002
pit-0.001
isdom-0.001
ement-0.001
 oven-0.001
Token can
Token ID460
Feature activation-2.15e-02

Pos logit contributions
azes+0.031
ement+0.028
mmm+0.027
ires+0.027
ummy+0.027
Neg logit contributions
sc-0.013
ch-0.012
B-0.011
ny-0.011
ched-0.011
Token help
Token ID1037
Feature activation+3.97e-03

Pos logit contributions
azes+0.055
ement+0.052
pit+0.049
ae+0.048
isdom+0.048
Neg logit contributions
ch-0.018
bo-0.018
 hurt-0.016
B-0.015
ny-0.014
Token each
Token ID1123
Feature activation-8.13e-02

Pos logit contributions
ch+0.004
B+0.004
sc+0.004
bo+0.003
l+0.003
Neg logit contributions
azes-0.009
pit-0.009
ement-0.009
castle-0.008
ocolate-0.008
Token other
Token ID584
Feature activation+3.72e-03

Pos logit contributions
azes+0.186
pit+0.162
mmm+0.160
ement+0.159
isdom+0.158
Neg logit contributions
ch-0.085
sc-0.085
B-0.085
 subs-0.079
bo-0.073
Token fight
Token ID1907
Feature activation-3.19e-03

Pos logit contributions
sc+0.005
ch+0.005
B+0.004
ny+0.004
bo+0.004
Neg logit contributions
azes-0.008
ement-0.007
mmm-0.007
isdom-0.007
ires-0.007
Token the
Token ID262
Feature activation-7.03e-03

Pos logit contributions
azes+0.008
pit+0.007
ement+0.007
ae+0.007
isdom+0.006
Neg logit contributions
ch-0.003
sc-0.003
B-0.003
ny-0.003
s-0.003
Token bad
Token ID2089
Feature activation+1.68e-03

Pos logit contributions
azes+0.013
pit+0.011
 fridge+0.011
 oven+0.011
ires+0.011
Neg logit contributions
B-0.010
ch-0.010
sc-0.010
 subs-0.010
ched-0.009
Token guys
Token ID3730
Feature activation-3.06e-04

Pos logit contributions
ch+0.003
sc+0.003
B+0.002
 subs+0.002
bo+0.002
Neg logit contributions
azes-0.003
pit-0.003
mmm-0.003
ement-0.003
ires-0.003
Token think
Token ID892
Feature activation-1.35e-02

Pos logit contributions
sc+0.019
ny+0.019
ched+0.018
 scrap+0.018
bo+0.018
Neg logit contributions
azes-0.034
ires-0.029
isdom-0.029
pit-0.029
ement-0.029
Token.
Token ID13
Feature activation+3.74e-03

Pos logit contributions
azes+0.023
pit+0.023
castle+0.020
eria+0.020
 Majesty+0.020
Neg logit contributions
ch-0.021
l-0.020
sc-0.019
s-0.019
ched-0.018
Token They
Token ID1119
Feature activation-9.32e-03

Pos logit contributions
ch+0.005
B+0.005
sc+0.005
 subs+0.004
bo+0.004
Neg logit contributions
azes-0.008
pit-0.007
isdom-0.007
ement-0.007
castle-0.007
Token look
Token ID804
Feature activation-6.36e-03

Pos logit contributions
azes+0.020
ement+0.017
pit+0.017
ummy+0.017
ires+0.017
Neg logit contributions
ny-0.011
 scrap-0.011
sc-0.011
ch-0.011
ched-0.011
Token at
Token ID379
Feature activation-3.55e-02

Pos logit contributions
azes+0.012
mmm+0.011
ocolate+0.011
pit+0.011
castle+0.011
Neg logit contributions
sc-0.009
ch-0.009
B-0.008
 scrap-0.008
ched-0.008
Token each
Token ID1123
Feature activation-8.13e-02

Pos logit contributions
azes+0.071
pit+0.060
mmm+0.059
castle+0.057
isdom+0.057
Neg logit contributions
ch-0.048
B-0.047
sc-0.046
 scrap-0.044
bo-0.044
Token other
Token ID584
Feature activation+7.76e-03

Pos logit contributions
azes+0.189
pit+0.163
mmm+0.162
isdom+0.161
ement+0.161
Neg logit contributions
sc-0.083
B-0.082
ch-0.082
 subs-0.076
bo-0.072
Token.
Token ID13
Feature activation+6.77e-03

Pos logit contributions
sc+0.008
B+0.007
ch+0.007
s+0.007
bo+0.007
Neg logit contributions
azes-0.018
mmm-0.016
isdom-0.016
pit-0.015
castle-0.015
Token They
Token ID1119
Feature activation-1.14e-02

Pos logit contributions
B+0.007
sc+0.007
ch+0.007
Free+0.006
 subs+0.006
Neg logit contributions
azes-0.016
pit-0.014
castle-0.014
isdom-0.014
ement-0.014
Token have
Token ID423
Feature activation+5.39e-03

Pos logit contributions
azes+0.026
ummy+0.023
ement+0.023
ocolate+0.023
pit+0.023
Neg logit contributions
 scrap-0.012
 went-0.011
l-0.011
 st-0.011
bo-0.011
Token an
Token ID281
Feature activation-9.77e-03

Pos logit contributions
ch+0.007
B+0.007
 scrap+0.007
sc+0.006
 subs+0.006
Neg logit contributions
azes-0.011
ement-0.010
pit-0.010
mmm-0.010
ires-0.009
Token tomato
Token ID24240
Feature activation-2.47e-02

Pos logit contributions
azes+0.049
mmm+0.038
isdom+0.038
ocolate+0.037
eria+0.037
Neg logit contributions
ch-0.044
l-0.042
B-0.042
fl-0.041
sc-0.041
Token.
Token ID13
Feature activation+1.14e-02

Pos logit contributions
azes+0.064
isdom+0.059
mmm+0.057
 Majesty+0.056
idge+0.055
Neg logit contributions
sc-0.019
s-0.018
l-0.017
fl-0.017
ch-0.017
Token They
Token ID1119
Feature activation-9.90e-03

Pos logit contributions
sc+0.011
B+0.011
ch+0.011
Free+0.010
 subs+0.010
Neg logit contributions
azes-0.026
pit-0.024
ae-0.023
castle-0.023
 throne-0.023
Token looked
Token ID3114
Feature activation-1.27e-02

Pos logit contributions
azes+0.025
ocolate+0.023
ummy+0.023
ires+0.023
mmm+0.023
Neg logit contributions
 scrap-0.008
B-0.008
 hurt-0.007
ched-0.007
 went-0.007
Token at
Token ID379
Feature activation-2.76e-02

Pos logit contributions
azes+0.025
ocolate+0.021
mmm+0.021
pit+0.021
ae+0.021
Neg logit contributions
sc-0.018
ch-0.018
B-0.017
ched-0.017
 scrap-0.017
Token each
Token ID1123
Feature activation-8.06e-02

Pos logit contributions
azes+0.062
pit+0.055
mmm+0.053
eria+0.052
isdom+0.052
Neg logit contributions
ch-0.030
B-0.030
sc-0.029
 scrap-0.027
 subs-0.027
Token other
Token ID584
Feature activation+1.09e-02

Pos logit contributions
azes+0.174
pit+0.151
 oven+0.149
ires+0.149
isdom+0.148
Neg logit contributions
ch-0.092
sc-0.092
B-0.091
 subs-0.089
bo-0.081
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
sc+0.011
ch+0.011
B+0.010
 subs+0.010
ched+0.010
Neg logit contributions
azes-0.025
isdom-0.022
mmm-0.022
pit-0.022
ires-0.021
Token They
Token ID1119
Feature activation-8.70e-03

Pos logit contributions
sc+0.010
B+0.010
ch+0.010
Free+0.009
Spot+0.008
Neg logit contributions
azes-0.024
pit-0.022
castle-0.022
ae-0.022
 throne-0.022
Token felt
Token ID2936
Feature activation+4.10e-03

Pos logit contributions
azes+0.022
ocolate+0.020
mmm+0.020
ummy+0.020
ires+0.020
Neg logit contributions
 scrap-0.007
B-0.007
 subs-0.006
ched-0.006
ch-0.006
Token bad
Token ID2089
Feature activation+4.46e-03

Pos logit contributions
ch+0.007
B+0.007
sc+0.007
 subs+0.007
ched+0.006
Neg logit contributions
azes-0.007
pit-0.005
isdom-0.005
mmm-0.005
ement-0.005
Token to
Token ID284
Feature activation-2.84e-03

Pos logit contributions
azes+0.021
 writing+0.018
 patiently+0.018
 counting+0.018
 reading+0.018
Neg logit contributions
sc-0.019
text-0.019
ch-0.019
 subs-0.018
B-0.018
Token climb
Token ID12080
Feature activation-1.01e-02

Pos logit contributions
azes+0.006
pit+0.005
 fridge+0.005
 oven+0.005
 writing+0.005
Neg logit contributions
sc-0.004
ch-0.004
B-0.004
 subs-0.003
text-0.003
Token.
Token ID13
Feature activation+3.03e-03

Pos logit contributions
azes+0.019
pit+0.017
 oven+0.017
ement+0.016
 patiently+0.016
Neg logit contributions
B-0.014
 subs-0.014
sc-0.013
ch-0.013
text-0.013
Token They
Token ID1119
Feature activation-1.56e-02

Pos logit contributions
sc+0.004
ch+0.004
B+0.004
 subs+0.004
text+0.003
Neg logit contributions
azes-0.006
isdom-0.005
pit-0.005
castle-0.005
ement-0.005
Token help
Token ID1037
Feature activation+8.44e-03

Pos logit contributions
azes+0.034
ement+0.029
ires+0.029
pit+0.028
mmm+0.028
Neg logit contributions
ch-0.019
sc-0.019
B-0.018
 subs-0.018
bo-0.017
Token each
Token ID1123
Feature activation-7.99e-02

Pos logit contributions
sc+0.015
ch+0.015
B+0.015
 subs+0.015
ched+0.014
Neg logit contributions
azes-0.013
 oven-0.010
pit-0.010
ires-0.010
 fridge-0.010
Token other
Token ID584
Feature activation+1.41e-02

Pos logit contributions
azes+0.150
 oven+0.136
pit+0.133
 fridge+0.132
ires+0.129
Neg logit contributions
sc-0.106
 subs-0.103
ch-0.103
B-0.102
ched-0.098
Token when
Token ID618
Feature activation+2.54e-02

Pos logit contributions
ch+0.019
sc+0.019
B+0.019
 subs+0.018
ched+0.017
Neg logit contributions
azes-0.028
pit-0.024
mmm-0.023
isdom-0.023
ires-0.023
Token the
Token ID262
Feature activation-2.91e-03

Pos logit contributions
ch+0.027
B+0.026
 subs+0.025
sc+0.025
 scrap+0.023
Neg logit contributions
azes-0.057
isdom-0.051
pit-0.051
mmm-0.050
ement-0.050
Token ground
Token ID2323
Feature activation+6.23e-03

Pos logit contributions
azes+0.005
isdom+0.004
mmm+0.004
pit+0.004
ires+0.004
Neg logit contributions
ch-0.005
B-0.005
sc-0.005
 subs-0.005
 scrap-0.004
Token is
Token ID318
Feature activation-2.99e-03

Pos logit contributions
ch+0.007
sc+0.007
B+0.007
 subs+0.007
ched+0.006
Neg logit contributions
azes-0.013
pit-0.012
isdom-0.011
mmm-0.011
ires-0.011
Token.
Token ID13
Feature activation+5.77e-03

Pos logit contributions
s+0.002
sc+0.002
l+0.002
fl+0.002
B+0.002
Neg logit contributions
azes-0.008
isdom-0.007
 Majesty-0.007
mmm-0.007
ocolate-0.007
Token Mom
Token ID11254
Feature activation-2.32e-02

Pos logit contributions
sc+0.008
ch+0.008
B+0.008
 subs+0.007
text+0.007
Neg logit contributions
azes-0.011
isdom-0.010
pit-0.010
castle-0.010
mmm-0.009
Token said
Token ID531
Feature activation-7.26e-03

Pos logit contributions
azes+0.066
ocolate+0.058
ires+0.058
sing+0.058
mmm+0.057
Neg logit contributions
 went-0.015
bo-0.014
B-0.014
s-0.014
ny-0.013
Token they
Token ID484
Feature activation-1.10e-04

Pos logit contributions
azes+0.016
pit+0.014
ires+0.014
sing+0.014
isdom+0.014
Neg logit contributions
ch-0.008
B-0.008
sc-0.008
l-0.007
 subs-0.007
Token could
Token ID714
Feature activation+3.46e-03

Pos logit contributions
azes+0.000
pit+0.000
ires+0.000
mmm+0.000
ement+0.000
Neg logit contributions
 went-0.000
 had-0.000
 got-0.000
B-0.000
ched-0.000
Token each
Token ID1123
Feature activation-7.98e-02

Pos logit contributions
 take+0.003
 went+0.003
bo+0.002
 let+0.002
s+0.002
Neg logit contributions
azes-0.009
ement-0.008
pit-0.008
otion-0.007
isdom-0.007
Token choose
Token ID3853
Feature activation-2.67e-03

Pos logit contributions
azes+0.185
sing+0.152
ement+0.152
pit+0.149
mmm+0.149
Neg logit contributions
sc-0.089
B-0.088
ch-0.083
ny-0.082
bo-0.080
Token one
Token ID530
Feature activation-3.77e-02

Pos logit contributions
azes+0.006
sing+0.005
pit+0.005
 studio+0.005
isdom+0.005
Neg logit contributions
l-0.003
fl-0.003
sc-0.003
B-0.003
s-0.003
Token fruit
Token ID8234
Feature activation-2.84e-02

Pos logit contributions
azes+0.100
 Majesty+0.084
 balcony+0.083
sing+0.083
isdom+0.082
Neg logit contributions
sc-0.034
s-0.030
ch-0.030
fl-0.028
ny-0.028
Token to
Token ID284
Feature activation+1.92e-03

Pos logit contributions
azes+0.072
isdom+0.061
sing+0.059
 Majesty+0.059
 workshop+0.059
Neg logit contributions
sc-0.030
s-0.026
fl-0.025
ch-0.025
ny-0.024
Token take
Token ID1011
Feature activation+2.35e-02

Pos logit contributions
 went+0.002
 take+0.002
 st+0.001
 away+0.001
 hurt+0.001
Neg logit contributions
azes-0.005
eria-0.004
apping-0.004
ement-0.004
pit-0.004
Token was
Token ID373
Feature activation-1.98e-02

Pos logit contributions
azes+0.020
pit+0.017
mmm+0.017
isdom+0.017
ires+0.017
Neg logit contributions
ch-0.009
B-0.009
sc-0.009
 subs-0.009
ched-0.008
Token.
Token ID13
Feature activation+3.32e-03

Pos logit contributions
azes+0.043
pit+0.036
ement+0.036
mmm+0.035
isdom+0.035
Neg logit contributions
ch-0.025
B-0.024
sc-0.023
 subs-0.023
ched-0.022
Token They
Token ID1119
Feature activation-1.55e-02

Pos logit contributions
B+0.004
sc+0.004
ch+0.004
 subs+0.004
Br+0.004
Neg logit contributions
azes-0.007
pit-0.006
castle-0.006
isdom-0.006
ement-0.006
Token looked
Token ID3114
Feature activation-2.48e-02

Pos logit contributions
azes+0.037
ement+0.032
ires+0.031
pit+0.031
mmm+0.031
Neg logit contributions
B-0.016
sc-0.016
ch-0.015
 subs-0.015
ched-0.014
Token at
Token ID379
Feature activation-3.35e-02

Pos logit contributions
azes+0.053
mmm+0.046
ocolate+0.045
ae+0.045
pit+0.044
Neg logit contributions
sc-0.032
ch-0.032
B-0.029
ched-0.029
 scrap-0.029
Token each
Token ID1123
Feature activation-7.96e-02

Pos logit contributions
azes+0.078
pit+0.068
mmm+0.068
eria+0.066
castle+0.065
Neg logit contributions
B-0.035
ch-0.035
sc-0.032
 scrap-0.032
bo-0.031
Token other
Token ID584
Feature activation+5.49e-03

Pos logit contributions
azes+0.185
pit+0.159
mmm+0.157
isdom+0.157
ement+0.156
Neg logit contributions
sc-0.083
ch-0.083
B-0.082
 subs-0.076
ched-0.072
Token and
Token ID290
Feature activation+6.34e-03

Pos logit contributions
sc+0.006
s+0.006
ch+0.005
B+0.005
ny+0.005
Neg logit contributions
azes-0.012
mmm-0.011
isdom-0.011
pit-0.010
castle-0.010
Token felt
Token ID2936
Feature activation-4.02e-03

Pos logit contributions
B+0.007
sc+0.007
 scrap+0.007
ch+0.007
 subs+0.007
Neg logit contributions
azes-0.014
mmm-0.012
ires-0.012
ocolate-0.012
isdom-0.012
Token sorry
Token ID7926
Feature activation+4.53e-03

Pos logit contributions
azes+0.007
pit+0.006
mmm+0.006
isdom+0.006
ement+0.005
Neg logit contributions
ch-0.007
sc-0.006
B-0.006
 subs-0.006
ched-0.006
Token.
Token ID13
Feature activation+3.14e-03

Pos logit contributions
ch+0.005
sc+0.005
B+0.005
 subs+0.004
fl+0.004
Neg logit contributions
azes-0.010
pit-0.009
isdom-0.009
castle-0.009
 balcony-0.009
Token hoop
Token ID44792
Feature activation-3.26e-03

Pos logit contributions
sc+0.002
ch+0.002
B+0.002
 subs+0.002
 scrap+0.001
Neg logit contributions
azes-0.002
mmm-0.002
isdom-0.002
pit-0.002
ement-0.002
Token.
Token ID13
Feature activation+3.95e-03

Pos logit contributions
azes+0.008
mmm+0.008
isdom+0.008
ocolate+0.007
mmmm+0.007
Neg logit contributions
sc-0.003
s-0.003
ch-0.002
l-0.002
ched-0.002
Token They
Token ID1119
Feature activation-1.12e-02

Pos logit contributions
sc+0.004
ch+0.004
B+0.004
 subs+0.004
text+0.003
Neg logit contributions
azes-0.009
pit-0.008
castle-0.008
isdom-0.008
 balcony-0.008
Token look
Token ID804
Feature activation-1.43e-02

Pos logit contributions
azes+0.024
ummy+0.021
mmm+0.021
ocolate+0.021
ement+0.021
Neg logit contributions
 scrap-0.013
sc-0.013
ched-0.013
ny-0.013
ch-0.012
Token at
Token ID379
Feature activation-3.73e-02

Pos logit contributions
azes+0.028
pit+0.023
ocolate+0.023
mmm+0.023
isdom+0.022
Neg logit contributions
sc-0.021
ch-0.021
B-0.020
ched-0.019
 subs-0.019
Token each
Token ID1123
Feature activation-7.96e-02

Pos logit contributions
azes+0.079
pit+0.069
mmm+0.067
castle+0.065
 balcony+0.065
Neg logit contributions
ch-0.046
B-0.046
sc-0.045
 scrap-0.042
ched-0.041
Token other
Token ID584
Feature activation+1.11e-02

Pos logit contributions
azes+0.159
pit+0.139
ires+0.137
 oven+0.136
ement+0.135
Neg logit contributions
ch-0.101
 subs-0.100
B-0.099
sc-0.097
bo-0.092
Token.
Token ID13
Feature activation+4.57e-03

Pos logit contributions
sc+0.013
ch+0.012
B+0.012
s+0.011
ny+0.011
Neg logit contributions
azes-0.024
mmm-0.021
isdom-0.021
ocolate-0.021
pit-0.021
Token
Token ID198
Feature activation-2.75e-03

Pos logit contributions
sc+0.005
B+0.005
ch+0.005
Free+0.004
W+0.004
Neg logit contributions
azes-0.011
pit-0.009
castle-0.009
isdom-0.009
 balcony-0.009
Token
Token ID198
Feature activation-6.75e-03

Pos logit contributions
azes+0.007
pit+0.006
castle+0.006
 oven+0.006
ement+0.006
Neg logit contributions
B-0.003
ch-0.002
sc-0.002
ched-0.002
bo-0.002
Token"
Token ID1
Feature activation+2.52e-02

Pos logit contributions
azes+0.014
mmm+0.012
pit+0.012
ires+0.012
isdom+0.012
Neg logit contributions
 subs-0.008
sc-0.008
ch-0.008
B-0.008
fl-0.008
Token stick
Token ID4859
Feature activation-6.28e-03

Pos logit contributions
ch+0.006
sc+0.005
B+0.005
fl+0.005
 subs+0.005
Neg logit contributions
azes-0.007
mmm-0.006
ires-0.006
isdom-0.006
ement-0.006
Token.
Token ID13
Feature activation+6.20e-03

Pos logit contributions
azes+0.016
mmm+0.014
isdom+0.014
 Majesty+0.013
ires+0.013
Neg logit contributions
sc-0.006
ch-0.006
B-0.005
ny-0.005
ched-0.005
Token They
Token ID1119
Feature activation-8.75e-03

Pos logit contributions
ch+0.007
sc+0.007
B+0.006
 subs+0.006
Free+0.006
Neg logit contributions
azes-0.014
pit-0.013
castle-0.012
isdom-0.012
ement-0.012
Token look
Token ID804
Feature activation-2.58e-03

Pos logit contributions
azes+0.019
ement+0.016
ummy+0.016
pit+0.016
ires+0.016
Neg logit contributions
 scrap-0.010
ch-0.010
sc-0.010
bo-0.009
ched-0.009
Token at
Token ID379
Feature activation-3.40e-02

Pos logit contributions
azes+0.005
ocolate+0.004
pit+0.004
mmm+0.004
isdom+0.004
Neg logit contributions
ch-0.004
sc-0.004
B-0.004
ched-0.003
 scrap-0.003
Token each
Token ID1123
Feature activation-7.92e-02

Pos logit contributions
azes+0.078
pit+0.068
mmm+0.068
isdom+0.066
castle+0.066
Neg logit contributions
ch-0.035
B-0.034
sc-0.034
 subs-0.031
 scrap-0.030
Token other
Token ID584
Feature activation+6.45e-03

Pos logit contributions
azes+0.152
pit+0.135
 oven+0.134
 fridge+0.132
ires+0.131
Neg logit contributions
 subs-0.103
ch-0.102
sc-0.102
B-0.101
text-0.094
Token.
Token ID13
Feature activation+6.41e-03

Pos logit contributions
sc+0.007
ch+0.007
B+0.007
 subs+0.006
ched+0.006
Neg logit contributions
azes-0.014
mmm-0.013
isdom-0.013
pit-0.012
ires-0.012
Token
Token ID198
Feature activation-6.58e-03

Pos logit contributions
B+0.007
ch+0.007
sc+0.007
 subs+0.006
Br+0.006
Neg logit contributions
azes-0.014
pit-0.013
castle-0.013
ement-0.013
isdom-0.013
Token
Token ID198
Feature activation-8.60e-03

Pos logit contributions
azes+0.016
pit+0.014
castle+0.014
ement+0.014
 oven+0.014
Neg logit contributions
B-0.006
ch-0.006
sc-0.006
ched-0.005
bo-0.005
TokenThey
Token ID2990
Feature activation-1.04e-02

Pos logit contributions
azes+0.016
mmm+0.014
ires+0.014
isdom+0.013
eria+0.013
Neg logit contributions
 subs-0.012
sc-0.012
ch-0.012
fl-0.011
 scrap-0.011
Token
Token ID198
Feature activation+6.69e-03

Pos logit contributions
sc+0.014
ch+0.014
 subs+0.014
B+0.013
fl+0.013
Neg logit contributions
azes-0.021
ires-0.018
mmm-0.018
isdom-0.018
pit-0.017
Token"
Token ID1
Feature activation+2.46e-02

Pos logit contributions
 subs+0.010
fl+0.010
sc+0.010
ch+0.010
 scrap+0.010
Neg logit contributions
azes-0.011
pit-0.010
ires-0.010
 patiently-0.010
mmm-0.010
TokenTom
Token ID13787
Feature activation-2.15e-03

Pos logit contributions
sc+0.030
ch+0.030
 subs+0.030
text+0.028
ched+0.028
Neg logit contributions
azes-0.049
pit-0.043
mmm-0.042
ires-0.042
ocolate-0.041
Token,
Token ID11
Feature activation-2.22e-02

Pos logit contributions
azes+0.004
ires+0.003
pit+0.003
ement+0.003
isdom+0.003
Neg logit contributions
ch-0.003
B-0.003
 subs-0.003
sc-0.003
bo-0.003
Token what
Token ID644
Feature activation+1.26e-02

Pos logit contributions
azes+0.042
ement+0.036
isdom+0.035
pit+0.035
ires+0.035
Neg logit contributions
ch-0.033
B-0.031
sc-0.031
 subs-0.030
Fl-0.029
Token are
Token ID389
Feature activation-7.87e-02

Pos logit contributions
ch+0.017
sc+0.017
B+0.017
 subs+0.016
bo+0.016
Neg logit contributions
azes-0.025
pit-0.021
isdom-0.020
mmm-0.020
ires-0.020
Token you
Token ID345
Feature activation-4.36e-02

Pos logit contributions
azes+0.147
pit+0.125
ement+0.122
isdom+0.121
mmm+0.120
Neg logit contributions
ch-0.114
sc-0.113
B-0.113
 subs-0.108
ched-0.103
Token doing
Token ID1804
Feature activation-3.19e-02

Pos logit contributions
azes+0.077
ement+0.074
ues+0.071
 throne+0.070
pit+0.069
Neg logit contributions
ch-0.058
 st-0.055
B-0.055
 hurt-0.053
 subs-0.053
Token with
Token ID351
Feature activation-3.74e-02

Pos logit contributions
azes+0.059
pit+0.051
 balcony+0.049
eria+0.049
ement+0.049
Neg logit contributions
ch-0.045
B-0.043
sc-0.042
bo-0.040
ched-0.039
Token that
Token ID326
Feature activation-4.42e-02

Pos logit contributions
azes+0.092
pit+0.084
ement+0.081
castle+0.080
sing+0.079
Neg logit contributions
ch-0.036
B-0.032
 scrap-0.031
Fl-0.029
 st-0.028
Token worm
Token ID23849
Feature activation-4.90e-02

Pos logit contributions
azes+0.078
ement+0.062
ires+0.061
isdom+0.060
pit+0.060
Neg logit contributions
ch-0.073
sc-0.070
B-0.069
 subs-0.067
bo-0.067
Token say
Token ID910
Feature activation-1.93e-02

Pos logit contributions
ch+0.007
sc+0.007
B+0.007
 subs+0.007
ched+0.007
Neg logit contributions
azes-0.010
pit-0.008
isdom-0.008
mmm-0.008
ires-0.008
Token "
Token ID366
Feature activation+5.19e-03

Pos logit contributions
azes+0.030
pit+0.025
mmm+0.024
isdom+0.024
ires+0.024
Neg logit contributions
ch-0.034
B-0.034
sc-0.034
 subs-0.033
ched-0.031
Tokensorry
Token ID41599
Feature activation-2.81e-02

Pos logit contributions
sc+0.006
ch+0.006
text+0.006
 subs+0.006
ched+0.006
Neg logit contributions
azes-0.009
mmm-0.009
ement-0.009
ae-0.008
ires-0.008
Token"
Token ID1
Feature activation+2.05e-02

Pos logit contributions
azes+0.038
mmm+0.034
ocolate+0.033
ummy+0.033
ement+0.032
Neg logit contributions
sc-0.050
 scrap-0.049
 subs-0.049
B-0.049
ch-0.048
Token to
Token ID284
Feature activation-7.63e-03

Pos logit contributions
ch+0.032
B+0.031
sc+0.031
 subs+0.030
ched+0.029
Neg logit contributions
azes-0.036
pit-0.030
mmm-0.029
isdom-0.029
ires-0.029
Token each
Token ID1123
Feature activation-7.81e-02

Pos logit contributions
azes+0.013
pit+0.011
isdom+0.011
mmm+0.011
ires+0.011
Neg logit contributions
ch-0.012
sc-0.012
B-0.012
 subs-0.012
ched-0.011
Token other
Token ID584
Feature activation+1.48e-02

Pos logit contributions
azes+0.119
 oven+0.110
ocolate+0.110
 baker+0.105
ae+0.105
Neg logit contributions
 subs-0.124
sc-0.116
fl-0.112
ch-0.112
bo-0.112
Token.
Token ID13
Feature activation-2.48e-03

Pos logit contributions
sc+0.021
ch+0.021
B+0.021
 subs+0.021
ched+0.019
Neg logit contributions
azes-0.027
pit-0.023
 oven-0.023
 fridge-0.023
 writing-0.022
Token They
Token ID1119
Feature activation-1.07e-02

Pos logit contributions
azes+0.006
castle+0.005
pit+0.005
isdom+0.005
ement+0.005
Neg logit contributions
sc-0.002
B-0.002
ch-0.002
Fl-0.002
 subs-0.002
Token hug
Token ID16225
Feature activation-1.34e-02

Pos logit contributions
azes+0.019
 oven+0.016
pit+0.015
 fridge+0.015
isdom+0.015
Neg logit contributions
B-0.017
ch-0.016
sc-0.016
 subs-0.016
fl-0.015
Token and
Token ID290
Feature activation+9.46e-03

Pos logit contributions
azes+0.021
pit+0.017
 oven+0.017
isdom+0.017
mmm+0.017
Neg logit contributions
sc-0.023
ch-0.023
B-0.023
 subs-0.022
ched-0.021
Token,
Token ID11
Feature activation-3.37e-02

Pos logit contributions
azes+0.026
pit+0.021
isdom+0.021
ires+0.021
mmm+0.021
Neg logit contributions
ch-0.027
sc-0.027
B-0.027
 subs-0.026
bo-0.025
Token sweet
Token ID6029
Feature activation-2.96e-02

Pos logit contributions
azes+0.069
isdom+0.058
pit+0.058
ement+0.058
 balcony+0.057
Neg logit contributions
B-0.044
ch-0.044
sc-0.042
 subs-0.040
Fl-0.039
Tokenie
Token ID494
Feature activation-3.46e-03

Pos logit contributions
azes+0.051
pit+0.042
 fridge+0.041
isdom+0.041
mmm+0.041
Neg logit contributions
sc-0.049
ch-0.048
B-0.048
 subs-0.046
ched-0.044
Token.
Token ID13
Feature activation-9.03e-03

Pos logit contributions
azes+0.006
pit+0.005
 oven+0.005
 fridge+0.005
isdom+0.005
Neg logit contributions
B-0.005
sc-0.005
ch-0.005
 subs-0.005
ched-0.005
Token What
Token ID1867
Feature activation-2.03e-03

Pos logit contributions
azes+0.017
pit+0.015
castle+0.015
ement+0.014
isdom+0.014
Neg logit contributions
ch-0.013
B-0.013
sc-0.013
 subs-0.012
Fl-0.012
Token are
Token ID389
Feature activation-7.81e-02

Pos logit contributions
azes+0.004
pit+0.004
ement+0.004
ires+0.004
castle+0.004
Neg logit contributions
ch-0.003
B-0.002
sc-0.002
 subs-0.002
ched-0.002
Token you
Token ID345
Feature activation-3.84e-02

Pos logit contributions
azes+0.145
pit+0.124
ement+0.120
isdom+0.120
mmm+0.119
Neg logit contributions
ch-0.114
B-0.113
sc-0.113
 subs-0.108
bo-0.102
Token expecting
Token ID12451
Feature activation-4.67e-02

Pos logit contributions
azes+0.070
ement+0.067
isdom+0.063
ues+0.063
pit+0.063
Neg logit contributions
ch-0.051
 st-0.049
B-0.047
fl-0.046
 hurt-0.046
Token?"
Token ID1701
Feature activation-1.78e-02

Pos logit contributions
azes+0.096
pit+0.093
castle+0.085
ement+0.085
 Majesty+0.084
Neg logit contributions
ch-0.070
sc-0.061
B-0.058
 on-0.058
 st-0.058
Token Dad
Token ID17415
Feature activation-2.14e-02

Pos logit contributions
azes+0.031
pit+0.030
castle+0.029
ement+0.028
eria+0.027
Neg logit contributions
B-0.026
 scrap-0.025
ch-0.025
bo-0.024
 subs-0.024
Token says
Token ID1139
Feature activation-1.35e-03

Pos logit contributions
azes+0.039
mmm+0.034
pit+0.033
isdom+0.033
castle+0.033
Neg logit contributions
ch-0.031
sc-0.031
B-0.030
bo-0.030
 subs-0.029
Token looks
Token ID3073
Feature activation-2.05e-02

Pos logit contributions
sc+0.009
s+0.009
bo+0.009
ch+0.009
ny+0.008
Neg logit contributions
azes-0.019
ocolate-0.018
mmm-0.018
isdom-0.018
ruits-0.018
Token.
Token ID13
Feature activation+5.73e-03

Pos logit contributions
azes+0.038
mmm+0.032
pit+0.032
ocolate+0.032
isdom+0.032
Neg logit contributions
sc-0.030
ch-0.030
B-0.029
 subs-0.028
ched-0.028
Token She
Token ID1375
Feature activation-6.75e-03

Pos logit contributions
ch+0.007
sc+0.007
B+0.007
Free+0.006
W+0.006
Neg logit contributions
castle-0.012
azes-0.011
pit-0.011
isdom-0.011
ement-0.010
Token sees
Token ID7224
Feature activation-3.46e-02

Pos logit contributions
ocolate+0.015
azes+0.015
mmm+0.014
eria+0.014
ummy+0.014
Neg logit contributions
 scrap-0.007
ch-0.006
bo-0.006
l-0.006
sc-0.006
Token the
Token ID262
Feature activation-1.95e-02

Pos logit contributions
azes+0.089
mmm+0.086
isdom+0.084
castle+0.083
pit+0.081
Neg logit contributions
l-0.024
ch-0.022
B-0.021
 scrap-0.020
s-0.018
Token ele
Token ID9766
Feature activation-7.80e-02

Pos logit contributions
azes+0.040
isdom+0.034
pit+0.033
mmm+0.033
ires+0.033
Neg logit contributions
ch-0.025
sc-0.025
B-0.025
 subs-0.024
bo-0.023
Tokenventh
Token ID20987
Feature activation+1.67e-02

Pos logit contributions
azes+0.142
isdom+0.138
pit+0.134
 fridge+0.132
ires+0.130
Neg logit contributions
B-0.101
ch-0.101
sc-0.098
text-0.095
bo-0.089
Token pengu
Token ID38373
Feature activation-3.71e-02

Pos logit contributions
ch+0.023
B+0.023
sc+0.023
 subs+0.022
ched+0.021
Neg logit contributions
azes-0.032
pit-0.027
 oven-0.026
ires-0.026
isdom-0.026
Tokenin
Token ID259
Feature activation-5.84e-03

Pos logit contributions
azes+0.069
 fridge+0.062
mmm+0.061
isdom+0.059
sing+0.059
Neg logit contributions
ch-0.054
B-0.054
sc-0.053
bo-0.051
l-0.049
Token.
Token ID13
Feature activation+2.12e-03

Pos logit contributions
azes+0.014
mmm+0.013
isdom+0.012
ement+0.012
 balcony+0.012
Neg logit contributions
ch-0.006
sc-0.005
B-0.005
bo-0.005
ched-0.005
Token She
Token ID1375
Feature activation-2.07e-03

Pos logit contributions
ch+0.002
sc+0.002
B+0.002
Fl+0.002
W+0.002
Neg logit contributions
castle-0.005
azes-0.004
pit-0.004
isdom-0.004
 workshop-0.004
Token
Token ID198
Feature activation-6.01e-03

Pos logit contributions
ch+0.000
B+0.000
sc+0.000
 subs+0.000
bo+0.000
Neg logit contributions
azes-0.000
pit-0.000
isdom-0.000
ement-0.000
mmm-0.000
Token
Token ID198
Feature activation-9.89e-03

Pos logit contributions
azes+0.012
ires+0.010
mmm+0.010
isdom+0.010
 fridge+0.010
Neg logit contributions
sc-0.008
 subs-0.008
ch-0.008
B-0.008
fl-0.007
TokenThey
Token ID2990
Feature activation-1.48e-02

Pos logit contributions
azes+0.014
mmm+0.012
ires+0.012
eria+0.012
 gods+0.011
Neg logit contributions
 subs-0.019
burn-0.017
fl-0.017
sc-0.016
ch-0.016
Token look
Token ID804
Feature activation-3.58e-03

Pos logit contributions
azes+0.029
pit+0.024
ires+0.024
ement+0.024
mmm+0.024
Neg logit contributions
ch-0.020
sc-0.020
B-0.020
 subs-0.019
ched-0.018
Token at
Token ID379
Feature activation-3.32e-02

Pos logit contributions
azes+0.005
pit+0.004
ires+0.004
ement+0.004
 oven+0.004
Neg logit contributions
B-0.006
ch-0.006
sc-0.006
 subs-0.006
bo-0.006
Token each
Token ID1123
Feature activation-7.76e-02

Pos logit contributions
azes+0.076
pit+0.067
mmm+0.066
isdom+0.065
ement+0.065
Neg logit contributions
ch-0.034
B-0.034
sc-0.033
 subs-0.031
bo-0.030
Token other
Token ID584
Feature activation+1.65e-02

Pos logit contributions
azes+0.110
 oven+0.102
pit+0.098
 fridge+0.098
ires+0.096
Neg logit contributions
 subs-0.132
sc-0.124
ch-0.121
bo-0.117
beh-0.117
Token.
Token ID13
Feature activation+4.53e-03

Pos logit contributions
ch+0.021
sc+0.021
B+0.021
 subs+0.020
ched+0.019
Neg logit contributions
azes-0.034
pit-0.029
isdom-0.028
ires-0.028
ement-0.028
Token They
Token ID1119
Feature activation-1.15e-02

Pos logit contributions
B+0.005
ch+0.005
sc+0.005
 subs+0.005
Fl+0.004
Neg logit contributions
azes-0.010
pit-0.009
castle-0.009
ement-0.009
isdom-0.009
Token hug
Token ID16225
Feature activation-1.21e-02

Pos logit contributions
azes+0.023
pit+0.020
mmm+0.020
ement+0.020
ocolate+0.020
Neg logit contributions
ch-0.014
 scrap-0.014
sc-0.014
 subs-0.014
B-0.014
Token the
Token ID262
Feature activation-1.41e-02

Pos logit contributions
azes+0.022
pit+0.018
isdom+0.018
ires+0.018
 oven+0.018
Neg logit contributions
sc-0.019
ch-0.019
B-0.018
 subs-0.018
ched-0.017
Token and
Token ID290
Feature activation+1.26e-02

Pos logit contributions
azes+0.052
isdom+0.048
ocolate+0.048
mmm+0.047
 oven+0.047
Neg logit contributions
fl-0.020
B-0.019
 subs-0.018
s-0.018
ch-0.018
Token grateful
Token ID14066
Feature activation+2.81e-03

Pos logit contributions
ch+0.026
B+0.026
sc+0.026
 subs+0.024
text+0.023
Neg logit contributions
azes-0.017
 writing-0.014
 balcony-0.013
sing-0.013
 patiently-0.013
Token.
Token ID13
Feature activation-8.28e-03

Pos logit contributions
ch+0.003
sc+0.003
B+0.003
 subs+0.003
fl+0.003
Neg logit contributions
azes-0.006
pit-0.005
isdom-0.005
mmm-0.005
ement-0.005
Token They
Token ID1119
Feature activation-1.66e-02

Pos logit contributions
azes+0.015
ires+0.013
mmm+0.012
 fridge+0.012
isdom+0.012
Neg logit contributions
sc-0.013
ch-0.012
 subs-0.012
B-0.012
ched-0.012
Token loved
Token ID6151
Feature activation-2.99e-02

Pos logit contributions
azes+0.040
pit+0.034
ires+0.034
ement+0.034
isdom+0.034
Neg logit contributions
B-0.016
sc-0.016
ch-0.016
 subs-0.015
ched-0.014
Token each
Token ID1123
Feature activation-7.75e-02

Pos logit contributions
azes+0.074
pit+0.068
ement+0.068
mmm+0.066
eria+0.065
Neg logit contributions
fl-0.025
ch-0.024
B-0.023
 subs-0.022
 st-0.020
Token other
Token ID584
Feature activation+1.53e-02

Pos logit contributions
azes+0.171
pit+0.149
 oven+0.146
ires+0.146
isdom+0.146
Neg logit contributions
ch-0.086
sc-0.086
B-0.085
 subs-0.081
ched-0.075
Token very
Token ID845
Feature activation-6.92e-04

Pos logit contributions
sc+0.018
B+0.017
ch+0.017
bo+0.016
 subs+0.016
Neg logit contributions
azes-0.034
isdom-0.029
ires-0.029
mmm-0.029
pit-0.029
Token much
Token ID881
Feature activation+1.70e-02

Pos logit contributions
azes+0.001
pit+0.001
 fridge+0.001
 oven+0.001
isdom+0.001
Neg logit contributions
B-0.001
ch-0.001
sc-0.001
bo-0.001
text-0.001
Token.
Token ID13
Feature activation-3.99e-03

Pos logit contributions
sc+0.014
B+0.013
ch+0.013
l+0.012
ny+0.012
Neg logit contributions
azes-0.043
isdom-0.038
pit-0.038
ocolate-0.037
mmm-0.037
Token<|endoftext|>
Token ID50256
Feature activation-4.86e-03

Pos logit contributions
azes+0.008
ires+0.007
mmm+0.007
pit+0.007
isdom+0.007
Neg logit contributions
sc-0.005
ch-0.005
 subs-0.005
B-0.005
ched-0.005
Token,
Token ID11
Feature activation-3.09e-02

Pos logit contributions
azes+0.142
isdom+0.123
eria+0.121
sing+0.121
 workshop+0.119
Neg logit contributions
ch-0.096
sc-0.091
l-0.088
 subs-0.088
B-0.087
Token sixteen
Token ID27913
Feature activation-9.45e-02

Pos logit contributions
azes+0.046
isdom+0.036
pit+0.035
ement+0.035
ires+0.035
Neg logit contributions
ch-0.059
sc-0.057
B-0.055
 subs-0.055
bo-0.054
Token,
Token ID11
Feature activation-3.62e-02

Pos logit contributions
azes+0.183
isdom+0.159
eria+0.158
sing+0.157
 workshop+0.154
Neg logit contributions
ch-0.125
sc-0.122
B-0.119
fl-0.119
 subs-0.118
Token seventeen
Token ID38741
Feature activation-6.57e-02

Pos logit contributions
azes+0.058
isdom+0.046
pit+0.045
ires+0.044
ement+0.044
Neg logit contributions
ch-0.066
sc-0.064
B-0.062
 subs-0.061
bo-0.060
Token,
Token ID11
Feature activation-2.90e-02

Pos logit contributions
azes+0.126
eria+0.113
sing+0.111
 balcony+0.111
isdom+0.111
Neg logit contributions
ch-0.084
fl-0.079
l-0.078
 subs-0.077
bo-0.077
Token eighteen
Token ID29095
Feature activation-7.73e-02

Pos logit contributions
azes+0.045
isdom+0.035
pit+0.034
ement+0.034
ires+0.034
Neg logit contributions
ch-0.054
sc-0.052
B-0.050
 subs-0.050
bo-0.050
Token,
Token ID11
Feature activation-2.65e-02

Pos logit contributions
azes+0.158
isdom+0.144
eria+0.144
 workshop+0.143
sing+0.141
Neg logit contributions
l-0.085
ch-0.085
s-0.082
fl-0.081
 subs-0.080
Token nineteen
Token ID43063
Feature activation-6.14e-02

Pos logit contributions
azes+0.041
isdom+0.033
ement+0.032
ires+0.031
pit+0.031
Neg logit contributions
ch-0.049
sc-0.047
 subs-0.046
B-0.045
bo-0.045
Token,
Token ID11
Feature activation-2.99e-02

Pos logit contributions
azes+0.132
eria+0.118
sing+0.118
ocolate+0.117
 balcony+0.115
Neg logit contributions
l-0.066
s-0.060
ch-0.059
sc-0.058
 subs-0.057
Token twenty
Token ID8208
Feature activation-4.33e-02

Pos logit contributions
azes+0.044
isdom+0.036
eria+0.033
ocolate+0.033
 journal+0.033
Neg logit contributions
ch-0.056
sc-0.052
l-0.052
bo-0.052
fl-0.052
Token,
Token ID11
Feature activation-3.49e-02

Pos logit contributions
azes+0.083
 fridge+0.070
isdom+0.070
pit+0.069
sing+0.069
Neg logit contributions
ch-0.060
sc-0.058
B-0.057
 subs-0.055
bo-0.054
Token,
Token ID11
Feature activation-1.38e-02

Pos logit contributions
azes+0.043
isdom+0.039
eria+0.037
ocolate+0.036
ires+0.036
Neg logit contributions
ch-0.026
sc-0.024
 subs-0.023
bo-0.023
fl-0.023
Token fourteen
Token ID29167
Feature activation-5.41e-02

Pos logit contributions
azes+0.023
isdom+0.019
ement+0.018
eria+0.018
 workshop+0.017
Neg logit contributions
ch-0.024
sc-0.023
bo-0.022
B-0.022
 subs-0.022
Token,
Token ID11
Feature activation-2.06e-02

Pos logit contributions
azes+0.110
isdom+0.102
eria+0.100
sing+0.097
ocolate+0.097
Neg logit contributions
ch-0.061
 subs-0.060
l-0.060
s-0.057
 scrap-0.057
Token fifteen
Token ID17280
Feature activation-5.85e-02

Pos logit contributions
azes+0.031
isdom+0.026
ement+0.024
pit+0.024
 studio+0.024
Neg logit contributions
ch-0.039
sc-0.036
bo-0.036
 subs-0.036
B-0.035
Token,
Token ID11
Feature activation-2.49e-02

Pos logit contributions
azes+0.120
isdom+0.111
sing+0.107
eria+0.106
ocolate+0.104
Neg logit contributions
ch-0.067
l-0.064
 subs-0.061
s-0.060
sc-0.060
Token sixteen
Token ID27913
Feature activation-7.73e-02

Pos logit contributions
azes+0.039
isdom+0.032
ement+0.030
pit+0.029
ires+0.029
Neg logit contributions
ch-0.047
sc-0.045
bo-0.043
B-0.042
 subs-0.042
Token,
Token ID11
Feature activation-2.89e-02

Pos logit contributions
azes+0.161
isdom+0.151
sing+0.149
eria+0.148
itor+0.145
Neg logit contributions
l-0.081
s-0.079
ch-0.079
 subs-0.076
fl-0.073
Token seventeen
Token ID38741
Feature activation-5.89e-02

Pos logit contributions
azes+0.048
isdom+0.039
ires+0.037
pit+0.037
ement+0.037
Neg logit contributions
ch-0.052
sc-0.050
bo-0.047
B-0.046
 subs-0.046
Token,
Token ID11
Feature activation-2.40e-02

Pos logit contributions
azes+0.114
eria+0.106
isdom+0.105
sing+0.105
 balcony+0.104
Neg logit contributions
ch-0.069
l-0.065
fl-0.065
bo-0.064
s-0.064
Token eighteen
Token ID29095
Feature activation-6.88e-02

Pos logit contributions
azes+0.039
isdom+0.032
ement+0.030
pit+0.030
ires+0.029
Neg logit contributions
ch-0.044
sc-0.041
bo-0.040
 subs-0.039
B-0.039
Token,
Token ID11
Feature activation-2.26e-02

Pos logit contributions
azes+0.142
isdom+0.138
ocolate+0.135
eria+0.133
itor+0.133
Neg logit contributions
s-0.070
l-0.069
ch-0.066
 subs-0.062
fl-0.060
Token They
Token ID1119
Feature activation-4.69e-03

Pos logit contributions
ch+0.002
sc+0.002
B+0.002
 subs+0.001
ched+0.001
Neg logit contributions
azes-0.002
pit-0.002
isdom-0.002
mmm-0.002
ires-0.002
Token learned
Token ID4499
Feature activation+2.27e-02

Pos logit contributions
azes+0.007
 patiently+0.006
isdom+0.006
 fridge+0.006
 oven+0.006
Neg logit contributions
ch-0.008
sc-0.008
B-0.008
 subs-0.007
fl-0.007
Token to
Token ID284
Feature activation+8.24e-03

Pos logit contributions
sc+0.032
ch+0.032
B+0.031
 subs+0.031
bo+0.029
Neg logit contributions
azes-0.043
ires-0.036
isdom-0.036
pit-0.036
 oven-0.036
Token recommend
Token ID4313
Feature activation-1.31e-02

Pos logit contributions
sc+0.012
ch+0.012
B+0.012
 subs+0.012
ched+0.011
Neg logit contributions
azes-0.015
pit-0.012
isdom-0.012
mmm-0.012
 oven-0.012
Token to
Token ID284
Feature activation-9.41e-03

Pos logit contributions
azes+0.027
pit+0.025
mmm+0.023
ement+0.023
isdom+0.023
Neg logit contributions
ch-0.017
B-0.016
sc-0.016
 subs-0.015
ched-0.014
Token each
Token ID1123
Feature activation-7.73e-02

Pos logit contributions
azes+0.019
pit+0.016
ement+0.016
mmm+0.016
ires+0.016
Neg logit contributions
ch-0.012
B-0.012
sc-0.012
 subs-0.011
bo-0.011
Token other
Token ID584
Feature activation-5.54e-04

Pos logit contributions
azes+0.136
 oven+0.125
 fridge+0.120
ires+0.118
 office+0.117
Neg logit contributions
 subs-0.112
sc-0.111
ch-0.105
ched-0.104
B-0.103
Token what
Token ID644
Feature activation+2.79e-02

Pos logit contributions
azes+0.001
pit+0.001
isdom+0.001
ires+0.001
 oven+0.001
Neg logit contributions
ch-0.001
sc-0.001
B-0.001
 subs-0.001
bo-0.001
Token was
Token ID373
Feature activation-4.31e-02

Pos logit contributions
sc+0.040
ch+0.038
B+0.038
 subs+0.037
bo+0.035
Neg logit contributions
azes-0.054
 writing-0.046
isdom-0.045
ires-0.045
 oven-0.044
Token best
Token ID1266
Feature activation-5.09e-02

Pos logit contributions
azes+0.078
pit+0.066
mmm+0.064
ement+0.064
isdom+0.064
Neg logit contributions
ch-0.067
sc-0.066
B-0.065
 subs-0.062
ched-0.060
Token for
Token ID329
Feature activation-2.24e-02

Pos logit contributions
azes+0.103
pit+0.088
isdom+0.086
mmm+0.086
ement+0.086
Neg logit contributions
ch-0.067
sc-0.066
B-0.066
 subs-0.063
ched-0.059
Token Then
Token ID3244
Feature activation+2.05e-02

Pos logit contributions
sc+0.010
ch+0.010
B+0.009
text+0.009
Fl+0.009
Neg logit contributions
azes-0.022
castle-0.021
pit-0.021
isdom-0.020
ement-0.020
Token they
Token ID484
Feature activation-8.71e-03

Pos logit contributions
sc+0.032
B+0.032
ch+0.032
 subs+0.030
ched+0.029
Neg logit contributions
azes-0.036
pit-0.030
 writing-0.030
 oven-0.030
mmm-0.030
Token try
Token ID1949
Feature activation-7.08e-03

Pos logit contributions
azes+0.019
ement+0.017
mmm+0.017
isdom+0.016
ummy+0.016
Neg logit contributions
sc-0.010
ch-0.010
 scrap-0.010
bo-0.010
 subs-0.009
Token to
Token ID284
Feature activation-5.89e-03

Pos logit contributions
azes+0.020
mmm+0.019
ement+0.018
eria+0.018
isdom+0.018
Neg logit contributions
s-0.003
l-0.003
 off-0.003
fl-0.003
ch-0.003
Token find
Token ID1064
Feature activation-3.76e-02

Pos logit contributions
azes+0.014
ement+0.013
ires+0.012
isdom+0.012
ummy+0.012
Neg logit contributions
B-0.006
 hurt-0.006
bo-0.006
ch-0.006
ny-0.005
Token each
Token ID1123
Feature activation-7.72e-02

Pos logit contributions
azes+0.101
pit+0.089
eria+0.085
castle+0.085
mmm+0.084
Neg logit contributions
ch-0.030
 away-0.027
ched-0.025
B-0.025
 off-0.024
Token other
Token ID584
Feature activation+1.09e-02

Pos logit contributions
azes+0.192
mmm+0.163
ement+0.161
isdom+0.161
 balcony+0.159
Neg logit contributions
sc-0.072
B-0.072
ch-0.070
ny-0.066
 subs-0.061
Token's
Token ID338
Feature activation+1.91e-03

Pos logit contributions
sc+0.011
B+0.010
ch+0.010
ny+0.010
bo+0.009
Neg logit contributions
azes-0.026
mmm-0.023
isdom-0.023
ires-0.022
ement-0.022
Token fruit
Token ID8234
Feature activation-1.81e-02

Pos logit contributions
fl+0.003
 scrap+0.003
 subs+0.002
ch+0.002
 hurts+0.002
Neg logit contributions
azes-0.004
mmm-0.003
otion-0.003
ifting-0.003
ement-0.003
Token.
Token ID13
Feature activation+1.07e-02

Pos logit contributions
azes+0.045
isdom+0.039
mmm+0.038
ement+0.037
ires+0.037
Neg logit contributions
sc-0.018
ch-0.016
B-0.016
ny-0.016
 subs-0.015
Token They
Token ID1119
Feature activation-4.04e-03

Pos logit contributions
sc+0.013
ch+0.012
B+0.012
Fl+0.011
text+0.011
Neg logit contributions
azes-0.023
castle-0.021
pit-0.021
isdom-0.021
ement-0.021
Token for
Token ID329
Feature activation-3.29e-02

Pos logit contributions
azes+0.025
mmm+0.020
pit+0.020
isdom+0.020
ocolate+0.019
Neg logit contributions
sc-0.024
ch-0.023
B-0.023
 subs-0.022
ched-0.021
Token treasures
Token ID29561
Feature activation+3.20e-03

Pos logit contributions
azes+0.069
castle+0.060
pit+0.060
mmm+0.058
ires+0.057
Neg logit contributions
B-0.043
ch-0.040
sc-0.039
 subs-0.037
fl-0.036
Token.
Token ID13
Feature activation+1.12e-02

Pos logit contributions
sc+0.004
B+0.003
ch+0.003
 subs+0.003
ched+0.003
Neg logit contributions
azes-0.007
mmm-0.006
isdom-0.006
ires-0.006
ocolate-0.006
Token
Token ID198
Feature activation+1.75e-02

Pos logit contributions
ch+0.010
sc+0.010
B+0.010
Spot+0.008
text+0.008
Neg logit contributions
azes-0.027
castle-0.026
pit-0.025
ement-0.024
 patiently-0.024
Token
Token ID198
Feature activation+1.17e-02

Pos logit contributions
B+0.016
ch+0.015
sc+0.015
ched+0.013
bo+0.012
Neg logit contributions
azes-0.043
pit-0.040
castle-0.039
 fridge-0.038
 oven-0.038
TokenWhen
Token ID2215
Feature activation+5.90e-02

Pos logit contributions
sc+0.015
ch+0.015
B+0.014
 subs+0.014
bo+0.013
Neg logit contributions
azes-0.023
pit-0.020
mmm-0.020
isdom-0.020
ires-0.020
Token it
Token ID340
Feature activation+2.24e-02

Pos logit contributions
sc+0.084
B+0.082
ch+0.081
 subs+0.078
bo+0.077
Neg logit contributions
azes-0.111
 oven-0.095
mmm-0.095
pit-0.093
isdom-0.093
Token was
Token ID373
Feature activation+1.22e-02

Pos logit contributions
sc+0.028
burn+0.028
B+0.027
Br+0.027
 scrap+0.027
Neg logit contributions
azes-0.042
pit-0.039
ement-0.037
castle-0.036
 patiently-0.036
Token time
Token ID640
Feature activation+4.47e-02

Pos logit contributions
ch+0.010
B+0.010
sc+0.010
 subs+0.009
ched+0.008
Neg logit contributions
azes-0.031
pit-0.027
isdom-0.027
ement-0.027
ires-0.027
Token to
Token ID284
Feature activation+7.93e-03

Pos logit contributions
ch+0.047
sc+0.046
B+0.044
 subs+0.044
ched+0.041
Neg logit contributions
azes-0.102
pit-0.089
isdom-0.088
mmm-0.088
ires-0.088
Token go
Token ID467
Feature activation+3.11e-03

Pos logit contributions
sc+0.012
B+0.012
ch+0.011
ched+0.011
 subs+0.010
Neg logit contributions
azes-0.015
 fridge-0.013
castle-0.013
mmm-0.013
 oven-0.013
Token broken
Token ID5445
Feature activation+2.29e-02

Pos logit contributions
azes+0.017
 oven+0.014
 writing+0.014
 fridge+0.014
pit+0.014
Neg logit contributions
sc-0.016
B-0.016
ch-0.015
 subs-0.015
text-0.015
Token and
Token ID290
Feature activation+4.62e-02

Pos logit contributions
ch+0.031
sc+0.031
B+0.031
 subs+0.029
bo+0.027
Neg logit contributions
azes-0.046
pit-0.039
isdom-0.038
mmm-0.038
ires-0.038
Token it
Token ID340
Feature activation+3.89e-02

Pos logit contributions
sc+0.070
B+0.069
ch+0.068
 subs+0.066
ched+0.063
Neg logit contributions
azes-0.082
 fridge-0.071
 oven-0.070
pit-0.068
 writing-0.068
Token didn
Token ID1422
Feature activation+7.10e-02

Pos logit contributions
Br+0.054
B+0.052
Ar+0.051
sc+0.051
ch+0.050
Neg logit contributions
azes-0.067
 fridge-0.067
 oven-0.064
 desk-0.062
 writing-0.060
Token't
Token ID470
Feature activation+4.78e-02

Pos logit contributions
sc+0.136
fl+0.136
B+0.131
Br+0.128
 subs+0.126
Neg logit contributions
azes-0.095
mmm-0.076
ummy-0.075
ires-0.074
 oven-0.073
Token work
Token ID670
Feature activation+7.33e-02

Pos logit contributions
sc+0.067
B+0.067
ch+0.066
 subs+0.064
ched+0.060
Neg logit contributions
azes-0.090
isdom-0.075
 fridge-0.075
pit-0.075
mmm-0.074
Token anymore
Token ID7471
Feature activation+2.64e-02

Pos logit contributions
B+0.123
Br+0.122
ch+0.122
sc+0.122
fl+0.119
Neg logit contributions
azes-0.109
 writing-0.095
 patiently-0.093
ement-0.090
 fridge-0.088
Token.
Token ID13
Feature activation+1.58e-02

Pos logit contributions
B+0.027
ch+0.027
sc+0.026
 subs+0.025
ched+0.024
Neg logit contributions
azes-0.060
pit-0.054
 oven-0.053
ires-0.053
ement-0.053
Token Lily
Token ID20037
Feature activation+3.11e-02

Pos logit contributions
ch+0.023
sc+0.023
 subs+0.023
B+0.022
ched+0.021
Neg logit contributions
azes-0.029
 fridge-0.025
pit-0.024
ement-0.024
mmm-0.024
Token was
Token ID373
Feature activation-2.56e-03

Pos logit contributions
ch+0.032
sc+0.032
B+0.031
 subs+0.030
ched+0.027
Neg logit contributions
azes-0.071
pit-0.062
 fridge-0.061
isdom-0.061
 oven-0.061
Token sad
Token ID6507
Feature activation+7.43e-03

Pos logit contributions
azes+0.005
pit+0.004
mmm+0.004
isdom+0.004
ires+0.004
Neg logit contributions
ch-0.003
B-0.003
 subs-0.003
sc-0.003
 scrap-0.003
Token history
Token ID2106
Feature activation+8.76e-03

Pos logit contributions
azes+0.011
mmm+0.009
ement+0.009
pit+0.009
isdom+0.009
Neg logit contributions
ch-0.007
B-0.007
fl-0.007
 subs-0.007
sc-0.007
Token.
Token ID13
Feature activation+6.78e-03

Pos logit contributions
sc+0.007
ch+0.007
B+0.006
l+0.006
 subs+0.006
Neg logit contributions
azes-0.021
sing-0.019
mmm-0.019
pit-0.019
ocolate-0.019
Token Every
Token ID3887
Feature activation+4.11e-02

Pos logit contributions
B+0.008
ch+0.008
sc+0.007
Ch+0.006
Fl+0.006
Neg logit contributions
castle-0.015
pit-0.015
azes-0.014
 patiently-0.014
 Majesty-0.014
Token day
Token ID1110
Feature activation+4.91e-02

Pos logit contributions
ch+0.032
sc+0.027
B+0.026
 subs+0.025
ny+0.023
Neg logit contributions
azes-0.111
mmm-0.099
pit-0.099
ires-0.099
ement-0.097
Token and
Token ID290
Feature activation+2.08e-02

Pos logit contributions
sc+0.087
B+0.087
ch+0.087
 subs+0.084
bo+0.081
Neg logit contributions
azes-0.076
pit-0.060
mmm-0.060
isdom-0.059
ires-0.059
Token night
Token ID1755
Feature activation+5.98e-02

Pos logit contributions
ch+0.023
sc+0.022
B+0.022
 subs+0.020
 scrap+0.020
Neg logit contributions
azes-0.046
pit-0.041
ement-0.041
castle-0.038
mmm-0.038
Token,
Token ID11
Feature activation-4.34e-03

Pos logit contributions
sc+0.111
B+0.111
 subs+0.108
ch+0.108
bo+0.105
Neg logit contributions
azes-0.087
 writing-0.069
 oven-0.067
mmm-0.067
 fridge-0.067
Token a
Token ID257
Feature activation+3.03e-02

Pos logit contributions
azes+0.010
pit+0.010
ement+0.009
castle+0.009
isdom+0.009
Neg logit contributions
ch-0.004
B-0.004
sc-0.004
 subs-0.003
Free-0.003
Token spray
Token ID11662
Feature activation+2.00e-02

Pos logit contributions
ch+0.029
B+0.028
sc+0.026
 subs+0.026
 scrap+0.023
Neg logit contributions
azes-0.075
pit-0.065
ires-0.065
isdom-0.064
mmm-0.064
Token of
Token ID286
Feature activation+2.42e-02

Pos logit contributions
ch+0.026
sc+0.026
B+0.025
 subs+0.024
ched+0.023
Neg logit contributions
azes-0.042
mmm-0.036
pit-0.035
isdom-0.035
ires-0.035
Token water
Token ID1660
Feature activation-2.62e-03

Pos logit contributions
ch+0.026
B+0.023
 scrap+0.021
sc+0.020
 st+0.020
Neg logit contributions
azes-0.057
pit-0.054
ae-0.051
mmm-0.050
ement-0.050
Token their
Token ID511
Feature activation+2.17e-02

Pos logit contributions
sc+0.011
ch+0.011
B+0.011
 subs+0.011
bo+0.010
Neg logit contributions
azes-0.011
 oven-0.010
isdom-0.009
ires-0.009
mmm-0.009
Token medicine
Token ID9007
Feature activation+3.69e-02

Pos logit contributions
ch+0.034
B+0.034
sc+0.033
Br+0.033
ched+0.033
Neg logit contributions
azes-0.033
 fridge-0.033
 oven-0.032
 refrigerator-0.032
 pasta-0.031
Token and
Token ID290
Feature activation+1.34e-02

Pos logit contributions
sc+0.045
ch+0.043
B+0.042
 subs+0.040
ched+0.039
Neg logit contributions
azes-0.083
mmm-0.070
isdom-0.069
pit-0.069
ires-0.069
Token stuck
Token ID7819
Feature activation+3.19e-02

Pos logit contributions
ch+0.017
sc+0.017
B+0.017
 subs+0.016
bo+0.015
Neg logit contributions
azes-0.027
pit-0.023
 oven-0.023
isdom-0.023
 fridge-0.023
Token a
Token ID257
Feature activation+2.48e-02

Pos logit contributions
ch+0.043
sc+0.043
B+0.043
 subs+0.041
ched+0.038
Neg logit contributions
azes-0.063
pit-0.053
isdom-0.052
mmm-0.052
ires-0.052
Token therm
Token ID21969
Feature activation+6.28e-02

Pos logit contributions
ch+0.035
sc+0.035
B+0.035
 subs+0.033
bo+0.031
Neg logit contributions
azes-0.048
pit-0.040
isdom-0.040
ires-0.039
mmm-0.039
Tokenometer
Token ID15635
Feature activation+2.79e-02

Pos logit contributions
ch+0.100
sc+0.098
B+0.098
 subs+0.095
bo+0.092
Neg logit contributions
azes-0.111
pit-0.092
isdom-0.089
mmm-0.089
ement-0.089
Token in
Token ID287
Feature activation+1.44e-02

Pos logit contributions
B+0.036
sc+0.035
ch+0.035
 subs+0.034
text+0.032
Neg logit contributions
azes-0.055
pit-0.048
 oven-0.048
 fridge-0.047
castle-0.046
Token their
Token ID511
Feature activation-1.25e-03

Pos logit contributions
ch+0.016
sc+0.016
B+0.016
 subs+0.015
ched+0.014
Neg logit contributions
azes-0.031
pit-0.027
 oven-0.027
 fridge-0.027
mmm-0.027
Token mouth
Token ID5422
Feature activation+1.39e-02

Pos logit contributions
azes+0.002
pit+0.002
mmm+0.002
isdom+0.002
ires+0.002
Neg logit contributions
ch-0.002
sc-0.002
B-0.002
 subs-0.002
bo-0.002
Token.
Token ID13
Feature activation+4.98e-03

Pos logit contributions
ch+0.016
sc+0.016
B+0.016
 subs+0.015
ched+0.015
Neg logit contributions
azes-0.030
isdom-0.026
pit-0.026
mmm-0.026
ires-0.026
Token have
Token ID423
Feature activation+7.71e-03

Pos logit contributions
ch+0.028
sc+0.028
B+0.027
 subs+0.026
ched+0.025
Neg logit contributions
azes-0.051
pit-0.044
ement-0.044
ires-0.043
mmm-0.043
Token to
Token ID284
Feature activation+1.75e-02

Pos logit contributions
ch+0.009
B+0.008
 subs+0.008
sc+0.008
 scrap+0.008
Neg logit contributions
azes-0.017
ement-0.015
pit-0.015
mmm-0.015
ires-0.015
Token pay
Token ID1414
Feature activation+4.76e-03

Pos logit contributions
ch+0.022
B+0.021
bo+0.021
sc+0.021
 subs+0.020
Neg logit contributions
azes-0.036
ement-0.032
pit-0.031
ires-0.031
eria-0.029
Token one
Token ID530
Feature activation-2.05e-02

Pos logit contributions
ch+0.007
sc+0.006
B+0.006
bo+0.006
ched+0.006
Neg logit contributions
azes-0.010
pit-0.009
castle-0.008
ement-0.008
 balcony-0.008
Token dollar
Token ID8872
Feature activation-9.73e-03

Pos logit contributions
azes+0.044
pit+0.036
isdom+0.036
ement+0.035
 balcony+0.035
Neg logit contributions
ch-0.027
sc-0.027
B-0.025
bo-0.023
 subs-0.023
Token per
Token ID583
Feature activation+6.02e-02

Pos logit contributions
azes+0.023
pit+0.019
 balcony+0.019
ement+0.019
isdom+0.019
Neg logit contributions
sc-0.011
ch-0.011
B-0.010
bo-0.010
fl-0.009
Token cup
Token ID6508
Feature activation+8.54e-03

Pos logit contributions
B+0.077
 subs+0.073
 scrap+0.071
text+0.070
Br+0.068
Neg logit contributions
azes-0.112
ires-0.104
 oven-0.103
isdom-0.102
mmm-0.101
Token.
Token ID13
Feature activation+3.08e-03

Pos logit contributions
sc+0.010
ch+0.009
B+0.008
bo+0.008
ched+0.008
Neg logit contributions
azes-0.020
isdom-0.017
pit-0.017
 balcony-0.017
 Majesty-0.017
Token Or
Token ID1471
Feature activation+9.26e-03

Pos logit contributions
ch+0.004
sc+0.004
B+0.004
bo+0.004
Fl+0.004
Neg logit contributions
azes-0.006
pit-0.005
castle-0.005
 balcony-0.005
ement-0.005
Token you
Token ID345
Feature activation+2.32e-02

Pos logit contributions
ch+0.014
sc+0.013
B+0.013
 subs+0.012
bo+0.012
Neg logit contributions
azes-0.017
pit-0.015
ement-0.015
castle-0.014
 balcony-0.014
Token can
Token ID460
Feature activation+1.24e-02

Pos logit contributions
ch+0.023
sc+0.023
B+0.022
 subs+0.020
ched+0.020
Neg logit contributions
azes-0.056
pit-0.049
ement-0.048
ires-0.048
mmm-0.047
Token,
Token ID11
Feature activation-1.51e-02

Pos logit contributions
azes+0.005
pit+0.005
 workshop+0.005
ires+0.004
sing+0.004
Neg logit contributions
ch-0.003
B-0.003
ched-0.003
sc-0.003
l-0.003
Token "
Token ID366
Feature activation-4.81e-03

Pos logit contributions
azes+0.034
pit+0.030
castle+0.029
ement+0.029
isdom+0.029
Neg logit contributions
ch-0.018
B-0.017
sc-0.016
Fl-0.015
fl-0.015
TokenL
Token ID43
Feature activation+4.99e-03

Pos logit contributions
azes+0.009
mmm+0.009
ocolate+0.008
isdom+0.008
ae+0.008
Neg logit contributions
 subs-0.006
ch-0.006
sc-0.006
text-0.005
 scrap-0.005
Tokenily
Token ID813
Feature activation+1.60e-02

Pos logit contributions
ch+0.010
B+0.010
sc+0.010
 subs+0.009
ched+0.009
Neg logit contributions
azes-0.007
pit-0.006
 balcony-0.005
 fridge-0.005
castle-0.005
Token,
Token ID11
Feature activation-1.99e-02

Pos logit contributions
ch+0.023
B+0.022
sc+0.022
 subs+0.022
ched+0.021
Neg logit contributions
azes-0.031
ires-0.026
ement-0.025
pit-0.025
 balcony-0.025
Token you
Token ID345
Feature activation+1.70e-02

Pos logit contributions
azes+0.045
ement+0.040
pit+0.039
ires+0.037
castle+0.037
Neg logit contributions
ch-0.025
Fl-0.021
sc-0.021
B-0.021
 subs-0.021
Token have
Token ID423
Feature activation+5.64e-03

Pos logit contributions
ch+0.017
B+0.016
sc+0.016
ched+0.015
 subs+0.015
Neg logit contributions
azes-0.041
pit-0.036
ement-0.036
ires-0.036
castle-0.035
Token to
Token ID284
Feature activation+1.25e-02

Pos logit contributions
ch+0.006
B+0.005
sc+0.005
 subs+0.005
 scrap+0.005
Neg logit contributions
azes-0.014
pit-0.012
ement-0.012
ires-0.012
mmm-0.011
Token eat
Token ID4483
Feature activation-2.10e-02

Pos logit contributions
ch+0.015
bo+0.014
B+0.014
sc+0.014
ny+0.014
Neg logit contributions
azes-0.027
ement-0.023
pit-0.023
ires-0.023
eria-0.022
Token your
Token ID534
Feature activation-2.01e-03

Pos logit contributions
azes+0.048
pit+0.041
sing+0.039
castle+0.039
 balcony+0.038
Neg logit contributions
ch-0.026
sc-0.026
B-0.025
Fl-0.023
 away-0.023
Token broccoli
Token ID44653
Feature activation-1.04e-03

Pos logit contributions
azes+0.003
pit+0.003
otion+0.003
eria+0.002
mmm+0.002
Neg logit contributions
 scrap-0.003
ch-0.003
sc-0.003
fl-0.003
B-0.003
Token it
Token ID340
Feature activation-1.53e-02

Pos logit contributions
azes+0.005
mmm+0.004
isdom+0.004
pit+0.004
ires+0.004
Neg logit contributions
B-0.001
ch-0.001
sc-0.001
 subs-0.001
 scrap-0.001
Token and
Token ID290
Feature activation+1.03e-02

Pos logit contributions
azes+0.037
mmm+0.034
isdom+0.033
ocolate+0.033
mmmm+0.032
Neg logit contributions
sc-0.016
B-0.014
ch-0.014
ched-0.013
s-0.013
Token saw
Token ID2497
Feature activation-2.48e-02

Pos logit contributions
ch+0.013
B+0.013
sc+0.013
 subs+0.012
bo+0.011
Neg logit contributions
azes-0.021
pit-0.018
isdom-0.018
 fridge-0.018
 oven-0.018
Token the
Token ID262
Feature activation+8.62e-04

Pos logit contributions
azes+0.053
mmm+0.047
pit+0.045
isdom+0.045
ocolate+0.045
Neg logit contributions
B-0.030
ch-0.030
sc-0.029
ched-0.028
 scrap-0.027
Token pepper
Token ID13385
Feature activation+1.11e-02

Pos logit contributions
ch+0.001
sc+0.001
B+0.001
 subs+0.001
 scrap+0.001
Neg logit contributions
azes-0.002
mmm-0.001
isdom-0.001
ement-0.001
ires-0.001
Token resting
Token ID19186
Feature activation+9.68e-03

Pos logit contributions
ch+0.013
sc+0.012
 subs+0.012
ched+0.012
B+0.011
Neg logit contributions
azes-0.026
isdom-0.022
mmm-0.021
ires-0.021
ocolate-0.021
Token in
Token ID287
Feature activation+3.60e-03

Pos logit contributions
sc+0.010
ch+0.010
B+0.010
 subs+0.009
ched+0.008
Neg logit contributions
azes-0.023
isdom-0.020
mmm-0.020
pit-0.019
ires-0.019
Token a
Token ID257
Feature activation+7.43e-03

Pos logit contributions
B+0.003
ch+0.003
sc+0.003
 subs+0.002
 scrap+0.002
Neg logit contributions
azes-0.010
isdom-0.008
ires-0.008
pit-0.008
mmm-0.008
Token bush
Token ID24484
Feature activation-2.48e-02

Pos logit contributions
ch+0.012
sc+0.011
B+0.011
 subs+0.010
 scrap+0.010
Neg logit contributions
azes-0.015
ires-0.012
isdom-0.012
mmm-0.012
pit-0.011
Token.
Token ID13
Feature activation+2.58e-04

Pos logit contributions
azes+0.063
isdom+0.054
ires+0.054
mmm+0.053
 Majesty+0.053
Neg logit contributions
sc-0.023
ch-0.022
B-0.020
ny-0.019
ched-0.019
Token He
Token ID679
Feature activation-9.53e-03

Pos logit contributions
B+0.000
sc+0.000
ch+0.000
Fl+0.000
Spot+0.000
Neg logit contributions
azes-0.001
castle-0.000
pit-0.000
ement-0.000
isdom-0.000
Token said
Token ID531
Feature activation-4.84e-04

Pos logit contributions
B+0.015
ch+0.015
sc+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.041
pit-0.036
ires-0.036
mmm-0.035
ement-0.035
Token sorry
Token ID7926
Feature activation+2.49e-03

Pos logit contributions
azes+0.001
pit+0.001
ires+0.001
ement+0.001
isdom+0.001
Neg logit contributions
ch-0.001
B-0.001
 subs-0.001
sc-0.001
ched-0.001
Token and
Token ID290
Feature activation+1.48e-02

Pos logit contributions
ch+0.003
B+0.003
sc+0.003
 subs+0.003
bo+0.003
Neg logit contributions
azes-0.005
pit-0.004
ement-0.004
mmm-0.004
isdom-0.004
Token helped
Token ID4193
Feature activation+2.38e-02

Pos logit contributions
sc+0.018
ch+0.018
B+0.017
 subs+0.016
text+0.016
Neg logit contributions
azes-0.030
 patiently-0.027
 oven-0.027
 fridge-0.027
 writing-0.026
Token Lily
Token ID20037
Feature activation+1.16e-02

Pos logit contributions
sc+0.041
ch+0.040
 subs+0.040
B+0.039
fl+0.038
Neg logit contributions
azes-0.037
 fridge-0.031
 oven-0.030
 patiently-0.030
 writing-0.030
Token print
Token ID3601
Feature activation+1.38e-02

Pos logit contributions
ch+0.016
sc+0.016
B+0.016
 subs+0.015
ched+0.014
Neg logit contributions
azes-0.023
pit-0.020
mmm-0.019
ement-0.019
ires-0.019
Token her
Token ID607
Feature activation+2.01e-02

Pos logit contributions
ch+0.016
sc+0.016
B+0.015
ched+0.015
 subs+0.015
Neg logit contributions
azes-0.031
mmm-0.027
isdom-0.026
pit-0.026
 balcony-0.026
Token drawings
Token ID23388
Feature activation+2.46e-02

Pos logit contributions
ch+0.039
B+0.039
sc+0.038
 subs+0.037
ched+0.036
Neg logit contributions
azes-0.029
pit-0.023
isdom-0.022
 oven-0.022
 fridge-0.022
Token.
Token ID13
Feature activation+9.95e-03

Pos logit contributions
sc+0.023
 subs+0.021
ched+0.021
ch+0.021
B+0.021
Neg logit contributions
azes-0.062
isdom-0.054
mmm-0.054
ires-0.053
 Majesty-0.052
Token
Token ID220
Feature activation+1.31e-02

Pos logit contributions
sc+0.015
ch+0.015
B+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.019
pit-0.016
 fridge-0.015
ires-0.015
mmm-0.015
Token
Token ID198
Feature activation+1.03e-02

Pos logit contributions
B+0.013
sc+0.013
ch+0.013
 subs+0.012
ched+0.011
Neg logit contributions
azes-0.030
pit-0.027
castle-0.026
 oven-0.026
isdom-0.026
Token a
Token ID257
Feature activation+3.39e-02

Pos logit contributions
sc+0.047
ch+0.046
 subs+0.046
B+0.044
ched+0.042
Neg logit contributions
azes-0.052
 writing-0.042
 fridge-0.042
mmm-0.042
pit-0.042
Token time
Token ID640
Feature activation-9.89e-05

Pos logit contributions
sc+0.032
ch+0.032
B+0.032
 subs+0.031
ched+0.028
Neg logit contributions
azes-0.079
pit-0.069
mmm-0.068
isdom-0.068
 oven-0.068
Token,
Token ID11
Feature activation-3.03e-02

Pos logit contributions
azes+0.000
 oven+0.000
isdom+0.000
ires+0.000
pit+0.000
Neg logit contributions
B-0.000
ch-0.000
sc-0.000
 subs-0.000
text-0.000
Token there
Token ID612
Feature activation+3.44e-02

Pos logit contributions
azes+0.058
pit+0.049
isdom+0.048
mmm+0.048
ires+0.048
Neg logit contributions
ch-0.042
sc-0.042
B-0.042
 subs-0.040
ched-0.038
Token was
Token ID373
Feature activation-8.42e-03

Pos logit contributions
ch+0.026
sc+0.026
B+0.025
 subs+0.024
bo+0.021
Neg logit contributions
azes-0.088
pit-0.078
isdom-0.077
mmm-0.077
ires-0.077
Token a
Token ID257
Feature activation+2.51e-02

Pos logit contributions
azes+0.017
pit+0.015
isdom+0.014
ement+0.014
ires+0.014
Neg logit contributions
B-0.011
ch-0.011
sc-0.011
 subs-0.010
bo-0.010
Token mighty
Token ID18680
Feature activation+3.48e-03

Pos logit contributions
B+0.025
ch+0.025
sc+0.022
 subs+0.021
bo+0.020
Neg logit contributions
azes-0.061
pit-0.053
ement-0.052
isdom-0.052
mmm-0.052
Token land
Token ID1956
Feature activation+1.14e-02

Pos logit contributions
ch+0.004
B+0.004
ny+0.003
 k+0.003
 st+0.003
Neg logit contributions
azes-0.008
ires-0.007
mmm-0.007
ement-0.007
ocolate-0.007
Token.
Token ID13
Feature activation+1.13e-02

Pos logit contributions
B+0.016
ch+0.016
sc+0.016
 subs+0.016
text+0.015
Neg logit contributions
azes-0.021
pit-0.018
 oven-0.018
ires-0.018
castle-0.018
Token In
Token ID554
Feature activation+1.70e-02

Pos logit contributions
ch+0.015
B+0.014
sc+0.013
Free+0.012
 subs+0.012
Neg logit contributions
azes-0.023
pit-0.022
castle-0.022
ement-0.021
 patiently-0.021
Token this
Token ID428
Feature activation-2.13e-02

Pos logit contributions
ch+0.014
B+0.013
 subs+0.013
sc+0.012
text+0.012
Neg logit contributions
azes-0.042
pit-0.037
ement-0.037
isdom-0.036
ruits-0.036
Token
Token ID198
Feature activation-4.43e-03

Pos logit contributions
ch+0.001
B+0.001
sc+0.001
 subs+0.001
bo+0.001
Neg logit contributions
azes-0.002
pit-0.002
isdom-0.002
ires-0.002
 oven-0.002
Token
Token ID198
Feature activation-1.08e-02

Pos logit contributions
azes+0.010
pit+0.009
isdom+0.008
 oven+0.008
ires+0.008
Neg logit contributions
B-0.005
ch-0.005
sc-0.005
 subs-0.004
ched-0.004
TokenThe
Token ID464
Feature activation+1.65e-02

Pos logit contributions
azes+0.023
mmm+0.020
pit+0.020
ocolate+0.020
ae+0.020
Neg logit contributions
sc-0.012
 subs-0.012
ch-0.012
fl-0.011
B-0.011
Token silly
Token ID14397
Feature activation+6.18e-03

Pos logit contributions
B+0.019
ch+0.018
sc+0.018
 subs+0.018
 scrap+0.017
Neg logit contributions
azes-0.038
mmm-0.032
isdom-0.031
ues-0.030
castle-0.030
Token refrigerator
Token ID30500
Feature activation+6.40e-03

Pos logit contributions
ny+0.009
fl+0.008
B+0.008
 subs+0.008
 went+0.008
Neg logit contributions
azes-0.012
mmm-0.010
ues-0.010
ifting-0.009
ents-0.009
Token and
Token ID290
Feature activation+2.38e-02

Pos logit contributions
 subs+0.004
sc+0.004
ch+0.004
bo+0.004
B+0.004
Neg logit contributions
azes-0.018
mmm-0.016
ires-0.016
isdom-0.015
ruits-0.015
Token the
Token ID262
Feature activation+2.51e-02

Pos logit contributions
sc+0.030
ch+0.030
B+0.030
 subs+0.028
ched+0.027
Neg logit contributions
azes-0.049
pit-0.043
 oven-0.042
 writing-0.042
 fridge-0.041
Token little
Token ID1310
Feature activation+2.68e-02

Pos logit contributions
B+0.023
sc+0.021
 subs+0.021
 scrap+0.021
ch+0.020
Neg logit contributions
azes-0.064
mmm-0.056
isdom-0.055
ues-0.054
ires-0.052
Token girl
Token ID2576
Feature activation-3.58e-04

Pos logit contributions
B+0.021
sc+0.020
ch+0.020
 subs+0.019
fl+0.017
Neg logit contributions
azes-0.071
mmm-0.062
isdom-0.062
ires-0.061
castle-0.060
Token became
Token ID2627
Feature activation-8.36e-03

Pos logit contributions
azes+0.001
ires+0.001
mmm+0.001
pit+0.001
isdom+0.001
Neg logit contributions
ch-0.000
sc-0.000
 subs-0.000
B-0.000
bo-0.000
Token the
Token ID262
Feature activation+1.30e-02

Pos logit contributions
azes+0.014
pit+0.012
 writing+0.012
 oven+0.012
 patiently+0.012
Neg logit contributions
ch-0.013
sc-0.013
B-0.013
 subs-0.012
bo-0.012
Token with
Token ID351
Feature activation-1.88e-02

Pos logit contributions
azes+0.007
mmm+0.006
ummy+0.006
isdom+0.006
ement+0.006
Neg logit contributions
sc-0.003
B-0.003
ch-0.003
ched-0.002
l-0.002
Token their
Token ID511
Feature activation+7.87e-03

Pos logit contributions
azes+0.044
mmm+0.039
pit+0.039
isdom+0.038
ement+0.038
Neg logit contributions
B-0.019
ch-0.018
sc-0.018
 scrap-0.016
 subs-0.016
Token new
Token ID649
Feature activation+5.61e-03

Pos logit contributions
ch+0.012
sc+0.012
B+0.012
 subs+0.011
fl+0.011
Neg logit contributions
azes-0.014
mmm-0.012
isdom-0.011
ement-0.011
pit-0.011
Token things
Token ID1243
Feature activation+1.33e-02

Pos logit contributions
ch+0.007
sc+0.007
fl+0.007
B+0.006
 scrap+0.006
Neg logit contributions
azes-0.012
mmm-0.011
ifting-0.010
ement-0.010
otion-0.010
Token.
Token ID13
Feature activation+1.99e-03

Pos logit contributions
sc+0.010
B+0.009
ch+0.009
s+0.008
ny+0.008
Neg logit contributions
azes-0.036
isdom-0.033
mmm-0.032
 Majesty-0.031
ocolate-0.031
Token They
Token ID1119
Feature activation-4.79e-04

Pos logit contributions
ch+0.002
B+0.002
sc+0.002
 subs+0.002
Br+0.002
Neg logit contributions
azes-0.004
pit-0.004
ement-0.004
castle-0.004
 oven-0.004
Token are
Token ID389
Feature activation-3.83e-02

Pos logit contributions
azes+0.001
ement+0.001
mmm+0.001
ires+0.001
isdom+0.001
Neg logit contributions
sc-0.000
ch-0.000
B-0.000
 scrap-0.000
 subs-0.000
Token very
Token ID845
Feature activation-3.41e-03

Pos logit contributions
azes+0.071
ement+0.062
mmm+0.062
isdom+0.061
pit+0.058
Neg logit contributions
ch-0.055
B-0.053
 subs-0.053
sc-0.052
 scrap-0.050
Token happy
Token ID3772
Feature activation-1.64e-02

Pos logit contributions
azes+0.006
pit+0.005
isdom+0.005
ement+0.005
mmm+0.005
Neg logit contributions
ch-0.005
B-0.005
sc-0.005
 subs-0.005
bo-0.005
Token.
Token ID13
Feature activation-3.60e-03

Pos logit contributions
azes+0.039
mmm+0.037
isdom+0.036
ement+0.036
ocolate+0.036
Neg logit contributions
B-0.015
fl-0.014
ch-0.013
s-0.012
sc-0.012
Token They
Token ID1119
Feature activation-8.79e-03

Pos logit contributions
azes+0.007
ires+0.006
pit+0.006
mmm+0.006
isdom+0.006
Neg logit contributions
sc-0.005
ch-0.005
B-0.004
 subs-0.004
ched-0.004
Token happy
Token ID3772
Feature activation-1.16e-02

Pos logit contributions
B+0.001
ch+0.001
sc+0.001
 subs+0.001
ched+0.001
Neg logit contributions
azes-0.001
pit-0.001
isdom-0.001
 oven-0.001
 fridge-0.001
Token and
Token ID290
Feature activation+2.00e-02

Pos logit contributions
azes+0.024
ement+0.021
ocolate+0.021
ruits+0.021
pit+0.021
Neg logit contributions
B-0.014
ch-0.014
 subs-0.014
fl-0.013
sc-0.013
Token he
Token ID339
Feature activation-4.04e-03

Pos logit contributions
sc+0.034
ch+0.034
B+0.033
text+0.030
bo+0.030
Neg logit contributions
azes-0.034
 writing-0.030
 patiently-0.028
 fridge-0.027
 balcony-0.027
Token thanked
Token ID26280
Feature activation-4.44e-03

Pos logit contributions
azes+0.008
 patiently+0.007
 writing+0.007
castle+0.006
isdom+0.006
Neg logit contributions
sc-0.006
ch-0.005
B-0.005
fl-0.005
 subs-0.005
Token his
Token ID465
Feature activation+6.13e-03

Pos logit contributions
azes+0.010
pit+0.009
ires+0.008
ement+0.008
mmm+0.008
Neg logit contributions
B-0.005
ch-0.005
sc-0.005
 subs-0.005
bo-0.005
Token mom
Token ID1995
Feature activation-2.02e-02

Pos logit contributions
B+0.009
 subs+0.009
ch+0.009
 scrap+0.009
sc+0.008
Neg logit contributions
azes-0.011
mmm-0.009
pit-0.009
isdom-0.009
ocolate-0.009
Tokenmy
Token ID1820
Feature activation-9.69e-03

Pos logit contributions
azes+0.039
pit+0.032
mmm+0.032
ires+0.032
isdom+0.032
Neg logit contributions
ch-0.028
sc-0.028
B-0.028
 subs-0.026
bo-0.026
Token for
Token ID329
Feature activation-1.50e-02

Pos logit contributions
azes+0.017
pit+0.014
mmm+0.014
ires+0.014
isdom+0.014
Neg logit contributions
sc-0.015
ch-0.015
B-0.015
 subs-0.014
bo-0.014
Token getting
Token ID1972
Feature activation+8.19e-03

Pos logit contributions
azes+0.025
pit+0.021
ires+0.020
mmm+0.020
ement+0.020
Neg logit contributions
B-0.025
ch-0.025
sc-0.025
 subs-0.024
ched-0.023
Token him
Token ID683
Feature activation-5.93e-03

Pos logit contributions
sc+0.012
ch+0.012
B+0.012
 subs+0.012
text+0.011
Neg logit contributions
azes-0.015
 writing-0.012
 patiently-0.012
 oven-0.012
 fridge-0.012
Token a
Token ID257
Feature activation+1.62e-02

Pos logit contributions
azes+0.010
 writing+0.009
 fridge+0.009
 patiently+0.009
isdom+0.009
Neg logit contributions
sc-0.009
B-0.009
ch-0.009
 subs-0.008
Br-0.008
Token bear
Token ID6842
Feature activation+4.77e-03

Pos logit contributions
sc+0.007
ch+0.007
B+0.007
 subs+0.007
ched+0.006
Neg logit contributions
azes-0.013
pit-0.011
 oven-0.011
 fridge-0.011
ement-0.011
Token looked
Token ID3114
Feature activation-2.77e-02

Pos logit contributions
B+0.005
ch+0.004
sc+0.004
 subs+0.004
bo+0.004
Neg logit contributions
azes-0.012
ires-0.010
pit-0.010
mmm-0.010
 fridge-0.010
Token up
Token ID510
Feature activation-1.66e-02

Pos logit contributions
azes+0.064
mmm+0.055
ocolate+0.055
ae+0.054
pit+0.054
Neg logit contributions
sc-0.032
ch-0.031
B-0.030
ched-0.029
 subs-0.028
Token and
Token ID290
Feature activation+1.32e-02

Pos logit contributions
azes+0.032
isdom+0.027
pit+0.027
mmm+0.026
 oven+0.026
Neg logit contributions
B-0.024
ch-0.023
sc-0.023
 subs-0.023
ched-0.021
Token saw
Token ID2497
Feature activation-3.31e-02

Pos logit contributions
B+0.014
ch+0.014
sc+0.013
 subs+0.013
bo+0.013
Neg logit contributions
azes-0.030
ocolate-0.027
ires-0.026
 oven-0.026
ruits-0.026
Token a
Token ID257
Feature activation-5.33e-03

Pos logit contributions
azes+0.072
pit+0.063
mmm+0.062
ruits+0.060
castle+0.060
Neg logit contributions
B-0.041
ch-0.039
sc-0.037
ched-0.037
 subs-0.036
Token friendly
Token ID8030
Feature activation-7.72e-03

Pos logit contributions
azes+0.012
ires+0.011
isdom+0.010
mmm+0.010
pit+0.010
Neg logit contributions
ch-0.006
B-0.006
sc-0.006
 subs-0.006
 scrap-0.005
Token ot
Token ID30972
Feature activation-1.15e-02

Pos logit contributions
azes+0.018
ires+0.016
mmm+0.016
pit+0.016
ruits+0.015
Neg logit contributions
ch-0.009
B-0.008
 subs-0.007
ny-0.007
sc-0.007
Tokenter
Token ID353
Feature activation-1.05e-02

Pos logit contributions
azes+0.016
pit+0.014
castle+0.013
ement+0.013
 fridge+0.013
Neg logit contributions
sc-0.021
ch-0.020
 subs-0.020
B-0.020
text-0.020
Token.
Token ID13
Feature activation-6.33e-03

Pos logit contributions
azes+0.019
pit+0.016
isdom+0.016
ement+0.016
 oven+0.015
Neg logit contributions
sc-0.016
ch-0.016
B-0.015
 subs-0.015
ched-0.014
Token "
Token ID366
Feature activation-2.11e-03

Pos logit contributions
azes+0.013
pit+0.011
castle+0.011
isdom+0.010
ires+0.010
Neg logit contributions
ch-0.009
sc-0.009
B-0.008
 subs-0.008
bo-0.008
Token is
Token ID318
Feature activation-1.14e-02

Pos logit contributions
B+0.006
sc+0.005
 subs+0.005
ch+0.005
Br+0.005
Neg logit contributions
azes-0.007
 oven-0.006
 fridge-0.006
 writing-0.006
 refrigerator-0.006
Token hurt
Token ID5938
Feature activation+1.61e-02

Pos logit contributions
azes+0.024
 oven+0.020
 writing+0.020
pit+0.020
 fridge+0.020
Neg logit contributions
sc-0.015
B-0.014
ch-0.014
 subs-0.013
fl-0.013
Token.
Token ID13
Feature activation+1.14e-02

Pos logit contributions
sc+0.024
ch+0.023
B+0.023
 subs+0.023
ched+0.021
Neg logit contributions
azes-0.030
pit-0.025
ires-0.025
ement-0.025
mmm-0.024
Token
Token ID198
Feature activation+1.23e-02

Pos logit contributions
ch+0.014
sc+0.014
B+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.024
pit-0.020
isdom-0.020
mmm-0.020
ires-0.020
Token
Token ID198
Feature activation+4.42e-03

Pos logit contributions
sc+0.014
ch+0.014
B+0.014
 subs+0.014
ched+0.013
Neg logit contributions
azes-0.027
pit-0.023
mmm-0.022
isdom-0.022
ires-0.022
TokenTom
Token ID13787
Feature activation-2.83e-03

Pos logit contributions
sc+0.006
 subs+0.006
fl+0.006
burn+0.006
text+0.006
Neg logit contributions
azes-0.009
mmm-0.007
isdom-0.007
ires-0.007
ocolate-0.007
Token is
Token ID318
Feature activation-6.46e-03

Pos logit contributions
azes+0.005
pit+0.004
 oven+0.004
 fridge+0.004
isdom+0.004
Neg logit contributions
sc-0.004
B-0.004
ch-0.004
 subs-0.004
ched-0.004
Token scared
Token ID12008
Feature activation+2.47e-03

Pos logit contributions
azes+0.012
 writing+0.010
 oven+0.009
 fridge+0.009
sing+0.009
Neg logit contributions
sc-0.010
B-0.010
ch-0.010
 subs-0.009
fl-0.009
Token.
Token ID13
Feature activation+5.23e-03

Pos logit contributions
ch+0.003
sc+0.003
B+0.003
 subs+0.003
ched+0.003
Neg logit contributions
azes-0.005
pit-0.004
isdom-0.004
mmm-0.004
ires-0.004
Token He
Token ID679
Feature activation-1.59e-03

Pos logit contributions
ch+0.006
B+0.006
sc+0.006
 subs+0.006
bo+0.006
Neg logit contributions
azes-0.011
pit-0.010
castle-0.010
isdom-0.010
ement-0.009
Token runs
Token ID4539
Feature activation+1.61e-02

Pos logit contributions
azes+0.003
 oven+0.002
 fridge+0.002
ires+0.002
 writing+0.002
Neg logit contributions
B-0.003
sc-0.002
 subs-0.002
ch-0.002
Br-0.002
Token a
Token ID257
Feature activation+2.62e-02

Pos logit contributions
sc+0.044
ch+0.041
 subs+0.041
ched+0.040
B+0.038
Neg logit contributions
azes-0.037
 writing-0.031
 fridge-0.030
 oven-0.030
 university-0.029
Token fancy
Token ID14996
Feature activation+1.45e-02

Pos logit contributions
B+0.033
ch+0.032
sc+0.031
 subs+0.030
bo+0.029
Neg logit contributions
azes-0.056
mmm-0.048
ement-0.047
isdom-0.047
pit-0.047
Token cabin
Token ID9351
Feature activation-8.83e-03

Pos logit contributions
B+0.026
ch+0.025
sc+0.024
ny+0.024
 subs+0.024
Neg logit contributions
azes-0.024
mmm-0.020
ocolate-0.019
ement-0.018
isdom-0.018
Token in
Token ID287
Feature activation+1.24e-02

Pos logit contributions
azes+0.020
mmm+0.018
isdom+0.018
pit+0.017
ement+0.017
Neg logit contributions
sc-0.009
B-0.009
ch-0.009
 subs-0.008
bo-0.008
Token the
Token ID262
Feature activation-6.54e-03

Pos logit contributions
sc+0.018
ch+0.018
 subs+0.017
B+0.017
ched+0.017
Neg logit contributions
azes-0.023
 fridge-0.019
 oven-0.019
 writing-0.019
ires-0.019
Token woods
Token ID16479
Feature activation-7.49e-03

Pos logit contributions
azes+0.011
mmm+0.009
isdom+0.009
pit+0.009
ement+0.009
Neg logit contributions
B-0.011
ch-0.011
sc-0.011
 subs-0.010
bo-0.010
Token.
Token ID13
Feature activation+1.10e-02

Pos logit contributions
azes+0.017
mmm+0.016
isdom+0.015
ement+0.015
pit+0.015
Neg logit contributions
sc-0.007
B-0.007
ch-0.007
 subs-0.006
bo-0.006
Token She
Token ID1375
Feature activation+1.39e-02

Pos logit contributions
B+0.015
ch+0.015
sc+0.015
Fl+0.014
|+0.014
Neg logit contributions
azes-0.021
castle-0.020
pit-0.019
isdom-0.018
 Majesty-0.018
Token had
Token ID550
Feature activation+9.17e-03

Pos logit contributions
B+0.020
ch+0.019
sc+0.019
 subs+0.019
bo+0.018
Neg logit contributions
azes-0.027
pit-0.023
mmm-0.022
ement-0.022
isdom-0.022
Token a
Token ID257
Feature activation+2.45e-02

Pos logit contributions
sc+0.015
ch+0.015
B+0.014
 subs+0.014
bo+0.013
Neg logit contributions
azes-0.016
 writing-0.013
 fridge-0.013
pit-0.013
 oven-0.013
Token pet
Token ID4273
Feature activation+1.10e-02

Pos logit contributions
B+0.024
ch+0.023
 subs+0.022
sc+0.022
 scrap+0.021
Neg logit contributions
azes-0.062
mmm-0.053
ement-0.053
ires-0.052
ocolate-0.052
Token Lily
Token ID20037
Feature activation+1.38e-02

Pos logit contributions
azes+0.044
pit+0.038
isdom+0.037
ement+0.037
castle+0.037
Neg logit contributions
B-0.035
ch-0.035
sc-0.035
 subs-0.033
 scrap-0.032
Token said
Token ID531
Feature activation+2.57e-04

Pos logit contributions
sc+0.014
B+0.014
ch+0.013
 subs+0.013
ched+0.012
Neg logit contributions
azes-0.032
ires-0.028
isdom-0.028
ement-0.028
pit-0.028
Token.
Token ID13
Feature activation-3.98e-03

Pos logit contributions
ch+0.000
B+0.000
sc+0.000
 subs+0.000
ched+0.000
Neg logit contributions
azes-0.001
pit-0.000
isdom-0.000
mmm-0.000
ires-0.000
Token "
Token ID366
Feature activation+2.51e-03

Pos logit contributions
azes+0.008
castle+0.007
isdom+0.007
pit+0.007
mmm+0.007
Neg logit contributions
B-0.006
ch-0.006
sc-0.006
 subs-0.005
ny-0.005
TokenCan
Token ID6090
Feature activation-1.93e-02

Pos logit contributions
B+0.003
sc+0.003
ch+0.003
 subs+0.002
bo+0.002
Neg logit contributions
azes-0.006
pit-0.005
ement-0.005
isdom-0.005
 oven-0.005
Token we
Token ID356
Feature activation-4.93e-02

Pos logit contributions
azes+0.037
pit+0.031
isdom+0.031
mmm+0.031
ires+0.031
Neg logit contributions
ch-0.027
sc-0.027
B-0.027
 subs-0.026
bo-0.024
Token take
Token ID1011
Feature activation+1.89e-02

Pos logit contributions
azes+0.106
pit+0.090
ement+0.090
isdom+0.088
ires+0.088
Neg logit contributions
ch-0.060
B-0.059
sc-0.058
 subs-0.056
bo-0.055
Token some
Token ID617
Feature activation-5.13e-03

Pos logit contributions
sc+0.029
B+0.029
ch+0.029
 subs+0.029
ched+0.027
Neg logit contributions
azes-0.032
 oven-0.027
isdom-0.027
pit-0.027
mmm-0.027
Token?"
Token ID1701
Feature activation-2.37e-02

Pos logit contributions
azes+0.010
pit+0.008
isdom+0.008
ement+0.008
mmm+0.008
Neg logit contributions
ch-0.007
sc-0.007
B-0.007
 subs-0.007
bo-0.007
Token
Token ID198
Feature activation+1.43e-03

Pos logit contributions
castle+0.064
azes+0.064
pit+0.063
 throne+0.060
ement+0.060
Neg logit contributions
B-0.014
ch-0.011
s-0.009
L-0.009
Spot-0.009
Token
Token ID198
Feature activation+1.15e-03

Pos logit contributions
sc+0.002
ch+0.002
B+0.002
 subs+0.002
ched+0.002
Neg logit contributions
azes-0.003
ires-0.003
pit-0.003
mmm-0.003
isdom-0.003
Token He
Token ID679
Feature activation-1.68e-02

Pos logit contributions
azes+0.044
castle+0.041
isdom+0.040
ocolate+0.040
sing+0.040
Neg logit contributions
B-0.020
sc-0.019
ch-0.018
 subs-0.017
ny-0.017
Token asked
Token ID1965
Feature activation-2.13e-02

Pos logit contributions
azes+0.039
ires+0.034
pit+0.034
ocolate+0.034
 oven+0.034
Neg logit contributions
B-0.017
ch-0.017
sc-0.017
 subs-0.017
ched-0.015
Token Annie
Token ID25860
Feature activation-8.97e-03

Pos logit contributions
azes+0.050
mmm+0.045
castle+0.044
sing+0.044
pit+0.043
Neg logit contributions
B-0.020
l-0.020
sc-0.019
ch-0.018
 subs-0.018
Token what
Token ID644
Feature activation+5.39e-03

Pos logit contributions
azes+0.019
mmm+0.016
ement+0.016
ires+0.016
pit+0.016
Neg logit contributions
sc-0.011
ch-0.011
B-0.010
ched-0.010
bo-0.010
Token she
Token ID673
Feature activation-2.28e-02

Pos logit contributions
B+0.006
ch+0.005
 subs+0.005
sc+0.005
l+0.005
Neg logit contributions
azes-0.013
pit-0.011
ires-0.011
mmm-0.011
sing-0.011
Token was
Token ID373
Feature activation-4.96e-02

Pos logit contributions
azes+0.061
ocolate+0.055
pit+0.054
 oven+0.054
ires+0.053
Neg logit contributions
B-0.016
ch-0.016
sc-0.015
 subs-0.015
ched-0.014
Token doing
Token ID1804
Feature activation-2.26e-02

Pos logit contributions
azes+0.089
pit+0.079
ocolate+0.078
ement+0.078
mmm+0.078
Neg logit contributions
fl-0.070
ched-0.063
ch-0.063
 hurt-0.062
sc-0.062
Token under
Token ID739
Feature activation-1.25e-02

Pos logit contributions
azes+0.055
sing+0.048
pit+0.046
mmm+0.046
 oven+0.046
Neg logit contributions
sc-0.020
B-0.018
fl-0.018
ched-0.017
 on-0.016
Token the
Token ID262
Feature activation-1.76e-02

Pos logit contributions
azes+0.030
pit+0.027
castle+0.026
mmm+0.026
ires+0.026
Neg logit contributions
B-0.011
ch-0.010
sc-0.010
 scrap-0.010
 subs-0.010
Token post
Token ID1281
Feature activation-2.07e-02

Pos logit contributions
azes+0.029
pit+0.023
ires+0.023
isdom+0.022
mmm+0.022
Neg logit contributions
ch-0.032
B-0.031
sc-0.030
 subs-0.030
fl-0.029
Token.
Token ID13
Feature activation-1.47e-02

Pos logit contributions
azes+0.043
ires+0.036
mmm+0.036
isdom+0.035
ruits+0.035
Neg logit contributions
sc-0.027
ch-0.025
bo-0.024
l-0.024
 subs-0.024
Token,
Token ID11
Feature activation-1.71e-02

Pos logit contributions
ch+0.008
B+0.008
sc+0.008
 subs+0.008
ched+0.008
Neg logit contributions
azes-0.009
pit-0.007
ires-0.007
isdom-0.007
 fridge-0.007
Token "
Token ID366
Feature activation-9.83e-03

Pos logit contributions
azes+0.032
pit+0.026
ires+0.026
isdom+0.026
ement+0.025
Neg logit contributions
B-0.027
ch-0.027
sc-0.026
 subs-0.025
bo-0.024
TokenOf
Token ID5189
Feature activation-3.16e-03

Pos logit contributions
azes+0.016
mmm+0.014
ires+0.014
 patiently+0.014
pit+0.014
Neg logit contributions
 subs-0.014
ch-0.014
text-0.014
 scrap-0.014
sc-0.014
Token course
Token ID1781
Feature activation-2.93e-02

Pos logit contributions
ement+0.002
azes+0.002
 oven+0.002
mmm+0.002
isdom+0.001
Neg logit contributions
sc-0.008
 subs-0.008
ch-0.008
ched-0.008
 scrap-0.008
Token!
Token ID0
Feature activation-2.26e-02

Pos logit contributions
azes+0.050
pit+0.041
mmm+0.041
ement+0.041
isdom+0.041
Neg logit contributions
sc-0.047
ch-0.047
B-0.047
 subs-0.045
ched-0.043
Token Thank
Token ID6952
Feature activation-4.36e-02

Pos logit contributions
azes+0.040
pit+0.033
mmm+0.032
isdom+0.032
ires+0.032
Neg logit contributions
B-0.035
sc-0.035
ch-0.035
 subs-0.034
ched-0.032
Token you
Token ID345
Feature activation-8.56e-03

Pos logit contributions
azes+0.059
mmm+0.051
ocolate+0.047
isdom+0.046
ummy+0.046
Neg logit contributions
sc-0.084
 subs-0.084
 scrap-0.082
ch-0.081
ched-0.079
Token for
Token ID329
Feature activation-1.42e-02

Pos logit contributions
azes+0.016
pit+0.013
ires+0.012
isdom+0.012
ement+0.012
Neg logit contributions
ch-0.014
sc-0.013
B-0.013
 subs-0.012
text-0.012
Token offering
Token ID6011
Feature activation-2.79e-02

Pos logit contributions
azes+0.022
pit+0.019
ement+0.018
ires+0.018
castle+0.017
Neg logit contributions
B-0.026
ch-0.025
sc-0.025
 subs-0.024
ched-0.023
Token to
Token ID284
Feature activation-8.05e-03

Pos logit contributions
azes+0.063
pit+0.055
castle+0.052
sing+0.052
ement+0.050
Neg logit contributions
ch-0.037
sc-0.036
B-0.036
ched-0.032
text-0.031
Token help
Token ID1037
Feature activation+8.26e-03

Pos logit contributions
azes+0.017
pit+0.014
ement+0.014
ires+0.014
castle+0.014
Neg logit contributions
ch-0.011
B-0.011
sc-0.010
 subs-0.010
bo-0.009
Token but
Token ID475
Feature activation-1.20e-02

Pos logit contributions
azes+0.018
isdom+0.015
mmm+0.015
ires+0.015
ement+0.015
Neg logit contributions
ch-0.014
sc-0.014
B-0.014
 subs-0.013
 scrap-0.013
Token he
Token ID339
Feature activation-4.72e-03

Pos logit contributions
azes+0.026
isdom+0.023
pit+0.022
ement+0.022
mmm+0.022
Neg logit contributions
ch-0.015
B-0.014
sc-0.014
 subs-0.014
 scrap-0.013
Token cannot
Token ID2314
Feature activation+8.88e-03

Pos logit contributions
azes+0.009
mmm+0.008
isdom+0.008
pit+0.008
ocolate+0.008
Neg logit contributions
ch-0.006
sc-0.006
ched-0.006
 scrap-0.006
bo-0.006
Token grab
Token ID5552
Feature activation-7.34e-03

Pos logit contributions
ch+0.011
B+0.011
sc+0.010
bo+0.010
 subs+0.010
Neg logit contributions
azes-0.019
pit-0.017
ement-0.016
isdom-0.016
 oven-0.016
Token the
Token ID262
Feature activation-1.02e-02

Pos logit contributions
azes+0.014
pit+0.012
ires+0.012
ement+0.012
 oven+0.012
Neg logit contributions
sc-0.010
ch-0.010
B-0.010
 subs-0.010
ched-0.009
Token o
Token ID267
Feature activation-4.46e-02

Pos logit contributions
azes+0.018
pit+0.015
isdom+0.015
mmm+0.015
ires+0.015
Neg logit contributions
ch-0.016
sc-0.016
B-0.016
 subs-0.015
bo-0.014
Tokenat
Token ID265
Feature activation-1.09e-02

Pos logit contributions
azes+0.061
pit+0.047
isdom+0.046
mmm+0.046
ires+0.046
Neg logit contributions
ch-0.087
sc-0.087
B-0.087
 subs-0.084
bo-0.081
Token.
Token ID13
Feature activation+7.51e-03

Pos logit contributions
azes+0.022
pit+0.019
 oven+0.019
mmm+0.019
ement+0.018
Neg logit contributions
ch-0.014
B-0.014
sc-0.014
 subs-0.013
ched-0.012
Token
Token ID198
Feature activation+1.26e-02

Pos logit contributions
B+0.009
ch+0.009
sc+0.009
 subs+0.008
bo+0.008
Neg logit contributions
azes-0.016
castle-0.014
pit-0.014
isdom-0.014
ement-0.014
Token
Token ID198
Feature activation+6.80e-03

Pos logit contributions
sc+0.015
ch+0.015
B+0.015
 subs+0.015
ched+0.013
Neg logit contributions
azes-0.026
pit-0.022
mmm-0.022
ires-0.022
isdom-0.022
TokenTom
Token ID13787
Feature activation+5.96e-03

Pos logit contributions
sc+0.010
 subs+0.010
ch+0.010
B+0.010
fl+0.010
Neg logit contributions
azes-0.012
mmm-0.010
pit-0.010
ires-0.010
ae-0.010
Token a
Token ID257
Feature activation-7.21e-04

Pos logit contributions
azes+0.092
mmm+0.084
isdom+0.083
ement+0.080
pit+0.080
Neg logit contributions
ch-0.042
B-0.041
sc-0.037
fl-0.036
 scrap-0.035
Token big
Token ID1263
Feature activation-1.81e-02

Pos logit contributions
azes+0.002
isdom+0.002
mmm+0.002
ires+0.001
ement+0.001
Neg logit contributions
ch-0.001
B-0.001
sc-0.001
 subs-0.001
 scrap-0.001
Token board
Token ID3096
Feature activation-2.38e-02

Pos logit contributions
azes+0.041
isdom+0.036
ires+0.035
mmm+0.034
idge+0.034
Neg logit contributions
ch-0.024
sc-0.021
 st-0.020
B-0.020
 k-0.019
Token at
Token ID379
Feature activation-2.63e-02

Pos logit contributions
azes+0.053
isdom+0.048
mmm+0.047
 Majesty+0.046
ires+0.045
Neg logit contributions
sc-0.029
ch-0.028
B-0.025
 subs-0.024
ny-0.024
Token the
Token ID262
Feature activation-2.34e-02

Pos logit contributions
azes+0.071
isdom+0.064
mmm+0.064
ement+0.064
pit+0.062
Neg logit contributions
ch-0.016
B-0.016
sc-0.016
 subs-0.013
fl-0.013
Token edge
Token ID5743
Feature activation-5.01e-02

Pos logit contributions
azes+0.039
isdom+0.033
mmm+0.032
ement+0.031
pit+0.031
Neg logit contributions
ch-0.040
sc-0.040
B-0.039
 subs-0.037
bo-0.036
Token of
Token ID286
Feature activation+9.29e-03

Pos logit contributions
azes+0.115
isdom+0.102
mmm+0.101
ement+0.098
 Majesty+0.098
Neg logit contributions
ch-0.053
sc-0.052
B-0.050
 subs-0.046
bo-0.045
Token the
Token ID262
Feature activation-1.78e-02

Pos logit contributions
B+0.006
ch+0.006
sc+0.005
Fl+0.004
 subs+0.004
Neg logit contributions
azes-0.026
isdom-0.023
mmm-0.023
ement-0.023
sing-0.022
Token pool
Token ID5933
Feature activation-2.60e-02

Pos logit contributions
azes+0.024
pit+0.019
 oven+0.019
 fridge+0.019
ires+0.019
Neg logit contributions
B-0.034
ch-0.034
sc-0.034
 subs-0.034
ched-0.032
Token.
Token ID13
Feature activation+6.77e-04

Pos logit contributions
azes+0.070
isdom+0.066
mmm+0.063
 Majesty+0.062
ement+0.060
Neg logit contributions
sc-0.022
ch-0.018
ny-0.017
B-0.016
s-0.016
Token She
Token ID1375
Feature activation+4.17e-03

Pos logit contributions
ch+0.001
B+0.001
sc+0.001
 subs+0.001
fl+0.001
Neg logit contributions
azes-0.001
isdom-0.001
castle-0.001
ement-0.001
 writing-0.001