TOP ACTIVATIONS
MAX = 0.198

sad . She threw a tant rum and cried loudly .
kept crying and throwing a tant rum . Finally
to cry and threw a tant rum . She didn 't
listen . She threw a tant rum in the store .
She cried and threw a tant rum . Her mom tried
and started to throw a tant rum . Suddenly , she
her feet and threw a tant rum . She grabbed the
. He kept throwing a tant rum . His mom was
and so she threw a tant rum when her parents said
surprise . He threw a tant rum . He kicked and
stubborn . She throws a tant rum . She kicks and
vegetables laughed at the boast ful onion . One of them
out loud and enjoyed the sensation of being so high up
his feet and threw a tant rum . "
Suddenly , she noticed a lull aby playing in the distance
being careless . The sorrow ful little boy started to cry
being careless . The sorrow ful little boy started to cry
and returned to his rest ful spot beneath the tree with
, he remembered the sorrow ful look the older man had
and the lightning struck the hyd rant . Suddenly water was

MIN ACTIVATIONS
MIN = -0.166

They are powerful . <|endoftext|> One day , Tom and Lily
They are careful . <|endoftext|> One morning , a little ot
shoes are better . <|endoftext|> One day , a little girl
and to share . <|endoftext|> One day , a little boy
not bother anyone . <|endoftext|> One day , a boy named
They are lucky . <|endoftext|> One day , Tom and his
. They hoped . <|endoftext|> One day , a baby k
They were mighty . <|endoftext|> One day , a boy and
the needle was . <|endoftext|> One day , a boy named
y ummy too . <|endoftext|> One day , a woman went
she was smiling all <|endoftext|> One day , Sally was outside
might get hurt . <|endoftext|> One day , a little boy
protect their skin . <|endoftext|> One day , Lily and Tom
what you have . <|endoftext|> One day , a boy and
of their work . <|endoftext|> One day , Lily and Max
cake is inside . <|endoftext|> One day , a boy named
They are happy . <|endoftext|> One day , a boy named
other and waited . <|endoftext|> One day , a big storm
belong to them . <|endoftext|> One day , a little boy
love to paint . <|endoftext|> One day , a boy went

INTERVAL 0.107 - 0.198
CONTAINS 0.030%

a big smile on his face . <|endoftext|> Once there was
with a smile on her face . What an amazing adventure
stubborn . She throws a tant rum . She kicks and
and jump in the p udd les . One day ,
work . Ol iver smiled and be amed

INTERVAL 0.016 - 0.107
CONTAINS 25.485%

a few days , they got the book back and it
night , but eventually fell asleep . The monster came back
colors filled the room with joy . Cl ara
. It had a very important job - to power a
the university was coming true . <|endoftext|> Once there was a

INTERVAL -0.075 - 0.016
CONTAINS 73.664%

was making his thick toast in the kitchen . She put
inside . <|endoftext|> John and his Dad went to the park
buttons and levers did . He said she could help him
We will come back soon . Be a good boy ."
. Sara and Tom pretend the rug is a map of

INTERVAL -0.166 - -0.075
CONTAINS 0.821%

bear . You are very pretty . But I will make
, I want to go first !" Ben says . He
the time . One day , L ola was
One summer , things were different . Jack 's dad was
, she went to the zoo with her mom and dad
Token sad
Token ID6507
Feature activation+2.46e-02

Pos logit contributions
ends+0.008
 counts+0.008
 Or+0.007
One+0.007
 cannot+0.007
Neg logit contributions
 tune-0.010
 sensation-0.010
 dolphin-0.009
 shark-0.009
 Dolphin-0.009
Token.
Token ID13
Feature activation+1.30e-02

Pos logit contributions
 tune+0.063
 sensation+0.061
 dolphin+0.059
 shark+0.057
 Dolphin+0.057
Neg logit contributions
ends-0.023
 counts-0.022
One-0.021
 Or-0.021
 Maybe-0.016
Token She
Token ID1375
Feature activation-8.47e-03

Pos logit contributions
 tune+0.023
 sensation+0.023
 face+0.021
 Sister+0.020
 expression+0.019
Neg logit contributions
ends-0.020
 counts-0.020
One-0.018
 ours-0.018
 Or-0.017
Token threw
Token ID9617
Feature activation-6.18e-03

Pos logit contributions
 counts+0.012
ends+0.012
One+0.012
 Or+0.012
 has+0.010
Neg logit contributions
 sensation-0.017
 tune-0.017
 dolphin-0.015
 shark-0.015
 Dolphin-0.015
Token a
Token ID257
Feature activation+3.75e-03

Pos logit contributions
 counts+0.011
ends+0.011
One+0.010
 Or+0.010
 Ready+0.009
Neg logit contributions
 tune-0.011
 sensation-0.011
 dolphin-0.010
 shark-0.009
 Dolphin-0.009
Token tant
Token ID24246
Feature activation+1.98e-01

Pos logit contributions
 sensation+0.005
 tune+0.005
urg+0.004
 Dolphin+0.004
 dolphin+0.004
Neg logit contributions
ends-0.008
One-0.008
 counts-0.008
 Or-0.007
 Ready-0.007
Tokenrum
Token ID6582
Feature activation+2.16e-02

Pos logit contributions
 tune+0.320
 sensation+0.315
 dolphin+0.287
 shark+0.273
 chill+0.257
Neg logit contributions
 counts-0.363
ends-0.361
One-0.344
 Or-0.332
 Ready-0.299
Token and
Token ID290
Feature activation+3.24e-03

Pos logit contributions
 tune+0.045
 sensation+0.045
 dolphin+0.041
 shark+0.040
 chill+0.038
Neg logit contributions
ends-0.029
 counts-0.029
One-0.028
 Or-0.027
 Ready-0.024
Token cried
Token ID16896
Feature activation+6.15e-04

Pos logit contributions
 tune+0.007
 sensation+0.006
 dolphin+0.006
 shark+0.006
 Dolphin+0.006
Neg logit contributions
 counts-0.005
ends-0.005
 Or-0.004
One-0.004
 cannot-0.004
Token loudly
Token ID23112
Feature activation+1.63e-03

Pos logit contributions
 tune+0.002
 sensation+0.002
 dolphin+0.002
 shark+0.002
 Dolphin+0.001
Neg logit contributions
ends-0.001
 Or-0.001
One-0.001
 counts-0.000
house-0.000
Token.
Token ID13
Feature activation+8.73e-03

Pos logit contributions
 tune+0.004
 sensation+0.004
 dolphin+0.004
 shark+0.004
 Dolphin+0.004
Neg logit contributions
 counts-0.002
ends-0.002
One-0.001
 Or-0.001
 Ready-0.001
Token kept
Token ID4030
Feature activation+2.12e-02

Pos logit contributions
 counts+0.013
ends+0.012
One+0.012
 Or+0.011
+0.010
Neg logit contributions
 sensation-0.016
 tune-0.016
 dolphin-0.014
 shark-0.014
 Dolphin-0.014
Token crying
Token ID13774
Feature activation+4.84e-02

Pos logit contributions
 sensation+0.045
 tune+0.045
 dolphin+0.042
 shark+0.041
 Dolphin+0.040
Neg logit contributions
ends-0.026
 counts-0.025
One-0.025
 Or-0.022
 here-0.021
Token and
Token ID290
Feature activation-8.90e-03

Pos logit contributions
 tune+0.120
 sensation+0.117
 dolphin+0.114
 Dolphin+0.111
 shark+0.110
Neg logit contributions
ends-0.048
 counts-0.046
 Or-0.045
One-0.044
 Maybe-0.035
Token throwing
Token ID9644
Feature activation+9.41e-03

Pos logit contributions
 counts+0.012
ends+0.012
One+0.012
 Or+0.011
 cannot+0.010
Neg logit contributions
 tune-0.018
 dolphin-0.017
 sensation-0.017
 shark-0.017
 Dolphin-0.016
Token a
Token ID257
Feature activation-3.50e-03

Pos logit contributions
 tune+0.020
 sensation+0.019
 dolphin+0.018
 shark+0.017
 Dolphin+0.017
Neg logit contributions
 counts-0.013
ends-0.013
One-0.013
 Or-0.012
 Ready-0.010
Token tant
Token ID24246
Feature activation+1.97e-01

Pos logit contributions
One+0.007
ends+0.007
 counts+0.007
 Or+0.007
house+0.007
Neg logit contributions
 sensation-0.004
 tune-0.004
 Dolphin-0.004
 dolphin-0.004
 streaming-0.003
Tokenrum
Token ID6582
Feature activation+1.34e-02

Pos logit contributions
 tune+0.312
 sensation+0.306
 dolphin+0.277
 shark+0.265
 chill+0.248
Neg logit contributions
 counts-0.368
ends-0.367
One-0.353
 Or-0.340
 Ready-0.311
Token.
Token ID13
Feature activation+1.16e-02

Pos logit contributions
 tune+0.031
 sensation+0.030
 dolphin+0.028
 shark+0.027
 chill+0.026
Neg logit contributions
ends-0.016
 counts-0.016
One-0.015
 Or-0.014
 Ready-0.012
Token
Token ID198
Feature activation+1.03e-02

Pos logit contributions
 tune+0.023
 sensation+0.023
 dolphin+0.020
 streaming+0.019
 face+0.019
Neg logit contributions
 counts-0.017
ends-0.017
One-0.015
 Or-0.015
 ours-0.014
Token
Token ID198
Feature activation+2.06e-02

Pos logit contributions
 tune+0.019
 sensation+0.018
 dolphin+0.016
 face+0.016
 streamed+0.016
Neg logit contributions
 counts-0.016
ends-0.015
 ours-0.014
 Or-0.013
 Ready-0.013
TokenFinally
Token ID11158
Feature activation-2.70e-02

Pos logit contributions
 tune+0.038
 sensation+0.035
 face+0.031
 lull+0.031
 streamed+0.031
Neg logit contributions
 counts-0.036
ends-0.032
 ours-0.030
 Or-0.029
 Ready-0.028
Token to
Token ID284
Feature activation+1.41e-02

Pos logit contributions
 counts+0.003
ends+0.003
One+0.003
 Or+0.003
 Ready+0.002
Neg logit contributions
 sensation-0.008
 tune-0.008
 dolphin-0.008
 shark-0.008
 Dolphin-0.007
Token cry
Token ID3960
Feature activation+3.72e-02

Pos logit contributions
 sensation+0.027
 tune+0.027
 dolphin+0.026
 shark+0.025
 Dolphin+0.024
Neg logit contributions
 counts-0.021
One-0.020
 Or-0.019
ends-0.019
-0.017
Token and
Token ID290
Feature activation+5.77e-03

Pos logit contributions
 sensation+0.096
 tune+0.092
 dolphin+0.091
 shark+0.090
 Dolphin+0.090
Neg logit contributions
 Or-0.032
ends-0.032
One-0.029
 counts-0.027
 Maybe-0.024
Token threw
Token ID9617
Feature activation-1.54e-02

Pos logit contributions
 sensation+0.012
 tune+0.012
 dolphin+0.011
 shark+0.010
 Dolphin+0.010
Neg logit contributions
 counts-0.008
ends-0.008
 Or-0.008
One-0.008
 Ready-0.007
Token a
Token ID257
Feature activation-5.15e-04

Pos logit contributions
 counts+0.024
ends+0.024
One+0.023
 Or+0.022
 Ready+0.020
Neg logit contributions
 tune-0.029
 sensation-0.029
 dolphin-0.026
 shark-0.025
 streaming-0.024
Token tant
Token ID24246
Feature activation+1.96e-01

Pos logit contributions
ends+0.001
One+0.001
 Or+0.001
 counts+0.001
 Ready+0.001
Neg logit contributions
 sensation-0.001
 tune-0.001
urg-0.001
 Dolphin-0.001
lee-0.001
Tokenrum
Token ID6582
Feature activation+2.29e-02

Pos logit contributions
 tune+0.316
 sensation+0.311
 dolphin+0.283
 shark+0.270
 Dolphin+0.254
Neg logit contributions
 counts-0.358
ends-0.356
One-0.340
 Or-0.329
 Ready-0.295
Token.
Token ID13
Feature activation+2.41e-02

Pos logit contributions
 tune+0.053
 sensation+0.053
 dolphin+0.049
 shark+0.047
 chill+0.046
Neg logit contributions
ends-0.026
 counts-0.026
One-0.024
 Or-0.023
 Ready-0.020
Token She
Token ID1375
Feature activation-3.69e-03

Pos logit contributions
 tune+0.052
 sensation+0.052
 dolphin+0.048
 shark+0.047
 chill+0.045
Neg logit contributions
 counts-0.031
ends-0.031
One-0.029
 Or-0.028
 Ready-0.025
Token didn
Token ID1422
Feature activation+2.97e-02

Pos logit contributions
 counts+0.006
ends+0.005
One+0.005
 Or+0.005
 cannot+0.004
Neg logit contributions
 sensation-0.008
 tune-0.007
 dolphin-0.007
 Dolphin-0.006
andel-0.006
Token't
Token ID470
Feature activation-3.44e-02

Pos logit contributions
 sensation+0.059
 tune+0.059
 dolphin+0.055
 shark+0.054
 streaming+0.054
Neg logit contributions
 counts-0.039
ends-0.037
One-0.032
 Or-0.031
 ours-0.031
Token listen
Token ID6004
Feature activation+1.12e-02

Pos logit contributions
 tune+0.023
 sensation+0.023
 dolphin+0.021
 shark+0.020
 streaming+0.020
Neg logit contributions
 counts-0.012
ends-0.012
One-0.012
 Or-0.011
 Ready-0.010
Token.
Token ID13
Feature activation+2.62e-03

Pos logit contributions
 tune+0.022
 sensation+0.021
 dolphin+0.019
 shark+0.018
 face+0.018
Neg logit contributions
 counts-0.017
ends-0.017
One-0.016
 Or-0.015
 Ready-0.015
Token She
Token ID1375
Feature activation-1.96e-02

Pos logit contributions
 sensation+0.005
 tune+0.005
 face+0.005
 dolphin+0.005
 Sister+0.005
Neg logit contributions
ends-0.003
 counts-0.003
One-0.003
 ours-0.003
 belongs-0.003
Token threw
Token ID9617
Feature activation-1.25e-02

Pos logit contributions
 counts+0.030
ends+0.030
One+0.028
 Or+0.028
 Ready+0.024
Neg logit contributions
 tune-0.039
 sensation-0.038
 dolphin-0.034
 shark-0.033
 streaming-0.032
Token a
Token ID257
Feature activation+8.87e-04

Pos logit contributions
 counts+0.021
ends+0.020
One+0.020
 Or+0.019
 Ready+0.017
Neg logit contributions
 tune-0.023
 sensation-0.022
 dolphin-0.020
 shark-0.020
 chill-0.019
Token tant
Token ID24246
Feature activation+1.92e-01

Pos logit contributions
 sensation+0.001
 tune+0.001
 streaming+0.001
urg+0.001
lee+0.001
Neg logit contributions
ends-0.002
One-0.002
 counts-0.002
 Or-0.002
 Ready-0.002
Tokenrum
Token ID6582
Feature activation+1.13e-02

Pos logit contributions
 tune+0.310
 sensation+0.306
 dolphin+0.277
 shark+0.263
 chill+0.250
Neg logit contributions
 counts-0.352
ends-0.351
One-0.336
 Or-0.325
 Ready-0.292
Token in
Token ID287
Feature activation+2.17e-02

Pos logit contributions
 tune+0.022
 sensation+0.022
 dolphin+0.019
 shark+0.019
 streaming+0.019
Neg logit contributions
 counts-0.017
ends-0.017
One-0.016
 Or-0.015
 Ready-0.015
Token the
Token ID262
Feature activation+9.88e-03

Pos logit contributions
 tune+0.039
 sensation+0.038
 dolphin+0.035
 shark+0.034
 Dolphin+0.032
Neg logit contributions
 counts-0.037
One-0.035
ends-0.035
 Or-0.033
 Ready-0.031
Token store
Token ID3650
Feature activation-6.35e-02

Pos logit contributions
 tune+0.017
 sensation+0.017
ous+0.016
 streaming+0.016
urg+0.015
Neg logit contributions
One-0.015
 counts-0.015
 has-0.015
 Ready-0.014
ends-0.014
Token.
Token ID13
Feature activation-3.26e-03

Pos logit contributions
ends+0.088
 counts+0.088
One+0.085
 Or+0.079
 Ready+0.074
Neg logit contributions
 tune-0.132
 sensation-0.129
 dolphin-0.117
 shark-0.113
 face-0.111
Token She
Token ID1375
Feature activation-1.02e-02

Pos logit contributions
 tune+0.026
 sensation+0.025
 face+0.024
 Sister+0.023
 dolphin+0.022
Neg logit contributions
ends-0.016
 counts-0.016
One-0.015
 ours-0.014
 Or-0.013
Token cried
Token ID16896
Feature activation+8.97e-03

Pos logit contributions
 counts+0.015
ends+0.015
One+0.015
 Or+0.014
 Ready+0.012
Neg logit contributions
 sensation-0.021
 tune-0.020
 dolphin-0.018
 shark-0.018
 streaming-0.017
Token and
Token ID290
Feature activation+7.99e-03

Pos logit contributions
 sensation+0.024
 dolphin+0.024
 shark+0.023
 tune+0.023
 Dolphin+0.023
Neg logit contributions
ends-0.007
 Or-0.007
One-0.006
 counts-0.006
 or-0.004
Token threw
Token ID9617
Feature activation-2.01e-02

Pos logit contributions
 tune+0.019
andel+0.019
 sensation+0.019
 shark+0.019
 Dolphin+0.019
Neg logit contributions
 counts-0.007
ends-0.007
 Or-0.006
One-0.006
 she-0.005
Token a
Token ID257
Feature activation-5.01e-03

Pos logit contributions
 counts+0.030
ends+0.029
One+0.028
 Or+0.027
 Ready+0.025
Neg logit contributions
 tune-0.040
 sensation-0.039
 dolphin-0.036
 shark-0.035
 Dolphin-0.034
Token tant
Token ID24246
Feature activation+1.89e-01

Pos logit contributions
ends+0.010
One+0.010
 Or+0.010
 counts+0.009
 Ready+0.009
Neg logit contributions
 sensation-0.007
 tune-0.006
urg-0.006
 Ned-0.006
 Dolphin-0.006
Tokenrum
Token ID6582
Feature activation+7.50e-03

Pos logit contributions
 tune+0.305
 sensation+0.302
 dolphin+0.274
 shark+0.261
 chill+0.247
Neg logit contributions
 counts-0.345
ends-0.343
One-0.328
 Or-0.317
 Ready-0.284
Token.
Token ID13
Feature activation+4.56e-03

Pos logit contributions
 tune+0.018
 sensation+0.017
 dolphin+0.016
 shark+0.016
 chill+0.015
Neg logit contributions
ends-0.008
 counts-0.008
One-0.008
 Or-0.007
 Ready-0.007
Token Her
Token ID2332
Feature activation-1.71e-04

Pos logit contributions
 tune+0.009
 sensation+0.009
 dolphin+0.008
 shark+0.008
 chill+0.007
Neg logit contributions
ends-0.007
 counts-0.007
One-0.007
 Or-0.006
 Ready-0.006
Token mom
Token ID1995
Feature activation-9.99e-03

Pos logit contributions
 counts+0.000
ends+0.000
One+0.000
 Or+0.000
house+0.000
Neg logit contributions
 tune-0.000
 sensation-0.000
 dolphin-0.000
 Dolphin-0.000
 streaming-0.000
Token tried
Token ID3088
Feature activation-2.23e-02

Pos logit contributions
ends+0.014
 counts+0.014
One+0.013
 Or+0.013
house+0.012
Neg logit contributions
 sensation-0.021
 tune-0.020
 dolphin-0.019
 shark-0.018
 Dolphin-0.018
Token and
Token ID290
Feature activation+7.77e-03

Pos logit contributions
 tune+0.011
 sensation+0.010
 dolphin+0.010
 shark+0.009
 chill+0.009
Neg logit contributions
 counts-0.005
ends-0.005
One-0.005
 Or-0.005
 Ready-0.004
Token started
Token ID2067
Feature activation-2.60e-02

Pos logit contributions
 tune+0.016
 sensation+0.016
 dolphin+0.014
 Dolphin+0.014
 shark+0.014
Neg logit contributions
ends-0.011
 counts-0.011
 Or-0.010
One-0.009
 cannot-0.009
Token to
Token ID284
Feature activation+1.42e-03

Pos logit contributions
ends+0.025
 counts+0.024
One+0.024
 Or+0.021
 cannot+0.017
Neg logit contributions
 tune-0.064
 sensation-0.063
 dolphin-0.060
 shark-0.058
 Dolphin-0.058
Token throw
Token ID3714
Feature activation-1.57e-02

Pos logit contributions
 tune+0.003
 sensation+0.003
andel+0.003
 dolphin+0.003
 Dolphin+0.003
Neg logit contributions
 counts-0.002
One-0.002
 Or-0.002
ends-0.002
-0.001
Token a
Token ID257
Feature activation-1.03e-02

Pos logit contributions
 counts+0.025
ends+0.024
One+0.024
 Or+0.022
 Ready+0.020
Neg logit contributions
 tune-0.030
 sensation-0.029
 dolphin-0.027
 shark-0.026
 Dolphin-0.025
Token tant
Token ID24246
Feature activation+1.89e-01

Pos logit contributions
ends+0.019
One+0.018
 counts+0.018
 Or+0.017
 ours+0.016
Neg logit contributions
 sensation-0.016
 tune-0.016
 streaming-0.015
 streamed-0.014
 tunes-0.013
Tokenrum
Token ID6582
Feature activation+7.91e-03

Pos logit contributions
 tune+0.298
 sensation+0.293
 dolphin+0.263
 shark+0.251
 chill+0.237
Neg logit contributions
 counts-0.354
ends-0.353
One-0.339
 Or-0.328
 Ready-0.301
Token.
Token ID13
Feature activation+1.32e-02

Pos logit contributions
 tune+0.018
 sensation+0.018
 dolphin+0.016
 shark+0.016
 streaming+0.015
Neg logit contributions
 counts-0.009
ends-0.009
One-0.009
 Or-0.008
 Ready-0.008
Token Suddenly
Token ID24975
Feature activation-2.06e-02

Pos logit contributions
 tune+0.028
 sensation+0.027
 dolphin+0.025
 shark+0.024
 streaming+0.023
Neg logit contributions
ends-0.018
 counts-0.018
One-0.017
 Or-0.017
 Ready-0.015
Token,
Token ID11
Feature activation+3.07e-02

Pos logit contributions
ends+0.032
 counts+0.031
One+0.030
 Or+0.029
 Ready+0.027
Neg logit contributions
 tune-0.039
 sensation-0.038
 dolphin-0.035
 shark-0.034
 streaming-0.033
Token she
Token ID673
Feature activation+1.69e-02

Pos logit contributions
 tune+0.080
 sensation+0.077
 dolphin+0.072
 shark+0.070
 Dolphin+0.068
Neg logit contributions
ends-0.028
 counts-0.027
One-0.026
 Or-0.024
 Ready-0.021
Token her
Token ID607
Feature activation-1.29e-02

Pos logit contributions
 sensation+0.060
 tune+0.059
 dolphin+0.057
 shark+0.055
 Dolphin+0.054
Neg logit contributions
 counts-0.037
ends-0.034
 Or-0.032
One-0.031
 Ready-0.028
Token feet
Token ID3625
Feature activation+5.55e-02

Pos logit contributions
ends+0.023
 counts+0.022
 Or+0.020
 cannot+0.019
One+0.019
Neg logit contributions
 sensation-0.022
 tune-0.020
 dolphin-0.019
 Dolphin-0.018
 streamed-0.018
Token and
Token ID290
Feature activation-3.37e-03

Pos logit contributions
 tune+0.124
 sensation+0.122
 dolphin+0.115
 shark+0.112
 Dolphin+0.108
Neg logit contributions
ends-0.068
 counts-0.066
One-0.063
 Or-0.060
 Ready-0.052
Token threw
Token ID9617
Feature activation-2.23e-02

Pos logit contributions
 counts+0.005
ends+0.004
 Or+0.004
One+0.004
 cannot+0.004
Neg logit contributions
 tune-0.007
 sensation-0.007
 dolphin-0.006
 Dolphin-0.006
andel-0.006
Token a
Token ID257
Feature activation-4.38e-03

Pos logit contributions
 counts+0.034
ends+0.033
One+0.032
 Or+0.030
 Ready+0.028
Neg logit contributions
 tune-0.043
 sensation-0.043
 dolphin-0.039
 shark-0.038
 Dolphin-0.036
Token tant
Token ID24246
Feature activation+1.89e-01

Pos logit contributions
ends+0.011
One+0.010
 counts+0.010
 Or+0.010
 Ready+0.010
Neg logit contributions
 sensation-0.004
 tune-0.004
 streaming-0.003
 Dolphin-0.003
 dolphin-0.003
Tokenrum
Token ID6582
Feature activation+3.21e-03

Pos logit contributions
 tune+0.303
 sensation+0.299
 dolphin+0.271
 shark+0.258
 potato+0.243
Neg logit contributions
 counts-0.347
ends-0.345
One-0.330
 Or-0.319
 Ready-0.288
Token.
Token ID13
Feature activation+4.85e-03

Pos logit contributions
 tune+0.007
 sensation+0.007
 dolphin+0.007
 shark+0.007
 chill+0.006
Neg logit contributions
ends-0.004
 counts-0.004
One-0.003
 Or-0.003
 Ready-0.003
Token She
Token ID1375
Feature activation-1.19e-02

Pos logit contributions
 tune+0.011
 sensation+0.011
 dolphin+0.010
 shark+0.010
 chill+0.010
Neg logit contributions
ends-0.006
 counts-0.006
 Or-0.006
One-0.005
 Ready-0.005
Token grabbed
Token ID13176
Feature activation+2.43e-02

Pos logit contributions
 counts+0.017
ends+0.016
One+0.015
 Or+0.015
 cannot+0.014
Neg logit contributions
 sensation-0.025
 tune-0.025
 Dolphin-0.022
 dolphin-0.022
andel-0.022
Token the
Token ID262
Feature activation-1.52e-03

Pos logit contributions
 tune+0.052
 sensation+0.051
 dolphin+0.046
 shark+0.045
 lull+0.044
Neg logit contributions
 counts-0.033
ends-0.033
One-0.031
 Or-0.028
 Ready-0.027
Token.
Token ID13
Feature activation-7.26e-03

Pos logit contributions
 tune+0.023
 sensation+0.023
 dolphin+0.021
 shark+0.021
 streaming+0.020
Neg logit contributions
 counts-0.013
ends-0.013
One-0.012
 Or-0.011
 Ready-0.010
Token He
Token ID679
Feature activation-1.22e-02

Pos logit contributions
 counts+0.012
ends+0.012
One+0.011
 Or+0.011
 Ready+0.010
Neg logit contributions
 tune-0.014
 sensation-0.013
 dolphin-0.012
 shark-0.012
 streaming-0.011
Token kept
Token ID4030
Feature activation+1.89e-02

Pos logit contributions
 counts+0.018
ends+0.017
 Or+0.016
One+0.016
 has+0.014
Neg logit contributions
 tune-0.026
 sensation-0.025
 dolphin-0.023
 Dolphin-0.022
 shark-0.022
Token throwing
Token ID9644
Feature activation-1.21e-03

Pos logit contributions
 tune+0.040
 sensation+0.040
 dolphin+0.037
 lull+0.035
 shark+0.035
Neg logit contributions
ends-0.024
 counts-0.024
One-0.022
 Or-0.020
 ours-0.019
Token a
Token ID257
Feature activation-5.91e-03

Pos logit contributions
 counts+0.002
ends+0.002
One+0.002
 Or+0.002
 Ready+0.002
Neg logit contributions
 tune-0.002
 sensation-0.002
 dolphin-0.002
 shark-0.002
 streaming-0.002
Token tant
Token ID24246
Feature activation+1.88e-01

Pos logit contributions
ends+0.009
 counts+0.009
One+0.009
 Or+0.009
 Ready+0.008
Neg logit contributions
 sensation-0.011
 tune-0.010
 streaming-0.009
 dolphin-0.009
 Dolphin-0.009
Tokenrum
Token ID6582
Feature activation+2.82e-03

Pos logit contributions
 tune+0.293
 sensation+0.289
 dolphin+0.257
 shark+0.245
 streaming+0.231
Neg logit contributions
ends-0.356
 counts-0.356
One-0.342
 Or-0.331
 Ready-0.306
Token.
Token ID13
Feature activation-1.53e-02

Pos logit contributions
 sensation+0.006
 tune+0.006
 dolphin+0.005
 shark+0.005
 face+0.005
Neg logit contributions
 counts-0.004
ends-0.004
One-0.004
 Or-0.003
 Ready-0.003
Token His
Token ID2399
Feature activation+6.34e-03

Pos logit contributions
ends+0.026
 counts+0.025
One+0.024
 Or+0.023
 Ready+0.022
Neg logit contributions
 tune-0.027
 sensation-0.027
 dolphin-0.024
 shark-0.023
 streaming-0.022
Token mom
Token ID1995
Feature activation-2.04e-02

Pos logit contributions
 tune+0.012
 sensation+0.012
 dolphin+0.011
 streaming+0.011
 Dolphin+0.010
Neg logit contributions
ends-0.009
 counts-0.009
 Or-0.009
One-0.008
 cannot-0.008
Token was
Token ID373
Feature activation-2.11e-02

Pos logit contributions
 counts+0.030
ends+0.030
 Or+0.028
One+0.028
 Ready+0.024
Neg logit contributions
 tune-0.041
 sensation-0.041
 dolphin-0.037
 shark-0.036
 Dolphin-0.035
Token and
Token ID290
Feature activation-1.84e-02

Pos logit contributions
 counts+0.007
ends+0.007
One+0.007
 Or+0.006
 Ready+0.005
Neg logit contributions
 tune-0.014
 sensation-0.014
 dolphin-0.013
 shark-0.013
 chill-0.012
Token so
Token ID523
Feature activation+2.04e-04

Pos logit contributions
 counts+0.027
ends+0.027
One+0.026
 Or+0.024
 cannot+0.023
Neg logit contributions
 tune-0.037
 sensation-0.036
 dolphin-0.033
 shark-0.033
 Dolphin-0.031
Token she
Token ID673
Feature activation-1.64e-02

Pos logit contributions
 tune+0.001
 sensation+0.001
 dolphin+0.000
 shark+0.000
 potato+0.000
Neg logit contributions
 counts-0.000
ends-0.000
One-0.000
 Or-0.000
 Ready-0.000
Token threw
Token ID9617
Feature activation-1.77e-02

Pos logit contributions
One+0.022
 counts+0.021
ends+0.020
 Or+0.019
+0.018
Neg logit contributions
 sensation-0.036
 tune-0.034
andel-0.030
ulp-0.030
 streaming-0.030
Token a
Token ID257
Feature activation-4.02e-03

Pos logit contributions
 counts+0.026
ends+0.026
One+0.025
 Or+0.023
 Ready+0.021
Neg logit contributions
 tune-0.036
 sensation-0.035
 dolphin-0.032
 shark-0.031
 streaming-0.030
Token tant
Token ID24246
Feature activation+1.86e-01

Pos logit contributions
ends+0.005
One+0.005
 Or+0.005
 counts+0.005
 ours+0.005
Neg logit contributions
 sensation-0.008
 tune-0.007
urg-0.007
 streaming-0.007
lee-0.007
Tokenrum
Token ID6582
Feature activation+1.30e-02

Pos logit contributions
 tune+0.304
 sensation+0.302
 dolphin+0.274
 shark+0.260
 chill+0.247
Neg logit contributions
 counts-0.337
ends-0.334
One-0.318
 Or-0.306
 Ready-0.271
Token when
Token ID618
Feature activation+1.20e-02

Pos logit contributions
 tune+0.029
 sensation+0.029
 dolphin+0.027
 shark+0.026
 chill+0.025
Neg logit contributions
ends-0.016
 counts-0.016
One-0.015
 Or-0.014
 Ready-0.012
Token her
Token ID607
Feature activation-5.82e-03

Pos logit contributions
 tune+0.027
 sensation+0.027
 dolphin+0.025
 shark+0.024
 streaming+0.024
Neg logit contributions
ends-0.014
 counts-0.014
One-0.013
 Or-0.013
 Ready-0.011
Token parents
Token ID3397
Feature activation+4.15e-02

Pos logit contributions
ends+0.007
 counts+0.007
 Or+0.007
One+0.007
 cannot+0.007
Neg logit contributions
 sensation-0.012
 tune-0.012
 lull-0.010
hing-0.010
 streaming-0.010
Token said
Token ID531
Feature activation-3.75e-02

Pos logit contributions
 sensation+0.089
 tune+0.089
 dolphin+0.077
 shark+0.076
 lull+0.073
Neg logit contributions
ends-0.056
 counts-0.056
One-0.053
 Or-0.050
-0.045
Token surprise
Token ID5975
Feature activation+2.12e-03

Pos logit contributions
ends+0.015
 counts+0.014
One+0.014
 Or+0.013
 Ready+0.012
Neg logit contributions
 tune-0.014
 sensation-0.014
 dolphin-0.013
 shark-0.012
 potato-0.012
Token.
Token ID13
Feature activation-2.03e-02

Pos logit contributions
 tune+0.005
 sensation+0.005
 dolphin+0.004
 shark+0.004
 face+0.004
Neg logit contributions
ends-0.003
 counts-0.003
One-0.003
 Or-0.002
 Ready-0.002
Token He
Token ID679
Feature activation-1.38e-02

Pos logit contributions
ends+0.038
 counts+0.038
One+0.036
 ours+0.033
 Or+0.033
Neg logit contributions
 sensation-0.030
 tune-0.029
 shark-0.026
 face-0.025
 dolphin-0.025
Token threw
Token ID9617
Feature activation-1.50e-02

Pos logit contributions
 counts+0.021
ends+0.020
One+0.020
 Or+0.019
 Ready+0.017
Neg logit contributions
 tune-0.028
 sensation-0.027
 dolphin-0.024
 shark-0.023
 streaming-0.023
Token a
Token ID257
Feature activation+6.48e-04

Pos logit contributions
 counts+0.026
ends+0.026
One+0.025
 Or+0.024
 Ready+0.022
Neg logit contributions
 tune-0.026
 sensation-0.026
 dolphin-0.023
 shark-0.022
 streaming-0.021
Token tant
Token ID24246
Feature activation+1.82e-01

Pos logit contributions
 sensation+0.001
 tune+0.001
 streaming+0.001
 Dolphin+0.001
 dolphin+0.001
Neg logit contributions
ends-0.001
One-0.001
 counts-0.001
 Or-0.001
 Ready-0.001
Tokenrum
Token ID6582
Feature activation+3.95e-03

Pos logit contributions
 tune+0.279
 sensation+0.274
 dolphin+0.242
 shark+0.231
 streaming+0.218
Neg logit contributions
ends-0.350
 counts-0.350
One-0.337
 Or-0.325
 Ready-0.304
Token.
Token ID13
Feature activation-2.09e-02

Pos logit contributions
 tune+0.008
 sensation+0.008
 dolphin+0.008
 shark+0.007
 streaming+0.007
Neg logit contributions
 counts-0.005
ends-0.005
One-0.005
 Or-0.005
 Ready-0.004
Token He
Token ID679
Feature activation-1.13e-02

Pos logit contributions
ends+0.034
 counts+0.034
One+0.032
 Or+0.031
 Ready+0.029
Neg logit contributions
 tune-0.039
 sensation-0.038
 dolphin-0.035
 shark-0.034
 streaming-0.032
Token kicked
Token ID12165
Feature activation+2.84e-03

Pos logit contributions
 counts+0.016
ends+0.016
One+0.015
 Or+0.014
 cannot+0.013
Neg logit contributions
 tune-0.024
 sensation-0.023
 dolphin-0.021
 Dolphin-0.020
 streaming-0.020
Token and
Token ID290
Feature activation+8.85e-03

Pos logit contributions
 tune+0.005
 sensation+0.005
 dolphin+0.004
 shark+0.004
 streaming+0.004
Neg logit contributions
 counts-0.005
ends-0.005
One-0.005
 Or-0.005
 Ready-0.004
Token stubborn
Token ID24444
Feature activation-1.30e-02

Pos logit contributions
ends+0.045
 counts+0.045
One+0.043
 Or+0.041
 Ready+0.038
Neg logit contributions
 tune-0.056
 sensation-0.056
 dolphin-0.050
 shark-0.049
 streaming-0.047
Token.
Token ID13
Feature activation-2.98e-02

Pos logit contributions
ends+0.018
 counts+0.017
One+0.017
 Or+0.016
 Ready+0.016
Neg logit contributions
 tune-0.027
 sensation-0.027
 dolphin-0.024
 shark-0.023
 face-0.023
Token She
Token ID1375
Feature activation-3.48e-02

Pos logit contributions
ends+0.034
 counts+0.033
One+0.031
 ours+0.027
 Or+0.026
Neg logit contributions
 sensation-0.067
 tune-0.067
 dolphin-0.061
 face-0.059
 shark-0.059
Token throws
Token ID12542
Feature activation-1.26e-02

Pos logit contributions
ends+0.043
One+0.043
 counts+0.039
 Or+0.038
 ours+0.037
Neg logit contributions
 sensation-0.069
 tune-0.067
 face-0.066
 dolphin-0.065
 shark-0.064
Token a
Token ID257
Feature activation+1.09e-04

Pos logit contributions
 counts+0.020
ends+0.020
One+0.019
 Or+0.018
 Ready+0.017
Neg logit contributions
 tune-0.024
 sensation-0.023
 dolphin-0.021
 shark-0.020
 streaming-0.019
Token tant
Token ID24246
Feature activation+1.68e-01

Pos logit contributions
 sensation+0.000
 tune+0.000
 streaming+0.000
 dolphin+0.000
 Dolphin+0.000
Neg logit contributions
ends-0.000
One-0.000
 counts-0.000
 Or-0.000
 Ready-0.000
Tokenrum
Token ID6582
Feature activation-8.92e-03

Pos logit contributions
 tune+0.256
 sensation+0.250
 dolphin+0.221
 shark+0.211
 streaming+0.199
Neg logit contributions
ends-0.327
 counts-0.326
One-0.314
 Or-0.304
 Ready-0.286
Token.
Token ID13
Feature activation-3.18e-02

Pos logit contributions
 counts+0.013
One+0.013
ends+0.013
 Ready+0.012
 Or+0.012
Neg logit contributions
 sensation-0.017
 tune-0.017
 dolphin-0.015
 face-0.015
 streaming-0.015
Token She
Token ID1375
Feature activation-3.37e-02

Pos logit contributions
ends+0.036
 counts+0.036
One+0.033
 Or+0.029
 ours+0.028
Neg logit contributions
 sensation-0.072
 tune-0.072
 dolphin-0.066
 shark-0.064
 face-0.063
Token kicks
Token ID17364
Feature activation-3.17e-03

Pos logit contributions
ends+0.043
One+0.043
 counts+0.039
 Or+0.038
 Ready+0.036
Neg logit contributions
 sensation-0.068
 tune-0.067
 shark-0.065
 dolphin-0.065
 face-0.063
Token and
Token ID290
Feature activation+9.20e-03

Pos logit contributions
 counts+0.005
ends+0.005
One+0.005
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.006
 sensation-0.006
 dolphin-0.005
 shark-0.005
 streaming-0.005
Token vegetables
Token ID13701
Feature activation+3.81e-02

Pos logit contributions
 tune+0.035
 sensation+0.035
 dolphin+0.030
 shark+0.029
 streaming+0.029
Neg logit contributions
 counts-0.038
ends-0.038
One-0.037
 Or-0.036
 Ready-0.033
Token laughed
Token ID13818
Feature activation+2.12e-02

Pos logit contributions
 tune+0.072
 sensation+0.070
 dolphin+0.064
 shark+0.061
 streaming+0.059
Neg logit contributions
 counts-0.060
ends-0.060
One-0.057
 Or-0.055
 Ready-0.050
Token at
Token ID379
Feature activation-6.02e-03

Pos logit contributions
 sensation+0.049
 tune+0.049
 dolphin+0.045
 shark+0.044
 chill+0.043
Neg logit contributions
ends-0.024
 counts-0.023
One-0.021
 Or-0.021
 Ready-0.017
Token the
Token ID262
Feature activation-2.78e-03

Pos logit contributions
 counts+0.008
One+0.008
ends+0.008
 Or+0.008
 Ready+0.007
Neg logit contributions
 tune-0.012
 sensation-0.012
 chill-0.010
 dolphin-0.010
 shark-0.010
Token boast
Token ID28020
Feature activation+6.47e-02

Pos logit contributions
 counts+0.005
One+0.005
ends+0.005
 Or+0.004
 Ready+0.004
Neg logit contributions
 tune-0.004
 sensation-0.004
 streaming-0.004
 streamed-0.004
 dolphin-0.004
Tokenful
Token ID913
Feature activation+1.68e-01

Pos logit contributions
 tune+0.106
 sensation+0.104
 dolphin+0.095
 shark+0.094
 chill+0.092
Neg logit contributions
ends-0.114
 counts-0.108
One-0.104
 Or-0.103
one-0.090
Token onion
Token ID21670
Feature activation+6.84e-02

Pos logit contributions
 sensation+0.308
 tune+0.303
 streaming+0.277
 streamed+0.271
 dolphin+0.270
Neg logit contributions
 counts-0.253
ends-0.239
One-0.239
 Or-0.232
 has-0.207
Token.
Token ID13
Feature activation+7.01e-03

Pos logit contributions
 tune+0.158
 sensation+0.158
 dolphin+0.147
 shark+0.144
 Dolphin+0.143
Neg logit contributions
 counts-0.071
ends-0.066
 Or-0.063
One-0.062
house-0.053
Token One
Token ID1881
Feature activation-1.00e-01

Pos logit contributions
 tune+0.013
 sensation+0.013
 dolphin+0.012
 shark+0.011
 Dolphin+0.011
Neg logit contributions
ends-0.011
 counts-0.011
One-0.010
 Or-0.010
 ours-0.009
Token of
Token ID286
Feature activation+8.25e-02

Pos logit contributions
 counts+0.153
One+0.151
ends+0.136
 Or+0.132
+0.132
Neg logit contributions
 tune-0.187
 sensation-0.176
 streaming-0.164
 dolphin-0.159
 streamed-0.156
Token them
Token ID606
Feature activation+4.14e-02

Pos logit contributions
 tune+0.165
 sensation+0.157
 chill+0.138
 dolphin+0.136
 shark+0.135
Neg logit contributions
One-0.127
 counts-0.116
ends-0.108
 Or-0.100
-0.096
Token out
Token ID503
Feature activation+3.38e-02

Pos logit contributions
ends+0.016
 counts+0.015
 Or+0.014
One+0.013
 Ready+0.013
Neg logit contributions
 sensation-0.026
 tune-0.025
 dolphin-0.023
 shark-0.022
 chill-0.022
Token loud
Token ID7812
Feature activation+3.84e-02

Pos logit contributions
 sensation+0.079
 tune+0.076
 dolphin+0.072
 Sister+0.071
 Dolphin+0.071
Neg logit contributions
ends-0.034
One-0.034
 counts-0.033
 Or-0.032
-0.028
Token and
Token ID290
Feature activation+1.67e-02

Pos logit contributions
 sensation+0.087
 tune+0.084
 dolphin+0.078
 streamed+0.077
 chill+0.076
Neg logit contributions
 counts-0.043
ends-0.037
 Or-0.034
One-0.031
 has-0.030
Token enjoyed
Token ID8359
Feature activation+2.53e-02

Pos logit contributions
 sensation+0.032
 tune+0.031
 dolphin+0.028
 shark+0.028
 potato+0.027
Neg logit contributions
 counts-0.026
ends-0.026
One-0.024
 Or-0.023
 Ready-0.022
Token the
Token ID262
Feature activation+7.03e-03

Pos logit contributions
 tune+0.047
 sensation+0.046
 dolphin+0.042
 shark+0.040
 streamed+0.040
Neg logit contributions
ends-0.040
 counts-0.039
One-0.038
 Or-0.034
 Ready-0.033
Token sensation
Token ID18098
Feature activation+1.57e-01

Pos logit contributions
 tune+0.013
 sensation+0.013
 dolphin+0.012
 shark+0.011
 chill+0.011
Neg logit contributions
ends-0.011
 counts-0.011
One-0.011
 Or-0.010
 Ready-0.009
Token of
Token ID286
Feature activation+5.22e-02

Pos logit contributions
 tune+0.334
 sensation+0.314
 dolphin+0.312
 Dolphin+0.298
 shark+0.298
Neg logit contributions
 counts-0.200
ends-0.191
One-0.181
 Or-0.165
-0.162
Token being
Token ID852
Feature activation+6.64e-02

Pos logit contributions
 tune+0.099
 sensation+0.093
 dolphin+0.090
 shark+0.086
 Dolphin+0.085
Neg logit contributions
 counts-0.077
One-0.074
ends-0.071
 ours-0.067
 Ready-0.066
Token so
Token ID523
Feature activation-3.98e-03

Pos logit contributions
 tune+0.130
 sensation+0.129
 dolphin+0.118
 shark+0.115
 potato+0.109
Neg logit contributions
ends-0.101
 counts-0.099
One-0.092
 Or-0.086
 Ready-0.080
Token high
Token ID1029
Feature activation-2.12e-02

Pos logit contributions
ends+0.006
 counts+0.006
One+0.006
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.008
 sensation-0.008
 dolphin-0.007
 shark-0.007
 potato-0.007
Token up
Token ID510
Feature activation-4.47e-02

Pos logit contributions
 counts+0.026
ends+0.025
One+0.024
 Or+0.022
 Ready+0.020
Neg logit contributions
 tune-0.047
 sensation-0.046
 dolphin-0.043
 shark-0.041
 Dolphin-0.040
Token his
Token ID465
Feature activation-2.74e-03

Pos logit contributions
 tune+0.082
 sensation+0.078
 chill+0.070
 lull+0.069
 dolphin+0.069
Neg logit contributions
 counts-0.069
One-0.069
ends-0.067
 Or-0.064
"-0.060
Token feet
Token ID3625
Feature activation+5.00e-02

Pos logit contributions
ends+0.005
 counts+0.005
 Or+0.005
One+0.005
 ours+0.005
Neg logit contributions
 sensation-0.004
 tune-0.004
 dolphin-0.003
 streamed-0.003
 streaming-0.003
Token and
Token ID290
Feature activation-1.75e-02

Pos logit contributions
 tune+0.101
 sensation+0.098
 dolphin+0.088
 shark+0.085
 streaming+0.084
Neg logit contributions
 counts-0.074
ends-0.072
One-0.068
 Or-0.066
 Ready-0.062
Token threw
Token ID9617
Feature activation-2.61e-02

Pos logit contributions
 counts+0.026
ends+0.025
 Or+0.022
One+0.021
 has+0.020
Neg logit contributions
 tune-0.035
 sensation-0.034
andel-0.031
 dolphin-0.031
 Dolphin-0.030
Token a
Token ID257
Feature activation-1.20e-02

Pos logit contributions
 counts+0.044
ends+0.043
One+0.043
 Or+0.040
 Ready+0.038
Neg logit contributions
 tune-0.046
 sensation-0.045
 dolphin-0.041
 shark-0.038
 Dolphin-0.037
Token tant
Token ID24246
Feature activation+1.53e-01

Pos logit contributions
ends+0.019
One+0.018
 counts+0.018
 Or+0.018
 Ready+0.016
Neg logit contributions
 sensation-0.021
 tune-0.021
 Dolphin-0.018
 streaming-0.018
 dolphin-0.018
Tokenrum
Token ID6582
Feature activation-1.73e-02

Pos logit contributions
 tune+0.237
 sensation+0.232
 dolphin+0.206
 shark+0.197
 streaming+0.185
Neg logit contributions
ends-0.292
 counts-0.292
One-0.281
 Or-0.271
 Ready-0.253
Token.
Token ID13
Feature activation-2.60e-02

Pos logit contributions
ends+0.020
 counts+0.019
One+0.018
 Or+0.017
 Ready+0.015
Neg logit contributions
 tune-0.040
 sensation-0.040
 dolphin-0.037
 shark-0.036
 chill-0.034
Token
Token ID198
Feature activation-1.21e-02

Pos logit contributions
ends+0.032
 counts+0.031
One+0.031
 Or+0.030
 Ready+0.026
Neg logit contributions
 tune-0.060
 sensation-0.058
 dolphin-0.053
 shark-0.051
 streaming-0.051
Token
Token ID198
Feature activation+9.75e-05

Pos logit contributions
 counts+0.020
ends+0.018
 ours+0.016
 belongs+0.016
 Or+0.016
Neg logit contributions
 tune-0.021
 sensation-0.020
 dolphin-0.020
 shark-0.019
 Sister-0.018
Token"
Token ID1
Feature activation-4.54e-02

Pos logit contributions
 tune+0.000
 sensation+0.000
 dolphin+0.000
 shark+0.000
 Sister+0.000
Neg logit contributions
 counts-0.000
ends-0.000
 belongs-0.000
 ours-0.000
 Or-0.000
TokenSuddenly
Token ID38582
Feature activation-1.17e-02

Pos logit contributions
 tune+0.046
 sensation+0.043
 dolphin+0.039
 lull+0.037
 Dolphin+0.035
Neg logit contributions
 counts-0.046
ends-0.041
 belongs-0.038
 ours-0.038
 Or-0.037
Token,
Token ID11
Feature activation+3.98e-02

Pos logit contributions
 counts+0.016
ends+0.015
One+0.015
 Or+0.014
 Ready+0.013
Neg logit contributions
 tune-0.025
 sensation-0.025
 dolphin-0.023
 shark-0.022
 streaming-0.021
Token she
Token ID673
Feature activation+2.23e-02

Pos logit contributions
 tune+0.088
 sensation+0.085
 dolphin+0.077
 shark+0.075
 streaming+0.073
Neg logit contributions
 counts-0.050
ends-0.050
One-0.048
 Or-0.045
 Ready-0.042
Token noticed
Token ID6810
Feature activation+4.58e-02

Pos logit contributions
 tune+0.041
 sensation+0.040
 dolphin+0.036
 shark+0.035
 streaming+0.033
Neg logit contributions
 counts-0.037
ends-0.037
One-0.035
 Or-0.034
 Ready-0.031
Token a
Token ID257
Feature activation+1.10e-02

Pos logit contributions
 tune+0.107
 sensation+0.103
 dolphin+0.094
 shark+0.093
 Dolphin+0.092
Neg logit contributions
 counts-0.050
One-0.049
ends-0.048
 Or-0.044
 Ready-0.039
Token lull
Token ID42858
Feature activation+1.53e-01

Pos logit contributions
 sensation+0.020
 tune+0.019
 streamed+0.018
 streaming+0.018
ened+0.017
Neg logit contributions
One-0.017
 Or-0.015
ends-0.015
 counts-0.015
 Ready-0.015
Tokenaby
Token ID3930
Feature activation+6.27e-02

Pos logit contributions
 tant+0.263
 tune+0.254
 lobster+0.252
 sensation+0.249
 gul+0.237
Neg logit contributions
ends-0.198
house-0.196
"-0.177
 Or-0.169
ider-0.166
Token playing
Token ID2712
Feature activation+3.94e-02

Pos logit contributions
 sensation+0.143
 tune+0.142
 dolphin+0.131
 shark+0.130
 Dolphin+0.125
Neg logit contributions
 counts-0.068
ends-0.067
One-0.061
 Or-0.059
-0.055
Token in
Token ID287
Feature activation+1.99e-02

Pos logit contributions
 tune+0.086
 sensation+0.083
 dolphin+0.082
 Dolphin+0.079
 shark+0.079
Neg logit contributions
 counts-0.045
ends-0.043
One-0.042
 Or-0.041
house-0.037
Token the
Token ID262
Feature activation+1.88e-02

Pos logit contributions
 tune+0.039
 sensation+0.037
 dolphin+0.035
 shark+0.034
 Dolphin+0.033
Neg logit contributions
 counts-0.030
One-0.030
ends-0.028
 Or-0.027
 Ready-0.025
Token distance
Token ID5253
Feature activation+3.59e-02

Pos logit contributions
 tune+0.034
 sensation+0.034
 dolphin+0.031
 shark+0.030
 streaming+0.029
Neg logit contributions
One-0.030
 counts-0.029
ends-0.027
 Or-0.026
 Ready-0.026
Token being
Token ID852
Feature activation+1.80e-02

Pos logit contributions
 sensation+0.040
 tune+0.039
 shark+0.038
 dolphin+0.037
 Dolphin+0.035
Neg logit contributions
 counts-0.027
One-0.027
 Or-0.025
ends-0.025
 ours-0.022
Token careless
Token ID36138
Feature activation+1.86e-03

Pos logit contributions
 tune+0.037
 sensation+0.036
 dolphin+0.033
 shark+0.032
 streaming+0.030
Neg logit contributions
ends-0.026
 counts-0.026
One-0.024
 Or-0.023
 Ready-0.021
Token.
Token ID13
Feature activation-7.91e-04

Pos logit contributions
 tune+0.004
 sensation+0.004
 dolphin+0.004
 shark+0.004
 streaming+0.004
Neg logit contributions
 counts-0.002
ends-0.002
One-0.002
 Or-0.002
 Ready-0.002
Token The
Token ID383
Feature activation-4.18e-04

Pos logit contributions
ends+0.001
 counts+0.001
One+0.001
 ours+0.001
 Or+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 shark-0.001
 Dolphin-0.001
Token sorrow
Token ID24140
Feature activation+9.79e-02

Pos logit contributions
 counts+0.001
One+0.001
ends+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 streaming-0.001
 streamed-0.001
Tokenful
Token ID913
Feature activation+1.53e-01

Pos logit contributions
 tune+0.217
 sensation+0.214
 dolphin+0.193
 shark+0.187
 Dolphin+0.185
Neg logit contributions
ends-0.122
 counts-0.119
 Or-0.107
 has-0.103
One-0.103
Token little
Token ID1310
Feature activation+3.63e-02

Pos logit contributions
 sensation+0.252
 tune+0.251
 dolphin+0.224
 streaming+0.219
 Dolphin+0.213
Neg logit contributions
 counts-0.256
One-0.241
 Or-0.240
ends-0.240
 has-0.221
Token boy
Token ID2933
Feature activation+3.09e-02

Pos logit contributions
 tune+0.057
 sensation+0.056
 dolphin+0.049
 streaming+0.047
 streamed+0.046
Neg logit contributions
 counts-0.066
One-0.062
ends-0.062
 Or-0.061
 Ready-0.057
Token started
Token ID2067
Feature activation-1.15e-02

Pos logit contributions
 sensation+0.064
 tune+0.064
 dolphin+0.059
 Dolphin+0.056
 shark+0.056
Neg logit contributions
 counts-0.043
ends-0.041
 Or-0.039
One-0.038
 has-0.036
Token to
Token ID284
Feature activation+6.23e-03

Pos logit contributions
 counts+0.012
ends+0.012
One+0.011
 Or+0.011
+0.010
Neg logit contributions
 sensation-0.026
 tune-0.026
 dolphin-0.025
 shark-0.024
 Dolphin-0.024
Token cry
Token ID3960
Feature activation+3.30e-02

Pos logit contributions
 tune+0.011
 sensation+0.011
 dolphin+0.011
 shark+0.010
 Dolphin+0.010
Neg logit contributions
 counts-0.009
One-0.009
ends-0.009
 Or-0.009
-0.008
Token being
Token ID852
Feature activation+1.80e-02

Pos logit contributions
 sensation+0.040
 tune+0.039
 shark+0.038
 dolphin+0.037
 Dolphin+0.035
Neg logit contributions
 counts-0.027
One-0.027
 Or-0.025
ends-0.025
 ours-0.022
Token careless
Token ID36138
Feature activation+1.86e-03

Pos logit contributions
 tune+0.037
 sensation+0.036
 dolphin+0.033
 shark+0.032
 streaming+0.030
Neg logit contributions
ends-0.026
 counts-0.026
One-0.024
 Or-0.023
 Ready-0.021
Token.
Token ID13
Feature activation-7.91e-04

Pos logit contributions
 tune+0.004
 sensation+0.004
 dolphin+0.004
 shark+0.004
 streaming+0.004
Neg logit contributions
 counts-0.002
ends-0.002
One-0.002
 Or-0.002
 Ready-0.002
Token The
Token ID383
Feature activation-4.18e-04

Pos logit contributions
ends+0.001
 counts+0.001
One+0.001
 ours+0.001
 Or+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 shark-0.001
 Dolphin-0.001
Token sorrow
Token ID24140
Feature activation+9.79e-02

Pos logit contributions
 counts+0.001
One+0.001
ends+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 streaming-0.001
 streamed-0.001
Tokenful
Token ID913
Feature activation+1.53e-01

Pos logit contributions
 tune+0.217
 sensation+0.214
 dolphin+0.193
 shark+0.187
 Dolphin+0.185
Neg logit contributions
ends-0.122
 counts-0.119
 Or-0.107
 has-0.103
One-0.103
Token little
Token ID1310
Feature activation+3.63e-02

Pos logit contributions
 sensation+0.252
 tune+0.251
 dolphin+0.224
 streaming+0.219
 Dolphin+0.213
Neg logit contributions
 counts-0.256
One-0.241
 Or-0.240
ends-0.240
 has-0.221
Token boy
Token ID2933
Feature activation+3.09e-02

Pos logit contributions
 tune+0.057
 sensation+0.056
 dolphin+0.049
 streaming+0.047
 streamed+0.046
Neg logit contributions
 counts-0.066
One-0.062
ends-0.062
 Or-0.061
 Ready-0.057
Token started
Token ID2067
Feature activation-1.15e-02

Pos logit contributions
 sensation+0.064
 tune+0.064
 dolphin+0.059
 Dolphin+0.056
 shark+0.056
Neg logit contributions
 counts-0.043
ends-0.041
 Or-0.039
One-0.038
 has-0.036
Token to
Token ID284
Feature activation+6.23e-03

Pos logit contributions
 counts+0.012
ends+0.012
One+0.011
 Or+0.011
+0.010
Neg logit contributions
 sensation-0.026
 tune-0.026
 dolphin-0.025
 shark-0.024
 Dolphin-0.024
Token cry
Token ID3960
Feature activation+3.30e-02

Pos logit contributions
 tune+0.011
 sensation+0.011
 dolphin+0.011
 shark+0.010
 Dolphin+0.010
Neg logit contributions
 counts-0.009
One-0.009
ends-0.009
 Or-0.009
-0.008
Token and
Token ID290
Feature activation+3.74e-03

Pos logit contributions
 counts+0.006
ends+0.006
One+0.005
 Or+0.005
 Ready+0.004
Neg logit contributions
 tune-0.012
 sensation-0.011
 dolphin-0.010
 shark-0.010
 streamed-0.010
Token returned
Token ID4504
Feature activation-2.16e-02

Pos logit contributions
 tune+0.007
 sensation+0.007
 dolphin+0.006
 shark+0.006
 streaming+0.006
Neg logit contributions
 counts-0.006
ends-0.006
One-0.005
 Or-0.005
 Ready-0.005
Token to
Token ID284
Feature activation+3.03e-02

Pos logit contributions
 counts+0.033
ends+0.032
One+0.031
 Or+0.028
 Ready+0.027
Neg logit contributions
 tune-0.042
 sensation-0.042
 dolphin-0.037
 shark-0.036
 streamed-0.035
Token his
Token ID465
Feature activation+2.52e-02

Pos logit contributions
 sensation+0.051
 tune+0.051
 streamed+0.045
 dolphin+0.045
 streaming+0.044
Neg logit contributions
 counts-0.053
One-0.051
ends-0.050
 Ready-0.047
 Or-0.046
Token rest
Token ID1334
Feature activation+8.01e-02

Pos logit contributions
 tune+0.043
 sensation+0.043
 dolphin+0.038
 shark+0.036
 streaming+0.036
Neg logit contributions
 counts-0.044
ends-0.043
One-0.042
 Or-0.040
 Ready-0.038
Tokenful
Token ID913
Feature activation+1.51e-01

Pos logit contributions
 tune+0.192
 sensation+0.180
 dolphin+0.176
 shark+0.171
 Dolphin+0.169
Neg logit contributions
ends-0.080
 counts-0.077
One-0.072
 Or-0.061
house-0.057
Token spot
Token ID4136
Feature activation-5.53e-03

Pos logit contributions
 tune+0.282
 sensation+0.281
 dolphin+0.260
 shark+0.248
 Dolphin+0.244
Neg logit contributions
 counts-0.234
One-0.221
ends-0.217
 Or-0.205
house-0.193
Token beneath
Token ID11061
Feature activation+2.75e-02

Pos logit contributions
ends+0.007
 counts+0.006
One+0.006
 Or+0.006
house+0.005
Neg logit contributions
 tune-0.012
 sensation-0.012
 dolphin-0.011
 shark-0.011
 Dolphin-0.011
Token the
Token ID262
Feature activation+2.83e-02

Pos logit contributions
 tune+0.051
 sensation+0.050
 dolphin+0.046
 chill+0.043
 shark+0.042
Neg logit contributions
One-0.041
 counts-0.041
ends-0.036
 Or-0.035
-0.035
Token tree
Token ID5509
Feature activation-2.23e-02

Pos logit contributions
 tune+0.057
 sensation+0.057
 dolphin+0.052
 shark+0.050
 streaming+0.048
Neg logit contributions
 counts-0.041
ends-0.041
One-0.039
 Or-0.037
 Ready-0.034
Token with
Token ID351
Feature activation+1.92e-02

Pos logit contributions
 counts+0.026
ends+0.026
One+0.024
 Or+0.023
 Ready+0.020
Neg logit contributions
 tune-0.051
 sensation-0.051
 dolphin-0.047
 shark-0.046
 streaming-0.044
Token,
Token ID11
Feature activation+4.52e-02

Pos logit contributions
 tune+0.030
 sensation+0.030
 dolphin+0.027
 shark+0.026
 potato+0.025
Neg logit contributions
 counts-0.025
ends-0.024
One-0.024
 Or-0.022
 Ready-0.021
Token he
Token ID339
Feature activation+3.65e-02

Pos logit contributions
 tune+0.098
 sensation+0.096
 dolphin+0.093
 potato+0.087
 shark+0.086
Neg logit contributions
 counts-0.058
ends-0.055
One-0.053
 Or-0.050
 Ready-0.046
Token remembered
Token ID12086
Feature activation+3.16e-02

Pos logit contributions
 sensation+0.065
 tune+0.065
 dolphin+0.058
 shark+0.056
 potato+0.053
Neg logit contributions
 counts-0.063
ends-0.061
One-0.060
 Or-0.057
 Ready-0.052
Token the
Token ID262
Feature activation+1.55e-02

Pos logit contributions
 tune+0.070
 sensation+0.068
 streamed+0.059
andel+0.059
 dolphin+0.059
Neg logit contributions
 counts-0.039
One-0.039
ends-0.036
 Or-0.032
-0.031
Token sorrow
Token ID24140
Feature activation+1.04e-01

Pos logit contributions
 tune+0.028
 sensation+0.028
 dolphin+0.025
 shark+0.024
 streaming+0.024
Neg logit contributions
 counts-0.026
ends-0.025
One-0.025
 Or-0.023
 Ready-0.022
Tokenful
Token ID913
Feature activation+1.50e-01

Pos logit contributions
 tune+0.221
 sensation+0.219
 dolphin+0.197
 shark+0.193
 Dolphin+0.187
Neg logit contributions
ends-0.137
 counts-0.130
One-0.120
 Or-0.119
-0.115
Token look
Token ID804
Feature activation+2.36e-02

Pos logit contributions
 tune+0.279
 sensation+0.267
 dolphin+0.251
 Dolphin+0.241
 streaming+0.239
Neg logit contributions
 counts-0.241
ends-0.223
One-0.219
 Or-0.215
 Ready-0.191
Token the
Token ID262
Feature activation-1.50e-02

Pos logit contributions
 tune+0.050
 sensation+0.049
 dolphin+0.045
 shark+0.043
 Dolphin+0.042
Neg logit contributions
 counts-0.032
ends-0.032
One-0.030
 Or-0.028
 Ready-0.025
Token older
Token ID4697
Feature activation-1.93e-02

Pos logit contributions
 counts+0.023
ends+0.023
One+0.022
 Or+0.021
 Ready+0.020
Neg logit contributions
 tune-0.028
 sensation-0.028
 dolphin-0.026
 shark-0.025
 streaming-0.024
Token man
Token ID582
Feature activation-4.99e-03

Pos logit contributions
ends+0.037
 counts+0.036
One+0.035
 Or+0.034
 Ready+0.032
Neg logit contributions
 tune-0.031
 sensation-0.030
 dolphin-0.026
 shark-0.025
 chill-0.024
Token had
Token ID550
Feature activation+1.49e-04

Pos logit contributions
 counts+0.006
ends+0.006
One+0.006
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.011
 sensation-0.011
 dolphin-0.010
 shark-0.010
 chill-0.010
Token and
Token ID290
Feature activation+1.77e-02

Pos logit contributions
 tune+0.026
 sensation+0.024
 dolphin+0.024
 Dolphin+0.023
 shark+0.023
Neg logit contributions
One-0.020
 counts-0.020
ends-0.018
 Or-0.018
 Ready-0.017
Token the
Token ID262
Feature activation+1.45e-03

Pos logit contributions
 tune+0.037
 sensation+0.035
 dolphin+0.033
 Dolphin+0.031
andel+0.031
Neg logit contributions
 counts-0.026
ends-0.023
One-0.022
 Or-0.021
 has-0.021
Token lightning
Token ID14357
Feature activation+5.58e-02

Pos logit contributions
 tune+0.003
 sensation+0.003
 Dolphin+0.002
 dolphin+0.002
 streaming+0.002
Neg logit contributions
 counts-0.002
One-0.002
ends-0.002
 Or-0.002
house-0.002
Token struck
Token ID7425
Feature activation-1.44e-02

Pos logit contributions
 tune+0.102
 sensation+0.102
 Dolphin+0.101
 dolphin+0.100
 lobster+0.097
Neg logit contributions
 counts-0.082
ends-0.074
One-0.073
house-0.072
 Or-0.070
Token the
Token ID262
Feature activation+3.46e-03

Pos logit contributions
 counts+0.019
One+0.019
ends+0.018
 Or+0.017
 Ready+0.016
Neg logit contributions
 tune-0.029
 sensation-0.028
 dolphin-0.028
 shark-0.026
 Dolphin-0.026
Token hyd
Token ID7409
Feature activation+1.49e-01

Pos logit contributions
 tune+0.006
 sensation+0.006
 streaming+0.005
 dolphin+0.005
 streamed+0.005
Neg logit contributions
 counts-0.005
ends-0.005
One-0.005
 Or-0.005
 Ready-0.005
Tokenrant
Token ID5250
Feature activation+2.25e-02

Pos logit contributions
 tune+0.271
 sensation+0.268
 dolphin+0.264
 shark+0.249
 Dolphin+0.243
Neg logit contributions
 counts-0.232
ends-0.229
One-0.218
 Or-0.207
 belongs-0.196
Token.
Token ID13
Feature activation+9.53e-03

Pos logit contributions
 tune+0.050
 sensation+0.050
 dolphin+0.047
 Dolphin+0.046
 Sister+0.045
Neg logit contributions
 counts-0.026
ends-0.022
One-0.022
 Or-0.022
house-0.021
Token Suddenly
Token ID24975
Feature activation-1.89e-02

Pos logit contributions
 tune+0.019
 sensation+0.019
 dolphin+0.018
 shark+0.017
 streaming+0.016
Neg logit contributions
 counts-0.014
One-0.014
ends-0.014
 Or-0.013
 Ready-0.012
Token water
Token ID1660
Feature activation+6.07e-03

Pos logit contributions
 counts+0.025
ends+0.024
One+0.023
 Or+0.022
 Ready+0.020
Neg logit contributions
 tune-0.041
 sensation-0.041
 dolphin-0.037
 shark-0.035
 streaming-0.035
Token was
Token ID373
Feature activation-1.32e-02

Pos logit contributions
 tune+0.014
 sensation+0.013
 manner+0.012
 toddler+0.012
 tunes+0.012
Neg logit contributions
 counts-0.007
ends-0.006
 has-0.006
 cannot-0.006
One-0.006
Token They
Token ID1119
Feature activation-1.98e-02

Pos logit contributions
 counts+0.047
ends+0.047
One+0.042
 belongs+0.040
 ours+0.040
Neg logit contributions
 tune-0.050
 sensation-0.047
 dolphin-0.047
 shark-0.045
 Sister-0.045
Token are
Token ID389
Feature activation-2.27e-02

Pos logit contributions
ends+0.029
 counts+0.028
One+0.027
 Or+0.026
 Ready+0.024
Neg logit contributions
 tune-0.040
 sensation-0.039
 dolphin-0.036
 shark-0.035
 chill-0.033
Token powerful
Token ID3665
Feature activation-4.27e-02

Pos logit contributions
ends+0.037
 counts+0.035
One+0.033
 Or+0.032
 Ready+0.029
Neg logit contributions
 tune-0.043
 sensation-0.043
 dolphin-0.038
 shark-0.037
 chill-0.036
Token.
Token ID13
Feature activation-4.27e-02

Pos logit contributions
ends+0.060
One+0.060
 counts+0.057
 Or+0.054
 Ready+0.052
Neg logit contributions
 tune-0.089
 sensation-0.086
 dolphin-0.080
 shark-0.079
 tunes-0.074
Token<|endoftext|>
Token ID50256
Feature activation-4.35e-02

Pos logit contributions
ends+0.065
 counts+0.065
One+0.057
 ours+0.054
 belongs+0.054
Neg logit contributions
 tune-0.077
 sensation-0.074
 dolphin-0.072
 shark-0.069
 Sister-0.068
TokenOne
Token ID3198
Feature activation-1.58e-01

Pos logit contributions
ends+0.060
 counts+0.056
 Or+0.052
One+0.052
 ours+0.049
Neg logit contributions
 sensation-0.089
 tune-0.087
 dolphin-0.082
 shark-0.082
 Dolphin-0.077
Token day
Token ID1110
Feature activation+1.78e-02

Pos logit contributions
 counts+0.271
One+0.269
ends+0.258
 Or+0.236
 has+0.218
Neg logit contributions
 tune-0.285
 sensation-0.272
 streaming-0.248
 dolphin-0.247
 shark-0.240
Token,
Token ID11
Feature activation-3.64e-03

Pos logit contributions
 sensation+0.028
 tune+0.028
 dolphin+0.027
 shark+0.027
 streamed+0.025
Neg logit contributions
 counts-0.030
ends-0.030
One-0.029
 Or-0.028
 Ready-0.027
Token Tom
Token ID4186
Feature activation-3.27e-02

Pos logit contributions
 counts+0.006
ends+0.006
 Or+0.005
One+0.005
 Ready+0.005
Neg logit contributions
 sensation-0.006
 tune-0.006
 dolphin-0.006
 shark-0.006
 chill-0.005
Token and
Token ID290
Feature activation+2.20e-02

Pos logit contributions
ends+0.055
One+0.051
 counts+0.048
 Or+0.048
 Ready+0.048
Neg logit contributions
 sensation-0.057
 tune-0.055
 dolphin-0.053
 shark-0.052
 potato-0.049
Token Lily
Token ID20037
Feature activation-4.54e-02

Pos logit contributions
 tune+0.035
 sensation+0.035
 dolphin+0.030
 shark+0.029
 streaming+0.028
Neg logit contributions
 counts-0.041
ends-0.041
One-0.040
 Or-0.039
 ours-0.035
Token They
Token ID1119
Feature activation-1.53e-02

Pos logit contributions
 counts+0.048
ends+0.047
 belongs+0.041
 ours+0.041
One+0.040
Neg logit contributions
 dolphin-0.039
 tune-0.037
 Dolphin-0.037
 Sister-0.037
 shark-0.035
Token are
Token ID389
Feature activation-2.10e-02

Pos logit contributions
 counts+0.020
ends+0.020
One+0.019
 Or+0.018
 Ready+0.017
Neg logit contributions
 tune-0.033
 sensation-0.032
 dolphin-0.030
 shark-0.029
 streaming-0.028
Token careful
Token ID8161
Feature activation+2.10e-03

Pos logit contributions
ends+0.034
 counts+0.032
One+0.031
 Or+0.029
 ours+0.027
Neg logit contributions
 tune-0.040
 sensation-0.040
 dolphin-0.035
 shark-0.034
 tant-0.033
Token.
Token ID13
Feature activation-3.80e-02

Pos logit contributions
 tune+0.004
 sensation+0.004
 face+0.004
 dolphin+0.004
 shark+0.003
Neg logit contributions
 counts-0.003
One-0.003
ends-0.003
 Ready-0.003
 Or-0.003
Token<|endoftext|>
Token ID50256
Feature activation-4.59e-02

Pos logit contributions
 counts+0.060
ends+0.058
One+0.052
 belongs+0.050
 ours+0.050
Neg logit contributions
 tune-0.066
 dolphin-0.064
 sensation-0.063
 Dolphin-0.060
 Sister-0.060
TokenOne
Token ID3198
Feature activation-1.58e-01

Pos logit contributions
ends+0.060
 counts+0.057
One+0.053
 Or+0.052
 ours+0.048
Neg logit contributions
 sensation-0.097
 tune-0.096
 dolphin-0.091
 shark-0.090
 Dolphin-0.085
Token morning
Token ID3329
Feature activation-7.51e-04

Pos logit contributions
One+0.257
 counts+0.254
ends+0.237
 Or+0.215
 has+0.207
Neg logit contributions
 tune-0.301
 sensation-0.283
 streaming-0.269
 dolphin-0.258
 streamed-0.252
Token,
Token ID11
Feature activation-3.37e-03

Pos logit contributions
 counts+0.001
ends+0.001
One+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 shark-0.001
 potato-0.001
Token a
Token ID257
Feature activation-1.72e-02

Pos logit contributions
 counts+0.006
ends+0.005
One+0.005
 Or+0.005
 Ready+0.005
Neg logit contributions
 sensation-0.006
 tune-0.006
 dolphin-0.005
 shark-0.005
 face-0.005
Token little
Token ID1310
Feature activation+2.99e-02

Pos logit contributions
One+0.031
ends+0.030
 Or+0.029
 counts+0.027
 has+0.026
Neg logit contributions
 sensation-0.030
 tune-0.028
 streaming-0.027
ened-0.025
 streamed-0.025
Token ot
Token ID30972
Feature activation+3.43e-03

Pos logit contributions
 sensation+0.054
 tune+0.052
 streaming+0.045
 streamed+0.044
 lull+0.043
Neg logit contributions
One-0.048
 counts-0.048
ends-0.046
 Or-0.045
house-0.041
Token shoes
Token ID10012
Feature activation-3.97e-02

Pos logit contributions
 counts+0.029
ends+0.029
One+0.027
 Or+0.027
 Ready+0.025
Neg logit contributions
 tune-0.030
 sensation-0.030
 dolphin-0.026
 streaming-0.025
 shark-0.025
Token are
Token ID389
Feature activation-4.69e-02

Pos logit contributions
ends+0.068
 counts+0.068
One+0.066
 Or+0.063
 Ready+0.060
Neg logit contributions
 tune-0.068
 sensation-0.067
 dolphin-0.060
 shark-0.058
 streaming-0.055
Token better
Token ID1365
Feature activation-5.08e-02

Pos logit contributions
 counts+0.066
ends+0.065
One+0.061
 Or+0.061
 Ready+0.055
Neg logit contributions
 tune-0.097
 sensation-0.096
 dolphin-0.088
 shark-0.085
 streaming-0.083
Token.
Token ID13
Feature activation-3.97e-02

Pos logit contributions
 Ready+0.075
 counts+0.075
One+0.075
 Or+0.069
ends+0.069
Neg logit contributions
 tune-0.097
 sensation-0.089
 dolphin-0.085
 shark-0.082
 face-0.080
Token<|endoftext|>
Token ID50256
Feature activation-4.49e-02

Pos logit contributions
 counts+0.069
ends+0.067
One+0.059
 belongs+0.059
 ours+0.059
Neg logit contributions
 tune-0.058
 dolphin-0.057
 Sister-0.056
 Dolphin-0.055
 sensation-0.054
TokenOne
Token ID3198
Feature activation-1.58e-01

Pos logit contributions
ends+0.064
 counts+0.059
 Or+0.056
One+0.055
 ours+0.054
Neg logit contributions
 sensation-0.088
 tune-0.085
 dolphin-0.082
 shark-0.082
 Dolphin-0.078
Token day
Token ID1110
Feature activation+1.87e-02

Pos logit contributions
One+0.262
 counts+0.260
ends+0.242
 Or+0.223
 has+0.209
Neg logit contributions
 tune-0.293
 sensation-0.278
 streaming-0.259
 dolphin-0.254
 shark-0.246
Token,
Token ID11
Feature activation-4.10e-03

Pos logit contributions
 sensation+0.031
 tune+0.031
 dolphin+0.030
 shark+0.029
 streamed+0.028
Neg logit contributions
 counts-0.031
ends-0.030
One-0.029
 Or-0.029
 Ready-0.027
Token a
Token ID257
Feature activation-1.71e-02

Pos logit contributions
 counts+0.007
ends+0.006
 Or+0.006
One+0.006
 Ready+0.006
Neg logit contributions
 sensation-0.007
 tune-0.007
 dolphin-0.007
 shark-0.007
 Dolphin-0.006
Token little
Token ID1310
Feature activation+3.13e-02

Pos logit contributions
One+0.032
ends+0.032
 Or+0.030
 counts+0.028
 has+0.027
Neg logit contributions
 sensation-0.029
 tune-0.026
 streaming-0.026
ened-0.023
 streamed-0.023
Token girl
Token ID2576
Feature activation+1.31e-02

Pos logit contributions
 sensation+0.056
 tune+0.054
 streaming+0.048
 streamed+0.046
 lull+0.045
Neg logit contributions
One-0.050
 counts-0.050
ends-0.049
 Or-0.047
house-0.043
Token and
Token ID290
Feature activation-1.98e-02

Pos logit contributions
ends+0.082
 counts+0.080
One+0.077
 Or+0.073
 Ready+0.067
Neg logit contributions
 tune-0.124
 sensation-0.122
 dolphin-0.112
 shark-0.109
 face-0.105
Token to
Token ID284
Feature activation+7.01e-03

Pos logit contributions
 counts+0.027
ends+0.026
One+0.025
 Or+0.025
 Ready+0.022
Neg logit contributions
 sensation-0.042
 tune-0.042
 dolphin-0.039
 shark-0.037
 Dolphin-0.036
Token share
Token ID2648
Feature activation-1.24e-02

Pos logit contributions
 sensation+0.014
 tune+0.014
 dolphin+0.013
 shark+0.013
 streamed+0.013
Neg logit contributions
One-0.010
 counts-0.010
 Or-0.009
ends-0.009
 ours-0.008
Token.
Token ID13
Feature activation-3.11e-02

Pos logit contributions
ends+0.019
 counts+0.019
One+0.018
 Or+0.017
 Ready+0.017
Neg logit contributions
 tune-0.024
 sensation-0.023
 dolphin-0.021
 face-0.021
 shark-0.021
Token<|endoftext|>
Token ID50256
Feature activation-4.20e-02

Pos logit contributions
 counts+0.056
ends+0.053
 belongs+0.048
 ours+0.046
One+0.045
Neg logit contributions
 tune-0.044
 dolphin-0.043
 Sister-0.043
 Dolphin-0.042
 sensation-0.041
TokenOne
Token ID3198
Feature activation-1.57e-01

Pos logit contributions
ends+0.055
 counts+0.052
One+0.048
 Or+0.048
 ours+0.044
Neg logit contributions
 sensation-0.089
 tune-0.088
 dolphin-0.082
 shark-0.081
 streamed-0.077
Token day
Token ID1110
Feature activation+1.47e-02

Pos logit contributions
One+0.251
 counts+0.248
ends+0.230
 Or+0.211
 has+0.205
Neg logit contributions
 tune-0.305
 sensation-0.286
 streaming-0.274
 dolphin-0.264
 streamed-0.256
Token,
Token ID11
Feature activation-2.12e-03

Pos logit contributions
 tune+0.025
 sensation+0.025
 dolphin+0.024
 shark+0.024
 streamed+0.023
Neg logit contributions
 counts-0.024
ends-0.024
One-0.022
 Or-0.022
 Ready-0.021
Token a
Token ID257
Feature activation-1.76e-02

Pos logit contributions
 counts+0.003
ends+0.003
 Or+0.003
One+0.003
 Ready+0.003
Neg logit contributions
 sensation-0.004
 tune-0.004
 dolphin-0.003
 shark-0.003
 chill-0.003
Token little
Token ID1310
Feature activation+3.04e-02

Pos logit contributions
One+0.032
ends+0.032
 Or+0.030
 counts+0.029
 has+0.027
Neg logit contributions
 sensation-0.030
 tune-0.028
 streaming-0.027
ened-0.025
 streamed-0.025
Token boy
Token ID2933
Feature activation+1.11e-02

Pos logit contributions
 sensation+0.055
 tune+0.054
 streaming+0.047
 streamed+0.046
 lull+0.045
Neg logit contributions
One-0.048
 counts-0.048
ends-0.046
 Or-0.045
house-0.042
Token not
Token ID407
Feature activation-1.07e-02

Pos logit contributions
 tune+0.018
 sensation+0.018
 dolphin+0.017
 shark+0.016
 face+0.016
Neg logit contributions
 counts-0.010
ends-0.009
 Or-0.009
One-0.009
 Ready-0.008
Token bother
Token ID11393
Feature activation-2.27e-03

Pos logit contributions
 counts+0.015
ends+0.014
One+0.014
 Or+0.013
 Ready+0.012
Neg logit contributions
 sensation-0.022
 tune-0.022
 dolphin-0.020
 shark-0.020
 Dolphin-0.019
Token anyone
Token ID2687
Feature activation-2.53e-02

Pos logit contributions
 counts+0.003
One+0.003
ends+0.003
 Or+0.003
 ours+0.003
Neg logit contributions
 sensation-0.005
 tune-0.005
 grin-0.004
 streamed-0.004
 chill-0.004
Token.
Token ID13
Feature activation-2.92e-02

Pos logit contributions
ends+0.035
 counts+0.034
One+0.034
 Or+0.032
 Ready+0.031
Neg logit contributions
 tune-0.053
 sensation-0.050
 dolphin-0.047
 shark-0.046
 face-0.044
Token<|endoftext|>
Token ID50256
Feature activation-4.48e-02

Pos logit contributions
 counts+0.045
ends+0.044
One+0.039
 belongs+0.038
 Or+0.037
Neg logit contributions
 tune-0.053
 sensation-0.051
 dolphin-0.049
 shark-0.047
 Sister-0.047
TokenOne
Token ID3198
Feature activation-1.57e-01

Pos logit contributions
ends+0.056
 counts+0.054
One+0.050
 Or+0.048
 ours+0.043
Neg logit contributions
 sensation-0.098
 tune-0.098
 dolphin-0.091
 shark-0.089
 Dolphin-0.085
Token day
Token ID1110
Feature activation+1.86e-02

Pos logit contributions
One+0.245
 counts+0.241
ends+0.222
 Or+0.198
 has+0.195
Neg logit contributions
 tune-0.306
 sensation-0.287
 streaming-0.279
 dolphin-0.266
 streamed-0.260
Token,
Token ID11
Feature activation-1.65e-03

Pos logit contributions
 tune+0.038
 sensation+0.037
 dolphin+0.034
 shark+0.033
 streaming+0.032
Neg logit contributions
 counts-0.027
ends-0.026
One-0.025
 Or-0.024
 Ready-0.022
Token a
Token ID257
Feature activation-1.72e-02

Pos logit contributions
 counts+0.002
ends+0.002
One+0.002
 Or+0.002
 Ready+0.002
Neg logit contributions
 tune-0.003
 sensation-0.003
 dolphin-0.003
 shark-0.003
 streamed-0.003
Token boy
Token ID2933
Feature activation+1.35e-02

Pos logit contributions
One+0.032
ends+0.032
 Or+0.030
 counts+0.028
 has+0.028
Neg logit contributions
 sensation-0.029
 tune-0.026
 streaming-0.026
ened-0.024
 streamed-0.024
Token named
Token ID3706
Feature activation-5.12e-02

Pos logit contributions
 tune+0.025
 sensation+0.024
 dolphin+0.022
 shark+0.022
 potato+0.020
Neg logit contributions
ends-0.022
 counts-0.021
One-0.021
 Or-0.020
 Ready-0.019
Token They
Token ID1119
Feature activation-2.06e-02

Pos logit contributions
 counts+0.064
ends+0.063
One+0.058
 Or+0.054
 ours+0.052
Neg logit contributions
 tune-0.094
 sensation-0.091
 dolphin-0.088
 shark-0.085
 Dolphin-0.082
Token are
Token ID389
Feature activation-2.57e-02

Pos logit contributions
ends+0.025
 counts+0.025
One+0.024
 Or+0.023
 Ready+0.020
Neg logit contributions
 tune-0.046
 sensation-0.045
 dolphin-0.042
 shark-0.040
 chill-0.039
Token lucky
Token ID9670
Feature activation-2.33e-02

Pos logit contributions
ends+0.043
 counts+0.042
One+0.041
 Or+0.039
 Ready+0.036
Neg logit contributions
 tune-0.047
 sensation-0.046
 dolphin-0.041
 shark-0.039
 streaming-0.038
Token.
Token ID13
Feature activation-4.57e-02

Pos logit contributions
ends+0.031
 counts+0.030
One+0.029
 Or+0.027
 Ready+0.025
Neg logit contributions
 tune-0.050
 sensation-0.049
 dolphin-0.046
 shark-0.046
 potato-0.043
Token<|endoftext|>
Token ID50256
Feature activation-5.07e-02

Pos logit contributions
 counts+0.063
ends+0.063
One+0.057
 Or+0.053
 ours+0.052
Neg logit contributions
 tune-0.091
 sensation-0.088
 dolphin-0.087
 shark-0.083
 Dolphin-0.080
TokenOne
Token ID3198
Feature activation-1.56e-01

Pos logit contributions
ends+0.072
 counts+0.066
 Or+0.062
One+0.061
 ours+0.059
Neg logit contributions
 sensation-0.102
 tune-0.099
 dolphin-0.096
 shark-0.095
 Dolphin-0.090
Token day
Token ID1110
Feature activation+1.04e-02

Pos logit contributions
One+0.251
 counts+0.248
ends+0.233
 Or+0.208
 has+0.204
Neg logit contributions
 tune-0.298
 sensation-0.288
 streaming-0.266
 dolphin-0.256
 streamed-0.254
Token,
Token ID11
Feature activation-8.90e-03

Pos logit contributions
 dolphin+0.015
 shark+0.014
 sensation+0.014
 tune+0.014
 streamed+0.013
Neg logit contributions
 counts-0.019
ends-0.019
 Or-0.018
One-0.018
 Ready-0.017
Token Tom
Token ID4186
Feature activation-3.41e-02

Pos logit contributions
 counts+0.015
ends+0.014
 Or+0.013
One+0.013
 Ready+0.013
Neg logit contributions
 sensation-0.015
 tune-0.015
 dolphin-0.014
 shark-0.014
 face-0.013
Token and
Token ID290
Feature activation+1.44e-02

Pos logit contributions
ends+0.048
One+0.043
 Ready+0.040
 Or+0.040
 counts+0.038
Neg logit contributions
 sensation-0.068
 tune-0.066
 dolphin-0.066
 shark-0.066
 potato-0.062
Token his
Token ID465
Feature activation-7.55e-03

Pos logit contributions
 tune+0.024
 sensation+0.023
 dolphin+0.021
 shark+0.020
 streaming+0.019
Neg logit contributions
 counts-0.026
ends-0.026
One-0.025
 Or-0.025
 Ready-0.023
Token.
Token ID13
Feature activation-2.67e-02

Pos logit contributions
ends+0.006
 counts+0.005
One+0.005
 Or+0.005
 someday+0.004
Neg logit contributions
 tune-0.017
 sensation-0.017
 dolphin-0.016
 shark-0.016
 Dolphin-0.016
Token They
Token ID1119
Feature activation-9.83e-03

Pos logit contributions
 counts+0.036
ends+0.036
One+0.034
 Or+0.032
 Ready+0.029
Neg logit contributions
 tune-0.056
 sensation-0.055
 dolphin-0.050
 shark-0.049
 streaming-0.047
Token hoped
Token ID10719
Feature activation-1.14e-02

Pos logit contributions
 counts+0.014
One+0.013
ends+0.013
 Or+0.012
 cannot+0.011
Neg logit contributions
 tune-0.021
 sensation-0.021
 dolphin-0.019
 shark-0.018
 Dolphin-0.018
Token.
Token ID13
Feature activation-3.85e-02

Pos logit contributions
 counts+0.014
ends+0.013
One+0.012
 Or+0.012
 Ready+0.011
Neg logit contributions
 tune-0.026
 sensation-0.026
 dolphin-0.023
 streamed-0.022
 shark-0.022
Token<|endoftext|>
Token ID50256
Feature activation-4.75e-02

Pos logit contributions
 counts+0.057
ends+0.055
One+0.051
 ours+0.047
 Or+0.047
Neg logit contributions
 tune-0.074
 sensation-0.073
 dolphin-0.067
 shark-0.064
 Dolphin-0.063
TokenOne
Token ID3198
Feature activation-1.56e-01

Pos logit contributions
ends+0.070
 counts+0.063
 Or+0.060
One+0.060
 ours+0.060
Neg logit contributions
 sensation-0.093
 tune-0.088
 shark-0.086
 dolphin-0.086
 Dolphin-0.081
Token day
Token ID1110
Feature activation+1.46e-02

Pos logit contributions
 counts+0.255
One+0.247
ends+0.231
 Or+0.212
 has+0.208
Neg logit contributions
 tune-0.298
 sensation-0.278
 streaming-0.260
 dolphin-0.252
 streamed-0.244
Token,
Token ID11
Feature activation-3.98e-03

Pos logit contributions
 sensation+0.024
 tune+0.024
 dolphin+0.024
 shark+0.023
 streamed+0.022
Neg logit contributions
 counts-0.024
ends-0.024
One-0.023
 Or-0.022
 Ready-0.021
Token a
Token ID257
Feature activation-1.89e-02

Pos logit contributions
 counts+0.007
ends+0.006
 Or+0.006
One+0.006
 Ready+0.005
Neg logit contributions
 sensation-0.007
 tune-0.007
 dolphin-0.006
 shark-0.006
 streamed-0.006
Token baby
Token ID5156
Feature activation+2.96e-02

Pos logit contributions
One+0.035
ends+0.034
 Or+0.033
 counts+0.032
 has+0.030
Neg logit contributions
 sensation-0.032
 tune-0.029
 streaming-0.028
 streamed-0.026
ened-0.026
Token k
Token ID479
Feature activation-2.16e-02

Pos logit contributions
 tune+0.058
 sensation+0.057
 dolphin+0.050
 shark+0.049
 streaming+0.047
Neg logit contributions
 counts-0.046
ends-0.044
One-0.043
 Or-0.041
 Ready-0.038
Token They
Token ID1119
Feature activation-1.06e-02

Pos logit contributions
ends+0.039
 counts+0.039
One+0.036
 Or+0.035
 Ready+0.031
Neg logit contributions
 tune-0.065
 sensation-0.065
 dolphin-0.059
 shark-0.058
 streaming-0.055
Token were
Token ID547
Feature activation-1.19e-02

Pos logit contributions
 counts+0.016
ends+0.016
One+0.015
 Or+0.014
 has+0.012
Neg logit contributions
 tune-0.021
 sensation-0.021
 shark-0.019
 dolphin-0.019
 potato-0.018
Token mighty
Token ID18680
Feature activation-5.54e-02

Pos logit contributions
ends+0.019
 counts+0.019
One+0.018
 Or+0.018
 Ready+0.016
Neg logit contributions
 tune-0.022
 sensation-0.022
 dolphin-0.020
 shark-0.019
 streaming-0.018
Token.
Token ID13
Feature activation-4.15e-02

Pos logit contributions
ends+0.095
One+0.094
 counts+0.092
 Or+0.088
 Ready+0.084
Neg logit contributions
 tune-0.097
 sensation-0.092
 dolphin-0.086
 shark-0.083
 face-0.078
Token<|endoftext|>
Token ID50256
Feature activation-4.54e-02

Pos logit contributions
ends+0.050
 counts+0.050
One+0.047
 Or+0.044
 Ready+0.040
Neg logit contributions
 tune-0.095
 sensation-0.093
 dolphin-0.086
 shark-0.083
 streaming-0.081
TokenOne
Token ID3198
Feature activation-1.56e-01

Pos logit contributions
ends+0.057
 counts+0.054
One+0.050
 Or+0.049
 ours+0.044
Neg logit contributions
 sensation-0.099
 tune-0.098
 dolphin-0.092
 shark-0.091
 streamed-0.086
Token day
Token ID1110
Feature activation+1.02e-02

Pos logit contributions
One+0.241
 counts+0.240
ends+0.224
 Or+0.202
 has+0.196
Neg logit contributions
 tune-0.308
 sensation-0.295
 streaming-0.276
 dolphin-0.264
 streamed-0.260
Token,
Token ID11
Feature activation-7.36e-03

Pos logit contributions
 sensation+0.017
 tune+0.017
 dolphin+0.016
 shark+0.016
 streamed+0.015
Neg logit contributions
 counts-0.017
ends-0.017
 Or-0.016
One-0.016
 Ready-0.015
Token a
Token ID257
Feature activation-1.86e-02

Pos logit contributions
 counts+0.012
ends+0.012
 Or+0.011
One+0.011
 Ready+0.010
Neg logit contributions
 sensation-0.013
 tune-0.012
 dolphin-0.011
 shark-0.011
 streamed-0.011
Token boy
Token ID2933
Feature activation+6.62e-03

Pos logit contributions
One+0.032
ends+0.031
 Or+0.030
 counts+0.029
 Ready+0.027
Neg logit contributions
 sensation-0.034
 tune-0.031
 streaming-0.031
ened-0.029
 streamed-0.029
Token and
Token ID290
Feature activation+3.84e-04

Pos logit contributions
 tune+0.013
 sensation+0.012
 dolphin+0.011
 shark+0.011
 streaming+0.010
Neg logit contributions
ends-0.010
 counts-0.010
One-0.010
 Or-0.009
 Ready-0.009
Token the
Token ID262
Feature activation-1.39e-02

Pos logit contributions
 tune+0.000
 sensation+0.000
 dolphin+0.000
 shark+0.000
 streaming+0.000
Neg logit contributions
 counts-0.000
ends-0.000
One-0.000
 Or-0.000
 Ready-0.000
Token needle
Token ID17598
Feature activation+3.17e-02

Pos logit contributions
ends+0.025
 counts+0.024
One+0.023
 Or+0.022
 Ready+0.021
Neg logit contributions
 tune-0.024
 sensation-0.024
 dolphin-0.021
 shark-0.020
 potato-0.019
Token was
Token ID373
Feature activation-3.34e-02

Pos logit contributions
 tune+0.060
 sensation+0.059
 dolphin+0.054
 shark+0.052
 streaming+0.050
Neg logit contributions
ends-0.049
 counts-0.049
One-0.047
 Or-0.045
 Ready-0.041
Token.
Token ID13
Feature activation-3.78e-02

Pos logit contributions
ends+0.048
 counts+0.047
One+0.045
 Or+0.043
 Ready+0.040
Neg logit contributions
 tune-0.068
 sensation-0.068
 dolphin-0.062
 shark-0.061
 streaming-0.059
Token<|endoftext|>
Token ID50256
Feature activation-5.02e-02

Pos logit contributions
 counts+0.053
ends+0.053
One+0.050
 Or+0.046
 Ready+0.043
Neg logit contributions
 tune-0.076
 sensation-0.076
 dolphin-0.070
 shark-0.068
 Dolphin-0.064
TokenOne
Token ID3198
Feature activation-1.56e-01

Pos logit contributions
ends+0.066
 counts+0.061
One+0.058
 Or+0.057
 ours+0.052
Neg logit contributions
 sensation-0.107
 tune-0.105
 dolphin-0.099
 shark-0.098
 streamed-0.093
Token day
Token ID1110
Feature activation+1.32e-02

Pos logit contributions
 counts+0.240
One+0.237
ends+0.214
 Or+0.200
 has+0.199
Neg logit contributions
 tune-0.311
 sensation-0.290
 streaming-0.275
 dolphin-0.263
 tant-0.259
Token,
Token ID11
Feature activation-7.89e-03

Pos logit contributions
 dolphin+0.019
 sensation+0.019
 shark+0.019
 tune+0.019
 streamed+0.017
Neg logit contributions
 counts-0.024
ends-0.023
One-0.022
 Or-0.022
 Ready-0.021
Token a
Token ID257
Feature activation-1.88e-02

Pos logit contributions
 counts+0.013
ends+0.012
One+0.012
 Or+0.012
 Ready+0.011
Neg logit contributions
 sensation-0.014
 tune-0.013
 dolphin-0.013
 shark-0.012
 face-0.011
Token boy
Token ID2933
Feature activation+8.90e-03

Pos logit contributions
One+0.035
ends+0.033
 Or+0.033
 counts+0.031
 Ready+0.029
Neg logit contributions
 sensation-0.032
 tune-0.029
 streaming-0.029
 streamed-0.027
ened-0.026
Token named
Token ID3706
Feature activation-4.84e-02

Pos logit contributions
 tune+0.016
 sensation+0.015
 dolphin+0.014
 shark+0.014
 streamed+0.013
Neg logit contributions
ends-0.015
 counts-0.014
One-0.014
 Or-0.013
 Ready-0.013
Token y
Token ID331
Feature activation-4.03e-02

Pos logit contributions
 counts+0.067
ends+0.067
One+0.064
 Or+0.061
 Ready+0.056
Neg logit contributions
 tune-0.088
 sensation-0.087
 dolphin-0.079
 shark-0.076
 streaming-0.073
Tokenummy
Token ID13513
Feature activation-6.95e-02

Pos logit contributions
ends+0.099
 counts+0.097
One+0.093
 Or+0.092
 Ready+0.089
Neg logit contributions
 tune-0.042
 sensation-0.042
 dolphin-0.034
 shark-0.032
 potato-0.031
Token too
Token ID1165
Feature activation-6.52e-02

Pos logit contributions
ends+0.102
One+0.097
 counts+0.095
 Or+0.092
 Ready+0.088
Neg logit contributions
 tune-0.143
 sensation-0.141
 dolphin-0.129
 shark-0.127
 potato-0.122
Token.
Token ID13
Feature activation-2.66e-02

Pos logit contributions
ends+0.097
 counts+0.096
 Or+0.091
One+0.090
 Ready+0.087
Neg logit contributions
 tune-0.126
 sensation-0.121
 dolphin-0.117
 shark-0.113
 potato-0.108
Token<|endoftext|>
Token ID50256
Feature activation-4.55e-02

Pos logit contributions
 counts+0.042
ends+0.041
One+0.037
 belongs+0.034
 ours+0.034
Neg logit contributions
 tune-0.047
 sensation-0.045
 dolphin-0.044
 shark-0.042
 Dolphin-0.042
TokenOne
Token ID3198
Feature activation-1.56e-01

Pos logit contributions
ends+0.065
 counts+0.061
 Or+0.057
One+0.056
 ours+0.055
Neg logit contributions
 sensation-0.090
 tune-0.087
 dolphin-0.084
 shark-0.083
 Dolphin-0.079
Token day
Token ID1110
Feature activation+1.70e-02

Pos logit contributions
 counts+0.260
One+0.259
ends+0.245
 Or+0.224
 has+0.210
Neg logit contributions
 tune-0.286
 sensation-0.272
 streaming-0.254
 dolphin-0.246
 streamed-0.240
Token,
Token ID11
Feature activation-2.45e-03

Pos logit contributions
 tune+0.034
 sensation+0.034
 dolphin+0.031
 shark+0.030
 streamed+0.029
Neg logit contributions
 counts-0.024
ends-0.024
One-0.023
 Or-0.022
 Ready-0.020
Token a
Token ID257
Feature activation-1.76e-02

Pos logit contributions
 counts+0.004
ends+0.004
One+0.003
 Or+0.003
 Ready+0.003
Neg logit contributions
 sensation-0.005
 tune-0.005
 dolphin-0.004
 shark-0.004
 face-0.004
Token woman
Token ID2415
Feature activation+6.86e-03

Pos logit contributions
One+0.033
ends+0.033
 Or+0.031
 counts+0.030
 has+0.029
Neg logit contributions
 sensation-0.029
 tune-0.026
 streaming-0.026
 streamed-0.023
ened-0.023
Token went
Token ID1816
Feature activation-4.71e-03

Pos logit contributions
 tune+0.010
 sensation+0.009
 dolphin+0.009
 shark+0.008
 streamed+0.008
Neg logit contributions
ends-0.014
One-0.013
 ours-0.012
 Ready-0.012
 counts-0.012
Token she
Token ID673
Feature activation+3.85e-03

Pos logit contributions
 tune+0.029
 sensation+0.029
 dolphin+0.026
 shark+0.025
 chill+0.025
Neg logit contributions
 counts-0.013
ends-0.013
One-0.011
 Or-0.011
 Ready-0.009
Token was
Token ID373
Feature activation-3.69e-03

Pos logit contributions
 tune+0.007
 sensation+0.007
 dolphin+0.006
 shark+0.006
 streaming+0.006
Neg logit contributions
 counts-0.006
ends-0.006
One-0.006
 Or-0.006
 Ready-0.005
Token smiling
Token ID16755
Feature activation+7.09e-03

Pos logit contributions
ends+0.006
 counts+0.005
One+0.005
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.007
 sensation-0.007
 dolphin-0.006
 shark-0.006
 potato-0.006
Token all
Token ID477
Feature activation-2.48e-02

Pos logit contributions
 sensation+0.017
 tune+0.017
 chill+0.015
 dolphin+0.015
 shark+0.015
Neg logit contributions
ends-0.007
 counts-0.006
One-0.006
 Or-0.006
 Maybe-0.005
Token<|endoftext|>
Token ID50256
Feature activation-3.92e-02

Pos logit contributions
 counts+0.042
ends+0.040
One+0.040
 Or+0.036
 ours+0.034
Neg logit contributions
 tune-0.044
 sensation-0.043
 dolphin-0.040
 shark-0.039
 potato-0.037
TokenOne
Token ID3198
Feature activation-1.55e-01

Pos logit contributions
ends+0.048
 counts+0.046
One+0.043
 Or+0.042
 ours+0.037
Neg logit contributions
 sensation-0.087
 tune-0.087
 dolphin-0.080
 shark-0.078
 streaming-0.075
Token day
Token ID1110
Feature activation+1.37e-02

Pos logit contributions
One+0.264
 counts+0.262
ends+0.252
 Or+0.226
+0.213
Neg logit contributions
 tune-0.281
 sensation-0.268
 dolphin-0.245
 shark-0.239
 streaming-0.236
Token,
Token ID11
Feature activation-5.39e-04

Pos logit contributions
 tune+0.029
 sensation+0.029
 dolphin+0.026
 shark+0.026
 streaming+0.025
Neg logit contributions
ends-0.018
 counts-0.018
One-0.017
 Or-0.016
 Ready-0.015
Token Sally
Token ID25737
Feature activation-1.10e-02

Pos logit contributions
 counts+0.001
ends+0.001
One+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.001
 sensation-0.001
 dolphin-0.001
 shark-0.001
 streaming-0.001
Token was
Token ID373
Feature activation-2.44e-03

Pos logit contributions
ends+0.018
 counts+0.017
One+0.017
 Or+0.016
 Ready+0.015
Neg logit contributions
 tune-0.020
 sensation-0.020
 dolphin-0.018
 shark-0.018
 potato-0.017
Token outside
Token ID2354
Feature activation-3.74e-03

Pos logit contributions
 counts+0.004
ends+0.004
One+0.004
 Or+0.004
 Ready+0.003
Neg logit contributions
 tune-0.004
 sensation-0.004
 dolphin-0.004
 shark-0.004
 streaming-0.004
Token might
Token ID1244
Feature activation-3.59e-02

Pos logit contributions
 tune+0.025
 sensation+0.025
 dolphin+0.023
 shark+0.023
 face+0.022
Neg logit contributions
ends-0.012
 counts-0.012
One-0.012
 Or-0.011
 Ready-0.010
Token get
Token ID651
Feature activation-2.55e-02

Pos logit contributions
ends+0.056
 counts+0.055
 Or+0.050
One+0.050
 Ready+0.047
Neg logit contributions
 tune-0.069
 sensation-0.067
 dolphin-0.060
 shark-0.058
 face-0.058
Token hurt
Token ID5938
Feature activation-2.60e-02

Pos logit contributions
 counts+0.041
ends+0.040
One+0.039
 Or+0.038
 Ready+0.035
Neg logit contributions
 tune-0.047
 sensation-0.046
 dolphin-0.042
 shark-0.041
 streaming-0.040
Token.
Token ID13
Feature activation-2.83e-02

Pos logit contributions
ends+0.033
 counts+0.031
One+0.028
 Or+0.026
 Ready+0.026
Neg logit contributions
 tune-0.059
 sensation-0.057
 dolphin-0.053
 shark-0.052
 potato-0.050
Token<|endoftext|>
Token ID50256
Feature activation-4.35e-02

Pos logit contributions
ends+0.042
 counts+0.040
 ours+0.035
One+0.034
 belongs+0.033
Neg logit contributions
 tune-0.050
 sensation-0.048
 Sister-0.047
 Potato-0.046
 Dolphin-0.045
TokenOne
Token ID3198
Feature activation-1.55e-01

Pos logit contributions
ends+0.057
 counts+0.054
One+0.050
 Or+0.050
 ours+0.045
Neg logit contributions
 sensation-0.092
 tune-0.091
 dolphin-0.085
 shark-0.084
 Dolphin-0.079
Token day
Token ID1110
Feature activation+1.86e-02

Pos logit contributions
One+0.258
 counts+0.257
ends+0.237
 Or+0.219
 has+0.205
Neg logit contributions
 tune-0.288
 sensation-0.274
 streaming-0.254
 dolphin-0.251
 shark-0.242
Token,
Token ID11
Feature activation-3.91e-03

Pos logit contributions
 tune+0.037
 sensation+0.037
 dolphin+0.034
 shark+0.033
 streaming+0.031
Neg logit contributions
 counts-0.027
ends-0.027
One-0.025
 Or-0.024
 Ready-0.022
Token a
Token ID257
Feature activation-1.51e-02

Pos logit contributions
 counts+0.006
ends+0.006
One+0.005
 Or+0.005
 Ready+0.005
Neg logit contributions
 sensation-0.008
 tune-0.008
 dolphin-0.007
 shark-0.007
 chill-0.006
Token little
Token ID1310
Feature activation+3.06e-02

Pos logit contributions
One+0.028
ends+0.028
 Or+0.026
 counts+0.025
 has+0.024
Neg logit contributions
 sensation-0.025
 tune-0.023
 streaming-0.023
ened-0.021
 streamed-0.021
Token boy
Token ID2933
Feature activation+1.23e-02

Pos logit contributions
 sensation+0.056
 tune+0.054
 streaming+0.048
 streamed+0.047
 lull+0.045
Neg logit contributions
One-0.048
 counts-0.047
ends-0.045
 Or-0.045
house-0.041
Token protect
Token ID1805
Feature activation+2.41e-02

Pos logit contributions
 counts+0.053
ends+0.049
One+0.045
 Or+0.044
 has+0.042
Neg logit contributions
 tune-0.059
 sensation-0.058
 dolphin-0.051
 shark-0.050
 streaming-0.049
Token their
Token ID511
Feature activation-1.73e-02

Pos logit contributions
 sensation+0.048
 tune+0.048
 dolphin+0.041
 shark+0.041
 chill+0.041
Neg logit contributions
 counts-0.035
ends-0.034
One-0.033
 Or-0.031
 ours-0.030
Token skin
Token ID4168
Feature activation+2.63e-02

Pos logit contributions
 counts+0.028
ends+0.028
One+0.026
 Or+0.025
 Ready+0.023
Neg logit contributions
 sensation-0.031
 tune-0.031
 dolphin-0.028
 shark-0.027
 streaming-0.027
Token.
Token ID13
Feature activation-3.52e-02

Pos logit contributions
 tune+0.062
 sensation+0.061
 dolphin+0.057
 shark+0.055
 streaming+0.053
Neg logit contributions
ends-0.029
 counts-0.028
One-0.027
 Or-0.025
 Ready-0.022
Token<|endoftext|>
Token ID50256
Feature activation-4.14e-02

Pos logit contributions
 counts+0.057
ends+0.053
One+0.050
 belongs+0.049
 ours+0.049
Neg logit contributions
 tune-0.057
 sensation-0.057
 dolphin-0.053
 Sister-0.052
 shark-0.051
TokenOne
Token ID3198
Feature activation-1.55e-01

Pos logit contributions
ends+0.058
 counts+0.054
 Or+0.050
One+0.050
 ours+0.048
Neg logit contributions
 sensation-0.083
 tune-0.081
 dolphin-0.077
 shark-0.077
 Dolphin-0.073
Token day
Token ID1110
Feature activation+1.86e-02

Pos logit contributions
 counts+0.262
One+0.260
ends+0.248
 Or+0.225
 has+0.211
Neg logit contributions
 tune-0.285
 sensation-0.271
 streaming-0.250
 dolphin-0.247
 shark-0.240
Token,
Token ID11
Feature activation-4.73e-04

Pos logit contributions
 sensation+0.031
 tune+0.031
 dolphin+0.030
 shark+0.029
 streamed+0.028
Neg logit contributions
 counts-0.031
ends-0.030
One-0.029
 Or-0.029
 Ready-0.027
Token Lily
Token ID20037
Feature activation-4.32e-02

Pos logit contributions
 counts+0.001
ends+0.001
One+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 sensation-0.001
 tune-0.001
 dolphin-0.001
 shark-0.001
 chill-0.001
Token and
Token ID290
Feature activation+2.16e-02

Pos logit contributions
ends+0.066
 counts+0.063
One+0.062
 Or+0.059
 Ready+0.056
Neg logit contributions
 tune-0.084
 sensation-0.083
 dolphin-0.075
 shark-0.072
 potato-0.070
Token Tom
Token ID4186
Feature activation-2.98e-02

Pos logit contributions
 tune+0.047
 sensation+0.047
 dolphin+0.040
 shark+0.040
 chill+0.039
Neg logit contributions
ends-0.029
 counts-0.028
One-0.027
 Or-0.026
 ours-0.024
Token what
Token ID644
Feature activation+2.94e-02

Pos logit contributions
 sensation+0.015
 tune+0.015
 dolphin+0.013
 shark+0.013
 streamed+0.013
Neg logit contributions
 counts-0.008
ends-0.008
One-0.008
 Or-0.008
 Ready-0.007
Token you
Token ID345
Feature activation+2.37e-02

Pos logit contributions
 tune+0.071
 sensation+0.070
 dolphin+0.067
 shark+0.066
 potato+0.064
Neg logit contributions
ends-0.030
 counts-0.027
One-0.027
 Or-0.024
 Ready-0.022
Token have
Token ID423
Feature activation-3.31e-02

Pos logit contributions
 tune+0.054
 sensation+0.053
 dolphin+0.049
 shark+0.048
 streaming+0.046
Neg logit contributions
ends-0.028
 counts-0.028
One-0.027
 Or-0.025
 Ready-0.023
Token.
Token ID13
Feature activation-2.37e-02

Pos logit contributions
 counts+0.044
ends+0.044
One+0.042
 Or+0.040
 Ready+0.039
Neg logit contributions
 tune-0.069
 sensation-0.067
 dolphin-0.062
 shark-0.060
 face-0.059
Token<|endoftext|>
Token ID50256
Feature activation-4.20e-02

Pos logit contributions
ends+0.035
 counts+0.034
 ours+0.030
One+0.029
 belongs+0.029
Neg logit contributions
 tune-0.040
 Potato-0.039
 Sister-0.039
 Dolphin-0.039
 sensation-0.038
TokenOne
Token ID3198
Feature activation-1.55e-01

Pos logit contributions
ends+0.053
 counts+0.051
One+0.047
 Or+0.046
 ours+0.041
Neg logit contributions
 sensation-0.091
 tune-0.091
 dolphin-0.084
 shark-0.083
 Dolphin-0.079
Token day
Token ID1110
Feature activation+1.84e-02

Pos logit contributions
One+0.247
 counts+0.244
ends+0.222
 Or+0.204
 has+0.197
Neg logit contributions
 tune-0.299
 sensation-0.282
 streaming-0.269
 dolphin-0.260
 streamed-0.252
Token,
Token ID11
Feature activation-4.94e-03

Pos logit contributions
 tune+0.037
 sensation+0.036
 dolphin+0.034
 shark+0.033
 streaming+0.031
Neg logit contributions
 counts-0.026
ends-0.026
One-0.025
 Or-0.024
 Ready-0.022
Token a
Token ID257
Feature activation-1.63e-02

Pos logit contributions
 counts+0.008
ends+0.007
 Or+0.007
One+0.007
 Ready+0.006
Neg logit contributions
 sensation-0.009
 tune-0.009
 dolphin-0.008
 shark-0.008
 chill-0.008
Token boy
Token ID2933
Feature activation+1.41e-02

Pos logit contributions
One+0.030
ends+0.029
 Or+0.028
 counts+0.026
 has+0.026
Neg logit contributions
 sensation-0.028
 tune-0.026
 streaming-0.025
ened-0.023
 streamed-0.023
Token and
Token ID290
Feature activation+6.34e-03

Pos logit contributions
 tune+0.025
 sensation+0.025
 dolphin+0.023
 shark+0.022
 potato+0.021
Neg logit contributions
ends-0.024
 counts-0.023
One-0.022
 Or-0.021
 Ready-0.020
Token of
Token ID286
Feature activation+1.23e-02

Pos logit contributions
ends+0.042
 counts+0.042
One+0.040
 Or+0.038
 Ready+0.036
Neg logit contributions
 tune-0.044
 sensation-0.043
 dolphin-0.039
 shark-0.038
 potato-0.035
Token their
Token ID511
Feature activation+4.74e-04

Pos logit contributions
 tune+0.022
 sensation+0.022
 dolphin+0.019
 shark+0.019
 chill+0.018
Neg logit contributions
ends-0.021
 counts-0.021
One-0.020
 Or-0.019
 Ready-0.018
Token work
Token ID670
Feature activation+2.45e-03

Pos logit contributions
 tune+0.001
 sensation+0.001
 dolphin+0.001
 shark+0.001
 streaming+0.001
Neg logit contributions
 counts-0.001
ends-0.001
One-0.001
 Or-0.001
 Ready-0.001
Token.
Token ID13
Feature activation-3.09e-02

Pos logit contributions
 tune+0.006
 sensation+0.006
 dolphin+0.005
 shark+0.005
 streaming+0.005
Neg logit contributions
 counts-0.003
ends-0.003
One-0.003
 Or-0.003
 Ready-0.002
Token<|endoftext|>
Token ID50256
Feature activation-4.42e-02

Pos logit contributions
 counts+0.047
ends+0.045
One+0.041
 Or+0.038
 ours+0.038
Neg logit contributions
 tune-0.056
 sensation-0.056
 dolphin-0.055
 shark-0.052
 Dolphin-0.051
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.056
 counts+0.053
One+0.049
 Or+0.048
 ours+0.044
Neg logit contributions
 sensation-0.096
 tune-0.095
 dolphin-0.090
 shark-0.088
 streamed-0.084
Token day
Token ID1110
Feature activation+1.31e-02

Pos logit contributions
One+0.242
 counts+0.242
ends+0.225
 Or+0.203
 has+0.197
Neg logit contributions
 tune-0.301
 sensation-0.286
 streaming-0.266
 dolphin-0.257
 tant-0.253
Token,
Token ID11
Feature activation-6.04e-03

Pos logit contributions
 dolphin+0.020
 sensation+0.020
 shark+0.019
 tune+0.019
 streamed+0.018
Neg logit contributions
 counts-0.023
ends-0.022
 Or-0.022
One-0.022
 Ready-0.021
Token Lily
Token ID20037
Feature activation-4.63e-02

Pos logit contributions
 counts+0.010
ends+0.010
 Or+0.009
One+0.009
 Ready+0.009
Neg logit contributions
 sensation-0.010
 tune-0.010
 dolphin-0.009
 shark-0.009
 face-0.008
Token and
Token ID290
Feature activation+1.58e-02

Pos logit contributions
ends+0.076
One+0.071
 counts+0.071
 Or+0.067
 Ready+0.065
Neg logit contributions
 tune-0.085
 sensation-0.084
 dolphin-0.077
 shark-0.074
 potato-0.071
Token Max
Token ID5436
Feature activation+7.49e-03

Pos logit contributions
 tune+0.031
 sensation+0.030
 dolphin+0.027
 shark+0.026
 streaming+0.025
Neg logit contributions
ends-0.024
 counts-0.024
One-0.023
 Or-0.022
 Ready-0.021
Token cake
Token ID12187
Feature activation-5.52e-02

Pos logit contributions
ends+0.034
 counts+0.034
One+0.031
 Or+0.030
 Ready+0.027
Neg logit contributions
 tune-0.064
 sensation-0.064
 dolphin-0.059
 shark-0.057
 streaming-0.055
Token is
Token ID318
Feature activation-7.64e-02

Pos logit contributions
ends+0.072
 counts+0.070
One+0.070
 Or+0.064
 Ready+0.062
Neg logit contributions
 tune-0.118
 sensation-0.118
 dolphin-0.107
 shark-0.104
 face-0.103
Token inside
Token ID2641
Feature activation-5.53e-02

Pos logit contributions
 counts+0.113
ends+0.111
One+0.107
 Or+0.102
 Ready+0.095
Neg logit contributions
 tune-0.152
 sensation-0.151
 dolphin-0.137
 shark-0.132
 potato-0.127
Token.
Token ID13
Feature activation-2.94e-02

Pos logit contributions
ends+0.069
 counts+0.069
One+0.065
 Or+0.061
 Ready+0.055
Neg logit contributions
 tune-0.123
 sensation-0.121
 dolphin-0.111
 shark-0.108
 streaming-0.104
Token<|endoftext|>
Token ID50256
Feature activation-3.66e-02

Pos logit contributions
ends+0.047
 counts+0.046
One+0.042
 ours+0.040
 belongs+0.039
Neg logit contributions
 tune-0.045
 sensation-0.044
 dolphin-0.044
 Sister-0.044
 shark-0.042
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.050
 counts+0.047
One+0.044
 Or+0.044
 ours+0.041
Neg logit contributions
 sensation-0.074
 tune-0.073
 dolphin-0.068
 shark-0.068
 Dolphin-0.064
Token day
Token ID1110
Feature activation+2.53e-02

Pos logit contributions
 counts+0.274
One+0.270
ends+0.262
 Or+0.244
 Ready+0.225
Neg logit contributions
 tune-0.267
 sensation-0.254
 dolphin-0.233
 streaming-0.225
 shark-0.223
Token,
Token ID11
Feature activation-3.97e-03

Pos logit contributions
 tune+0.051
 sensation+0.050
 dolphin+0.046
 shark+0.045
 streaming+0.043
Neg logit contributions
 counts-0.036
ends-0.036
One-0.034
 Or-0.033
 Ready-0.030
Token a
Token ID257
Feature activation-1.71e-02

Pos logit contributions
 counts+0.006
ends+0.006
 Or+0.006
One+0.006
 Ready+0.005
Neg logit contributions
 sensation-0.007
 tune-0.007
 dolphin-0.007
 shark-0.007
 face-0.006
Token boy
Token ID2933
Feature activation+1.42e-02

Pos logit contributions
One+0.032
ends+0.032
 Or+0.030
 counts+0.028
 has+0.028
Neg logit contributions
 sensation-0.028
 tune-0.026
 streaming-0.025
ened-0.023
 streamed-0.023
Token named
Token ID3706
Feature activation-4.59e-02

Pos logit contributions
 tune+0.026
 sensation+0.025
 dolphin+0.023
 shark+0.022
 potato+0.021
Neg logit contributions
ends-0.024
 counts-0.023
One-0.022
 Or-0.021
 Ready-0.020
Token They
Token ID1119
Feature activation-9.09e-03

Pos logit contributions
 counts+0.031
ends+0.030
One+0.028
 Or+0.026
 belongs+0.025
Neg logit contributions
 tune-0.041
 sensation-0.040
 dolphin-0.038
 shark-0.036
 Dolphin-0.035
Token are
Token ID389
Feature activation-2.40e-02

Pos logit contributions
ends+0.012
 counts+0.012
One+0.011
 Or+0.011
 Ready+0.010
Neg logit contributions
 tune-0.020
 sensation-0.019
 dolphin-0.018
 shark-0.017
 chill-0.017
Token happy
Token ID3772
Feature activation-1.58e-02

Pos logit contributions
 counts+0.041
ends+0.041
One+0.040
 Or+0.038
 Ready+0.035
Neg logit contributions
 tune-0.042
 sensation-0.041
 dolphin-0.037
 shark-0.036
 streaming-0.034
Token.
Token ID13
Feature activation-2.87e-02

Pos logit contributions
ends+0.025
One+0.024
 counts+0.024
 Or+0.022
 Ready+0.021
Neg logit contributions
 tune-0.031
 sensation-0.029
 dolphin-0.028
 shark-0.028
 face-0.025
Token<|endoftext|>
Token ID50256
Feature activation-4.50e-02

Pos logit contributions
 counts+0.044
ends+0.043
One+0.039
 belongs+0.037
 ours+0.036
Neg logit contributions
 tune-0.052
 dolphin-0.049
 sensation-0.049
 shark-0.046
 Dolphin-0.046
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.066
 counts+0.061
 Or+0.057
One+0.057
 ours+0.055
Neg logit contributions
 sensation-0.087
 tune-0.085
 shark-0.082
 dolphin-0.081
 Dolphin-0.077
Token day
Token ID1110
Feature activation+1.75e-02

Pos logit contributions
One+0.244
 counts+0.240
ends+0.221
 Or+0.200
 has+0.197
Neg logit contributions
 tune-0.298
 sensation-0.285
 streaming-0.276
 streamed-0.259
 dolphin-0.257
Token,
Token ID11
Feature activation-3.75e-03

Pos logit contributions
 tune+0.028
 sensation+0.028
 dolphin+0.028
 shark+0.027
 streamed+0.025
Neg logit contributions
 counts-0.029
ends-0.029
One-0.028
 Or-0.028
 Ready-0.026
Token a
Token ID257
Feature activation-1.60e-02

Pos logit contributions
 counts+0.006
ends+0.006
 Or+0.006
One+0.006
 Ready+0.005
Neg logit contributions
 sensation-0.006
 tune-0.006
 dolphin-0.006
 shark-0.006
 face-0.005
Token boy
Token ID2933
Feature activation+1.36e-02

Pos logit contributions
One+0.030
ends+0.030
 Or+0.028
 counts+0.028
 has+0.026
Neg logit contributions
 sensation-0.026
 tune-0.024
 streaming-0.024
 streamed-0.022
ened-0.021
Token named
Token ID3706
Feature activation-4.74e-02

Pos logit contributions
 tune+0.023
 sensation+0.022
 dolphin+0.021
 shark+0.020
 potato+0.019
Neg logit contributions
ends-0.024
One-0.022
 counts-0.022
 Or-0.021
 Ready-0.021
Token other
Token ID584
Feature activation-6.65e-02

Pos logit contributions
 sensation+0.132
 tune+0.130
 dolphin+0.116
 shark+0.115
 streaming+0.114
Neg logit contributions
 counts-0.089
ends-0.087
One-0.087
 Or-0.079
 Ready-0.074
Token and
Token ID290
Feature activation-7.01e-03

Pos logit contributions
ends+0.082
 counts+0.082
One+0.076
 Or+0.073
 Ready+0.064
Neg logit contributions
 sensation-0.148
 tune-0.147
 dolphin-0.135
 shark-0.131
 streaming-0.127
Token waited
Token ID13488
Feature activation-3.61e-02

Pos logit contributions
 counts+0.009
ends+0.008
 Or+0.007
 has+0.007
 Maybe+0.006
Neg logit contributions
 tune-0.016
 sensation-0.015
andel-0.015
opher-0.014
 Dolphin-0.014
Token.
Token ID13
Feature activation-2.40e-02

Pos logit contributions
ends+0.039
 counts+0.035
One+0.034
 Or+0.033
 Ready+0.027
Neg logit contributions
 tune-0.086
 sensation-0.085
 dolphin-0.080
 shark-0.078
 Dolphin-0.076
Token<|endoftext|>
Token ID50256
Feature activation-3.87e-02

Pos logit contributions
 counts+0.039
ends+0.037
 ours+0.034
 belongs+0.034
One+0.034
Neg logit contributions
 tune-0.040
 sensation-0.039
 dolphin-0.038
 Dolphin-0.036
 shark-0.036
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.052
 counts+0.049
One+0.045
 Or+0.045
 ours+0.042
Neg logit contributions
 sensation-0.080
 tune-0.079
 dolphin-0.074
 shark-0.074
 Dolphin-0.070
Token day
Token ID1110
Feature activation+1.93e-02

Pos logit contributions
 counts+0.247
One+0.245
ends+0.227
 Or+0.208
 has+0.203
Neg logit contributions
 tune-0.295
 sensation-0.275
 streaming-0.260
 dolphin-0.254
 shark-0.244
Token,
Token ID11
Feature activation-7.94e-04

Pos logit contributions
 tune+0.038
 sensation+0.037
 dolphin+0.035
 shark+0.034
 streamed+0.032
Neg logit contributions
 counts-0.028
ends-0.028
One-0.027
 Or-0.026
 Ready-0.024
Token a
Token ID257
Feature activation-1.63e-02

Pos logit contributions
 counts+0.001
ends+0.001
One+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.002
 sensation-0.002
 dolphin-0.001
 shark-0.001
 streamed-0.001
Token big
Token ID1263
Feature activation+1.30e-02

Pos logit contributions
One+0.030
ends+0.030
 Or+0.028
 counts+0.027
 has+0.026
Neg logit contributions
 sensation-0.028
 tune-0.025
 streaming-0.024
ened-0.022
 streamed-0.022
Token storm
Token ID6388
Feature activation+5.48e-02

Pos logit contributions
 sensation+0.026
 tune+0.025
 streaming+0.023
 streamed+0.023
 lull+0.021
Neg logit contributions
 counts-0.018
One-0.017
 has-0.017
house-0.016
 Or-0.016
Token belong
Token ID5594
Feature activation-2.71e-02

Pos logit contributions
ends+0.007
 counts+0.007
One+0.007
 Or+0.006
 Ready+0.006
Neg logit contributions
 tune-0.012
 sensation-0.012
 dolphin-0.011
 shark-0.010
 streaming-0.010
Token to
Token ID284
Feature activation-2.14e-02

Pos logit contributions
One+0.052
ends+0.050
 counts+0.050
 Or+0.048
 Ready+0.048
Neg logit contributions
 tune-0.036
 shark-0.035
 sensation-0.033
 dolphin-0.033
 face-0.031
Token them
Token ID606
Feature activation-1.54e-02

Pos logit contributions
 counts+0.032
ends+0.032
One+0.030
 Or+0.029
 Ready+0.027
Neg logit contributions
 tune-0.042
 sensation-0.042
 dolphin-0.038
 shark-0.037
 streaming-0.035
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
 counts+0.022
ends+0.022
One+0.021
 Or+0.019
 Ready+0.019
Neg logit contributions
 tune-0.030
 sensation-0.029
 dolphin-0.027
 shark-0.027
 face-0.026
Token<|endoftext|>
Token ID50256
Feature activation-4.23e-02

Pos logit contributions
 counts+0.039
ends+0.036
One+0.032
 belongs+0.032
 ours+0.030
Neg logit contributions
 tune-0.030
 Sister-0.029
 sensation-0.029
 dolphin-0.028
 Ned-0.028
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.056
 counts+0.054
One+0.050
 Or+0.049
 ours+0.045
Neg logit contributions
 sensation-0.088
 tune-0.087
 dolphin-0.082
 shark-0.081
 Dolphin-0.077
Token day
Token ID1110
Feature activation+2.07e-02

Pos logit contributions
One+0.251
 counts+0.246
ends+0.229
 Or+0.210
 has+0.199
Neg logit contributions
 tune-0.292
 sensation-0.276
 streaming-0.260
 dolphin-0.252
 streamed-0.245
Token,
Token ID11
Feature activation-8.79e-04

Pos logit contributions
 tune+0.039
 sensation+0.039
 dolphin+0.036
 shark+0.035
 streamed+0.034
Neg logit contributions
 counts-0.031
ends-0.031
One-0.029
 Or-0.029
 Ready-0.026
Token a
Token ID257
Feature activation-1.52e-02

Pos logit contributions
 counts+0.001
ends+0.001
 Or+0.001
One+0.001
 Ready+0.001
Neg logit contributions
 sensation-0.002
 tune-0.002
 dolphin-0.001
 shark-0.001
 face-0.001
Token little
Token ID1310
Feature activation+3.15e-02

Pos logit contributions
One+0.029
ends+0.029
 Or+0.027
 counts+0.026
 Ready+0.025
Neg logit contributions
 sensation-0.025
 tune-0.023
 streaming-0.022
 streamed-0.020
ened-0.020
Token boy
Token ID2933
Feature activation+1.33e-02

Pos logit contributions
 sensation+0.057
 tune+0.055
 streaming+0.048
 streamed+0.047
 lull+0.046
Neg logit contributions
One-0.050
 counts-0.050
ends-0.048
 Or-0.047
house-0.043
Token love
Token ID1842
Feature activation+4.70e-03

Pos logit contributions
ends+0.025
 counts+0.025
One+0.024
 Or+0.023
 Ready+0.021
Neg logit contributions
 tune-0.034
 sensation-0.034
 dolphin-0.032
 shark-0.030
 face-0.030
Token to
Token ID284
Feature activation-1.73e-02

Pos logit contributions
 tune+0.009
 sensation+0.009
 dolphin+0.008
 shark+0.008
 streamed+0.007
Neg logit contributions
 counts-0.007
ends-0.007
One-0.007
 Or-0.007
 Ready-0.006
Token paint
Token ID7521
Feature activation-5.22e-02

Pos logit contributions
 counts+0.027
One+0.026
ends+0.025
 Or+0.024
 Ready+0.021
Neg logit contributions
 sensation-0.034
 tune-0.034
 dolphin-0.030
 shark-0.029
 streaming-0.029
Token.
Token ID13
Feature activation-3.58e-02

Pos logit contributions
ends+0.073
 counts+0.070
One+0.068
 Or+0.065
 Ready+0.061
Neg logit contributions
 tune-0.108
 sensation-0.108
 dolphin-0.097
 shark-0.094
 face-0.092
Token<|endoftext|>
Token ID50256
Feature activation-4.35e-02

Pos logit contributions
ends+0.053
 counts+0.053
One+0.048
 ours+0.046
 Or+0.045
Neg logit contributions
 tune-0.066
 sensation-0.065
 dolphin-0.062
 shark-0.059
 Dolphin-0.057
TokenOne
Token ID3198
Feature activation-1.54e-01

Pos logit contributions
ends+0.055
 counts+0.052
One+0.048
 Or+0.047
 ours+0.042
Neg logit contributions
 sensation-0.095
 tune-0.094
 dolphin-0.089
 shark-0.087
 Dolphin-0.083
Token day
Token ID1110
Feature activation+1.21e-02

Pos logit contributions
One+0.254
 counts+0.253
ends+0.242
 Or+0.220
 has+0.207
Neg logit contributions
 tune-0.285
 sensation-0.271
 streaming-0.247
 dolphin-0.246
 shark-0.243
Token,
Token ID11
Feature activation-4.37e-03

Pos logit contributions
 dolphin+0.019
 sensation+0.019
 tune+0.018
 shark+0.018
 streamed+0.017
Neg logit contributions
 counts-0.021
ends-0.020
 Or-0.020
One-0.020
 Ready-0.019
Token a
Token ID257
Feature activation-2.02e-02

Pos logit contributions
 counts+0.007
ends+0.007
One+0.007
 Or+0.006
 Ready+0.006
Neg logit contributions
 sensation-0.008
 tune-0.007
 dolphin-0.007
 shark-0.007
 face-0.006
Token boy
Token ID2933
Feature activation+5.99e-03

Pos logit contributions
One+0.036
ends+0.035
 Or+0.034
 counts+0.032
 has+0.030
Neg logit contributions
 sensation-0.035
 streaming-0.033
 tune-0.032
 streamed-0.030
ened-0.030
Token went
Token ID1816
Feature activation-7.93e-03

Pos logit contributions
 tune+0.010
 sensation+0.010
 dolphin+0.009
 shark+0.009
 potato+0.009
Neg logit contributions
ends-0.010
One-0.010
 counts-0.010
 Or-0.009
 Ready-0.009
Token a
Token ID257
Feature activation+8.79e-03

Pos logit contributions
 tune+0.042
 sensation+0.040
urg+0.036
 dolphin+0.036
uc+0.036
Neg logit contributions
One-0.019
 counts-0.019
ends-0.017
-0.015
 ours-0.014
Token big
Token ID1263
Feature activation+3.17e-02

Pos logit contributions
 sensation+0.016
 tune+0.015
urg+0.015
ened+0.014
 tunes+0.014
Neg logit contributions
One-0.013
ends-0.013
 Or-0.012
-0.012
 counts-0.011
Token smile
Token ID8212
Feature activation+8.90e-02

Pos logit contributions
 sensation+0.047
 tune+0.046
 tunes+0.040
urg+0.039
 Dolphin+0.039
Neg logit contributions
 counts-0.057
One-0.055
 has-0.053
ends-0.052
 Or-0.052
Token on
Token ID319
Feature activation+3.75e-02

Pos logit contributions
 sensation+0.219
 tune+0.214
 shark+0.210
 dolphin+0.206
 tant+0.205
Neg logit contributions
ends-0.062
One-0.055
 counts-0.051
-0.050
"-0.047
Token his
Token ID465
Feature activation+2.71e-02

Pos logit contributions
 sensation+0.069
 tune+0.066
 Dolphin+0.064
 dolphin+0.064
 toddler+0.063
Neg logit contributions
One-0.054
 counts-0.047
"-0.044
ends-0.043
 Or-0.041
Token face
Token ID1986
Feature activation+1.20e-01

Pos logit contributions
 sensation+0.030
 Dolphin+0.028
 streamed+0.028
 Ned+0.027
 shark+0.027
Neg logit contributions
 Or-0.053
ends-0.052
 counts-0.050
-0.049
One-0.047
Token.
Token ID13
Feature activation+2.49e-02

Pos logit contributions
 sensation+0.314
 tune+0.313
 dolphin+0.307
 shark+0.305
 Dolphin+0.295
Neg logit contributions
ends-0.077
One-0.074
 counts-0.074
-0.057
 Or-0.056
Token<|endoftext|>
Token ID50256
Feature activation-2.71e-02

Pos logit contributions
 tune+0.057
 sensation+0.056
 dolphin+0.052
 shark+0.050
 chill+0.049
Neg logit contributions
ends-0.030
 counts-0.030
One-0.029
 Or-0.028
 Ready-0.025
TokenOnce
Token ID7454
Feature activation+4.82e-03

Pos logit contributions
ends+0.035
 counts+0.033
One+0.031
 Or+0.031
 ours+0.028
Neg logit contributions
 sensation-0.058
 tune-0.058
 dolphin-0.054
 shark-0.053
 streamed-0.051
Token there
Token ID612
Feature activation-1.29e-02

Pos logit contributions
 tune+0.010
 sensation+0.010
 dolphin+0.009
 shark+0.009
 streaming+0.009
Neg logit contributions
 counts-0.006
ends-0.006
One-0.006
 Or-0.006
 Ready-0.005
Token was
Token ID373
Feature activation+1.84e-02

Pos logit contributions
 counts+0.018
ends+0.018
One+0.017
 Or+0.016
 Ready+0.015
Neg logit contributions
 tune-0.027
 sensation-0.026
 dolphin-0.024
 shark-0.023
 potato-0.022
Token with
Token ID351
Feature activation+1.71e-02

Pos logit contributions
 tune+0.073
 sensation+0.071
 dolphin+0.065
 shark+0.063
 streaming+0.061
Neg logit contributions
ends-0.046
 counts-0.046
One-0.044
 Or-0.042
 Ready-0.038
Token a
Token ID257
Feature activation+8.05e-03

Pos logit contributions
 tune+0.036
 sensation+0.035
 dolphin+0.031
 shark+0.031
 Dolphin+0.030
Neg logit contributions
 counts-0.024
ends-0.023
One-0.023
 Or-0.021
 Ready-0.019
Token smile
Token ID8212
Feature activation+8.86e-02

Pos logit contributions
 sensation+0.014
 tune+0.014
 streaming+0.012
 streamed+0.012
urg+0.012
Neg logit contributions
ends-0.013
One-0.013
 counts-0.012
 Or-0.012
 Ready-0.011
Token on
Token ID319
Feature activation+3.58e-02

Pos logit contributions
 sensation+0.208
 tune+0.206
 shark+0.194
 dolphin+0.189
 tant+0.187
Neg logit contributions
ends-0.086
 counts-0.082
One-0.075
 has-0.066
-0.065
Token her
Token ID607
Feature activation+7.89e-03

Pos logit contributions
 sensation+0.084
 tune+0.083
 shark+0.081
 dolphin+0.078
 Dolphin+0.076
Neg logit contributions
One-0.038
 counts-0.035
ends-0.034
 Or-0.030
She-0.027
Token face
Token ID1986
Feature activation+1.13e-01

Pos logit contributions
 sensation+0.009
 streamed+0.008
 tant+0.008
 toddler+0.008
 Sister+0.007
Neg logit contributions
ends-0.015
 counts-0.014
 Or-0.014
 next-0.014
 has-0.014
Token.
Token ID13
Feature activation+1.37e-02

Pos logit contributions
 tune+0.286
 sensation+0.285
 shark+0.270
 dolphin+0.269
 Dolphin+0.258
Neg logit contributions
 counts-0.096
ends-0.095
One-0.088
 Or-0.079
-0.066
Token What
Token ID1867
Feature activation+2.71e-02

Pos logit contributions
 tune+0.029
 sensation+0.028
 dolphin+0.026
 shark+0.025
 streaming+0.024
Neg logit contributions
 counts-0.018
ends-0.018
One-0.017
 Or-0.016
 ours-0.015
Token an
Token ID281
Feature activation-8.50e-03

Pos logit contributions
 tune+0.051
 sensation+0.051
 dolphin+0.047
 shark+0.045
 face+0.043
Neg logit contributions
ends-0.041
 counts-0.040
One-0.038
 Or-0.037
 Ready-0.035
Token amazing
Token ID4998
Feature activation+1.24e-02

Pos logit contributions
ends+0.017
 counts+0.016
One+0.015
 Or+0.015
 Ready+0.014
Neg logit contributions
 tune-0.013
 sensation-0.013
 dolphin-0.012
 shark-0.011
 face-0.011
Token adventure
Token ID8855
Feature activation+5.43e-03

Pos logit contributions
 sensation+0.021
 tune+0.020
 streaming+0.019
 streamed+0.019
 dolphin+0.017
Neg logit contributions
 counts-0.021
One-0.020
ends-0.019
 Or-0.018
house-0.017
Token stubborn
Token ID24444
Feature activation-1.30e-02

Pos logit contributions
ends+0.045
 counts+0.045
One+0.043
 Or+0.041
 Ready+0.038
Neg logit contributions
 tune-0.056
 sensation-0.056
 dolphin-0.050
 shark-0.049
 streaming-0.047
Token.
Token ID13
Feature activation-2.98e-02

Pos logit contributions
ends+0.018
 counts+0.017
One+0.017
 Or+0.016
 Ready+0.016
Neg logit contributions
 tune-0.027
 sensation-0.027
 dolphin-0.024
 shark-0.023
 face-0.023
Token She
Token ID1375
Feature activation-3.48e-02

Pos logit contributions
ends+0.034
 counts+0.033
One+0.031
 ours+0.027
 Or+0.026
Neg logit contributions
 sensation-0.067
 tune-0.067
 dolphin-0.061
 face-0.059
 shark-0.059
Token throws
Token ID12542
Feature activation-1.26e-02

Pos logit contributions
ends+0.043
One+0.043
 counts+0.039
 Or+0.038
 ours+0.037
Neg logit contributions
 sensation-0.069
 tune-0.067
 face-0.066
 dolphin-0.065
 shark-0.064
Token a
Token ID257
Feature activation+1.09e-04

Pos logit contributions
 counts+0.020
ends+0.020
One+0.019
 Or+0.018
 Ready+0.017
Neg logit contributions
 tune-0.024
 sensation-0.023
 dolphin-0.021
 shark-0.020
 streaming-0.019
Token tant
Token ID24246
Feature activation+1.68e-01

Pos logit contributions
 sensation+0.000
 tune+0.000
 streaming+0.000
 dolphin+0.000
 Dolphin+0.000
Neg logit contributions
ends-0.000
One-0.000
 counts-0.000
 Or-0.000
 Ready-0.000
Tokenrum
Token ID6582
Feature activation-8.92e-03

Pos logit contributions
 tune+0.256
 sensation+0.250
 dolphin+0.221
 shark+0.211
 streaming+0.199
Neg logit contributions
ends-0.327
 counts-0.326
One-0.314
 Or-0.304
 Ready-0.286
Token.
Token ID13
Feature activation-3.18e-02

Pos logit contributions
 counts+0.013
One+0.013
ends+0.013
 Ready+0.012
 Or+0.012
Neg logit contributions
 sensation-0.017
 tune-0.017
 dolphin-0.015
 face-0.015
 streaming-0.015
Token She
Token ID1375
Feature activation-3.37e-02

Pos logit contributions
ends+0.036
 counts+0.036
One+0.033
 Or+0.029
 ours+0.028
Neg logit contributions
 sensation-0.072
 tune-0.072
 dolphin-0.066
 shark-0.064
 face-0.063
Token kicks
Token ID17364
Feature activation-3.17e-03

Pos logit contributions
ends+0.043
One+0.043
 counts+0.039
 Or+0.038
 Ready+0.036
Neg logit contributions
 sensation-0.068
 tune-0.067
 shark-0.065
 dolphin-0.065
 face-0.063
Token and
Token ID290
Feature activation+9.20e-03

Pos logit contributions
 counts+0.005
ends+0.005
One+0.005
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.006
 sensation-0.006
 dolphin-0.005
 shark-0.005
 streaming-0.005
Token and
Token ID290
Feature activation-9.38e-04

Pos logit contributions
ends+0.011
 counts+0.011
One+0.011
 Or+0.010
 Ready+0.009
Neg logit contributions
 tune-0.019
 sensation-0.019
 dolphin-0.017
 shark-0.017
 chill-0.016
Token jump
Token ID4391
Feature activation-1.86e-02

Pos logit contributions
 counts+0.001
ends+0.001
One+0.001
 Or+0.001
 Ready+0.001
Neg logit contributions
 tune-0.002
 sensation-0.002
 dolphin-0.002
 shark-0.002
 streaming-0.002
Token in
Token ID287
Feature activation+3.79e-03

Pos logit contributions
ends+0.031
 counts+0.031
One+0.030
 Or+0.028
 Ready+0.026
Neg logit contributions
 tune-0.034
 sensation-0.033
 dolphin-0.030
 shark-0.029
 streaming-0.027
Token the
Token ID262
Feature activation-4.89e-03

Pos logit contributions
 sensation+0.005
 tune+0.005
 dolphin+0.004
 shark+0.004
 face+0.004
Neg logit contributions
ends-0.009
 counts-0.008
 Or-0.008
One-0.008
 Ready-0.007
Token p
Token ID279
Feature activation+1.56e-02

Pos logit contributions
 counts+0.008
ends+0.008
One+0.008
 Or+0.008
 Ready+0.007
Neg logit contributions
 tune-0.009
 sensation-0.008
 dolphin-0.008
 shark-0.007
 streaming-0.007
Tokenudd
Token ID4185
Feature activation+1.09e-01

Pos logit contributions
 sensation+0.018
 tune+0.018
 dolphin+0.016
 face+0.016
lee+0.015
Neg logit contributions
 counts-0.034
One-0.032
ends-0.031
 Or-0.031
 Ready-0.031
Tokenles
Token ID829
Feature activation+3.18e-02

Pos logit contributions
 tune+0.170
 sensation+0.165
 dolphin+0.143
 shark+0.142
 Dolphin+0.141
Neg logit contributions
 counts-0.197
One-0.192
ends-0.186
 belongs-0.176
 ours-0.168
Token.
Token ID13
Feature activation+1.76e-03

Pos logit contributions
 tune+0.072
 sensation+0.071
 dolphin+0.065
 shark+0.063
 streaming+0.061
Neg logit contributions
 counts-0.039
ends-0.039
One-0.037
 Or-0.035
 Ready-0.032
Token One
Token ID1881
Feature activation-1.01e-01

Pos logit contributions
 tune+0.004
 sensation+0.004
 face+0.003
 stomach+0.003
 dolphin+0.003
Neg logit contributions
ends-0.002
 counts-0.002
 ours-0.002
 Or-0.002
ider-0.002
Token day
Token ID1110
Feature activation+4.05e-02

Pos logit contributions
One+0.162
 counts+0.153
ends+0.147
 Or+0.137
house+0.128
Neg logit contributions
 tune-0.191
 sensation-0.186
 streaming-0.170
 dolphin-0.167
 shark-0.166
Token,
Token ID11
Feature activation+2.57e-02

Pos logit contributions
 tune+0.085
 sensation+0.084
 dolphin+0.077
 shark+0.074
 streaming+0.071
Neg logit contributions
ends-0.056
 counts-0.056
One-0.053
 Or-0.050
 Ready-0.046
Token work
Token ID670
Feature activation+2.07e-02

Pos logit contributions
 sensation+0.142
 tune+0.141
 dolphin+0.131
 shark+0.125
 streaming+0.121
Neg logit contributions
 counts-0.121
ends-0.115
One-0.113
 Or-0.105
 Ready-0.096
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
 tune+0.050
 sensation+0.049
 dolphin+0.046
 shark+0.045
 potato+0.043
Neg logit contributions
 counts-0.022
ends-0.021
One-0.019
 Or-0.018
 Ready-0.015
Token
Token ID220
Feature activation+9.25e-03

Pos logit contributions
 tune+0.028
 sensation+0.028
 dolphin+0.025
 shark+0.024
 chill+0.024
Neg logit contributions
ends-0.019
 counts-0.018
One-0.018
 Or-0.017
 Ready-0.016
Token
Token ID198
Feature activation+1.62e-02

Pos logit contributions
 tune+0.019
 sensation+0.019
 dolphin+0.017
 shark+0.017
 Dolphin+0.016
Neg logit contributions
 counts-0.012
ends-0.012
One-0.011
 Or-0.010
 ours-0.010
Token
Token ID198
Feature activation+2.64e-02

Pos logit contributions
 tune+0.038
 sensation+0.038
 dolphin+0.035
 shark+0.034
 streaming+0.033
Neg logit contributions
ends-0.018
 counts-0.018
One-0.017
 Or-0.016
 Ready-0.014
TokenOl
Token ID30098
Feature activation+1.19e-01

Pos logit contributions
 tune+0.054
 sensation+0.053
 dolphin+0.048
 shark+0.047
 streaming+0.045
Neg logit contributions
ends-0.038
 counts-0.038
One-0.036
 Or-0.034
 Ready-0.032
Tokeniver
Token ID1428
Feature activation+5.86e-02

Pos logit contributions
 tune+0.205
 dolphin+0.199
 sensation+0.197
 shark+0.192
 Dolphin+0.177
Neg logit contributions
 counts-0.201
One-0.199
ends-0.194
 Or-0.190
 Ready-0.168
Token smiled
Token ID13541
Feature activation-1.93e-02

Pos logit contributions
 sensation+0.116
andel+0.115
 tune+0.115
 Dolphin+0.110
 dolphin+0.110
Neg logit contributions
 counts-0.061
ends-0.060
 has-0.058
house-0.056
-0.055
Token and
Token ID290
Feature activation+1.16e-02

Pos logit contributions
ends+0.014
 Or+0.012
One+0.011
 counts+0.011
+0.010
Neg logit contributions
 sensation-0.051
 tune-0.048
 shark-0.046
roach-0.046
 dolphin-0.046
Token be
Token ID307
Feature activation+1.61e-02

Pos logit contributions
 sensation+0.027
 tune+0.026
andel+0.025
 tant+0.024
ascal+0.024
Neg logit contributions
 counts-0.013
ends-0.013
-0.012
 has-0.011
 Or-0.010
Tokenamed
Token ID2434
Feature activation-2.94e-03

Pos logit contributions
 tune+0.028
 sensation+0.027
 dolphin+0.025
 shark+0.025
 potato+0.023
Neg logit contributions
ends-0.029
 counts-0.027
One-0.027
 Or-0.025
 Ready-0.023
Token a
Token ID257
Feature activation+1.96e-02

Pos logit contributions
ends+0.024
 counts+0.023
One+0.022
 Or+0.021
 Ready+0.019
Neg logit contributions
 tune-0.037
 sensation-0.037
 dolphin-0.033
 shark-0.033
 chill-0.031
Token few
Token ID1178
Feature activation+5.64e-02

Pos logit contributions
 sensation+0.038
 tune+0.035
 dolphin+0.033
 streaming+0.033
 shark+0.032
Neg logit contributions
One-0.032
ends-0.030
 counts-0.028
 Or-0.027
 Ready-0.026
Token days
Token ID1528
Feature activation+3.88e-02

Pos logit contributions
 tune+0.098
 sensation+0.097
 dolphin+0.085
 shark+0.082
 chill+0.079
Neg logit contributions
ends-0.098
 counts-0.094
One-0.092
 Or-0.090
 Ready-0.084
Token,
Token ID11
Feature activation+2.06e-02

Pos logit contributions
 tune+0.077
 sensation+0.076
 dolphin+0.069
 shark+0.067
 streaming+0.065
Neg logit contributions
ends-0.056
 counts-0.056
One-0.054
 Or-0.051
 Ready-0.047
Token they
Token ID484
Feature activation+2.89e-02

Pos logit contributions
 tune+0.053
 sensation+0.052
 dolphin+0.047
 shark+0.047
 chill+0.046
Neg logit contributions
ends-0.019
 counts-0.017
One-0.017
 Or-0.016
house-0.015
Token got
Token ID1392
Feature activation+2.47e-02

Pos logit contributions
 tune+0.064
 sensation+0.063
 dolphin+0.057
 shark+0.056
 Dolphin+0.053
Neg logit contributions
 counts-0.038
ends-0.038
One-0.036
 Or-0.034
house-0.031
Token the
Token ID262
Feature activation-1.61e-02

Pos logit contributions
 tune+0.043
 sensation+0.043
 dolphin+0.039
 shark+0.037
 face+0.036
Neg logit contributions
 counts-0.042
ends-0.041
One-0.040
 Or-0.038
 Ready-0.036
Token book
Token ID1492
Feature activation-1.21e-02

Pos logit contributions
ends+0.027
 counts+0.026
One+0.025
 Or+0.025
atters+0.023
Neg logit contributions
 tune-0.029
 sensation-0.029
 face-0.025
 dolphin-0.024
 lull-0.024
Token back
Token ID736
Feature activation-3.99e-03

Pos logit contributions
 counts+0.019
ends+0.019
One+0.019
 Ready+0.017
 Or+0.017
Neg logit contributions
 tune-0.022
 sensation-0.021
 face-0.019
 streaming-0.018
 chill-0.018
Token and
Token ID290
Feature activation+1.15e-03

Pos logit contributions
 counts+0.006
ends+0.006
One+0.006
 Or+0.005
 Ready+0.005
Neg logit contributions
 tune-0.008
 sensation-0.008
 dolphin-0.007
 shark-0.007
 streaming-0.007
Token it
Token ID340
Feature activation+2.71e-02

Pos logit contributions
 tune+0.002
 sensation+0.002
 dolphin+0.002
 shark+0.002
 Dolphin+0.002
Neg logit contributions
 counts-0.002
ends-0.002
One-0.002
 Or-0.002
 Ready-0.002
Token night
Token ID1755
Feature activation-8.43e-03

Pos logit contributions
 counts+0.036
ends+0.035
One+0.034
 Or+0.033
 Ready+0.031
Neg logit contributions
 tune-0.027
 sensation-0.026
 dolphin-0.024
 shark-0.022
 potato-0.021
Token,
Token ID11
Feature activation+8.37e-03

Pos logit contributions
ends+0.010
 counts+0.009
One+0.009
 Or+0.008
house+0.007
Neg logit contributions
 sensation-0.019
 tune-0.019
 dolphin-0.018
 shark-0.018
 streamed-0.017
Token but
Token ID475
Feature activation+2.75e-02

Pos logit contributions
 tune+0.013
 sensation+0.013
 dolphin+0.013
 shark+0.012
 Dolphin+0.012
Neg logit contributions
ends-0.015
 counts-0.015
 Or-0.014
One-0.014
 Ready-0.012
Token eventually
Token ID4191
Feature activation-4.94e-02

Pos logit contributions
 tune+0.065
 sensation+0.064
 dolphin+0.059
 shark+0.058
 potato+0.056
Neg logit contributions
ends-0.031
 counts-0.031
One-0.029
 Or-0.027
 Ready-0.024
Token fell
Token ID3214
Feature activation+1.10e-03

Pos logit contributions
 counts+0.063
ends+0.062
One+0.058
 Or+0.054
 Ready+0.049
Neg logit contributions
 tune-0.111
 sensation-0.109
 dolphin-0.100
 shark-0.097
 chill-0.093
Token asleep
Token ID16039
Feature activation+3.97e-02

Pos logit contributions
 tune+0.002
 sensation+0.002
 dolphin+0.001
 shark+0.001
 potato+0.001
Neg logit contributions
 counts-0.002
ends-0.002
One-0.002
 Or-0.002
 Ready-0.002
Token.
Token ID13
Feature activation+2.22e-02

Pos logit contributions
 sensation+0.092
 tune+0.092
 dolphin+0.091
 shark+0.087
 potato+0.087
Neg logit contributions
ends-0.041
 counts-0.038
One-0.036
 Or-0.036
house-0.031
Token The
Token ID383
Feature activation+1.34e-02

Pos logit contributions
 tune+0.045
 sensation+0.045
 dolphin+0.040
 shark+0.039
 streaming+0.038
Neg logit contributions
 counts-0.032
ends-0.032
One-0.030
 Or-0.028
 Ready-0.026
Token monster
Token ID9234
Feature activation+4.05e-02

Pos logit contributions
 tune+0.025
 sensation+0.025
 dolphin+0.022
 shark+0.021
 streaming+0.021
Neg logit contributions
 counts-0.021
ends-0.021
One-0.021
 Or-0.019
 Ready-0.018
Token came
Token ID1625
Feature activation+1.88e-02

Pos logit contributions
 sensation+0.079
 tune+0.077
 dolphin+0.071
 Dolphin+0.068
 potato+0.068
Neg logit contributions
 counts-0.062
ends-0.058
One-0.056
 Or-0.055
house-0.050
Token back
Token ID736
Feature activation-1.04e-02

Pos logit contributions
 tune+0.035
 sensation+0.034
 dolphin+0.031
 shark+0.030
 streaming+0.029
Neg logit contributions
 counts-0.031
ends-0.030
One-0.029
 Or-0.028
 Ready-0.026
Token colors
Token ID7577
Feature activation+8.15e-03

Pos logit contributions
 tune+0.074
 sensation+0.074
 dolphin+0.068
 shark+0.066
 Dolphin+0.063
Neg logit contributions
 counts-0.065
ends-0.062
One-0.060
 Or-0.057
 Ready-0.053
Token filled
Token ID5901
Feature activation+1.97e-02

Pos logit contributions
 tune+0.016
 sensation+0.015
 shark+0.014
 dolphin+0.014
 Dolphin+0.014
Neg logit contributions
 counts-0.012
ends-0.012
One-0.011
 Or-0.010
 has-0.010
Token the
Token ID262
Feature activation-6.91e-03

Pos logit contributions
 tune+0.040
 sensation+0.039
 shark+0.037
 dolphin+0.037
 streaming+0.035
Neg logit contributions
One-0.026
 counts-0.026
ends-0.026
 Or-0.022
 ours-0.020
Token room
Token ID2119
Feature activation+1.91e-03

Pos logit contributions
ends+0.012
 counts+0.012
One+0.012
 Or+0.011
 Ready+0.010
Neg logit contributions
 tune-0.012
 sensation-0.012
 dolphin-0.010
 shark-0.010
 face-0.010
Token with
Token ID351
Feature activation-9.46e-03

Pos logit contributions
 tune+0.004
 sensation+0.004
 dolphin+0.003
 shark+0.003
 streaming+0.003
Neg logit contributions
ends-0.003
 counts-0.003
One-0.003
 Or-0.003
 Ready-0.002
Token joy
Token ID8716
Feature activation+4.10e-02

Pos logit contributions
 counts+0.014
ends+0.014
One+0.013
 Or+0.012
 Ready+0.012
Neg logit contributions
 tune-0.018
 sensation-0.018
 shark-0.016
 dolphin-0.016
 Dolphin-0.016
Token.
Token ID13
Feature activation+1.93e-02

Pos logit contributions
 tune+0.099
 sensation+0.099
 shark+0.095
 dolphin+0.095
 Dolphin+0.092
Neg logit contributions
ends-0.038
 counts-0.037
One-0.029
 has-0.028
 Or-0.027
Token
Token ID198
Feature activation+1.81e-02

Pos logit contributions
 sensation+0.038
 tune+0.038
 dolphin+0.035
 shark+0.033
 Dolphin+0.033
Neg logit contributions
 counts-0.027
ends-0.027
One-0.025
 Or-0.024
 ours-0.022
Token
Token ID198
Feature activation+2.76e-02

Pos logit contributions
 tune+0.038
 sensation+0.038
 dolphin+0.034
 shark+0.033
 streaming+0.032
Neg logit contributions
ends-0.025
 counts-0.025
One-0.024
 Or-0.022
 Ready-0.020
TokenCl
Token ID2601
Feature activation+8.44e-03

Pos logit contributions
 tune+0.051
 sensation+0.049
 dolphin+0.045
 lull+0.041
 shark+0.041
Neg logit contributions
 counts-0.047
ends-0.043
 Or-0.041
One-0.039
 Ready-0.038
Tokenara
Token ID3301
Feature activation+9.93e-03

Pos logit contributions
 tune+0.019
 sensation+0.019
 dolphin+0.017
 shark+0.017
 streamed+0.016
Neg logit contributions
 counts-0.010
ends-0.010
One-0.009
 Or-0.009
 Ready-0.008
Token.
Token ID13
Feature activation+9.15e-03

Pos logit contributions
 counts+0.009
ends+0.009
One+0.009
 Or+0.008
house+0.007
Neg logit contributions
 sensation-0.020
 tune-0.019
 dolphin-0.018
 shark-0.017
 chill-0.017
Token It
Token ID632
Feature activation+3.67e-03

Pos logit contributions
 tune+0.018
 sensation+0.017
 dolphin+0.016
 shark+0.015
 potato+0.015
Neg logit contributions
 counts-0.014
One-0.014
ends-0.013
 Or-0.013
 Ready-0.012
Token had
Token ID550
Feature activation+5.71e-03

Pos logit contributions
 tune+0.008
 sensation+0.008
 dolphin+0.007
 shark+0.007
 streaming+0.007
Neg logit contributions
 counts-0.005
ends-0.005
One-0.005
 Or-0.005
 belongs-0.004
Token a
Token ID257
Feature activation-1.55e-02

Pos logit contributions
 tune+0.013
 sensation+0.013
 dolphin+0.012
 shark+0.012
 Dolphin+0.011
Neg logit contributions
 counts-0.007
ends-0.007
One-0.006
 Or-0.006
 Ready-0.005
Token very
Token ID845
Feature activation+6.87e-02

Pos logit contributions
ends+0.021
One+0.021
 Or+0.020
 counts+0.020
 has+0.018
Neg logit contributions
 sensation-0.031
 tune-0.030
 streaming-0.027
 streamed-0.027
iness-0.027
Token important
Token ID1593
Feature activation+4.13e-02

Pos logit contributions
 tune+0.132
 sensation+0.131
 dolphin+0.115
 streaming+0.115
 streamed+0.111
Neg logit contributions
ends-0.107
 counts-0.103
One-0.100
 Or-0.094
 ours-0.087
Token job
Token ID1693
Feature activation+3.69e-02

Pos logit contributions
 tune+0.077
 sensation+0.077
lee+0.074
 streaming+0.072
 dolphin+0.072
Neg logit contributions
 counts-0.058
One-0.055
 Or-0.054
ends-0.053
 Ready-0.048
Token -
Token ID532
Feature activation-3.02e-02

Pos logit contributions
 tune+0.080
 sensation+0.079
 dolphin+0.073
 shark+0.070
 chill+0.068
Neg logit contributions
 counts-0.048
ends-0.047
One-0.044
 Or-0.042
 Ready-0.038
Token to
Token ID284
Feature activation+4.14e-03

Pos logit contributions
ends+0.048
 counts+0.047
One+0.045
 Or+0.043
 Ready+0.039
Neg logit contributions
 tune-0.056
 sensation-0.055
 dolphin-0.051
 shark-0.050
 potato-0.047
Token power
Token ID1176
Feature activation-6.56e-03

Pos logit contributions
 sensation+0.008
 tune+0.008
 dolphin+0.008
 streaming+0.007
 shark+0.007
Neg logit contributions
One-0.006
 counts-0.006
ends-0.006
 Or-0.006
 Ready-0.005
Token a
Token ID257
Feature activation-2.40e-02

Pos logit contributions
 counts+0.009
ends+0.009
One+0.008
 Or+0.008
 Ready+0.007
Neg logit contributions
 tune-0.014
 sensation-0.014
 dolphin-0.013
 shark-0.012
 chill-0.012
Token the
Token ID262
Feature activation-6.55e-03

Pos logit contributions
 tune+0.010
 sensation+0.010
 dolphin+0.009
 shark+0.009
 streaming+0.009
Neg logit contributions
 counts-0.009
ends-0.009
One-0.009
 Or-0.008
 Ready-0.008
Token university
Token ID6403
Feature activation-6.13e-02

Pos logit contributions
ends+0.009
 counts+0.009
One+0.008
 Or+0.008
 ours+0.007
Neg logit contributions
 tune-0.013
 sensation-0.013
 dolphin-0.012
 face-0.011
 shark-0.011
Token was
Token ID373
Feature activation-2.38e-02

Pos logit contributions
ends+0.091
 counts+0.086
One+0.084
 Or+0.080
 Ready+0.077
Neg logit contributions
 tune-0.123
 sensation-0.115
 dolphin-0.103
 shark-0.100
 face-0.098
Token coming
Token ID2406
Feature activation+3.87e-02

Pos logit contributions
ends+0.034
 counts+0.034
One+0.032
 Or+0.031
 Ready+0.028
Neg logit contributions
 tune-0.048
 sensation-0.047
 dolphin-0.043
 shark-0.042
 streaming-0.040
Token true
Token ID2081
Feature activation-5.95e-02

Pos logit contributions
 tune+0.076
 sensation+0.073
 dolphin+0.066
 shark+0.064
 face+0.063
Neg logit contributions
 counts-0.059
ends-0.059
One-0.054
 Or-0.053
 Ready-0.049
Token.
Token ID13
Feature activation+2.10e-02

Pos logit contributions
ends+0.072
 counts+0.071
One+0.067
 Or+0.064
 Ready+0.057
Neg logit contributions
 tune-0.134
 sensation-0.133
 dolphin-0.122
 shark-0.119
 streaming-0.114
Token<|endoftext|>
Token ID50256
Feature activation-3.13e-02

Pos logit contributions
 tune+0.044
 sensation+0.043
 dolphin+0.040
 shark+0.038
 chill+0.037
Neg logit contributions
 counts-0.029
ends-0.029
One-0.028
 Or-0.027
 Ready-0.024
TokenOnce
Token ID7454
Feature activation+2.84e-04

Pos logit contributions
ends+0.039
 counts+0.038
One+0.035
 Or+0.034
 ours+0.030
Neg logit contributions
 sensation-0.068
 tune-0.068
 dolphin-0.063
 shark-0.062
 streamed-0.059
Token there
Token ID612
Feature activation-1.73e-02

Pos logit contributions
 tune+0.001
 sensation+0.001
 dolphin+0.001
 shark+0.001
 potato+0.001
Neg logit contributions
 counts-0.000
ends-0.000
One-0.000
 Or-0.000
 Ready-0.000
Token was
Token ID373
Feature activation+1.28e-02

Pos logit contributions
 counts+0.024
ends+0.023
One+0.023
 Or+0.021
 Ready+0.019
Neg logit contributions
 tune-0.037
 sensation-0.035
 dolphin-0.033
 shark-0.031
 potato-0.031
Token a
Token ID257
Feature activation-2.90e-02

Pos logit contributions
 sensation+0.023
 tune+0.023
 dolphin+0.021
 shark+0.020
 potato+0.019
Neg logit contributions
 counts-0.021
ends-0.021
 Or-0.019
One-0.019
 Ready-0.018
Token was
Token ID373
Feature activation-1.36e-02

Pos logit contributions
ends+0.030
 counts+0.030
One+0.029
 Or+0.028
 Ready+0.026
Neg logit contributions
 tune-0.033
 sensation-0.032
 dolphin-0.030
 shark-0.028
 potato-0.027
Token making
Token ID1642
Feature activation+7.59e-02

Pos logit contributions
ends+0.020
One+0.018
 counts+0.017
 Or+0.016
house+0.015
Neg logit contributions
 tune-0.028
 sensation-0.028
 dolphin-0.026
 shark-0.025
 Dolphin-0.024
Token his
Token ID465
Feature activation-1.02e-02

Pos logit contributions
 tune+0.147
 sensation+0.138
 streamed+0.130
 streaming+0.126
 tant+0.124
Neg logit contributions
ends-0.109
One-0.107
 counts-0.106
 Or-0.098
 you-0.090
Token thick
Token ID6546
Feature activation+3.05e-02

Pos logit contributions
 counts+0.018
ends+0.017
One+0.017
 Or+0.016
 Ready+0.015
Neg logit contributions
 tune-0.017
 sensation-0.017
 dolphin-0.016
 shark-0.015
 streaming-0.015
Token toast
Token ID27805
Feature activation+3.46e-02

Pos logit contributions
 tune+0.049
 sensation+0.048
 dolphin+0.044
 shark+0.042
 Dolphin+0.042
Neg logit contributions
 counts-0.056
ends-0.052
One-0.051
 Or-0.050
house-0.046
Token in
Token ID287
Feature activation+7.34e-04

Pos logit contributions
 tune+0.075
 sensation+0.074
 dolphin+0.068
 shark+0.066
 streaming+0.064
Neg logit contributions
ends-0.044
 counts-0.044
One-0.042
 Or-0.040
 Ready-0.035
Token the
Token ID262
Feature activation-2.01e-03

Pos logit contributions
 tune+0.001
 sensation+0.001
 dolphin+0.001
 shark+0.001
 Dolphin+0.001
Neg logit contributions
One-0.001
 counts-0.001
ends-0.001
 Or-0.001
 ours-0.001
Token kitchen
Token ID9592
Feature activation+1.17e-02

Pos logit contributions
One+0.003
 counts+0.003
 Or+0.002
ends+0.002
 Ready+0.002
Neg logit contributions
 tune-0.004
 sensation-0.004
 dolphin-0.004
 streaming-0.004
 shark-0.004
Token.
Token ID13
Feature activation+9.27e-03

Pos logit contributions
 tune+0.028
 sensation+0.028
 dolphin+0.026
 shark+0.025
 streamed+0.025
Neg logit contributions
 counts-0.012
ends-0.012
One-0.011
 Or-0.011
 Ready-0.009
Token She
Token ID1375
Feature activation-1.12e-02

Pos logit contributions
 tune+0.021
 sensation+0.021
 dolphin+0.019
 shark+0.019
 Dolphin+0.018
Neg logit contributions
 counts-0.011
ends-0.011
One-0.011
 Or-0.010
 Ready-0.009
Token put
Token ID1234
Feature activation+1.51e-02

Pos logit contributions
ends+0.018
 counts+0.018
One+0.017
 Or+0.017
 Ready+0.015
Neg logit contributions
 tune-0.021
 sensation-0.020
 dolphin-0.019
 shark-0.018
 streaming-0.017
Token inside
Token ID2641
Feature activation-3.54e-02

Pos logit contributions
 counts+0.015
ends+0.015
One+0.014
 Or+0.013
 Ready+0.012
Neg logit contributions
 tune-0.027
 sensation-0.026
 dolphin-0.024
 shark-0.023
 streaming-0.022
Token.
Token ID13
Feature activation+2.59e-02

Pos logit contributions
 counts+0.043
ends+0.041
One+0.040
 Or+0.037
 Ready+0.033
Neg logit contributions
 tune-0.080
 sensation-0.078
 dolphin-0.072
 shark-0.070
 chill-0.068
Token<|endoftext|>
Token ID50256
Feature activation-2.71e-02

Pos logit contributions
 tune+0.061
 sensation+0.061
 dolphin+0.056
 shark+0.054
 chill+0.053
Neg logit contributions
 counts-0.029
ends-0.029
One-0.027
 Or-0.026
 Ready-0.023
TokenJohn
Token ID7554
Feature activation-1.95e-02

Pos logit contributions
ends+0.031
 counts+0.031
One+0.029
 Or+0.027
 Ready+0.024
Neg logit contributions
 tune-0.062
 sensation-0.062
 dolphin-0.057
 shark-0.056
 streaming-0.054
Token and
Token ID290
Feature activation+1.92e-02

Pos logit contributions
 counts+0.029
ends+0.029
One+0.027
 Or+0.026
 Ready+0.024
Neg logit contributions
 tune-0.039
 sensation-0.038
 dolphin-0.035
 shark-0.034
 streaming-0.032
Token his
Token ID465
Feature activation+9.44e-04

Pos logit contributions
 sensation+0.036
 tune+0.034
andel+0.029
 tant+0.029
 chill+0.028
Neg logit contributions
 counts-0.033
One-0.030
 Or-0.029
ends-0.029
-0.028
Token Dad
Token ID17415
Feature activation+1.36e-02

Pos logit contributions
 sensation+0.002
 tune+0.002
 lull+0.002
 streamed+0.002
 streaming+0.002
Neg logit contributions
ends-0.001
One-0.001
 Or-0.001
 counts-0.001
 Ready-0.001
Token went
Token ID1816
Feature activation+4.71e-04

Pos logit contributions
 tune+0.026
 sensation+0.025
 dolphin+0.022
 shark+0.021
 chill+0.021
Neg logit contributions
 counts-0.022
ends-0.021
One-0.021
 Or-0.020
 Ready-0.018
Token to
Token ID284
Feature activation+2.19e-02

Pos logit contributions
 tune+0.001
 sensation+0.001
 dolphin+0.001
 shark+0.001
 potato+0.001
Neg logit contributions
 counts-0.001
ends-0.001
One-0.001
 Or-0.001
 Ready-0.001
Token the
Token ID262
Feature activation-1.84e-02

Pos logit contributions
 sensation+0.045
 tune+0.044
 streamed+0.038
 streaming+0.038
 chill+0.037
Neg logit contributions
 counts-0.032
One-0.032
ends-0.030
 Ready-0.029
 Or-0.029
Token park
Token ID3952
Feature activation-6.78e-02

Pos logit contributions
One+0.024
 counts+0.023
ends+0.023
 Or+0.021
 Ready+0.021
Neg logit contributions
 sensation-0.040
 tune-0.040
 streaming-0.035
 streamed-0.035
 dolphin-0.034
Token buttons
Token ID12163
Feature activation-2.46e-02

Pos logit contributions
ends+0.053
 counts+0.052
One+0.050
 Or+0.049
 Ready+0.046
Neg logit contributions
 tune-0.047
 sensation-0.045
 dolphin-0.041
 shark-0.038
 chill-0.036
Token and
Token ID290
Feature activation-1.72e-02

Pos logit contributions
 counts+0.038
ends+0.038
One+0.036
 Or+0.034
 Ready+0.031
Neg logit contributions
 tune-0.048
 sensation-0.047
 dolphin-0.042
 shark-0.041
 potato-0.039
Token levers
Token ID48872
Feature activation-2.94e-02

Pos logit contributions
 counts+0.027
ends+0.026
 Or+0.024
One+0.023
 Ready+0.021
Neg logit contributions
 sensation-0.034
 tune-0.034
 dolphin-0.030
 shark-0.029
 Dolphin-0.029
Token did
Token ID750
Feature activation-1.34e-02

Pos logit contributions
 counts+0.038
ends+0.038
One+0.035
 Or+0.034
 Ready+0.030
Neg logit contributions
 tune-0.064
 sensation-0.063
 dolphin-0.058
 shark-0.056
 potato-0.054
Token.
Token ID13
Feature activation-6.59e-03

Pos logit contributions
 counts+0.019
ends+0.019
One+0.018
 Or+0.018
 Ready+0.017
Neg logit contributions
 sensation-0.026
 tune-0.026
 face-0.022
 dolphin-0.022
 shark-0.022
Token He
Token ID679
Feature activation-7.75e-03

Pos logit contributions
ends+0.009
 counts+0.009
One+0.009
 Or+0.008
 Ready+0.007
Neg logit contributions
 tune-0.014
 sensation-0.014
 dolphin-0.013
 shark-0.012
 chill-0.012
Token said
Token ID531
Feature activation-2.08e-02

Pos logit contributions
ends+0.013
 counts+0.013
One+0.013
 Or+0.012
 Ready+0.011
Neg logit contributions
 tune-0.014
 sensation-0.013
 dolphin-0.012
 shark-0.012
 potato-0.011
Token she
Token ID673
Feature activation-1.63e-02

Pos logit contributions
ends+0.025
 counts+0.024
 Or+0.022
One+0.022
 Ready+0.018
Neg logit contributions
 tune-0.048
 sensation-0.048
 dolphin-0.042
 chill-0.041
 shark-0.041
Token could
Token ID714
Feature activation+1.28e-02

Pos logit contributions
 counts+0.028
ends+0.028
One+0.027
 Or+0.026
 Ready+0.026
Neg logit contributions
 tune-0.027
 sensation-0.026
 dolphin-0.024
 shark-0.022
 face-0.022
Token help
Token ID1037
Feature activation+4.10e-02

Pos logit contributions
 tune+0.023
 sensation+0.022
 dolphin+0.021
 shark+0.020
 face+0.019
Neg logit contributions
ends-0.021
 counts-0.021
One-0.020
 Or-0.019
 Ready-0.018
Token him
Token ID683
Feature activation+1.35e-02

Pos logit contributions
 sensation+0.085
 tune+0.085
 dolphin+0.078
 shark+0.077
 streamed+0.075
Neg logit contributions
 counts-0.054
ends-0.052
One-0.051
 Or-0.051
 Ready-0.043
Token We
Token ID775
Feature activation+7.29e-04

Pos logit contributions
ends+0.028
 counts+0.027
One+0.025
 Or+0.022
 ours+0.021
Neg logit contributions
 tune-0.047
 sensation-0.045
 dolphin-0.044
 shark-0.043
 Dolphin-0.043
Token will
Token ID481
Feature activation-1.20e-03

Pos logit contributions
 tune+0.002
 sensation+0.002
 dolphin+0.001
 shark+0.001
 chill+0.001
Neg logit contributions
ends-0.001
 counts-0.001
 Or-0.001
One-0.001
 Ready-0.001
Token come
Token ID1282
Feature activation-1.68e-02

Pos logit contributions
ends+0.002
 counts+0.002
 Or+0.002
One+0.002
 Ready+0.002
Neg logit contributions
 tune-0.002
 sensation-0.002
 dolphin-0.002
 face-0.002
 shark-0.002
Token back
Token ID736
Feature activation-3.63e-02

Pos logit contributions
 counts+0.033
One+0.030
 Or+0.029
 belongs+0.029
ends+0.029
Neg logit contributions
 tune-0.024
 dolphin-0.021
 face-0.020
 shark-0.019
 Dolphin-0.019
Token soon
Token ID2582
Feature activation-9.55e-02

Pos logit contributions
 counts+0.049
ends+0.048
One+0.045
 Or+0.045
atters+0.042
Neg logit contributions
 tune-0.071
 dolphin-0.064
 sensation-0.062
 shark-0.062
 Dolphin-0.059
Token.
Token ID13
Feature activation-2.85e-02

Pos logit contributions
 counts+0.100
ends+0.100
One+0.091
 Or+0.090
 Ready+0.085
Neg logit contributions
 tune-0.227
 sensation-0.213
 dolphin-0.204
 face-0.196
 shark-0.195
Token Be
Token ID1355
Feature activation+2.75e-02

Pos logit contributions
ends+0.036
 counts+0.035
One+0.033
 belongs+0.028
 ours+0.028
Neg logit contributions
 tune-0.060
 sensation-0.056
 dolphin-0.055
 shark-0.054
 Dolphin-0.054
Token a
Token ID257
Feature activation-3.74e-02

Pos logit contributions
 tune+0.052
 sensation+0.052
 dolphin+0.047
 shark+0.045
 streaming+0.043
Neg logit contributions
ends-0.043
 counts-0.043
One-0.041
 Or-0.039
 Ready-0.036
Token good
Token ID922
Feature activation+5.51e-02

Pos logit contributions
ends+0.065
 counts+0.064
One+0.063
 Or+0.061
 Ready+0.057
Neg logit contributions
 sensation-0.064
 tune-0.063
 dolphin-0.055
 streaming-0.053
 shark-0.053
Token boy
Token ID2933
Feature activation+3.28e-03

Pos logit contributions
 tune+0.084
 sensation+0.083
 dolphin+0.073
 shark+0.070
 streaming+0.066
Neg logit contributions
ends-0.107
 counts-0.107
One-0.103
 Or-0.100
 Ready-0.093
Token."
Token ID526
Feature activation-2.07e-02

Pos logit contributions
 tune+0.007
 dolphin+0.007
 sensation+0.007
 shark+0.007
 potato+0.006
Neg logit contributions
One-0.004
ends-0.004
 counts-0.003
 Or-0.003
 Ready-0.003
Token.
Token ID13
Feature activation-1.80e-02

Pos logit contributions
ends+0.031
One+0.031
 counts+0.030
 Or+0.029
 Ready+0.028
Neg logit contributions
 sensation-0.040
 tune-0.039
 dolphin-0.037
 shark-0.035
 streaming-0.034
Token Sara
Token ID24799
Feature activation-2.21e-02

Pos logit contributions
 counts+0.029
ends+0.028
 Or+0.025
One+0.025
house+0.024
Neg logit contributions
 sensation-0.030
 tune-0.029
 dolphin-0.028
 Sister-0.027
 Ned-0.027
Token and
Token ID290
Feature activation+1.47e-02

Pos logit contributions
One+0.036
ends+0.033
 ours+0.032
 Ready+0.031
 Or+0.030
Neg logit contributions
 tune-0.037
 dolphin-0.035
 sensation-0.034
 face-0.032
 potato-0.032
Token Tom
Token ID4186
Feature activation-2.46e-02

Pos logit contributions
 tune+0.037
 sensation+0.036
 dolphin+0.034
 shark+0.033
 streaming+0.032
Neg logit contributions
ends-0.014
 counts-0.014
One-0.013
 Or-0.012
 Ready-0.011
Token pretend
Token ID16614
Feature activation-4.33e-02

Pos logit contributions
ends+0.029
 counts+0.028
One+0.027
 Or+0.026
 Ready+0.023
Neg logit contributions
 tune-0.056
 sensation-0.054
 dolphin-0.051
 shark-0.049
 face-0.048
Token the
Token ID262
Feature activation-2.05e-02

Pos logit contributions
 counts+0.061
ends+0.059
One+0.055
 Or+0.054
 Ready+0.048
Neg logit contributions
 sensation-0.089
 tune-0.089
 dolphin-0.079
 streaming-0.076
 shark-0.076
Token rug
Token ID14477
Feature activation-2.93e-02

Pos logit contributions
ends+0.038
 counts+0.037
 Or+0.035
One+0.035
 ours+0.032
Neg logit contributions
 sensation-0.034
 tune-0.033
 dolphin-0.029
 shark-0.028
 chill-0.027
Token is
Token ID318
Feature activation-5.69e-02

Pos logit contributions
ends+0.034
 counts+0.032
One+0.031
 Or+0.029
 Ready+0.027
Neg logit contributions
 tune-0.067
 sensation-0.067
 dolphin-0.062
 shark-0.060
 streaming-0.058
Token a
Token ID257
Feature activation-6.19e-02

Pos logit contributions
ends+0.081
 counts+0.080
One+0.076
 Or+0.072
 Ready+0.066
Neg logit contributions
 tune-0.117
 sensation-0.115
 dolphin-0.104
 shark-0.101
 chill-0.098
Token map
Token ID3975
Feature activation-7.75e-02

Pos logit contributions
ends+0.108
 counts+0.108
One+0.104
 Or+0.100
 Ready+0.093
Neg logit contributions
 tune-0.106
 sensation-0.104
 dolphin-0.093
 shark-0.089
 streaming-0.085
Token of
Token ID286
Feature activation+1.16e-02

Pos logit contributions
ends+0.128
One+0.123
 counts+0.120
 Or+0.116
 Ready+0.110
Neg logit contributions
 tune-0.140
 dolphin-0.131
 sensation-0.131
 shark-0.125
 face-0.117
Token bear
Token ID6842
Feature activation-2.27e-02

Pos logit contributions
ends+0.022
 counts+0.022
One+0.021
 Or+0.020
 Ready+0.019
Neg logit contributions
 tune-0.015
 sensation-0.014
 dolphin-0.013
 shark-0.012
 potato-0.011
Token.
Token ID13
Feature activation-2.61e-02

Pos logit contributions
ends+0.032
 counts+0.031
One+0.030
 Or+0.028
 Ready+0.027
Neg logit contributions
 tune-0.046
 sensation-0.044
 dolphin-0.042
 shark-0.041
 potato-0.040
Token You
Token ID921
Feature activation+4.20e-03

Pos logit contributions
ends+0.033
One+0.032
 counts+0.030
 ours+0.025
 Or+0.024
Neg logit contributions
 sensation-0.055
 tune-0.055
 dolphin-0.052
 shark-0.050
 Dolphin-0.049
Token are
Token ID389
Feature activation-3.23e-02

Pos logit contributions
 tune+0.008
 sensation+0.008
 dolphin+0.008
 shark+0.008
 face+0.008
Neg logit contributions
ends-0.006
One-0.005
 counts-0.005
 Or-0.005
 Ready-0.005
Token very
Token ID845
Feature activation-2.63e-02

Pos logit contributions
 counts+0.053
One+0.052
ends+0.051
 Or+0.051
 Maybe+0.046
Neg logit contributions
 sensation-0.055
 tune-0.055
 dolphin-0.051
 shark-0.049
 face-0.048
Token pretty
Token ID2495
Feature activation-8.40e-02

Pos logit contributions
 counts+0.048
ends+0.046
One+0.046
 Or+0.045
 Ready+0.043
Neg logit contributions
 tune-0.041
 sensation-0.041
 dolphin-0.038
 shark-0.037
 face-0.035
Token.
Token ID13
Feature activation-4.30e-02

Pos logit contributions
ends+0.127
 counts+0.122
One+0.122
 Or+0.115
 Ready+0.114
Neg logit contributions
 tune-0.163
 sensation-0.156
 dolphin-0.151
 shark-0.150
 potato-0.141
Token But
Token ID887
Feature activation-3.02e-02

Pos logit contributions
ends+0.059
One+0.055
 counts+0.051
 ours+0.046
house+0.041
Neg logit contributions
 tune-0.085
 sensation-0.084
 dolphin-0.080
 Dolphin-0.077
 shark-0.077
Token I
Token ID314
Feature activation-2.03e-02

Pos logit contributions
ends+0.032
 counts+0.031
One+0.029
 Or+0.027
atters+0.023
Neg logit contributions
 tune-0.072
 sensation-0.071
 dolphin-0.067
 shark-0.065
 face-0.063
Token will
Token ID481
Feature activation-3.23e-03

Pos logit contributions
 counts+0.025
ends+0.025
One+0.024
 Or+0.022
 Ready+0.022
Neg logit contributions
 sensation-0.043
 tune-0.043
 dolphin-0.040
 shark-0.038
 face-0.037
Token make
Token ID787
Feature activation+3.08e-02

Pos logit contributions
ends+0.005
 counts+0.005
One+0.005
 Or+0.004
 Ready+0.004
Neg logit contributions
 tune-0.006
 sensation-0.006
 dolphin-0.006
 face-0.005
 shark-0.005
Token,
Token ID11
Feature activation-2.80e-02

Pos logit contributions
 tune+0.032
 sensation+0.032
 dolphin+0.029
 shark+0.028
 streaming+0.028
Neg logit contributions
 counts-0.019
ends-0.018
One-0.018
 Or-0.016
 Ready-0.015
Token I
Token ID314
Feature activation-2.95e-02

Pos logit contributions
 counts+0.030
ends+0.030
One+0.028
 Or+0.027
 Ready+0.024
Neg logit contributions
 tune-0.066
 sensation-0.066
 dolphin-0.058
 streaming-0.058
 streamed-0.057
Token want
Token ID765
Feature activation-1.96e-02

Pos logit contributions
 counts+0.037
ends+0.035
One+0.035
 Or+0.033
 Ready+0.030
Neg logit contributions
 tune-0.062
 sensation-0.062
 dolphin-0.058
 shark-0.056
 chill-0.054
Token to
Token ID284
Feature activation-6.67e-03

Pos logit contributions
ends+0.034
 counts+0.031
 Or+0.030
One+0.030
 Ready+0.027
Neg logit contributions
 tune-0.035
 sensation-0.033
 dolphin-0.033
 shark-0.032
 potato-0.031
Token go
Token ID467
Feature activation-3.10e-02

Pos logit contributions
ends+0.011
 counts+0.011
One+0.010
 Or+0.010
 Ready+0.009
Neg logit contributions
 tune-0.012
 sensation-0.012
 dolphin-0.011
 shark-0.011
 face-0.010
Token first
Token ID717
Feature activation-1.12e-01

Pos logit contributions
 counts+0.048
ends+0.047
 Or+0.047
One+0.043
 Ready+0.043
Neg logit contributions
 tune-0.057
 shark-0.052
 dolphin-0.051
 sensation-0.050
 face-0.047
Token!"
Token ID2474
Feature activation-2.79e-02

Pos logit contributions
ends+0.143
 Or+0.127
 counts+0.127
 Ready+0.121
One+0.113
Neg logit contributions
 tune-0.244
 sensation-0.233
 shark-0.218
 dolphin-0.218
 face-0.206
Token Ben
Token ID3932
Feature activation+4.10e-02

Pos logit contributions
ends+0.049
 counts+0.048
One+0.046
 Or+0.044
 Ready+0.042
Neg logit contributions
 tune-0.048
 sensation-0.047
 dolphin-0.044
 shark-0.042
 face-0.040
Token says
Token ID1139
Feature activation-2.11e-02

Pos logit contributions
 tune+0.087
 sensation+0.085
 dolphin+0.083
 shark+0.081
 face+0.079
Neg logit contributions
ends-0.051
One-0.050
 counts-0.045
 Or-0.042
 ours-0.041
Token.
Token ID13
Feature activation-8.21e-03

Pos logit contributions
 counts+0.028
ends+0.027
One+0.027
 Or+0.024
 Ready+0.023
Neg logit contributions
 tune-0.046
 sensation-0.045
 dolphin-0.041
 shark-0.040
 face-0.039
Token He
Token ID679
Feature activation-1.19e-02

Pos logit contributions
ends+0.013
 counts+0.013
One+0.012
 Or+0.012
 Ready+0.011
Neg logit contributions
 tune-0.016
 sensation-0.015
 dolphin-0.014
 shark-0.013
 streaming-0.013
Token the
Token ID262
Feature activation-4.62e-02

Pos logit contributions
 counts+0.061
ends+0.061
One+0.058
 Or+0.056
 Ready+0.052
Neg logit contributions
 tune-0.057
 sensation-0.055
 dolphin-0.050
 shark-0.048
 streaming-0.045
Token time
Token ID640
Feature activation+2.86e-03

Pos logit contributions
ends+0.081
 counts+0.078
 Or+0.076
One+0.075
 Ready+0.069
Neg logit contributions
 tune-0.081
 sensation-0.079
 dolphin-0.071
 shark-0.067
 chill-0.065
Token.
Token ID13
Feature activation+1.44e-03

Pos logit contributions
 tune+0.006
 sensation+0.006
 dolphin+0.005
 shark+0.005
 streaming+0.005
Neg logit contributions
 counts-0.004
ends-0.004
One-0.004
 Or-0.004
 Ready-0.003
Token
Token ID198
Feature activation+1.83e-03

Pos logit contributions
 sensation+0.003
 tune+0.003
 dolphin+0.002
 face+0.002
 Dolphin+0.002
Neg logit contributions
ends-0.002
 counts-0.002
 ours-0.002
One-0.002
 Or-0.002
Token
Token ID198
Feature activation+1.31e-02

Pos logit contributions
 tune+0.004
 sensation+0.003
 dolphin+0.003
 shark+0.003
 face+0.003
Neg logit contributions
 counts-0.003
ends-0.003
 Or-0.002
 ours-0.002
 Ready-0.002
TokenOne
Token ID3198
Feature activation-1.07e-01

Pos logit contributions
 tune+0.035
 sensation+0.033
 lull+0.031
 dolphin+0.031
 rhythm+0.031
Neg logit contributions
 counts-0.010
 ours-0.009
ends-0.008
 Or-0.007
 belongs-0.007
Token day
Token ID1110
Feature activation+3.85e-02

Pos logit contributions
One+0.155
 counts+0.148
ends+0.138
 Or+0.125
+0.125
Neg logit contributions
 tune-0.213
 sensation-0.209
 streaming-0.199
 streamed-0.188
 tant-0.188
Token,
Token ID11
Feature activation+3.08e-02

Pos logit contributions
 tune+0.091
 sensation+0.089
 dolphin+0.081
 shark+0.079
 chill+0.076
Neg logit contributions
ends-0.047
 counts-0.045
One-0.042
 Or-0.040
 Ready-0.036
Token L
Token ID406
Feature activation+1.64e-02

Pos logit contributions
 tune+0.072
 sensation+0.070
 shark+0.064
 dolphin+0.064
 chill+0.062
Neg logit contributions
ends-0.039
 counts-0.034
One-0.034
 Or-0.032
 Ready-0.029
Tokenola
Token ID5708
Feature activation+1.78e-02

Pos logit contributions
 tune+0.024
 sensation+0.024
 dolphin+0.021
 shark+0.020
 streaming+0.019
Neg logit contributions
 counts-0.032
ends-0.032
One-0.031
 Or-0.030
 Ready-0.028
Token was
Token ID373
Feature activation+4.87e-03

Pos logit contributions
 tune+0.032
 sensation+0.031
 dolphin+0.030
 shark+0.028
 potato+0.027
Neg logit contributions
ends-0.029
 counts-0.028
One-0.027
 Or-0.026
 Ready-0.025
TokenOne
Token ID3198
Feature activation-1.10e-01

Pos logit contributions
 tune+0.034
 sensation+0.033
 dolphin+0.032
 shark+0.030
 potato+0.030
Neg logit contributions
 counts-0.007
ends-0.007
 Or-0.005
 ours-0.004
 Ready-0.004
Token summer
Token ID3931
Feature activation+3.25e-02

Pos logit contributions
One+0.168
 counts+0.158
ends+0.157
 Or+0.143
house+0.140
Neg logit contributions
 tune-0.214
 sensation-0.212
 streaming-0.199
 streamed-0.189
 dolphin-0.185
Token,
Token ID11
Feature activation+3.04e-02

Pos logit contributions
 tune+0.056
 sensation+0.055
 dolphin+0.049
 shark+0.047
 streaming+0.046
Neg logit contributions
 counts-0.059
ends-0.055
One-0.055
 Or-0.051
 Ready-0.049
Token things
Token ID1243
Feature activation+6.00e-03

Pos logit contributions
 tune+0.053
 sensation+0.052
 streaming+0.044
 lull+0.043
 chill+0.043
Neg logit contributions
One-0.052
ends-0.051
 counts-0.051
 Or-0.047
 ours-0.046
Token were
Token ID547
Feature activation-7.41e-03

Pos logit contributions
 tune+0.013
 sensation+0.012
 dolphin+0.011
 Dolphin+0.010
 shark+0.010
Neg logit contributions
 counts-0.009
ends-0.008
One-0.008
 Or-0.007
 belongs-0.007
Token different
Token ID1180
Feature activation-7.99e-02

Pos logit contributions
ends+0.011
 counts+0.011
One+0.011
 Or+0.010
 Ready+0.010
Neg logit contributions
 tune-0.014
 sensation-0.014
 dolphin-0.013
 shark-0.012
 streaming-0.012
Token.
Token ID13
Feature activation+6.30e-03

Pos logit contributions
 counts+0.105
ends+0.101
One+0.096
 Or+0.091
 Ready+0.081
Neg logit contributions
 tune-0.175
 sensation-0.170
 dolphin-0.159
 shark-0.154
 Dolphin-0.148
Token Jack
Token ID3619
Feature activation+1.94e-02

Pos logit contributions
 tune+0.011
 sensation+0.010
 dolphin+0.009
 lull+0.009
 shark+0.009
Neg logit contributions
 counts-0.011
One-0.011
 Or-0.011
ends-0.011
 Ready-0.010
Token's
Token ID338
Feature activation+1.69e-02

Pos logit contributions
 tune+0.040
 sensation+0.040
 dolphin+0.036
 shark+0.035
 streaming+0.034
Neg logit contributions
 counts-0.027
ends-0.027
One-0.026
 Or-0.024
 Ready-0.022
Token dad
Token ID9955
Feature activation+1.80e-02

Pos logit contributions
 sensation+0.031
 tune+0.030
 streaming+0.027
 streamed+0.027
 dolphin+0.026
Neg logit contributions
ends-0.026
 counts-0.025
 Or-0.024
One-0.024
house-0.022
Token was
Token ID373
Feature activation-1.30e-02

Pos logit contributions
 tune+0.033
 sensation+0.032
 dolphin+0.029
 shark+0.028
 potato+0.027
Neg logit contributions
ends-0.029
 counts-0.029
One-0.028
 Or-0.027
 Ready-0.025
Token,
Token ID11
Feature activation+2.46e-02

Pos logit contributions
 tune+0.077
 sensation+0.076
 dolphin+0.070
 shark+0.067
 streaming+0.065
Neg logit contributions
ends-0.055
 counts-0.055
One-0.052
 Or-0.050
 Ready-0.046
Token she
Token ID673
Feature activation+1.33e-02

Pos logit contributions
 tune+0.057
 sensation+0.057
 dolphin+0.053
 shark+0.051
 potato+0.050
Neg logit contributions
 counts-0.029
ends-0.028
One-0.025
 Or-0.025
 Ready-0.022
Token went
Token ID1816
Feature activation+7.58e-03

Pos logit contributions
 tune+0.022
 sensation+0.021
 dolphin+0.020
 shark+0.019
 potato+0.018
Neg logit contributions
ends-0.024
 counts-0.024
One-0.023
 Or-0.022
 Ready-0.021
Token to
Token ID284
Feature activation+2.48e-02

Pos logit contributions
 tune+0.014
 sensation+0.014
 dolphin+0.013
 shark+0.012
 potato+0.012
Neg logit contributions
 counts-0.012
ends-0.012
One-0.011
 Or-0.011
 Ready-0.010
Token the
Token ID262
Feature activation-5.74e-03

Pos logit contributions
 sensation+0.045
 tune+0.045
 dolphin+0.038
 shark+0.038
 streaming+0.037
Neg logit contributions
 counts-0.042
One-0.041
ends-0.041
 Or-0.038
 Ready-0.036
Token zoo
Token ID26626
Feature activation-7.56e-02

Pos logit contributions
One+0.007
 counts+0.007
ends+0.007
 Or+0.006
 Ready+0.006
Neg logit contributions
 sensation-0.013
 tune-0.013
 streaming-0.012
 streamed-0.012
 dolphin-0.011
Token with
Token ID351
Feature activation+1.40e-03

Pos logit contributions
ends+0.117
 counts+0.116
One+0.110
 Or+0.108
 Ready+0.104
Neg logit contributions
 tune-0.144
 sensation-0.137
 dolphin-0.127
 shark-0.122
 face-0.120
Token her
Token ID607
Feature activation-7.13e-03

Pos logit contributions
 tune+0.002
 sensation+0.002
 dolphin+0.002
 shark+0.002
 face+0.002
Neg logit contributions
ends-0.002
 counts-0.002
One-0.002
 Or-0.002
 Ready-0.002
Token mom
Token ID1995
Feature activation-1.40e-02

Pos logit contributions
ends+0.011
 counts+0.011
One+0.011
 Or+0.010
house+0.009
Neg logit contributions
 sensation-0.014
 tune-0.014
 streamed-0.011
 streaming-0.011
 dolphin-0.011
Token and
Token ID290
Feature activation+7.59e-03

Pos logit contributions
ends+0.014
 counts+0.014
 Or+0.012
One+0.012
 Ready+0.011
Neg logit contributions
 tune-0.034
 sensation-0.033
 dolphin-0.032
 potato-0.031
 shark-0.030
Token dad
Token ID9955
Feature activation+1.44e-02

Pos logit contributions
 sensation+0.012
 tune+0.012
 dolphin+0.010
 shark+0.010
 streaming+0.009
Neg logit contributions
ends-0.015
 counts-0.014
One-0.014
 Or-0.014
 Ready-0.013