TOP ACTIVATIONS
MAX = 10.336

tree with leaves and a coconut ," Tom said . He
how can we get the coconut ?" Tom had
reached out and grabbed a coconut . He was so happy
, it 's a real coconut !" Tom said . He
and showed his mom the coconut . She was proud of
picked a coconut . The coconut was heavy , but she
the tree and picked a coconut . The coconut was heavy
the tree to get a coconut , but she was scared
, he saw a big coconut tree with lots of coc
food . He found some coconut on the beach which he
You did not share the peach . I hate you ."
I should have respected the temple . It is not a
if she could talk about microsc opes . Her
did you not share the peach ?" She was
They run away from the temple . They promise to never
that you will eat your peach , and rest your knee
stick . She hits a peach . The peach falls down
She pointed to the peach . It was brown and
hits a peach . The peach falls down . Ben catches
had to know all about microsc opes .

MIN ACTIVATIONS
MIN = -11.543

- a brand new bath robe . It was soft ,
Tim saw that his bath robe was too tight . He
to wear her cozy bath robe after taking a bath .
had a lovely blue bath robe . Lucy loved to play
to wear her cozy bath robe after taking a bath .
to wear his soft bath robe after bath time . He
; he has a bath robe too ." Her
loved to play make - bel ieve . Today she was
they decided to be mail carriers and deliver letters to their
. We love being mail carriers . Can we do it
they always threw the bath robe in the air when they
put on her lovely bath robe and feel so fit and
gave him an ancient bath robe that belonged to his grand
she put her new bath robe on . <|endoftext|> Once upon
You are very good mail carriers . You made your friends
! I found a bath robe !" Lily looked at the
put on her cozy bath robe and went to bed with
she had a charming bath robe to wear , and she
He wore his big bath robe because it was cold .
put on the new bath robe and it was a perfect

INTERVAL 4.866 - 10.336
CONTAINS 0.482%

still did not want to touch the snakes , but he
of the dark forest . Jane was never able to go
ly at Bob and said : " Don 't
who was asking him the question . The le
€ ™ re the best vendor â € , said Midnight

INTERVAL -0.604 - 4.866
CONTAINS 72.323%

and said , " That looks very nice Jack . You
my heard the noise . She came to the kitchen and
, red apple . She was very happy .
very sad . But the princess didn 't
the fairy find her way home . The fairy was grateful

INTERVAL -6.074 - -0.604
CONTAINS 27.087%

big and fluffy that looked like a bear .
. It was scared and so it started to hope that
I know that you can make something beautiful out of something
� � € � � The mountains look very clear today
used the flashlight to show him the way home .

INTERVAL -11.543 - -6.074
CONTAINS 0.107%

. I have some band ages and some medicine . I
made a loud wh ir ring sound . He wanted to
. He snaps the snow man 's arm . He breaks
. She saw the fire man working hard to put out
water and make more rain bows . They laugh and play
Token tree
Token ID5509
Feature activation+2.26e+00

Pos logit contributions
ours+0.434
 coconut+0.422
 Enjoy+0.412
oral+0.382
rust+0.370
Neg logit contributions
aces-0.564
amas-0.551
coat-0.550
robe-0.529
 forehead-0.528
Token with
Token ID351
Feature activation-8.89e-01

Pos logit contributions
ours+0.142
 coconut+0.139
 Enjoy+0.132
oral+0.123
rust+0.121
Neg logit contributions
aces-0.195
amas-0.192
coat-0.191
robe-0.185
 forehead-0.185
Token leaves
Token ID5667
Feature activation+1.33e-01

Pos logit contributions
aces+0.071
amas+0.066
 forehead+0.065
coat+0.065
drops+0.065
Neg logit contributions
ours-0.059
oral-0.057
 Enjoy-0.056
 coconut-0.054
rust-0.052
Token and
Token ID290
Feature activation+3.08e-01

Pos logit contributions
ours+0.008
 coconut+0.008
 Enjoy+0.008
oral+0.007
rust+0.007
Neg logit contributions
aces-0.012
amas-0.012
coat-0.011
robe-0.011
 forehead-0.011
Token a
Token ID257
Feature activation-3.29e-01

Pos logit contributions
ours+0.023
 Enjoy+0.022
 coconut+0.022
oral+0.021
rust+0.021
Neg logit contributions
aces-0.023
coat-0.021
amas-0.021
drops-0.021
 forehead-0.021
Token coconut
Token ID20132
Feature activation+1.03e+01

Pos logit contributions
coat+0.024
aces+0.024
 forehead+0.023
amas+0.023
drops+0.023
Neg logit contributions
ours-0.025
 Enjoy-0.023
 coconut-0.022
oral-0.022
rust-0.022
Token,"
Token ID553
Feature activation-5.28e-01

Pos logit contributions
ours+0.681
 coconut+0.663
 Enjoy+0.644
oral+0.598
rust+0.580
Neg logit contributions
aces-0.853
amas-0.837
coat-0.835
robe-0.802
 forehead-0.801
Token Tom
Token ID4186
Feature activation-1.48e-01

Pos logit contributions
aces+0.038
 forehead+0.037
stand+0.036
coat+0.036
amas+0.034
Neg logit contributions
ours-0.039
 coconut-0.038
 Enjoy-0.038
oral-0.037
grown-0.033
Token said
Token ID531
Feature activation-5.18e-01

Pos logit contributions
aces+0.013
coat+0.013
amas+0.013
 forehead+0.013
robe+0.013
Neg logit contributions
ours-0.009
 coconut-0.009
 Enjoy-0.008
oral-0.008
rust-0.007
Token.
Token ID13
Feature activation+2.75e-01

Pos logit contributions
amas+0.050
coat+0.050
aces+0.049
drops+0.048
bridge+0.047
Neg logit contributions
ours-0.027
 Enjoy-0.024
 coconut-0.023
oral-0.021
 asks-0.021
Token He
Token ID679
Feature activation+3.83e-03

Pos logit contributions
ours+0.018
 coconut+0.018
 Enjoy+0.017
rous+0.016
rust+0.015
Neg logit contributions
aces-0.023
amas-0.023
coat-0.022
robe-0.022
stand-0.022
Token how
Token ID703
Feature activation+8.03e-01

Pos logit contributions
ours+0.074
oral+0.074
 Enjoy+0.073
 coconut+0.072
rust+0.068
Neg logit contributions
aces-0.080
 forehead-0.075
amas-0.075
drops-0.074
robe-0.073
Token can
Token ID460
Feature activation+6.71e-01

Pos logit contributions
 coconut+0.050
ours+0.050
 Enjoy+0.049
rust+0.046
iw+0.045
Neg logit contributions
aces-0.068
amas-0.066
coat-0.065
 forehead-0.063
robe-0.063
Token we
Token ID356
Feature activation+2.76e-01

Pos logit contributions
ours+0.042
 coconut+0.041
 Enjoy+0.040
oral+0.037
rust+0.036
Neg logit contributions
aces-0.058
amas-0.057
coat-0.057
robe-0.055
 forehead-0.055
Token get
Token ID651
Feature activation+1.11e+00

Pos logit contributions
ours+0.020
 coconut+0.019
 Enjoy+0.019
oral+0.018
rust+0.017
Neg logit contributions
aces-0.021
coat-0.021
amas-0.020
drops-0.020
 forehead-0.019
Token the
Token ID262
Feature activation-2.25e-02

Pos logit contributions
oral+0.081
ours+0.080
 Enjoy+0.078
 coconut+0.076
arium+0.073
Neg logit contributions
aces-0.075
 forehead-0.072
amas-0.072
drops-0.072
coat-0.071
Token coconut
Token ID20132
Feature activation+1.03e+01

Pos logit contributions
aces+0.002
coat+0.002
 forehead+0.002
drops+0.002
amas+0.002
Neg logit contributions
ours-0.002
 Enjoy-0.001
oral-0.001
rust-0.001
 coconut-0.001
Token?"
Token ID1701
Feature activation+3.09e+00

Pos logit contributions
 coconut+0.640
ours+0.629
 Enjoy+0.557
oral+0.529
iw+0.524
Neg logit contributions
aces-0.893
coat-0.887
amas-0.886
 forehead-0.864
stand-0.845
Token
Token ID198
Feature activation+9.75e-02

Pos logit contributions
 coconut+0.225
ours+0.215
oral+0.199
rust+0.199
 Enjoy+0.198
Neg logit contributions
aces-0.238
amas-0.232
stand-0.232
coat-0.229
drops-0.215
Token
Token ID198
Feature activation+8.18e-02

Pos logit contributions
 coconut+0.006
ours+0.006
 Enjoy+0.005
rust+0.005
iw+0.005
Neg logit contributions
aces-0.009
amas-0.008
stand-0.008
coat-0.008
robe-0.008
TokenTom
Token ID13787
Feature activation-1.59e+00

Pos logit contributions
ours+0.005
 coconut+0.005
 Enjoy+0.005
oral+0.004
rust+0.004
Neg logit contributions
aces-0.008
amas-0.007
coat-0.007
 forehead-0.007
robe-0.007
Token had
Token ID550
Feature activation-3.54e-01

Pos logit contributions
aces+0.130
coat+0.127
amas+0.125
drops+0.119
 forehead+0.119
Neg logit contributions
ours-0.108
 Enjoy-0.107
 coconut-0.106
oral-0.098
iw-0.092
Token reached
Token ID4251
Feature activation-4.54e-01

Pos logit contributions
aces+0.016
coat+0.016
amas+0.015
 forehead+0.015
drops+0.015
Neg logit contributions
ours-0.013
 coconut-0.012
 Enjoy-0.012
oral-0.011
rust-0.011
Token out
Token ID503
Feature activation-1.37e+00

Pos logit contributions
aces+0.035
coat+0.033
amas+0.033
 forehead+0.032
robe+0.032
Neg logit contributions
ours-0.033
 coconut-0.032
 Enjoy-0.032
oral-0.030
rust-0.029
Token and
Token ID290
Feature activation+1.79e-01

Pos logit contributions
aces+0.114
coat+0.110
amas+0.108
 forehead+0.106
drops+0.105
Neg logit contributions
ours-0.091
 Enjoy-0.087
 coconut-0.086
oral-0.081
rust-0.079
Token grabbed
Token ID13176
Feature activation+1.21e-01

Pos logit contributions
ours+0.012
 coconut+0.011
 Enjoy+0.011
oral+0.010
rust+0.010
Neg logit contributions
aces-0.015
amas-0.015
coat-0.015
 forehead-0.014
robe-0.014
Token a
Token ID257
Feature activation-4.14e-01

Pos logit contributions
ours+0.008
oral+0.008
 Enjoy+0.008
 coconut+0.007
rust+0.007
Neg logit contributions
aces-0.010
coat-0.010
amas-0.009
drops-0.009
 forehead-0.009
Token coconut
Token ID20132
Feature activation+1.03e+01

Pos logit contributions
aces+0.031
coat+0.030
amas+0.030
 forehead+0.028
robe+0.028
Neg logit contributions
ours-0.031
 coconut-0.030
 Enjoy-0.030
oral-0.028
rust-0.027
Token.
Token ID13
Feature activation-9.22e-03

Pos logit contributions
 coconut+0.650
ours+0.639
 Enjoy+0.591
iw+0.539
oral+0.537
Neg logit contributions
amas-0.886
aces-0.876
coat-0.869
robe-0.845
 forehead-0.836
Token He
Token ID679
Feature activation-1.66e-01

Pos logit contributions
aces+0.001
amas+0.001
coat+0.001
robe+0.001
drops+0.001
Neg logit contributions
ours-0.001
 coconut-0.001
oral-0.001
 Enjoy-0.001
rust-0.001
Token was
Token ID373
Feature activation+6.82e-01

Pos logit contributions
aces+0.013
coat+0.012
amas+0.012
drops+0.012
 forehead+0.011
Neg logit contributions
ours-0.012
 Enjoy-0.012
oral-0.011
 coconut-0.011
 Sometimes-0.011
Token so
Token ID523
Feature activation-2.31e+00

Pos logit contributions
 coconut+0.049
ours+0.047
 Enjoy+0.045
 Joe+0.041
iw+0.040
Neg logit contributions
amas-0.056
robe-0.053
coat-0.052
aces-0.051
stand-0.050
Token happy
Token ID3772
Feature activation-1.12e+00

Pos logit contributions
aces+0.156
coat+0.149
amas+0.147
drops+0.143
 forehead+0.142
Neg logit contributions
ours-0.189
 coconut-0.180
 Enjoy-0.179
oral-0.175
rust-0.167
Token,
Token ID11
Feature activation+2.96e-02

Pos logit contributions
ours+0.157
 coconut+0.155
 Enjoy+0.154
oral+0.149
rust+0.141
Neg logit contributions
aces-0.200
amas-0.195
coat-0.193
 forehead-0.186
drops-0.185
Token it
Token ID340
Feature activation+1.28e+00

Pos logit contributions
ours+0.002
 coconut+0.002
 Enjoy+0.002
oral+0.002
rust+0.002
Neg logit contributions
aces-0.002
coat-0.002
amas-0.002
drops-0.002
robe-0.002
Token's
Token ID338
Feature activation-1.39e+00

Pos logit contributions
ours+0.079
 coconut+0.076
 Enjoy+0.072
rust+0.063
 volcano+0.063
Neg logit contributions
aces-0.113
coat-0.112
amas-0.112
 forehead-0.109
robe-0.107
Token a
Token ID257
Feature activation-4.89e-01

Pos logit contributions
aces+0.116
coat+0.114
amas+0.112
 forehead+0.110
robe+0.109
Neg logit contributions
ours-0.092
 coconut-0.092
 Enjoy-0.088
rust-0.076
 swamp-0.076
Token real
Token ID1103
Feature activation+7.08e-01

Pos logit contributions
coat+0.037
robe+0.035
amas+0.035
 forehead+0.035
drops+0.034
Neg logit contributions
ours-0.037
 Enjoy-0.033
oral-0.033
rust-0.032
 coconut-0.031
Token coconut
Token ID20132
Feature activation+1.03e+01

Pos logit contributions
ours+0.050
oral+0.046
 Enjoy+0.044
rust+0.044
 Sometimes+0.042
Neg logit contributions
coat-0.055
aces-0.054
drops-0.053
amas-0.053
robe-0.052
Token!"
Token ID2474
Feature activation-1.67e+00

Pos logit contributions
ours+0.693
 coconut+0.692
 Enjoy+0.642
oral+0.593
iw+0.588
Neg logit contributions
aces-0.823
amas-0.823
coat-0.819
robe-0.792
 forehead-0.787
Token Tom
Token ID4186
Feature activation-2.58e-01

Pos logit contributions
aces+0.140
amas+0.135
coat+0.134
stand+0.134
drops+0.130
Neg logit contributions
ours-0.107
 coconut-0.104
 Enjoy-0.101
oral-0.095
rust-0.092
Token said
Token ID531
Feature activation-5.76e-01

Pos logit contributions
amas+0.024
aces+0.024
coat+0.024
robe+0.024
 forehead+0.023
Neg logit contributions
ours-0.013
 coconut-0.013
 Enjoy-0.012
rust-0.011
 asks-0.011
Token.
Token ID13
Feature activation+1.87e-01

Pos logit contributions
amas+0.055
coat+0.055
aces+0.054
robe+0.052
drops+0.052
Neg logit contributions
ours-0.030
 coconut-0.028
 Enjoy-0.027
rust-0.023
 asks-0.023
Token He
Token ID679
Feature activation-1.03e-01

Pos logit contributions
ours+0.013
 coconut+0.013
 Enjoy+0.012
oral+0.011
rous+0.011
Neg logit contributions
aces-0.015
amas-0.015
coat-0.014
robe-0.014
stand-0.014
Token and
Token ID290
Feature activation+1.60e-01

Pos logit contributions
 coconut+0.123
ours+0.119
 Enjoy+0.114
iw+0.104
arium+0.102
Neg logit contributions
aces-0.184
amas-0.183
coat-0.176
robe-0.176
 forehead-0.173
Token showed
Token ID3751
Feature activation+5.47e-01

Pos logit contributions
ours+0.011
 coconut+0.011
 Enjoy+0.011
oral+0.011
rust+0.010
Neg logit contributions
aces-0.012
coat-0.012
amas-0.012
drops-0.012
 forehead-0.012
Token his
Token ID465
Feature activation+9.48e-01

Pos logit contributions
ours+0.034
oral+0.032
 coconut+0.032
 Enjoy+0.031
iw+0.028
Neg logit contributions
aces-0.048
amas-0.045
coat-0.044
robe-0.044
 forehead-0.044
Token mom
Token ID1995
Feature activation-3.82e-01

Pos logit contributions
ours+0.075
 Enjoy+0.074
oral+0.064
arium+0.064
 arose+0.062
Neg logit contributions
aces-0.067
 forehead-0.064
amas-0.059
coat-0.059
drops-0.058
Token the
Token ID262
Feature activation-1.64e-02

Pos logit contributions
aces+0.029
coat+0.029
stand+0.028
amas+0.027
robe+0.027
Neg logit contributions
ours-0.026
oral-0.025
 coconut-0.025
 Enjoy-0.025
iw-0.022
Token coconut
Token ID20132
Feature activation+1.03e+01

Pos logit contributions
aces+0.001
 forehead+0.001
amas+0.001
coat+0.001
drops+0.001
Neg logit contributions
ours-0.001
 Enjoy-0.001
arium-0.001
 arose-0.001
 Sometimes-0.001
Token.
Token ID13
Feature activation+7.45e-03

Pos logit contributions
ours+0.712
 Enjoy+0.697
 coconut+0.669
oral+0.651
rust+0.621
Neg logit contributions
aces-0.826
coat-0.779
amas-0.775
drops-0.761
 forehead-0.756
Token She
Token ID1375
Feature activation-3.02e-01

Pos logit contributions
ours+0.001
 coconut+0.000
 Enjoy+0.000
oral+0.000
rous+0.000
Neg logit contributions
amas-0.001
aces-0.001
robe-0.001
coat-0.001
stand-0.001
Token was
Token ID373
Feature activation+6.14e-01

Pos logit contributions
aces+0.023
coat+0.022
amas+0.022
 forehead+0.021
robe+0.021
Neg logit contributions
ours-0.022
 Enjoy-0.021
 coconut-0.020
oral-0.020
 Sometimes-0.019
Token proud
Token ID6613
Feature activation-1.19e-01

Pos logit contributions
 coconut+0.043
ours+0.043
 Enjoy+0.042
rust+0.037
iw+0.037
Neg logit contributions
amas-0.049
aces-0.047
coat-0.047
robe-0.046
 forehead-0.044
Token of
Token ID286
Feature activation+9.37e-02

Pos logit contributions
aces+0.010
coat+0.009
amas+0.009
stand+0.009
drops+0.009
Neg logit contributions
ours-0.008
 coconut-0.008
 Enjoy-0.008
oral-0.007
rust-0.007
Token picked
Token ID6497
Feature activation+2.19e+00

Pos logit contributions
ours+0.006
 Enjoy+0.005
oral+0.005
 coconut+0.005
rust+0.005
Neg logit contributions
aces-0.006
coat-0.006
drops-0.006
amas-0.006
 forehead-0.006
Token a
Token ID257
Feature activation-5.42e-01

Pos logit contributions
ours+0.175
oral+0.172
 coconut+0.168
 Enjoy+0.167
rust+0.160
Neg logit contributions
aces-0.149
amas-0.134
coat-0.133
drops-0.129
 forehead-0.129
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
coat+0.033
 forehead+0.032
amas+0.032
brush+0.031
stand+0.030
Neg logit contributions
ours-0.046
 Enjoy-0.042
 Sometimes-0.040
oral-0.040
rust-0.040
Token.
Token ID13
Feature activation+3.74e-02

Pos logit contributions
ours+0.685
 Enjoy+0.669
 coconut+0.637
oral+0.625
rust+0.598
Neg logit contributions
aces-0.843
coat-0.803
amas-0.797
drops-0.781
 forehead-0.774
Token The
Token ID383
Feature activation+3.39e-01

Pos logit contributions
 coconut+0.003
ours+0.003
oral+0.002
 Enjoy+0.002
rust+0.002
Neg logit contributions
aces-0.003
amas-0.003
coat-0.003
robe-0.003
drops-0.003
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
ours+0.029
 Enjoy+0.028
arium+0.025
 arose+0.025
rill+0.024
Neg logit contributions
 forehead-0.022
aces-0.021
drops-0.021
coat-0.020
 cheek-0.019
Token was
Token ID373
Feature activation+5.65e-01

Pos logit contributions
ours+0.523
 coconut+0.520
 Enjoy+0.474
oral+0.436
rust+0.424
Neg logit contributions
aces-0.991
amas-0.984
coat-0.980
 forehead-0.947
robe-0.946
Token heavy
Token ID4334
Feature activation-1.35e+00

Pos logit contributions
 coconut+0.040
ours+0.037
 Enjoy+0.035
rust+0.033
iw+0.033
Neg logit contributions
amas-0.047
aces-0.045
coat-0.045
robe-0.044
stand-0.044
Token,
Token ID11
Feature activation-9.89e-02

Pos logit contributions
amas+0.121
aces+0.121
coat+0.118
robe+0.117
 forehead+0.115
Neg logit contributions
ours-0.077
 coconut-0.077
 Enjoy-0.071
iw-0.064
grown-0.063
Token but
Token ID475
Feature activation-2.29e+00

Pos logit contributions
amas+0.009
aces+0.009
robe+0.009
coat+0.008
bridge+0.008
Neg logit contributions
 coconut-0.006
ours-0.005
 Sometimes-0.005
 Enjoy-0.005
iw-0.005
Token she
Token ID673
Feature activation-1.42e+00

Pos logit contributions
aces+0.180
coat+0.173
amas+0.171
drops+0.169
robe+0.168
Neg logit contributions
ours-0.162
 coconut-0.155
 Enjoy-0.154
oral-0.147
rust-0.140
Token the
Token ID262
Feature activation-9.82e-02

Pos logit contributions
ours+0.003
 Enjoy+0.003
oral+0.003
 coconut+0.003
rust+0.003
Neg logit contributions
aces-0.003
coat-0.003
amas-0.003
stand-0.003
 forehead-0.003
Token tree
Token ID5509
Feature activation+1.98e+00

Pos logit contributions
drops+0.006
 forehead+0.006
coat+0.006
aces+0.006
amas+0.006
Neg logit contributions
ours-0.008
 Enjoy-0.008
 arose-0.007
 whose-0.006
 asks-0.006
Token and
Token ID290
Feature activation+7.85e-02

Pos logit contributions
ours+0.116
 coconut+0.113
 Enjoy+0.107
iw+0.097
oral+0.097
Neg logit contributions
aces-0.174
amas-0.172
robe-0.168
coat-0.167
 forehead-0.163
Token picked
Token ID6497
Feature activation+2.19e+00

Pos logit contributions
ours+0.006
 Enjoy+0.005
oral+0.005
 coconut+0.005
rust+0.005
Neg logit contributions
aces-0.006
coat-0.006
drops-0.006
amas-0.006
 forehead-0.006
Token a
Token ID257
Feature activation-5.42e-01

Pos logit contributions
ours+0.175
oral+0.172
 coconut+0.168
 Enjoy+0.167
rust+0.160
Neg logit contributions
aces-0.149
amas-0.134
coat-0.133
drops-0.129
 forehead-0.129
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
coat+0.033
 forehead+0.032
amas+0.032
brush+0.031
stand+0.030
Neg logit contributions
ours-0.046
 Enjoy-0.042
 Sometimes-0.040
oral-0.040
rust-0.040
Token.
Token ID13
Feature activation+3.74e-02

Pos logit contributions
ours+0.685
 Enjoy+0.669
 coconut+0.637
oral+0.625
rust+0.598
Neg logit contributions
aces-0.843
coat-0.803
amas-0.797
drops-0.781
 forehead-0.774
Token The
Token ID383
Feature activation+3.39e-01

Pos logit contributions
 coconut+0.003
ours+0.003
oral+0.002
 Enjoy+0.002
rust+0.002
Neg logit contributions
aces-0.003
amas-0.003
coat-0.003
robe-0.003
drops-0.003
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
ours+0.029
 Enjoy+0.028
arium+0.025
 arose+0.025
rill+0.024
Neg logit contributions
 forehead-0.022
aces-0.021
drops-0.021
coat-0.020
 cheek-0.019
Token was
Token ID373
Feature activation+5.65e-01

Pos logit contributions
ours+0.523
 coconut+0.520
 Enjoy+0.474
oral+0.436
rust+0.424
Neg logit contributions
aces-0.991
amas-0.984
coat-0.980
 forehead-0.947
robe-0.946
Token heavy
Token ID4334
Feature activation-1.35e+00

Pos logit contributions
 coconut+0.040
ours+0.037
 Enjoy+0.035
rust+0.033
iw+0.033
Neg logit contributions
amas-0.047
aces-0.045
coat-0.045
robe-0.044
stand-0.044
Token the
Token ID262
Feature activation-9.88e-02

Pos logit contributions
ours+0.016
 Enjoy+0.015
 coconut+0.015
oral+0.015
rust+0.014
Neg logit contributions
aces-0.019
coat-0.019
drops-0.018
amas-0.018
stand-0.018
Token tree
Token ID5509
Feature activation+1.99e+00

Pos logit contributions
coat+0.007
aces+0.007
drops+0.007
amas+0.007
 forehead+0.007
Neg logit contributions
ours-0.007
 Enjoy-0.007
oral-0.007
rust-0.006
 coconut-0.006
Token to
Token ID284
Feature activation+4.67e-01

Pos logit contributions
ours+0.129
 coconut+0.127
 Enjoy+0.121
oral+0.111
rust+0.110
Neg logit contributions
aces-0.165
amas-0.163
coat-0.160
robe-0.156
 forehead-0.156
Token get
Token ID651
Feature activation+1.01e+00

Pos logit contributions
ours+0.038
oral+0.036
 coconut+0.034
rust+0.033
 Enjoy+0.032
Neg logit contributions
 forehead-0.030
aces-0.030
coat-0.029
stand-0.029
drops-0.028
Token a
Token ID257
Feature activation-5.39e-01

Pos logit contributions
ours+0.072
oral+0.068
 coconut+0.067
 Enjoy+0.066
arium+0.062
Neg logit contributions
aces-0.078
coat-0.075
 forehead-0.074
drops-0.073
amas-0.073
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
coat+0.041
 forehead+0.038
brush+0.036
stand+0.035
amas+0.034
Neg logit contributions
ours-0.042
 Enjoy-0.037
oral-0.037
rust-0.037
grown-0.035
Token,
Token ID11
Feature activation-1.12e-01

Pos logit contributions
 coconut+0.613
ours+0.568
 Enjoy+0.512
iw+0.491
 crab+0.483
Neg logit contributions
amas-0.940
coat-0.915
aces-0.911
robe-0.897
 forehead-0.881
Token but
Token ID475
Feature activation-2.32e+00

Pos logit contributions
aces+0.007
amas+0.007
coat+0.006
robe+0.006
 forehead+0.006
Neg logit contributions
ours-0.010
 coconut-0.010
 Enjoy-0.010
oral-0.009
rust-0.009
Token she
Token ID673
Feature activation-1.41e+00

Pos logit contributions
aces+0.189
coat+0.182
amas+0.182
 forehead+0.180
robe+0.177
Neg logit contributions
ours-0.156
 Enjoy-0.151
 coconut-0.151
oral-0.143
rust-0.137
Token was
Token ID373
Feature activation+5.46e-01

Pos logit contributions
aces+0.096
coat+0.090
amas+0.088
 forehead+0.084
robe+0.084
Neg logit contributions
 Enjoy-0.121
ours-0.113
oral-0.106
 coconut-0.104
rous-0.104
Token scared
Token ID12008
Feature activation+5.88e-01

Pos logit contributions
ours+0.043
 Enjoy+0.041
 coconut+0.041
oral+0.039
rust+0.037
Neg logit contributions
aces-0.040
coat-0.038
 forehead-0.037
amas-0.037
drops-0.036
Token,
Token ID11
Feature activation-2.11e-01

Pos logit contributions
ours+0.096
 coconut+0.096
 Enjoy+0.090
oral+0.083
rust+0.079
Neg logit contributions
aces-0.172
amas-0.170
coat-0.168
 forehead-0.163
robe-0.161
Token he
Token ID339
Feature activation-3.07e-01

Pos logit contributions
aces+0.018
amas+0.017
coat+0.017
robe+0.017
 forehead+0.017
Neg logit contributions
ours-0.014
 coconut-0.013
 Enjoy-0.013
oral-0.012
rust-0.012
Token saw
Token ID2497
Feature activation-4.00e-01

Pos logit contributions
aces+0.028
coat+0.027
amas+0.027
robe+0.026
 forehead+0.026
Neg logit contributions
ours-0.018
 Enjoy-0.017
 coconut-0.017
oral-0.016
rust-0.015
Token a
Token ID257
Feature activation-4.87e-01

Pos logit contributions
aces+0.029
robe+0.027
bridge+0.027
coat+0.027
 forehead+0.027
Neg logit contributions
ours-0.031
 Enjoy-0.029
oral-0.028
 coconut-0.028
iw-0.027
Token big
Token ID1263
Feature activation-9.90e-01

Pos logit contributions
coat+0.033
amas+0.033
 forehead+0.032
stand+0.031
robe+0.031
Neg logit contributions
ours-0.039
 Enjoy-0.037
 coconut-0.035
oral-0.034
rust-0.034
Token coconut
Token ID20132
Feature activation+1.02e+01

Pos logit contributions
coat+0.078
amas+0.077
aces+0.075
 forehead+0.075
drops+0.074
Neg logit contributions
ours-0.068
 Enjoy-0.064
oral-0.061
 coconut-0.061
rust-0.058
Token tree
Token ID5509
Feature activation+2.04e+00

Pos logit contributions
ours+0.764
 Enjoy+0.738
 coconut+0.729
oral+0.690
rust+0.674
Neg logit contributions
aces-0.753
coat-0.729
amas-0.728
drops-0.697
 forehead-0.695
Token with
Token ID351
Feature activation-1.17e+00

Pos logit contributions
ours+0.148
 Enjoy+0.143
 coconut+0.142
oral+0.134
rust+0.129
Neg logit contributions
aces-0.156
coat-0.153
amas-0.152
drops-0.145
robe-0.145
Token lots
Token ID6041
Feature activation+1.40e+00

Pos logit contributions
aces+0.094
amas+0.088
coat+0.088
bridge+0.086
 forehead+0.086
Neg logit contributions
ours-0.081
 Enjoy-0.075
oral-0.074
 coconut-0.072
rust-0.068
Token of
Token ID286
Feature activation+2.55e-02

Pos logit contributions
ours+0.102
 Enjoy+0.097
oral+0.092
rust+0.091
 coconut+0.089
Neg logit contributions
aces-0.111
coat-0.101
drops-0.099
bridge-0.098
amas-0.098
Token coc
Token ID8954
Feature activation+3.88e+00

Pos logit contributions
ours+0.002
arium+0.002
oral+0.002
 Enjoy+0.002
rust+0.002
Neg logit contributions
aces-0.002
bridge-0.002
 forehead-0.002
 strings-0.002
brush-0.002
Token food
Token ID2057
Feature activation-2.37e-01

Pos logit contributions
aces+0.082
 forehead+0.080
stand+0.077
coat+0.076
amas+0.076
Neg logit contributions
ours-0.094
oral-0.087
 Enjoy-0.082
 coconut-0.080
iw-0.079
Token.
Token ID13
Feature activation-2.38e-01

Pos logit contributions
amas+0.022
aces+0.022
coat+0.021
 forehead+0.021
drops+0.020
Neg logit contributions
ours-0.013
 coconut-0.013
 Enjoy-0.013
oral-0.011
rust-0.011
Token He
Token ID679
Feature activation-2.45e-01

Pos logit contributions
aces+0.018
amas+0.018
robe+0.017
coat+0.017
 forehead+0.016
Neg logit contributions
ours-0.018
 coconut-0.018
oral-0.017
 Enjoy-0.017
rust-0.016
Token found
Token ID1043
Feature activation-2.43e+00

Pos logit contributions
coat+0.019
aces+0.018
 forehead+0.018
amas+0.018
robe+0.018
Neg logit contributions
ours-0.018
 Enjoy-0.017
 coconut-0.017
oral-0.016
iw-0.015
Token some
Token ID617
Feature activation+4.37e-01

Pos logit contributions
aces+0.196
amas+0.188
coat+0.187
robe+0.183
 forehead+0.182
Neg logit contributions
ours-0.166
 coconut-0.156
 Enjoy-0.155
oral-0.150
rust-0.142
Token coconut
Token ID20132
Feature activation+1.01e+01

Pos logit contributions
ours+0.031
 Enjoy+0.029
oral+0.028
 coconut+0.028
rust+0.027
Neg logit contributions
aces-0.034
coat-0.032
amas-0.032
robe-0.031
 forehead-0.031
Token on
Token ID319
Feature activation+7.26e-01

Pos logit contributions
ours+0.710
 Enjoy+0.700
oral+0.658
 coconut+0.649
rust+0.615
Neg logit contributions
aces-0.807
coat-0.767
amas-0.756
drops-0.744
robe-0.740
Token the
Token ID262
Feature activation-1.15e-01

Pos logit contributions
ours+0.064
 Enjoy+0.058
oral+0.057
iw+0.051
rust+0.051
Neg logit contributions
stand-0.044
aces-0.043
amas-0.042
coat-0.040
bridge-0.040
Token beach
Token ID10481
Feature activation+2.00e+00

Pos logit contributions
aces+0.010
coat+0.010
amas+0.009
robe+0.009
drops+0.009
Neg logit contributions
 coconut-0.008
ours-0.007
 Enjoy-0.007
oral-0.007
rust-0.006
Token which
Token ID543
Feature activation-1.20e+00

Pos logit contributions
 coconut+0.122
ours+0.120
 Enjoy+0.113
oral+0.101
arium+0.101
Neg logit contributions
aces-0.173
amas-0.172
coat-0.168
 forehead-0.165
drops-0.161
Token he
Token ID339
Feature activation-3.39e-01

Pos logit contributions
aces+0.095
robe+0.094
coat+0.092
amas+0.090
drops+0.089
Neg logit contributions
ours-0.082
 Enjoy-0.077
oral-0.073
 coconut-0.070
rust-0.066
Token You
Token ID921
Feature activation-1.45e-01

Pos logit contributions
 coconut+0.027
ours+0.026
oral+0.025
 Enjoy+0.025
rust+0.024
Neg logit contributions
aces-0.022
amas-0.021
coat-0.021
robe-0.020
stand-0.020
Token did
Token ID750
Feature activation+7.83e-01

Pos logit contributions
aces+0.013
amas+0.012
coat+0.012
robe+0.012
 forehead+0.012
Neg logit contributions
ours-0.009
 coconut-0.009
 Enjoy-0.009
oral-0.008
rust-0.008
Token not
Token ID407
Feature activation-6.67e-01

Pos logit contributions
ours+0.071
 coconut+0.070
 Enjoy+0.067
oral+0.066
rust+0.063
Neg logit contributions
aces-0.045
stand-0.044
amas-0.043
coat-0.042
 forehead-0.041
Token share
Token ID2648
Feature activation+2.42e+00

Pos logit contributions
coat+0.049
aces+0.048
stand+0.046
drops+0.046
amas+0.045
Neg logit contributions
ours-0.051
 coconut-0.048
oral-0.046
arium-0.044
 Enjoy-0.044
Token the
Token ID262
Feature activation+3.54e-01

Pos logit contributions
ours+0.160
 coconut+0.155
 Enjoy+0.153
oral+0.143
rust+0.137
Neg logit contributions
aces-0.202
amas-0.196
coat-0.194
stand-0.187
robe-0.187
Token peach
Token ID47565
Feature activation+8.57e+00

Pos logit contributions
ours+0.026
 coconut+0.025
 Enjoy+0.025
oral+0.023
rust+0.023
Neg logit contributions
aces-0.027
amas-0.026
coat-0.026
robe-0.025
 forehead-0.025
Token.
Token ID13
Feature activation+3.20e-01

Pos logit contributions
ours+0.560
 coconut+0.544
 Enjoy+0.530
oral+0.493
rust+0.478
Neg logit contributions
aces-0.714
amas-0.699
coat-0.697
robe-0.670
 forehead-0.668
Token I
Token ID314
Feature activation-3.61e-01

Pos logit contributions
ours+0.025
 coconut+0.025
 Enjoy+0.024
oral+0.023
rust+0.022
Neg logit contributions
aces-0.023
amas-0.022
coat-0.021
robe-0.021
 forehead-0.021
Token hate
Token ID5465
Feature activation+2.69e+00

Pos logit contributions
coat+0.027
drops+0.026
aces+0.025
robe+0.024
ze+0.024
Neg logit contributions
 coconut-0.027
ours-0.027
arium-0.024
oral-0.024
 Enjoy-0.024
Token you
Token ID345
Feature activation+2.29e-01

Pos logit contributions
ours+0.203
 Enjoy+0.195
oral+0.192
 coconut+0.190
arium+0.182
Neg logit contributions
aces-0.197
amas-0.189
stand-0.185
drops-0.178
coat-0.178
Token."
Token ID526
Feature activation-1.43e+00

Pos logit contributions
ours+0.013
 coconut+0.013
 Enjoy+0.013
oral+0.011
rust+0.011
Neg logit contributions
aces-0.021
coat-0.020
amas-0.020
robe-0.020
bridge-0.020
Token I
Token ID314
Feature activation-3.46e-01

Pos logit contributions
 coconut+0.024
ours+0.024
oral+0.022
 Enjoy+0.022
rust+0.022
Neg logit contributions
aces-0.023
amas-0.022
coat-0.022
 forehead-0.022
robe-0.022
Token should
Token ID815
Feature activation+5.07e+00

Pos logit contributions
coat+0.026
drops+0.025
robe+0.023
aces+0.023
brush+0.022
Neg logit contributions
 coconut-0.027
ours-0.026
oral-0.024
 Enjoy-0.024
arium-0.023
Token have
Token ID423
Feature activation+2.54e+00

Pos logit contributions
ours+0.398
 coconut+0.391
oral+0.362
 Enjoy+0.358
iw+0.349
Neg logit contributions
coat-0.348
aces-0.346
amas-0.338
 forehead-0.336
stand-0.335
Token respected
Token ID14462
Feature activation+2.34e+00

Pos logit contributions
ours+0.192
 coconut+0.183
 Enjoy+0.177
oral+0.174
arium+0.164
Neg logit contributions
aces-0.186
coat-0.186
amas-0.180
 forehead-0.177
stand-0.173
Token the
Token ID262
Feature activation+1.89e-01

Pos logit contributions
ours+0.160
 coconut+0.154
 Enjoy+0.151
oral+0.142
rust+0.135
Neg logit contributions
aces-0.191
amas-0.183
coat-0.183
stand-0.177
 forehead-0.175
Token temple
Token ID12505
Feature activation+8.52e+00

Pos logit contributions
ours+0.014
 Enjoy+0.014
 coconut+0.013
rust+0.013
arium+0.012
Neg logit contributions
coat-0.014
 forehead-0.014
amas-0.013
drops-0.013
robe-0.013
Token.
Token ID13
Feature activation+3.38e-01

Pos logit contributions
ours+0.475
 coconut+0.457
 Enjoy+0.415
oral+0.406
iw+0.390
Neg logit contributions
amas-0.772
coat-0.757
aces-0.755
robe-0.738
 cartridge-0.721
Token It
Token ID632
Feature activation-5.30e-02

Pos logit contributions
 coconut+0.025
ours+0.025
 Enjoy+0.023
oral+0.023
rust+0.022
Neg logit contributions
aces-0.026
amas-0.025
coat-0.024
 forehead-0.024
robe-0.024
Token is
Token ID318
Feature activation+8.83e-01

Pos logit contributions
aces+0.005
amas+0.005
coat+0.005
robe+0.005
stand+0.005
Neg logit contributions
ours-0.003
 coconut-0.003
 Enjoy-0.003
oral-0.003
rust-0.002
Token not
Token ID407
Feature activation-7.48e-01

Pos logit contributions
ours+0.060
 coconut+0.059
 Enjoy+0.058
oral+0.053
rust+0.052
Neg logit contributions
aces-0.071
amas-0.070
coat-0.070
robe-0.067
drops-0.066
Token a
Token ID257
Feature activation-3.34e-01

Pos logit contributions
amas+0.066
coat+0.062
aces+0.062
robe+0.061
stand+0.061
Neg logit contributions
 coconut-0.045
ours-0.044
 Enjoy-0.044
rust-0.039
 whose-0.038
Token if
Token ID611
Feature activation+9.88e-01

Pos logit contributions
ours+0.110
 coconut+0.110
 Enjoy+0.102
rust+0.097
oral+0.096
Neg logit contributions
aces-0.140
amas-0.137
coat-0.134
 forehead-0.133
robe-0.132
Token she
Token ID673
Feature activation-1.41e+00

Pos logit contributions
ours+0.085
 coconut+0.082
 Enjoy+0.078
oral+0.075
rust+0.075
Neg logit contributions
aces-0.060
amas-0.059
 forehead-0.058
coat-0.058
robe-0.057
Token could
Token ID714
Feature activation-1.40e+00

Pos logit contributions
aces+0.090
coat+0.077
robe+0.076
brush+0.075
 forehead+0.075
Neg logit contributions
ours-0.127
 coconut-0.120
 Enjoy-0.118
oral-0.115
rous-0.114
Token talk
Token ID1561
Feature activation+5.11e-01

Pos logit contributions
aces+0.104
coat+0.101
amas+0.100
robe+0.097
 forehead+0.096
Neg logit contributions
ours-0.107
 coconut-0.103
 Enjoy-0.099
oral-0.094
rust-0.094
Token about
Token ID546
Feature activation+1.53e+00

Pos logit contributions
ours+0.029
 coconut+0.028
 Enjoy+0.028
oral+0.025
rust+0.024
Neg logit contributions
aces-0.047
amas-0.046
coat-0.046
robe-0.044
 forehead-0.044
Token microsc
Token ID21145
Feature activation+8.52e+00

Pos logit contributions
ours+0.109
 Enjoy+0.103
 coconut+0.102
oral+0.099
rust+0.094
Neg logit contributions
aces-0.121
coat-0.113
robe-0.112
stand-0.112
amas-0.111
Tokenopes
Token ID13920
Feature activation-2.73e+00

Pos logit contributions
ours+0.700
 coconut+0.671
 Enjoy+0.632
rust+0.620
oral+0.605
Neg logit contributions
coat-0.554
 forehead-0.533
stand-0.518
robe-0.510
drops-0.501
Token.
Token ID13
Feature activation-1.29e-01

Pos logit contributions
amas+0.237
aces+0.236
coat+0.233
 forehead+0.226
robe+0.225
Neg logit contributions
ours-0.163
 coconut-0.158
 Enjoy-0.154
oral-0.141
rust-0.135
Token
Token ID198
Feature activation-1.88e-01

Pos logit contributions
amas+0.010
 forehead+0.009
aces+0.009
robe+0.009
stand+0.009
Neg logit contributions
 coconut-0.010
ours-0.009
rust-0.009
 Enjoy-0.008
iw-0.008
Token
Token ID198
Feature activation-2.00e-01

Pos logit contributions
stand+0.015
amas+0.015
aces+0.014
 forehead+0.014
coat+0.014
Neg logit contributions
ours-0.013
 coconut-0.013
 Enjoy-0.011
iw-0.011
rust-0.011
TokenHer
Token ID9360
Feature activation-2.32e-01

Pos logit contributions
amas+0.013
aces+0.013
 forehead+0.012
coat+0.012
stand+0.012
Neg logit contributions
ours-0.017
 coconut-0.016
 Enjoy-0.015
iw-0.015
rust-0.015
Token did
Token ID750
Feature activation+7.17e-01

Pos logit contributions
ours+0.027
 coconut+0.026
 Enjoy+0.026
oral+0.024
rust+0.023
Neg logit contributions
aces-0.036
amas-0.035
coat-0.035
robe-0.034
 forehead-0.033
Token you
Token ID345
Feature activation+1.94e-01

Pos logit contributions
ours+0.045
 coconut+0.044
 Enjoy+0.043
oral+0.039
rust+0.038
Neg logit contributions
aces-0.062
coat-0.061
amas-0.061
robe-0.059
drops-0.058
Token not
Token ID407
Feature activation-7.19e-01

Pos logit contributions
ours+0.013
 coconut+0.013
 Enjoy+0.013
oral+0.012
rust+0.012
Neg logit contributions
aces-0.016
amas-0.015
coat-0.015
robe-0.015
 forehead-0.014
Token share
Token ID2648
Feature activation+2.38e+00

Pos logit contributions
coat+0.058
aces+0.057
amas+0.056
stand+0.055
drops+0.054
Neg logit contributions
ours-0.050
 coconut-0.048
 Enjoy-0.045
oral-0.044
iw-0.041
Token the
Token ID262
Feature activation+1.73e-01

Pos logit contributions
ours+0.152
 coconut+0.147
 Enjoy+0.145
oral+0.136
rust+0.129
Neg logit contributions
aces-0.203
amas-0.199
coat-0.197
stand-0.189
robe-0.189
Token peach
Token ID47565
Feature activation+8.51e+00

Pos logit contributions
ours+0.012
 coconut+0.011
 Enjoy+0.011
oral+0.010
rust+0.010
Neg logit contributions
aces-0.014
coat-0.014
amas-0.014
 forehead-0.013
robe-0.013
Token?"
Token ID1701
Feature activation+3.16e+00

Pos logit contributions
ours+0.411
 coconut+0.394
 Enjoy+0.383
oral+0.345
rust+0.330
Neg logit contributions
aces-0.856
amas-0.840
coat-0.838
robe-0.811
 forehead-0.810
Token
Token ID198
Feature activation+2.12e-01

Pos logit contributions
 coconut+0.235
ours+0.221
 Enjoy+0.206
rust+0.204
iw+0.191
Neg logit contributions
stand-0.245
amas-0.241
aces-0.229
 forehead-0.228
robe-0.223
Token
Token ID198
Feature activation+2.01e-01

Pos logit contributions
 coconut+0.013
ours+0.013
 Enjoy+0.012
rust+0.011
iw+0.011
Neg logit contributions
stand-0.018
amas-0.018
aces-0.017
 forehead-0.017
robe-0.017
TokenShe
Token ID3347
Feature activation-1.41e+00

Pos logit contributions
ours+0.014
 coconut+0.014
 Enjoy+0.013
oral+0.012
rust+0.012
Neg logit contributions
aces-0.016
amas-0.016
coat-0.015
stand-0.015
 forehead-0.015
Token was
Token ID373
Feature activation+8.31e-01

Pos logit contributions
aces+0.110
coat+0.107
amas+0.106
 forehead+0.103
robe+0.102
Neg logit contributions
 coconut-0.099
 Enjoy-0.098
ours-0.097
oral-0.089
rust-0.087
Token They
Token ID1119
Feature activation+1.16e+00

Pos logit contributions
 coconut+0.000
ours+0.000
 Enjoy+0.000
oral+0.000
rust+0.000
Neg logit contributions
amas-0.001
aces-0.001
coat-0.001
drops-0.001
robe-0.001
Token run
Token ID1057
Feature activation+6.35e-02

Pos logit contributions
ours+0.086
 coconut+0.081
 Enjoy+0.078
oral+0.076
iw+0.073
Neg logit contributions
aces-0.088
coat-0.086
 forehead-0.084
drops-0.084
amas-0.081
Token away
Token ID1497
Feature activation-1.22e+00

Pos logit contributions
ours+0.003
 Enjoy+0.003
 coconut+0.003
 Sometimes+0.002
oral+0.002
Neg logit contributions
aces-0.007
amas-0.006
 forehead-0.006
coat-0.006
bridge-0.006
Token from
Token ID422
Feature activation-1.74e-01

Pos logit contributions
aces+0.078
coat+0.077
 forehead+0.076
amas+0.075
drops+0.075
Neg logit contributions
ours-0.104
 coconut-0.101
 Enjoy-0.100
rust-0.093
oral-0.093
Token the
Token ID262
Feature activation+3.80e-01

Pos logit contributions
 forehead+0.011
drops+0.011
coat+0.011
aces+0.010
stand+0.010
Neg logit contributions
oral-0.014
ours-0.014
 Enjoy-0.014
 coconut-0.012
rust-0.012
Token temple
Token ID12505
Feature activation+8.51e+00

Pos logit contributions
ours+0.028
 Enjoy+0.027
 coconut+0.026
oral+0.025
rust+0.024
Neg logit contributions
coat-0.028
aces-0.028
amas-0.028
 forehead-0.027
robe-0.027
Token.
Token ID13
Feature activation+1.17e-01

Pos logit contributions
ours+0.515
 coconut+0.494
 Enjoy+0.467
oral+0.448
iw+0.420
Neg logit contributions
aces-0.741
amas-0.739
coat-0.727
robe-0.705
 forehead-0.700
Token They
Token ID1119
Feature activation+1.18e+00

Pos logit contributions
 coconut+0.008
ours+0.008
oral+0.008
rust+0.007
iw+0.007
Neg logit contributions
amas-0.009
stand-0.009
robe-0.009
aces-0.008
coat-0.008
Token promise
Token ID6991
Feature activation+3.01e+00

Pos logit contributions
ours+0.087
 coconut+0.083
 Enjoy+0.080
oral+0.078
rust+0.074
Neg logit contributions
aces-0.089
coat-0.088
drops-0.085
 forehead-0.084
amas-0.083
Token to
Token ID284
Feature activation+7.32e-01

Pos logit contributions
ours+0.137
 coconut+0.132
 Enjoy+0.130
oral+0.116
rust+0.111
Neg logit contributions
aces-0.318
amas-0.304
coat-0.301
bridge-0.297
robe-0.294
Token never
Token ID1239
Feature activation-5.74e-01

Pos logit contributions
ours+0.055
 coconut+0.054
 Enjoy+0.052
oral+0.049
rust+0.048
Neg logit contributions
aces-0.054
coat-0.053
amas-0.053
 forehead-0.051
robe-0.050
Token that
Token ID326
Feature activation-5.77e-01

Pos logit contributions
aces+0.027
amas+0.027
coat+0.026
drops+0.025
robe+0.025
Neg logit contributions
ours-0.018
 coconut-0.017
 Enjoy-0.017
oral-0.016
rust-0.015
Token you
Token ID345
Feature activation+2.08e-01

Pos logit contributions
aces+0.042
coat+0.040
amas+0.040
drops+0.039
 forehead+0.039
Neg logit contributions
ours-0.043
 coconut-0.041
 Enjoy-0.041
oral-0.040
rust-0.038
Token will
Token ID481
Feature activation-5.52e-01

Pos logit contributions
ours+0.013
 coconut+0.012
 Enjoy+0.012
oral+0.011
rust+0.011
Neg logit contributions
aces-0.018
coat-0.018
amas-0.018
 forehead-0.017
robe-0.017
Token eat
Token ID4483
Feature activation+3.40e+00

Pos logit contributions
aces+0.048
amas+0.047
coat+0.046
robe+0.045
 forehead+0.044
Neg logit contributions
 Enjoy-0.034
 coconut-0.033
ours-0.033
oral-0.031
rust-0.029
Token your
Token ID534
Feature activation+1.42e+00

Pos logit contributions
ours+0.234
 coconut+0.222
 Enjoy+0.218
oral+0.212
rust+0.202
Neg logit contributions
aces-0.276
amas-0.264
coat-0.260
 forehead-0.253
stand-0.251
Token peach
Token ID47565
Feature activation+8.48e+00

Pos logit contributions
ours+0.114
 Enjoy+0.106
 coconut+0.102
oral+0.100
rust+0.099
Neg logit contributions
aces-0.099
 forehead-0.098
amas-0.097
coat-0.095
robe-0.092
Token,
Token ID11
Feature activation+1.92e-01

Pos logit contributions
ours+0.486
 coconut+0.482
 Enjoy+0.444
oral+0.418
iw+0.412
Neg logit contributions
amas-0.744
aces-0.737
coat-0.736
robe-0.714
 forehead-0.708
Token and
Token ID290
Feature activation+2.97e-01

Pos logit contributions
ours+0.014
 coconut+0.014
 Enjoy+0.013
oral+0.012
rust+0.012
Neg logit contributions
aces-0.015
amas-0.014
coat-0.014
robe-0.014
drops-0.014
Token rest
Token ID1334
Feature activation-7.14e-01

Pos logit contributions
ours+0.022
 coconut+0.021
oral+0.021
rust+0.020
 Enjoy+0.020
Neg logit contributions
aces-0.021
 forehead-0.021
stand-0.020
drops-0.020
coat-0.020
Token your
Token ID534
Feature activation+1.37e+00

Pos logit contributions
aces+0.060
amas+0.060
coat+0.059
drops+0.057
robe+0.056
Neg logit contributions
ours-0.046
 coconut-0.044
 Enjoy-0.044
oral-0.040
rust-0.038
Token knee
Token ID10329
Feature activation-2.42e+00

Pos logit contributions
ours+0.096
 coconut+0.092
 Enjoy+0.091
oral+0.084
rust+0.083
Neg logit contributions
aces-0.110
amas-0.107
coat-0.107
 forehead-0.105
robe-0.102
Token stick
Token ID4859
Feature activation-8.55e-02

Pos logit contributions
ours+0.022
 Enjoy+0.021
 coconut+0.020
rust+0.019
oral+0.019
Neg logit contributions
aces-0.025
coat-0.024
amas-0.024
drops-0.024
 forehead-0.024
Token.
Token ID13
Feature activation+2.34e-01

Pos logit contributions
amas+0.008
coat+0.008
robe+0.007
 forehead+0.007
aces+0.007
Neg logit contributions
ours-0.005
 coconut-0.004
 Enjoy-0.004
 Sometimes-0.004
 asks-0.004
Token She
Token ID1375
Feature activation-9.05e-02

Pos logit contributions
 coconut+0.019
ours+0.019
oral+0.018
 Enjoy+0.017
rust+0.017
Neg logit contributions
aces-0.016
amas-0.015
coat-0.015
robe-0.015
stand-0.014
Token hits
Token ID7127
Feature activation-8.31e-01

Pos logit contributions
aces+0.008
coat+0.007
drops+0.007
amas+0.007
robe+0.007
Neg logit contributions
 coconut-0.006
 Enjoy-0.006
ours-0.006
oral-0.005
rust-0.005
Token a
Token ID257
Feature activation-3.69e-01

Pos logit contributions
aces+0.075
amas+0.072
coat+0.072
 forehead+0.070
robe+0.070
Neg logit contributions
ours-0.050
 coconut-0.048
 Enjoy-0.047
oral-0.044
rust-0.042
Token peach
Token ID47565
Feature activation+8.47e+00

Pos logit contributions
aces+0.029
amas+0.029
coat+0.029
 forehead+0.027
robe+0.027
Neg logit contributions
ours-0.026
 coconut-0.025
 Enjoy-0.024
oral-0.023
rust-0.022
Token.
Token ID13
Feature activation+2.36e-01

Pos logit contributions
ours+0.528
 coconut+0.514
 Enjoy+0.474
oral+0.448
iw+0.436
Neg logit contributions
amas-0.732
coat-0.721
aces-0.715
robe-0.694
 forehead-0.692
Token The
Token ID383
Feature activation+7.36e-01

Pos logit contributions
 coconut+0.017
ours+0.017
rust+0.016
oral+0.016
 Enjoy+0.015
Neg logit contributions
aces-0.018
amas-0.018
stand-0.017
coat-0.017
robe-0.017
Token peach
Token ID47565
Feature activation+8.45e+00

Pos logit contributions
ours+0.033
 Enjoy+0.031
oral+0.027
rust+0.027
 coconut+0.027
Neg logit contributions
aces-0.078
coat-0.075
 forehead-0.075
amas-0.074
drops-0.074
Token falls
Token ID8953
Feature activation-6.32e-01

Pos logit contributions
ours+0.504
 coconut+0.479
 Enjoy+0.450
oral+0.424
iw+0.403
Neg logit contributions
coat-0.754
amas-0.753
aces-0.750
robe-0.726
 forehead-0.723
Token down
Token ID866
Feature activation-1.04e+00

Pos logit contributions
bes+0.052
amas+0.052
 forehead+0.051
aces+0.051
coat+0.051
Neg logit contributions
ours-0.039
 Sometimes-0.037
 coconut-0.036
 Enjoy-0.034
 asks-0.033
Token
Token ID198
Feature activation+1.21e-01

Pos logit contributions
ours+0.011
 coconut+0.011
 Enjoy+0.010
 seafood+0.010
rust+0.009
Neg logit contributions
stand-0.010
amas-0.009
aces-0.009
bes-0.009
coat-0.009
TokenShe
Token ID3347
Feature activation-1.45e+00

Pos logit contributions
ours+0.011
 coconut+0.010
 Enjoy+0.010
add+0.009
 Crab+0.009
Neg logit contributions
stand-0.007
coat-0.007
aces-0.007
bes-0.006
amas-0.006
Token pointed
Token ID6235
Feature activation+1.92e+00

Pos logit contributions
aces+0.112
coat+0.107
amas+0.102
 forehead+0.102
drops+0.102
Neg logit contributions
 coconut-0.102
 Enjoy-0.098
oral-0.097
rust-0.094
ours-0.092
Token to
Token ID284
Feature activation+7.15e-01

Pos logit contributions
 coconut+0.129
ours+0.128
 Enjoy+0.126
oral+0.125
rust+0.118
Neg logit contributions
aces-0.154
coat-0.142
 forehead-0.139
amas-0.137
drops-0.136
Token the
Token ID262
Feature activation+1.45e-01

Pos logit contributions
ours+0.058
oral+0.053
 Enjoy+0.052
 coconut+0.051
iw+0.048
Neg logit contributions
aces-0.048
coat-0.045
stand-0.045
 forehead-0.044
amas-0.043
Token peach
Token ID47565
Feature activation+8.46e+00

Pos logit contributions
ours+0.009
 Enjoy+0.009
oral+0.008
rust+0.008
 coconut+0.008
Neg logit contributions
aces-0.012
coat-0.012
 forehead-0.012
amas-0.012
drops-0.012
Token.
Token ID13
Feature activation+3.52e-01

Pos logit contributions
ours+0.577
 Enjoy+0.559
 coconut+0.557
oral+0.512
rust+0.503
Neg logit contributions
aces-0.690
coat-0.663
amas-0.663
 forehead-0.639
robe-0.639
Token It
Token ID632
Feature activation-3.42e-02

Pos logit contributions
 coconut+0.030
ours+0.029
rust+0.028
oral+0.028
 Enjoy+0.027
Neg logit contributions
aces-0.023
robe-0.021
amas-0.021
coat-0.021
stand-0.021
Token was
Token ID373
Feature activation+8.04e-01

Pos logit contributions
aces+0.003
coat+0.003
robe+0.003
drops+0.002
bridge+0.002
Neg logit contributions
 Enjoy-0.002
ours-0.002
 coconut-0.002
oral-0.002
rous-0.002
Token brown
Token ID7586
Feature activation-1.72e+00

Pos logit contributions
ours+0.057
 coconut+0.056
 Enjoy+0.054
oral+0.050
rust+0.050
Neg logit contributions
aces-0.063
amas-0.062
coat-0.062
robe-0.059
 forehead-0.058
Token and
Token ID290
Feature activation+3.89e-01

Pos logit contributions
amas+0.151
coat+0.148
aces+0.147
robe+0.144
 forehead+0.143
Neg logit contributions
ours-0.102
 coconut-0.100
 Enjoy-0.095
iw-0.084
oral-0.083
Token hits
Token ID7127
Feature activation-8.31e-01

Pos logit contributions
aces+0.008
coat+0.007
drops+0.007
amas+0.007
robe+0.007
Neg logit contributions
 coconut-0.006
 Enjoy-0.006
ours-0.006
oral-0.005
rust-0.005
Token a
Token ID257
Feature activation-3.69e-01

Pos logit contributions
aces+0.075
amas+0.072
coat+0.072
 forehead+0.070
robe+0.070
Neg logit contributions
ours-0.050
 coconut-0.048
 Enjoy-0.047
oral-0.044
rust-0.042
Token peach
Token ID47565
Feature activation+8.47e+00

Pos logit contributions
aces+0.029
amas+0.029
coat+0.029
 forehead+0.027
robe+0.027
Neg logit contributions
ours-0.026
 coconut-0.025
 Enjoy-0.024
oral-0.023
rust-0.022
Token.
Token ID13
Feature activation+2.36e-01

Pos logit contributions
ours+0.528
 coconut+0.514
 Enjoy+0.474
oral+0.448
iw+0.436
Neg logit contributions
amas-0.732
coat-0.721
aces-0.715
robe-0.694
 forehead-0.692
Token The
Token ID383
Feature activation+7.36e-01

Pos logit contributions
 coconut+0.017
ours+0.017
rust+0.016
oral+0.016
 Enjoy+0.015
Neg logit contributions
aces-0.018
amas-0.018
stand-0.017
coat-0.017
robe-0.017
Token peach
Token ID47565
Feature activation+8.45e+00

Pos logit contributions
ours+0.033
 Enjoy+0.031
oral+0.027
rust+0.027
 coconut+0.027
Neg logit contributions
aces-0.078
coat-0.075
 forehead-0.075
amas-0.074
drops-0.074
Token falls
Token ID8953
Feature activation-6.32e-01

Pos logit contributions
ours+0.504
 coconut+0.479
 Enjoy+0.450
oral+0.424
iw+0.403
Neg logit contributions
coat-0.754
amas-0.753
aces-0.750
robe-0.726
 forehead-0.723
Token down
Token ID866
Feature activation-1.04e+00

Pos logit contributions
bes+0.052
amas+0.052
 forehead+0.051
aces+0.051
coat+0.051
Neg logit contributions
ours-0.039
 Sometimes-0.037
 coconut-0.036
 Enjoy-0.034
 asks-0.033
Token.
Token ID13
Feature activation+2.58e-01

Pos logit contributions
aces+0.087
amas+0.085
coat+0.085
robe+0.082
 forehead+0.081
Neg logit contributions
ours-0.068
 coconut-0.066
 Enjoy-0.065
oral-0.060
rust-0.058
Token Ben
Token ID3932
Feature activation-4.52e-01

Pos logit contributions
 coconut+0.018
ours+0.018
rust+0.017
oral+0.017
 Enjoy+0.016
Neg logit contributions
aces-0.020
amas-0.020
stand-0.019
robe-0.019
coat-0.019
Token catches
Token ID17591
Feature activation-2.41e-01

Pos logit contributions
aces+0.039
coat+0.039
amas+0.038
robe+0.037
 forehead+0.037
Neg logit contributions
ours-0.029
 coconut-0.027
 Enjoy-0.026
oral-0.024
rust-0.023
Token had
Token ID550
Feature activation-5.01e-01

Pos logit contributions
aces+0.117
coat+0.113
amas+0.111
robe+0.109
 forehead+0.107
Neg logit contributions
 coconut-0.094
ours-0.094
 Enjoy-0.092
oral-0.085
rust-0.083
Token to
Token ID284
Feature activation+4.49e-01

Pos logit contributions
amas+0.046
aces+0.045
coat+0.044
 forehead+0.044
robe+0.044
Neg logit contributions
 coconut-0.028
ours-0.027
 Enjoy-0.027
grown-0.024
oral-0.024
Token know
Token ID760
Feature activation+2.15e+00

Pos logit contributions
 coconut+0.026
ours+0.026
 Enjoy+0.026
oral+0.023
rust+0.022
Neg logit contributions
aces-0.040
amas-0.039
coat-0.039
robe-0.038
drops-0.037
Token all
Token ID477
Feature activation+2.46e+00

Pos logit contributions
ours+0.092
 coconut+0.089
 Enjoy+0.086
oral+0.076
rust+0.074
Neg logit contributions
aces-0.228
amas-0.222
coat-0.222
robe-0.216
 forehead-0.215
Token about
Token ID546
Feature activation+1.48e+00

Pos logit contributions
ours+0.138
 coconut+0.134
 Enjoy+0.130
oral+0.120
rust+0.117
Neg logit contributions
aces-0.228
coat-0.222
amas-0.221
 forehead-0.214
robe-0.214
Token microsc
Token ID21145
Feature activation+8.45e+00

Pos logit contributions
ours+0.102
 coconut+0.097
 Enjoy+0.096
oral+0.092
rust+0.090
Neg logit contributions
aces-0.120
coat-0.114
amas-0.113
robe-0.111
stand-0.110
Tokenopes
Token ID13920
Feature activation-2.80e+00

Pos logit contributions
ours+0.702
 coconut+0.670
 Enjoy+0.636
rust+0.607
oral+0.599
Neg logit contributions
coat-0.542
 forehead-0.533
robe-0.505
stand-0.504
drops-0.487
Token.
Token ID13
Feature activation+1.38e-02

Pos logit contributions
amas+0.259
 forehead+0.252
coat+0.250
aces+0.249
robe+0.243
Neg logit contributions
ours-0.150
 coconut-0.147
 Enjoy-0.141
grown-0.125
oral-0.123
Token
Token ID220
Feature activation+3.98e-02

Pos logit contributions
 coconut+0.001
ours+0.001
rust+0.001
 Enjoy+0.001
oral+0.001
Neg logit contributions
amas-0.001
aces-0.001
robe-0.001
 forehead-0.001
stand-0.001
Token
Token ID198
Feature activation-3.17e-01

Pos logit contributions
ours+0.002
 coconut+0.002
 Enjoy+0.002
oral+0.001
rust+0.001
Neg logit contributions
aces-0.004
coat-0.004
amas-0.004
 forehead-0.004
robe-0.004
Token
Token ID198
Feature activation-3.29e-01

Pos logit contributions
aces+0.027
amas+0.027
stand+0.026
coat+0.025
robe+0.025
Neg logit contributions
ours-0.021
 coconut-0.020
 Enjoy-0.018
iw-0.017
rust-0.016
Token -
Token ID532
Feature activation+1.33e+00

Pos logit contributions
amas+0.125
aces+0.123
coat+0.120
drops+0.118
 forehead+0.117
Neg logit contributions
ours-0.067
 Enjoy-0.065
 coconut-0.065
oral-0.058
arium-0.055
Token a
Token ID257
Feature activation-5.80e-01

Pos logit contributions
ours+0.086
 coconut+0.083
 Enjoy+0.082
oral+0.075
rust+0.073
Neg logit contributions
aces-0.112
amas-0.109
coat-0.109
robe-0.105
 forehead-0.105
Token brand
Token ID4508
Feature activation-4.82e+00

Pos logit contributions
aces+0.043
coat+0.042
amas+0.042
 forehead+0.040
robe+0.040
Neg logit contributions
ours-0.043
 coconut-0.042
 Enjoy-0.041
oral-0.038
rust-0.038
Token new
Token ID649
Feature activation+1.26e+00

Pos logit contributions
robe+0.334
 forehead+0.332
coat+0.330
amas+0.329
brush+0.317
Neg logit contributions
ours-0.360
 coconut-0.343
 Enjoy-0.329
rust-0.321
oral-0.315
Token bath
Token ID7837
Feature activation-2.60e+00

Pos logit contributions
ours+0.095
 coconut+0.089
 Enjoy+0.089
oral+0.084
rust+0.082
Neg logit contributions
coat-0.092
aces-0.092
amas-0.089
robe-0.088
 forehead-0.087
Tokenrobe
Token ID25481
Feature activation-1.15e+01

Pos logit contributions
aces+0.189
 forehead+0.183
amas+0.183
stand+0.174
drops+0.171
Neg logit contributions
ours-0.173
 Enjoy-0.169
 coconut-0.164
 arose-0.160
arium-0.157
Token.
Token ID13
Feature activation-4.21e-01

Pos logit contributions
amas+1.127
aces+1.110
 forehead+1.077
drops+1.066
stand+1.045
Neg logit contributions
 Enjoy-0.535
ours-0.488
 coconut-0.476
arium-0.454
grown-0.447
Token It
Token ID632
Feature activation-9.51e-01

Pos logit contributions
aces+0.035
amas+0.034
coat+0.034
stand+0.033
robe+0.033
Neg logit contributions
ours-0.028
 coconut-0.028
oral-0.025
 Enjoy-0.025
rust-0.024
Token was
Token ID373
Feature activation+3.23e-01

Pos logit contributions
amas+0.103
aces+0.103
coat+0.100
 forehead+0.097
drops+0.097
Neg logit contributions
ours-0.039
 Enjoy-0.035
 coconut-0.034
 asks-0.030
rust-0.030
Token soft
Token ID2705
Feature activation-4.11e+00

Pos logit contributions
ours+0.021
 Enjoy+0.020
 coconut+0.020
grown+0.018
 Sometimes+0.018
Neg logit contributions
amas-0.027
aces-0.027
coat-0.026
drops-0.026
 forehead-0.025
Token,
Token ID11
Feature activation-2.81e-01

Pos logit contributions
aces+0.437
amas+0.418
bridge+0.409
stand+0.409
drops+0.406
Neg logit contributions
 Enjoy-0.152
ours-0.150
 coconut-0.145
grown-0.132
 Sometimes-0.127
Token Tim
Token ID5045
Feature activation+1.81e+00

Pos logit contributions
ours+0.018
 coconut+0.018
 Enjoy+0.017
oral+0.016
rust+0.015
Neg logit contributions
aces-0.026
amas-0.025
coat-0.025
robe-0.024
 forehead-0.024
Token saw
Token ID2497
Feature activation-4.59e-01

Pos logit contributions
ours+0.115
 coconut+0.110
 Enjoy+0.109
oral+0.100
rust+0.099
Neg logit contributions
aces-0.154
amas-0.152
coat-0.148
robe-0.145
drops-0.145
Token that
Token ID326
Feature activation-1.25e+00

Pos logit contributions
aces+0.037
coat+0.036
amas+0.036
robe+0.035
 forehead+0.035
Neg logit contributions
ours-0.031
 Enjoy-0.030
 coconut-0.029
oral-0.028
iw-0.026
Token his
Token ID465
Feature activation+8.35e-01

Pos logit contributions
coat+0.102
aces+0.102
robe+0.100
amas+0.100
stand+0.099
Neg logit contributions
ours-0.084
 Enjoy-0.079
 coconut-0.075
oral-0.074
iw-0.068
Token bath
Token ID7837
Feature activation-2.39e+00

Pos logit contributions
ours+0.060
 Enjoy+0.058
 coconut+0.056
oral+0.054
rust+0.051
Neg logit contributions
aces-0.065
coat-0.064
amas-0.063
 forehead-0.062
robe-0.062
Tokenrobe
Token ID25481
Feature activation-1.13e+01

Pos logit contributions
aces+0.135
 forehead+0.134
bridge+0.128
 click+0.125
ring+0.122
Neg logit contributions
ours-0.196
 arose-0.184
 swamp-0.174
 shrimp-0.173
 crab-0.172
Token was
Token ID373
Feature activation+3.79e-01

Pos logit contributions
aces+1.274
amas+1.262
 forehead+1.247
drops+1.223
coat+1.207
Neg logit contributions
ours-0.306
 coconut-0.297
 Enjoy-0.262
 arose-0.261
:-0.236
Token too
Token ID1165
Feature activation+1.06e+00

Pos logit contributions
 coconut+0.021
ours+0.020
 Enjoy+0.019
grown+0.019
 swamp+0.018
Neg logit contributions
aces-0.034
amas-0.032
drops-0.032
 forehead-0.032
coat-0.031
Token tight
Token ID5381
Feature activation+1.58e+00

Pos logit contributions
ours+0.073
 coconut+0.072
 Enjoy+0.069
oral+0.064
rust+0.063
Neg logit contributions
aces-0.087
amas-0.083
coat-0.082
 forehead-0.081
drops-0.080
Token.
Token ID13
Feature activation-8.83e-02

Pos logit contributions
:+0.047
ours+0.047
 Enjoy+0.040
 arose+0.039
grown+0.036
Neg logit contributions
bridge-0.176
aces-0.168
 forehead-0.164
amed-0.163
amas-0.163
Token He
Token ID679
Feature activation-2.91e-01

Pos logit contributions
amas+0.007
aces+0.007
coat+0.007
robe+0.006
stand+0.006
Neg logit contributions
ours-0.006
 coconut-0.006
oral-0.006
 Enjoy-0.006
rous-0.006
Token to
Token ID284
Feature activation+7.86e-01

Pos logit contributions
ours+0.147
 Enjoy+0.143
 coconut+0.143
oral+0.134
iw+0.126
Neg logit contributions
aces-0.140
coat-0.139
amas-0.137
robe-0.131
 forehead-0.128
Token wear
Token ID5806
Feature activation-9.81e-01

Pos logit contributions
ours+0.053
 coconut+0.051
 Enjoy+0.050
oral+0.047
rust+0.045
Neg logit contributions
aces-0.063
coat-0.062
amas-0.062
 forehead-0.060
robe-0.060
Token her
Token ID607
Feature activation-1.12e+00

Pos logit contributions
amas+0.086
drops+0.084
aces+0.083
coat+0.083
 forehead+0.081
Neg logit contributions
ours-0.056
 Enjoy-0.056
 coconut-0.050
 Sometimes-0.048
rous-0.047
Token cozy
Token ID37438
Feature activation-3.30e+00

Pos logit contributions
amas+0.082
aces+0.081
drops+0.079
coat+0.078
bridge+0.075
Neg logit contributions
ours-0.080
 coconut-0.078
 Enjoy-0.075
oral-0.070
 whose-0.070
Token bath
Token ID7837
Feature activation-2.37e+00

Pos logit contributions
aces+0.199
drops+0.190
amas+0.184
bridge+0.184
coat+0.177
Neg logit contributions
 coconut-0.275
ours-0.275
 Enjoy-0.261
 swamp-0.253
arium-0.252
Tokenrobe
Token ID25481
Feature activation-1.13e+01

Pos logit contributions
 click+0.119
bridge+0.114
drops+0.114
 forehead+0.111
aces+0.108
Neg logit contributions
ours-0.192
arium-0.188
?-0.184
 swamp-0.180
 shrimp-0.179
Token after
Token ID706
Feature activation+9.28e-01

Pos logit contributions
drops+1.122
amas+1.115
aces+1.081
 forehead+1.059
bridge+1.051
Neg logit contributions
 Enjoy-0.422
arium-0.422
ours-0.412
?-0.377
:-0.376
Token taking
Token ID2263
Feature activation-2.43e-01

Pos logit contributions
ours+0.061
 coconut+0.058
 Enjoy+0.058
oral+0.054
rust+0.052
Neg logit contributions
aces-0.077
coat-0.076
amas-0.075
 forehead-0.072
drops-0.072
Token a
Token ID257
Feature activation-6.12e-01

Pos logit contributions
amas+0.022
aces+0.022
coat+0.022
drops+0.021
bridge+0.021
Neg logit contributions
 Enjoy-0.013
ours-0.012
 Sometimes-0.011
grown-0.011
 coconut-0.011
Token bath
Token ID7837
Feature activation-2.46e+00

Pos logit contributions
aces+0.052
amas+0.047
drops+0.045
bridge+0.045
coat+0.045
Neg logit contributions
 coconut-0.041
 Enjoy-0.041
ours-0.040
oral-0.038
arium-0.036
Token.
Token ID13
Feature activation-2.68e-01

Pos logit contributions
aces+0.232
bridge+0.222
 forehead+0.218
drops+0.212
 click+0.212
Neg logit contributions
ours-0.110
 Enjoy-0.102
?-0.101
 Sometimes-0.095
 arose-0.095
Token had
Token ID550
Feature activation-5.04e-01

Pos logit contributions
coat+0.035
aces+0.035
amas+0.034
robe+0.033
drops+0.033
Neg logit contributions
ours-0.031
 coconut-0.030
 Enjoy-0.029
oral-0.027
rust-0.026
Token a
Token ID257
Feature activation-5.74e-01

Pos logit contributions
amas+0.049
aces+0.047
 forehead+0.047
coat+0.046
drops+0.046
Neg logit contributions
 coconut-0.025
 Enjoy-0.025
ours-0.024
grown-0.022
 Sometimes-0.020
Token lovely
Token ID14081
Feature activation-7.77e-01

Pos logit contributions
coat+0.038
robe+0.035
amas+0.035
 forehead+0.035
aces+0.034
Neg logit contributions
ours-0.049
 Enjoy-0.046
 coconut-0.045
oral-0.044
rust-0.044
Token blue
Token ID4171
Feature activation-1.45e+00

Pos logit contributions
aces+0.057
amas+0.055
coat+0.055
robe+0.053
 forehead+0.053
Neg logit contributions
ours-0.059
 coconut-0.057
 Enjoy-0.056
oral-0.053
rust-0.051
Token bath
Token ID7837
Feature activation-2.26e+00

Pos logit contributions
coat+0.083
aces+0.083
amas+0.081
robe+0.078
 forehead+0.077
Neg logit contributions
ours-0.132
 coconut-0.127
 Enjoy-0.127
oral-0.121
rust-0.120
Tokenrobe
Token ID25481
Feature activation-1.13e+01

Pos logit contributions
amas+0.106
 forehead+0.105
aces+0.103
 click+0.096
bridge+0.096
Neg logit contributions
ours-0.201
 Enjoy-0.191
 swamp-0.188
arium-0.187
:-0.186
Token.
Token ID13
Feature activation+9.88e-02

Pos logit contributions
amas+1.197
aces+1.149
 forehead+1.138
drops+1.125
bes+1.106
Neg logit contributions
 Enjoy-0.446
ours-0.389
arium-0.366
:-0.363
 named-0.348
Token Lucy
Token ID22162
Feature activation+9.51e-01

Pos logit contributions
 coconut+0.007
ours+0.007
oral+0.006
 Enjoy+0.006
rust+0.006
Neg logit contributions
amas-0.008
aces-0.008
coat-0.008
robe-0.007
stand-0.007
Token loved
Token ID6151
Feature activation-7.01e-01

Pos logit contributions
ours+0.051
 coconut+0.048
 Enjoy+0.047
rust+0.041
oral+0.041
Neg logit contributions
aces-0.090
amas-0.089
coat-0.087
 forehead-0.085
robe-0.085
Token to
Token ID284
Feature activation+3.37e-01

Pos logit contributions
aces+0.061
coat+0.060
amas+0.060
robe+0.058
 forehead+0.058
Neg logit contributions
ours-0.043
 coconut-0.042
 Enjoy-0.041
oral-0.038
rust-0.037
Token play
Token ID711
Feature activation-8.14e-02

Pos logit contributions
ours+0.021
 Enjoy+0.020
 coconut+0.019
 Sometimes+0.017
oral+0.017
Neg logit contributions
aces-0.030
amas-0.029
coat-0.028
drops-0.028
 forehead-0.027
Token to
Token ID284
Feature activation+4.31e-01

Pos logit contributions
coat+0.052
amas+0.051
aces+0.051
stand+0.049
drops+0.048
Neg logit contributions
ours-0.051
oral-0.049
 coconut-0.048
 Enjoy-0.047
rust-0.045
Token wear
Token ID5806
Feature activation-1.06e+00

Pos logit contributions
ours+0.029
 coconut+0.027
 Enjoy+0.027
oral+0.026
rust+0.025
Neg logit contributions
aces-0.036
coat-0.035
amas-0.035
 forehead-0.034
robe-0.034
Token her
Token ID607
Feature activation-1.35e+00

Pos logit contributions
 forehead+0.111
drops+0.111
amas+0.110
bridge+0.106
aces+0.105
Neg logit contributions
 what-0.043
 Enjoy-0.040
ours-0.038
 shrimp-0.038
rous-0.035
Token cozy
Token ID37438
Feature activation-3.21e+00

Pos logit contributions
drops+0.123
amas+0.122
bridge+0.118
aces+0.118
 forehead+0.112
Neg logit contributions
ours-0.070
 whose-0.064
 coconut-0.064
 shrimp-0.062
 Enjoy-0.061
Token bath
Token ID7837
Feature activation-2.28e+00

Pos logit contributions
aces+0.205
drops+0.204
bridge+0.197
amed+0.183
amas+0.183
Neg logit contributions
 coconut-0.256
ours-0.249
 swamp-0.239
 Enjoy-0.238
 oy-0.235
Tokenrobe
Token ID25481
Feature activation-1.13e+01

Pos logit contributions
 click+0.108
bridge+0.103
plets+0.101
mic+0.098
 forehead+0.096
Neg logit contributions
arium-0.192
?-0.192
:-0.187
 swamp-0.184
 Sometimes-0.184
Token after
Token ID706
Feature activation+1.12e+00

Pos logit contributions
drops+1.229
amas+1.213
bridge+1.184
 forehead+1.174
bes+1.160
Neg logit contributions
arium-0.333
:-0.318
 Enjoy-0.311
?-0.305
 Sometimes-0.297
Token taking
Token ID2263
Feature activation-2.62e-02

Pos logit contributions
ours+0.077
 coconut+0.074
 Enjoy+0.073
oral+0.068
rust+0.066
Neg logit contributions
aces-0.091
coat-0.089
amas-0.088
drops-0.085
 forehead-0.084
Token a
Token ID257
Feature activation-5.58e-01

Pos logit contributions
amas+0.002
 forehead+0.002
coat+0.002
bridge+0.002
drops+0.002
Neg logit contributions
 Enjoy-0.001
ours-0.001
 Sometimes-0.001
grown-0.001
 what-0.001
Token bath
Token ID7837
Feature activation-2.29e+00

Pos logit contributions
aces+0.048
amas+0.044
drops+0.043
bridge+0.043
coat+0.042
Neg logit contributions
ours-0.037
 Enjoy-0.036
 coconut-0.035
oral-0.033
 Sometimes-0.032
Token.
Token ID13
Feature activation+3.95e-02

Pos logit contributions
 forehead+0.232
bridge+0.228
aces+0.226
 click+0.221
amas+0.219
Neg logit contributions
ours-0.078
?-0.077
 Enjoy-0.073
arium-0.070
 swamp-0.069
Token to
Token ID284
Feature activation+4.38e-01

Pos logit contributions
amas+0.048
coat+0.047
aces+0.047
stand+0.046
 forehead+0.044
Neg logit contributions
ours-0.048
oral-0.045
 coconut-0.043
 Enjoy-0.043
rust-0.041
Token wear
Token ID5806
Feature activation-1.03e+00

Pos logit contributions
ours+0.030
 coconut+0.028
 Enjoy+0.027
oral+0.026
rust+0.025
Neg logit contributions
aces-0.036
coat-0.036
amas-0.035
 forehead-0.034
robe-0.034
Token his
Token ID465
Feature activation+9.09e-01

Pos logit contributions
drops+0.122
 forehead+0.119
amas+0.118
aces+0.116
coat+0.115
Neg logit contributions
 Enjoy-0.029
 what-0.027
 shrimp-0.025
ours-0.025
 seafood-0.023
Token soft
Token ID2705
Feature activation-3.81e+00

Pos logit contributions
ours+0.054
 coconut+0.054
 Enjoy+0.051
 shrimp+0.049
 crab+0.048
Neg logit contributions
drops-0.077
amas-0.075
aces-0.075
coat-0.071
bridge-0.071
Token bath
Token ID7837
Feature activation-2.26e+00

Pos logit contributions
drops+0.242
aces+0.237
bridge+0.224
amas+0.216
amed+0.216
Neg logit contributions
ours-0.300
 coconut-0.297
 shrimp-0.280
 Enjoy-0.280
 oy-0.279
Tokenrobe
Token ID25481
Feature activation-1.12e+01

Pos logit contributions
 click+0.091
bridge+0.087
 forehead+0.087
drops+0.085
plets+0.084
Neg logit contributions
ours-0.204
arium-0.202
?-0.199
 Sometimes-0.198
 shrimp-0.197
Token after
Token ID706
Feature activation+1.14e+00

Pos logit contributions
drops+1.199
amas+1.182
 forehead+1.146
bes+1.135
bridge+1.134
Neg logit contributions
 Enjoy-0.379
arium-0.347
 Sometimes-0.344
:-0.331
?-0.321
Token bath
Token ID7837
Feature activation-2.27e+00

Pos logit contributions
ours+0.073
 coconut+0.071
 Enjoy+0.069
oral+0.064
rust+0.062
Neg logit contributions
aces-0.097
amas-0.095
coat-0.095
robe-0.091
 forehead-0.091
Token time
Token ID640
Feature activation-1.00e+00

Pos logit contributions
 forehead+0.229
bridge+0.228
 click+0.219
aces+0.218
drops+0.215
Neg logit contributions
ours-0.091
 Enjoy-0.079
 Sometimes-0.077
 arose-0.073
?-0.072
Token.
Token ID13
Feature activation-2.02e-01

Pos logit contributions
bridge+0.107
aces+0.105
drops+0.104
coat+0.102
 forehead+0.102
Neg logit contributions
ours-0.038
 Enjoy-0.037
 coconut-0.034
 Sometimes-0.031
 Crab-0.030
Token He
Token ID679
Feature activation-2.70e-01

Pos logit contributions
amas+0.018
aces+0.017
coat+0.017
robe+0.017
stand+0.016
Neg logit contributions
 coconut-0.013
ours-0.013
oral-0.011
 Enjoy-0.011
rous-0.011
Token;
Token ID26
Feature activation+1.15e+00

Pos logit contributions
aces+0.015
amas+0.015
coat+0.015
 forehead+0.014
drops+0.014
Neg logit contributions
ours-0.010
 coconut-0.010
 Enjoy-0.009
oral-0.008
 Sometimes-0.008
Token he
Token ID339
Feature activation-1.19e-01

Pos logit contributions
ours+0.075
 coconut+0.073
oral+0.070
 Enjoy+0.069
rust+0.067
Neg logit contributions
aces-0.094
coat-0.091
amas-0.090
robe-0.089
stand-0.088
Token has
Token ID468
Feature activation+7.69e-01

Pos logit contributions
aces+0.010
coat+0.010
amas+0.010
robe+0.010
 forehead+0.010
Neg logit contributions
ours-0.007
 coconut-0.007
 Enjoy-0.007
oral-0.007
rust-0.006
Token a
Token ID257
Feature activation-3.65e-01

Pos logit contributions
ours+0.047
 coconut+0.046
 Enjoy+0.045
oral+0.041
rust+0.040
Neg logit contributions
aces-0.066
amas-0.066
coat-0.065
 forehead-0.063
robe-0.063
Token bath
Token ID7837
Feature activation-2.12e+00

Pos logit contributions
coat+0.025
aces+0.024
amas+0.024
robe+0.024
 forehead+0.024
Neg logit contributions
ours-0.029
 Enjoy-0.028
 coconut-0.027
oral-0.027
rust-0.026
Tokenrobe
Token ID25481
Feature activation-1.12e+01

Pos logit contributions
aces+0.144
amas+0.144
 forehead+0.140
stand+0.131
bridge+0.129
Neg logit contributions
ours-0.155
 coconut-0.144
 Enjoy-0.144
 crab-0.138
 shrimp-0.135
Token too
Token ID1165
Feature activation+1.42e+00

Pos logit contributions
amas+1.086
aces+1.065
 forehead+1.035
bes+1.004
drops+1.001
Neg logit contributions
 Enjoy-0.549
ours-0.519
 coconut-0.503
grown-0.449
 crab-0.437
Token."
Token ID526
Feature activation-1.69e+00

Pos logit contributions
ours+0.099
 coconut+0.097
 Enjoy+0.094
oral+0.090
rust+0.088
Neg logit contributions
aces-0.112
coat-0.110
amas-0.109
robe-0.105
stand-0.103
Token
Token ID198
Feature activation-1.70e-01

Pos logit contributions
stand+0.124
amas+0.118
aces+0.116
coat+0.112
 forehead+0.111
Neg logit contributions
ours-0.128
 coconut-0.126
oral-0.110
 Enjoy-0.109
rous-0.108
Token
Token ID198
Feature activation-2.04e-01

Pos logit contributions
stand+0.013
amas+0.012
coat+0.011
aces+0.011
bes+0.011
Neg logit contributions
ours-0.013
 coconut-0.012
 Enjoy-0.011
 seafood-0.010
 oy-0.010
TokenHer
Token ID9360
Feature activation-3.21e-01

Pos logit contributions
stand+0.013
coat+0.013
aces+0.012
amas+0.012
bridge+0.012
Neg logit contributions
ours-0.018
 coconut-0.016
 Enjoy-0.016
 volcano-0.014
arium-0.014
Token loved
Token ID6151
Feature activation-9.64e-01

Pos logit contributions
ours+0.012
 coconut+0.012
 Enjoy+0.012
oral+0.011
rust+0.010
Neg logit contributions
aces-0.019
coat-0.019
amas-0.019
robe-0.018
 forehead-0.018
Token to
Token ID284
Feature activation+1.07e-01

Pos logit contributions
aces+0.075
coat+0.075
amas+0.074
stand+0.072
drops+0.071
Neg logit contributions
ours-0.067
 coconut-0.063
oral-0.063
 Enjoy-0.061
rust-0.058
Token play
Token ID711
Feature activation-3.23e-01

Pos logit contributions
ours+0.007
 coconut+0.007
 Enjoy+0.007
oral+0.006
rust+0.006
Neg logit contributions
aces-0.009
coat-0.009
amas-0.009
 forehead-0.008
robe-0.008
Token make
Token ID787
Feature activation-3.83e+00

Pos logit contributions
 forehead+0.029
aces+0.028
amas+0.028
drops+0.028
coat+0.027
Neg logit contributions
 Enjoy-0.018
ours-0.017
 swamp-0.017
 coconut-0.016
 whose-0.016
Token-
Token ID12
Feature activation-4.72e-01

Pos logit contributions
coat+0.306
aces+0.304
 forehead+0.301
amas+0.298
robe+0.291
Neg logit contributions
ours-0.259
 Enjoy-0.238
 coconut-0.237
rust-0.221
 whose-0.212
Tokenbel
Token ID6667
Feature activation-1.12e+01

Pos logit contributions
amas+0.031
aces+0.031
 forehead+0.030
coat+0.029
robe+0.029
Neg logit contributions
ours-0.039
 coconut-0.038
 Enjoy-0.037
rust-0.036
oral-0.035
Tokenieve
Token ID12311
Feature activation-2.07e+00

Pos logit contributions
coat+0.892
robe+0.840
drops+0.839
amas+0.817
 forehead+0.805
Neg logit contributions
ours-0.814
 Enjoy-0.793
oral-0.759
 coconut-0.725
rous-0.715
Token.
Token ID13
Feature activation-2.26e-01

Pos logit contributions
aces+0.161
coat+0.161
amas+0.160
 forehead+0.157
drops+0.152
Neg logit contributions
ours-0.142
 coconut-0.138
 Enjoy-0.133
oral-0.126
rust-0.123
Token Today
Token ID6288
Feature activation+1.50e+00

Pos logit contributions
aces+0.020
amas+0.019
coat+0.019
robe+0.019
stand+0.019
Neg logit contributions
 coconut-0.014
ours-0.014
 Enjoy-0.013
oral-0.012
rust-0.011
Token she
Token ID673
Feature activation-1.61e+00

Pos logit contributions
ours+0.089
 coconut+0.086
 Enjoy+0.084
oral+0.078
rust+0.074
Neg logit contributions
aces-0.134
amas-0.131
coat-0.130
 forehead-0.126
robe-0.125
Token was
Token ID373
Feature activation+3.28e-01

Pos logit contributions
aces+0.141
coat+0.138
amas+0.136
robe+0.133
drops+0.131
Neg logit contributions
ours-0.100
 coconut-0.098
 Enjoy-0.095
oral-0.089
rust-0.084
Token they
Token ID484
Feature activation+5.25e-01

Pos logit contributions
coat+0.001
aces+0.001
amas+0.001
robe+0.001
stand+0.001
Neg logit contributions
ours-0.001
 coconut-0.001
 Enjoy-0.001
oral-0.001
rust-0.001
Token decided
Token ID3066
Feature activation+3.19e+00

Pos logit contributions
ours+0.041
 coconut+0.040
 Enjoy+0.039
oral+0.038
iw+0.036
Neg logit contributions
coat-0.035
aces-0.035
robe-0.035
 cartridge-0.033
amas-0.032
Token to
Token ID284
Feature activation+6.12e-01

Pos logit contributions
ours+0.193
 coconut+0.185
 Enjoy+0.183
oral+0.171
rust+0.164
Neg logit contributions
aces-0.280
coat-0.278
amas-0.273
 forehead-0.265
drops-0.265
Token be
Token ID307
Feature activation-1.20e+00

Pos logit contributions
ours+0.045
 coconut+0.044
 Enjoy+0.042
oral+0.041
iw+0.040
Neg logit contributions
aces-0.045
coat-0.045
amas-0.044
robe-0.043
stand-0.043
Token mail
Token ID6920
Feature activation-3.23e-01

Pos logit contributions
amas+0.115
aces+0.110
 forehead+0.106
coat+0.104
stand+0.103
Neg logit contributions
 Enjoy-0.069
ours-0.066
 coconut-0.061
grown-0.060
rous-0.058
Token carriers
Token ID16651
Feature activation-1.12e+01

Pos logit contributions
aces+0.029
amas+0.029
 forehead+0.029
coat+0.028
drops+0.028
Neg logit contributions
ours-0.018
 Enjoy-0.018
 coconut-0.017
 whose-0.016
 Sometimes-0.015
Token and
Token ID290
Feature activation+5.25e-02

Pos logit contributions
aces+0.982
amas+0.976
coat+0.962
 forehead+0.944
drops+0.941
Neg logit contributions
ours-0.669
 Enjoy-0.652
 coconut-0.638
oral-0.585
rust-0.568
Token deliver
Token ID5203
Feature activation-1.08e+00

Pos logit contributions
ours+0.004
 coconut+0.004
 Enjoy+0.004
oral+0.004
rust+0.004
Neg logit contributions
aces-0.004
amas-0.004
 forehead-0.004
coat-0.004
stand-0.003
Token letters
Token ID7475
Feature activation-1.80e+00

Pos logit contributions
aces+0.075
amas+0.072
coat+0.070
 forehead+0.068
robe+0.068
Neg logit contributions
ours-0.085
 coconut-0.082
 Enjoy-0.080
oral-0.078
rust-0.076
Token to
Token ID284
Feature activation+6.24e-01

Pos logit contributions
amas+0.195
coat+0.190
 forehead+0.189
aces+0.183
drops+0.183
Neg logit contributions
ours-0.069
 Enjoy-0.065
 coconut-0.065
 whose-0.056
 Sometimes-0.051
Token their
Token ID511
Feature activation+1.12e+00

Pos logit contributions
ours+0.039
 coconut+0.037
 Enjoy+0.036
oral+0.034
rust+0.033
Neg logit contributions
aces-0.053
amas-0.052
coat-0.051
 forehead-0.050
stand-0.050
Token.
Token ID13
Feature activation+2.06e-01

Pos logit contributions
aces+0.134
amas+0.131
coat+0.130
drops+0.127
robe+0.126
Neg logit contributions
ours-0.078
 Enjoy-0.078
 coconut-0.074
oral-0.068
 asks-0.064
Token We
Token ID775
Feature activation+9.14e-03

Pos logit contributions
ours+0.012
 Enjoy+0.012
 coconut+0.011
oral+0.010
 Sometimes+0.010
Neg logit contributions
aces-0.019
coat-0.019
amas-0.019
drops-0.018
 forehead-0.018
Token love
Token ID1842
Feature activation-2.21e-01

Pos logit contributions
 Enjoy+0.000
ours+0.000
 coconut+0.000
oral+0.000
rust+0.000
Neg logit contributions
amas-0.001
aces-0.001
stand-0.001
coat-0.001
robe-0.001
Token being
Token ID852
Feature activation+1.51e-01

Pos logit contributions
amas+0.018
aces+0.018
stand+0.017
coat+0.017
 forehead+0.017
Neg logit contributions
ours-0.015
 coconut-0.014
 Enjoy-0.013
oral-0.013
rust-0.013
Token mail
Token ID6920
Feature activation-3.12e-01

Pos logit contributions
 Enjoy+0.007
 whose+0.006
ours+0.006
 shrimp+0.006
 coconut+0.006
Neg logit contributions
drops-0.016
coat-0.016
amas-0.016
aces-0.016
 forehead-0.015
Token carriers
Token ID16651
Feature activation-1.12e+01

Pos logit contributions
coat+0.030
amas+0.030
aces+0.029
 forehead+0.029
drops+0.029
Neg logit contributions
 Enjoy-0.014
ours-0.013
 coconut-0.013
oral-0.012
 Crab-0.012
Token.
Token ID13
Feature activation+2.13e-01

Pos logit contributions
amas+0.994
coat+0.985
aces+0.983
drops+0.951
 forehead+0.946
Neg logit contributions
ours-0.639
 Enjoy-0.628
 coconut-0.614
oral-0.561
rust-0.538
Token Can
Token ID1680
Feature activation+2.03e+00

Pos logit contributions
ours+0.013
 coconut+0.012
 Enjoy+0.012
oral+0.011
rust+0.010
Neg logit contributions
aces-0.019
coat-0.019
amas-0.019
robe-0.018
 forehead-0.018
Token we
Token ID356
Feature activation+2.78e-01

Pos logit contributions
ours+0.125
 coconut+0.122
 Enjoy+0.118
oral+0.109
rust+0.106
Neg logit contributions
aces-0.177
amas-0.174
coat-0.173
 forehead-0.167
robe-0.167
Token do
Token ID466
Feature activation+9.15e-01

Pos logit contributions
ours+0.018
 Enjoy+0.018
 coconut+0.018
oral+0.017
rust+0.016
Neg logit contributions
aces-0.023
amas-0.022
coat-0.022
robe-0.021
drops-0.021
Token it
Token ID340
Feature activation+1.52e+00

Pos logit contributions
 coconut+0.059
ours+0.059
 Enjoy+0.054
iw+0.053
rust+0.053
Neg logit contributions
aces-0.074
amas-0.074
 forehead-0.072
stand-0.072
coat-0.070
Token they
Token ID484
Feature activation+2.19e-01

Pos logit contributions
aces+0.043
amas+0.042
coat+0.042
robe+0.040
 forehead+0.040
Neg logit contributions
ours-0.025
 coconut-0.025
 Enjoy-0.025
oral-0.021
rust-0.021
Token always
Token ID1464
Feature activation-9.77e-01

Pos logit contributions
ours+0.014
 coconut+0.014
 Enjoy+0.014
oral+0.013
rust+0.012
Neg logit contributions
aces-0.018
amas-0.018
coat-0.018
 forehead-0.017
robe-0.017
Token threw
Token ID9617
Feature activation+2.28e-01

Pos logit contributions
aces+0.083
amas+0.082
coat+0.081
robe+0.078
drops+0.077
Neg logit contributions
ours-0.062
 coconut-0.060
 Enjoy-0.059
oral-0.054
rust-0.053
Token the
Token ID262
Feature activation+2.74e-02

Pos logit contributions
ours+0.014
 coconut+0.013
 Enjoy+0.013
oral+0.012
rust+0.012
Neg logit contributions
aces-0.020
amas-0.020
coat-0.020
robe-0.019
 forehead-0.019
Token bath
Token ID7837
Feature activation-2.15e+00

Pos logit contributions
 coconut+0.002
ours+0.002
 Enjoy+0.002
oral+0.002
rust+0.002
Neg logit contributions
aces-0.002
amas-0.002
coat-0.002
robe-0.002
drops-0.002
Tokenrobe
Token ID25481
Feature activation-1.12e+01

Pos logit contributions
amas+0.158
 forehead+0.152
aces+0.149
bridge+0.145
stand+0.139
Neg logit contributions
ours-0.147
 coconut-0.135
 crab-0.132
 shrimp-0.129
 Enjoy-0.127
Token in
Token ID287
Feature activation+1.03e+00

Pos logit contributions
amas+1.030
aces+0.996
 forehead+0.963
bes+0.951
bridge+0.935
Neg logit contributions
ours-0.570
 Enjoy-0.560
 coconut-0.551
arium-0.510
fired-0.479
Token the
Token ID262
Feature activation-2.20e-01

Pos logit contributions
ours+0.056
 coconut+0.054
 Enjoy+0.053
oral+0.048
rust+0.046
Neg logit contributions
aces-0.098
amas-0.096
coat-0.096
robe-0.092
 forehead-0.092
Token air
Token ID1633
Feature activation-3.38e+00

Pos logit contributions
aces+0.017
amas+0.016
coat+0.016
 forehead+0.015
robe+0.015
Neg logit contributions
ours-0.016
 coconut-0.016
 Enjoy-0.015
oral-0.014
rust-0.014
Token when
Token ID618
Feature activation-2.09e+00

Pos logit contributions
amas+0.323
aces+0.322
 forehead+0.307
coat+0.302
robe+0.300
Neg logit contributions
 coconut-0.164
ours-0.164
 Enjoy-0.149
arium-0.136
 crab-0.131
Token they
Token ID484
Feature activation+8.33e-02

Pos logit contributions
amas+0.179
aces+0.178
coat+0.176
robe+0.173
stand+0.173
Neg logit contributions
ours-0.127
 coconut-0.122
 Enjoy-0.119
oral-0.112
rust-0.111
Token put
Token ID1234
Feature activation-1.73e+00

Pos logit contributions
aces+0.099
amas+0.096
coat+0.094
drops+0.091
robe+0.091
Neg logit contributions
ours-0.075
 Enjoy-0.075
 coconut-0.073
oral-0.067
rust-0.064
Token on
Token ID319
Feature activation+6.78e-01

Pos logit contributions
amas+0.175
aces+0.172
drops+0.172
coat+0.169
 forehead+0.166
Neg logit contributions
ours-0.082
 Enjoy-0.079
 coconut-0.067
rous-0.065
 Sometimes-0.064
Token her
Token ID607
Feature activation-1.46e+00

Pos logit contributions
ours+0.039
 Enjoy+0.037
 coconut+0.036
rous+0.033
rust+0.032
Neg logit contributions
aces-0.061
amas-0.061
coat-0.059
drops-0.059
bridge-0.058
Token lovely
Token ID14081
Feature activation-7.17e-01

Pos logit contributions
aces+0.146
drops+0.141
bridge+0.139
amas+0.139
coat+0.138
Neg logit contributions
ours-0.071
 Enjoy-0.064
 coconut-0.061
oral-0.061
 Sometimes-0.060
Token bath
Token ID7837
Feature activation-2.15e+00

Pos logit contributions
aces+0.064
drops+0.061
bridge+0.059
amas+0.059
coat+0.056
Neg logit contributions
ours-0.043
 Enjoy-0.040
 coconut-0.040
 shrimp-0.038
oral-0.038
Tokenrobe
Token ID25481
Feature activation-1.12e+01

Pos logit contributions
bridge+0.071
aces+0.067
 forehead+0.067
drops+0.065
ring+0.061
Neg logit contributions
ours-0.223
 Enjoy-0.210
arium-0.207
ry-0.206
 Sometimes-0.206
Token and
Token ID290
Feature activation+8.17e-02

Pos logit contributions
amas+1.119
drops+1.076
 forehead+1.074
aces+1.071
bridge+1.068
Neg logit contributions
 Enjoy-0.431
arium-0.412
ours-0.406
:-0.394
 coconut-0.368
Token feel
Token ID1254
Feature activation-1.85e+00

Pos logit contributions
ours+0.005
 Enjoy+0.005
 coconut+0.005
 asks+0.004
 Sometimes+0.004
Neg logit contributions
amas-0.007
aces-0.007
coat-0.007
bridge-0.006
robe-0.006
Token so
Token ID523
Feature activation-2.52e+00

Pos logit contributions
amas+0.179
aces+0.175
coat+0.172
drops+0.164
robe+0.156
Neg logit contributions
 Enjoy-0.112
ours-0.101
 coconut-0.092
rous-0.092
 Sometimes-0.090
Token fit
Token ID4197
Feature activation-4.71e+00

Pos logit contributions
aces+0.214
coat+0.212
amas+0.212
drops+0.198
robe+0.194
Neg logit contributions
ours-0.166
 Enjoy-0.165
 coconut-0.154
oral-0.140
rous-0.139
Token and
Token ID290
Feature activation+6.99e-02

Pos logit contributions
amas+0.479
coat+0.442
aces+0.441
 forehead+0.439
drops+0.432
Neg logit contributions
ours-0.202
 Enjoy-0.196
grown-0.194
 what-0.179
fired-0.167
Token gave
Token ID2921
Feature activation-1.28e+00

Pos logit contributions
aces+0.048
amas+0.047
coat+0.047
 forehead+0.045
robe+0.045
Neg logit contributions
ours-0.027
 coconut-0.025
 Enjoy-0.025
oral-0.023
rust-0.022
Token him
Token ID683
Feature activation-1.76e+00

Pos logit contributions
aces+0.112
coat+0.109
amas+0.109
robe+0.105
 forehead+0.104
Neg logit contributions
ours-0.079
 coconut-0.076
 Enjoy-0.074
oral-0.071
rust-0.066
Token an
Token ID281
Feature activation+1.05e+00

Pos logit contributions
aces+0.129
coat+0.124
robe+0.120
amas+0.119
 forehead+0.117
Neg logit contributions
ours-0.134
 coconut-0.129
 Enjoy-0.125
oral-0.124
rust-0.117
Token ancient
Token ID6156
Feature activation+4.04e-01

Pos logit contributions
ours+0.080
 coconut+0.078
 Enjoy+0.077
oral+0.072
rust+0.070
Neg logit contributions
aces-0.076
amas-0.075
coat-0.074
drops-0.072
robe-0.070
Token bath
Token ID7837
Feature activation-2.12e+00

Pos logit contributions
ours+0.034
 coconut+0.033
 Enjoy+0.033
oral+0.031
rust+0.030
Neg logit contributions
aces-0.026
amas-0.025
coat-0.025
robe-0.024
 forehead-0.024
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
aces+0.068
amas+0.066
 forehead+0.064
bridge+0.059
drops+0.057
Neg logit contributions
ours-0.231
 Enjoy-0.221
 arose-0.217
oral-0.213
 coconut-0.212
Token that
Token ID326
Feature activation-9.76e-01

Pos logit contributions
amas+1.164
aces+1.125
 forehead+1.117
drops+1.116
coat+1.085
Neg logit contributions
 Enjoy-0.472
ours-0.431
 coconut-0.396
:-0.361
arium-0.352
Token belonged
Token ID19611
Feature activation-2.02e+00

Pos logit contributions
aces+0.093
amas+0.093
coat+0.091
 forehead+0.090
drops+0.087
Neg logit contributions
ours-0.051
 coconut-0.050
 Enjoy-0.048
rust-0.043
 Crab-0.041
Token to
Token ID284
Feature activation+4.88e-01

Pos logit contributions
amas+0.223
 forehead+0.221
drops+0.218
aces+0.217
bridge+0.212
Neg logit contributions
ours-0.069
 Enjoy-0.068
 coconut-0.065
grown-0.060
rust-0.054
Token his
Token ID465
Feature activation+9.61e-01

Pos logit contributions
ours+0.028
 coconut+0.027
 Enjoy+0.026
oral+0.024
rust+0.023
Neg logit contributions
aces-0.045
amas-0.044
coat-0.044
robe-0.042
 forehead-0.042
Token grand
Token ID4490
Feature activation-2.58e+00

Pos logit contributions
ours+0.071
 coconut+0.067
 Enjoy+0.067
oral+0.061
rust+0.060
Neg logit contributions
coat-0.072
robe-0.071
amas-0.069
 forehead-0.068
aces-0.067
Token she
Token ID673
Feature activation-1.54e+00

Pos logit contributions
amas+0.074
aces+0.072
 forehead+0.071
coat+0.071
robe+0.069
Neg logit contributions
ours-0.049
 Enjoy-0.045
 coconut-0.044
grown-0.042
oral-0.041
Token put
Token ID1234
Feature activation-1.65e+00

Pos logit contributions
aces+0.125
coat+0.123
amas+0.121
robe+0.117
drops+0.117
Neg logit contributions
ours-0.104
 coconut-0.102
 Enjoy-0.098
oral-0.093
rust-0.089
Token her
Token ID607
Feature activation-1.42e+00

Pos logit contributions
amas+0.149
aces+0.145
drops+0.143
 forehead+0.143
coat+0.140
Neg logit contributions
ours-0.093
 Enjoy-0.092
 coconut-0.082
rous-0.077
iw-0.077
Token new
Token ID649
Feature activation+1.76e+00

Pos logit contributions
amas+0.116
aces+0.115
bridge+0.112
drops+0.111
coat+0.109
Neg logit contributions
ours-0.094
 Enjoy-0.088
 coconut-0.087
 shrimp-0.079
 crab-0.078
Token bath
Token ID7837
Feature activation-2.09e+00

Pos logit contributions
ours+0.125
 coconut+0.119
 Enjoy+0.119
oral+0.108
rust+0.107
Neg logit contributions
aces-0.140
amas-0.136
drops-0.131
bridge-0.128
coat-0.128
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
amas+0.079
 forehead+0.077
bridge+0.068
stand+0.068
 click+0.067
Neg logit contributions
ours-0.204
 Enjoy-0.198
arium-0.191
 swamp-0.187
in-0.187
Token on
Token ID319
Feature activation+7.11e-01

Pos logit contributions
amas+1.073
drops+1.003
 forehead+0.996
aces+0.990
bes+0.965
Neg logit contributions
 Enjoy-0.517
arium-0.462
ours-0.452
:-0.440
 coconut-0.437
Token.
Token ID13
Feature activation-1.70e-01

Pos logit contributions
ours+0.045
 coconut+0.044
 Enjoy+0.043
oral+0.040
rust+0.038
Neg logit contributions
aces-0.060
amas-0.059
coat-0.058
 forehead-0.057
drops-0.056
Token<|endoftext|>
Token ID50256
Feature activation+4.85e-01

Pos logit contributions
aces+0.016
amas+0.015
robe+0.015
coat+0.015
stand+0.015
Neg logit contributions
ours-0.010
 coconut-0.010
 Enjoy-0.008
oral-0.008
rust-0.008
TokenOnce
Token ID7454
Feature activation-2.12e+00

Pos logit contributions
ours+0.027
 coconut+0.026
 Enjoy+0.023
rust+0.022
oral+0.022
Neg logit contributions
aces-0.046
amas-0.046
robe-0.046
coat-0.045
stand-0.044
Token upon
Token ID2402
Feature activation+1.70e+00

Pos logit contributions
aces+0.185
coat+0.184
robe+0.184
amas+0.179
stand+0.178
Neg logit contributions
 Enjoy-0.122
 coconut-0.120
ours-0.112
rust-0.103
iw-0.099
TokenYou
Token ID1639
Feature activation-7.17e-01

Pos logit contributions
aces+0.122
amas+0.119
coat+0.118
 forehead+0.116
robe+0.113
Neg logit contributions
 Enjoy-0.079
ours-0.075
oral-0.075
 coconut-0.072
rust-0.071
Token are
Token ID389
Feature activation-1.53e+00

Pos logit contributions
amas+0.068
coat+0.067
aces+0.066
drops+0.064
stand+0.063
Neg logit contributions
 Enjoy-0.041
 coconut-0.038
ours-0.037
 asks-0.033
oral-0.033
Token very
Token ID845
Feature activation+2.08e+00

Pos logit contributions
coat+0.153
aces+0.151
amas+0.150
drops+0.146
 forehead+0.141
Neg logit contributions
 Enjoy-0.084
ours-0.074
 coconut-0.074
 Crab-0.065
rust-0.064
Token good
Token ID922
Feature activation+1.47e+00

Pos logit contributions
ours+0.148
 Enjoy+0.144
 coconut+0.143
oral+0.130
rust+0.129
Neg logit contributions
aces-0.165
amas-0.161
coat-0.160
robe-0.152
 forehead-0.152
Token mail
Token ID6920
Feature activation-2.70e-01

Pos logit contributions
ours+0.088
 Enjoy+0.083
 coconut+0.082
oral+0.076
rust+0.072
Neg logit contributions
aces-0.132
coat-0.129
amas-0.129
robe-0.125
 forehead-0.124
Token carriers
Token ID16651
Feature activation-1.11e+01

Pos logit contributions
aces+0.028
amas+0.028
coat+0.027
 forehead+0.027
drops+0.026
Neg logit contributions
 Enjoy-0.011
ours-0.010
 coconut-0.009
 whose-0.009
 Joe-0.009
Token.
Token ID13
Feature activation+1.77e-01

Pos logit contributions
aces+0.984
amas+0.980
coat+0.964
 forehead+0.938
drops+0.936
Neg logit contributions
ours-0.655
 Enjoy-0.641
 coconut-0.633
oral-0.574
rust-0.550
Token You
Token ID921
Feature activation-4.00e-01

Pos logit contributions
ours+0.013
 coconut+0.013
 Enjoy+0.012
oral+0.011
rust+0.011
Neg logit contributions
aces-0.014
amas-0.013
coat-0.013
robe-0.013
 forehead-0.012
Token made
Token ID925
Feature activation-4.25e+00

Pos logit contributions
amas+0.039
coat+0.038
aces+0.037
stand+0.036
 forehead+0.036
Neg logit contributions
 Enjoy-0.022
 coconut-0.019
ours-0.019
 asks-0.018
oral-0.017
Token your
Token ID534
Feature activation+1.34e+00

Pos logit contributions
aces+0.396
coat+0.391
amas+0.384
robe+0.375
drops+0.374
Neg logit contributions
ours-0.241
 coconut-0.233
 Enjoy-0.230
oral-0.207
rust-0.202
Token friends
Token ID2460
Feature activation+1.12e+00

Pos logit contributions
ours+0.083
 coconut+0.081
 Enjoy+0.079
oral+0.073
rust+0.070
Neg logit contributions
aces-0.117
amas-0.115
coat-0.114
robe-0.110
drops-0.110
Token!
Token ID0
Feature activation-7.21e-02

Pos logit contributions
ours+0.021
 Enjoy+0.020
 coconut+0.019
 Joe+0.017
grown+0.016
Neg logit contributions
aces-0.043
amas-0.042
 forehead-0.042
drops-0.041
coat-0.041
Token I
Token ID314
Feature activation-6.63e-01

Pos logit contributions
aces+0.006
coat+0.006
robe+0.006
amas+0.005
stand+0.005
Neg logit contributions
 coconut-0.005
ours-0.005
oral-0.005
 Enjoy-0.005
rust-0.005
Token found
Token ID1043
Feature activation-2.39e+00

Pos logit contributions
coat+0.044
robe+0.043
aces+0.041
 cartridge+0.040
drops+0.039
Neg logit contributions
ours-0.054
oral-0.052
 coconut-0.052
 Enjoy-0.048
rust-0.045
Token a
Token ID257
Feature activation-5.36e-01

Pos logit contributions
aces+0.193
amas+0.187
coat+0.184
robe+0.178
 forehead+0.178
Neg logit contributions
ours-0.165
 coconut-0.159
 Enjoy-0.155
oral-0.150
rust-0.144
Token bath
Token ID7837
Feature activation-2.24e+00

Pos logit contributions
coat+0.037
robe+0.035
amas+0.034
 forehead+0.034
stand+0.033
Neg logit contributions
ours-0.045
 Enjoy-0.041
oral-0.040
 coconut-0.040
rust-0.039
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
aces+0.162
 forehead+0.151
amas+0.151
stand+0.145
bridge+0.144
Neg logit contributions
ours-0.166
 Enjoy-0.158
 coconut-0.157
 swamp-0.145
 crab-0.145
Token!"
Token ID2474
Feature activation-1.63e+00

Pos logit contributions
aces+1.056
amas+1.043
 forehead+1.008
bes+0.996
drops+0.979
Neg logit contributions
 Enjoy-0.594
ours-0.556
 coconut-0.528
iw-0.467
grown-0.459
Token Lily
Token ID20037
Feature activation+4.50e-01

Pos logit contributions
aces+0.171
coat+0.166
amas+0.165
drops+0.162
 forehead+0.160
Neg logit contributions
ours-0.070
 Enjoy-0.067
 coconut-0.065
oral-0.056
 Sometimes-0.056
Token looked
Token ID3114
Feature activation+1.72e+00

Pos logit contributions
ours+0.028
 coconut+0.027
 Enjoy+0.027
oral+0.025
rust+0.024
Neg logit contributions
aces-0.039
amas-0.038
coat-0.038
 forehead-0.037
robe-0.036
Token at
Token ID379
Feature activation+2.76e+00

Pos logit contributions
 coconut+0.129
oral+0.128
ours+0.128
 Enjoy+0.128
rust+0.114
Neg logit contributions
aces-0.122
coat-0.121
 forehead-0.118
amas-0.117
drops-0.114
Token the
Token ID262
Feature activation-7.15e-02

Pos logit contributions
ours+0.158
 coconut+0.153
 Enjoy+0.149
oral+0.136
rust+0.133
Neg logit contributions
aces-0.254
amas-0.249
coat-0.248
robe-0.239
 forehead-0.239
Token put
Token ID1234
Feature activation-1.59e+00

Pos logit contributions
ours+0.029
 coconut+0.028
 Enjoy+0.027
oral+0.025
rust+0.025
Neg logit contributions
aces-0.039
coat-0.038
amas-0.038
robe-0.037
 forehead-0.037
Token on
Token ID319
Feature activation+7.82e-01

Pos logit contributions
amas+0.157
aces+0.155
coat+0.153
drops+0.152
 forehead+0.151
Neg logit contributions
ours-0.079
 Enjoy-0.075
 coconut-0.072
iw-0.063
rous-0.063
Token her
Token ID607
Feature activation-1.36e+00

Pos logit contributions
ours+0.048
 coconut+0.046
 Enjoy+0.045
oral+0.042
rust+0.041
Neg logit contributions
aces-0.068
amas-0.067
coat-0.067
robe-0.064
 forehead-0.064
Token cozy
Token ID37438
Feature activation-3.11e+00

Pos logit contributions
aces+0.106
drops+0.102
amas+0.100
coat+0.099
bridge+0.098
Neg logit contributions
ours-0.091
 coconut-0.088
 Enjoy-0.086
 crab-0.082
 shrimp-0.081
Token bath
Token ID7837
Feature activation-2.10e+00

Pos logit contributions
aces+0.210
drops+0.201
bridge+0.191
amas+0.183
coat+0.178
Neg logit contributions
 coconut-0.247
ours-0.239
 swamp-0.231
 Enjoy-0.230
 crab-0.225
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
drops+0.067
 click+0.067
bridge+0.066
hy+0.066
 forehead+0.065
Neg logit contributions
ours-0.212
arium-0.203
ry-0.197
 Enjoy-0.196
 Sometimes-0.195
Token and
Token ID290
Feature activation-7.70e-03

Pos logit contributions
drops+1.090
 forehead+1.081
aces+1.080
amas+1.079
bridge+1.063
Neg logit contributions
arium-0.422
ours-0.417
 Enjoy-0.390
:-0.366
 Sometimes-0.363
Token went
Token ID1816
Feature activation-1.86e-01

Pos logit contributions
aces+0.001
amas+0.001
coat+0.001
bridge+0.001
drops+0.001
Neg logit contributions
ours-0.000
 Enjoy-0.000
 coconut-0.000
 asks-0.000
rust-0.000
Token to
Token ID284
Feature activation+5.05e-01

Pos logit contributions
aces+0.019
amas+0.018
coat+0.018
drops+0.017
 forehead+0.017
Neg logit contributions
ours-0.009
 Enjoy-0.009
 coconut-0.008
oral-0.007
 Sometimes-0.007
Token bed
Token ID3996
Feature activation-2.92e+00

Pos logit contributions
 coconut+0.051
ours+0.051
oral+0.049
rust+0.047
iw+0.047
Neg logit contributions
stand-0.024
 forehead-0.024
amas-0.023
robe-0.022
coat-0.022
Token with
Token ID351
Feature activation-1.15e+00

Pos logit contributions
aces+0.294
coat+0.276
drops+0.274
bridge+0.272
 pads+0.262
Neg logit contributions
ours-0.141
 Enjoy-0.124
arium-0.117
oral-0.115
 whose-0.110
Token she
Token ID673
Feature activation-1.51e+00

Pos logit contributions
aces+0.128
amas+0.124
coat+0.124
robe+0.120
 forehead+0.119
Neg logit contributions
ours-0.067
 coconut-0.064
 Enjoy-0.064
oral-0.056
grown-0.053
Token had
Token ID550
Feature activation-5.28e-01

Pos logit contributions
aces+0.156
amas+0.154
coat+0.147
bridge+0.141
robe+0.141
Neg logit contributions
 Enjoy-0.062
 asks-0.055
ours-0.053
 crab-0.050
lly-0.050
Token a
Token ID257
Feature activation-5.85e-01

Pos logit contributions
amas+0.053
drops+0.053
 forehead+0.051
coat+0.050
aces+0.050
Neg logit contributions
 Enjoy-0.026
 coconut-0.023
grown-0.023
rous-0.022
ours-0.021
Token charming
Token ID23332
Feature activation-1.44e+00

Pos logit contributions
coat+0.044
aces+0.043
amas+0.043
 forehead+0.042
robe+0.042
Neg logit contributions
ours-0.043
 coconut-0.040
 Enjoy-0.039
oral-0.037
rust-0.036
Token bath
Token ID7837
Feature activation-2.12e+00

Pos logit contributions
aces+0.120
amas+0.117
coat+0.117
robe+0.112
 forehead+0.112
Neg logit contributions
ours-0.095
 coconut-0.092
 Enjoy-0.090
oral-0.084
rust-0.081
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
amas+0.190
aces+0.187
 forehead+0.180
stand+0.174
bridge+0.174
Neg logit contributions
 Enjoy-0.106
ours-0.101
 coconut-0.096
arium-0.096
 crab-0.093
Token to
Token ID284
Feature activation+3.63e-01

Pos logit contributions
amas+1.229
aces+1.188
bes+1.160
 forehead+1.131
drops+1.094
Neg logit contributions
 Enjoy-0.419
:-0.337
arium-0.316
 coconut-0.315
grown-0.314
Token wear
Token ID5806
Feature activation-1.04e+00

Pos logit contributions
 Enjoy+0.022
ours+0.021
 coconut+0.021
oral+0.019
rust+0.018
Neg logit contributions
aces-0.032
amas-0.032
coat-0.031
drops-0.030
robe-0.029
Token,
Token ID11
Feature activation-1.72e-01

Pos logit contributions
amas+0.104
drops+0.101
aces+0.101
 forehead+0.100
coat+0.097
Neg logit contributions
 Enjoy-0.042
 what-0.038
grown-0.037
:-0.036
ours-0.036
Token and
Token ID290
Feature activation-5.25e-03

Pos logit contributions
aces+0.016
amas+0.015
coat+0.014
robe+0.014
drops+0.014
Neg logit contributions
 Enjoy-0.010
ours-0.009
 coconut-0.009
 whose-0.009
 Sometimes-0.008
Token she
Token ID673
Feature activation-1.52e+00

Pos logit contributions
aces+0.000
amas+0.000
coat+0.000
robe+0.000
 forehead+0.000
Neg logit contributions
ours-0.000
 Enjoy-0.000
 coconut-0.000
oral-0.000
rust-0.000
Token He
Token ID679
Feature activation-2.20e-01

Pos logit contributions
coat+0.009
amas+0.009
aces+0.009
robe+0.009
drops+0.009
Neg logit contributions
ours-0.008
 coconut-0.008
 Enjoy-0.008
oral-0.008
rust-0.007
Token wore
Token ID12408
Feature activation-9.15e-01

Pos logit contributions
coat+0.018
aces+0.018
robe+0.017
amas+0.017
drops+0.016
Neg logit contributions
ours-0.015
 Enjoy-0.014
 coconut-0.014
oral-0.014
iw-0.012
Token his
Token ID465
Feature activation+8.95e-01

Pos logit contributions
amas+0.095
 forehead+0.091
coat+0.090
drops+0.089
aces+0.089
Neg logit contributions
 coconut-0.044
ours-0.042
 Enjoy-0.040
rust-0.037
 Sometimes-0.034
Token big
Token ID1263
Feature activation-1.05e+00

Pos logit contributions
 coconut+0.060
ours+0.056
 crab+0.052
rust+0.050
 Enjoy+0.050
Neg logit contributions
amas-0.075
drops-0.070
stand-0.070
coat-0.069
 forehead-0.068
Token bath
Token ID7837
Feature activation-2.07e+00

Pos logit contributions
aces+0.078
amas+0.077
coat+0.076
drops+0.074
 forehead+0.073
Neg logit contributions
ours-0.077
 coconut-0.076
 Enjoy-0.072
oral-0.068
rust-0.067
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
 forehead+0.083
amas+0.074
stand+0.068
drops+0.066
 click+0.064
Neg logit contributions
ours-0.209
 coconut-0.208
 swamp-0.207
arium-0.202
 crab-0.199
Token because
Token ID780
Feature activation+7.55e-01

Pos logit contributions
drops+1.101
amas+1.101
aces+1.066
bridge+1.060
 forehead+1.058
Neg logit contributions
 Enjoy-0.468
 coconut-0.452
arium-0.416
ours-0.411
:-0.400
Token it
Token ID340
Feature activation+1.25e+00

Pos logit contributions
ours+0.044
 coconut+0.042
 Enjoy+0.042
oral+0.039
rust+0.037
Neg logit contributions
aces-0.068
coat-0.065
robe-0.065
amas-0.064
bridge-0.063
Token was
Token ID373
Feature activation+4.72e-01

Pos logit contributions
 coconut+0.052
ours+0.051
rust+0.047
 swamp+0.043
 Enjoy+0.042
Neg logit contributions
amas-0.132
 forehead-0.130
aces-0.127
coat-0.124
stand-0.124
Token cold
Token ID4692
Feature activation+5.53e-01

Pos logit contributions
 coconut+0.031
ours+0.030
 Enjoy+0.029
rust+0.027
grown+0.026
Neg logit contributions
amas-0.040
aces-0.039
coat-0.038
 forehead-0.037
drops-0.037
Token.
Token ID13
Feature activation-1.27e-01

Pos logit contributions
 coconut+0.022
 Enjoy+0.022
ours+0.022
 Sometimes+0.022
grown+0.020
Neg logit contributions
 forehead-0.055
aces-0.054
bes-0.053
amas-0.052
bridge-0.052
Token put
Token ID1234
Feature activation-1.68e+00

Pos logit contributions
aces+0.016
coat+0.016
amas+0.016
 forehead+0.015
robe+0.015
Neg logit contributions
ours-0.011
 coconut-0.010
 Enjoy-0.010
oral-0.009
rust-0.009
Token on
Token ID319
Feature activation+6.94e-01

Pos logit contributions
amas+0.169
aces+0.166
drops+0.165
coat+0.163
bridge+0.159
Neg logit contributions
ours-0.083
 Enjoy-0.076
 coconut-0.067
iw-0.063
 Sometimes-0.063
Token the
Token ID262
Feature activation-4.34e-02

Pos logit contributions
ours+0.036
 Enjoy+0.034
 coconut+0.034
 Sometimes+0.029
iw+0.028
Neg logit contributions
aces-0.067
amas-0.065
coat-0.064
bridge-0.064
drops-0.064
Token new
Token ID649
Feature activation+1.71e+00

Pos logit contributions
aces+0.004
drops+0.004
amas+0.004
bridge+0.004
coat+0.004
Neg logit contributions
ours-0.002
 shrimp-0.002
 Sometimes-0.002
 Enjoy-0.002
oral-0.002
Token bath
Token ID7837
Feature activation-2.14e+00

Pos logit contributions
ours+0.115
 Enjoy+0.107
 coconut+0.106
 shrimp+0.102
 Sometimes+0.101
Neg logit contributions
aces-0.141
drops-0.132
amas-0.130
bridge-0.129
coat-0.122
Tokenrobe
Token ID25481
Feature activation-1.11e+01

Pos logit contributions
aces+0.071
bridge+0.070
 forehead+0.070
drops+0.064
ring+0.060
Neg logit contributions
ours-0.230
 Enjoy-0.212
ry-0.209
 Sometimes-0.208
arium-0.208
Token and
Token ID290
Feature activation+5.83e-02

Pos logit contributions
amas+1.064
aces+1.035
drops+1.035
 forehead+1.028
coat+1.014
Neg logit contributions
ours-0.534
 Enjoy-0.512
 coconut-0.480
arium-0.477
 Sometimes-0.434
Token it
Token ID340
Feature activation+1.16e+00

Pos logit contributions
ours+0.003
 coconut+0.003
 Enjoy+0.003
 asks+0.003
rust+0.003
Neg logit contributions
aces-0.005
amas-0.005
coat-0.005
 forehead-0.005
bridge-0.005
Token was
Token ID373
Feature activation+5.18e-01

Pos logit contributions
ours+0.039
 arose+0.032
 asks+0.030
arium+0.029
lly+0.029
Neg logit contributions
amas-0.127
aces-0.126
 forehead-0.122
bridge-0.116
coat-0.115
Token a
Token ID257
Feature activation-5.48e-01

Pos logit contributions
ours+0.031
 coconut+0.029
 Enjoy+0.029
 what+0.026
 Sometimes+0.026
Neg logit contributions
aces-0.045
amas-0.045
drops-0.043
 forehead-0.042
coat-0.042
Token perfect
Token ID2818
Feature activation-3.13e+00

Pos logit contributions
aces+0.043
amas+0.042
coat+0.041
robe+0.039
drops+0.039
Neg logit contributions
ours-0.039
 coconut-0.038
 Enjoy-0.037
oral-0.035
rust-0.034
Token still
Token ID991
Feature activation+9.06e-01

Pos logit contributions
aces+0.110
coat+0.107
drops+0.100
robe+0.100
bridge+0.099
Neg logit contributions
ours-0.124
 Enjoy-0.119
 coconut-0.119
oral-0.108
iw-0.108
Token did
Token ID750
Feature activation+4.61e-01

Pos logit contributions
ours+0.062
 coconut+0.061
 Enjoy+0.059
oral+0.058
rust+0.054
Neg logit contributions
aces-0.074
coat-0.072
 forehead-0.069
drops-0.069
robe-0.068
Token not
Token ID407
Feature activation-1.01e+00

Pos logit contributions
ours+0.035
 coconut+0.034
 Enjoy+0.032
rust+0.031
oral+0.031
Neg logit contributions
aces-0.034
amas-0.032
stand-0.032
coat-0.032
 forehead-0.031
Token want
Token ID765
Feature activation-4.38e-01

Pos logit contributions
aces+0.084
coat+0.082
amas+0.080
drops+0.078
 forehead+0.078
Neg logit contributions
ours-0.068
 coconut-0.065
 Enjoy-0.063
oral-0.059
rust-0.057
Token to
Token ID284
Feature activation+5.41e-01

Pos logit contributions
aces+0.033
stand+0.031
coat+0.029
drops+0.029
robe+0.029
Neg logit contributions
ours-0.033
oral-0.031
 coconut-0.030
 Enjoy-0.030
rust-0.029
Token touch
Token ID3638
Feature activation+5.00e+00

Pos logit contributions
ours+0.043
 coconut+0.040
 Enjoy+0.039
oral+0.038
iw+0.037
Neg logit contributions
aces-0.038
coat-0.038
stand-0.037
drops-0.037
amas-0.037
Token the
Token ID262
Feature activation+4.61e-02

Pos logit contributions
ours+0.329
 Enjoy+0.320
 coconut+0.314
oral+0.310
rust+0.285
Neg logit contributions
aces-0.425
amas-0.395
coat-0.393
robe-0.385
drops-0.384
Token snakes
Token ID26042
Feature activation+2.33e+00

Pos logit contributions
ours+0.003
 Enjoy+0.003
 coconut+0.003
rust+0.003
oral+0.003
Neg logit contributions
aces-0.004
drops-0.004
amas-0.004
coat-0.004
robe-0.004
Token,
Token ID11
Feature activation+1.15e-02

Pos logit contributions
ours+0.135
 coconut+0.133
 Enjoy+0.127
oral+0.116
rust+0.113
Neg logit contributions
amas-0.208
aces-0.207
coat-0.205
 forehead-0.198
robe-0.197
Token but
Token ID475
Feature activation-2.10e+00

Pos logit contributions
ours+0.001
 coconut+0.001
 Enjoy+0.001
oral+0.001
rust+0.001
Neg logit contributions
aces-0.001
coat-0.001
amas-0.001
drops-0.001
 forehead-0.001
Token he
Token ID339
Feature activation-3.18e-01

Pos logit contributions
aces+0.151
robe+0.140
 forehead+0.138
coat+0.135
drops+0.134
Neg logit contributions
ours-0.161
oral-0.155
 Enjoy-0.153
 coconut-0.148
iw-0.141
Token of
Token ID286
Feature activation+5.90e-02

Pos logit contributions
aces+0.134
amas+0.131
coat+0.131
robe+0.126
 forehead+0.125
Neg logit contributions
ours-0.095
 coconut-0.092
 Enjoy-0.090
oral-0.083
rust-0.081
Token the
Token ID262
Feature activation-4.89e-02

Pos logit contributions
ours+0.004
 Enjoy+0.004
oral+0.004
 coconut+0.004
rust+0.004
Neg logit contributions
coat-0.004
aces-0.004
stand-0.004
amas-0.004
 forehead-0.004
Token dark
Token ID3223
Feature activation+1.54e+00

Pos logit contributions
amas+0.003
 forehead+0.003
coat+0.003
aces+0.003
robe+0.003
Neg logit contributions
ours-0.004
 Enjoy-0.003
 coconut-0.003
iw-0.003
oral-0.003
Token forest
Token ID8222
Feature activation+2.51e+00

Pos logit contributions
ours+0.097
 Enjoy+0.092
 coconut+0.090
oral+0.083
rust+0.080
Neg logit contributions
aces-0.136
coat-0.132
amas-0.132
robe-0.128
drops-0.128
Token.
Token ID13
Feature activation+1.33e-01

Pos logit contributions
 coconut+0.114
ours+0.108
 Enjoy+0.101
grown+0.097
arium+0.097
Neg logit contributions
amas-0.252
aces-0.247
coat-0.241
 forehead-0.238
robe-0.236
Token Jane
Token ID12091
Feature activation+4.97e+00

Pos logit contributions
ours+0.009
 coconut+0.008
rust+0.008
oral+0.008
 crab+0.007
Neg logit contributions
aces-0.011
amas-0.011
coat-0.010
 forehead-0.010
drops-0.010
Token was
Token ID373
Feature activation+5.11e-01

Pos logit contributions
 coconut+0.253
ours+0.251
 Enjoy+0.233
oral+0.218
rust+0.212
Neg logit contributions
amas-0.475
aces-0.475
coat-0.465
robe-0.455
 forehead-0.445
Token never
Token ID1239
Feature activation-9.55e-01

Pos logit contributions
 coconut+0.031
 Enjoy+0.031
ours+0.030
grown+0.029
 what+0.028
Neg logit contributions
amas-0.044
aces-0.043
robe-0.041
coat-0.041
drops-0.039
Token able
Token ID1498
Feature activation-4.12e+00

Pos logit contributions
amas+0.091
aces+0.087
robe+0.085
coat+0.084
drops+0.082
Neg logit contributions
 coconut-0.051
grown-0.049
ours-0.048
 Enjoy-0.047
fired-0.045
Token to
Token ID284
Feature activation+4.41e-01

Pos logit contributions
aces+0.355
coat+0.346
amas+0.337
 forehead+0.332
drops+0.330
Neg logit contributions
ours-0.263
 coconut-0.244
 Enjoy-0.243
oral-0.234
rust-0.221
Token go
Token ID467
Feature activation-1.32e+00

Pos logit contributions
ours+0.033
 coconut+0.031
 Enjoy+0.031
oral+0.029
rust+0.028
Neg logit contributions
aces-0.033
coat-0.033
amas-0.032
drops-0.031
 forehead-0.031
Tokenly
Token ID306
Feature activation-3.35e+00

Pos logit contributions
aces+0.277
amas+0.273
bridge+0.269
coat+0.265
 cartridge+0.263
Neg logit contributions
 coconut-0.214
 Joe-0.209
ours-0.204
 Enjoy-0.200
 oy-0.199
Token at
Token ID379
Feature activation+2.75e+00

Pos logit contributions
amas+0.298
coat+0.294
aces+0.294
robe+0.290
bridge+0.277
Neg logit contributions
ours-0.193
 coconut-0.192
 Enjoy-0.180
grown-0.164
oral-0.159
Token Bob
Token ID5811
Feature activation+3.56e+00

Pos logit contributions
ours+0.193
oral+0.175
 Enjoy+0.170
 coconut+0.170
iw+0.162
Neg logit contributions
aces-0.221
 forehead-0.214
amas-0.204
coat-0.201
stand-0.197
Token and
Token ID290
Feature activation+2.02e-01

Pos logit contributions
ours+0.208
 coconut+0.206
 Enjoy+0.199
oral+0.172
rust+0.171
Neg logit contributions
aces-0.310
coat-0.307
amas-0.307
robe-0.300
bridge-0.297
Token said
Token ID531
Feature activation-6.61e-01

Pos logit contributions
ours+0.014
 Enjoy+0.013
 coconut+0.013
oral+0.013
rust+0.012
Neg logit contributions
aces-0.016
coat-0.016
 forehead-0.016
amas-0.015
drops-0.015
Token:
Token ID25
Feature activation+5.15e+00

Pos logit contributions
aces+0.056
coat+0.054
amas+0.054
 forehead+0.052
robe+0.052
Neg logit contributions
ours-0.043
 coconut-0.041
 Enjoy-0.040
oral-0.038
rust-0.037
Token
Token ID198
Feature activation-2.94e-02

Pos logit contributions
ours+0.434
 coconut+0.412
rous+0.389
rust+0.381
oral+0.378
Neg logit contributions
stand-0.328
amas-0.327
aces-0.313
robe-0.306
coat-0.304
Token
Token ID198
Feature activation-3.87e-02

Pos logit contributions
stand+0.002
amas+0.002
aces+0.002
coat+0.002
bridge+0.002
Neg logit contributions
ours-0.002
 coconut-0.002
rust-0.002
rous-0.002
 Enjoy-0.002
Token"
Token ID1
Feature activation-1.86e+00

Pos logit contributions
aces+0.002
coat+0.002
amas+0.002
stand+0.002
drops+0.002
Neg logit contributions
ours-0.003
 coconut-0.003
 Enjoy-0.003
rust-0.003
arium-0.003
TokenDon
Token ID3987
Feature activation-2.59e+00

Pos logit contributions
aces+0.150
coat+0.147
amas+0.146
robe+0.140
 forehead+0.140
Neg logit contributions
ours-0.129
 coconut-0.123
 Enjoy-0.120
oral-0.112
rust-0.108
Token't
Token ID470
Feature activation-2.01e+00

Pos logit contributions
aces+0.179
coat+0.174
amas+0.168
robe+0.164
 forehead+0.162
Neg logit contributions
ours-0.207
 Enjoy-0.206
 coconut-0.204
oral-0.197
 Sometimes-0.183
Token who
Token ID508
Feature activation+3.73e+00

Pos logit contributions
aces+0.008
coat+0.008
stand+0.008
amas+0.008
 forehead+0.008
Neg logit contributions
ours-0.009
 coconut-0.009
oral-0.009
 Enjoy-0.009
rust-0.008
Token was
Token ID373
Feature activation+5.55e-01

Pos logit contributions
ours+0.279
 coconut+0.261
 Enjoy+0.259
oral+0.249
rust+0.227
Neg logit contributions
coat-0.276
aces-0.276
robe-0.269
amas-0.265
stand-0.263
Token asking
Token ID4737
Feature activation-4.49e-01

Pos logit contributions
ours+0.042
 coconut+0.040
 Enjoy+0.040
oral+0.039
rust+0.036
Neg logit contributions
aces-0.041
coat-0.040
amas-0.039
 forehead-0.038
robe-0.038
Token him
Token ID683
Feature activation-1.72e+00

Pos logit contributions
aces+0.038
amas+0.037
coat+0.037
 forehead+0.036
robe+0.036
Neg logit contributions
ours-0.029
 coconut-0.028
 Enjoy-0.027
oral-0.025
rust-0.025
Token the
Token ID262
Feature activation-8.71e-02

Pos logit contributions
aces+0.144
robe+0.136
coat+0.136
 forehead+0.133
amas+0.132
Neg logit contributions
ours-0.111
 coconut-0.107
 Enjoy-0.100
oral-0.100
rust-0.098
Token question
Token ID1808
Feature activation+6.10e+00

Pos logit contributions
coat+0.007
aces+0.006
robe+0.006
 forehead+0.006
drops+0.006
Neg logit contributions
ours-0.007
 coconut-0.006
 Enjoy-0.006
rous-0.006
rust-0.005
Token.
Token ID13
Feature activation-1.16e-01

Pos logit contributions
ours+0.379
 coconut+0.367
 Enjoy+0.359
oral+0.332
rust+0.322
Neg logit contributions
aces-0.526
amas-0.514
coat-0.513
robe-0.494
 forehead-0.493
Token
Token ID198
Feature activation-8.98e-02

Pos logit contributions
aces+0.009
amas+0.009
coat+0.008
stand+0.008
robe+0.008
Neg logit contributions
ours-0.009
 coconut-0.009
oral-0.008
 Enjoy-0.008
rust-0.008
Token
Token ID198
Feature activation-9.54e-02

Pos logit contributions
aces+0.008
coat+0.008
amas+0.008
stand+0.007
robe+0.007
Neg logit contributions
ours-0.006
 coconut-0.005
 Enjoy-0.005
oral-0.005
rust-0.005
TokenThe
Token ID464
Feature activation-1.10e+00

Pos logit contributions
aces+0.008
amas+0.008
coat+0.008
 forehead+0.008
robe+0.008
Neg logit contributions
 Enjoy-0.006
 coconut-0.006
ours-0.006
oral-0.005
rust-0.005
Token le
Token ID443
Feature activation+5.21e+00

Pos logit contributions
coat+0.080
drops+0.077
 forehead+0.076
robe+0.075
aces+0.074
Neg logit contributions
ours-0.086
 Enjoy-0.077
 coconut-0.075
arium-0.073
rous-0.073
Token€
Token ID26391
Feature activation-1.96e+00

Pos logit contributions
aces+0.060
amas+0.059
coat+0.057
robe+0.055
 forehead+0.055
Neg logit contributions
 Enjoy-0.075
ours-0.070
 coconut-0.069
oral-0.068
rust-0.066
Tokenâ„¢
Token ID8151
Feature activation-5.52e+00

Pos logit contributions
 forehead+0.176
aces+0.174
coat+0.165
robe+0.165
drops+0.164
Neg logit contributions
ours-0.116
 Enjoy-0.113
 coconut-0.105
iw-0.095
oral-0.095
Tokenre
Token ID260
Feature activation-6.83e-02

Pos logit contributions
aces+0.528
coat+0.518
amas+0.516
robe+0.499
 forehead+0.495
Neg logit contributions
ours-0.293
 coconut-0.282
 Enjoy-0.274
oral-0.249
rust-0.241
Token the
Token ID262
Feature activation-4.05e-02

Pos logit contributions
coat+0.007
amas+0.007
aces+0.006
drops+0.006
stand+0.006
Neg logit contributions
 Enjoy-0.004
 coconut-0.003
ours-0.003
 Joe-0.003
 what-0.003
Token best
Token ID1266
Feature activation-1.35e+00

Pos logit contributions
amas+0.003
aces+0.003
coat+0.003
 forehead+0.003
drops+0.003
Neg logit contributions
 coconut-0.003
 Enjoy-0.003
ours-0.003
oral-0.003
rust-0.003
Token vendor
Token ID18371
Feature activation+5.13e+00

Pos logit contributions
aces+0.106
coat+0.105
amas+0.104
robe+0.102
 forehead+0.101
Neg logit contributions
ours-0.094
 coconut-0.092
 Enjoy-0.088
oral-0.084
rust-0.084
Tokenâ
Token ID22940
Feature activation-9.13e-01

Pos logit contributions
ours+0.301
 Enjoy+0.298
 coconut+0.297
iw+0.260
oral+0.255
Neg logit contributions
amas-0.457
aces-0.444
coat-0.441
 forehead-0.426
drops-0.422
Token€
Token ID26391
Feature activation-1.94e+00

Pos logit contributions
aces+0.062
amas+0.061
coat+0.058
 forehead+0.058
bridge+0.057
Neg logit contributions
 Enjoy-0.073
ours-0.069
 coconut-0.066
oral-0.066
rust-0.064
Token,
Token ID11
Feature activation-1.47e-01

Pos logit contributions
 forehead+0.171
aces+0.171
coat+0.162
amas+0.162
drops+0.162
Neg logit contributions
ours-0.114
 Enjoy-0.113
 coconut-0.106
oral-0.096
grown-0.095
Token said
Token ID531
Feature activation-6.95e-01

Pos logit contributions
aces+0.012
amas+0.012
coat+0.012
robe+0.011
 forehead+0.011
Neg logit contributions
ours-0.010
 coconut-0.009
 Enjoy-0.009
oral-0.008
rust-0.008
Token Midnight
Token ID27960
Feature activation-5.25e+00

Pos logit contributions
aces+0.061
amas+0.060
coat+0.060
drops+0.057
robe+0.056
Neg logit contributions
 Enjoy-0.043
ours-0.042
 coconut-0.039
oral-0.037
 Sometimes-0.035
Token and
Token ID290
Feature activation-1.15e-01

Pos logit contributions
ours+0.020
 coconut+0.020
 Enjoy+0.019
 Crab+0.019
rust+0.018
Neg logit contributions
coat-0.047
aces-0.046
amas-0.045
drops-0.044
bes-0.044
Token said
Token ID531
Feature activation-8.52e-01

Pos logit contributions
aces+0.010
amas+0.010
coat+0.010
robe+0.009
 forehead+0.009
Neg logit contributions
ours-0.007
 coconut-0.007
 Enjoy-0.007
oral-0.006
rust-0.006
Token,
Token ID11
Feature activation-2.24e-01

Pos logit contributions
aces+0.076
amas+0.075
coat+0.075
robe+0.072
 forehead+0.072
Neg logit contributions
ours-0.051
 coconut-0.049
 Enjoy-0.048
oral-0.044
rust-0.042
Token "
Token ID366
Feature activation-1.63e+00

Pos logit contributions
aces+0.020
coat+0.020
amas+0.019
robe+0.019
drops+0.018
Neg logit contributions
ours-0.013
 Enjoy-0.013
 coconut-0.013
oral-0.012
rust-0.011
TokenThat
Token ID2504
Feature activation+1.59e+00

Pos logit contributions
aces+0.145
amas+0.142
coat+0.141
 forehead+0.139
robe+0.136
Neg logit contributions
 Enjoy-0.099
 coconut-0.097
ours-0.095
oral-0.089
rust-0.088
Token looks
Token ID3073
Feature activation+2.20e+00

Pos logit contributions
ours+0.124
 coconut+0.120
 Enjoy+0.119
oral+0.112
rust+0.109
Neg logit contributions
aces-0.113
amas-0.110
coat-0.109
robe-0.105
 forehead-0.104
Token very
Token ID845
Feature activation+1.96e+00

Pos logit contributions
ours+0.162
 coconut+0.160
 Enjoy+0.157
rust+0.143
oral+0.141
Neg logit contributions
aces-0.165
amas-0.164
coat-0.162
robe-0.155
 forehead-0.155
Token nice
Token ID3621
Feature activation-7.55e-01

Pos logit contributions
ours+0.180
oral+0.171
 coconut+0.170
 Enjoy+0.167
 arose+0.160
Neg logit contributions
aces-0.103
coat-0.095
drops-0.095
robe-0.094
 forehead-0.094
Token Jack
Token ID3619
Feature activation+1.35e+00

Pos logit contributions
aces+0.064
amas+0.064
coat+0.063
 forehead+0.061
robe+0.060
Neg logit contributions
ours-0.048
 coconut-0.047
 Enjoy-0.046
oral-0.041
rust-0.040
Token.
Token ID13
Feature activation-2.30e-01

Pos logit contributions
ours+0.079
 coconut+0.077
 Enjoy+0.076
oral+0.066
rust+0.065
Neg logit contributions
amas-0.117
coat-0.117
drops-0.115
aces-0.115
 forehead-0.112
Token You
Token ID921
Feature activation-6.35e-01

Pos logit contributions
aces+0.015
amas+0.015
robe+0.014
coat+0.014
stand+0.014
Neg logit contributions
 coconut-0.019
ours-0.018
oral-0.017
rous-0.017
rust-0.017
Tokenmy
Token ID1820
Feature activation-6.35e-01

Pos logit contributions
aces+0.145
coat+0.143
amas+0.143
 forehead+0.138
drops+0.137
Neg logit contributions
ours-0.107
 coconut-0.106
 Enjoy-0.103
oral-0.095
rust-0.094
Token heard
Token ID2982
Feature activation+1.51e+00

Pos logit contributions
aces+0.056
amas+0.054
coat+0.054
robe+0.052
 forehead+0.052
Neg logit contributions
ours-0.039
 coconut-0.038
 Enjoy-0.037
oral-0.034
rust-0.033
Token the
Token ID262
Feature activation+2.34e-01

Pos logit contributions
ours+0.088
 coconut+0.085
 Enjoy+0.083
oral+0.077
rust+0.074
Neg logit contributions
aces-0.136
amas-0.134
coat-0.133
 forehead-0.129
stand-0.129
Token noise
Token ID7838
Feature activation-2.02e+00

Pos logit contributions
 coconut+0.019
ours+0.019
 Enjoy+0.018
oral+0.018
 volcano+0.017
Neg logit contributions
aces-0.016
coat-0.015
amas-0.014
robe-0.014
 cartridge-0.014
Token.
Token ID13
Feature activation-1.15e-01

Pos logit contributions
aces+0.203
amas+0.196
coat+0.196
robe+0.195
 forehead+0.186
Neg logit contributions
ours-0.096
 coconut-0.086
oral-0.080
 Enjoy-0.079
 arose-0.079
Token She
Token ID1375
Feature activation-1.97e-01

Pos logit contributions
amas+0.007
aces+0.006
coat+0.006
robe+0.006
drops+0.006
Neg logit contributions
 coconut-0.010
ours-0.010
 Enjoy-0.010
rust-0.010
oral-0.009
Token came
Token ID1625
Feature activation-1.57e+00

Pos logit contributions
aces+0.014
coat+0.014
amas+0.014
drops+0.014
 forehead+0.014
Neg logit contributions
ours-0.015
 Enjoy-0.014
 coconut-0.014
oral-0.013
rust-0.013
Token to
Token ID284
Feature activation+7.41e-01

Pos logit contributions
aces+0.121
amas+0.119
coat+0.119
 forehead+0.113
robe+0.113
Neg logit contributions
ours-0.113
 coconut-0.110
 Enjoy-0.108
oral-0.101
rust-0.098
Token the
Token ID262
Feature activation+2.71e-01

Pos logit contributions
ours+0.049
 coconut+0.046
 Enjoy+0.044
oral+0.042
rust+0.040
Neg logit contributions
coat-0.061
aces-0.060
amas-0.060
 forehead-0.059
stand-0.059
Token kitchen
Token ID9592
Feature activation+9.00e-01

Pos logit contributions
ours+0.020
 coconut+0.019
 Enjoy+0.019
rust+0.018
oral+0.018
Neg logit contributions
amas-0.020
aces-0.020
coat-0.020
 forehead-0.019
robe-0.019
Token and
Token ID290
Feature activation+2.00e-01

Pos logit contributions
ours+0.046
 coconut+0.041
 Enjoy+0.039
 temple+0.036
 Sometimes+0.036
Neg logit contributions
aces-0.091
coat-0.085
robe-0.083
bridge-0.083
amas-0.083
Token,
Token ID11
Feature activation+6.64e-02

Pos logit contributions
aces+0.073
coat+0.072
amas+0.070
drops+0.067
 forehead+0.067
Neg logit contributions
ours-0.069
 Enjoy-0.067
oral-0.064
 coconut-0.063
rust-0.063
Token red
Token ID2266
Feature activation-1.82e+00

Pos logit contributions
 coconut+0.005
ours+0.005
 Enjoy+0.004
iw+0.004
oral+0.004
Neg logit contributions
aces-0.005
amas-0.005
robe-0.005
coat-0.005
stand-0.005
Token apple
Token ID17180
Feature activation+2.93e+00

Pos logit contributions
aces+0.149
coat+0.146
amas+0.145
drops+0.137
 forehead+0.136
Neg logit contributions
ours-0.122
 Enjoy-0.115
 coconut-0.113
rust-0.111
oral-0.110
Token.
Token ID13
Feature activation+2.94e-01

Pos logit contributions
 coconut+0.176
ours+0.171
 Enjoy+0.156
iw+0.145
oral+0.140
Neg logit contributions
amas-0.264
aces-0.256
coat-0.255
 forehead-0.251
robe-0.246
Token She
Token ID1375
Feature activation-2.12e-01

Pos logit contributions
 coconut+0.018
ours+0.018
 Enjoy+0.017
oral+0.017
rust+0.017
Neg logit contributions
aces-0.025
amas-0.025
coat-0.025
stand-0.024
drops-0.023
Token was
Token ID373
Feature activation+6.80e-01

Pos logit contributions
aces+0.018
coat+0.017
amas+0.016
drops+0.016
robe+0.016
Neg logit contributions
ours-0.014
 Enjoy-0.014
 coconut-0.014
oral-0.013
rous-0.012
Token very
Token ID845
Feature activation+2.12e+00

Pos logit contributions
 coconut+0.039
ours+0.037
 Enjoy+0.037
iw+0.032
 what+0.032
Neg logit contributions
amas-0.064
robe-0.060
 cartridge-0.059
stand-0.059
coat-0.058
Token happy
Token ID3772
Feature activation-1.12e+00

Pos logit contributions
ours+0.170
 coconut+0.164
 Enjoy+0.162
oral+0.156
rust+0.149
Neg logit contributions
aces-0.147
coat-0.143
amas-0.140
 forehead-0.136
drops-0.136
Token.
Token ID13
Feature activation+2.63e-01

Pos logit contributions
aces+0.089
coat+0.087
amas+0.087
robe+0.083
drops+0.083
Neg logit contributions
ours-0.079
 coconut-0.077
 Enjoy-0.075
oral-0.071
rust-0.069
Token
Token ID198
Feature activation+1.09e-01

Pos logit contributions
 coconut+0.018
ours+0.017
rust+0.016
oral+0.015
 Enjoy+0.015
Neg logit contributions
aces-0.022
amas-0.021
coat-0.021
stand-0.021
robe-0.020
Token
Token ID198
Feature activation+9.67e-02

Pos logit contributions
ours+0.007
 coconut+0.007
 Enjoy+0.006
rust+0.006
iw+0.006
Neg logit contributions
stand-0.009
aces-0.009
amas-0.009
coat-0.009
robe-0.008
Token very
Token ID845
Feature activation+1.95e+00

Pos logit contributions
ours+0.031
 coconut+0.030
 Enjoy+0.029
rust+0.026
oral+0.026
Neg logit contributions
aces-0.035
amas-0.035
coat-0.034
robe-0.033
drops-0.033
Token sad
Token ID6507
Feature activation+6.44e-01

Pos logit contributions
ours+0.150
 coconut+0.148
 Enjoy+0.147
oral+0.141
rust+0.133
Neg logit contributions
aces-0.137
 forehead-0.133
amas-0.132
coat-0.131
robe-0.126
Token.
Token ID13
Feature activation-1.44e-01

Pos logit contributions
ours+0.043
 coconut+0.041
 Enjoy+0.040
oral+0.038
rust+0.036
Neg logit contributions
aces-0.053
amas-0.052
coat-0.052
 forehead-0.050
robe-0.050
Token
Token ID220
Feature activation-1.24e-02

Pos logit contributions
aces+0.012
amas+0.012
coat+0.011
robe+0.011
stand+0.011
Neg logit contributions
ours-0.010
 coconut-0.010
 Enjoy-0.009
oral-0.009
rust-0.008
Token
Token ID198
Feature activation-2.16e-01

Pos logit contributions
aces+0.001
amas+0.001
coat+0.001
robe+0.001
 forehead+0.001
Neg logit contributions
ours-0.001
 coconut-0.001
 Enjoy-0.001
oral-0.000
rust-0.000
Token
Token ID198
Feature activation-2.20e-01

Pos logit contributions
stand+0.017
amas+0.017
aces+0.017
coat+0.017
robe+0.016
Neg logit contributions
ours-0.015
 coconut-0.014
 Enjoy-0.013
rust-0.012
oral-0.012
TokenBut
Token ID1537
Feature activation+8.13e-01

Pos logit contributions
aces+0.017
amas+0.017
coat+0.017
robe+0.016
 forehead+0.016
Neg logit contributions
ours-0.016
 coconut-0.015
 Enjoy-0.015
oral-0.014
rust-0.014
Token the
Token ID262
Feature activation-1.33e-01

Pos logit contributions
ours+0.073
 coconut+0.071
 Enjoy+0.069
oral+0.068
rust+0.065
Neg logit contributions
aces-0.049
 forehead-0.047
coat-0.045
amas-0.044
drops-0.043
Token princess
Token ID21752
Feature activation-1.35e+00

Pos logit contributions
aces+0.010
amas+0.010
coat+0.010
drops+0.009
robe+0.009
Neg logit contributions
 coconut-0.010
ours-0.010
 Enjoy-0.009
oral-0.009
rust-0.009
Token didn
Token ID1422
Feature activation+3.97e-02

Pos logit contributions
aces+0.116
coat+0.113
amas+0.111
robe+0.108
 forehead+0.108
Neg logit contributions
ours-0.087
 coconut-0.084
 Enjoy-0.083
oral-0.075
rust-0.072
Token't
Token ID470
Feature activation-2.09e+00

Pos logit contributions
ours+0.003
 coconut+0.003
 Enjoy+0.003
oral+0.003
rust+0.003
Neg logit contributions
aces-0.003
coat-0.002
amas-0.002
 forehead-0.002
drops-0.002
Token the
Token ID262
Feature activation+7.71e-02

Pos logit contributions
aces+0.263
coat+0.257
amas+0.257
robe+0.247
 forehead+0.247
Neg logit contributions
ours-0.157
 coconut-0.152
 Enjoy-0.150
oral-0.136
rust-0.132
Token fairy
Token ID25607
Feature activation-1.60e+00

Pos logit contributions
 coconut+0.005
ours+0.005
 Enjoy+0.005
oral+0.004
rust+0.004
Neg logit contributions
aces-0.007
amas-0.006
coat-0.006
robe-0.006
drops-0.006
Token find
Token ID1064
Feature activation-3.02e-01

Pos logit contributions
amas+0.134
aces+0.133
coat+0.132
drops+0.127
robe+0.124
Neg logit contributions
ours-0.102
 coconut-0.099
 Enjoy-0.096
oral-0.087
arium-0.086
Token her
Token ID607
Feature activation-1.26e+00

Pos logit contributions
coat+0.028
aces+0.028
amas+0.028
drops+0.027
robe+0.026
Neg logit contributions
ours-0.017
 coconut-0.017
 Enjoy-0.016
oral-0.014
rust-0.014
Token way
Token ID835
Feature activation-1.80e+00

Pos logit contributions
aces+0.107
amas+0.106
coat+0.105
drops+0.103
bridge+0.100
Neg logit contributions
 coconut-0.079
ours-0.078
 Enjoy-0.074
oral-0.069
 Sometimes-0.067
Token home
Token ID1363
Feature activation+1.17e+00

Pos logit contributions
aces+0.188
amas+0.185
bridge+0.180
coat+0.180
drops+0.176
Neg logit contributions
 coconut-0.079
ours-0.075
 Enjoy-0.072
 Sometimes-0.066
arium-0.063
Token.
Token ID13
Feature activation+4.46e-01

Pos logit contributions
 coconut+0.048
ours+0.048
 Enjoy+0.044
arium+0.041
 Sometimes+0.040
Neg logit contributions
amas-0.118
aces-0.117
coat-0.115
bridge-0.112
 cartridge-0.112
Token The
Token ID383
Feature activation+4.78e-01

Pos logit contributions
ours+0.027
 coconut+0.026
 Enjoy+0.025
oral+0.023
rust+0.022
Neg logit contributions
aces-0.040
coat-0.039
amas-0.039
 forehead-0.037
robe-0.037
Token fairy
Token ID25607
Feature activation-1.61e+00

Pos logit contributions
 coconut+0.029
ours+0.028
 Enjoy+0.028
oral+0.027
rust+0.025
Neg logit contributions
aces-0.044
amas-0.042
coat-0.041
robe-0.040
drops-0.040
Token was
Token ID373
Feature activation+6.79e-01

Pos logit contributions
aces+0.147
amas+0.147
coat+0.144
drops+0.139
robe+0.138
Neg logit contributions
ours-0.090
 coconut-0.088
 Enjoy-0.083
oral-0.079
rust-0.075
Token grateful
Token ID14066
Feature activation-2.00e+00

Pos logit contributions
ours+0.036
 coconut+0.034
 Enjoy+0.033
 asks+0.033
grown+0.032
Neg logit contributions
amas-0.064
drops-0.062
coat-0.062
aces-0.061
robe-0.059
Token big
Token ID1263
Feature activation-1.02e+00

Pos logit contributions
ours+0.145
 coconut+0.141
 Enjoy+0.140
oral+0.131
rust+0.127
Neg logit contributions
aces-0.151
coat-0.148
amas-0.145
robe-0.142
drops-0.141
Token and
Token ID290
Feature activation+2.12e-01

Pos logit contributions
amas+0.092
coat+0.088
aces+0.088
 forehead+0.086
robe+0.084
Neg logit contributions
ours-0.062
 coconut-0.059
 Enjoy-0.058
grown-0.051
oral-0.051
Token fluffy
Token ID39145
Feature activation-2.75e+00

Pos logit contributions
ours+0.016
 coconut+0.016
 Enjoy+0.015
oral+0.014
rust+0.014
Neg logit contributions
amas-0.016
aces-0.016
coat-0.015
robe-0.014
 forehead-0.014
Token that
Token ID326
Feature activation-9.05e-01

Pos logit contributions
amas+0.246
aces+0.244
 forehead+0.237
coat+0.237
robe+0.228
Neg logit contributions
ours-0.157
 coconut-0.152
 Enjoy-0.143
grown-0.132
 crab-0.126
Token looked
Token ID3114
Feature activation+1.74e+00

Pos logit contributions
amas+0.082
aces+0.081
coat+0.080
 forehead+0.078
drops+0.076
Neg logit contributions
ours-0.053
 coconut-0.052
 Enjoy-0.049
rust-0.045
oral-0.045
Token like
Token ID588
Feature activation-2.45e+00

Pos logit contributions
ours+0.149
 coconut+0.144
 Enjoy+0.143
oral+0.137
rust+0.132
Neg logit contributions
aces-0.110
coat-0.106
amas-0.102
robe-0.101
 forehead-0.100
Token a
Token ID257
Feature activation-4.37e-01

Pos logit contributions
amas+0.220
aces+0.213
coat+0.213
 forehead+0.208
drops+0.204
Neg logit contributions
ours-0.146
 coconut-0.143
 Enjoy-0.137
oral-0.121
grown-0.119
Token bear
Token ID6842
Feature activation-1.77e+00

Pos logit contributions
aces+0.029
coat+0.028
amas+0.028
robe+0.027
 forehead+0.027
Neg logit contributions
ours-0.036
 coconut-0.035
 Enjoy-0.034
oral-0.032
rust-0.032
Token.
Token ID13
Feature activation+2.02e-01

Pos logit contributions
amas+0.154
aces+0.154
coat+0.150
 forehead+0.148
robe+0.144
Neg logit contributions
ours-0.107
 coconut-0.103
 Enjoy-0.101
oral-0.089
rust-0.087
Token
Token ID198
Feature activation+2.98e-02

Pos logit contributions
 coconut+0.014
ours+0.014
oral+0.013
 Enjoy+0.013
rous+0.012
Neg logit contributions
robe-0.015
aces-0.015
stand-0.015
amas-0.015
 cartridge-0.014
Token
Token ID198
Feature activation+2.93e-02

Pos logit contributions
 coconut+0.002
ours+0.002
 Enjoy+0.002
iw+0.002
rust+0.002
Neg logit contributions
amas-0.002
stand-0.002
aces-0.002
robe-0.002
coat-0.002
Token.
Token ID13
Feature activation+8.98e-02

Pos logit contributions
amas+0.443
robe+0.412
 cartridge+0.411
bridge+0.403
aces+0.403
Neg logit contributions
 coconut-0.134
:-0.120
 whose-0.109
ours-0.106
 what-0.105
Token It
Token ID632
Feature activation-3.36e-01

Pos logit contributions
ours+0.006
 coconut+0.006
 Enjoy+0.006
oral+0.005
rust+0.005
Neg logit contributions
aces-0.008
amas-0.007
coat-0.007
robe-0.007
 forehead-0.007
Token was
Token ID373
Feature activation+6.82e-01

Pos logit contributions
aces+0.030
amas+0.030
coat+0.029
robe+0.028
stand+0.028
Neg logit contributions
ours-0.020
 coconut-0.020
 Enjoy-0.018
oral-0.018
rust-0.017
Token scared
Token ID12008
Feature activation+7.36e-01

Pos logit contributions
 coconut+0.041
ours+0.040
 Enjoy+0.038
grown+0.036
 what+0.036
Neg logit contributions
amas-0.060
aces-0.060
robe-0.057
coat-0.057
stand-0.055
Token and
Token ID290
Feature activation+2.25e-01

Pos logit contributions
ours+0.036
 coconut+0.034
 Sometimes+0.033
 Enjoy+0.033
grown+0.032
Neg logit contributions
aces-0.073
amas-0.073
coat-0.071
robe-0.071
 forehead-0.069
Token so
Token ID523
Feature activation-2.31e+00

Pos logit contributions
ours+0.014
 coconut+0.013
 Enjoy+0.013
oral+0.012
rust+0.012
Neg logit contributions
aces-0.020
amas-0.019
coat-0.019
robe-0.018
drops-0.018
Token it
Token ID340
Feature activation+1.42e+00

Pos logit contributions
aces+0.210
amas+0.206
coat+0.206
robe+0.198
 forehead+0.198
Neg logit contributions
ours-0.134
 coconut-0.130
 Enjoy-0.127
oral-0.116
rust-0.112
Token started
Token ID2067
Feature activation+4.27e-01

Pos logit contributions
ours+0.088
 coconut+0.087
 Enjoy+0.083
oral+0.078
rust+0.076
Neg logit contributions
aces-0.121
amas-0.121
coat-0.118
robe-0.114
stand-0.113
Token to
Token ID284
Feature activation+5.90e-01

Pos logit contributions
ours+0.022
 coconut+0.021
 Enjoy+0.020
oral+0.018
rust+0.018
Neg logit contributions
aces-0.042
amas-0.041
coat-0.041
robe-0.040
 forehead-0.039
Token hope
Token ID2911
Feature activation-1.96e+00

Pos logit contributions
 coconut+0.042
ours+0.040
 Enjoy+0.039
 swamp+0.037
 Sometimes+0.036
Neg logit contributions
aces-0.047
amas-0.045
robe-0.043
coat-0.043
bridge-0.041
Token that
Token ID326
Feature activation-8.17e-01

Pos logit contributions
amas+0.200
aces+0.200
coat+0.198
robe+0.192
 forehead+0.187
Neg logit contributions
ours-0.087
 coconut-0.087
 Enjoy-0.083
oral-0.076
 Sometimes-0.075
Token I
Token ID314
Feature activation-6.35e-01

Pos logit contributions
 Enjoy+0.012
ours+0.011
 coconut+0.010
 asks+0.010
 Joe+0.009
Neg logit contributions
coat-0.024
aces-0.024
amas-0.024
robe-0.023
amed-0.023
Token know
Token ID760
Feature activation+2.32e+00

Pos logit contributions
aces+0.055
coat+0.054
robe+0.052
amas+0.052
drops+0.052
Neg logit contributions
 coconut-0.039
ours-0.039
 Enjoy-0.036
rust-0.034
oral-0.034
Token that
Token ID326
Feature activation-8.69e-01

Pos logit contributions
ours+0.122
 coconut+0.119
 Enjoy+0.111
oral+0.108
rust+0.105
Neg logit contributions
aces-0.223
amas-0.215
coat-0.214
 forehead-0.209
robe-0.206
Token you
Token ID345
Feature activation+1.76e-02

Pos logit contributions
aces+0.074
amas+0.072
coat+0.071
robe+0.069
 forehead+0.069
Neg logit contributions
ours-0.055
 coconut-0.053
 Enjoy-0.052
oral-0.049
rust-0.048
Token can
Token ID460
Feature activation+6.59e-01

Pos logit contributions
ours+0.001
 coconut+0.001
 Enjoy+0.001
oral+0.001
rust+0.001
Neg logit contributions
aces-0.002
coat-0.001
amas-0.001
robe-0.001
 forehead-0.001
Token make
Token ID787
Feature activation-3.30e+00

Pos logit contributions
ours+0.046
 coconut+0.045
 Enjoy+0.044
oral+0.041
rust+0.040
Neg logit contributions
aces-0.052
amas-0.050
coat-0.050
robe-0.048
 forehead-0.048
Token something
Token ID1223
Feature activation+2.13e+00

Pos logit contributions
aces+0.298
coat+0.294
amas+0.291
robe+0.282
drops+0.282
Neg logit contributions
ours-0.194
 coconut-0.186
 Enjoy-0.182
oral-0.166
rust-0.162
Token beautiful
Token ID4950
Feature activation-9.00e-01

Pos logit contributions
ours+0.147
 coconut+0.143
 Enjoy+0.140
oral+0.130
rust+0.126
Neg logit contributions
aces-0.170
amas-0.166
coat-0.165
robe-0.158
 forehead-0.158
Token out
Token ID503
Feature activation-1.42e+00

Pos logit contributions
aces+0.079
coat+0.079
amas+0.079
 forehead+0.077
drops+0.075
Neg logit contributions
ours-0.052
 Enjoy-0.051
 coconut-0.050
oral-0.045
 Sometimes-0.043
Token of
Token ID286
Feature activation+1.67e-01

Pos logit contributions
aces+0.124
amas+0.121
coat+0.121
robe+0.117
 forehead+0.116
Neg logit contributions
ours-0.087
 coconut-0.084
 Enjoy-0.082
oral-0.076
rust-0.074
Token something
Token ID1223
Feature activation+2.13e+00

Pos logit contributions
ours+0.010
 coconut+0.009
 Enjoy+0.009
oral+0.008
rust+0.008
Neg logit contributions
aces-0.015
amas-0.015
coat-0.015
 forehead-0.014
robe-0.014
Token �
Token ID6184
Feature activation+1.99e+00

Pos logit contributions
aces+0.008
coat+0.008
amas+0.008
robe+0.008
 forehead+0.008
Neg logit contributions
ours-0.006
 coconut-0.006
 Enjoy-0.006
oral-0.005
rust-0.005
Token�
Token ID95
Feature activation-1.19e+00

Pos logit contributions
ours+0.146
 coconut+0.142
 Enjoy+0.139
oral+0.130
rust+0.127
Neg logit contributions
aces-0.150
amas-0.147
coat-0.146
robe-0.140
 forehead-0.140
Token€
Token ID26391
Feature activation-2.00e+00

Pos logit contributions
aces+0.096
coat+0.096
 forehead+0.095
amas+0.095
robe+0.092
Neg logit contributions
 Enjoy-0.077
ours-0.075
oral-0.068
 coconut-0.066
 Sometimes-0.063
Token�
Token ID129
Feature activation-2.42e+00

Pos logit contributions
 forehead+0.169
aces+0.163
robe+0.156
coat+0.156
 cartridge+0.153
Neg logit contributions
ours-0.133
 Enjoy-0.128
 coconut-0.116
 Sometimes-0.112
oral-0.112
Token�
Token ID241
Feature activation-4.22e+00

Pos logit contributions
aces+0.177
robe+0.177
coat+0.176
amas+0.174
 forehead+0.168
Neg logit contributions
ours-0.176
 Enjoy-0.175
 coconut-0.164
oral-0.156
rust-0.153
TokenThe
Token ID464
Feature activation-1.11e+00

Pos logit contributions
coat+0.335
aces+0.330
amas+0.317
drops+0.315
stand+0.312
Neg logit contributions
ours-0.297
 coconut-0.282
 Enjoy-0.261
oral-0.255
rust-0.254
Token mountains
Token ID12269
Feature activation-3.54e-01

Pos logit contributions
coat+0.103
amas+0.103
aces+0.102
 forehead+0.098
drops+0.097
Neg logit contributions
 Enjoy-0.060
 coconut-0.058
ours-0.058
oral-0.056
 temple-0.051
Token look
Token ID804
Feature activation+3.89e-01

Pos logit contributions
amas+0.034
aces+0.033
coat+0.033
 forehead+0.032
stand+0.032
Neg logit contributions
ours-0.019
 coconut-0.018
 Enjoy-0.017
oral-0.016
 Sometimes-0.014
Token very
Token ID845
Feature activation+1.99e+00

Pos logit contributions
ours+0.025
 Enjoy+0.024
 coconut+0.023
 Sometimes+0.021
grown+0.021
Neg logit contributions
amas-0.034
coat-0.033
aces-0.033
 forehead-0.032
robe-0.032
Token clear
Token ID1598
Feature activation-1.00e+00

Pos logit contributions
ours+0.161
 coconut+0.158
 Enjoy+0.155
oral+0.150
rust+0.144
Neg logit contributions
aces-0.130
amas-0.125
coat-0.124
drops-0.122
 forehead-0.120
Token today
Token ID1909
Feature activation+1.58e+00

Pos logit contributions
coat+0.092
amas+0.092
aces+0.091
 forehead+0.089
robe+0.089
Neg logit contributions
ours-0.056
 coconut-0.053
 Enjoy-0.052
grown-0.046
 Sometimes-0.045
Token used
Token ID973
Feature activation+1.16e+00

Pos logit contributions
aces+0.019
coat+0.018
amas+0.018
 forehead+0.018
drops+0.017
Neg logit contributions
ours-0.016
 coconut-0.015
 Enjoy-0.015
oral-0.014
rust-0.013
Token the
Token ID262
Feature activation+1.16e-01

Pos logit contributions
 coconut+0.047
ours+0.045
 Enjoy+0.044
 Sometimes+0.034
rust+0.034
Neg logit contributions
amas-0.127
aces-0.125
coat-0.124
drops-0.124
robe-0.122
Token flashlight
Token ID38371
Feature activation-1.03e+00

Pos logit contributions
 coconut+0.008
 Enjoy+0.007
ours+0.007
oral+0.007
 question+0.007
Neg logit contributions
amas-0.009
aces-0.009
coat-0.009
robe-0.009
drops-0.009
Token to
Token ID284
Feature activation+4.80e-01

Pos logit contributions
amas+0.109
aces+0.104
coat+0.101
bridge+0.100
bes+0.100
Neg logit contributions
 coconut-0.041
ours-0.038
fired-0.034
 Enjoy-0.033
arium-0.032
Token show
Token ID905
Feature activation+1.60e+00

Pos logit contributions
 coconut+0.030
 Enjoy+0.029
ours+0.027
oral+0.025
 temple+0.025
Neg logit contributions
aces-0.042
amas-0.041
coat-0.039
robe-0.039
drops-0.038
Token him
Token ID683
Feature activation-1.70e+00

Pos logit contributions
ours+0.090
 Enjoy+0.090
 coconut+0.089
oral+0.078
rust+0.075
Neg logit contributions
coat-0.148
amas-0.145
aces-0.145
drops-0.139
robe-0.138
Token the
Token ID262
Feature activation+4.63e-02

Pos logit contributions
aces+0.159
coat+0.149
amas+0.149
 forehead+0.149
robe+0.147
Neg logit contributions
ours-0.096
 coconut-0.090
 Enjoy-0.089
oral-0.084
rust-0.082
Token way
Token ID835
Feature activation-1.80e+00

Pos logit contributions
ours+0.003
 coconut+0.003
 Enjoy+0.003
oral+0.003
rust+0.003
Neg logit contributions
aces-0.004
amas-0.004
coat-0.004
robe-0.003
 forehead-0.003
Token home
Token ID1363
Feature activation+1.16e+00

Pos logit contributions
aces+0.162
amas+0.159
coat+0.155
robe+0.150
drops+0.148
Neg logit contributions
 coconut-0.107
ours-0.104
 Enjoy-0.100
 Joe-0.091
 Sometimes-0.090
Token.
Token ID13
Feature activation+1.58e-02

Pos logit contributions
 coconut+0.055
ours+0.054
 Enjoy+0.050
 Joe+0.046
 temple+0.045
Neg logit contributions
amas-0.112
aces-0.110
coat-0.110
 cartridge-0.106
 forehead-0.105
Token
Token ID220
Feature activation+5.08e-02

Pos logit contributions
ours+0.001
 coconut+0.001
rust+0.001
oral+0.001
rous+0.001
Neg logit contributions
amas-0.001
aces-0.001
stand-0.001
robe-0.001
bridge-0.001
Token.
Token ID13
Feature activation+2.66e-01

Pos logit contributions
aces+0.162
amas+0.162
coat+0.154
robe+0.152
 forehead+0.151
Neg logit contributions
ours-0.102
 coconut-0.096
 Enjoy-0.092
 Sometimes-0.083
iw-0.082
Token I
Token ID314
Feature activation-5.83e-01

Pos logit contributions
ours+0.017
 coconut+0.016
 Enjoy+0.015
oral+0.015
rust+0.014
Neg logit contributions
aces-0.023
amas-0.022
coat-0.022
robe-0.021
 forehead-0.021
Token have
Token ID423
Feature activation+2.27e+00

Pos logit contributions
coat+0.046
aces+0.042
robe+0.042
drops+0.042
brush+0.040
Neg logit contributions
ours-0.042
 coconut-0.040
oral-0.037
 Enjoy-0.036
rust-0.036
Token some
Token ID617
Feature activation+5.04e-01

Pos logit contributions
ours+0.144
 coconut+0.139
 Enjoy+0.136
oral+0.128
rust+0.123
Neg logit contributions
aces-0.195
coat-0.189
amas-0.189
 forehead-0.182
robe-0.181
Token band
Token ID4097
Feature activation-1.75e+00

Pos logit contributions
ours+0.037
 coconut+0.035
 Enjoy+0.035
oral+0.033
rust+0.032
Neg logit contributions
aces-0.039
coat-0.038
amas-0.037
 forehead-0.036
robe-0.036
Tokenages
Token ID1095
Feature activation-7.63e+00

Pos logit contributions
amas+0.128
 cartridge+0.114
 forehead+0.113
aces+0.109
coat+0.107
Neg logit contributions
 Enjoy-0.133
ours-0.128
 coconut-0.117
iw-0.115
rill-0.114
Token and
Token ID290
Feature activation+2.55e-01

Pos logit contributions
amas+0.700
aces+0.684
coat+0.675
 forehead+0.652
drops+0.651
Neg logit contributions
ours-0.430
 Enjoy-0.416
 coconut-0.402
oral-0.347
iw-0.344
Token some
Token ID617
Feature activation+4.98e-01

Pos logit contributions
ours+0.017
 coconut+0.016
 Enjoy+0.016
oral+0.015
rust+0.014
Neg logit contributions
aces-0.021
amas-0.021
coat-0.021
robe-0.020
drops-0.020
Token medicine
Token ID9007
Feature activation-9.78e-01

Pos logit contributions
ours+0.035
 coconut+0.034
 Enjoy+0.034
oral+0.031
rust+0.030
Neg logit contributions
aces-0.039
amas-0.038
coat-0.038
robe-0.036
 forehead-0.036
Token.
Token ID13
Feature activation+2.69e-01

Pos logit contributions
amas+0.087
aces+0.085
coat+0.085
 forehead+0.082
drops+0.081
Neg logit contributions
ours-0.058
 Enjoy-0.055
 coconut-0.054
oral-0.049
arium-0.046
Token I
Token ID314
Feature activation-5.90e-01

Pos logit contributions
ours+0.018
 coconut+0.017
 Enjoy+0.016
oral+0.016
rust+0.016
Neg logit contributions
aces-0.022
amas-0.022
coat-0.021
 forehead-0.021
robe-0.021
Token made
Token ID925
Feature activation-4.35e+00

Pos logit contributions
aces+0.119
amas+0.119
 forehead+0.115
coat+0.113
drops+0.110
Neg logit contributions
 coconut-0.056
ours-0.056
 arose-0.048
 Enjoy-0.047
 asks-0.047
Token a
Token ID257
Feature activation-4.59e-01

Pos logit contributions
coat+0.473
aces+0.469
amas+0.454
drops+0.443
robe+0.441
Neg logit contributions
ours-0.181
 coconut-0.173
 Enjoy-0.168
 Joe-0.151
 Sometimes-0.148
Token loud
Token ID7812
Feature activation-9.47e-01

Pos logit contributions
aces+0.034
coat+0.031
amas+0.030
robe+0.030
drops+0.029
Neg logit contributions
 coconut-0.036
ours-0.036
 Enjoy-0.034
oral-0.033
rust-0.032
Token wh
Token ID348
Feature activation-2.58e+00

Pos logit contributions
aces+0.073
coat+0.068
robe+0.067
amas+0.064
 cartridge+0.063
Neg logit contributions
ours-0.068
 coconut-0.067
 Enjoy-0.065
 volcano-0.063
 question-0.062
Tokenir
Token ID343
Feature activation+3.84e+00

Pos logit contributions
aces+0.209
amas+0.205
amed+0.194
coat+0.193
bridge+0.193
Neg logit contributions
 Enjoy-0.183
 coconut-0.176
ours-0.162
 Crab-0.154
 Sometimes-0.153
Tokenring
Token ID1806
Feature activation-6.11e+00

Pos logit contributions
iles+0.260
:+0.259
oral+0.257
 Joe+0.255
 coconut+0.253
Neg logit contributions
coat-0.262
 cartridge-0.260
robe-0.256
amas-0.252
bes-0.228
Token sound
Token ID2128
Feature activation-2.93e+00

Pos logit contributions
aces+0.510
 cartridge+0.503
coat+0.480
amas+0.476
robe+0.469
Neg logit contributions
ours-0.338
 arose-0.320
 volcano-0.306
 coconut-0.305
 Sometimes-0.300
Token.
Token ID13
Feature activation+2.68e-02

Pos logit contributions
amas+0.334
aces+0.323
coat+0.322
robe+0.321
 cartridge+0.312
Neg logit contributions
 what-0.085
:-0.084
 Joe-0.080
 why-0.077
ours-0.075
Token He
Token ID679
Feature activation-2.40e-01

Pos logit contributions
ours+0.002
 coconut+0.002
oral+0.002
rous+0.002
rust+0.002
Neg logit contributions
amas-0.002
aces-0.002
stand-0.002
robe-0.002
coat-0.002
Token wanted
Token ID2227
Feature activation+1.48e+00

Pos logit contributions
aces+0.018
coat+0.018
drops+0.018
amas+0.018
 forehead+0.017
Neg logit contributions
 Enjoy-0.017
ours-0.017
 coconut-0.016
oral-0.016
iw-0.015
Token to
Token ID284
Feature activation+6.14e-01

Pos logit contributions
ours+0.097
 coconut+0.090
 Enjoy+0.089
oral+0.089
rust+0.084
Neg logit contributions
aces-0.124
coat-0.122
stand-0.119
amas-0.119
 forehead-0.117
Token.
Token ID13
Feature activation+1.61e-01

Pos logit contributions
ours+0.190
 coconut+0.184
 Enjoy+0.179
oral+0.167
rust+0.159
Neg logit contributions
aces-0.295
amas-0.289
coat-0.286
robe-0.278
 forehead-0.274
Token He
Token ID679
Feature activation-4.54e-02

Pos logit contributions
 coconut+0.011
ours+0.011
 Enjoy+0.011
oral+0.010
rust+0.010
Neg logit contributions
aces-0.013
amas-0.012
 forehead-0.012
robe-0.012
coat-0.012
Token snaps
Token ID23429
Feature activation+1.96e+00

Pos logit contributions
aces+0.003
coat+0.003
drops+0.003
 forehead+0.003
amas+0.003
Neg logit contributions
 coconut-0.003
oral-0.003
 Enjoy-0.003
ours-0.003
rust-0.003
Token the
Token ID262
Feature activation+5.35e-02

Pos logit contributions
ours+0.110
 coconut+0.107
 Enjoy+0.106
oral+0.099
rust+0.094
Neg logit contributions
aces-0.182
coat-0.175
amas-0.175
robe-0.170
 forehead-0.170
Token snow
Token ID6729
Feature activation-6.74e-01

Pos logit contributions
ours+0.003
 coconut+0.003
 Enjoy+0.003
oral+0.003
rust+0.003
Neg logit contributions
aces-0.005
amas-0.004
coat-0.004
drops-0.004
robe-0.004
Tokenman
Token ID805
Feature activation-7.45e+00

Pos logit contributions
aces+0.045
coat+0.044
amas+0.043
stand+0.041
 cartridge+0.040
Neg logit contributions
 coconut-0.054
ours-0.051
 volcano-0.050
 crab-0.046
 crocod-0.046
Token's
Token ID338
Feature activation-1.25e+00

Pos logit contributions
aces+0.597
coat+0.592
amas+0.591
robe+0.574
drops+0.553
Neg logit contributions
ours-0.508
 coconut-0.500
 Enjoy-0.479
oral-0.437
rust-0.425
Token arm
Token ID3211
Feature activation-6.21e-01

Pos logit contributions
aces+0.083
coat+0.082
amas+0.082
drops+0.078
robe+0.078
Neg logit contributions
 coconut-0.103
ours-0.101
 Enjoy-0.095
oral-0.091
iw-0.086
Token.
Token ID13
Feature activation+1.61e-01

Pos logit contributions
amas+0.063
coat+0.061
robe+0.060
 forehead+0.059
 cartridge+0.059
Neg logit contributions
ours-0.025
 coconut-0.024
 Sometimes-0.021
 whose-0.019
:-0.019
Token He
Token ID679
Feature activation-3.44e-02

Pos logit contributions
 coconut+0.011
ours+0.011
 Enjoy+0.010
oral+0.010
rust+0.010
Neg logit contributions
aces-0.013
amas-0.013
robe-0.012
 forehead-0.012
coat-0.012
Token breaks
Token ID9457
Feature activation+1.75e+00

Pos logit contributions
aces+0.003
drops+0.002
coat+0.002
 forehead+0.002
amas+0.002
Neg logit contributions
oral-0.002
 coconut-0.002
 Enjoy-0.002
ours-0.002
rust-0.002
Token.
Token ID13
Feature activation+3.18e-02

Pos logit contributions
amas+0.157
aces+0.155
 forehead+0.151
bridge+0.149
bes+0.148
Neg logit contributions
ours-0.060
 arose-0.054
 coconut-0.053
 Joe-0.052
 Enjoy-0.050
Token She
Token ID1375
Feature activation-4.12e-01

Pos logit contributions
ours+0.002
 coconut+0.002
 Enjoy+0.002
rust+0.002
oral+0.002
Neg logit contributions
amas-0.002
aces-0.002
coat-0.002
 forehead-0.002
drops-0.002
Token saw
Token ID2497
Feature activation-4.69e-01

Pos logit contributions
aces+0.033
coat+0.033
amas+0.032
 forehead+0.031
robe+0.031
Neg logit contributions
ours-0.028
 Enjoy-0.027
 coconut-0.027
oral-0.025
rust-0.023
Token the
Token ID262
Feature activation-1.81e-01

Pos logit contributions
aces+0.042
amas+0.042
coat+0.042
 forehead+0.040
robe+0.040
Neg logit contributions
ours-0.028
 coconut-0.027
 Enjoy-0.026
oral-0.024
rust-0.023
Token fire
Token ID2046
Feature activation+1.54e+00

Pos logit contributions
aces+0.017
coat+0.016
amas+0.016
robe+0.016
stand+0.016
Neg logit contributions
ours-0.010
 coconut-0.010
 Enjoy-0.010
oral-0.009
 Sometimes-0.008
Tokenman
Token ID805
Feature activation-7.81e+00

Pos logit contributions
ours+0.113
 coconut+0.109
 Enjoy+0.103
oral+0.099
 volcano+0.098
Neg logit contributions
aces-0.120
coat-0.115
amas-0.112
 forehead-0.107
stand-0.107
Token working
Token ID1762
Feature activation-3.84e-01

Pos logit contributions
aces+0.636
coat+0.620
amas+0.620
robe+0.595
 forehead+0.590
Neg logit contributions
ours-0.528
 coconut-0.512
 Enjoy-0.500
oral-0.466
rust-0.450
Token hard
Token ID1327
Feature activation+2.86e+00

Pos logit contributions
aces+0.033
amas+0.032
 forehead+0.032
coat+0.032
robe+0.031
Neg logit contributions
ours-0.023
 coconut-0.022
 Enjoy-0.021
grown-0.020
oral-0.019
Token to
Token ID284
Feature activation+3.75e-01

Pos logit contributions
ours+0.154
 Enjoy+0.139
 coconut+0.139
 Sometimes+0.131
grown+0.131
Neg logit contributions
aces-0.260
amas-0.256
coat-0.256
 forehead-0.255
bridge-0.249
Token put
Token ID1234
Feature activation-1.72e+00

Pos logit contributions
 coconut+0.027
ours+0.027
 Enjoy+0.027
oral+0.024
rust+0.024
Neg logit contributions
aces-0.029
amas-0.028
coat-0.027
robe-0.026
 forehead-0.026
Token out
Token ID503
Feature activation-1.53e+00

Pos logit contributions
aces+0.151
amas+0.150
coat+0.148
 forehead+0.146
drops+0.142
Neg logit contributions
ours-0.101
 coconut-0.096
 Enjoy-0.096
 Sometimes-0.083
grown-0.083
Token water
Token ID1660
Feature activation+1.05e+00

Pos logit contributions
ours+0.272
 coconut+0.265
 Enjoy+0.260
oral+0.244
rust+0.237
Neg logit contributions
aces-0.281
coat-0.273
amas-0.273
robe-0.261
 forehead-0.261
Token and
Token ID290
Feature activation+2.62e-01

Pos logit contributions
ours+0.065
 coconut+0.060
 Enjoy+0.056
oral+0.051
iw+0.050
Neg logit contributions
amas-0.094
aces-0.091
 forehead-0.089
coat-0.087
bridge-0.086
Token make
Token ID787
Feature activation-3.12e+00

Pos logit contributions
ours+0.018
 coconut+0.017
 Enjoy+0.017
oral+0.015
rust+0.015
Neg logit contributions
amas-0.021
aces-0.021
coat-0.021
robe-0.020
 forehead-0.020
Token more
Token ID517
Feature activation+3.71e+00

Pos logit contributions
aces+0.268
coat+0.264
amas+0.263
robe+0.251
 forehead+0.251
Neg logit contributions
ours-0.200
 coconut-0.189
 Enjoy-0.185
oral-0.170
rust-0.169
Token rain
Token ID6290
Feature activation-1.22e+00

Pos logit contributions
ours+0.310
 coconut+0.303
 Enjoy+0.298
oral+0.281
rust+0.275
Neg logit contributions
aces-0.243
amas-0.236
coat-0.235
robe-0.224
 forehead-0.223
Tokenbows
Token ID25435
Feature activation-7.62e+00

Pos logit contributions
amas+0.065
aces+0.063
 cartridge+0.059
 forehead+0.058
bridge+0.055
Neg logit contributions
ours-0.114
 Sometimes-0.101
 coconut-0.100
 swamp-0.098
 temple-0.097
Token.
Token ID13
Feature activation+1.95e-01

Pos logit contributions
amas+0.853
coat+0.842
 forehead+0.819
aces+0.814
robe+0.790
Neg logit contributions
ours-0.260
 coconut-0.255
 volcano-0.222
 temple-0.214
 Enjoy-0.200
Token They
Token ID1119
Feature activation+1.27e+00

Pos logit contributions
 coconut+0.013
ours+0.013
oral+0.011
 Enjoy+0.011
rous+0.011
Neg logit contributions
aces-0.016
robe-0.016
amas-0.016
stand-0.015
coat-0.015
Token laugh
Token ID6487
Feature activation+1.17e-01

Pos logit contributions
ours+0.089
 coconut+0.087
 Enjoy+0.085
oral+0.079
rust+0.077
Neg logit contributions
aces-0.100
coat-0.098
amas-0.097
 forehead-0.094
robe-0.094
Token and
Token ID290
Feature activation+2.59e-01

Pos logit contributions
ours+0.004
 Enjoy+0.004
 Sometimes+0.004
 what+0.004
:+0.004
Neg logit contributions
coat-0.013
amas-0.013
aces-0.013
drops-0.012
bes-0.012
Token play
Token ID711
Feature activation+1.95e-01

Pos logit contributions
 Enjoy+0.015
ours+0.015
 coconut+0.015
 asks+0.013
rust+0.013
Neg logit contributions
amas-0.023
aces-0.023
coat-0.022
bridge-0.022
robe-0.021