TOP ACTIVATIONS
MAX = 5.314

It was a sm al ll toy car that looked just
" Hi , O ll ie !" Sara says .
Mom took the dog to Mr . Lee 's house .
said , " Hi O ll ie ". <|endoftext|> Once ,
" Good morning , O ll ie !" said his mother
Are you okay , O ll ie ?" she asked .
Thank you , God . Amen ." He opened
" Good job , O ll ie !" his mother praised
charming oct opus named O ll ie . O ll ie
and boots and ran to Mr . Lee 's shop .
. He said , " if people hurt the universe ,
happy oct opus named O ll ie . O ll ie
and brought it back to Mr . Johnson . He was
healthy ot ter named O ll ie . O ll ie
furry oct opus named O ll ie . O ll ie
saved ! He thanked O ll ie and went back to
little ot ter named O ll ie . O ll ie
a good friend , O ll ie ." O ll ie
make it whole again . Amen ." Then ,
" OK , O ll ie , show them what

MIN ACTIVATIONS
MIN = -6.415

the tree to its once former beauty . He was happy
from her , but Kay lee was determined to catch it
my had told him long ago . The bird spread out
started to chase after Kay lee . She screamed and ran
know you shouldn â t speed ? â
bag so we don 't accidentally touch it again ."
to be restored to its former glory ."
. And if she did accidentally knock over her cup ,
no , I must have accidentally spilled some water on it
quietly so he wouldn 't accidentally hit the wall or disturb
, so she wouldn 't accidentally break anything again . <|endoftext|>
's mom realized she had accidentally used salt instead of sugar
one he had when he accidentally reversed . <|endoftext|> Once upon
But Jimmy had accidentally put too much of a
and worse until Tim my accidentally ran into the fence and
behind her , but she accidentally tri pped and the pipe
to get the pot but accidentally dropped it on the floor
while , but then she accidentally knocked over a tower she
." But Lily accidentally dropped the gem and it
pepper in it when she accidentally spilled some on her hand

INTERVAL 2.381 - 5.314
CONTAINS 1.045%

fish ." O ll ie was sad , but
. They say , " OK , Mom . We will
Sh h , Tom , don 't be rude !" Jen
a mom my , a daddy and a baby . The
The bird said , " Yes , let 's play !"

INTERVAL -0.551 - 2.381
CONTAINS 71.217%

!" L ily and Ben did not want to
. We can share our toys and have fun ."
." He goes to the pig . The pig
's raining ! It 's not fair !" His
I 'm glad you enjoyed it . Do you want to

INTERVAL -3.483 - -0.551
CONTAINS 27.694%

her a snack and said , " Now you can play
sorry . He asks if he can play with them .
" Mom my , can I have that toy ?"
ground ! â The bird was scared and didn â
and sad . Lily and her mom were happy to have

INTERVAL -6.415 - -3.483
CONTAINS 0.043%

hours , until Tim my accidentally cut his finger on a
when we don â t know what to do
if you don â t understand why , sometimes
. One day , she accidentally dropped her cup and it
he just couldn â t work it out .
Token It
Token ID632
Feature activation+1.11e+00

Pos logit contributions
adies+0.017
 carriers+0.015
 lantern+0.015
Mom+0.015
 ho+0.014
Neg logit contributions
arre-0.017
 fault-0.014
wheel-0.014
ravings-0.013
igsaw-0.013
Token was
Token ID373
Feature activation+1.44e+00

Pos logit contributions
adies+0.026
 lantern+0.024
Mom+0.023
 carriers+0.023
 ho+0.021
Neg logit contributions
arre-0.041
igsaw-0.038
ravings-0.036
 fault-0.036
 courtroom-0.035
Token a
Token ID257
Feature activation+1.03e+00

Pos logit contributions
adies+0.030
 lantern+0.029
 ho+0.028
Mom+0.027
 carriers+0.027
Neg logit contributions
arre-0.060
igsaw-0.054
 courtroom-0.050
ravings-0.049
wheel-0.049
Token sm
Token ID895
Feature activation+4.61e-01

Pos logit contributions
adies+0.028
 lantern+0.025
 carriers+0.025
Mom+0.024
 ho+0.023
Neg logit contributions
arre-0.034
igsaw-0.030
ravings-0.029
wheel-0.029
 fault-0.029
Tokenal
Token ID282
Feature activation+2.90e-01

Pos logit contributions
adies+0.018
 carriers+0.017
 lantern+0.016
 ho+0.016
Mom+0.016
Neg logit contributions
arre-0.010
igsaw-0.008
ravings-0.008
 courtroom-0.008
wheel-0.008
Tokenll
Token ID297
Feature activation+5.31e+00

Pos logit contributions
adies+0.008
 lantern+0.007
le+0.007
 carriers+0.007
 ho+0.007
Neg logit contributions
arre-0.010
igsaw-0.009
 courtroom-0.008
 fault-0.008
ravings-0.008
Token toy
Token ID13373
Feature activation+7.55e-01

Pos logit contributions
 carriers+0.136
adies+0.132
Mom+0.114
works+0.112
Ho+0.112
Neg logit contributions
wheel-0.153
 supposed-0.150
arre-0.146
 about-0.145
igsaw-0.145
Token car
Token ID1097
Feature activation+1.56e+00

Pos logit contributions
adies+0.018
 carriers+0.017
 lantern+0.016
 ho+0.015
 aliens+0.015
Neg logit contributions
arre-0.029
igsaw-0.026
 courtroom-0.024
ravings-0.023
 fault-0.023
Token that
Token ID326
Feature activation-2.93e-01

Pos logit contributions
adies+0.031
 carriers+0.030
 ho+0.027
 lantern+0.026
le+0.025
Neg logit contributions
arre-0.065
igsaw-0.059
ravings-0.054
 courtroom-0.053
 fault-0.053
Token looked
Token ID3114
Feature activation+1.11e+00

Pos logit contributions
arre+0.011
igsaw+0.010
ravings+0.009
 courtroom+0.009
 fault+0.009
Neg logit contributions
adies-0.007
 lantern-0.006
 carriers-0.006
Mom-0.006
 ho-0.006
Token just
Token ID655
Feature activation+4.86e-01

Pos logit contributions
adies+0.036
 carriers+0.032
 lantern+0.032
Mom+0.031
 ho+0.030
Neg logit contributions
arre-0.031
igsaw-0.027
ravings-0.026
 courtroom-0.025
wheel-0.025
Token
Token ID198
Feature activation+5.72e-01

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 courtroom+0.003
wheel+0.003
Neg logit contributions
adies-0.004
Mom-0.003
 lantern-0.003
 carriers-0.003
Dad-0.003
Token"
Token ID1
Feature activation+4.26e-02

Pos logit contributions
adies+0.016
 carriers+0.014
 lantern+0.014
Mom+0.013
 ho+0.013
Neg logit contributions
arre-0.019
igsaw-0.016
ravings-0.016
 courtroom-0.016
wheel-0.016
TokenHi
Token ID17250
Feature activation+1.39e+00

Pos logit contributions
adies+0.001
 lantern+0.001
 carriers+0.001
works+0.001
 sle+0.001
Neg logit contributions
arre-0.001
wheel-0.001
 winning-0.001
Em-0.001
 accidentally-0.001
Token,
Token ID11
Feature activation+5.49e-01

Pos logit contributions
adies+0.047
 carriers+0.045
 Amen+0.043
 lantern+0.042
works+0.041
Neg logit contributions
arre-0.033
 accidentally-0.030
 winning-0.029
igsaw-0.027
 court-0.027
Token O
Token ID440
Feature activation+1.81e+00

Pos logit contributions
adies+0.017
 carriers+0.016
 lantern+0.015
Mom+0.015
 ho+0.015
Neg logit contributions
arre-0.016
igsaw-0.014
ravings-0.013
 courtroom-0.013
wheel-0.013
Tokenll
Token ID297
Feature activation+5.24e+00

Pos logit contributions
adies+0.078
 carriers+0.077
 lantern+0.072
 aliens+0.072
 Amen+0.071
Neg logit contributions
arre-0.031
igsaw-0.022
wheel-0.022
ravings-0.021
 accidentally-0.020
Tokenie
Token ID494
Feature activation+7.61e-01

Pos logit contributions
 carriers+0.155
adies+0.147
works+0.131
 aliens+0.130
 lantern+0.127
Neg logit contributions
 accidentally-0.131
arre-0.129
wheel-0.123
 having-0.120
 about-0.118
Token!"
Token ID2474
Feature activation-1.01e+00

Pos logit contributions
adies+0.022
 carriers+0.020
 lantern+0.019
Mom+0.019
works+0.018
Neg logit contributions
arre-0.022
igsaw-0.020
 accidentally-0.019
 court-0.019
ravings-0.019
Token Sara
Token ID24799
Feature activation-8.90e-01

Pos logit contributions
arre+0.030
igsaw+0.026
ravings+0.025
 fault+0.025
 courtroom+0.025
Neg logit contributions
adies-0.032
 lantern-0.029
 carriers-0.028
Mom-0.028
 ho-0.027
Token says
Token ID1139
Feature activation+1.07e+00

Pos logit contributions
arre+0.031
igsaw+0.028
ravings+0.027
 courtroom+0.027
wheel+0.027
Neg logit contributions
adies-0.022
 carriers-0.019
 lantern-0.019
Mom-0.018
 ho-0.018
Token.
Token ID13
Feature activation+3.34e-01

Pos logit contributions
adies+0.038
 carriers+0.034
works+0.034
 lantern+0.034
Mom+0.033
Neg logit contributions
arre-0.024
 about-0.022
 winning-0.022
 court-0.022
wheel-0.022
Token Mom
Token ID11254
Feature activation+2.89e-01

Pos logit contributions
adies+0.065
 carriers+0.057
 lantern+0.055
Mom+0.053
le+0.052
Neg logit contributions
arre-0.068
igsaw-0.061
 winning-0.060
 court-0.059
ravings-0.059
Token took
Token ID1718
Feature activation+2.95e-01

Pos logit contributions
adies+0.009
 lantern+0.008
 carriers+0.008
 Amen+0.008
 comet+0.008
Neg logit contributions
arre-0.008
wheel-0.007
 accidentally-0.007
 about-0.007
ravings-0.007
Token the
Token ID262
Feature activation+1.29e-01

Pos logit contributions
adies+0.009
works+0.008
 lantern+0.007
 carriers+0.007
 Amen+0.007
Neg logit contributions
arre-0.008
igsaw-0.007
 fault-0.007
 court-0.007
 courtroom-0.007
Token dog
Token ID3290
Feature activation+1.15e+00

Pos logit contributions
adies+0.004
 lantern+0.003
 carriers+0.003
el+0.003
works+0.003
Neg logit contributions
arre-0.004
 court-0.003
 courtroom-0.003
igsaw-0.003
 fault-0.003
Token to
Token ID284
Feature activation-1.44e-02

Pos logit contributions
adies+0.032
 carriers+0.030
 lantern+0.029
 ho+0.028
le+0.027
Neg logit contributions
arre-0.037
igsaw-0.032
ravings-0.031
 courtroom-0.031
wheel-0.031
Token Mr
Token ID1770
Feature activation+5.11e+00

Pos logit contributions
arre+0.001
igsaw+0.000
ravings+0.000
wheel+0.000
 fault+0.000
Neg logit contributions
adies-0.000
 lantern-0.000
 carriers-0.000
Mom-0.000
 ho-0.000
Token.
Token ID13
Feature activation+9.48e-01

Pos logit contributions
adies+0.193
Run+0.164
 lantern+0.160
Ho+0.158
works+0.157
Neg logit contributions
 fault-0.105
arre-0.099
 court-0.098
ravings-0.093
 courtroom-0.085
Token Lee
Token ID5741
Feature activation+8.28e-01

Pos logit contributions
adies+0.035
le+0.030
works+0.030
 carriers+0.029
 lantern+0.029
Neg logit contributions
arre-0.022
 court-0.020
 fault-0.020
ravings-0.020
 winning-0.019
Token's
Token ID338
Feature activation+6.27e-01

Pos logit contributions
adies+0.018
 lantern+0.017
 ho+0.016
 carriers+0.016
le+0.016
Neg logit contributions
arre-0.032
igsaw-0.030
ravings-0.028
 fault-0.027
 courtroom-0.027
Token house
Token ID2156
Feature activation-8.26e-01

Pos logit contributions
adies+0.013
 carriers+0.013
 lantern+0.012
 ho+0.012
Mom+0.011
Neg logit contributions
arre-0.025
igsaw-0.022
wheel-0.022
 accidentally-0.021
ravings-0.020
Token.
Token ID13
Feature activation+5.24e-01

Pos logit contributions
arre+0.027
igsaw+0.024
ravings+0.024
 courtroom+0.024
wheel+0.023
Neg logit contributions
adies-0.022
 carriers-0.020
 lantern-0.019
Mom-0.018
 ho-0.018
Token said
Token ID531
Feature activation-3.85e-01

Pos logit contributions
 ho+0.002
 lantern+0.002
adies+0.002
 wave+0.002
 carriers+0.002
Neg logit contributions
arre-0.005
igsaw-0.005
 fault-0.004
ravings-0.004
 courtroom-0.004
Token,
Token ID11
Feature activation-1.67e+00

Pos logit contributions
arre+0.013
igsaw+0.011
ravings+0.011
 courtroom+0.011
wheel+0.011
Neg logit contributions
adies-0.010
 carriers-0.009
 lantern-0.009
Mom-0.009
 ho-0.008
Token "
Token ID366
Feature activation-4.72e-01

Pos logit contributions
arre+0.046
igsaw+0.043
 fault+0.040
ravings+0.039
 courtroom+0.038
Neg logit contributions
adies-0.054
Mom-0.050
 lantern-0.050
 ho-0.050
 carriers-0.048
TokenHi
Token ID17250
Feature activation+9.32e-01

Pos logit contributions
arre+0.013
wheel+0.011
 winning+0.011
igsaw+0.011
Em+0.011
Neg logit contributions
adies-0.015
 lantern-0.013
 carriers-0.013
 sle-0.012
 aliens-0.012
Token O
Token ID440
Feature activation+2.03e+00

Pos logit contributions
adies+0.031
 carriers+0.029
 lantern+0.026
works+0.026
 Amen+0.026
Neg logit contributions
arre-0.023
 accidentally-0.020
igsaw-0.020
 winning-0.020
ravings-0.019
Tokenll
Token ID297
Feature activation+5.09e+00

Pos logit contributions
 carriers+0.082
adies+0.080
 aliens+0.077
 Amen+0.073
 lantern+0.072
Neg logit contributions
arre-0.038
igsaw-0.032
ott-0.030
ravings-0.028
 colony-0.028
Tokenie
Token ID494
Feature activation+8.00e-01

Pos logit contributions
adies+0.132
 carriers+0.129
Run+0.110
works+0.109
 aliens+0.108
Neg logit contributions
arre-0.146
 accidentally-0.136
igsaw-0.135
wheel-0.134
 winning-0.133
Token".
Token ID1911
Feature activation-1.07e+00

Pos logit contributions
adies+0.022
 carriers+0.020
 lantern+0.019
Mom+0.018
works+0.018
Neg logit contributions
arre-0.025
igsaw-0.022
ravings-0.022
 winning-0.022
wheel-0.022
Token<|endoftext|>
Token ID50256
Feature activation-1.48e+00

Pos logit contributions
arre+0.035
igsaw+0.031
 courtroom+0.030
ravings+0.030
 fault+0.030
Neg logit contributions
adies-0.029
Mom-0.028
 lantern-0.028
el-0.027
 ho-0.027
TokenOnce
Token ID7454
Feature activation+2.71e-01

Pos logit contributions
arre+0.036
igsaw+0.031
ravings+0.030
 courtroom+0.029
wheel+0.029
Neg logit contributions
adies-0.052
 carriers-0.047
 lantern-0.047
Mom-0.045
 ho-0.044
Token,
Token ID11
Feature activation-1.07e+00

Pos logit contributions
adies+0.007
 lantern+0.006
Mom+0.006
 carriers+0.006
 ho+0.006
Neg logit contributions
arre-0.009
 courtroom-0.008
ravings-0.008
wheel-0.008
igsaw-0.008
Token"
Token ID1
Feature activation-1.83e-01

Pos logit contributions
adies+0.004
 carriers+0.004
 lantern+0.003
works+0.003
 comet+0.003
Neg logit contributions
arre-0.003
 accidentally-0.003
wheel-0.003
 court-0.003
ravings-0.003
TokenGood
Token ID10248
Feature activation+2.04e+00

Pos logit contributions
arre+0.004
Em+0.004
 accidentally+0.004
wheel+0.004
 basketball+0.004
Neg logit contributions
adies-0.006
works-0.005
 carriers-0.005
 sle-0.005
 lantern-0.005
Token morning
Token ID3329
Feature activation+5.75e-02

Pos logit contributions
adies+0.070
 carriers+0.065
Mom+0.064
 lantern+0.060
Dad+0.059
Neg logit contributions
arre-0.050
 fault-0.046
ravings-0.042
 accidentally-0.040
wheel-0.040
Token,
Token ID11
Feature activation+2.64e-01

Pos logit contributions
adies+0.002
 carriers+0.002
 lantern+0.001
Mom+0.001
works+0.001
Neg logit contributions
arre-0.002
igsaw-0.001
 accidentally-0.001
 court-0.001
ravings-0.001
Token O
Token ID440
Feature activation+1.55e+00

Pos logit contributions
adies+0.009
 carriers+0.008
 lantern+0.008
Mom+0.007
 ho+0.007
Neg logit contributions
arre-0.007
igsaw-0.006
ravings-0.006
 courtroom-0.006
wheel-0.006
Tokenll
Token ID297
Feature activation+4.91e+00

Pos logit contributions
adies+0.069
 carriers+0.068
 Amen+0.063
 aliens+0.063
 lantern+0.062
Neg logit contributions
arre-0.027
ravings-0.018
igsaw-0.018
 accidentally-0.018
wheel-0.017
Tokenie
Token ID494
Feature activation+7.45e-01

Pos logit contributions
 carriers+0.155
adies+0.142
works+0.130
Run+0.130
 Amen+0.125
Neg logit contributions
arre-0.117
 accidentally-0.113
 done-0.108
 having-0.108
 about-0.106
Token!"
Token ID2474
Feature activation-1.29e+00

Pos logit contributions
adies+0.021
 carriers+0.019
 lantern+0.018
Mom+0.018
works+0.018
Neg logit contributions
arre-0.022
 accidentally-0.020
ravings-0.019
 court-0.019
igsaw-0.019
Token said
Token ID531
Feature activation+7.76e-01

Pos logit contributions
arre+0.028
 accidentally+0.024
 court+0.023
wheel+0.023
ravings+0.023
Neg logit contributions
adies-0.048
 carriers-0.045
 lantern-0.042
works-0.041
le-0.040
Token his
Token ID465
Feature activation+6.90e-01

Pos logit contributions
adies+0.028
 carriers+0.026
 lantern+0.025
works+0.024
Mom+0.024
Neg logit contributions
arre-0.017
 accidentally-0.015
 court-0.015
 winning-0.015
igsaw-0.015
Token mother
Token ID2802
Feature activation+3.18e-01

Pos logit contributions
adies+0.019
 carriers+0.016
le+0.015
works+0.015
 lantern+0.014
Neg logit contributions
arre-0.023
ravings-0.020
 fault-0.020
 courtroom-0.020
igsaw-0.020
TokenAre
Token ID8491
Feature activation+2.36e+00

Pos logit contributions
arre+0.005
wheel+0.004
 accidentally+0.004
Em+0.004
ravings+0.004
Neg logit contributions
adies-0.005
 carriers-0.005
 lantern-0.004
works-0.004
 sle-0.004
Token you
Token ID345
Feature activation-7.78e-01

Pos logit contributions
adies+0.061
 carriers+0.057
 lantern+0.054
le+0.050
works+0.049
Neg logit contributions
arre-0.079
 winning-0.070
 accidentally-0.070
igsaw-0.070
wheel-0.069
Token okay
Token ID8788
Feature activation+1.50e+00

Pos logit contributions
arre+0.025
igsaw+0.024
 courtroom+0.023
ravings+0.023
 court+0.023
Neg logit contributions
adies-0.021
 lantern-0.018
 ho-0.018
 carriers-0.018
Mom-0.017
Token,
Token ID11
Feature activation-8.06e-02

Pos logit contributions
adies+0.039
 carriers+0.034
 lantern+0.034
Mom+0.032
 ho+0.032
Neg logit contributions
arre-0.051
igsaw-0.045
ravings-0.044
 courtroom-0.044
wheel-0.044
Token O
Token ID440
Feature activation+1.68e+00

Pos logit contributions
arre+0.003
igsaw+0.002
ravings+0.002
 courtroom+0.002
wheel+0.002
Neg logit contributions
adies-0.002
 carriers-0.002
 lantern-0.002
Mom-0.002
 ho-0.002
Tokenll
Token ID297
Feature activation+4.85e+00

Pos logit contributions
 carriers+0.072
adies+0.072
 aliens+0.067
 lantern+0.065
Mom+0.064
Neg logit contributions
arre-0.029
 fault-0.021
ravings-0.021
igsaw-0.020
 accidentally-0.020
Tokenie
Token ID494
Feature activation+4.11e-01

Pos logit contributions
 carriers+0.143
adies+0.141
Land+0.121
Run+0.120
works+0.117
Neg logit contributions
arre-0.116
 about-0.112
wheel-0.111
 accidentally-0.111
 having-0.104
Token?"
Token ID1701
Feature activation-1.68e+00

Pos logit contributions
adies+0.010
 lantern+0.008
 carriers+0.008
le+0.008
Mom+0.008
Neg logit contributions
arre-0.015
igsaw-0.014
ravings-0.013
 courtroom-0.013
 winning-0.013
Token she
Token ID673
Feature activation-1.39e-01

Pos logit contributions
arre+0.058
igsaw+0.051
 fault+0.049
ravings+0.049
 courtroom+0.046
Neg logit contributions
adies-0.046
Mom-0.044
 lantern-0.043
 ho-0.042
Dad-0.041
Token asked
Token ID1965
Feature activation+2.09e-01

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 fault+0.003
 courtroom+0.003
Neg logit contributions
adies-0.005
 carriers-0.004
Mom-0.004
 lantern-0.004
 ho-0.004
Token.
Token ID13
Feature activation+7.16e-01

Pos logit contributions
adies+0.007
 carriers+0.006
works+0.006
 lantern+0.006
 comet+0.006
Neg logit contributions
arre-0.005
 accidentally-0.005
 court-0.005
 winning-0.005
wheel-0.005
Token Thank
Token ID6952
Feature activation+3.54e+00

Pos logit contributions
adies+0.023
 lantern+0.020
le+0.020
agles+0.020
 carriers+0.019
Neg logit contributions
arre-0.021
 By-0.019
 Unfortunately-0.019
 winning-0.018
 courtroom-0.018
Token you
Token ID345
Feature activation+1.45e+00

Pos logit contributions
 flashlight+0.129
adies+0.127
 lantern+0.126
lights+0.125
 telescope+0.125
Neg logit contributions
 court-0.054
 being-0.051
 class-0.050
 practicing-0.050
arre-0.049
Token,
Token ID11
Feature activation+1.01e+00

Pos logit contributions
adies+0.050
 carriers+0.045
 lantern+0.045
Mom+0.043
le+0.043
Neg logit contributions
arre-0.036
ravings-0.029
 court-0.029
igsaw-0.029
 fault-0.029
Token God
Token ID1793
Feature activation+3.81e+00

Pos logit contributions
adies+0.026
 carriers+0.022
 lantern+0.022
Mom+0.021
 ho+0.021
Neg logit contributions
arre-0.035
igsaw-0.032
ravings-0.031
 courtroom-0.031
wheel-0.031
Token.
Token ID13
Feature activation+1.17e+00

Pos logit contributions
adies+0.100
 carriers+0.088
 lantern+0.086
Mom+0.082
 ho+0.081
Neg logit contributions
arre-0.129
igsaw-0.116
ravings-0.112
 courtroom-0.112
wheel-0.110
Token Amen
Token ID34717
Feature activation+4.84e+00

Pos logit contributions
adies+0.029
le+0.025
 lantern+0.025
agles+0.024
 carriers+0.024
Neg logit contributions
arre-0.038
 By-0.035
 winning-0.034
 Working-0.034
 courtroom-0.034
Token."
Token ID526
Feature activation+1.37e-01

Pos logit contributions
adies+0.145
 lantern+0.135
 carriers+0.132
 comet+0.128
Ho+0.126
Neg logit contributions
arre-0.130
 court-0.112
 about-0.112
 winning-0.112
ravings-0.112
Token
Token ID198
Feature activation+9.25e-01

Pos logit contributions
adies+0.005
 carriers+0.004
 lantern+0.004
Mom+0.004
 ho+0.004
Neg logit contributions
arre-0.004
igsaw-0.003
ravings-0.003
wheel-0.003
 fault-0.003
Token
Token ID198
Feature activation+1.41e+00

Pos logit contributions
adies+0.034
 carriers+0.031
 lantern+0.031
Mom+0.030
 ho+0.029
Neg logit contributions
arre-0.021
igsaw-0.018
ravings-0.017
 courtroom-0.017
wheel-0.017
TokenHe
Token ID1544
Feature activation+2.10e+00

Pos logit contributions
adies+0.038
 carriers+0.032
 lantern+0.032
works+0.031
 ho+0.031
Neg logit contributions
arre-0.045
 court-0.041
 courtroom-0.041
igsaw-0.040
ravings-0.039
Token opened
Token ID4721
Feature activation-1.16e-01

Pos logit contributions
adies+0.057
 carriers+0.050
 lantern+0.050
Mom+0.048
 ho+0.047
Neg logit contributions
arre-0.069
igsaw-0.062
ravings-0.060
 fault-0.060
 courtroom-0.059
Token"
Token ID1
Feature activation+1.40e-01

Pos logit contributions
adies+0.011
 carriers+0.010
Mom+0.010
 lantern+0.010
 ho+0.009
Neg logit contributions
arre-0.019
igsaw-0.017
 fault-0.017
ravings-0.016
 courtroom-0.016
TokenGood
Token ID10248
Feature activation+2.34e+00

Pos logit contributions
adies+0.004
 lantern+0.004
 sle+0.004
le+0.004
 ho+0.004
Neg logit contributions
arre-0.004
wheel-0.004
Em-0.004
 winning-0.003
 basketball-0.003
Token job
Token ID1693
Feature activation+1.67e-01

Pos logit contributions
adies+0.074
 carriers+0.065
Mom+0.064
 lantern+0.062
Ho+0.060
Neg logit contributions
arre-0.065
 fault-0.059
igsaw-0.056
 winning-0.055
ravings-0.055
Token,
Token ID11
Feature activation+1.76e-01

Pos logit contributions
adies+0.003
 ho+0.003
 lantern+0.002
 carriers+0.002
works+0.002
Neg logit contributions
arre-0.007
 courtroom-0.007
ravings-0.006
igsaw-0.006
wheel-0.006
Token O
Token ID440
Feature activation+1.55e+00

Pos logit contributions
adies+0.005
 lantern+0.004
 ho+0.004
Mom+0.004
 carriers+0.004
Neg logit contributions
arre-0.006
 courtroom-0.005
igsaw-0.005
wheel-0.005
ravings-0.005
Tokenll
Token ID297
Feature activation+4.82e+00

Pos logit contributions
adies+0.065
 carriers+0.064
 lantern+0.060
 aliens+0.059
Mom+0.058
Neg logit contributions
arre-0.029
igsaw-0.022
ravings-0.021
wheel-0.020
 accidentally-0.020
Tokenie
Token ID494
Feature activation+5.12e-01

Pos logit contributions
adies+0.129
 carriers+0.128
 Amen+0.107
Run+0.107
 lantern+0.106
Neg logit contributions
arre-0.141
wheel-0.128
 accidentally-0.125
ravings-0.121
 having-0.120
Token!"
Token ID2474
Feature activation-1.07e+00

Pos logit contributions
adies+0.012
 carriers+0.011
 lantern+0.010
Mom+0.010
 ho+0.010
Neg logit contributions
arre-0.019
igsaw-0.017
ravings-0.016
 courtroom-0.016
wheel-0.016
Token his
Token ID465
Feature activation+1.02e+00

Pos logit contributions
arre+0.036
igsaw+0.032
ravings+0.031
 fault+0.030
 courtroom+0.030
Neg logit contributions
adies-0.029
 lantern-0.028
Mom-0.027
 carriers-0.026
 ho-0.026
Token mother
Token ID2802
Feature activation-1.70e-01

Pos logit contributions
adies+0.024
 carriers+0.018
works+0.018
le+0.017
 lantern+0.016
Neg logit contributions
arre-0.038
 fault-0.035
 accidentally-0.035
ravings-0.034
 courtroom-0.034
Token praised
Token ID15342
Feature activation+2.47e+00

Pos logit contributions
arre+0.005
igsaw+0.004
ravings+0.004
 fault+0.004
 courtroom+0.004
Neg logit contributions
adies-0.005
 carriers-0.005
 lantern-0.005
Mom-0.005
 ho-0.005
Token charming
Token ID23332
Feature activation+2.61e-01

Pos logit contributions
adies+0.035
 carriers+0.031
le+0.030
works+0.030
el+0.030
Neg logit contributions
 accidentally-0.023
arre-0.023
 basketball-0.022
 court-0.022
 winning-0.021
Token oct
Token ID19318
Feature activation+1.68e+00

Pos logit contributions
adies+0.007
 carriers+0.006
Mom+0.006
works+0.006
 lantern+0.006
Neg logit contributions
arre-0.008
igsaw-0.008
 court-0.008
 accidentally-0.008
 courtroom-0.008
Tokenopus
Token ID25790
Feature activation+1.15e+00

Pos logit contributions
adies+0.057
 carriers+0.053
 lantern+0.051
Mom+0.050
 ho+0.049
Neg logit contributions
arre-0.043
igsaw-0.038
 court-0.038
wheel-0.037
 courtroom-0.037
Token named
Token ID3706
Feature activation-7.28e-02

Pos logit contributions
adies+0.033
 carriers+0.029
 lantern+0.029
Mom+0.027
 ho+0.027
Neg logit contributions
arre-0.037
igsaw-0.032
ravings-0.031
 courtroom-0.031
 fault-0.031
Token O
Token ID440
Feature activation+1.67e+00

Pos logit contributions
arre+0.002
igsaw+0.001
 court+0.001
 courtroom+0.001
 winning+0.001
Neg logit contributions
adies-0.003
 carriers-0.002
 lantern-0.002
Mom-0.002
le-0.002
Tokenll
Token ID297
Feature activation+4.81e+00

Pos logit contributions
adies+0.065
 carriers+0.062
 lantern+0.059
Mom+0.057
 Amen+0.056
Neg logit contributions
arre-0.036
igsaw-0.029
 court-0.028
ravings-0.028
 accidentally-0.028
Tokenie
Token ID494
Feature activation+1.46e+00

Pos logit contributions
 carriers+0.125
adies+0.122
Run+0.104
works+0.104
Mom+0.102
Neg logit contributions
arre-0.148
 accidentally-0.139
 fault-0.134
 about-0.131
 having-0.129
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
adies+0.042
 carriers+0.037
 lantern+0.036
Mom+0.035
works+0.035
Neg logit contributions
arre-0.044
igsaw-0.039
ravings-0.039
 court-0.038
 courtroom-0.038
Token O
Token ID440
Feature activation+4.65e-01

Pos logit contributions
adies+0.000
 carriers+0.000
 lantern+0.000
Mom+0.000
 ho+0.000
Neg logit contributions
arre-0.000
igsaw-0.000
ravings-0.000
 courtroom-0.000
wheel-0.000
Tokenll
Token ID297
Feature activation+4.15e+00

Pos logit contributions
adies+0.017
 carriers+0.016
 lantern+0.016
Mom+0.015
 ho+0.015
Neg logit contributions
arre-0.011
igsaw-0.009
ravings-0.009
 courtroom-0.008
 court-0.008
Tokenie
Token ID494
Feature activation-1.77e-01

Pos logit contributions
adies+0.100
 carriers+0.099
Run+0.082
works+0.081
 lantern+0.079
Neg logit contributions
arre-0.142
 accidentally-0.126
 fault-0.125
igsaw-0.123
 about-0.122
Token and
Token ID290
Feature activation+1.99e+00

Pos logit contributions
adies+0.043
 carriers+0.040
 lantern+0.039
Mom+0.038
 ho+0.036
Neg logit contributions
arre-0.076
igsaw-0.069
ravings-0.068
 courtroom-0.067
 winning-0.066
Token boots
Token ID14412
Feature activation+2.32e-01

Pos logit contributions
adies+0.051
 carriers+0.041
works+0.041
 lantern+0.040
 Amen+0.040
Neg logit contributions
arre-0.065
 accidentally-0.061
 court-0.060
igsaw-0.058
wheel-0.057
Token and
Token ID290
Feature activation+1.61e+00

Pos logit contributions
adies+0.005
 carriers+0.004
 lantern+0.004
Mom+0.004
 ho+0.004
Neg logit contributions
arre-0.009
igsaw-0.009
 courtroom-0.008
ravings-0.008
 winning-0.008
Token ran
Token ID4966
Feature activation-6.43e-02

Pos logit contributions
adies+0.043
 carriers+0.037
 lantern+0.036
works+0.035
le+0.034
Neg logit contributions
arre-0.052
 accidentally-0.047
 court-0.046
igsaw-0.046
wheel-0.045
Token to
Token ID284
Feature activation+1.28e+00

Pos logit contributions
arre+0.002
ravings+0.002
igsaw+0.002
wheel+0.002
 fault+0.001
Neg logit contributions
adies-0.002
 carriers-0.002
 lantern-0.002
Mom-0.002
 ho-0.002
Token Mr
Token ID1770
Feature activation+4.80e+00

Pos logit contributions
adies+0.023
 lantern+0.023
Mom+0.021
 God+0.021
 ho+0.021
Neg logit contributions
arre-0.056
ravings-0.048
wheel-0.048
 fault-0.046
igsaw-0.046
Token.
Token ID13
Feature activation+9.41e-01

Pos logit contributions
adies+0.166
 lantern+0.142
Run+0.138
Ho+0.137
agles+0.135
Neg logit contributions
arre-0.115
 fault-0.110
ravings-0.104
 court-0.103
 accidentally-0.096
Token Lee
Token ID5741
Feature activation+2.81e-01

Pos logit contributions
adies+0.026
 carriers+0.023
 lantern+0.022
 ho+0.021
Mom+0.021
Neg logit contributions
arre-0.031
igsaw-0.027
wheel-0.027
ravings-0.026
 courtroom-0.026
Token's
Token ID338
Feature activation+3.57e-01

Pos logit contributions
adies+0.005
 carriers+0.005
 lantern+0.005
 ho+0.005
le+0.004
Neg logit contributions
arre-0.012
igsaw-0.011
 fault-0.010
ravings-0.010
wheel-0.010
Token shop
Token ID6128
Feature activation-2.31e-01

Pos logit contributions
 lantern+0.008
adies+0.008
 carriers+0.008
 ho+0.007
 sle+0.007
Neg logit contributions
arre-0.013
igsaw-0.012
wheel-0.012
 By-0.011
 accidentally-0.011
Token.
Token ID13
Feature activation+5.89e-01

Pos logit contributions
arre+0.008
igsaw+0.007
wheel+0.007
ravings+0.007
 courtroom+0.007
Neg logit contributions
adies-0.006
 carriers-0.005
 lantern-0.005
 ho-0.005
Mom-0.005
Token.
Token ID13
Feature activation-2.19e-02

Pos logit contributions
adies+0.014
 carriers+0.012
 lantern+0.012
Mom+0.012
 ho+0.011
Neg logit contributions
arre-0.019
igsaw-0.017
ravings-0.016
wheel-0.016
 courtroom-0.016
Token He
Token ID679
Feature activation+1.08e+00

Pos logit contributions
arre+0.001
wheel+0.001
ravings+0.001
 accidentally+0.001
 salon+0.001
Neg logit contributions
adies-0.001
Mom-0.001
 carriers-0.001
 Amen-0.001
 God-0.001
Token said
Token ID531
Feature activation+5.60e-01

Pos logit contributions
 lantern+0.023
adies+0.022
Mom+0.022
le+0.022
 ho+0.022
Neg logit contributions
igsaw-0.039
arre-0.038
ravings-0.035
 fault-0.035
 courtroom-0.035
Token,
Token ID11
Feature activation-1.55e+00

Pos logit contributions
adies+0.014
 lantern+0.013
 carriers+0.013
 God+0.012
 ho+0.012
Neg logit contributions
arre-0.019
igsaw-0.018
ravings-0.017
 fault-0.017
 courtroom-0.017
Token "
Token ID366
Feature activation-4.44e-01

Pos logit contributions
arre+0.046
 fault+0.044
igsaw+0.043
ravings+0.042
 courtroom+0.039
Neg logit contributions
adies-0.045
Mom-0.044
 lantern-0.043
 God-0.042
 Amen-0.042
Tokenif
Token ID361
Feature activation+4.75e+00

Pos logit contributions
arre+0.015
igsaw+0.014
ravings+0.013
 courtroom+0.013
 fault+0.013
Neg logit contributions
adies-0.012
 carriers-0.010
Mom-0.010
 lantern-0.010
Ho-0.009
Token people
Token ID661
Feature activation-2.98e-01

Pos logit contributions
adies+0.102
 carriers+0.087
Mom+0.084
 lantern+0.082
works+0.081
Neg logit contributions
arre-0.183
ravings-0.166
igsaw-0.163
 courtroom-0.161
 fault-0.160
Token hurt
Token ID5938
Feature activation-1.56e+00

Pos logit contributions
arre+0.012
igsaw+0.012
ravings+0.011
 fault+0.011
 courtroom+0.011
Neg logit contributions
adies-0.005
 ho-0.005
le-0.005
 lantern-0.005
works-0.004
Token the
Token ID262
Feature activation-3.43e-01

Pos logit contributions
arre+0.056
igsaw+0.051
ravings+0.050
wheel+0.050
 courtroom+0.049
Neg logit contributions
adies-0.037
 carriers-0.033
 lantern-0.032
Mom-0.031
 ho-0.030
Token universe
Token ID6881
Feature activation-1.10e+00

Pos logit contributions
arre+0.012
igsaw+0.011
ravings+0.010
 courtroom+0.010
 court+0.010
Neg logit contributions
adies-0.009
 carriers-0.008
 lantern-0.007
Mom-0.007
 ho-0.007
Token,
Token ID11
Feature activation-4.26e-01

Pos logit contributions
arre+0.041
ravings+0.037
 courtroom+0.037
igsaw+0.036
wheel+0.036
Neg logit contributions
adies-0.023
 lantern-0.021
 carriers-0.021
le-0.020
 ho-0.020
Token happy
Token ID3772
Feature activation+1.87e+00

Pos logit contributions
adies+0.035
 carriers+0.031
works+0.030
le+0.029
agles+0.029
Neg logit contributions
arre-0.021
 accidentally-0.020
 court-0.019
 basketball-0.019
 courtroom-0.019
Token oct
Token ID19318
Feature activation+1.34e+00

Pos logit contributions
adies+0.057
 carriers+0.050
works+0.048
le+0.046
 lantern+0.046
Neg logit contributions
arre-0.056
igsaw-0.050
 court-0.050
ravings-0.048
 courtroom-0.048
Tokenopus
Token ID25790
Feature activation+8.87e-01

Pos logit contributions
adies+0.046
 carriers+0.043
 lantern+0.042
Mom+0.041
 ho+0.040
Neg logit contributions
arre-0.033
igsaw-0.029
wheel-0.029
 court-0.029
 courtroom-0.028
Token named
Token ID3706
Feature activation-2.58e-01

Pos logit contributions
adies+0.026
 carriers+0.023
 lantern+0.023
Mom+0.022
 ho+0.022
Neg logit contributions
arre-0.027
igsaw-0.024
ravings-0.023
 courtroom-0.023
 fault-0.023
Token O
Token ID440
Feature activation+1.37e+00

Pos logit contributions
arre+0.006
igsaw+0.005
ravings+0.005
 courtroom+0.005
wheel+0.005
Neg logit contributions
adies-0.010
 carriers-0.009
 lantern-0.009
Mom-0.009
 ho-0.008
Tokenll
Token ID297
Feature activation+4.74e+00

Pos logit contributions
adies+0.054
 carriers+0.050
 lantern+0.049
Mom+0.047
 ho+0.047
Neg logit contributions
arre-0.029
igsaw-0.023
ravings-0.023
 court-0.022
 courtroom-0.022
Tokenie
Token ID494
Feature activation+1.42e+00

Pos logit contributions
adies+0.115
 carriers+0.110
 lantern+0.094
Mom+0.093
works+0.093
Neg logit contributions
arre-0.159
 accidentally-0.145
 fault-0.138
ravings-0.137
 having-0.135
Token.
Token ID13
Feature activation+2.07e-01

Pos logit contributions
adies+0.038
 carriers+0.033
 lantern+0.033
Mom+0.031
 ho+0.031
Neg logit contributions
arre-0.047
igsaw-0.042
ravings-0.041
 courtroom-0.041
wheel-0.040
Token O
Token ID440
Feature activation+4.54e-01

Pos logit contributions
adies+0.007
 carriers+0.006
 lantern+0.006
Mom+0.006
le+0.006
Neg logit contributions
arre-0.006
igsaw-0.005
ravings-0.005
 courtroom-0.005
 winning-0.005
Tokenll
Token ID297
Feature activation+4.31e+00

Pos logit contributions
adies+0.017
 carriers+0.015
 lantern+0.015
Mom+0.015
 ho+0.014
Neg logit contributions
arre-0.010
igsaw-0.009
ravings-0.008
 courtroom-0.008
wheel-0.008
Tokenie
Token ID494
Feature activation-4.84e-02

Pos logit contributions
adies+0.103
 carriers+0.098
works+0.082
Run+0.081
 lantern+0.081
Neg logit contributions
arre-0.152
 accidentally-0.134
 fault-0.132
igsaw-0.131
ravings-0.130
Token and
Token ID290
Feature activation+4.77e-02

Pos logit contributions
adies+0.005
 lantern+0.005
 carriers+0.005
 ho+0.005
le+0.004
Neg logit contributions
arre-0.008
ravings-0.008
igsaw-0.008
 courtroom-0.007
 winning-0.007
Token brought
Token ID3181
Feature activation+5.47e-01

Pos logit contributions
 lantern+0.001
 ho+0.001
adies+0.001
 carriers+0.001
Mom+0.001
Neg logit contributions
arre-0.002
igsaw-0.002
 fault-0.002
ravings-0.001
 winning-0.001
Token it
Token ID340
Feature activation+9.40e-01

Pos logit contributions
adies+0.016
 carriers+0.014
works+0.014
 lantern+0.013
ilot+0.013
Neg logit contributions
arre-0.016
igsaw-0.014
 accidentally-0.014
 court-0.014
 courtroom-0.014
Token back
Token ID736
Feature activation+8.55e-01

Pos logit contributions
adies+0.026
 carriers+0.023
 lantern+0.023
Mom+0.021
 ho+0.021
Neg logit contributions
arre-0.032
igsaw-0.028
 courtroom-0.028
ravings-0.027
 winning-0.026
Token to
Token ID284
Feature activation-1.39e-01

Pos logit contributions
adies+0.026
 lantern+0.024
 ho+0.023
 carriers+0.023
Mom+0.023
Neg logit contributions
arre-0.027
igsaw-0.023
ravings-0.022
 courtroom-0.022
wheel-0.022
Token Mr
Token ID1770
Feature activation+4.74e+00

Pos logit contributions
arre+0.005
igsaw+0.005
 fault+0.005
wheel+0.004
ravings+0.004
Neg logit contributions
adies-0.003
 God-0.003
 ho-0.003
 lantern-0.003
Mom-0.003
Token.
Token ID13
Feature activation+3.57e-01

Pos logit contributions
adies+0.136
 carriers+0.118
 lantern+0.117
Mom+0.113
 ho+0.110
Neg logit contributions
arre-0.147
igsaw-0.131
ravings-0.129
 court-0.128
 fault-0.128
Token Johnson
Token ID5030
Feature activation-5.43e-01

Pos logit contributions
adies+0.009
Mom+0.009
 carriers+0.009
 lantern+0.009
 ho+0.008
Neg logit contributions
arre-0.012
wheel-0.011
igsaw-0.010
ravings-0.010
Em-0.009
Token.
Token ID13
Feature activation+2.04e-01

Pos logit contributions
arre+0.025
igsaw+0.023
 courtroom+0.022
ravings+0.022
 fault+0.021
Neg logit contributions
 ho-0.008
 lantern-0.008
adies-0.008
le-0.007
 carriers-0.007
Token He
Token ID679
Feature activation+1.07e+00

Pos logit contributions
Mom+0.004
 Amen+0.004
 God+0.004
 sle+0.003
Dad+0.003
Neg logit contributions
arre-0.008
wheel-0.008
igsaw-0.007
 fault-0.007
 supposed-0.007
Token was
Token ID373
Feature activation+1.46e-01

Pos logit contributions
adies+0.026
 lantern+0.026
 ho+0.025
 carriers+0.025
Mom+0.024
Neg logit contributions
arre-0.038
igsaw-0.037
 fault-0.032
 courtroom-0.032
 winning-0.032
Token healthy
Token ID5448
Feature activation+1.35e+00

Pos logit contributions
adies+0.032
 carriers+0.028
le+0.027
works+0.027
 lantern+0.026
Neg logit contributions
arre-0.026
 accidentally-0.023
 court-0.023
 courtroom-0.022
 winning-0.022
Token ot
Token ID30972
Feature activation+1.75e+00

Pos logit contributions
adies+0.033
 lantern+0.029
 carriers+0.029
Mom+0.027
 ho+0.027
Neg logit contributions
arre-0.048
igsaw-0.043
wheel-0.041
ravings-0.041
 fault-0.041
Tokenter
Token ID353
Feature activation+2.47e-01

Pos logit contributions
adies+0.045
 lantern+0.040
 carriers+0.039
Mom+0.037
 Amen+0.036
Neg logit contributions
arre-0.059
wheel-0.052
 court-0.051
ravings-0.051
 winning-0.050
Token named
Token ID3706
Feature activation-2.70e-01

Pos logit contributions
adies+0.007
 lantern+0.006
 carriers+0.006
Mom+0.006
 ho+0.006
Neg logit contributions
arre-0.008
igsaw-0.008
ravings-0.007
 fault-0.007
 courtroom-0.007
Token O
Token ID440
Feature activation+1.64e+00

Pos logit contributions
arre+0.005
igsaw+0.004
 court+0.004
 winning+0.004
 accidentally+0.004
Neg logit contributions
adies-0.011
 carriers-0.010
 lantern-0.010
le-0.009
Mom-0.009
Tokenll
Token ID297
Feature activation+4.72e+00

Pos logit contributions
adies+0.061
 carriers+0.056
 lantern+0.055
Mom+0.053
 ho+0.052
Neg logit contributions
arre-0.038
igsaw-0.031
ravings-0.030
 courtroom-0.030
 court-0.029
Tokenie
Token ID494
Feature activation+1.39e+00

Pos logit contributions
adies+0.103
 carriers+0.096
 lantern+0.083
works+0.081
Mom+0.081
Neg logit contributions
arre-0.171
 accidentally-0.156
 fault-0.153
ravings-0.151
 court-0.150
Token.
Token ID13
Feature activation-7.14e-03

Pos logit contributions
adies+0.038
 carriers+0.034
 lantern+0.033
Mom+0.031
 ho+0.031
Neg logit contributions
arre-0.045
igsaw-0.040
ravings-0.039
 courtroom-0.039
 court-0.039
Token O
Token ID440
Feature activation+4.85e-01

Pos logit contributions
arre+0.000
igsaw+0.000
ravings+0.000
 courtroom+0.000
wheel+0.000
Neg logit contributions
adies-0.000
 carriers-0.000
 lantern-0.000
Mom-0.000
 ho-0.000
Tokenll
Token ID297
Feature activation+4.07e+00

Pos logit contributions
adies+0.017
 lantern+0.015
 carriers+0.015
Mom+0.015
le+0.015
Neg logit contributions
arre-0.012
igsaw-0.010
 courtroom-0.010
ravings-0.010
wheel-0.010
Tokenie
Token ID494
Feature activation-1.39e-01

Pos logit contributions
adies+0.085
 carriers+0.075
 lantern+0.065
works+0.063
Mom+0.062
Neg logit contributions
arre-0.157
 accidentally-0.140
 fault-0.139
ravings-0.139
igsaw-0.138
Token furry
Token ID40857
Feature activation+1.06e+00

Pos logit contributions
adies+0.033
 carriers+0.029
works+0.028
le+0.028
agles+0.027
Neg logit contributions
arre-0.023
 accidentally-0.021
 court-0.020
 basketball-0.020
 courtroom-0.020
Token oct
Token ID19318
Feature activation+1.55e+00

Pos logit contributions
adies+0.031
 carriers+0.028
 lantern+0.026
Mom+0.026
works+0.026
Neg logit contributions
arre-0.032
igsaw-0.028
 court-0.028
ravings-0.028
 accidentally-0.028
Tokenopus
Token ID25790
Feature activation+9.98e-01

Pos logit contributions
adies+0.053
 carriers+0.050
 lantern+0.048
Mom+0.047
works+0.045
Neg logit contributions
arre-0.039
igsaw-0.035
 court-0.034
wheel-0.034
 courtroom-0.033
Token named
Token ID3706
Feature activation-1.30e-01

Pos logit contributions
adies+0.032
 carriers+0.029
 lantern+0.028
Mom+0.027
 ho+0.027
Neg logit contributions
arre-0.028
igsaw-0.024
ravings-0.023
 courtroom-0.023
 court-0.023
Token O
Token ID440
Feature activation+1.38e+00

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 courtroom+0.002
 court+0.002
Neg logit contributions
adies-0.005
 carriers-0.004
 lantern-0.004
Mom-0.004
le-0.004
Tokenll
Token ID297
Feature activation+4.72e+00

Pos logit contributions
adies+0.053
 carriers+0.050
 lantern+0.048
Mom+0.047
 ho+0.046
Neg logit contributions
arre-0.030
igsaw-0.024
ravings-0.024
 court-0.024
 courtroom-0.023
Tokenie
Token ID494
Feature activation+1.42e+00

Pos logit contributions
 carriers+0.128
adies+0.126
works+0.108
Run+0.106
Mom+0.106
Neg logit contributions
arre-0.137
 accidentally-0.134
 about-0.121
 fault-0.120
 having-0.120
Token.
Token ID13
Feature activation+1.31e-02

Pos logit contributions
adies+0.042
 carriers+0.038
 lantern+0.036
works+0.035
Mom+0.035
Neg logit contributions
arre-0.042
ravings-0.036
 court-0.036
igsaw-0.036
 accidentally-0.036
Token O
Token ID440
Feature activation+3.68e-01

Pos logit contributions
adies+0.000
 carriers+0.000
 lantern+0.000
Mom+0.000
le+0.000
Neg logit contributions
arre-0.000
igsaw-0.000
ravings-0.000
 winning-0.000
 courtroom-0.000
Tokenll
Token ID297
Feature activation+4.17e+00

Pos logit contributions
adies+0.013
 carriers+0.012
 lantern+0.012
Mom+0.012
 ho+0.012
Neg logit contributions
arre-0.009
igsaw-0.007
ravings-0.007
 courtroom-0.007
wheel-0.007
Tokenie
Token ID494
Feature activation-4.05e-02

Pos logit contributions
adies+0.103
 carriers+0.101
Run+0.085
works+0.083
 Amen+0.081
Neg logit contributions
arre-0.141
 accidentally-0.127
 fault-0.123
ravings-0.121
igsaw-0.120
Token saved
Token ID7448
Feature activation+1.37e+00

Pos logit contributions
adies+0.028
 carriers+0.025
 lantern+0.025
 ho+0.024
Mom+0.023
Neg logit contributions
arre-0.033
igsaw-0.031
 salon-0.030
 fault-0.030
 courtroom-0.029
Token!
Token ID0
Feature activation-1.90e-01

Pos logit contributions
adies+0.038
 carriers+0.035
 lantern+0.034
 ho+0.033
Mom+0.032
Neg logit contributions
arre-0.044
igsaw-0.039
ravings-0.038
 courtroom-0.038
wheel-0.037
Token He
Token ID679
Feature activation+1.41e+00

Pos logit contributions
arre+0.008
 salon+0.008
 courtroom+0.007
 fault+0.007
 supposed+0.007
Neg logit contributions
adies-0.002
bye-0.002
Mom-0.002
Ho-0.002
 God-0.002
Token thanked
Token ID26280
Feature activation+2.76e+00

Pos logit contributions
adies+0.032
 lantern+0.032
 ho+0.032
Mom+0.030
Ho+0.030
Neg logit contributions
igsaw-0.049
arre-0.049
 fault-0.046
 courtroom-0.044
ravings-0.043
Token O
Token ID440
Feature activation+1.59e+00

Pos logit contributions
adies+0.088
 lantern+0.079
 carriers+0.079
Mom+0.077
 ho+0.075
Neg logit contributions
arre-0.080
igsaw-0.068
ravings-0.066
 courtroom-0.065
 fault-0.064
Tokenll
Token ID297
Feature activation+4.71e+00

Pos logit contributions
adies+0.070
 carriers+0.067
 lantern+0.065
 Amen+0.063
 aliens+0.063
Neg logit contributions
arre-0.029
ravings-0.020
igsaw-0.020
wheel-0.019
 accidentally-0.018
Tokenie
Token ID494
Feature activation+7.99e-01

Pos logit contributions
adies+0.107
 carriers+0.102
works+0.091
ilot+0.086
 lantern+0.086
Neg logit contributions
arre-0.159
wheel-0.149
igsaw-0.145
 court-0.145
ravings-0.143
Token and
Token ID290
Feature activation+5.57e-01

Pos logit contributions
adies+0.025
 carriers+0.023
 lantern+0.023
Mom+0.022
 ho+0.022
Neg logit contributions
arre-0.023
igsaw-0.020
 courtroom-0.019
ravings-0.019
 fault-0.018
Token went
Token ID1816
Feature activation-8.33e-01

Pos logit contributions
adies+0.014
 carriers+0.013
 lantern+0.013
Mom+0.012
 ho+0.012
Neg logit contributions
arre-0.019
igsaw-0.017
ravings-0.017
 fault-0.017
 courtroom-0.017
Token back
Token ID736
Feature activation+1.36e+00

Pos logit contributions
arre+0.028
igsaw+0.025
 fault+0.024
ravings+0.024
 courtroom+0.024
Neg logit contributions
adies-0.023
 lantern-0.020
 carriers-0.020
 ho-0.019
Mom-0.019
Token to
Token ID284
Feature activation+7.77e-01

Pos logit contributions
adies+0.042
 lantern+0.039
 ho+0.037
 carriers+0.037
 sle+0.036
Neg logit contributions
arre-0.041
igsaw-0.036
ravings-0.034
wheel-0.034
 fault-0.034
Token little
Token ID1310
Feature activation+9.14e-01

Pos logit contributions
adies+0.034
 carriers+0.030
le+0.028
 lantern+0.028
works+0.028
Neg logit contributions
arre-0.029
 accidentally-0.026
igsaw-0.026
 court-0.026
 courtroom-0.025
Token ot
Token ID30972
Feature activation+1.50e+00

Pos logit contributions
adies+0.028
 carriers+0.023
works+0.021
agles+0.021
Mom+0.021
Neg logit contributions
arre-0.027
 basketball-0.026
igsaw-0.025
 court-0.025
 accidentally-0.025
Tokenter
Token ID353
Feature activation+2.57e-02

Pos logit contributions
adies+0.041
 lantern+0.036
 carriers+0.036
Mom+0.034
 Amen+0.034
Neg logit contributions
arre-0.047
wheel-0.042
 court-0.041
 winning-0.041
 accidentally-0.040
Token named
Token ID3706
Feature activation-3.81e-01

Pos logit contributions
adies+0.001
 carriers+0.001
 lantern+0.001
Mom+0.001
 ho+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 courtroom-0.001
 court-0.001
Token O
Token ID440
Feature activation+1.33e+00

Pos logit contributions
arre+0.006
igsaw+0.005
 winning+0.005
 accidentally+0.005
 court+0.005
Neg logit contributions
adies-0.016
 carriers-0.015
 lantern-0.014
le-0.014
Mom-0.014
Tokenll
Token ID297
Feature activation+4.68e+00

Pos logit contributions
adies+0.050
 carriers+0.046
 lantern+0.045
Mom+0.044
 ho+0.043
Neg logit contributions
arre-0.030
igsaw-0.025
ravings-0.024
 courtroom-0.023
 court-0.023
Tokenie
Token ID494
Feature activation+1.17e+00

Pos logit contributions
adies+0.113
 carriers+0.109
works+0.092
Mom+0.090
 lantern+0.090
Neg logit contributions
arre-0.156
 accidentally-0.144
 fault-0.138
 having-0.135
ravings-0.134
Token.
Token ID13
Feature activation-2.09e-01

Pos logit contributions
adies+0.033
 carriers+0.029
 lantern+0.028
Mom+0.027
works+0.026
Neg logit contributions
arre-0.038
igsaw-0.033
ravings-0.033
 courtroom-0.032
 court-0.032
Token O
Token ID440
Feature activation+2.62e-01

Pos logit contributions
arre+0.006
igsaw+0.005
ravings+0.005
 courtroom+0.005
wheel+0.005
Neg logit contributions
adies-0.007
 carriers-0.006
 lantern-0.006
Mom-0.006
 ho-0.006
Tokenll
Token ID297
Feature activation+3.99e+00

Pos logit contributions
adies+0.009
 lantern+0.008
 carriers+0.008
Mom+0.008
le+0.008
Neg logit contributions
arre-0.006
igsaw-0.005
 courtroom-0.005
ravings-0.005
wheel-0.005
Tokenie
Token ID494
Feature activation-2.93e-01

Pos logit contributions
adies+0.092
 carriers+0.084
works+0.071
 lantern+0.070
Run+0.069
Neg logit contributions
arre-0.145
 accidentally-0.130
 fault-0.128
ravings-0.126
igsaw-0.123
Token a
Token ID257
Feature activation-1.11e-01

Pos logit contributions
adies+0.012
 carriers+0.010
 lantern+0.010
Mom+0.010
 ho+0.010
Neg logit contributions
arre-0.011
igsaw-0.010
ravings-0.009
 courtroom-0.009
wheel-0.009
Token good
Token ID922
Feature activation+1.62e-01

Pos logit contributions
arre+0.003
igsaw+0.003
 courtroom+0.003
 winning+0.003
ravings+0.003
Neg logit contributions
adies-0.004
 carriers-0.003
 lantern-0.003
le-0.003
works-0.003
Token friend
Token ID1545
Feature activation+5.21e-01

Pos logit contributions
adies+0.006
 carriers+0.005
 lantern+0.005
works+0.005
Mom+0.005
Neg logit contributions
arre-0.004
igsaw-0.004
ravings-0.004
 fault-0.004
 accidentally-0.004
Token,
Token ID11
Feature activation+3.69e-01

Pos logit contributions
adies+0.014
 carriers+0.012
 lantern+0.011
Mom+0.011
 ho+0.011
Neg logit contributions
arre-0.018
igsaw-0.016
 courtroom-0.016
ravings-0.016
 court-0.015
Token O
Token ID440
Feature activation+1.29e+00

Pos logit contributions
adies+0.010
 ho+0.010
 lantern+0.010
Mom+0.010
 carriers+0.009
Neg logit contributions
arre-0.012
 courtroom-0.010
ravings-0.010
seek-0.010
igsaw-0.010
Tokenll
Token ID297
Feature activation+4.68e+00

Pos logit contributions
adies+0.052
 carriers+0.050
 lantern+0.048
Mom+0.046
 sle+0.045
Neg logit contributions
arre-0.026
igsaw-0.021
ravings-0.020
wheel-0.019
 accidentally-0.019
Tokenie
Token ID494
Feature activation+6.93e-01

Pos logit contributions
 carriers+0.118
adies+0.117
 lantern+0.099
Run+0.098
works+0.097
Neg logit contributions
arre-0.145
 accidentally-0.133
wheel-0.132
 winning-0.129
ravings-0.129
Token."
Token ID526
Feature activation-9.78e-01

Pos logit contributions
adies+0.016
 carriers+0.014
 lantern+0.014
 ho+0.013
Mom+0.013
Neg logit contributions
arre-0.026
igsaw-0.023
 courtroom-0.023
ravings-0.023
wheel-0.022
Token O
Token ID440
Feature activation+4.12e-01

Pos logit contributions
arre+0.033
igsaw+0.031
 courtroom+0.031
ravings+0.029
 fault+0.028
Neg logit contributions
adies-0.024
Mom-0.023
 lantern-0.023
 ho-0.023
el-0.022
Tokenll
Token ID297
Feature activation+3.93e+00

Pos logit contributions
adies+0.013
 lantern+0.012
Mom+0.012
el+0.011
le+0.011
Neg logit contributions
arre-0.011
 courtroom-0.010
igsaw-0.010
 court-0.010
ravings-0.009
Tokenie
Token ID494
Feature activation-8.44e-01

Pos logit contributions
adies+0.088
 carriers+0.082
 lantern+0.072
works+0.069
Mom+0.066
Neg logit contributions
arre-0.140
 accidentally-0.125
igsaw-0.124
wheel-0.123
ravings-0.122
Token make
Token ID787
Feature activation+2.65e+00

Pos logit contributions
Mom+0.019
 carriers+0.018
 lantern+0.018
adies+0.018
 ho+0.018
Neg logit contributions
arre-0.039
igsaw-0.033
 colony-0.032
ravings-0.032
 fault-0.032
Token it
Token ID340
Feature activation+1.13e+00

Pos logit contributions
adies+0.050
Mom+0.046
 carriers+0.046
 lantern+0.045
el+0.041
Neg logit contributions
arre-0.106
igsaw-0.100
ravings-0.095
 courtroom-0.095
 court-0.091
Token whole
Token ID2187
Feature activation-9.89e-01

Pos logit contributions
 lantern+0.026
adies+0.026
 carriers+0.025
 ho+0.023
le+0.023
Neg logit contributions
arre-0.040
 courtroom-0.038
igsaw-0.036
ravings-0.035
 court-0.034
Token again
Token ID757
Feature activation+4.81e-01

Pos logit contributions
arre+0.036
igsaw+0.035
 courtroom+0.034
ravings+0.033
 court+0.032
Neg logit contributions
adies-0.020
 carriers-0.018
 lantern-0.018
Mom-0.018
le-0.017
Token.
Token ID13
Feature activation+6.46e-01

Pos logit contributions
adies+0.009
Mom+0.008
 lantern+0.008
 carriers+0.008
le+0.007
Neg logit contributions
arre-0.020
igsaw-0.019
 courtroom-0.018
wheel-0.017
ravings-0.017
Token Amen
Token ID34717
Feature activation+4.67e+00

Pos logit contributions
adies+0.019
 carriers+0.017
Mom+0.017
 lantern+0.017
 Amen+0.016
Neg logit contributions
arre-0.020
igsaw-0.017
ravings-0.017
 courtroom-0.017
wheel-0.017
Token."
Token ID526
Feature activation+1.33e-02

Pos logit contributions
adies+0.123
 carriers+0.107
 lantern+0.104
Mom+0.102
 ho+0.098
Neg logit contributions
arre-0.159
igsaw-0.145
ravings-0.138
 courtroom-0.138
wheel-0.137
Token
Token ID198
Feature activation+3.96e-01

Pos logit contributions
Mom+0.000
Dad+0.000
adies+0.000
 God+0.000
 lantern+0.000
Neg logit contributions
arre-0.000
igsaw-0.000
wheel-0.000
ravings-0.000
 supposed-0.000
Token
Token ID198
Feature activation+9.72e-01

Pos logit contributions
adies+0.014
Mom+0.013
 carriers+0.012
 lantern+0.012
Dad+0.012
Neg logit contributions
arre-0.010
igsaw-0.009
ravings-0.009
 fault-0.008
wheel-0.008
TokenThen
Token ID6423
Feature activation+6.07e-01

Pos logit contributions
adies+0.023
 carriers+0.020
 lantern+0.019
Mom+0.019
 ho+0.018
Neg logit contributions
arre-0.036
igsaw-0.032
ravings-0.032
 courtroom-0.031
 fault-0.031
Token,
Token ID11
Feature activation-2.07e-01

Pos logit contributions
adies+0.017
 carriers+0.015
 lantern+0.015
Mom+0.014
le+0.014
Neg logit contributions
arre-0.019
igsaw-0.017
 accidentally-0.016
 court-0.016
 courtroom-0.016
Token
Token ID198
Feature activation+1.01e+00

Pos logit contributions
adies+0.017
 carriers+0.015
Mom+0.015
 lantern+0.015
Ho+0.014
Neg logit contributions
arre-0.012
igsaw-0.010
ravings-0.010
wheel-0.009
 courtroom-0.009
Token"
Token ID1
Feature activation-1.69e-01

Pos logit contributions
adies+0.025
 lantern+0.021
 carriers+0.021
 ho+0.020
le+0.019
Neg logit contributions
arre-0.036
igsaw-0.032
 courtroom-0.031
 court-0.031
ravings-0.031
TokenOK
Token ID11380
Feature activation+3.30e+00

Pos logit contributions
arre+0.006
igsaw+0.005
wheel+0.005
 court+0.005
 winning+0.005
Neg logit contributions
adies-0.004
 lantern-0.004
 carriers-0.004
le-0.004
 ho-0.004
Token,
Token ID11
Feature activation-2.29e-01

Pos logit contributions
adies+0.094
 carriers+0.083
 lantern+0.081
Mom+0.077
Ho+0.076
Neg logit contributions
arre-0.102
igsaw-0.090
 court-0.088
 accidentally-0.087
 winning-0.087
Token O
Token ID440
Feature activation+1.58e+00

Pos logit contributions
arre+0.007
wheel+0.006
 winning+0.006
igsaw+0.006
 court+0.006
Neg logit contributions
adies-0.006
 carriers-0.005
 lantern-0.005
le-0.005
Mom-0.005
Tokenll
Token ID297
Feature activation+4.65e+00

Pos logit contributions
adies+0.049
 carriers+0.047
 lantern+0.045
Mom+0.044
 aliens+0.043
Neg logit contributions
arre-0.045
igsaw-0.039
wheel-0.037
 court-0.037
ravings-0.037
Tokenie
Token ID494
Feature activation+3.27e-01

Pos logit contributions
adies+0.114
 carriers+0.113
works+0.098
Run+0.097
 lantern+0.095
Neg logit contributions
arre-0.139
wheel-0.129
 done-0.129
 court-0.126
 accidentally-0.126
Token,
Token ID11
Feature activation-1.29e-01

Pos logit contributions
adies+0.008
 carriers+0.007
 lantern+0.007
Mom+0.007
 ho+0.007
Neg logit contributions
arre-0.011
igsaw-0.010
ravings-0.010
 courtroom-0.010
wheel-0.010
Token show
Token ID905
Feature activation-4.10e-02

Pos logit contributions
arre+0.004
 winning+0.004
wheel+0.004
igsaw+0.004
 court+0.004
Neg logit contributions
adies-0.003
 carriers-0.003
 lantern-0.003
Mom-0.003
le-0.003
Token them
Token ID606
Feature activation-1.41e-01

Pos logit contributions
arre+0.001
 court+0.001
igsaw+0.001
 winning+0.001
 courtroom+0.001
Neg logit contributions
adies-0.001
 carriers-0.001
 lantern-0.001
Mom-0.001
Ho-0.001
Token what
Token ID644
Feature activation+8.76e-01

Pos logit contributions
arre+0.005
 courtroom+0.004
igsaw+0.004
ravings+0.004
wheel+0.004
Neg logit contributions
adies-0.004
 carriers-0.003
 lantern-0.003
 ho-0.003
works-0.003
Token the
Token ID262
Feature activation-1.69e-01

Pos logit contributions
arre+0.014
ravings+0.012
igsaw+0.012
wheel+0.012
 courtroom+0.012
Neg logit contributions
adies-0.010
 lantern-0.009
 carriers-0.009
le-0.009
 ho-0.009
Token tree
Token ID5509
Feature activation-1.20e+00

Pos logit contributions
arre+0.007
igsaw+0.006
wheel+0.006
 Working+0.005
Em+0.005
Neg logit contributions
adies-0.004
 lantern-0.004
 ho-0.003
 comet-0.003
 sle-0.003
Token to
Token ID284
Feature activation-3.28e-01

Pos logit contributions
arre+0.047
ravings+0.041
 courtroom+0.041
Em+0.040
 gol+0.040
Neg logit contributions
 lantern-0.023
 ho-0.021
 carriers-0.021
el-0.019
le-0.019
Token its
Token ID663
Feature activation-2.50e-01

Pos logit contributions
arre+0.013
ravings+0.011
 fault+0.011
igsaw+0.011
 courtroom+0.011
Neg logit contributions
 lantern-0.007
adies-0.006
 God-0.006
 ho-0.006
 carriers-0.006
Token once
Token ID1752
Feature activation+1.94e-02

Pos logit contributions
arre+0.009
wheel+0.008
igsaw+0.008
 Working+0.007
Em+0.007
Neg logit contributions
 lantern-0.006
 ho-0.006
 sle-0.006
 carriers-0.006
adies-0.006
Token former
Token ID1966
Feature activation-6.22e+00

Pos logit contributions
 lantern+0.000
adies+0.000
le+0.000
 carriers+0.000
 ho+0.000
Neg logit contributions
arre-0.001
 courtroom-0.001
ravings-0.001
igsaw-0.001
 court-0.001
Token beauty
Token ID8737
Feature activation-5.16e-01

Pos logit contributions
arre+0.202
wheel+0.189
 accidentally+0.176
 basketball+0.169
 court+0.168
Neg logit contributions
adies-0.166
 lantern-0.156
 carriers-0.149
 ho-0.147
 sle-0.144
Token.
Token ID13
Feature activation+7.46e-02

Pos logit contributions
arre+0.022
 courtroom+0.020
ravings+0.019
 accidentally+0.019
igsaw+0.019
Neg logit contributions
adies-0.008
le-0.008
 carriers-0.008
 lantern-0.008
works-0.008
Token He
Token ID679
Feature activation+8.40e-01

Pos logit contributions
adies+0.002
Mom+0.002
Dad+0.002
 God+0.001
 Amen+0.001
Neg logit contributions
arre-0.003
wheel-0.002
ravings-0.002
 accidentally-0.002
 salon-0.002
Token was
Token ID373
Feature activation-1.67e-01

Pos logit contributions
 lantern+0.017
adies+0.017
 ho+0.017
le+0.016
Mom+0.016
Neg logit contributions
arre-0.032
igsaw-0.031
 fault-0.029
ravings-0.028
 courtroom-0.028
Token happy
Token ID3772
Feature activation+2.29e-01

Pos logit contributions
arre+0.006
igsaw+0.006
ravings+0.005
con+0.005
 court+0.005
Neg logit contributions
 lantern-0.004
 Amen-0.004
 ho-0.004
 close-0.004
Mom-0.004
Token from
Token ID422
Feature activation-1.24e-01

Pos logit contributions
arre+0.002
igsaw+0.002
 courtroom+0.002
ravings+0.002
wheel+0.002
Neg logit contributions
adies-0.001
 ho-0.001
 lantern-0.001
 carriers-0.001
le-0.001
Token her
Token ID607
Feature activation-6.74e-01

Pos logit contributions
arre+0.004
igsaw+0.003
ravings+0.003
wheel+0.003
 courtroom+0.003
Neg logit contributions
adies-0.004
 lantern-0.003
 carriers-0.003
 ho-0.003
Mom-0.003
Token,
Token ID11
Feature activation-1.38e+00

Pos logit contributions
arre+0.028
wheel+0.023
igsaw+0.023
 Working+0.023
ravings+0.023
Neg logit contributions
 ho-0.014
 lantern-0.014
adies-0.013
 sle-0.013
 carriers-0.012
Token but
Token ID475
Feature activation+4.33e-01

Pos logit contributions
arre+0.051
igsaw+0.049
ravings+0.044
 courtroom+0.042
 Working+0.042
Neg logit contributions
 ho-0.032
adies-0.031
 lantern-0.030
Mom-0.028
 sle-0.026
Token Kay
Token ID17356
Feature activation-2.17e+00

Pos logit contributions
adies+0.009
 ho+0.008
Mom+0.008
le+0.008
 lantern+0.008
Neg logit contributions
arre-0.018
igsaw-0.016
ravings-0.016
 fault-0.015
 courtroom-0.015
Tokenlee
Token ID7197
Feature activation-5.92e+00

Pos logit contributions
arre+0.089
igsaw+0.082
ravings+0.079
 winning+0.077
 court+0.077
Neg logit contributions
adies-0.044
le-0.036
 lantern-0.034
el-0.034
Mom-0.034
Token was
Token ID373
Feature activation-1.38e+00

Pos logit contributions
igsaw+0.225
arre+0.214
ravings+0.191
 courtroom+0.189
 fault+0.188
Neg logit contributions
adies-0.127
le-0.123
 ho-0.119
Mom-0.114
el-0.105
Token determined
Token ID5295
Feature activation-1.64e-01

Pos logit contributions
igsaw+0.057
arre+0.055
ravings+0.050
 disl+0.047
con+0.047
Neg logit contributions
 closer-0.029
 close-0.029
adies-0.028
 ho-0.026
Mom-0.024
Token to
Token ID284
Feature activation-1.46e+00

Pos logit contributions
arre+0.006
igsaw+0.006
ravings+0.005
 courtroom+0.005
Em+0.005
Neg logit contributions
adies-0.003
 ho-0.003
Mom-0.003
 lantern-0.003
le-0.003
Token catch
Token ID4929
Feature activation-5.71e-01

Pos logit contributions
arre+0.047
igsaw+0.042
ravings+0.040
 courtroom+0.040
 fault+0.039
Neg logit contributions
adies-0.041
 carriers-0.037
 lantern-0.036
Mom-0.035
 ho-0.034
Token it
Token ID340
Feature activation-3.05e-02

Pos logit contributions
arre+0.023
ravings+0.021
igsaw+0.021
 courtroom+0.020
 fault+0.020
Neg logit contributions
adies-0.011
le-0.010
 lantern-0.009
 carriers-0.009
Mom-0.009
Tokenmy
Token ID1820
Feature activation+2.80e-01

Pos logit contributions
adies+0.013
le+0.013
 lantern+0.013
el+0.012
 ho+0.012
Neg logit contributions
arre-0.025
igsaw-0.024
ravings-0.022
 fault-0.022
 courtroom-0.021
Token had
Token ID550
Feature activation-1.72e+00

Pos logit contributions
adies+0.006
 ho+0.006
le+0.006
 lantern+0.006
 carriers+0.005
Neg logit contributions
igsaw-0.010
arre-0.010
ravings-0.009
 courtroom-0.009
 court-0.009
Token told
Token ID1297
Feature activation-4.47e-01

Pos logit contributions
igsaw+0.053
arre+0.052
 courtroom+0.045
ravings+0.044
 colony+0.043
Neg logit contributions
adies-0.051
 ho-0.046
 lantern-0.045
el-0.042
Mom-0.041
Token him
Token ID683
Feature activation-5.92e-01

Pos logit contributions
arre+0.020
igsaw+0.018
ravings+0.017
 courtroom+0.017
wheel+0.017
Neg logit contributions
adies-0.008
 carriers-0.006
 lantern-0.006
 ho-0.006
le-0.006
Token long
Token ID890
Feature activation-2.29e+00

Pos logit contributions
arre+0.019
igsaw+0.016
ravings+0.016
 courtroom+0.016
 gol+0.015
Neg logit contributions
adies-0.017
 lantern-0.015
 carriers-0.015
 ho-0.014
le-0.014
Token ago
Token ID2084
Feature activation-5.52e+00

Pos logit contributions
arre+0.083
igsaw+0.074
wheel+0.072
 courtroom+0.072
Em+0.071
Neg logit contributions
adies-0.049
 ho-0.047
 lantern-0.045
le-0.042
el-0.042
Token.
Token ID13
Feature activation-1.28e-01

Pos logit contributions
arre+0.237
igsaw+0.218
 courtroom+0.204
ravings+0.204
wheel+0.194
Neg logit contributions
adies-0.095
 carriers-0.084
 lantern-0.080
 ho-0.079
Mom-0.073
Token The
Token ID383
Feature activation-1.57e-03

Pos logit contributions
arre+0.005
 supposed+0.005
 fault+0.005
wheel+0.004
ravings+0.004
Neg logit contributions
Mom-0.003
Dad-0.002
adies-0.002
 God-0.002
Run-0.002
Token bird
Token ID6512
Feature activation-1.17e+00

Pos logit contributions
arre+0.000
ravings+0.000
igsaw+0.000
 fault+0.000
wheel+0.000
Neg logit contributions
adies-0.000
 carriers-0.000
 lantern-0.000
Mom-0.000
 ho-0.000
Token spread
Token ID4104
Feature activation+1.66e-01

Pos logit contributions
igsaw+0.045
 fault+0.044
arre+0.044
ravings+0.042
 court+0.042
Neg logit contributions
le-0.024
 lantern-0.022
el-0.021
adies-0.020
 ho-0.020
Token out
Token ID503
Feature activation+4.26e-01

Pos logit contributions
adies+0.004
 carriers+0.004
 lantern+0.004
Mom+0.004
 ho+0.004
Neg logit contributions
arre-0.006
igsaw-0.005
ravings-0.005
 courtroom-0.005
 court-0.005
Token started
Token ID2067
Feature activation-6.21e-01

Pos logit contributions
arre+0.009
 fault+0.008
igsaw+0.008
ravings+0.008
 courtroom+0.007
Neg logit contributions
 ho-0.004
 lantern-0.004
adies-0.004
Mom-0.004
le-0.004
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
arre+0.022
igsaw+0.020
ravings+0.019
Em+0.018
 courtroom+0.017
Neg logit contributions
 ho-0.017
 lantern-0.015
 closer-0.014
 hello-0.014
 sle-0.014
Token chase
Token ID15505
Feature activation-6.87e-01

Pos logit contributions
arre+0.040
igsaw+0.037
ravings+0.037
 fault+0.035
 By+0.035
Neg logit contributions
 lantern-0.019
adies-0.019
 ho-0.018
le-0.016
 wave-0.016
Token after
Token ID706
Feature activation+9.11e-01

Pos logit contributions
arre+0.023
igsaw+0.020
ravings+0.020
 fault+0.019
 courtroom+0.019
Neg logit contributions
adies-0.019
 carriers-0.017
 lantern-0.016
Mom-0.016
 ho-0.015
Token Kay
Token ID17356
Feature activation-1.63e+00

Pos logit contributions
adies+0.025
 lantern+0.021
 carriers+0.021
 ho+0.020
Mom+0.020
Neg logit contributions
arre-0.032
ravings-0.028
igsaw-0.027
 fault-0.027
 courtroom-0.026
Tokenlee
Token ID7197
Feature activation-5.51e+00

Pos logit contributions
arre+0.054
ravings+0.047
igsaw+0.047
 winning+0.044
 courtroom+0.044
Neg logit contributions
adies-0.046
 carriers-0.039
 lantern-0.038
 ho-0.038
el-0.037
Token.
Token ID13
Feature activation-5.90e-01

Pos logit contributions
arre+0.248
igsaw+0.234
ravings+0.220
 courtroom+0.218
 fault+0.210
Neg logit contributions
 ho-0.086
adies-0.082
 carriers-0.067
le-0.065
bye-0.063
Token She
Token ID1375
Feature activation-6.82e-01

Pos logit contributions
arre+0.021
ravings+0.019
igsaw+0.018
 exercising+0.018
Em+0.017
Neg logit contributions
adies-0.015
Mom-0.013
Dad-0.012
 carriers-0.012
 hello-0.012
Token screamed
Token ID25421
Feature activation-1.97e-01

Pos logit contributions
arre+0.027
igsaw+0.027
 fault+0.026
ravings+0.025
 courtroom+0.023
Neg logit contributions
 ho-0.013
adies-0.012
le-0.012
Mom-0.012
 lantern-0.012
Token and
Token ID290
Feature activation-1.06e+00

Pos logit contributions
arre+0.007
igsaw+0.006
ravings+0.006
 courtroom+0.006
 fault+0.006
Neg logit contributions
adies-0.005
 ho-0.004
 lantern-0.004
 carriers-0.004
Mom-0.004
Token ran
Token ID4966
Feature activation-1.22e+00

Pos logit contributions
arre+0.043
igsaw+0.040
ravings+0.039
 fault+0.039
 courtroom+0.035
Neg logit contributions
 ho-0.023
adies-0.021
 lantern-0.020
Run-0.020
Mom-0.020
Token know
Token ID760
Feature activation-7.79e-01

Pos logit contributions
arre+0.033
igsaw+0.029
ravings+0.029
 courtroom+0.029
 court+0.028
Neg logit contributions
adies-0.022
 carriers-0.019
 lantern-0.018
Mom-0.018
 ho-0.017
Token you
Token ID345
Feature activation-2.11e+00

Pos logit contributions
arre+0.027
igsaw+0.024
 courtroom+0.023
ravings+0.022
wheel+0.022
Neg logit contributions
adies-0.022
 lantern-0.018
 carriers-0.018
 ho-0.018
le-0.018
Token shouldn
Token ID6584
Feature activation-2.33e+00

Pos logit contributions
arre+0.079
igsaw+0.072
ravings+0.070
 courtroom+0.070
 court+0.068
Neg logit contributions
adies-0.048
 carriers-0.041
 lantern-0.040
Mom-0.038
 ho-0.037
Tokenâ
Token ID22940
Feature activation-3.96e-01

Pos logit contributions
arre+0.076
ravings+0.070
igsaw+0.070
 fault+0.066
 courtroom+0.065
Neg logit contributions
adies-0.060
Mom-0.057
 carriers-0.051
el-0.051
 lantern-0.051
Token
Token ID26391
Feature activation-2.16e+00

Pos logit contributions
arre+0.011
 fault+0.010
 accidentally+0.010
ravings+0.010
igsaw+0.010
Neg logit contributions
adies-0.012
le-0.011
Mom-0.011
 carriers-0.010
works-0.010
Token
Token ID8151
Feature activation-5.39e+00

Pos logit contributions
arre+0.063
igsaw+0.058
 fault+0.055
ravings+0.055
wheel+0.052
Neg logit contributions
adies-0.062
 carriers-0.055
Mom-0.054
 ho-0.054
ilot-0.053
Tokent
Token ID83
Feature activation-1.24e+00

Pos logit contributions
arre+0.184
igsaw+0.171
ravings+0.170
 courtroom+0.170
 disl+0.161
Neg logit contributions
adies-0.125
Mom-0.107
ilot-0.103
 ho-0.098
 carriers-0.097
Token speed
Token ID2866
Feature activation-1.65e+00

Pos logit contributions
arre+0.047
igsaw+0.044
ravings+0.043
 fault+0.042
 courtroom+0.042
Neg logit contributions
adies-0.025
le-0.021
Mom-0.021
 ho-0.020
 lantern-0.020
Token?
Token ID30
Feature activation-1.29e+00

Pos logit contributions
arre+0.061
igsaw+0.057
ravings+0.057
 courtroom+0.054
 accidentally+0.053
Neg logit contributions
adies-0.034
le-0.032
 carriers-0.030
Mom-0.029
el-0.029
Tokenâ
Token ID22940
Feature activation+1.79e+00

Pos logit contributions
arre+0.041
ravings+0.039
 fault+0.038
igsaw+0.037
 supposed+0.035
Neg logit contributions
adies-0.034
Mom-0.032
 carriers-0.030
Ho-0.029
 Amen-0.029
Token
Token ID26391
Feature activation-3.61e-01

Pos logit contributions
adies+0.054
le+0.048
works+0.047
Mom+0.046
 carriers+0.045
Neg logit contributions
arre-0.052
 fault-0.047
 accidentally-0.046
ravings-0.045
igsaw-0.044
Token bag
Token ID6131
Feature activation-2.04e-01

Pos logit contributions
 lantern+0.026
adies+0.026
 sle+0.023
Mom+0.023
 ho+0.023
Neg logit contributions
arre-0.039
ravings-0.034
 fault-0.034
wheel-0.033
igsaw-0.033
Token so
Token ID523
Feature activation+5.46e-01

Pos logit contributions
arre+0.008
igsaw+0.008
 courtroom+0.007
ravings+0.007
 disl+0.007
Neg logit contributions
adies-0.004
 carriers-0.004
 lantern-0.004
le-0.003
Mom-0.003
Token we
Token ID356
Feature activation-4.52e-01

Pos logit contributions
adies+0.014
 carriers+0.012
 lantern+0.012
Mom+0.011
le+0.011
Neg logit contributions
arre-0.018
igsaw-0.017
ravings-0.016
 winning-0.016
 accidentally-0.016
Token don
Token ID836
Feature activation+1.12e+00

Pos logit contributions
arre+0.018
igsaw+0.017
ravings+0.016
 courtroom+0.016
wheel+0.016
Neg logit contributions
adies-0.009
 lantern-0.007
 carriers-0.007
Mom-0.007
 ho-0.007
Token't
Token ID470
Feature activation-1.57e+00

Pos logit contributions
adies+0.040
 carriers+0.037
 lantern+0.036
 ho+0.034
Mom+0.034
Neg logit contributions
arre-0.028
igsaw-0.023
wheel-0.022
 court-0.022
ravings-0.022
Token accidentally
Token ID14716
Feature activation-5.32e+00

Pos logit contributions
arre+0.061
igsaw+0.059
ravings+0.056
 courtroom+0.054
Em+0.053
Neg logit contributions
adies-0.031
le-0.028
 ho-0.027
 lantern-0.026
Mom-0.025
Token touch
Token ID3638
Feature activation-3.81e-01

Pos logit contributions
arre+0.191
igsaw+0.180
ravings+0.162
 courtroom+0.161
 By+0.159
Neg logit contributions
adies-0.133
 lantern-0.119
 ho-0.118
le-0.114
 carriers-0.109
Token it
Token ID340
Feature activation-3.95e-01

Pos logit contributions
arre+0.013
igsaw+0.011
ravings+0.011
 courtroom+0.011
 fault+0.011
Neg logit contributions
adies-0.010
 carriers-0.009
 lantern-0.008
Mom-0.008
 ho-0.008
Token again
Token ID757
Feature activation-3.83e-01

Pos logit contributions
arre+0.015
igsaw+0.014
 courtroom+0.014
ravings+0.013
 court+0.013
Neg logit contributions
adies-0.009
 lantern-0.008
le-0.007
 carriers-0.007
 ho-0.007
Token."
Token ID526
Feature activation-1.23e+00

Pos logit contributions
arre+0.015
igsaw+0.015
wheel+0.014
 courtroom+0.014
ravings+0.014
Neg logit contributions
adies-0.008
 lantern-0.007
 ho-0.006
 carriers-0.006
le-0.006
Token
Token ID198
Feature activation-1.17e-01

Pos logit contributions
 supposed+0.046
 courtroom+0.045
arre+0.045
ravings+0.044
 winning+0.044
Neg logit contributions
Mom-0.024
adies-0.020
Dad-0.019
Ho-0.017
 lantern-0.017
Token to
Token ID284
Feature activation-5.53e-01

Pos logit contributions
arre+0.043
igsaw+0.038
ravings+0.037
 courtroom+0.037
wheel+0.036
Neg logit contributions
adies-0.030
 carriers-0.028
 lantern-0.026
works-0.026
 ho-0.025
Token be
Token ID307
Feature activation+2.45e-01

Pos logit contributions
arre+0.022
igsaw+0.019
ravings+0.019
 fault+0.018
 courtroom+0.018
Neg logit contributions
 lantern-0.012
adies-0.011
 carriers-0.010
 sail-0.010
Mom-0.010
Token restored
Token ID15032
Feature activation-5.37e-01

Pos logit contributions
adies+0.006
 carriers+0.006
 lantern+0.006
 ho+0.006
 sailors+0.005
Neg logit contributions
arre-0.008
igsaw-0.008
ravings-0.007
 salon-0.007
wheel-0.007
Token to
Token ID284
Feature activation-3.95e-01

Pos logit contributions
arre+0.020
igsaw+0.018
 courtroom+0.018
wheel+0.018
ravings+0.018
Neg logit contributions
adies-0.012
 carriers-0.011
 lantern-0.011
le-0.010
 ho-0.010
Token its
Token ID663
Feature activation-6.91e-01

Pos logit contributions
arre+0.016
ravings+0.014
igsaw+0.014
wheel+0.013
 fault+0.013
Neg logit contributions
adies-0.008
 lantern-0.008
 God-0.008
 aliens-0.007
 carriers-0.007
Token former
Token ID1966
Feature activation-5.30e+00

Pos logit contributions
arre+0.022
wheel+0.020
igsaw+0.019
Em+0.018
 Working+0.017
Neg logit contributions
 lantern-0.019
adies-0.019
 ho-0.019
 carriers-0.018
 sle-0.017
Token glory
Token ID13476
Feature activation-1.51e+00

Pos logit contributions
arre+0.168
wheel+0.155
 accidentally+0.144
igsaw+0.137
 winning+0.137
Neg logit contributions
adies-0.145
 lantern-0.139
 carriers-0.133
 ho-0.127
 sle-0.127
Token."
Token ID526
Feature activation-1.16e+00

Pos logit contributions
arre+0.058
igsaw+0.051
 courtroom+0.051
ravings+0.051
wheel+0.050
Neg logit contributions
adies-0.035
 carriers-0.031
 lantern-0.030
works-0.028
Mom-0.027
Token
Token ID220
Feature activation+2.32e-01

Pos logit contributions
arre+0.038
ravings+0.035
 courtroom+0.035
igsaw+0.034
 fault+0.033
Neg logit contributions
Mom-0.030
adies-0.030
 lantern-0.028
el-0.028
Dad-0.027
Token
Token ID198
Feature activation-8.35e-02

Pos logit contributions
Mom+0.005
adies+0.004
ilot+0.004
Dad+0.004
el+0.004
Neg logit contributions
arre-0.008
 courtroom-0.008
igsaw-0.008
ravings-0.008
wheel-0.008
Token
Token ID198
Feature activation+3.77e-01

Pos logit contributions
arre+0.003
ravings+0.003
 fault+0.002
 courtroom+0.002
igsaw+0.002
Neg logit contributions
Mom-0.002
adies-0.002
Dad-0.002
 lantern-0.002
Ho-0.002
Token.
Token ID13
Feature activation-3.10e-01

Pos logit contributions
le+0.006
el+0.006
 ho+0.006
adies+0.005
Mom+0.005
Neg logit contributions
arre-0.021
igsaw-0.020
 fault-0.019
ravings-0.018
wheel-0.018
Token And
Token ID843
Feature activation+1.31e+00

Pos logit contributions
arre+0.010
wheel+0.009
 courtroom+0.009
igsaw+0.009
ravings+0.009
Neg logit contributions
Mom-0.008
adies-0.007
 God-0.007
Dad-0.006
 carriers-0.006
Token if
Token ID611
Feature activation+3.30e+00

Pos logit contributions
Mom+0.028
 God+0.026
adies+0.026
 ho+0.026
 lantern+0.026
Neg logit contributions
arre-0.049
ravings-0.043
igsaw-0.043
 courtroom-0.042
wheel-0.039
Token she
Token ID673
Feature activation-2.67e+00

Pos logit contributions
adies+0.083
Mom+0.072
 lantern+0.071
 carriers+0.070
 God+0.070
Neg logit contributions
arre-0.120
igsaw-0.104
ravings-0.101
 courtroom-0.097
wheel-0.097
Token did
Token ID750
Feature activation+2.12e-01

Pos logit contributions
arre+0.078
igsaw+0.070
 courtroom+0.068
ravings+0.067
 court+0.067
Neg logit contributions
adies-0.080
 lantern-0.071
le-0.070
 carriers-0.070
Mom-0.070
Token accidentally
Token ID14716
Feature activation-5.28e+00

Pos logit contributions
adies+0.004
le+0.004
 lantern+0.004
 ho+0.004
Mom+0.003
Neg logit contributions
arre-0.008
igsaw-0.008
ravings-0.007
 fault-0.007
wheel-0.007
Token knock
Token ID10643
Feature activation-3.04e+00

Pos logit contributions
arre+0.180
igsaw+0.174
ravings+0.158
 courtroom+0.158
 By+0.157
Neg logit contributions
 lantern-0.123
adies-0.119
le-0.119
 ho-0.119
Mom-0.110
Token over
Token ID625
Feature activation+7.32e-01

Pos logit contributions
arre+0.119
ravings+0.106
igsaw+0.104
 courtroom+0.103
wheel+0.101
Neg logit contributions
adies-0.067
Mom-0.058
 lantern-0.057
le-0.057
el-0.055
Token her
Token ID607
Feature activation+9.71e-01

Pos logit contributions
adies+0.020
 lantern+0.018
 carriers+0.018
Mom+0.017
le+0.017
Neg logit contributions
arre-0.024
ravings-0.021
wheel-0.021
igsaw-0.020
 courtroom-0.020
Token cup
Token ID6508
Feature activation-1.83e-01

Pos logit contributions
 lantern+0.022
 ho+0.021
 carriers+0.021
adies+0.021
Mom+0.021
Neg logit contributions
arre-0.036
wheel-0.032
 courtroom-0.031
igsaw-0.030
 winning-0.030
Token,
Token ID11
Feature activation-8.34e-01

Pos logit contributions
arre+0.006
 winning+0.005
 supposed+0.005
 losing+0.005
 courtroom+0.005
Neg logit contributions
adies-0.004
 carriers-0.004
 lantern-0.004
Mom-0.004
el-0.004
Token no
Token ID645
Feature activation+1.07e+00

Pos logit contributions
adies+0.028
 carriers+0.025
 lantern+0.024
Mom+0.023
Ho+0.023
Neg logit contributions
arre-0.029
 accidentally-0.026
igsaw-0.025
 fault-0.025
 winning-0.025
Token,
Token ID11
Feature activation-1.33e+00

Pos logit contributions
adies+0.022
 ho+0.019
 lantern+0.019
le+0.018
Mom+0.018
Neg logit contributions
arre-0.041
igsaw-0.038
 courtroom-0.038
ravings-0.037
 court-0.037
Token I
Token ID314
Feature activation-1.87e+00

Pos logit contributions
arre+0.045
igsaw+0.040
 courtroom+0.039
ravings+0.039
 court+0.038
Neg logit contributions
adies-0.035
 carriers-0.031
 lantern-0.031
Mom-0.030
 ho-0.029
Token must
Token ID1276
Feature activation-1.09e+00

Pos logit contributions
igsaw+0.064
 courtroom+0.058
 court+0.058
ravings+0.057
arre+0.056
Neg logit contributions
le-0.041
adies-0.041
Mom-0.038
 ho-0.038
 lantern-0.036
Token have
Token ID423
Feature activation-1.73e+00

Pos logit contributions
arre+0.035
igsaw+0.031
ravings+0.030
 courtroom+0.030
 winning+0.029
Neg logit contributions
adies-0.031
 lantern-0.027
 carriers-0.027
Mom-0.026
 ho-0.025
Token accidentally
Token ID14716
Feature activation-5.26e+00

Pos logit contributions
arre+0.056
igsaw+0.054
 courtroom+0.047
wheel+0.045
 court+0.045
Neg logit contributions
adies-0.048
Mom-0.045
 lantern-0.044
 ho-0.042
le-0.041
Token spilled
Token ID31896
Feature activation-2.94e+00

Pos logit contributions
arre+0.183
igsaw+0.175
 courtroom+0.155
 By+0.151
 Working+0.151
Neg logit contributions
adies-0.137
 ho-0.122
 lantern-0.119
le-0.116
Mom-0.111
Token some
Token ID617
Feature activation+1.02e-01

Pos logit contributions
arre+0.114
 courtroom+0.104
igsaw+0.103
ravings+0.099
 winning+0.097
Neg logit contributions
adies-0.061
Mom-0.055
 lantern-0.055
el-0.052
 ho-0.051
Token water
Token ID1660
Feature activation+4.66e-01

Pos logit contributions
adies+0.003
 carriers+0.003
 lantern+0.003
Mom+0.003
 ho+0.002
Neg logit contributions
arre-0.003
igsaw-0.003
 courtroom-0.003
ravings-0.003
 court-0.003
Token on
Token ID319
Feature activation-7.76e-01

Pos logit contributions
adies+0.009
 lantern+0.008
le+0.008
 carriers+0.008
 ho+0.008
Neg logit contributions
arre-0.018
igsaw-0.017
 courtroom-0.017
 winning-0.016
 Working-0.016
Token it
Token ID340
Feature activation+2.47e-01

Pos logit contributions
arre+0.031
igsaw+0.027
wheel+0.026
ravings+0.026
 courtroom+0.026
Neg logit contributions
adies-0.017
Mom-0.015
 lantern-0.015
el-0.014
 carriers-0.014
Token quietly
Token ID12703
Feature activation+1.67e-01

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 fault+0.003
 courtroom+0.003
Neg logit contributions
 lantern-0.002
 ho-0.001
 carriers-0.001
adies-0.001
 wave-0.001
Token so
Token ID523
Feature activation+3.61e-01

Pos logit contributions
 ho+0.004
 wave+0.003
 lantern+0.003
le+0.003
 sail+0.003
Neg logit contributions
arre-0.006
igsaw-0.006
ravings-0.006
Em-0.005
 courtroom-0.005
Token he
Token ID339
Feature activation-1.97e-01

Pos logit contributions
adies+0.008
 carriers+0.008
 ho+0.008
 lantern+0.008
Mom+0.007
Neg logit contributions
arre-0.013
igsaw-0.012
ravings-0.011
 courtroom-0.011
 court-0.011
Token wouldn
Token ID3636
Feature activation+6.43e-02

Pos logit contributions
igsaw+0.007
arre+0.007
 courtroom+0.006
 court+0.006
ravings+0.006
Neg logit contributions
adies-0.004
le-0.004
 ho-0.004
 lantern-0.004
Mom-0.004
Token't
Token ID470
Feature activation-8.75e-01

Pos logit contributions
adies+0.002
Mom+0.002
le+0.001
 ho+0.001
el+0.001
Neg logit contributions
arre-0.002
igsaw-0.002
ravings-0.002
 courtroom-0.002
 fault-0.002
Token accidentally
Token ID14716
Feature activation-5.23e+00

Pos logit contributions
igsaw+0.034
arre+0.033
ravings+0.031
 courtroom+0.030
 By+0.029
Neg logit contributions
le-0.016
 ho-0.016
 lantern-0.016
Mom-0.015
 wave-0.015
Token hit
Token ID2277
Feature activation-2.11e+00

Pos logit contributions
arre+0.180
igsaw+0.178
ravings+0.157
 By+0.156
 courtroom+0.150
Neg logit contributions
 ho-0.123
adies-0.122
 lantern-0.117
le-0.114
Mom-0.111
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
arre+0.085
ravings+0.077
igsaw+0.076
 courtroom+0.072
wheel+0.071
Neg logit contributions
adies-0.041
el-0.037
 lantern-0.036
Mom-0.036
 ho-0.035
Token wall
Token ID3355
Feature activation-8.09e-01

Pos logit contributions
adies+0.027
 lantern+0.026
 carriers+0.025
Mom+0.025
 ho+0.024
Neg logit contributions
arre-0.042
igsaw-0.038
ravings-0.036
 By-0.035
wheel-0.035
Token or
Token ID393
Feature activation+6.52e-02

Pos logit contributions
arre+0.029
 courtroom+0.027
ravings+0.026
igsaw+0.026
Em+0.025
Neg logit contributions
adies-0.017
le-0.016
 lantern-0.016
Mom-0.015
 ho-0.015
Token disturb
Token ID17037
Feature activation-7.92e-01

Pos logit contributions
 lantern+0.001
adies+0.001
 ho+0.001
Mom+0.001
 wave+0.001
Neg logit contributions
arre-0.003
igsaw-0.002
ravings-0.002
 Working-0.002
 By-0.002
Token,
Token ID11
Feature activation-5.90e-01

Pos logit contributions
arre+0.066
igsaw+0.064
 courtroom+0.057
Em+0.057
wheel+0.057
Neg logit contributions
adies-0.024
 carriers-0.023
 lantern-0.020
 ho-0.019
Mom-0.018
Token so
Token ID523
Feature activation+1.04e+00

Pos logit contributions
igsaw+0.023
arre+0.022
seek+0.020
 courtroom+0.020
 court+0.019
Neg logit contributions
 ho-0.012
 lantern-0.011
 God-0.011
adies-0.011
Mom-0.010
Token she
Token ID673
Feature activation-1.05e+00

Pos logit contributions
Mom+0.025
adies+0.025
 ho+0.024
 lantern+0.024
 carriers+0.024
Neg logit contributions
arre-0.036
igsaw-0.033
 courtroom-0.030
wheel-0.030
ravings-0.029
Token wouldn
Token ID3636
Feature activation+1.92e-01

Pos logit contributions
igsaw+0.036
arre+0.035
 courtroom+0.032
 court+0.032
 winning+0.031
Neg logit contributions
adies-0.025
 lantern-0.023
le-0.022
 carriers-0.022
Mom-0.022
Token't
Token ID470
Feature activation-8.21e-01

Pos logit contributions
adies+0.004
Mom+0.004
el+0.004
le+0.004
 ho+0.004
Neg logit contributions
arre-0.007
ravings-0.006
igsaw-0.006
 fault-0.006
 courtroom-0.006
Token accidentally
Token ID14716
Feature activation-5.22e+00

Pos logit contributions
igsaw+0.035
arre+0.034
ravings+0.031
 Working+0.031
 Grad+0.030
Neg logit contributions
 ho-0.014
 go-0.014
 lantern-0.013
 wave-0.013
 sail-0.012
Token break
Token ID2270
Feature activation-2.25e+00

Pos logit contributions
igsaw+0.186
arre+0.185
 By+0.165
ravings+0.161
 Working+0.158
Neg logit contributions
 ho-0.114
adies-0.113
 lantern-0.112
 go-0.107
le-0.107
Token anything
Token ID1997
Feature activation-6.27e-01

Pos logit contributions
arre+0.088
igsaw+0.078
ravings+0.078
wheel+0.075
 courtroom+0.074
Neg logit contributions
adies-0.045
 lantern-0.042
el-0.042
Mom-0.042
le-0.041
Token again
Token ID757
Feature activation-7.08e-01

Pos logit contributions
arre+0.025
igsaw+0.023
 Working+0.022
ravings+0.022
 courtroom+0.022
Neg logit contributions
le-0.012
 lantern-0.010
 closer-0.010
 ho-0.010
adies-0.009
Token.
Token ID13
Feature activation-3.65e-01

Pos logit contributions
arre+0.031
igsaw+0.029
ravings+0.027
wheel+0.027
 courtroom+0.027
Neg logit contributions
adies-0.010
le-0.010
el-0.009
 ho-0.009
 lantern-0.009
Token<|endoftext|>
Token ID50256
Feature activation-1.50e+00

Pos logit contributions
arre+0.013
wheel+0.012
igsaw+0.012
 courtroom+0.012
 winning+0.011
Neg logit contributions
Mom-0.008
adies-0.007
 God-0.007
 Amen-0.007
Ho-0.007
Token's
Token ID338
Feature activation-4.76e-01

Pos logit contributions
igsaw+0.050
 fault+0.046
arre+0.045
 court+0.043
 courtroom+0.043
Neg logit contributions
Mom-0.026
 ho-0.023
le-0.023
adies-0.022
 lantern-0.021
Token mom
Token ID1995
Feature activation-1.05e-01

Pos logit contributions
arre+0.020
igsaw+0.018
ravings+0.017
 courtroom+0.017
wheel+0.017
Neg logit contributions
adies-0.009
 carriers-0.008
 lantern-0.007
Mom-0.007
 ho-0.007
Token realized
Token ID6939
Feature activation-1.82e+00

Pos logit contributions
arre+0.003
igsaw+0.003
 fault+0.003
ravings+0.003
 court+0.003
Neg logit contributions
adies-0.003
Mom-0.003
 lantern-0.002
 carriers-0.002
le-0.002
Token she
Token ID673
Feature activation-1.97e+00

Pos logit contributions
arre+0.065
igsaw+0.060
seek+0.058
 fault+0.057
 colony+0.056
Neg logit contributions
adies-0.040
Mom-0.039
 God-0.039
 lantern-0.038
 comet-0.037
Token had
Token ID550
Feature activation-2.27e+00

Pos logit contributions
arre+0.064
igsaw+0.063
 court+0.058
ravings+0.058
 courtroom+0.058
Neg logit contributions
adies-0.049
le-0.045
Mom-0.043
 carriers-0.042
 lantern-0.042
Token accidentally
Token ID14716
Feature activation-5.18e+00

Pos logit contributions
igsaw+0.092
arre+0.076
 gol+0.075
 Working+0.074
 gym+0.073
Neg logit contributions
 ho-0.041
w-0.037
Mom-0.036
 come-0.035
 d-0.035
Token used
Token ID973
Feature activation-1.61e+00

Pos logit contributions
igsaw+0.181
arre+0.172
 By+0.153
 Working+0.152
 courtroom+0.151
Neg logit contributions
adies-0.120
 ho-0.118
le-0.115
 lantern-0.114
Mom-0.114
Token salt
Token ID8268
Feature activation-9.72e-01

Pos logit contributions
arre+0.071
 courtroom+0.064
seek+0.061
wheel+0.061
 court+0.060
Neg logit contributions
el-0.027
 ho-0.025
 lantern-0.024
Mom-0.022
 them-0.022
Token instead
Token ID2427
Feature activation+9.65e-01

Pos logit contributions
arre+0.033
 courtroom+0.033
igsaw+0.032
 court+0.030
 winning+0.030
Neg logit contributions
le-0.022
adies-0.022
 ho-0.021
 lantern-0.021
 carriers-0.020
Token of
Token ID286
Feature activation-1.74e+00

Pos logit contributions
adies+0.015
le+0.014
el+0.013
 ho+0.013
Mom+0.013
Neg logit contributions
arre-0.043
 courtroom-0.039
igsaw-0.038
 disl-0.037
ravings-0.037
Token sugar
Token ID7543
Feature activation-2.19e+00

Pos logit contributions
arre+0.068
seek+0.061
wheel+0.061
igsaw+0.060
 courtroom+0.059
Neg logit contributions
 ho-0.038
 lantern-0.035
el-0.035
 noses-0.031
 hus-0.031
Token one
Token ID530
Feature activation+1.65e+00

Pos logit contributions
arre+0.014
igsaw+0.012
ravings+0.012
wheel+0.012
 courtroom+0.011
Neg logit contributions
adies-0.009
 lantern-0.008
 carriers-0.008
 comet-0.007
 ho-0.007
Token he
Token ID339
Feature activation+5.33e-01

Pos logit contributions
 lantern+0.035
adies+0.035
le+0.034
 comet+0.032
 carriers+0.031
Neg logit contributions
arre-0.066
igsaw-0.056
 courtroom-0.055
ravings-0.055
wheel-0.053
Token had
Token ID550
Feature activation-1.24e+00

Pos logit contributions
adies+0.012
le+0.010
 ho+0.010
 lantern+0.010
 carriers+0.009
Neg logit contributions
arre-0.021
igsaw-0.019
 fault-0.018
ravings-0.018
 courtroom-0.017
Token when
Token ID618
Feature activation+1.38e+00

Pos logit contributions
arre+0.042
igsaw+0.040
ravings+0.035
wheel+0.035
 courtroom+0.034
Neg logit contributions
adies-0.034
 ho-0.030
 lantern-0.030
 carriers-0.027
le-0.027
Token he
Token ID339
Feature activation-9.79e-01

Pos logit contributions
adies+0.033
 ho+0.030
 lantern+0.030
 carriers+0.029
Mom+0.028
Neg logit contributions
arre-0.051
igsaw-0.043
ravings-0.042
wheel-0.042
 fault-0.041
Token accidentally
Token ID14716
Feature activation-5.18e+00

Pos logit contributions
arre+0.037
igsaw+0.036
 fault+0.033
ravings+0.032
 winning+0.032
Neg logit contributions
adies-0.022
le-0.020
 ho-0.019
 lantern-0.018
Mom-0.017
Token reversed
Token ID17687
Feature activation-1.63e+00

Pos logit contributions
arre+0.184
igsaw+0.172
ravings+0.155
 By+0.153
 Working+0.152
Neg logit contributions
adies-0.132
 lantern-0.122
 ho-0.121
le-0.112
 carriers-0.108
Token.
Token ID13
Feature activation+1.51e-01

Pos logit contributions
arre+0.068
ravings+0.060
igsaw+0.058
wheel+0.057
 courtroom+0.056
Neg logit contributions
adies-0.029
 lantern-0.029
 ho-0.028
le-0.028
el-0.026
Token<|endoftext|>
Token ID50256
Feature activation-1.34e+00

Pos logit contributions
Mom+0.003
adies+0.003
 Amen+0.003
 God+0.003
Dad+0.003
Neg logit contributions
arre-0.006
 fault-0.005
ravings-0.005
 supposed-0.005
wheel-0.005
TokenOnce
Token ID7454
Feature activation+6.48e-01

Pos logit contributions
arre+0.042
ravings+0.037
igsaw+0.036
 fault+0.035
 courtroom+0.035
Neg logit contributions
adies-0.039
Mom-0.035
 carriers-0.034
 lantern-0.033
Ho-0.032
Token upon
Token ID2402
Feature activation+2.35e-01

Pos logit contributions
adies+0.013
Mom+0.011
el+0.010
 lantern+0.010
 ho+0.010
Neg logit contributions
arre-0.026
wheel-0.024
ravings-0.024
 courtroom-0.024
seek-0.023
Token
Token ID198
Feature activation-2.86e-01

Pos logit contributions
adies+0.012
Mom+0.012
 carriers+0.010
ilot+0.010
Dad+0.010
Neg logit contributions
arre-0.014
 courtroom-0.014
igsaw-0.014
ravings-0.014
wheel-0.013
Token
Token ID198
Feature activation+3.11e-01

Pos logit contributions
arre+0.009
 fault+0.009
ravings+0.008
 courtroom+0.008
igsaw+0.008
Neg logit contributions
Mom-0.008
Dad-0.008
adies-0.008
Ho-0.007
 lantern-0.007
TokenBut
Token ID1537
Feature activation+1.37e+00

Pos logit contributions
Mom+0.007
adies+0.007
Dad+0.007
 carriers+0.006
Ho+0.006
Neg logit contributions
 fault-0.011
arre-0.011
igsaw-0.010
ravings-0.010
 courtroom-0.010
Token Jimmy
Token ID12963
Feature activation-3.89e-01

Pos logit contributions
adies+0.036
Mom+0.035
 ho+0.034
le+0.033
 lantern+0.032
Neg logit contributions
arre-0.045
igsaw-0.041
wheel-0.040
 fault-0.040
ravings-0.039
Token had
Token ID550
Feature activation-2.42e+00

Pos logit contributions
igsaw+0.016
arre+0.014
 fault+0.013
ravings+0.013
 courtroom+0.013
Neg logit contributions
le-0.007
 ho-0.007
Mom-0.007
el-0.007
adies-0.006
Token accidentally
Token ID14716
Feature activation-5.18e+00

Pos logit contributions
igsaw+0.090
arre+0.081
 disl+0.077
wheel+0.077
handed+0.076
Neg logit contributions
 ho-0.045
Mom-0.040
 sails-0.039
adies-0.038
Dad-0.038
Token put
Token ID1234
Feature activation-1.79e+00

Pos logit contributions
arre+0.179
igsaw+0.174
 courtroom+0.153
ravings+0.153
 By+0.152
Neg logit contributions
 lantern-0.121
adies-0.120
 ho-0.119
le-0.109
Mom-0.107
Token too
Token ID1165
Feature activation+5.12e-01

Pos logit contributions
arre+0.065
igsaw+0.058
 courtroom+0.058
ravings+0.058
 court+0.056
Neg logit contributions
adies-0.039
 lantern-0.038
el-0.037
 carriers-0.037
Mom-0.036
Token much
Token ID881
Feature activation-2.31e+00

Pos logit contributions
adies+0.013
 lantern+0.011
Mom+0.011
 carriers+0.011
 ho+0.011
Neg logit contributions
arre-0.018
igsaw-0.016
ravings-0.016
 courtroom-0.015
wheel-0.015
Token of
Token ID286
Feature activation-9.74e-01

Pos logit contributions
arre+0.078
 courtroom+0.072
igsaw+0.070
ravings+0.069
 winning+0.068
Neg logit contributions
adies-0.058
 lantern-0.054
le-0.051
 carriers-0.051
Mom-0.049
Token a
Token ID257
Feature activation-2.49e-01

Pos logit contributions
arre+0.039
 courtroom+0.033
ravings+0.033
wheel+0.032
igsaw+0.032
Neg logit contributions
 lantern-0.020
adies-0.019
 ho-0.018
el-0.018
Mom-0.018
Token and
Token ID290
Feature activation-9.03e-01

Pos logit contributions
arre+0.047
igsaw+0.046
 courtroom+0.046
wheel+0.045
ravings+0.045
Neg logit contributions
 ho-0.012
le-0.012
ies-0.011
adies-0.010
el-0.009
Token worse
Token ID4785
Feature activation-1.23e+00

Pos logit contributions
arre+0.032
igsaw+0.031
ravings+0.030
 fault+0.028
 courtroom+0.028
Neg logit contributions
adies-0.020
Mom-0.019
 carriers-0.018
Run-0.018
 lantern-0.017
Token until
Token ID1566
Feature activation-9.39e-01

Pos logit contributions
arre+0.057
igsaw+0.053
 courtroom+0.052
wheel+0.051
ravings+0.051
Neg logit contributions
 ho-0.016
adies-0.015
ies-0.014
 lantern-0.014
le-0.014
Token Tim
Token ID5045
Feature activation-1.52e+00

Pos logit contributions
arre+0.034
ravings+0.030
 fault+0.028
igsaw+0.028
wheel+0.028
Neg logit contributions
adies-0.022
Mom-0.022
 lantern-0.020
el-0.020
 carriers-0.019
Tokenmy
Token ID1820
Feature activation-1.69e+00

Pos logit contributions
igsaw+0.056
arre+0.053
 courtroom+0.051
 disl+0.051
 court+0.050
Neg logit contributions
adies-0.033
 ho-0.030
Mom-0.029
le-0.028
el-0.027
Token accidentally
Token ID14716
Feature activation-5.15e+00

Pos logit contributions
igsaw+0.068
arre+0.060
 disl+0.059
 fault+0.058
ravings+0.058
Neg logit contributions
le-0.031
Mom-0.029
 ho-0.029
adies-0.029
Dad-0.025
Token ran
Token ID4966
Feature activation-1.36e+00

Pos logit contributions
arre+0.164
igsaw+0.159
ravings+0.143
 By+0.140
 courtroom+0.139
Neg logit contributions
adies-0.141
 ho-0.130
 lantern-0.128
Mom-0.123
 carriers-0.122
Token into
Token ID656
Feature activation+2.86e-01

Pos logit contributions
arre+0.054
ravings+0.050
igsaw+0.049
 colony+0.048
 disl+0.047
Neg logit contributions
adies-0.024
 carriers-0.022
 ho-0.022
Mom-0.021
 closer-0.020
Token the
Token ID262
Feature activation+6.63e-01

Pos logit contributions
Mom+0.004
 lantern+0.004
el+0.004
 breath+0.004
 ho+0.003
Neg logit contributions
arre-0.013
ravings-0.012
 fault-0.011
wheel-0.011
seek-0.011
Token fence
Token ID13990
Feature activation-9.95e-01

Pos logit contributions
Mom+0.017
adies+0.016
 carriers+0.015
 ho+0.014
 lantern+0.014
Neg logit contributions
arre-0.021
wheel-0.020
igsaw-0.020
 fault-0.019
ravings-0.019
Token and
Token ID290
Feature activation-4.99e-01

Pos logit contributions
arre+0.045
igsaw+0.041
ravings+0.041
 disl+0.040
 courtroom+0.039
Neg logit contributions
adies-0.013
Mom-0.012
 lantern-0.011
el-0.011
 carriers-0.011
Token behind
Token ID2157
Feature activation-3.78e-01

Pos logit contributions
 lantern+0.003
adies+0.003
 ho+0.003
 carriers+0.003
le+0.003
Neg logit contributions
arre-0.005
 courtroom-0.004
ravings-0.004
igsaw-0.004
 fault-0.004
Token her
Token ID607
Feature activation-2.64e-01

Pos logit contributions
arre+0.013
ravings+0.011
igsaw+0.011
wheel+0.011
 courtroom+0.011
Neg logit contributions
adies-0.010
 lantern-0.009
 carriers-0.009
Mom-0.008
 ho-0.008
Token,
Token ID11
Feature activation-1.21e+00

Pos logit contributions
arre+0.010
igsaw+0.009
ravings+0.009
wheel+0.009
 courtroom+0.009
Neg logit contributions
 lantern-0.005
adies-0.005
 carriers-0.005
 ho-0.005
Mom-0.005
Token but
Token ID475
Feature activation+3.78e-01

Pos logit contributions
arre+0.040
igsaw+0.039
ravings+0.035
seek+0.034
 Working+0.033
Neg logit contributions
 ho-0.031
 lantern-0.030
adies-0.030
Mom-0.028
 wave-0.027
Token she
Token ID673
Feature activation-2.64e+00

Pos logit contributions
Mom+0.008
adies+0.008
 ho+0.007
le+0.006
el+0.006
Neg logit contributions
arre-0.015
igsaw-0.014
 fault-0.013
ravings-0.013
 courtroom-0.013
Token accidentally
Token ID14716
Feature activation-5.14e+00

Pos logit contributions
igsaw+0.106
arre+0.095
 fault+0.090
 courtroom+0.088
 Working+0.088
Neg logit contributions
le-0.054
Mom-0.054
 ho-0.054
adies-0.050
el-0.046
Token tri
Token ID1333
Feature activation-2.27e+00

Pos logit contributions
arre+0.175
igsaw+0.167
ravings+0.151
 courtroom+0.149
 By+0.145
Neg logit contributions
adies-0.136
 lantern-0.125
 ho-0.124
Mom-0.118
le-0.115
Tokenpped
Token ID1496
Feature activation-8.63e-01

Pos logit contributions
arre+0.087
igsaw+0.082
ravings+0.080
 courtroom+0.078
 fault+0.078
Neg logit contributions
adies-0.050
Mom-0.040
le-0.038
 ho-0.038
 carriers-0.038
Token and
Token ID290
Feature activation-1.17e+00

Pos logit contributions
arre+0.033
igsaw+0.031
ravings+0.030
 courtroom+0.030
 fault+0.028
Neg logit contributions
adies-0.017
Mom-0.017
 lantern-0.016
el-0.015
 ho-0.015
Token the
Token ID262
Feature activation-3.42e-01

Pos logit contributions
arre+0.049
igsaw+0.047
ravings+0.044
 courtroom+0.043
 fault+0.042
Neg logit contributions
 ho-0.020
adies-0.020
Mom-0.020
 lantern-0.017
el-0.017
Token pipe
Token ID12656
Feature activation-5.64e-01

Pos logit contributions
arre+0.010
igsaw+0.009
ravings+0.009
 courtroom+0.009
 court+0.009
Neg logit contributions
adies-0.010
 carriers-0.009
 lantern-0.009
Mom-0.009
 ho-0.008
Token to
Token ID284
Feature activation+4.54e-01

Pos logit contributions
arre+0.034
 fault+0.029
ravings+0.029
igsaw+0.028
 winning+0.027
Neg logit contributions
adies-0.035
 lantern-0.031
Mom-0.031
 carriers-0.030
 ho-0.029
Token get
Token ID651
Feature activation+2.02e-01

Pos logit contributions
adies+0.008
 lantern+0.008
 God+0.008
Mom+0.008
 ho+0.007
Neg logit contributions
arre-0.019
ravings-0.017
wheel-0.016
 fault-0.016
igsaw-0.016
Token the
Token ID262
Feature activation-5.12e-02

Pos logit contributions
adies+0.006
 carriers+0.005
 lantern+0.005
Mom+0.005
 ho+0.005
Neg logit contributions
arre-0.006
igsaw-0.006
ravings-0.005
 courtroom-0.005
wheel-0.005
Token pot
Token ID1787
Feature activation+5.28e-01

Pos logit contributions
arre+0.001
 fault+0.001
 accidentally+0.001
wheel+0.001
igsaw+0.001
Neg logit contributions
adies-0.002
 carriers-0.001
 Amen-0.001
Ho-0.001
 aliens-0.001
Token but
Token ID475
Feature activation+2.46e-02

Pos logit contributions
adies+0.012
 lantern+0.011
 carriers+0.011
le+0.010
 ho+0.010
Neg logit contributions
arre-0.019
 courtroom-0.017
 winning-0.017
igsaw-0.017
ravings-0.017
Token accidentally
Token ID14716
Feature activation-5.14e+00

Pos logit contributions
adies+0.001
Mom+0.001
 ho+0.000
 lantern+0.000
el+0.000
Neg logit contributions
arre-0.001
igsaw-0.001
 fault-0.001
 courtroom-0.001
ravings-0.001
Token dropped
Token ID5710
Feature activation-1.91e+00

Pos logit contributions
arre+0.159
igsaw+0.143
ravings+0.132
 courtroom+0.132
 winning+0.128
Neg logit contributions
adies-0.156
 lantern-0.142
 ho-0.136
Mom-0.135
 carriers-0.135
Token it
Token ID340
Feature activation+5.19e-03

Pos logit contributions
arre+0.082
igsaw+0.073
ravings+0.073
 courtroom+0.072
seek+0.071
Neg logit contributions
adies-0.033
 lantern-0.031
Mom-0.030
el-0.029
 ho-0.027
Token on
Token ID319
Feature activation-1.50e+00

Pos logit contributions
adies+0.000
le+0.000
el+0.000
 ho+0.000
 lantern+0.000
Neg logit contributions
arre-0.000
 courtroom-0.000
igsaw-0.000
 disl-0.000
ravings-0.000
Token the
Token ID262
Feature activation-6.26e-01

Pos logit contributions
arre+0.061
ravings+0.051
wheel+0.050
 fault+0.050
igsaw+0.049
Neg logit contributions
Mom-0.033
adies-0.031
 lantern-0.030
el-0.029
 carriers-0.028
Token floor
Token ID4314
Feature activation-1.76e+00

Pos logit contributions
arre+0.017
igsaw+0.014
ravings+0.014
 courtroom+0.014
 winning+0.013
Neg logit contributions
adies-0.021
 carriers-0.019
 lantern-0.019
Mom-0.018
 ho-0.018
Token while
Token ID981
Feature activation+1.02e+00

Pos logit contributions
le+0.004
adies+0.004
 ho+0.004
 lantern+0.004
Mom+0.004
Neg logit contributions
arre-0.010
ravings-0.009
 fault-0.009
 courtroom-0.009
igsaw-0.009
Token,
Token ID11
Feature activation-1.76e+00

Pos logit contributions
adies+0.021
 ho+0.020
Mom+0.019
 lantern+0.018
 carriers+0.017
Neg logit contributions
arre-0.040
igsaw-0.036
ravings-0.035
 fault-0.035
wheel-0.034
Token but
Token ID475
Feature activation+3.53e-01

Pos logit contributions
igsaw+0.066
arre+0.066
ravings+0.059
 fault+0.058
 courtroom+0.056
Neg logit contributions
 ho-0.039
adies-0.036
 lantern-0.033
Mom-0.030
 yes-0.030
Token then
Token ID788
Feature activation-1.13e+00

Pos logit contributions
Mom+0.007
adies+0.007
 ho+0.006
 God+0.006
le+0.006
Neg logit contributions
arre-0.014
igsaw-0.013
ravings-0.013
wheel-0.012
 fault-0.012
Token she
Token ID673
Feature activation-1.73e+00

Pos logit contributions
arre+0.039
igsaw+0.036
ravings+0.035
 fault+0.034
wheel+0.033
Neg logit contributions
adies-0.028
 lantern-0.026
 ho-0.025
Mom-0.025
 carriers-0.024
Token accidentally
Token ID14716
Feature activation-5.13e+00

Pos logit contributions
igsaw+0.058
arre+0.054
 fault+0.053
ravings+0.051
 winning+0.050
Neg logit contributions
adies-0.043
Mom-0.042
le-0.040
 ho-0.040
Dad-0.037
Token knocked
Token ID13642
Feature activation-2.54e+00

Pos logit contributions
arre+0.165
igsaw+0.159
ravings+0.141
 By+0.139
 courtroom+0.139
Neg logit contributions
adies-0.141
 lantern-0.133
 ho-0.130
Mom-0.124
 carriers-0.123
Token over
Token ID625
Feature activation+5.08e-01

Pos logit contributions
arre+0.099
ravings+0.097
 courtroom+0.094
igsaw+0.094
 disl+0.092
Neg logit contributions
Mom-0.054
el-0.048
adies-0.041
le-0.040
Dad-0.040
Token a
Token ID257
Feature activation+1.03e+00

Pos logit contributions
adies+0.012
 lantern+0.011
Mom+0.011
 carriers+0.011
 ho+0.010
Neg logit contributions
arre-0.019
ravings-0.016
wheel-0.016
 fault-0.016
igsaw-0.016
Token tower
Token ID10580
Feature activation+2.12e-01

Pos logit contributions
 lantern+0.022
 comet+0.020
 sle+0.020
Mom+0.019
adies+0.019
Neg logit contributions
arre-0.041
 fault-0.036
seek-0.036
ravings-0.035
 By-0.034
Token she
Token ID673
Feature activation-4.73e-03

Pos logit contributions
adies+0.003
le+0.003
el+0.003
Mom+0.003
 lantern+0.003
Neg logit contributions
arre-0.009
 courtroom-0.009
ravings-0.008
igsaw-0.008
 fault-0.008
Token."
Token ID526
Feature activation-1.14e+00

Pos logit contributions
arre+0.027
igsaw+0.024
 courtroom+0.023
ravings+0.023
wheel+0.023
Neg logit contributions
adies-0.020
 carriers-0.017
 lantern-0.017
Mom-0.016
 ho-0.016
Token
Token ID198
Feature activation-2.47e-01

Pos logit contributions
arre+0.041
igsaw+0.039
wheel+0.038
ravings+0.038
 courtroom+0.038
Neg logit contributions
Mom-0.027
adies-0.023
Dad-0.021
 lantern-0.019
el-0.018
Token
Token ID198
Feature activation+2.83e-01

Pos logit contributions
arre+0.007
igsaw+0.007
ravings+0.007
 courtroom+0.007
 fault+0.006
Neg logit contributions
adies-0.007
Mom-0.007
 lantern-0.006
Dad-0.006
 carriers-0.006
TokenBut
Token ID1537
Feature activation+1.32e+00

Pos logit contributions
adies+0.007
Mom+0.007
 carriers+0.006
 lantern+0.006
Dad+0.006
Neg logit contributions
arre-0.010
 fault-0.009
igsaw-0.009
ravings-0.009
 courtroom-0.009
Token Lily
Token ID20037
Feature activation-2.01e+00

Pos logit contributions
Mom+0.036
adies+0.036
 ho+0.032
le+0.031
 carriers+0.030
Neg logit contributions
arre-0.045
igsaw-0.042
ravings-0.039
 courtroom-0.039
 fault-0.038
Token accidentally
Token ID14716
Feature activation-5.12e+00

Pos logit contributions
igsaw+0.077
arre+0.066
 winning+0.065
 fault+0.065
 courtroom+0.063
Neg logit contributions
Mom-0.042
le-0.040
adies-0.040
el-0.039
 ho-0.037
Token dropped
Token ID5710
Feature activation-1.70e+00

Pos logit contributions
arre+0.148
igsaw+0.131
ravings+0.123
 courtroom+0.123
 court+0.120
Neg logit contributions
adies-0.162
 lantern-0.144
 carriers-0.144
Mom-0.140
 ho-0.138
Token the
Token ID262
Feature activation-7.97e-01

Pos logit contributions
arre+0.066
igsaw+0.059
ravings+0.059
wheel+0.057
 courtroom+0.056
Neg logit contributions
adies-0.036
 lantern-0.033
 carriers-0.031
Mom-0.031
el-0.030
Token gem
Token ID16840
Feature activation-6.64e-01

Pos logit contributions
arre+0.024
igsaw+0.022
 court+0.021
 courtroom+0.021
wheel+0.021
Neg logit contributions
adies-0.024
 carriers-0.021
 Amen-0.020
le-0.019
 lantern-0.019
Token and
Token ID290
Feature activation-5.24e-01

Pos logit contributions
arre+0.025
igsaw+0.022
ravings+0.022
 courtroom+0.022
 winning+0.021
Neg logit contributions
adies-0.014
 lantern-0.013
 carriers-0.013
 ho-0.012
le-0.012
Token it
Token ID340
Feature activation+1.66e-01

Pos logit contributions
arre+0.021
igsaw+0.019
ravings+0.018
 fault+0.018
 courtroom+0.018
Neg logit contributions
adies-0.011
 lantern-0.009
Mom-0.009
 carriers-0.009
 ho-0.009
Token pepper
Token ID13385
Feature activation-1.04e-01

Pos logit contributions
adies+0.013
 lantern+0.012
 carriers+0.011
Mom+0.011
 ho+0.011
Neg logit contributions
arre-0.015
igsaw-0.013
 courtroom-0.013
ravings-0.013
wheel-0.012
Token in
Token ID287
Feature activation+6.54e-01

Pos logit contributions
arre+0.004
igsaw+0.003
 courtroom+0.003
 winning+0.003
 court+0.003
Neg logit contributions
adies-0.002
 lantern-0.002
 carriers-0.002
 ho-0.002
le-0.002
Token it
Token ID340
Feature activation+4.26e-01

Pos logit contributions
adies+0.016
Mom+0.015
 lantern+0.015
el+0.014
 ho+0.014
Neg logit contributions
arre-0.024
ravings-0.021
wheel-0.020
igsaw-0.020
 courtroom-0.020
Token when
Token ID618
Feature activation+1.16e+00

Pos logit contributions
 lantern+0.006
le+0.006
 ho+0.006
adies+0.006
el+0.005
Neg logit contributions
arre-0.019
igsaw-0.018
 courtroom-0.017
 fault-0.017
 Working-0.017
Token she
Token ID673
Feature activation-2.36e+00

Pos logit contributions
adies+0.032
Mom+0.029
 lantern+0.029
 ho+0.028
 carriers+0.028
Neg logit contributions
arre-0.039
igsaw-0.033
ravings-0.032
wheel-0.032
 fault-0.032
Token accidentally
Token ID14716
Feature activation-5.12e+00

Pos logit contributions
arre+0.079
igsaw+0.078
 fault+0.071
 winning+0.069
 courtroom+0.069
Neg logit contributions
adies-0.062
le-0.055
Mom-0.055
 lantern-0.055
 ho-0.053
Token spilled
Token ID31896
Feature activation-2.73e+00

Pos logit contributions
arre+0.168
igsaw+0.161
 courtroom+0.145
 winning+0.141
ravings+0.141
Neg logit contributions
adies-0.142
 lantern-0.132
 ho-0.129
le-0.123
Mom-0.122
Token some
Token ID617
Feature activation+4.00e-01

Pos logit contributions
arre+0.102
 courtroom+0.094
igsaw+0.092
ravings+0.090
seek+0.088
Neg logit contributions
 lantern-0.057
adies-0.057
Mom-0.055
el-0.054
 ho-0.051
Token on
Token ID319
Feature activation-1.18e+00

Pos logit contributions
adies+0.010
 lantern+0.009
 carriers+0.009
Mom+0.008
le+0.008
Neg logit contributions
arre-0.014
igsaw-0.013
 courtroom-0.012
ravings-0.012
 winning-0.012
Token her
Token ID607
Feature activation-8.37e-01

Pos logit contributions
arre+0.050
ravings+0.042
wheel+0.042
 courtroom+0.042
igsaw+0.041
Neg logit contributions
Mom-0.023
 lantern-0.022
el-0.022
adies-0.021
 ho-0.019
Token hand
Token ID1021
Feature activation-1.79e-01

Pos logit contributions
arre+0.032
igsaw+0.028
 courtroom+0.027
wheel+0.027
 winning+0.027
Neg logit contributions
adies-0.019
 lantern-0.018
 carriers-0.017
 ho-0.017
Mom-0.016
Token fish
Token ID5916
Feature activation+6.26e-01

Pos logit contributions
adies+0.003
 carriers+0.003
Mom+0.003
 lantern+0.003
Ho+0.002
Neg logit contributions
arre-0.007
igsaw-0.007
ravings-0.007
 court-0.007
 courtroom-0.007
Token."
Token ID526
Feature activation-1.30e+00

Pos logit contributions
adies+0.012
le+0.010
 ho+0.010
 lantern+0.010
 carriers+0.009
Neg logit contributions
arre-0.025
igsaw-0.024
 courtroom-0.024
Em-0.023
ravings-0.023
Token
Token ID198
Feature activation-4.97e-01

Pos logit contributions
arre+0.047
igsaw+0.046
 courtroom+0.044
ravings+0.042
 fault+0.041
Neg logit contributions
adies-0.029
Mom-0.026
 lantern-0.026
el-0.026
 ho-0.025
Token
Token ID198
Feature activation+6.72e-02

Pos logit contributions
arre+0.014
igsaw+0.012
 courtroom+0.012
ravings+0.012
 fault+0.012
Neg logit contributions
adies-0.016
Mom-0.015
 lantern-0.014
 carriers-0.014
Dad-0.014
TokenO
Token ID46
Feature activation-5.29e-01

Pos logit contributions
adies+0.002
 carriers+0.002
 lantern+0.002
Mom+0.002
 ho+0.002
Neg logit contributions
arre-0.002
igsaw-0.002
ravings-0.002
 courtroom-0.002
wheel-0.002
Tokenll
Token ID297
Feature activation+3.37e+00

Pos logit contributions
arre+0.015
igsaw+0.013
ravings+0.013
wheel+0.012
 fault+0.012
Neg logit contributions
adies-0.017
 carriers-0.016
 lantern-0.015
Mom-0.015
 ho-0.014
Tokenie
Token ID494
Feature activation-1.00e+00

Pos logit contributions
adies+0.066
 carriers+0.060
 lantern+0.054
Mom+0.051
works+0.050
Neg logit contributions
arre-0.132
 accidentally-0.118
ravings-0.117
 fault-0.117
igsaw-0.115
Token was
Token ID373
Feature activation-6.75e-02

Pos logit contributions
igsaw+0.035
arre+0.034
 fault+0.032
 winning+0.031
ravings+0.031
Neg logit contributions
adies-0.024
le-0.022
Mom-0.022
 ho-0.021
 lantern-0.020
Token sad
Token ID6507
Feature activation-2.03e+00

Pos logit contributions
igsaw+0.003
con+0.003
enny+0.003
arre+0.003
 salon+0.003
Neg logit contributions
 close-0.001
Mom-0.001
adies-0.001
 ho-0.001
 closer-0.001
Token,
Token ID11
Feature activation-1.05e+00

Pos logit contributions
 courtroom+0.088
igsaw+0.087
arre+0.084
wheel+0.081
 disl+0.078
Neg logit contributions
le-0.029
 ho-0.026
adies-0.025
ies-0.025
el-0.022
Token but
Token ID475
Feature activation-3.02e-02

Pos logit contributions
igsaw+0.040
arre+0.040
 courtroom+0.035
ravings+0.035
 fault+0.035
Neg logit contributions
 ho-0.023
adies-0.022
 lantern-0.021
Mom-0.021
Run-0.020
Token.
Token ID13
Feature activation+3.09e-01

Pos logit contributions
adies+0.104
 carriers+0.092
 lantern+0.090
 sle+0.088
works+0.087
Neg logit contributions
arre-0.088
 winning-0.076
 about-0.076
 fault-0.074
 accidentally-0.073
Token They
Token ID1119
Feature activation+1.57e+00

Pos logit contributions
adies+0.009
 carriers+0.008
Mom+0.008
 lantern+0.008
Dad+0.008
Neg logit contributions
arre-0.009
igsaw-0.008
wheel-0.008
ravings-0.008
 courtroom-0.008
Token say
Token ID910
Feature activation+1.70e+00

Pos logit contributions
adies+0.034
 lantern+0.029
 carriers+0.028
 ho+0.028
Mom+0.027
Neg logit contributions
arre-0.059
igsaw-0.055
ravings-0.053
 courtroom-0.053
 winning-0.051
Token,
Token ID11
Feature activation-4.86e-01

Pos logit contributions
adies+0.045
 carriers+0.040
 lantern+0.039
Mom+0.037
 ho+0.037
Neg logit contributions
arre-0.057
igsaw-0.051
ravings-0.049
 courtroom-0.049
wheel-0.049
Token "
Token ID366
Feature activation+6.53e-02

Pos logit contributions
arre+0.014
igsaw+0.013
 fault+0.012
 courtroom+0.012
ravings+0.012
Neg logit contributions
adies-0.014
Mom-0.014
 lantern-0.013
 carriers-0.013
 ho-0.013
TokenOK
Token ID11380
Feature activation+3.69e+00

Pos logit contributions
adies+0.002
 lantern+0.001
 sle+0.001
 aliens+0.001
 carriers+0.001
Neg logit contributions
arre-0.002
wheel-0.002
ravings-0.002
 courtroom-0.002
 accidentally-0.002
Token,
Token ID11
Feature activation+1.69e-01

Pos logit contributions
adies+0.129
 carriers+0.115
Ho+0.115
 lantern+0.114
 sle+0.112
Neg logit contributions
arre-0.086
 winning-0.078
 accidentally-0.076
 court-0.074
 about-0.073
Token Mom
Token ID11254
Feature activation+3.06e+00

Pos logit contributions
adies+0.004
 carriers+0.003
 lantern+0.003
Mom+0.003
le+0.003
Neg logit contributions
arre-0.006
igsaw-0.006
ravings-0.006
 winning-0.006
wheel-0.006
Token.
Token ID13
Feature activation-4.13e-02

Pos logit contributions
adies+0.080
 carriers+0.070
 lantern+0.068
Mom+0.066
 ho+0.064
Neg logit contributions
arre-0.104
igsaw-0.093
ravings-0.090
 courtroom-0.090
wheel-0.089
Token We
Token ID775
Feature activation-5.82e-01

Pos logit contributions
arre+0.001
 fault+0.001
 winning+0.001
 By+0.001
igsaw+0.001
Neg logit contributions
adies-0.001
 lantern-0.001
le-0.001
agles-0.001
iri-0.001
Token will
Token ID481
Feature activation+2.26e+00

Pos logit contributions
arre+0.025
igsaw+0.023
ravings+0.022
 courtroom+0.022
wheel+0.022
Neg logit contributions
adies-0.010
 carriers-0.008
 lantern-0.008
Mom-0.008
 ho-0.008
TokenSh
Token ID2484
Feature activation+7.41e-01

Pos logit contributions
arre+0.008
wheel+0.007
igsaw+0.007
 court+0.007
 courtroom+0.007
Neg logit contributions
adies-0.006
 carriers-0.005
 lantern-0.005
le-0.005
 ho-0.005
Tokenh
Token ID71
Feature activation-1.47e-01

Pos logit contributions
adies+0.022
 carriers+0.019
 lantern+0.019
Mom+0.018
 ho+0.018
Neg logit contributions
arre-0.023
igsaw-0.020
ravings-0.020
 courtroom-0.019
wheel-0.019
Token,
Token ID11
Feature activation+2.40e-01

Pos logit contributions
arre+0.005
igsaw+0.004
ravings+0.004
wheel+0.004
 court+0.004
Neg logit contributions
adies-0.004
 carriers-0.004
 lantern-0.004
Mom-0.003
 ho-0.003
Token Tom
Token ID4186
Feature activation+2.10e+00

Pos logit contributions
adies+0.007
 carriers+0.006
 lantern+0.006
Mom+0.006
 ho+0.005
Neg logit contributions
arre-0.008
igsaw-0.007
ravings-0.007
 courtroom-0.007
wheel-0.007
Token,
Token ID11
Feature activation-8.94e-02

Pos logit contributions
adies+0.057
 carriers+0.051
 lantern+0.049
Mom+0.048
 ho+0.046
Neg logit contributions
arre-0.069
igsaw-0.060
ravings-0.059
 courtroom-0.059
 court-0.058
Token don
Token ID836
Feature activation+2.71e+00

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 courtroom+0.003
 fault+0.003
Neg logit contributions
adies-0.002
 lantern-0.002
 carriers-0.002
 ho-0.002
Mom-0.002
Token't
Token ID470
Feature activation+2.41e-02

Pos logit contributions
adies+0.118
 carriers+0.113
 lantern+0.113
le+0.109
maid+0.109
Neg logit contributions
arre-0.040
wheel-0.037
 winning-0.033
 By-0.033
 court-0.031
Token be
Token ID307
Feature activation-1.16e+00

Pos logit contributions
adies+0.001
 carriers+0.001
 lantern+0.001
Mom+0.000
 ho+0.000
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 courtroom-0.001
wheel-0.001
Token rude
Token ID22066
Feature activation-2.16e+00

Pos logit contributions
igsaw+0.043
arre+0.041
ravings+0.038
 Unfortunately+0.037
 By+0.037
Neg logit contributions
 ho-0.028
adies-0.025
le-0.024
 close-0.022
 carriers-0.022
Token!"
Token ID2474
Feature activation-1.21e+00

Pos logit contributions
arre+0.073
igsaw+0.064
ravings+0.063
wheel+0.063
 fault+0.063
Neg logit contributions
adies-0.056
 carriers-0.050
 lantern-0.049
Mom-0.047
 ho-0.045
Token Jen
Token ID13364
Feature activation-9.85e-01

Pos logit contributions
arre+0.038
igsaw+0.034
ravings+0.033
 courtroom+0.033
wheel+0.032
Neg logit contributions
adies-0.035
 carriers-0.031
 lantern-0.030
Mom-0.029
 ho-0.028
Token a
Token ID257
Feature activation+6.40e-01

Pos logit contributions
adies+0.014
 carriers+0.013
 lantern+0.013
works+0.012
 sle+0.012
Neg logit contributions
arre-0.012
igsaw-0.010
 accidentally-0.010
 court-0.010
 winning-0.010
Token mom
Token ID1995
Feature activation+2.45e+00

Pos logit contributions
adies+0.024
 carriers+0.021
works+0.020
agles+0.020
le+0.020
Neg logit contributions
arre-0.015
 accidentally-0.014
 court-0.013
 courtroom-0.013
 basketball-0.013
Tokenmy
Token ID1820
Feature activation+2.13e+00

Pos logit contributions
adies+0.081
 carriers+0.073
 lantern+0.070
 ho+0.067
Mom+0.067
Neg logit contributions
arre-0.067
igsaw-0.056
ravings-0.056
 courtroom-0.055
wheel-0.055
Token,
Token ID11
Feature activation-1.97e-01

Pos logit contributions
adies+0.049
 lantern+0.046
Mom+0.044
 carriers+0.044
 ho+0.043
Neg logit contributions
arre-0.078
igsaw-0.074
 fault-0.068
ravings-0.068
 courtroom-0.067
Token a
Token ID257
Feature activation+1.02e+00

Pos logit contributions
arre+0.005
igsaw+0.004
ravings+0.004
 courtroom+0.004
 court+0.004
Neg logit contributions
adies-0.007
 carriers-0.006
 lantern-0.006
Mom-0.006
le-0.006
Token daddy
Token ID35822
Feature activation+2.98e+00

Pos logit contributions
adies+0.031
 carriers+0.027
 lantern+0.026
Mom+0.025
le+0.025
Neg logit contributions
arre-0.030
igsaw-0.027
 courtroom-0.026
ravings-0.026
 accidentally-0.026
Token and
Token ID290
Feature activation+9.60e-01

Pos logit contributions
adies+0.084
 carriers+0.075
 lantern+0.075
Mom+0.072
 ho+0.070
Neg logit contributions
arre-0.095
igsaw-0.084
ravings-0.082
 courtroom-0.081
wheel-0.080
Token a
Token ID257
Feature activation+1.10e+00

Pos logit contributions
adies+0.033
works+0.029
le+0.028
 carriers+0.028
 lantern+0.028
Neg logit contributions
arre-0.023
 accidentally-0.022
igsaw-0.020
wheel-0.020
 court-0.020
Token baby
Token ID5156
Feature activation+1.22e+00

Pos logit contributions
adies+0.032
 carriers+0.028
 lantern+0.027
le+0.026
Mom+0.026
Neg logit contributions
arre-0.034
igsaw-0.030
 courtroom-0.030
ravings-0.030
 accidentally-0.029
Token.
Token ID13
Feature activation+1.80e-01

Pos logit contributions
adies+0.036
 carriers+0.032
 lantern+0.030
Mom+0.030
works+0.029
Neg logit contributions
arre-0.037
 accidentally-0.032
 court-0.031
ravings-0.031
 courtroom-0.031
Token The
Token ID383
Feature activation+2.41e-01

Pos logit contributions
adies+0.006
 carriers+0.005
 lantern+0.005
Mom+0.005
 ho+0.005
Neg logit contributions
arre-0.005
igsaw-0.004
wheel-0.004
ravings-0.004
 courtroom-0.004
Token The
Token ID383
Feature activation-1.04e-03

Pos logit contributions
Mom+0.016
adies+0.014
 ho+0.014
Ho+0.014
 carriers+0.014
Neg logit contributions
arre-0.023
ravings-0.020
igsaw-0.020
 fault-0.020
 supposed-0.019
Token bird
Token ID6512
Feature activation-7.11e-01

Pos logit contributions
arre+0.000
igsaw+0.000
ravings+0.000
wheel+0.000
 By+0.000
Neg logit contributions
adies-0.000
Mom-0.000
 lantern-0.000
 carriers-0.000
 ho-0.000
Token said
Token ID531
Feature activation+3.78e-01

Pos logit contributions
arre+0.026
igsaw+0.026
 courtroom+0.024
ravings+0.024
 court+0.023
Neg logit contributions
le-0.016
adies-0.016
 lantern-0.015
 ho-0.015
el-0.014
Token,
Token ID11
Feature activation-1.24e+00

Pos logit contributions
adies+0.010
 carriers+0.009
 lantern+0.009
Mom+0.008
 ho+0.008
Neg logit contributions
arre-0.013
igsaw-0.012
ravings-0.011
 courtroom-0.011
 court-0.011
Token "
Token ID366
Feature activation-4.76e-01

Pos logit contributions
arre+0.041
igsaw+0.040
ravings+0.037
 fault+0.036
handed+0.036
Neg logit contributions
 ho-0.033
Mom-0.031
adies-0.031
 lantern-0.030
 hello-0.029
TokenYes
Token ID5297
Feature activation+2.76e+00

Pos logit contributions
arre+0.017
igsaw+0.015
wheel+0.014
ravings+0.014
 courtroom+0.014
Neg logit contributions
adies-0.012
 carriers-0.011
 lantern-0.010
Mom-0.010
 ho-0.010
Token,
Token ID11
Feature activation-6.58e-01

Pos logit contributions
adies+0.087
 carriers+0.081
 lantern+0.078
Mom+0.073
 Amen+0.073
Neg logit contributions
arre-0.074
igsaw-0.065
 winning-0.065
 accidentally-0.065
 court-0.064
Token let
Token ID1309
Feature activation+2.03e+00

Pos logit contributions
arre+0.023
igsaw+0.020
ravings+0.019
 courtroom+0.019
 fault+0.019
Neg logit contributions
adies-0.017
 carriers-0.015
 lantern-0.015
Mom-0.015
 ho-0.014
Token's
Token ID338
Feature activation+3.83e-01

Pos logit contributions
adies+0.074
 carriers+0.069
 lantern+0.068
works+0.065
 sle+0.063
Neg logit contributions
arre-0.051
 accidentally-0.040
 courtroom-0.038
ravings-0.038
 court-0.037
Token play
Token ID711
Feature activation+3.72e-02

Pos logit contributions
adies+0.008
 lantern+0.007
 carriers+0.007
Mom+0.006
 ho+0.006
Neg logit contributions
arre-0.015
igsaw-0.014
ravings-0.013
 courtroom-0.013
 fault-0.013
Token!"
Token ID2474
Feature activation-9.55e-01

Pos logit contributions
adies+0.001
 carriers+0.001
le+0.001
Mom+0.001
 aliens+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 courtroom-0.001
 fault-0.001
Token!"
Token ID2474
Feature activation+4.10e-02

Pos logit contributions
adies+0.020
 carriers+0.017
 lantern+0.017
Mom+0.016
 ho+0.016
Neg logit contributions
arre-0.031
igsaw-0.027
ravings-0.027
 courtroom-0.027
wheel-0.026
Token
Token ID198
Feature activation+9.85e-01

Pos logit contributions
adies+0.001
Mom+0.001
Dad+0.001
 lantern+0.001
 carriers+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 fault-0.001
wheel-0.001
Token
Token ID198
Feature activation+1.49e+00

Pos logit contributions
adies+0.035
 carriers+0.032
 lantern+0.032
Mom+0.032
 ho+0.030
Neg logit contributions
arre-0.024
igsaw-0.021
ravings-0.020
 courtroom-0.019
 fault-0.019
TokenL
Token ID43
Feature activation-5.10e-01

Pos logit contributions
adies+0.041
 carriers+0.036
 lantern+0.035
 ho+0.033
works+0.033
Neg logit contributions
arre-0.048
igsaw-0.043
 court-0.042
 courtroom-0.042
wheel-0.041
Tokenily
Token ID813
Feature activation-1.03e+00

Pos logit contributions
arre+0.015
igsaw+0.013
 courtroom+0.013
ravings+0.013
 winning+0.013
Neg logit contributions
adies-0.016
 carriers-0.014
 lantern-0.013
Mom-0.013
le-0.013
Token and
Token ID290
Feature activation+1.50e-01

Pos logit contributions
igsaw+0.038
arre+0.038
 winning+0.036
ravings+0.035
 courtroom+0.035
Neg logit contributions
adies-0.021
 carriers-0.018
Mom-0.017
 lantern-0.017
le-0.017
Token Ben
Token ID3932
Feature activation+4.91e-01

Pos logit contributions
adies+0.004
 ho+0.004
Mom+0.003
 carriers+0.003
 lantern+0.003
Neg logit contributions
arre-0.005
igsaw-0.005
ravings-0.004
wheel-0.004
 courtroom-0.004
Token did
Token ID750
Feature activation+8.89e-01

Pos logit contributions
adies+0.013
 carriers+0.012
 lantern+0.011
Mom+0.011
 sle+0.010
Neg logit contributions
arre-0.017
 court-0.014
igsaw-0.014
 courtroom-0.014
ravings-0.014
Token not
Token ID407
Feature activation-4.74e-01

Pos logit contributions
adies+0.027
 lantern+0.023
 carriers+0.022
Mom+0.022
le+0.022
Neg logit contributions
arre-0.027
igsaw-0.024
 courtroom-0.023
ravings-0.023
wheel-0.022
Token want
Token ID765
Feature activation+1.40e-02

Pos logit contributions
arre+0.016
igsaw+0.014
ravings+0.014
 courtroom+0.014
wheel+0.013
Neg logit contributions
adies-0.013
 carriers-0.011
 lantern-0.011
Mom-0.011
 ho-0.010
Token to
Token ID284
Feature activation-1.53e+00

Pos logit contributions
adies+0.000
 carriers+0.000
 lantern+0.000
 aliens+0.000
Mom+0.000
Neg logit contributions
arre-0.000
igsaw-0.000
wheel-0.000
ravings-0.000
 courtroom-0.000
Token.
Token ID13
Feature activation+4.09e-01

Pos logit contributions
adies+0.020
 carriers+0.017
le+0.017
Mom+0.016
 ho+0.016
Neg logit contributions
arre-0.036
igsaw-0.035
 courtroom-0.034
ravings-0.032
wheel-0.032
Token We
Token ID775
Feature activation-8.53e-02

Pos logit contributions
adies+0.015
 lantern+0.014
le+0.014
 carriers+0.013
agles+0.013
Neg logit contributions
arre-0.008
 winning-0.007
igsaw-0.007
 By-0.007
 fault-0.007
Token can
Token ID460
Feature activation+1.69e+00

Pos logit contributions
arre+0.004
igsaw+0.004
ravings+0.004
 courtroom+0.003
wheel+0.003
Neg logit contributions
adies-0.001
 lantern-0.001
Mom-0.001
 carriers-0.001
 ho-0.001
Token share
Token ID2648
Feature activation+1.44e+00

Pos logit contributions
adies+0.053
 carriers+0.048
 lantern+0.046
 ho+0.044
 sle+0.044
Neg logit contributions
arre-0.049
igsaw-0.041
wheel-0.041
 winning-0.041
 court-0.041
Token our
Token ID674
Feature activation+2.54e-02

Pos logit contributions
adies+0.041
 carriers+0.036
 lantern+0.036
Mom+0.034
 ho+0.034
Neg logit contributions
arre-0.045
igsaw-0.039
 court-0.039
ravings-0.039
 courtroom-0.038
Token toys
Token ID14958
Feature activation+7.07e-01

Pos logit contributions
adies+0.001
 carriers+0.001
 lantern+0.001
Mom+0.001
 ho+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 courtroom-0.001
 court-0.001
Token and
Token ID290
Feature activation+9.89e-01

Pos logit contributions
adies+0.014
 carriers+0.014
Mom+0.013
 lantern+0.012
le+0.012
Neg logit contributions
arre-0.028
igsaw-0.026
 courtroom-0.025
ravings-0.025
 accidentally-0.024
Token have
Token ID423
Feature activation+9.00e-03

Pos logit contributions
adies+0.021
 carriers+0.021
 lantern+0.021
Mom+0.020
 ho+0.019
Neg logit contributions
arre-0.038
igsaw-0.034
ravings-0.032
 fault-0.032
 courtroom-0.032
Token fun
Token ID1257
Feature activation-4.99e-01

Pos logit contributions
adies+0.000
 carriers+0.000
 lantern+0.000
Mom+0.000
 ho+0.000
Neg logit contributions
arre-0.000
igsaw-0.000
ravings-0.000
wheel-0.000
 courtroom-0.000
Token."
Token ID526
Feature activation-7.60e-02

Pos logit contributions
arre+0.018
igsaw+0.017
 courtroom+0.017
ravings+0.016
wheel+0.016
Neg logit contributions
adies-0.012
 carriers-0.010
Mom-0.010
 lantern-0.010
le-0.009
Token
Token ID198
Feature activation+7.57e-01

Pos logit contributions
arre+0.002
igsaw+0.002
ravings+0.002
wheel+0.002
 supposed+0.002
Neg logit contributions
adies-0.002
Mom-0.002
 carriers-0.002
Dad-0.002
 lantern-0.002
Token."
Token ID526
Feature activation-3.54e-01

Pos logit contributions
arre+0.012
igsaw+0.011
 courtroom+0.011
 fault+0.010
 winning+0.010
Neg logit contributions
adies-0.007
le-0.006
 ho-0.006
 lantern-0.005
 carriers-0.005
Token
Token ID198
Feature activation+5.83e-01

Pos logit contributions
arre+0.011
igsaw+0.010
 winning+0.009
wheel+0.009
 courtroom+0.009
Neg logit contributions
adies-0.010
Mom-0.010
 carriers-0.009
 lantern-0.009
Dad-0.009
Token
Token ID198
Feature activation+1.03e+00

Pos logit contributions
adies+0.020
Mom+0.019
 carriers+0.019
 lantern+0.018
Ho+0.017
Neg logit contributions
arre-0.015
igsaw-0.013
 courtroom-0.012
ravings-0.012
 fault-0.012
TokenHe
Token ID1544
Feature activation+1.54e+00

Pos logit contributions
adies+0.029
 carriers+0.025
 lantern+0.025
Mom+0.023
 ho+0.023
Neg logit contributions
arre-0.034
igsaw-0.029
ravings-0.029
 courtroom-0.029
wheel-0.028
Token goes
Token ID2925
Feature activation-1.08e-01

Pos logit contributions
adies+0.042
 carriers+0.038
 lantern+0.037
Mom+0.035
works+0.034
Neg logit contributions
arre-0.050
igsaw-0.043
ravings-0.043
 courtroom-0.042
 court-0.042
Token to
Token ID284
Feature activation+1.01e+00

Pos logit contributions
arre+0.003
igsaw+0.003
 fault+0.002
ravings+0.002
 courtroom+0.002
Neg logit contributions
adies-0.004
 carriers-0.003
Mom-0.003
 lantern-0.003
 ho-0.003
Token the
Token ID262
Feature activation+9.09e-01

Pos logit contributions
adies+0.022
 lantern+0.021
Mom+0.021
 ho+0.020
 God+0.019
Neg logit contributions
arre-0.041
igsaw-0.035
ravings-0.034
wheel-0.033
 fault-0.033
Token pig
Token ID12967
Feature activation-3.03e-01

Pos logit contributions
adies+0.018
Mom+0.018
 carriers+0.018
 lantern+0.018
 ho+0.016
Neg logit contributions
arre-0.037
igsaw-0.035
 fault-0.032
wheel-0.032
 courtroom-0.031
Token.
Token ID13
Feature activation+9.43e-01

Pos logit contributions
arre+0.011
igsaw+0.011
 winning+0.010
 fault+0.010
wheel+0.010
Neg logit contributions
 lantern-0.006
adies-0.006
 carriers-0.006
 ho-0.006
 sle-0.005
Token The
Token ID383
Feature activation+5.03e-01

Pos logit contributions
adies+0.024
Mom+0.021
 carriers+0.020
 lantern+0.020
 God+0.019
Neg logit contributions
arre-0.036
igsaw-0.029
 supposed-0.029
 practicing-0.029
 fault-0.028
Token pig
Token ID12967
Feature activation-8.11e-01

Pos logit contributions
adies+0.012
 carriers+0.010
 lantern+0.010
Mom+0.009
 ho+0.009
Neg logit contributions
arre-0.019
igsaw-0.017
ravings-0.016
 courtroom-0.016
 court-0.016
Token's
Token ID338
Feature activation-6.30e-01

Pos logit contributions
arre+0.043
igsaw+0.042
 courtroom+0.041
 court+0.040
ravings+0.040
Neg logit contributions
adies-0.024
le-0.021
 lantern-0.020
 ho-0.020
 carriers-0.019
Token raining
Token ID43079
Feature activation-1.74e+00

Pos logit contributions
igsaw+0.021
arre+0.021
 courtroom+0.020
 gol+0.019
 court+0.019
Neg logit contributions
adies-0.015
 ho-0.014
 aliens-0.013
Mom-0.013
 lantern-0.013
Token!
Token ID0
Feature activation-6.66e-01

Pos logit contributions
arre+0.060
igsaw+0.053
 courtroom+0.053
ravings+0.053
 court+0.051
Neg logit contributions
adies-0.045
 carriers-0.038
 lantern-0.037
Mom-0.036
le-0.035
Token It
Token ID632
Feature activation+1.30e-01

Pos logit contributions
arre+0.023
igsaw+0.021
ravings+0.020
 fault+0.020
 court+0.020
Neg logit contributions
adies-0.017
Mom-0.015
 carriers-0.015
 Amen-0.014
 lantern-0.014
Token's
Token ID338
Feature activation+1.67e-01

Pos logit contributions
adies+0.003
 carriers+0.003
 lantern+0.003
 ho+0.003
Mom+0.003
Neg logit contributions
arre-0.004
igsaw-0.004
ravings-0.004
 courtroom-0.004
 court-0.004
Token not
Token ID407
Feature activation-3.74e-02

Pos logit contributions
adies+0.004
 ho+0.004
Mom+0.004
 aliens+0.004
 lantern+0.004
Neg logit contributions
igsaw-0.006
arre-0.005
 courtroom-0.005
 court-0.005
handed-0.005
Token fair
Token ID3148
Feature activation-1.82e+00

Pos logit contributions
arre+0.001
igsaw+0.001
 courtroom+0.001
ravings+0.001
 court+0.001
Neg logit contributions
adies-0.001
 ho-0.001
 lantern-0.001
 carriers-0.001
 aliens-0.001
Token!"
Token ID2474
Feature activation-1.61e+00

Pos logit contributions
arre+0.065
 courtroom+0.062
igsaw+0.061
 court+0.059
wheel+0.058
Neg logit contributions
adies-0.042
 carriers-0.037
 ho-0.037
 lantern-0.036
le-0.036
Token
Token ID198
Feature activation-5.83e-01

Pos logit contributions
arre+0.050
igsaw+0.048
 supposed+0.047
wheel+0.047
 winning+0.046
Neg logit contributions
Mom-0.042
 lantern-0.040
el-0.039
Dad-0.039
 ho-0.039
Token
Token ID198
Feature activation-1.65e-03

Pos logit contributions
arre+0.019
igsaw+0.017
ravings+0.017
 fault+0.016
 winning+0.016
Neg logit contributions
Mom-0.017
adies-0.015
Dad-0.015
 lantern-0.014
Ho-0.014
TokenHis
Token ID6653
Feature activation+6.85e-01

Pos logit contributions
arre+0.000
igsaw+0.000
 fault+0.000
ravings+0.000
wheel+0.000
Neg logit contributions
Mom-0.000
adies-0.000
 carriers-0.000
 lantern-0.000
Dad-0.000
Token I
Token ID314
Feature activation-1.01e+00

Pos logit contributions
arre+0.000
igsaw+0.000
ravings+0.000
 fault+0.000
 winning+0.000
Neg logit contributions
adies-0.000
 carriers-0.000
 lantern-0.000
Mom-0.000
le-0.000
Token'm
Token ID1101
Feature activation-9.54e-02

Pos logit contributions
arre+0.037
igsaw+0.035
 courtroom+0.034
ravings+0.034
 court+0.034
Neg logit contributions
adies-0.022
 lantern-0.020
le-0.020
Mom-0.019
 Amen-0.019
Token glad
Token ID9675
Feature activation+1.90e+00

Pos logit contributions
arre+0.003
igsaw+0.003
ravings+0.003
 courtroom+0.003
seek+0.003
Neg logit contributions
adies-0.003
 Amen-0.003
 lantern-0.003
 God-0.003
 ho-0.002
Token you
Token ID345
Feature activation-7.22e-01

Pos logit contributions
adies+0.041
 lantern+0.035
 carriers+0.035
Mom+0.034
 ho+0.033
Neg logit contributions
arre-0.075
igsaw-0.067
 courtroom-0.065
ravings-0.065
wheel-0.064
Token enjoyed
Token ID8359
Feature activation-1.71e-01

Pos logit contributions
arre+0.021
igsaw+0.018
ravings+0.018
wheel+0.018
 courtroom+0.018
Neg logit contributions
adies-0.022
 carriers-0.020
 lantern-0.019
Mom-0.019
 ho-0.018
Token it
Token ID340
Feature activation+2.77e-01

Pos logit contributions
arre+0.007
igsaw+0.006
wheel+0.006
 courtroom+0.005
ravings+0.005
Neg logit contributions
adies-0.004
 lantern-0.004
 ho-0.004
 carriers-0.003
Mom-0.003
Token.
Token ID13
Feature activation+4.80e-02

Pos logit contributions
adies+0.007
 carriers+0.006
 lantern+0.006
Mom+0.006
le+0.006
Neg logit contributions
arre-0.010
igsaw-0.009
 courtroom-0.009
ravings-0.009
 court-0.008
Token Do
Token ID2141
Feature activation+2.43e+00

Pos logit contributions
adies+0.002
 carriers+0.002
 lantern+0.002
le+0.001
 ho+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
ravings-0.001
 winning-0.001
wheel-0.001
Token you
Token ID345
Feature activation-9.60e-01

Pos logit contributions
adies+0.063
 carriers+0.058
 lantern+0.057
 sle+0.052
le+0.052
Neg logit contributions
arre-0.081
 winning-0.072
igsaw-0.071
 court-0.071
 accidentally-0.070
Token want
Token ID765
Feature activation+9.74e-01

Pos logit contributions
arre+0.028
igsaw+0.023
wheel+0.023
 accidentally+0.023
 fault+0.023
Neg logit contributions
adies-0.029
 carriers-0.028
 lantern-0.027
Mom-0.025
 sle-0.025
Token to
Token ID284
Feature activation-6.32e-01

Pos logit contributions
adies+0.035
 lantern+0.032
Mom+0.031
 carriers+0.031
le+0.030
Neg logit contributions
arre-0.026
igsaw-0.021
 courtroom-0.020
ravings-0.020
 fault-0.019
Token her
Token ID607
Feature activation-4.70e-01

Pos logit contributions
adies+0.012
le+0.010
 carriers+0.010
el+0.010
Mom+0.010
Neg logit contributions
arre-0.019
igsaw-0.017
ravings-0.017
wheel-0.017
 courtroom-0.016
Token a
Token ID257
Feature activation-3.09e-01

Pos logit contributions
arre+0.021
ravings+0.019
 winning+0.018
wheel+0.018
 courtroom+0.018
Neg logit contributions
adies-0.006
 lantern-0.006
 carriers-0.006
le-0.005
Mom-0.005
Token snack
Token ID26906
Feature activation+1.38e-01

Pos logit contributions
arre+0.012
 winning+0.011
ravings+0.011
seek+0.011
 losing+0.011
Neg logit contributions
 lantern-0.006
adies-0.005
 comet-0.005
 ho-0.005
Mom-0.005
Token and
Token ID290
Feature activation+2.68e-02

Pos logit contributions
 carriers+0.003
adies+0.002
Mom+0.002
 lantern+0.002
le+0.002
Neg logit contributions
arre-0.006
 courtroom-0.005
 winning-0.005
 losing-0.005
 disl-0.005
Token said
Token ID531
Feature activation-4.75e-01

Pos logit contributions
 lantern+0.001
 carriers+0.001
Mom+0.001
adies+0.001
 ho+0.001
Neg logit contributions
arre-0.001
igsaw-0.001
 fault-0.001
 courtroom-0.001
ravings-0.001
Token,
Token ID11
Feature activation-1.68e+00

Pos logit contributions
arre+0.020
ravings+0.018
 courtroom+0.018
igsaw+0.017
 fault+0.017
Neg logit contributions
adies-0.009
 lantern-0.008
 carriers-0.007
Mom-0.007
 hello-0.007
Token "
Token ID366
Feature activation-4.53e-01

Pos logit contributions
arre+0.054
igsaw+0.051
 fault+0.050
ravings+0.049
 courtroom+0.049
Neg logit contributions
Mom-0.044
 lantern-0.043
adies-0.042
 ho-0.042
 Amen-0.040
TokenNow
Token ID3844
Feature activation+5.35e-01

Pos logit contributions
arre+0.015
igsaw+0.013
ravings+0.013
 courtroom+0.013
 fault+0.013
Neg logit contributions
adies-0.012
Mom-0.011
 carriers-0.011
 lantern-0.010
Ho-0.010
Token you
Token ID345
Feature activation-1.27e+00

Pos logit contributions
adies+0.012
 lantern+0.010
 carriers+0.010
 ho+0.010
Mom+0.009
Neg logit contributions
arre-0.021
igsaw-0.018
ravings-0.018
 courtroom-0.018
wheel-0.018
Token can
Token ID460
Feature activation-2.31e-01

Pos logit contributions
arre+0.047
igsaw+0.043
ravings+0.042
 courtroom+0.042
 winning+0.041
Neg logit contributions
adies-0.029
 carriers-0.025
 lantern-0.024
Mom-0.023
 ho-0.023
Token play
Token ID711
Feature activation-4.77e-01

Pos logit contributions
arre+0.008
igsaw+0.007
 courtroom+0.007
 fault+0.007
ravings+0.007
Neg logit contributions
adies-0.006
Mom-0.005
 lantern-0.005
 carriers-0.005
le-0.005
Token sorry
Token ID7926
Feature activation+1.64e+00

Pos logit contributions
adies+0.010
 carriers+0.009
 lantern+0.009
Mom+0.008
 ho+0.008
Neg logit contributions
arre-0.014
igsaw-0.013
ravings-0.012
 courtroom-0.012
wheel-0.012
Token.
Token ID13
Feature activation+6.45e-01

Pos logit contributions
adies+0.045
 carriers+0.040
 lantern+0.039
Mom+0.038
 ho+0.037
Neg logit contributions
arre-0.054
igsaw-0.048
 courtroom-0.047
ravings-0.046
 court-0.045
Token He
Token ID679
Feature activation+1.71e+00

Pos logit contributions
adies+0.019
Mom+0.018
 carriers+0.018
 lantern+0.017
Dad+0.017
Neg logit contributions
arre-0.020
igsaw-0.018
ravings-0.017
wheel-0.017
 courtroom-0.016
Token asks
Token ID7893
Feature activation+7.96e-01

Pos logit contributions
adies+0.045
 carriers+0.039
 lantern+0.039
Mom+0.037
 ho+0.036
Neg logit contributions
arre-0.058
igsaw-0.052
ravings-0.050
 courtroom-0.050
wheel-0.050
Token if
Token ID611
Feature activation+3.56e+00

Pos logit contributions
adies+0.027
 carriers+0.025
 lantern+0.024
Mom+0.023
works+0.023
Neg logit contributions
arre-0.019
 court-0.017
igsaw-0.017
ravings-0.016
 winning-0.016
Token he
Token ID339
Feature activation-2.19e+00

Pos logit contributions
adies+0.111
 carriers+0.100
 lantern+0.098
Mom+0.095
 ho+0.093
Neg logit contributions
arre-0.101
igsaw-0.088
ravings-0.085
 courtroom-0.085
 court-0.084
Token can
Token ID460
Feature activation+2.15e-01

Pos logit contributions
arre+0.078
ravings+0.068
igsaw+0.068
 accidentally+0.067
wheel+0.067
Neg logit contributions
adies-0.054
 carriers-0.047
 lantern-0.046
Mom-0.043
 ho-0.042
Token play
Token ID711
Feature activation-3.06e-01

Pos logit contributions
adies+0.006
 carriers+0.006
 lantern+0.005
Mom+0.005
 ho+0.005
Neg logit contributions
arre-0.007
igsaw-0.006
 court-0.006
wheel-0.006
ravings-0.006
Token with
Token ID351
Feature activation+1.14e+00

Pos logit contributions
arre+0.011
 fault+0.010
igsaw+0.010
ravings+0.010
 courtroom+0.009
Neg logit contributions
adies-0.008
 carriers-0.007
Mom-0.006
 lantern-0.006
le-0.006
Token them
Token ID606
Feature activation+6.88e-01

Pos logit contributions
adies+0.033
 carriers+0.029
 lantern+0.029
Mom+0.027
works+0.027
Neg logit contributions
arre-0.035
igsaw-0.031
 court-0.030
ravings-0.030
 courtroom-0.030
Token.
Token ID13
Feature activation+6.80e-01

Pos logit contributions
adies+0.013
le+0.012
 carriers+0.012
Mom+0.011
 lantern+0.011
Neg logit contributions
arre-0.027
igsaw-0.026
 courtroom-0.026
ravings-0.024
 disl-0.024
Token "
Token ID366
Feature activation-6.11e-01

Pos logit contributions
arre+0.063
igsaw+0.060
ravings+0.057
 fault+0.056
handed+0.054
Neg logit contributions
adies-0.050
Mom-0.049
 ho-0.048
 lantern-0.047
Dad-0.044
TokenMom
Token ID29252
Feature activation+2.51e+00

Pos logit contributions
arre+0.023
wheel+0.020
igsaw+0.020
 courtroom+0.020
ravings+0.019
Neg logit contributions
adies-0.014
 lantern-0.012
 carriers-0.012
 ho-0.011
 sle-0.011
Tokenmy
Token ID1820
Feature activation+1.43e+00

Pos logit contributions
adies+0.059
Mom+0.051
 ho+0.051
 lantern+0.051
 carriers+0.050
Neg logit contributions
arre-0.089
igsaw-0.082
 courtroom-0.080
ravings-0.079
 court-0.078
Token,
Token ID11
Feature activation-2.98e-01

Pos logit contributions
adies+0.033
 lantern+0.028
 carriers+0.028
 ho+0.028
Mom+0.027
Neg logit contributions
arre-0.052
igsaw-0.048
ravings-0.046
 courtroom-0.046
wheel-0.045
Token can
Token ID460
Feature activation+1.64e+00

Pos logit contributions
arre+0.010
igsaw+0.009
ravings+0.009
 courtroom+0.009
wheel+0.008
Neg logit contributions
adies-0.008
 lantern-0.007
 carriers-0.007
Mom-0.007
 ho-0.007
Token I
Token ID314
Feature activation-1.45e+00

Pos logit contributions
adies+0.046
 carriers+0.040
 lantern+0.039
le+0.036
Ho+0.036
Neg logit contributions
arre-0.052
igsaw-0.047
wheel-0.046
 accidentally-0.046
 winning-0.046
Token have
Token ID423
Feature activation-1.47e+00

Pos logit contributions
arre+0.053
igsaw+0.048
ravings+0.047
 fault+0.046
 courtroom+0.045
Neg logit contributions
adies-0.034
Mom-0.031
 lantern-0.030
 ho-0.029
 carriers-0.029
Token that
Token ID326
Feature activation-9.62e-01

Pos logit contributions
arre+0.064
ravings+0.056
igsaw+0.056
 colony+0.054
isk+0.052
Neg logit contributions
adies-0.026
 lantern-0.024
Mom-0.024
 ho-0.022
 comet-0.021
Token toy
Token ID13373
Feature activation-6.02e-01

Pos logit contributions
arre+0.037
ravings+0.034
seek+0.033
 By+0.033
igsaw+0.032
Neg logit contributions
 comet-0.019
 lantern-0.018
Mom-0.017
adies-0.017
 ho-0.017
Token?"
Token ID1701
Feature activation-1.90e+00

Pos logit contributions
arre+0.022
igsaw+0.020
ravings+0.019
 courtroom+0.019
 fault+0.019
Neg logit contributions
adies-0.014
 carriers-0.013
Mom-0.012
 lantern-0.012
 ho-0.012
Token
Token ID220
Feature activation+9.51e-01

Pos logit contributions
 supposed+0.071
arre+0.069
igsaw+0.069
 fault+0.065
wheel+0.064
Neg logit contributions
Mom-0.045
Dad-0.036
el-0.033
 God-0.032
Parent-0.031
Token ground
Token ID2323
Feature activation-4.96e-01

Pos logit contributions
adies+0.008
Mom+0.008
 lantern+0.008
 comet+0.007
 carriers+0.007
Neg logit contributions
arre-0.010
igsaw-0.009
 fault-0.008
ravings-0.008
 By-0.008
Token!
Token ID0
Feature activation+6.83e-02

Pos logit contributions
arre+0.020
igsaw+0.018
 courtroom+0.018
ravings+0.018
Em+0.017
Neg logit contributions
adies-0.009
le-0.008
 lantern-0.008
works-0.008
 carriers-0.007
Tokenâ
Token ID22940
Feature activation+1.85e+00

Pos logit contributions
adies+0.002
Mom+0.002
 carriers+0.002
Dad+0.002
Ho+0.002
Neg logit contributions
arre-0.002
ravings-0.002
 fault-0.002
igsaw-0.002
 courtroom-0.002
Token
Token ID26391
Feature activation-1.64e-01

Pos logit contributions
adies+0.058
le+0.050
works+0.049
Mom+0.048
 carriers+0.047
Neg logit contributions
arre-0.054
 fault-0.049
ravings-0.048
 accidentally-0.047
igsaw-0.045
Token The
Token ID383
Feature activation-5.51e-02

Pos logit contributions
arre+0.005
igsaw+0.004
ravings+0.004
 fault+0.004
wheel+0.004
Neg logit contributions
adies-0.005
 carriers-0.004
Mom-0.004
 lantern-0.004
 ho-0.004
Token bird
Token ID6512
Feature activation-5.87e-01

Pos logit contributions
arre+0.002
ravings+0.002
igsaw+0.002
 fault+0.002
 accidentally+0.002
Neg logit contributions
adies-0.001
 lantern-0.001
Mom-0.001
 comet-0.001
 ho-0.001
Token was
Token ID373
Feature activation-1.29e-01

Pos logit contributions
arre+0.022
igsaw+0.022
ravings+0.021
 courtroom+0.020
 court+0.020
Neg logit contributions
adies-0.012
le-0.011
 ho-0.011
 lantern-0.011
el-0.010
Token scared
Token ID12008
Feature activation-1.29e+00

Pos logit contributions
igsaw+0.005
arre+0.005
ravings+0.004
enny+0.004
 court+0.004
Neg logit contributions
 ho-0.003
adies-0.003
 close-0.003
 wave-0.003
 closer-0.002
Token and
Token ID290
Feature activation-1.11e+00

Pos logit contributions
arre+0.048
igsaw+0.045
 courtroom+0.044
ravings+0.044
 court+0.042
Neg logit contributions
adies-0.029
le-0.025
 ho-0.025
Mom-0.024
 carriers-0.024
Token didn
Token ID1422
Feature activation-4.95e-01

Pos logit contributions
arre+0.043
igsaw+0.041
 fault+0.038
ravings+0.037
 courtroom+0.035
Neg logit contributions
 ho-0.024
adies-0.023
Run-0.022
 wave-0.021
 hello-0.020
Tokenâ
Token ID22940
Feature activation+5.52e-01

Pos logit contributions
arre+0.016
ravings+0.014
 courtroom+0.014
igsaw+0.014
 court+0.013
Neg logit contributions
adies-0.014
Mom-0.012
 carriers-0.012
 lantern-0.012
 ho-0.011
Token and
Token ID290
Feature activation-1.15e+00

Pos logit contributions
arre+0.093
igsaw+0.088
 courtroom+0.082
wheel+0.082
 Working+0.081
Neg logit contributions
le-0.050
 ho-0.046
adies-0.045
el-0.042
 lantern-0.041
Token sad
Token ID6507
Feature activation-2.00e+00

Pos logit contributions
arre+0.044
igsaw+0.039
 fault+0.037
ravings+0.037
 court+0.036
Neg logit contributions
 God-0.024
Mom-0.023
adies-0.023
 lantern-0.023
 carriers-0.022
Token.
Token ID13
Feature activation-6.85e-01

Pos logit contributions
arre+0.083
 courtroom+0.080
igsaw+0.079
wheel+0.076
ravings+0.073
Neg logit contributions
le-0.029
ies-0.026
 aliens-0.026
 ho-0.026
 lights-0.025
Token Lily
Token ID20037
Feature activation-1.41e+00

Pos logit contributions
arre+0.025
ravings+0.023
wheel+0.022
 fault+0.022
 salon+0.021
Neg logit contributions
Mom-0.017
Dad-0.013
 God-0.013
 Amen-0.012
adies-0.012
Token and
Token ID290
Feature activation-9.83e-01

Pos logit contributions
igsaw+0.057
 fault+0.051
arre+0.050
 winning+0.048
ravings+0.046
Neg logit contributions
Mom-0.029
le-0.026
 ho-0.026
 lantern-0.025
adies-0.024
Token her
Token ID607
Feature activation-1.17e+00

Pos logit contributions
arre+0.031
ravings+0.027
wheel+0.026
igsaw+0.026
 fault+0.026
Neg logit contributions
Mom-0.025
adies-0.025
 God-0.025
el-0.024
 carriers-0.024
Token mom
Token ID1995
Feature activation-2.72e-01

Pos logit contributions
arre+0.056
igsaw+0.051
wheel+0.051
 courtroom+0.049
 court+0.048
Neg logit contributions
 lantern-0.013
adies-0.013
Mom-0.012
 carriers-0.012
 ho-0.011
Token were
Token ID547
Feature activation-1.08e+00

Pos logit contributions
arre+0.009
igsaw+0.009
 fault+0.008
ravings+0.008
 courtroom+0.008
Neg logit contributions
adies-0.007
 lantern-0.006
Mom-0.006
 carriers-0.006
 ho-0.006
Token happy
Token ID3772
Feature activation-3.62e-01

Pos logit contributions
igsaw+0.040
con+0.037
arre+0.036
ravings+0.036
pill+0.034
Neg logit contributions
Mom-0.024
 lantern-0.024
 Amen-0.023
 close-0.022
 ho-0.022
Token to
Token ID284
Feature activation-7.24e-01

Pos logit contributions
arre+0.013
igsaw+0.012
 courtroom+0.012
ravings+0.012
wheel+0.011
Neg logit contributions
 lantern-0.008
 God-0.008
Mom-0.008
 carriers-0.007
 ho-0.007
Token have
Token ID423
Feature activation-3.87e-01

Pos logit contributions
arre+0.025
igsaw+0.022
ravings+0.021
 courtroom+0.021
wheel+0.021
Neg logit contributions
adies-0.019
 carriers-0.017
 lantern-0.017
Mom-0.016
 ho-0.015
Token hours
Token ID2250
Feature activation-2.31e+00

Pos logit contributions
arre+0.001
ravings+0.001
igsaw+0.000
 fault+0.000
wheel+0.000
Neg logit contributions
adies-0.000
 carriers-0.000
 lantern-0.000
Mom-0.000
 ho-0.000
Token,
Token ID11
Feature activation-8.23e-01

Pos logit contributions
arre+0.098
igsaw+0.094
ravings+0.090
 courtroom+0.090
 disl+0.086
Neg logit contributions
adies-0.039
 ho-0.034
 carriers-0.032
le-0.030
Mom-0.029
Token until
Token ID1566
Feature activation-2.66e-01

Pos logit contributions
arre+0.025
igsaw+0.023
ravings+0.022
 courtroom+0.021
 fault+0.020
Neg logit contributions
adies-0.024
 ho-0.022
 lantern-0.021
 carriers-0.021
Mom-0.020
Token Tim
Token ID5045
Feature activation-7.64e-01

Pos logit contributions
arre+0.008
igsaw+0.008
 accidentally+0.007
 court+0.007
 courtroom+0.007
Neg logit contributions
adies-0.007
 carriers-0.007
 lantern-0.006
Mom-0.006
le-0.006
Tokenmy
Token ID1820
Feature activation-1.02e+00

Pos logit contributions
arre+0.025
igsaw+0.025
 courtroom+0.023
ravings+0.023
 court+0.022
Neg logit contributions
adies-0.019
 ho-0.017
Mom-0.017
 carriers-0.016
 lantern-0.016
Token accidentally
Token ID14716
Feature activation-4.76e+00

Pos logit contributions
igsaw+0.037
arre+0.036
ravings+0.033
 fault+0.033
 winning+0.032
Neg logit contributions
adies-0.023
Mom-0.021
 ho-0.020
le-0.020
 carriers-0.020
Token cut
Token ID2005
Feature activation-1.24e+00

Pos logit contributions
arre+0.142
igsaw+0.127
ravings+0.119
 courtroom+0.116
 winning+0.113
Neg logit contributions
adies-0.146
 lantern-0.134
 carriers-0.132
 ho-0.129
Mom-0.125
Token his
Token ID465
Feature activation+5.75e-01

Pos logit contributions
arre+0.042
igsaw+0.038
ravings+0.037
 courtroom+0.037
 court+0.036
Neg logit contributions
adies-0.032
 carriers-0.029
 lantern-0.029
Mom-0.028
el-0.027
Token finger
Token ID7660
Feature activation-1.07e+00

Pos logit contributions
adies+0.014
 carriers+0.013
 lantern+0.013
 ho+0.012
Mom+0.012
Neg logit contributions
arre-0.020
igsaw-0.018
ravings-0.018
wheel-0.017
 courtroom-0.017
Token on
Token ID319
Feature activation-1.28e+00

Pos logit contributions
arre+0.042
igsaw+0.039
 courtroom+0.039
ravings+0.038
 court+0.037
Neg logit contributions
 lantern-0.020
le-0.019
el-0.019
adies-0.019
Mom-0.019
Token a
Token ID257
Feature activation-1.15e-01

Pos logit contributions
arre+0.052
ravings+0.045
igsaw+0.044
wheel+0.044
 courtroom+0.043
Neg logit contributions
 lantern-0.026
Mom-0.026
el-0.024
adies-0.024
 carriers-0.024
Token when
Token ID618
Feature activation+2.31e+00

Pos logit contributions
arre+0.021
ravings+0.019
igsaw+0.019
wheel+0.019
 courtroom+0.018
Neg logit contributions
adies-0.011
Mom-0.011
 carriers-0.010
 lantern-0.010
le-0.010
Token we
Token ID356
Feature activation-1.04e+00

Pos logit contributions
adies+0.039
 lantern+0.033
 carriers+0.033
Mom+0.032
 ho+0.032
Neg logit contributions
arre-0.101
igsaw-0.092
wheel-0.089
 courtroom-0.089
ravings-0.089
Token don
Token ID836
Feature activation+1.06e+00

Pos logit contributions
igsaw+0.041
arre+0.040
 courtroom+0.038
ravings+0.038
 fault+0.037
Neg logit contributions
adies-0.019
 ho-0.018
le-0.017
 lantern-0.017
Mom-0.015
Tokenâ
Token ID22940
Feature activation+2.04e-01

Pos logit contributions
adies+0.031
Mom+0.028
 carriers+0.027
 lantern+0.027
el+0.026
Neg logit contributions
arre-0.031
ravings-0.029
igsaw-0.029
 courtroom-0.028
 fault-0.028
Token
Token ID26391
Feature activation-1.06e+00

Pos logit contributions
adies+0.006
le+0.005
 Amen+0.005
Mom+0.005
 carriers+0.005
Neg logit contributions
arre-0.006
igsaw-0.005
 accidentally-0.005
ravings-0.005
 fault-0.005
Token
Token ID8151
Feature activation-4.20e+00

Pos logit contributions
arre+0.034
igsaw+0.034
ravings+0.031
wheel+0.031
 losing+0.030
Neg logit contributions
adies-0.026
ilot-0.023
 carriers-0.023
 ho-0.023
Mom-0.022
Tokent
Token ID83
Feature activation-1.06e+00

Pos logit contributions
igsaw+0.158
 courtroom+0.155
 accompl+0.153
ravings+0.152
arre+0.152
Neg logit contributions
adies-0.073
Mom-0.066
ilot-0.065
el-0.057
works-0.057
Token know
Token ID760
Feature activation-1.49e+00

Pos logit contributions
igsaw+0.038
 courtroom+0.037
ravings+0.036
arre+0.035
 fault+0.035
Neg logit contributions
le-0.023
adies-0.020
 wave-0.019
 go-0.018
ies-0.018
Token what
Token ID644
Feature activation+2.34e-01

Pos logit contributions
arre+0.054
igsaw+0.050
 courtroom+0.048
ravings+0.047
 court+0.046
Neg logit contributions
adies-0.036
 lantern-0.032
 carriers-0.031
works-0.030
le-0.030
Token to
Token ID284
Feature activation-1.37e+00

Pos logit contributions
adies+0.005
 lantern+0.004
 carriers+0.004
 comet+0.004
works+0.004
Neg logit contributions
arre-0.009
igsaw-0.009
ravings-0.008
 courtroom-0.008
wheel-0.008
Token do
Token ID466
Feature activation+6.41e-01

Pos logit contributions
arre+0.052
igsaw+0.047
ravings+0.046
 courtroom+0.046
 fault+0.044
Neg logit contributions
adies-0.029
 lantern-0.027
le-0.026
 carriers-0.026
Mom-0.025
Token if
Token ID611
Feature activation+3.02e+00

Pos logit contributions
 ho+0.024
adies+0.024
le+0.023
 God+0.023
 lantern+0.023
Neg logit contributions
arre-0.041
igsaw-0.038
wheel-0.037
ravings-0.037
 courtroom-0.036
Token you
Token ID345
Feature activation-2.31e+00

Pos logit contributions
adies+0.065
 carriers+0.056
 lantern+0.055
Mom+0.052
 ho+0.051
Neg logit contributions
arre-0.118
igsaw-0.106
ravings-0.104
wheel-0.103
 courtroom-0.102
Token don
Token ID836
Feature activation+6.88e-01

Pos logit contributions
arre+0.087
igsaw+0.082
 courtroom+0.079
ravings+0.079
 winning+0.077
Neg logit contributions
adies-0.049
le-0.042
 lantern-0.042
 ho-0.041
 carriers-0.040
Tokenâ
Token ID22940
Feature activation-1.76e-01

Pos logit contributions
adies+0.021
 carriers+0.019
Mom+0.019
 lantern+0.018
le+0.018
Neg logit contributions
arre-0.019
igsaw-0.018
ravings-0.017
 courtroom-0.017
 fault-0.017
Token
Token ID26391
Feature activation-2.09e+00

Pos logit contributions
arre+0.005
igsaw+0.004
 accidentally+0.004
ravings+0.004
 courtroom+0.004
Neg logit contributions
adies-0.005
 carriers-0.005
le-0.005
Mom-0.005
works-0.005
Token
Token ID8151
Feature activation-4.47e+00

Pos logit contributions
arre+0.056
igsaw+0.053
wheel+0.052
ravings+0.050
 courtroom+0.048
Neg logit contributions
adies-0.062
 carriers-0.059
ilot-0.055
Mom-0.055
 ho-0.054
Tokent
Token ID83
Feature activation-1.62e+00

Pos logit contributions
arre+0.156
ravings+0.150
igsaw+0.149
 courtroom+0.149
 disl+0.146
Neg logit contributions
adies-0.095
Mom-0.082
ilot-0.078
el-0.076
 ho-0.076
Token understand
Token ID1833
Feature activation-3.88e-01

Pos logit contributions
igsaw+0.061
arre+0.058
ravings+0.058
 courtroom+0.056
 batch+0.054
Neg logit contributions
le-0.033
 ho-0.029
 wave-0.029
adies-0.028
 go-0.027
Token why
Token ID1521
Feature activation-2.50e-01

Pos logit contributions
arre+0.014
igsaw+0.013
wheel+0.013
 courtroom+0.012
ravings+0.012
Neg logit contributions
adies-0.009
 lantern-0.008
 ho-0.008
le-0.008
 carriers-0.008
Token,
Token ID11
Feature activation-1.22e+00

Pos logit contributions
arre+0.010
igsaw+0.009
wheel+0.009
ravings+0.009
 courtroom+0.008
Neg logit contributions
adies-0.005
 ho-0.005
 lantern-0.004
 comet-0.004
 God-0.004
Token sometimes
Token ID3360
Feature activation+1.71e+00

Pos logit contributions
arre+0.049
igsaw+0.046
ravings+0.044
 courtroom+0.044
wheel+0.044
Neg logit contributions
adies-0.022
 ho-0.020
 lantern-0.019
 carriers-0.019
 God-0.018
Token.
Token ID13
Feature activation-2.29e-01

Pos logit contributions
arre+0.012
igsaw+0.011
ravings+0.010
 fault+0.010
 courtroom+0.010
Neg logit contributions
adies-0.007
Mom-0.006
 carriers-0.006
 lantern-0.006
 ho-0.006
Token One
Token ID1881
Feature activation+7.06e-01

Pos logit contributions
arre+0.007
igsaw+0.006
wheel+0.006
 courtroom+0.006
ravings+0.006
Neg logit contributions
adies-0.007
Mom-0.007
 lantern-0.006
 carriers-0.006
el-0.006
Token day
Token ID1110
Feature activation-3.23e-01

Pos logit contributions
adies+0.029
 carriers+0.027
 lantern+0.026
works+0.026
agles+0.025
Neg logit contributions
arre-0.013
igsaw-0.011
 about-0.010
 accidentally-0.010
ravings-0.010
Token,
Token ID11
Feature activation-9.00e-01

Pos logit contributions
arre+0.008
 accidentally+0.008
igsaw+0.007
 court+0.007
 winning+0.007
Neg logit contributions
adies-0.011
 carriers-0.010
 lantern-0.010
works-0.010
 Amen-0.009
Token she
Token ID673
Feature activation-5.97e-01

Pos logit contributions
arre+0.021
igsaw+0.018
ravings+0.017
 courtroom+0.017
 accidentally+0.017
Neg logit contributions
adies-0.033
 carriers-0.030
 lantern-0.030
Mom-0.029
 ho-0.028
Token accidentally
Token ID14716
Feature activation-4.22e+00

Pos logit contributions
arre+0.019
igsaw+0.017
ravings+0.016
 courtroom+0.016
wheel+0.016
Neg logit contributions
adies-0.017
 carriers-0.015
 lantern-0.015
Mom-0.014
 ho-0.014
Token dropped
Token ID5710
Feature activation-1.32e+00

Pos logit contributions
arre+0.115
igsaw+0.105
ravings+0.094
 courtroom+0.094
 winning+0.094
Neg logit contributions
adies-0.141
 lantern-0.126
Mom-0.124
 ho-0.123
 carriers-0.123
Token her
Token ID607
Feature activation-1.74e-01

Pos logit contributions
arre+0.050
igsaw+0.044
ravings+0.043
 courtroom+0.043
wheel+0.042
Neg logit contributions
adies-0.029
 lantern-0.029
Mom-0.028
el-0.028
 ho-0.027
Token cup
Token ID6508
Feature activation-7.42e-01

Pos logit contributions
arre+0.005
igsaw+0.004
 courtroom+0.004
ravings+0.004
wheel+0.004
Neg logit contributions
adies-0.006
 lantern-0.005
 carriers-0.005
Mom-0.005
 ho-0.005
Token and
Token ID290
Feature activation-4.89e-01

Pos logit contributions
arre+0.026
 winning+0.024
 disl+0.023
 courtroom+0.023
igsaw+0.023
Neg logit contributions
adies-0.017
Mom-0.017
 carriers-0.015
el-0.015
 lantern-0.015
Token it
Token ID340
Feature activation+3.60e-01

Pos logit contributions
arre+0.019
igsaw+0.017
 fault+0.016
ravings+0.016
 courtroom+0.016
Neg logit contributions
adies-0.011
 lantern-0.010
Mom-0.009
 carriers-0.009
 ho-0.009
Token he
Token ID339
Feature activation-1.52e+00

Pos logit contributions
 ho+0.007
 lantern+0.007
adies+0.007
Mom+0.007
 God+0.006
Neg logit contributions
arre-0.018
igsaw-0.016
ravings-0.016
 fault-0.015
wheel-0.014
Token just
Token ID655
Feature activation-1.02e+00

Pos logit contributions
igsaw+0.060
arre+0.058
 courtroom+0.051
ravings+0.051
 disl+0.050
Neg logit contributions
 ho-0.030
adies-0.030
le-0.029
Mom-0.028
 lantern-0.027
Token couldn
Token ID3521
Feature activation-9.24e-01

Pos logit contributions
arre+0.035
igsaw+0.032
ravings+0.030
 courtroom+0.029
 fault+0.029
Neg logit contributions
adies-0.028
 ho-0.025
 lantern-0.025
le-0.024
el-0.024
Tokenâ
Token ID22940
Feature activation+6.11e-01

Pos logit contributions
arre+0.029
ravings+0.026
 courtroom+0.025
igsaw+0.025
 fault+0.025
Neg logit contributions
adies-0.026
Mom-0.023
 carriers-0.022
el-0.022
 lantern-0.022
Token
Token ID26391
Feature activation-1.33e+00

Pos logit contributions
adies+0.020
le+0.017
Mom+0.017
 carriers+0.017
 lantern+0.016
Neg logit contributions
arre-0.017
 fault-0.015
ravings-0.015
igsaw-0.015
 accidentally-0.014
Token
Token ID8151
Feature activation-3.71e+00

Pos logit contributions
arre+0.034
igsaw+0.033
ravings+0.030
wheel+0.029
 courtroom+0.029
Neg logit contributions
adies-0.042
 carriers-0.039
Mom-0.038
ilot-0.038
 ho-0.038
Tokent
Token ID83
Feature activation-1.22e+00

Pos logit contributions
arre+0.124
ravings+0.117
igsaw+0.117
 courtroom+0.116
 disl+0.114
Neg logit contributions
adies-0.091
Mom-0.081
ilot-0.075
 ho-0.072
 carriers-0.071
Token work
Token ID670
Feature activation-2.58e+00

Pos logit contributions
arre+0.050
ravings+0.045
igsaw+0.044
 fault+0.044
 courtroom+0.043
Neg logit contributions
adies-0.025
le-0.024
 lantern-0.022
 ho-0.020
Mom-0.019
Token it
Token ID340
Feature activation-1.24e+00

Pos logit contributions
arre+0.116
ravings+0.105
 courtroom+0.103
 colony+0.103
igsaw+0.102
Neg logit contributions
adies-0.034
le-0.031
 carriers-0.031
 ho-0.031
 lantern-0.028
Token out
Token ID503
Feature activation-4.20e-01

Pos logit contributions
arre+0.057
igsaw+0.053
 courtroom+0.053
ravings+0.052
 disl+0.049
Neg logit contributions
adies-0.016
le-0.015
 ho-0.014
 carriers-0.013
el-0.013
Token.
Token ID13
Feature activation-3.64e-01

Pos logit contributions
arre+0.017
igsaw+0.015
ravings+0.015
 courtroom+0.015
wheel+0.014
Neg logit contributions
adies-0.009
 lantern-0.007
 carriers-0.007
le-0.007
 ho-0.007