TOP ACTIVATIONS
MAX = 0.343

found a lucky four - leaf cl over while picking flowers
excited . The four - leaf cl over was really lucky
andy found a four - leaf cl over ! She showed
and the lucky four - leaf cl over . They all
for being the best bal ancer in the competition .
steam . Hop sc ot ch thanked the
His name was Hop - Hop . Every day , he
very happy . Hop and the squirrel played the
silly bunny ?" Hop sc ot ch replied ,
row . Hop sc ot ch hopped to
Hop on my back , â
that too . Hop asked the butterfly , "
with you ." Hop listened carefully to the wise
of fun . Hop wanted to play a game
the world ". Hop was so excited . He
. He said , " Hop , hop , hop ,
love . Hop felt a little better knowing
to reach . Hop asked his friend , a
," she said . " Hop in ." R
around looking at the small fishes . And they waved to

MIN ACTIVATIONS
MIN = -0.307

climb up to the tree house in her backyard , but
great time in the play house . Tom and Sally were
would find a new doll house and that they would never
still play with the doll house with Lily . She will
and made a beautiful bird house with lots of seeds and
shot out of the tree house window and flew into the
dolls back in the doll house . They hugged each other
they could repair their doll house and make it look original
fell on Tim 's play house and broke it . Tim
He ran to the play house and jumped on the table
together in their old tree house . <|endoftext|> Once upon a
a window in the play house . She looks for the
of water in her bird house . " I can help
She had a big doll house with many rooms and furniture
and went to the tree house . Rose said sorry to
, he made a tree house out of tw igs just
who lived in a tree house . One day he had
's sleep in the tree house . It 's cozy and
He cried because his play house was broken . His mom
then played with the doll house . They had a lot

INTERVAL 0.181 - 0.343
CONTAINS 1.312%

The end . <|endoftext|> Once upon a time , there was
. " Thank you , Sara . You are
my will be mad if you make a mess ," Ben
. I was wondering if you could donate anything ?" The
easier and faster . Thank you very much . Here ,

INTERVAL 0.018 - 0.181
CONTAINS 32.049%

one I like ." Tim thought for a moment and then
He said , " I found a new house
wanted to print a picture of her big stone . She
his head and was very careful . He stepped around the
After eating , Lily felt very sleepy . She went

INTERVAL -0.144 - 0.018
CONTAINS 65.713%

cast . <|endoftext|> Tom and Sam are friends . They like
my loved it very much . From that day on ,
The little boy decided to explore some other places .
showed the wine to his mom , she was very happy
the new man . Her mom said , "

INTERVAL -0.307 - -0.144
CONTAINS 0.926%

loud noise . S ara came back and saw
bag and the pear was now ripe and delicious . Lily
kind and compassionate lady . One day she got a strange
forever and ever . <|endoftext|> One day there was a polite
't have any friends . One day she decided to find
Token found
Token ID1043
Feature activation-6.67e-02

Pos logit contributions
house+0.057
 handcuffs+0.056
ringe+0.056
packs+0.055
tr+0.054
Neg logit contributions
ter-0.062
fully-0.050
ya-0.050
ting-0.050
 Dolphin-0.050
Token a
Token ID257
Feature activation-1.88e-02

Pos logit contributions
house+0.138
packs+0.126
tr+0.121
 fort+0.120
oc+0.117
Neg logit contributions
ter-0.101
fully-0.084
ous-0.082
ting-0.080
 Dolphin-0.079
Token lucky
Token ID9670
Feature activation+5.25e-02

Pos logit contributions
house+0.034
packs+0.030
tr+0.029
 fort+0.029
oc+0.028
Neg logit contributions
ter-0.033
fully-0.029
ous-0.028
ting-0.027
 Dolphin-0.027
Token four
Token ID1440
Feature activation-1.02e-02

Pos logit contributions
ter+0.090
fully+0.080
ous+0.078
ting+0.077
 Dolphin+0.075
Neg logit contributions
house-0.097
packs-0.084
tr-0.083
 fort-0.083
oc-0.077
Token-
Token ID12
Feature activation+1.91e-02

Pos logit contributions
house+0.020
ringe+0.019
tr+0.019
packs+0.019
 handcuffs+0.019
Neg logit contributions
ter-0.015
fully-0.013
ya-0.012
izing-0.012
ous-0.012
Tokenleaf
Token ID33201
Feature activation+3.43e-01

Pos logit contributions
ter+0.030
fully+0.025
ous+0.025
 Dolphin+0.025
ting+0.024
Neg logit contributions
house-0.038
packs-0.034
tr-0.033
oc-0.032
 fort-0.032
Token cl
Token ID537
Feature activation+1.61e-01

Pos logit contributions
ter+0.530
fully+0.450
ous+0.437
 Dolphin+0.432
ancer+0.419
Neg logit contributions
house-0.690
packs-0.613
tr-0.600
oc-0.587
 fort-0.583
Tokenover
Token ID2502
Feature activation+4.11e-02

Pos logit contributions
ter+0.215
ous+0.166
fully+0.165
ting+0.162
ing+0.162
Neg logit contributions
house-0.356
 fort-0.342
 handcuffs-0.336
tr-0.336
packs-0.330
Token while
Token ID981
Feature activation-1.72e-02

Pos logit contributions
ter+0.065
fully+0.055
ting+0.053
ous+0.051
 Dolphin+0.049
Neg logit contributions
house-0.081
packs-0.076
tr-0.073
ringe-0.072
 fort-0.072
Token picking
Token ID10868
Feature activation+4.72e-02

Pos logit contributions
house+0.034
packs+0.030
tr+0.028
 fort+0.028
 handcuffs+0.027
Neg logit contributions
ter-0.028
fully-0.024
ous-0.023
 Dolphin-0.023
ting-0.022
Token flowers
Token ID12734
Feature activation+7.35e-02

Pos logit contributions
ter+0.075
ous+0.070
fully+0.068
ancer+0.065
izing+0.064
Neg logit contributions
house-0.087
packs-0.079
tr-0.074
 bars-0.073
 fort-0.073
Token excited
Token ID6568
Feature activation+3.14e-02

Pos logit contributions
ter+0.004
fully+0.003
ting+0.003
ous+0.003
ancer+0.003
Neg logit contributions
house-0.004
packs-0.004
tr-0.004
 fort-0.004
oc-0.004
Token.
Token ID13
Feature activation+8.23e-03

Pos logit contributions
ter+0.065
 Dolphin+0.060
ancer+0.060
redible+0.057
ternally+0.057
Neg logit contributions
house-0.044
packs-0.037
 fort-0.036
tr-0.032
 bars-0.031
Token The
Token ID383
Feature activation-3.41e-03

Pos logit contributions
ter+0.016
fully+0.014
ous+0.013
izing+0.013
ting+0.013
Neg logit contributions
house-0.013
packs-0.012
tr-0.011
oc-0.011
 fort-0.011
Token four
Token ID1440
Feature activation-9.29e-03

Pos logit contributions
house+0.005
 bars+0.005
 handcuffs+0.005
tr+0.005
 fort+0.004
Neg logit contributions
ter-0.006
fully-0.006
ous-0.005
ancer-0.005
 Joey-0.005
Token-
Token ID12
Feature activation+1.91e-02

Pos logit contributions
house+0.018
packs+0.016
tr+0.015
 fort+0.015
oc+0.015
Neg logit contributions
ter-0.015
fully-0.013
ous-0.013
 Dolphin-0.012
ting-0.012
Tokenleaf
Token ID33201
Feature activation+3.38e-01

Pos logit contributions
ter+0.031
izing+0.026
 Dolphin+0.025
ancer+0.025
ting+0.025
Neg logit contributions
house-0.036
tr-0.032
oc-0.032
ris-0.032
packs-0.031
Token cl
Token ID537
Feature activation+1.62e-01

Pos logit contributions
ter+0.528
ancer+0.460
fully+0.438
ous+0.437
 Dolphin+0.436
Neg logit contributions
house-0.627
oc-0.575
packs-0.539
ris-0.533
 bars-0.525
Tokenover
Token ID2502
Feature activation+3.89e-02

Pos logit contributions
ter+0.225
fully+0.186
ous+0.181
ting+0.175
 Dolphin+0.174
Neg logit contributions
house-0.356
packs-0.326
tr-0.315
 fort-0.311
oc-0.305
Token was
Token ID373
Feature activation-2.90e-02

Pos logit contributions
ter+0.061
fully+0.052
ous+0.051
ting+0.050
 Dolphin+0.049
Neg logit contributions
house-0.077
packs-0.070
tr-0.068
 fort-0.066
ringe-0.065
Token really
Token ID1107
Feature activation-1.44e-03

Pos logit contributions
house+0.060
packs+0.055
tr+0.053
 fort+0.052
oc+0.050
Neg logit contributions
ter-0.043
fully-0.036
ous-0.035
ting-0.034
 Dolphin-0.034
Token lucky
Token ID9670
Feature activation+5.41e-02

Pos logit contributions
house+0.003
packs+0.002
tr+0.002
 fort+0.002
oc+0.002
Neg logit contributions
ter-0.003
fully-0.002
ous-0.002
ting-0.002
ancer-0.002
Tokenandy
Token ID10757
Feature activation+1.37e-01

Pos logit contributions
ter+0.084
ting+0.080
ivated+0.079
ner+0.079
 Hop+0.078
Neg logit contributions
house-0.073
omo-0.069
 fort-0.067
ris-0.064
packs-0.063
Token found
Token ID1043
Feature activation-6.30e-02

Pos logit contributions
ter+0.288
fully+0.257
ous+0.252
ancer+0.251
izing+0.251
Neg logit contributions
house-0.207
packs-0.169
tr-0.159
 fort-0.159
oc-0.157
Token a
Token ID257
Feature activation-1.55e-02

Pos logit contributions
house+0.108
packs+0.094
 handcuffs+0.090
 fort+0.090
tr+0.089
Neg logit contributions
ter-0.106
ous-0.101
ancer-0.098
izing-0.097
fully-0.097
Token four
Token ID1440
Feature activation-9.82e-03

Pos logit contributions
house+0.026
packs+0.022
tr+0.022
 fort+0.021
oc+0.021
Neg logit contributions
ter-0.030
fully-0.026
ous-0.025
ting-0.025
 Dolphin-0.025
Token-
Token ID12
Feature activation+1.75e-02

Pos logit contributions
house+0.018
packs+0.016
tr+0.016
 fort+0.015
ringe+0.015
Neg logit contributions
ter-0.017
fully-0.014
ous-0.014
ting-0.014
 Dolphin-0.014
Tokenleaf
Token ID33201
Feature activation+3.37e-01

Pos logit contributions
ter+0.028
ous+0.023
fully+0.023
 Dolphin+0.022
izing+0.022
Neg logit contributions
house-0.034
packs-0.031
tr-0.031
oc-0.030
ris-0.029
Token cl
Token ID537
Feature activation+1.61e-01

Pos logit contributions
ter+0.543
ancer+0.462
fully+0.457
 Dolphin+0.453
ous+0.452
Neg logit contributions
house-0.640
oc-0.556
packs-0.548
tr-0.532
 fort-0.517
Tokenover
Token ID2502
Feature activation+3.70e-02

Pos logit contributions
ter+0.196
fully+0.157
ous+0.152
ting+0.146
 Dolphin+0.146
Neg logit contributions
house-0.382
packs-0.352
tr-0.343
 fort-0.338
oc-0.332
Token!
Token ID0
Feature activation+3.23e-02

Pos logit contributions
ter+0.059
fully+0.050
ting+0.048
ous+0.048
 Dolphin+0.047
Neg logit contributions
house-0.072
packs-0.065
tr-0.065
 fort-0.063
ringe-0.063
Token She
Token ID1375
Feature activation-3.67e-02

Pos logit contributions
ter+0.068
izing+0.059
fully+0.058
ble+0.057
gging+0.056
Neg logit contributions
house-0.048
packs-0.043
tr-0.039
oc-0.037
 fort-0.037
Token showed
Token ID3751
Feature activation-3.81e-02

Pos logit contributions
house+0.060
packs+0.052
tr+0.050
 fort+0.049
oc+0.048
Neg logit contributions
ter-0.071
fully-0.062
ous-0.061
ting-0.060
 Dolphin-0.060
Token and
Token ID290
Feature activation+8.68e-03

Pos logit contributions
ter+0.067
fully+0.057
ous+0.056
ting+0.053
 Dolphin+0.053
Neg logit contributions
house-0.082
packs-0.075
tr-0.073
ringe-0.073
oc-0.073
Token the
Token ID262
Feature activation+4.38e-02

Pos logit contributions
ter+0.014
fully+0.012
ous+0.012
 Dolphin+0.012
ting+0.011
Neg logit contributions
house-0.017
packs-0.015
tr-0.015
 fort-0.014
oc-0.014
Token lucky
Token ID9670
Feature activation+4.69e-02

Pos logit contributions
ter+0.074
fully+0.068
ous+0.067
ting+0.065
izing+0.064
Neg logit contributions
house-0.078
packs-0.070
 fort-0.069
tr-0.068
 handcuffs-0.063
Token four
Token ID1440
Feature activation-1.57e-02

Pos logit contributions
ter+0.084
fully+0.075
ting+0.074
ous+0.073
ancer+0.072
Neg logit contributions
house-0.081
packs-0.070
 fort-0.069
tr-0.068
oc-0.062
Token-
Token ID12
Feature activation+1.03e-02

Pos logit contributions
house+0.030
packs+0.028
tr+0.027
 fort+0.026
ringe+0.026
Neg logit contributions
ter-0.025
fully-0.022
ous-0.021
ting-0.020
izing-0.020
Tokenleaf
Token ID33201
Feature activation+3.31e-01

Pos logit contributions
ter+0.018
 Dolphin+0.015
izing+0.015
ting+0.015
fully+0.014
Neg logit contributions
house-0.019
tr-0.016
packs-0.016
oc-0.016
ris-0.016
Token cl
Token ID537
Feature activation+1.60e-01

Pos logit contributions
ter+0.537
fully+0.454
 Dolphin+0.446
ous+0.444
ancer+0.438
Neg logit contributions
house-0.634
packs-0.560
oc-0.542
tr-0.537
 fort-0.523
Tokenover
Token ID2502
Feature activation+2.80e-02

Pos logit contributions
ter+0.194
fully+0.155
ous+0.151
ting+0.144
 Dolphin+0.143
Neg logit contributions
house-0.379
packs-0.349
tr-0.342
 fort-0.339
oc-0.330
Token.
Token ID13
Feature activation+1.30e-03

Pos logit contributions
ter+0.048
fully+0.042
ous+0.041
ting+0.040
 Dolphin+0.039
Neg logit contributions
house-0.051
packs-0.046
tr-0.045
 fort-0.044
oc-0.043
Token They
Token ID1119
Feature activation-8.82e-02

Pos logit contributions
ter+0.002
fully+0.002
izing+0.002
ting+0.002
ous+0.002
Neg logit contributions
house-0.002
packs-0.002
tr-0.002
 fort-0.002
oc-0.002
Token all
Token ID477
Feature activation-3.77e-02

Pos logit contributions
house+0.161
packs+0.144
tr+0.141
 fort+0.138
ringe+0.136
Neg logit contributions
ter-0.154
fully-0.133
ous-0.130
ting-0.127
 Dolphin-0.126
Token for
Token ID329
Feature activation-6.32e-02

Pos logit contributions
house+0.139
packs+0.125
tr+0.121
 fort+0.118
oc+0.113
Neg logit contributions
ter-0.118
 Dolphin-0.098
fully-0.097
ting-0.097
ous-0.097
Token being
Token ID852
Feature activation+1.70e-01

Pos logit contributions
house+0.108
packs+0.095
 handcuffs+0.094
 fort+0.091
tr+0.087
Neg logit contributions
ter-0.106
ous-0.091
 Dolphin-0.090
ancer-0.087
izing-0.087
Token the
Token ID262
Feature activation+3.99e-02

Pos logit contributions
ter+0.259
fully+0.220
ous+0.216
ting+0.208
 Dolphin+0.208
Neg logit contributions
house-0.345
packs-0.311
tr-0.299
 fort-0.293
ringe-0.288
Token best
Token ID1266
Feature activation-4.45e-02

Pos logit contributions
ter+0.075
fully+0.068
ous+0.067
ting+0.065
 Dolphin+0.064
Neg logit contributions
house-0.065
packs-0.056
tr-0.056
 fort-0.054
ringe-0.052
Token bal
Token ID3652
Feature activation-6.70e-02

Pos logit contributions
house+0.077
packs+0.069
tr+0.066
oc+0.066
ringe+0.065
Neg logit contributions
ter-0.081
fully-0.071
ous-0.070
 Dolphin-0.067
ting-0.067
Tokenancer
Token ID8250
Feature activation+3.21e-01

Pos logit contributions
packs+0.157
house+0.156
tr+0.144
ringe+0.138
ris+0.134
Neg logit contributions
gging-0.058
izing-0.056
 eagle-0.055
ter-0.055
ya-0.055
Token in
Token ID287
Feature activation+1.00e-01

Pos logit contributions
ter+0.532
fully+0.449
ous+0.445
 Dolphin+0.445
izing+0.438
Neg logit contributions
house-0.602
packs-0.544
 fort-0.507
tr-0.504
oc-0.488
Token the
Token ID262
Feature activation+3.77e-02

Pos logit contributions
ter+0.119
fully+0.095
ous+0.089
ting+0.088
 Dolphin+0.088
Neg logit contributions
house-0.238
tr-0.222
packs-0.215
 fort-0.213
ringe-0.211
Token competition
Token ID5449
Feature activation+8.56e-02

Pos logit contributions
ter+0.071
fully+0.066
ous+0.064
izing+0.057
ting+0.057
Neg logit contributions
house-0.066
 fort-0.055
tr-0.055
packs-0.054
oc-0.053
Token.
Token ID13
Feature activation+2.65e-03

Pos logit contributions
ter+0.138
fully+0.120
ous+0.116
ting+0.112
 Dolphin+0.109
Neg logit contributions
house-0.162
packs-0.146
tr-0.145
 fort-0.142
ringe-0.142
Token
Token ID220
Feature activation+4.81e-03

Pos logit contributions
ter+0.005
fully+0.004
ous+0.004
ting+0.004
izing+0.004
Neg logit contributions
house-0.005
packs-0.004
tr-0.004
oc-0.004
 fort-0.004
Token steam
Token ID13324
Feature activation+3.68e-02

Pos logit contributions
ter+0.016
ancer+0.014
fully+0.014
ivated+0.014
ivating+0.014
Neg logit contributions
house-0.013
 fort-0.011
oc-0.011
packs-0.011
 bisc-0.010
Token.
Token ID13
Feature activation+6.64e-03

Pos logit contributions
ter+0.060
ya+0.054
 Dolphin+0.054
gging+0.054
ancer+0.053
Neg logit contributions
house-0.067
tr-0.059
 fort-0.059
packs-0.055
oc-0.052
Token
Token ID220
Feature activation+1.15e-02

Pos logit contributions
ter+0.013
fully+0.012
ous+0.011
izing+0.011
ya+0.011
Neg logit contributions
house-0.011
oc-0.009
 handcuffs-0.009
packs-0.009
tr-0.008
Token
Token ID198
Feature activation+1.23e-02

Pos logit contributions
ter+0.024
ya+0.022
fully+0.021
izing+0.021
ous+0.021
Neg logit contributions
house-0.018
packs-0.015
tr-0.015
oc-0.014
 handcuffs-0.014
Token
Token ID198
Feature activation+1.29e-02

Pos logit contributions
ter+0.026
ous+0.022
fully+0.022
izing+0.022
ting+0.021
Neg logit contributions
house-0.019
oc-0.017
packs-0.015
tr-0.015
 handcuffs-0.015
TokenHop
Token ID23483
Feature activation+3.20e-01

Pos logit contributions
ter+0.022
ous+0.020
fully+0.019
ya+0.018
ancer+0.018
Neg logit contributions
house-0.024
oc-0.021
tr-0.020
packs-0.020
ringe-0.019
Tokensc
Token ID1416
Feature activation-4.75e-03

Pos logit contributions
ter+0.615
ancer+0.585
ous+0.572
izing+0.566
 Him+0.560
Neg logit contributions
house-0.475
oc-0.433
packs-0.359
ris-0.345
 handcuffs-0.333
Tokenot
Token ID313
Feature activation-3.52e-02

Pos logit contributions
house+0.009
packs+0.008
tr+0.008
 fort+0.008
oc+0.008
Neg logit contributions
ter-0.008
fully-0.007
ous-0.007
ting-0.006
 Dolphin-0.006
Tokench
Token ID354
Feature activation+4.98e-02

Pos logit contributions
house+0.056
oc+0.050
tr+0.045
ris+0.044
packs+0.043
Neg logit contributions
ter-0.066
ancer-0.058
izing-0.057
 Dolphin-0.057
 Him-0.056
Token thanked
Token ID26280
Feature activation+6.48e-02

Pos logit contributions
ter+0.079
 Dolphin+0.073
izing+0.071
 Nathan+0.070
ancer+0.070
Neg logit contributions
house-0.092
oc-0.078
packs-0.076
 fort-0.075
tr-0.073
Token the
Token ID262
Feature activation+3.78e-02

Pos logit contributions
ter+0.111
ous+0.102
izing+0.096
ancer+0.095
fully+0.094
Neg logit contributions
house-0.108
oc-0.098
 handcuffs-0.097
 fort-0.094
ringe-0.093
Token His
Token ID2399
Feature activation-8.62e-03

Pos logit contributions
ter+0.016
ous+0.014
fully+0.014
izing+0.014
ting+0.014
Neg logit contributions
house-0.013
tr-0.010
 fort-0.010
packs-0.010
oc-0.010
Token name
Token ID1438
Feature activation+7.60e-05

Pos logit contributions
house+0.014
 fort+0.012
packs+0.012
tr+0.012
ringe+0.011
Neg logit contributions
ter-0.016
fully-0.015
ous-0.015
ting-0.015
izing-0.014
Token was
Token ID373
Feature activation-2.67e-02

Pos logit contributions
ter+0.000
fully+0.000
ous+0.000
 Dolphin+0.000
ting+0.000
Neg logit contributions
house-0.000
packs-0.000
tr-0.000
 fort-0.000
oc-0.000
Token Hop
Token ID9996
Feature activation+2.76e-01

Pos logit contributions
house+0.046
tr+0.038
 fort+0.038
oc+0.037
ringe+0.036
Neg logit contributions
izing-0.044
ter-0.044
fully-0.043
ous-0.042
ancer-0.040
Token-
Token ID12
Feature activation+2.31e-02

Pos logit contributions
ter+0.562
ous+0.518
redible+0.515
ting+0.504
fully+0.501
Neg logit contributions
house-0.420
oc-0.376
packs-0.342
 fort-0.325
tr-0.309
TokenHop
Token ID23483
Feature activation+3.17e-01

Pos logit contributions
ter+0.038
ous+0.033
fully+0.033
ancer+0.031
izing+0.031
Neg logit contributions
house-0.044
oc-0.039
tr-0.037
packs-0.037
ringe-0.036
Token.
Token ID13
Feature activation+8.98e-03

Pos logit contributions
ous+0.494
ter+0.478
redible+0.477
izing+0.452
fully+0.450
Neg logit contributions
house-0.599
oc-0.539
packs-0.486
ringe-0.459
 fort-0.446
Token Every
Token ID3887
Feature activation-6.72e-02

Pos logit contributions
ter+0.020
ous+0.018
fully+0.018
ya+0.018
illy+0.018
Neg logit contributions
house-0.012
oc-0.010
 fort-0.009
 case-0.009
packs-0.008
Token day
Token ID1110
Feature activation-5.36e-02

Pos logit contributions
house+0.088
packs+0.072
 fort+0.071
tr+0.071
oc+0.067
Neg logit contributions
ter-0.154
fully-0.135
ting-0.135
ous-0.135
izing-0.133
Token,
Token ID11
Feature activation+3.95e-02

Pos logit contributions
house+0.049
oc+0.046
packs+0.036
 case+0.036
 fort+0.034
Neg logit contributions
ter-0.124
redible-0.123
ous-0.117
ya-0.117
iful-0.116
Token he
Token ID339
Feature activation+1.03e-02

Pos logit contributions
ous+0.082
illy+0.078
iful+0.078
ying+0.077
redible+0.077
Neg logit contributions
house-0.044
 case-0.041
oc-0.039
 lot-0.038
 fort-0.037
Token very
Token ID845
Feature activation-3.66e-02

Pos logit contributions
house+0.039
packs+0.035
tr+0.033
 fort+0.033
oc+0.032
Neg logit contributions
ter-0.036
fully-0.031
ous-0.030
ting-0.029
 Dolphin-0.029
Token happy
Token ID3772
Feature activation+1.74e-02

Pos logit contributions
house+0.074
packs+0.067
tr+0.067
 fort+0.064
ringe+0.064
Neg logit contributions
ter-0.056
fully-0.047
ous-0.046
 Dolphin-0.045
ting-0.044
Token.
Token ID13
Feature activation+3.17e-03

Pos logit contributions
ter+0.032
ous+0.028
fully+0.028
 Dolphin+0.027
ancer+0.027
Neg logit contributions
house-0.030
packs-0.027
 fort-0.025
tr-0.024
oc-0.024
Token
Token ID198
Feature activation+7.52e-03

Pos logit contributions
ter+0.006
ous+0.006
heart+0.006
fully+0.006
ya+0.006
Neg logit contributions
house-0.005
oc-0.004
 fort-0.004
packs-0.004
tr-0.004
Token
Token ID198
Feature activation+7.98e-03

Pos logit contributions
ter+0.015
ous+0.013
ting+0.013
fully+0.012
izing+0.012
Neg logit contributions
house-0.012
oc-0.011
packs-0.010
 fort-0.010
tr-0.010
TokenHop
Token ID23483
Feature activation+3.14e-01

Pos logit contributions
ter+0.015
ous+0.014
fully+0.013
ancer+0.013
 Dolphin+0.012
Neg logit contributions
house-0.013
oc-0.012
 fort-0.011
tr-0.010
ris-0.010
Token and
Token ID290
Feature activation+1.76e-02

Pos logit contributions
ter+0.676
ous+0.641
ancer+0.621
fully+0.616
izing+0.611
Neg logit contributions
house-0.430
oc-0.355
packs-0.324
 fort-0.293
ris-0.291
Token the
Token ID262
Feature activation+4.93e-02

Pos logit contributions
ter+0.029
ous+0.028
fully+0.026
ancer+0.026
ting+0.025
Neg logit contributions
house-0.031
oc-0.028
packs-0.028
 fort-0.027
ringe-0.026
Token squirrel
Token ID33039
Feature activation+1.32e-01

Pos logit contributions
ter+0.066
fully+0.053
ous+0.051
 Dolphin+0.050
ting+0.050
Neg logit contributions
house-0.112
packs-0.103
tr-0.100
 fort-0.098
oc-0.096
Token played
Token ID2826
Feature activation-2.41e-02

Pos logit contributions
ter+0.219
fully+0.196
ting+0.184
 Hop+0.183
ous+0.179
Neg logit contributions
ringe-0.214
 handcuffs-0.214
house-0.213
packs-0.213
tr-0.207
Token the
Token ID262
Feature activation+4.93e-02

Pos logit contributions
house+0.042
packs+0.037
 fort+0.036
tr+0.035
oc+0.034
Neg logit contributions
ter-0.044
ous-0.039
fully-0.038
 Dolphin-0.037
ting-0.037
Token silly
Token ID14397
Feature activation-3.67e-03

Pos logit contributions
fully+0.045
ter+0.043
ivated+0.041
izing+0.040
ous+0.040
Neg logit contributions
house-0.053
 fort-0.048
oc-0.048
omo-0.048
tr-0.045
Token bunny
Token ID44915
Feature activation+1.44e-01

Pos logit contributions
house+0.007
 fort+0.007
oc+0.006
 choice+0.006
packs+0.006
Neg logit contributions
ivated-0.005
ter-0.005
fully-0.005
gging-0.005
ivating-0.005
Token?"
Token ID1701
Feature activation+3.88e-02

Pos logit contributions
ter+0.227
fully+0.201
ous+0.193
izing+0.188
ting+0.188
Neg logit contributions
house-0.288
packs-0.250
 fort-0.242
oc-0.240
tr-0.240
Token
Token ID198
Feature activation+6.84e-03

Pos logit contributions
ter+0.084
iful+0.081
ous+0.081
fully+0.079
izing+0.077
Neg logit contributions
house-0.052
oc-0.047
 handcuffs-0.041
 bars-0.037
tr-0.037
Token
Token ID198
Feature activation+7.48e-03

Pos logit contributions
ter+0.014
fully+0.012
ous+0.012
izing+0.011
ting+0.011
Neg logit contributions
house-0.011
oc-0.010
tr-0.009
packs-0.009
 handcuffs-0.009
TokenHop
Token ID23483
Feature activation+3.14e-01

Pos logit contributions
ter+0.013
fully+0.012
ous+0.012
ancer+0.011
iful+0.011
Neg logit contributions
house-0.014
oc-0.013
tr-0.011
ringe-0.011
 handcuffs-0.011
Tokensc
Token ID1416
Feature activation-1.22e-02

Pos logit contributions
ancer+0.641
ter+0.625
 Him+0.615
izing+0.598
ous+0.595
Neg logit contributions
house-0.414
oc-0.382
ris-0.286
tr-0.280
 handcuffs-0.277
Tokenot
Token ID313
Feature activation-3.95e-02

Pos logit contributions
house+0.023
packs+0.020
tr+0.019
 fort+0.019
oc+0.019
Neg logit contributions
ter-0.021
ous-0.018
fully-0.018
ting-0.017
 Dolphin-0.017
Tokench
Token ID354
Feature activation+4.50e-02

Pos logit contributions
house+0.055
oc+0.051
tr+0.046
ris+0.043
 bars+0.040
Neg logit contributions
ter-0.077
ancer-0.076
 Him-0.070
izing-0.070
iring-0.070
Token replied
Token ID8712
Feature activation+1.90e-02

Pos logit contributions
izing+0.067
ter+0.063
ya+0.062
 Dolphin+0.062
ivating+0.061
Neg logit contributions
house-0.082
packs-0.070
oc-0.068
tr-0.068
ringe-0.063
Token,
Token ID11
Feature activation+3.07e-02

Pos logit contributions
ancer+0.042
ter+0.039
iful+0.039
icated+0.038
izing+0.038
Neg logit contributions
house-0.024
oc-0.018
tr-0.017
 fort-0.015
 handcuffs-0.014
Tokenrow
Token ID808
Feature activation+1.16e-01

Pos logit contributions
ter+0.012
fully+0.010
ous+0.010
 Dolphin+0.010
ting+0.010
Neg logit contributions
house-0.016
packs-0.014
tr-0.014
 fort-0.013
ringe-0.013
Token.
Token ID13
Feature activation-2.38e-03

Pos logit contributions
ter+0.205
fully+0.176
ous+0.175
 Dolphin+0.172
ting+0.171
Neg logit contributions
house-0.207
packs-0.184
tr-0.176
 fort-0.174
oc-0.168
Token
Token ID220
Feature activation+6.98e-03

Pos logit contributions
house+0.004
packs+0.004
tr+0.004
 fort+0.004
oc+0.004
Neg logit contributions
ter-0.004
fully-0.004
ous-0.003
ting-0.003
 Dolphin-0.003
Token
Token ID198
Feature activation+3.52e-03

Pos logit contributions
ter+0.013
fully+0.011
ting+0.011
ous+0.011
izing+0.010
Neg logit contributions
house-0.013
packs-0.011
tr-0.011
 fort-0.010
oc-0.010
Token
Token ID198
Feature activation+4.79e-03

Pos logit contributions
ter+0.006
fully+0.005
ous+0.005
ting+0.005
 Dolphin+0.005
Neg logit contributions
house-0.007
packs-0.006
tr-0.006
 fort-0.006
oc-0.006
TokenHop
Token ID23483
Feature activation+3.13e-01

Pos logit contributions
ter+0.007
fully+0.006
 Dolphin+0.006
ting+0.006
izing+0.006
Neg logit contributions
house-0.009
packs-0.009
 fort-0.009
tr-0.009
ringe-0.008
Tokensc
Token ID1416
Feature activation-1.30e-02

Pos logit contributions
ter+0.638
ous+0.581
fully+0.560
ting+0.552
ancer+0.552
Neg logit contributions
house-0.482
packs-0.406
oc-0.403
tr-0.364
 fort-0.356
Tokenot
Token ID313
Feature activation-3.78e-02

Pos logit contributions
house+0.026
packs+0.023
tr+0.023
 fort+0.022
ringe+0.021
Neg logit contributions
ter-0.021
fully-0.018
ous-0.017
 Dolphin-0.017
ting-0.017
Tokench
Token ID354
Feature activation+4.58e-02

Pos logit contributions
house+0.062
oc+0.055
tr+0.050
packs+0.050
ris+0.048
Neg logit contributions
ter-0.071
izing-0.061
ancer-0.061
 Dolphin-0.061
ous-0.059
Token hopped
Token ID45230
Feature activation+7.46e-02

Pos logit contributions
ter+0.070
 Dolphin+0.064
ya+0.063
izing+0.063
ous+0.062
Neg logit contributions
house-0.089
packs-0.076
oc-0.076
 fort-0.072
tr-0.071
Token to
Token ID284
Feature activation-2.32e-02

Pos logit contributions
ter+0.132
ous+0.127
ancer+0.121
izing+0.118
gging+0.116
Neg logit contributions
house-0.125
 fort-0.105
tr-0.102
 handcuffs-0.100
packs-0.100
Token
Token ID6184
Feature activation+4.52e-02

Pos logit contributions
ter+0.008
fully+0.007
izing+0.007
ous+0.007
ble+0.006
Neg logit contributions
house-0.006
 handcuffs-0.005
packs-0.005
oc-0.005
tr-0.005
Token
Token ID95
Feature activation+9.67e-02

Pos logit contributions
ter+0.102
fully+0.101
ous+0.098
ancer+0.095
heart+0.094
Neg logit contributions
house-0.059
oc-0.046
atters-0.038
 fort-0.036
tr-0.036
Token
Token ID26391
Feature activation+1.65e-01

Pos logit contributions
ter+0.181
fully+0.164
ous+0.162
ancer+0.155
 Dolphin+0.152
Neg logit contributions
house-0.169
tr-0.140
oc-0.140
 fort-0.134
ringe-0.133
Token
Token ID129
Feature activation+1.11e-01

Pos logit contributions
ter+0.328
fully+0.293
ous+0.279
ting+0.276
 Dolphin+0.271
Neg logit contributions
house-0.273
packs-0.235
tr-0.223
oc-0.222
 fort-0.220
Token
Token ID241
Feature activation+4.17e-02

Pos logit contributions
ter+0.189
ous+0.161
fully+0.161
ancer+0.161
 Dolphin+0.153
Neg logit contributions
house-0.210
oc-0.178
tr-0.175
 fort-0.175
packs-0.174
TokenHop
Token ID23483
Feature activation+3.13e-01

Pos logit contributions
ter+0.068
fully+0.057
ous+0.056
ting+0.055
izing+0.054
Neg logit contributions
house-0.079
packs-0.072
oc-0.069
tr-0.069
ris-0.067
Token on
Token ID319
Feature activation-1.05e-01

Pos logit contributions
ter+0.564
fully+0.538
ous+0.524
 Him+0.509
ancer+0.507
Neg logit contributions
house-0.506
oc-0.443
packs-0.401
 handcuffs-0.396
ris-0.379
Token my
Token ID616
Feature activation-1.21e-03

Pos logit contributions
house+0.217
packs+0.194
oc+0.187
 fort+0.186
tr+0.185
Neg logit contributions
ter-0.156
fully-0.133
ous-0.132
ting-0.122
ancer-0.122
Token back
Token ID736
Feature activation-6.81e-02

Pos logit contributions
house+0.002
 fort+0.002
tr+0.002
oc+0.002
packs+0.002
Neg logit contributions
ter-0.002
fully-0.002
ancer-0.002
ting-0.002
ous-0.002
Token,
Token ID11
Feature activation+3.38e-02

Pos logit contributions
house+0.137
tr+0.125
packs+0.122
ringe+0.119
 fort+0.118
Neg logit contributions
ter-0.103
fully-0.086
ous-0.085
 Dolphin-0.082
ting-0.081
Tokenâ
Token ID22940
Feature activation+9.40e-02

Pos logit contributions
ter+0.052
fully+0.047
ous+0.046
izing+0.043
ting+0.042
Neg logit contributions
house-0.066
packs-0.057
 fort-0.056
oc-0.056
tr-0.056
Token that
Token ID326
Feature activation-3.89e-02

Pos logit contributions
ter+0.014
ancer+0.013
iful+0.013
ying+0.013
redible+0.013
Neg logit contributions
house-0.012
 bars-0.010
 handcuffs-0.010
 fort-0.009
tr-0.009
Token too
Token ID1165
Feature activation-1.65e-02

Pos logit contributions
house+0.073
packs+0.064
tr+0.063
 fort+0.062
oc+0.061
Neg logit contributions
ter-0.066
fully-0.056
ous-0.056
ting-0.054
 Dolphin-0.054
Token.
Token ID13
Feature activation-2.36e-03

Pos logit contributions
house+0.027
packs+0.023
 fort+0.022
oc+0.022
tr+0.021
Neg logit contributions
ter-0.033
fully-0.028
ting-0.028
 Dolphin-0.027
ous-0.027
Token
Token ID198
Feature activation+6.13e-03

Pos logit contributions
house+0.003
oc+0.003
 fort+0.003
packs+0.002
tr+0.002
Neg logit contributions
ter-0.005
ous-0.005
iful-0.005
fully-0.005
bling-0.005
Token
Token ID198
Feature activation+6.56e-03

Pos logit contributions
ter+0.012
ous+0.010
ting+0.010
fully+0.010
izing+0.009
Neg logit contributions
house-0.010
packs-0.009
oc-0.009
 fort-0.009
tr-0.009
TokenHop
Token ID23483
Feature activation+3.12e-01

Pos logit contributions
ter+0.009
ous+0.008
fully+0.007
ancer+0.007
 Dolphin+0.007
Neg logit contributions
house-0.014
oc-0.013
 fort-0.012
tr-0.012
ringe-0.012
Token asked
Token ID1965
Feature activation-2.14e-02

Pos logit contributions
ter+0.621
ous+0.584
izing+0.557
fully+0.556
ancer+0.555
Neg logit contributions
house-0.476
packs-0.403
oc-0.392
 fort-0.355
tr-0.354
Token the
Token ID262
Feature activation+4.40e-02

Pos logit contributions
house+0.034
oc+0.030
 fort+0.030
packs+0.029
 handcuffs+0.028
Neg logit contributions
ter-0.038
ous-0.037
izing-0.037
ivating-0.035
ivated-0.035
Token butterfly
Token ID35113
Feature activation+1.70e-01

Pos logit contributions
ter+0.051
ous+0.049
fully+0.048
 Joey+0.043
 Nathan+0.043
Neg logit contributions
house-0.097
 fort-0.088
tr-0.088
packs-0.083
 case-0.080
Token,
Token ID11
Feature activation+3.19e-02

Pos logit contributions
ter+0.325
izing+0.310
redible+0.302
ous+0.297
ivating+0.296
Neg logit contributions
house-0.279
packs-0.237
 fort-0.227
oc-0.212
tr-0.204
Token "
Token ID366
Feature activation+2.48e-02

Pos logit contributions
izing+0.069
heart+0.067
ternally+0.065
ime+0.065
ous+0.065
Neg logit contributions
house-0.034
tr-0.026
 named-0.026
 fort-0.024
 handcuffs-0.023
Token with
Token ID351
Feature activation-8.28e-02

Pos logit contributions
house+0.002
packs+0.002
oc+0.002
tr+0.002
 fort+0.002
Neg logit contributions
ter-0.002
ous-0.002
fully-0.002
 Dolphin-0.002
ting-0.002
Token you
Token ID345
Feature activation+1.87e-01

Pos logit contributions
house+0.163
packs+0.150
oc+0.141
 fort+0.133
ringe+0.130
Neg logit contributions
ter-0.128
ous-0.112
fully-0.107
 Dolphin-0.104
ancer-0.102
Token."
Token ID526
Feature activation-2.45e-03

Pos logit contributions
ter+0.374
fully+0.331
ous+0.329
 Dolphin+0.322
ting+0.318
Neg logit contributions
house-0.298
packs-0.262
 fort-0.236
oc-0.234
tr-0.230
Token
Token ID198
Feature activation+2.16e-03

Pos logit contributions
house+0.003
oc+0.003
 fort+0.002
 bars+0.002
packs+0.002
Neg logit contributions
ter-0.005
ous-0.005
bling-0.005
ying-0.005
iful-0.005
Token
Token ID198
Feature activation+2.83e-03

Pos logit contributions
ter+0.005
ous+0.004
ting+0.004
fully+0.004
ping+0.004
Neg logit contributions
house-0.003
oc-0.003
ris-0.002
packs-0.002
 fort-0.002
TokenHop
Token ID23483
Feature activation+3.12e-01

Pos logit contributions
ter+0.004
ous+0.004
ancer+0.004
fully+0.004
iddy+0.003
Neg logit contributions
house-0.006
oc-0.005
 fort-0.005
ris-0.005
tr-0.005
Token listened
Token ID16399
Feature activation+1.18e-01

Pos logit contributions
ter+0.662
ancer+0.651
ous+0.646
izing+0.604
 Him+0.601
Neg logit contributions
house-0.428
oc-0.369
packs-0.288
tr-0.279
ringe-0.267
Token carefully
Token ID7773
Feature activation+3.49e-02

Pos logit contributions
ter+0.271
ancer+0.232
redible+0.229
 Dolphin+0.228
ting+0.223
Neg logit contributions
house-0.157
tr-0.128
oc-0.124
 fort-0.120
packs-0.114
Token to
Token ID284
Feature activation-2.36e-02

Pos logit contributions
ter+0.062
 Dolphin+0.053
ous+0.052
ancer+0.052
fully+0.050
Neg logit contributions
house-0.063
packs-0.056
tr-0.053
 fort-0.053
oc-0.052
Token the
Token ID262
Feature activation+4.42e-02

Pos logit contributions
house+0.041
oc+0.036
 fort+0.035
packs+0.034
 handcuffs+0.034
Neg logit contributions
ter-0.039
ous-0.037
fully-0.034
iddy-0.033
izing-0.033
Token wise
Token ID10787
Feature activation+1.51e-01

Pos logit contributions
ter+0.069
fully+0.061
 Nathan+0.060
ancer+0.060
 Dolphin+0.059
Neg logit contributions
house-0.084
oc-0.076
 fort-0.075
tr-0.074
ringe-0.070
Token of
Token ID286
Feature activation+3.76e-02

Pos logit contributions
house+0.363
packs+0.336
tr+0.329
ringe+0.326
 fort+0.323
Neg logit contributions
ter-0.206
fully-0.178
ous-0.164
ting-0.157
ing-0.154
Token fun
Token ID1257
Feature activation-5.78e-02

Pos logit contributions
ter+0.072
fully+0.065
izing+0.060
gging+0.060
 Dolphin+0.060
Neg logit contributions
house-0.061
tr-0.054
packs-0.052
oc-0.050
ringe-0.049
Token.
Token ID13
Feature activation-1.63e-04

Pos logit contributions
house+0.116
tr+0.112
packs+0.109
ringe+0.107
 handcuffs+0.107
Neg logit contributions
ter-0.085
fully-0.075
izing-0.068
ting-0.068
ous-0.065
Token
Token ID198
Feature activation+4.58e-03

Pos logit contributions
house+0.000
oc+0.000
 fort+0.000
packs+0.000
tr+0.000
Neg logit contributions
ter-0.000
ous-0.000
fully-0.000
izing-0.000
redible-0.000
Token
Token ID198
Feature activation+4.97e-03

Pos logit contributions
ter+0.008
ous+0.007
fully+0.007
ting+0.007
 Dolphin+0.006
Neg logit contributions
house-0.008
packs-0.008
tr-0.007
 fort-0.007
oc-0.007
TokenHop
Token ID23483
Feature activation+3.12e-01

Pos logit contributions
ter+0.009
ous+0.008
fully+0.007
ancer+0.007
ting+0.007
Neg logit contributions
house-0.009
oc-0.008
tr-0.007
 fort-0.007
packs-0.007
Token wanted
Token ID2227
Feature activation-1.01e-02

Pos logit contributions
ter+0.652
ous+0.609
fully+0.581
ancer+0.576
izing+0.575
Neg logit contributions
house-0.454
oc-0.371
packs-0.371
 fort-0.336
tr-0.326
Token to
Token ID284
Feature activation-2.09e-02

Pos logit contributions
house+0.016
packs+0.014
 fort+0.013
tr+0.013
 handcuffs+0.012
Neg logit contributions
ter-0.020
ous-0.018
izing-0.018
fully-0.018
ancer-0.018
Token play
Token ID711
Feature activation-6.63e-02

Pos logit contributions
house+0.033
packs+0.031
tr+0.030
oc+0.028
 fort+0.028
Neg logit contributions
ter-0.040
fully-0.035
 Dolphin-0.034
 Hop-0.034
 Fly-0.033
Token a
Token ID257
Feature activation-1.69e-02

Pos logit contributions
house+0.123
packs+0.106
 fort+0.103
tr+0.103
oc+0.099
Neg logit contributions
ter-0.116
ous-0.101
fully-0.100
 Dolphin-0.097
ting-0.096
Token game
Token ID983
Feature activation-1.01e-01

Pos logit contributions
house+0.025
packs+0.023
tr+0.021
ringe+0.021
 handcuffs+0.020
Neg logit contributions
ter-0.035
fully-0.031
ous-0.030
 Dolphin-0.030
izing-0.030
Token the
Token ID262
Feature activation+4.53e-02

Pos logit contributions
ter+0.095
ous+0.082
fully+0.080
ancer+0.077
 Dolphin+0.074
Neg logit contributions
house-0.121
packs-0.112
oc-0.105
 fort-0.102
 handcuffs-0.095
Token world
Token ID995
Feature activation+1.17e-01

Pos logit contributions
ter+0.086
fully+0.078
ous+0.077
 Dolphin+0.070
ancer+0.070
Neg logit contributions
house-0.078
 fort-0.067
packs-0.067
oc-0.066
tr-0.062
Token".
Token ID1911
Feature activation+6.03e-02

Pos logit contributions
ter+0.210
fully+0.178
 Dolphin+0.175
ous+0.175
ancer+0.170
Neg logit contributions
house-0.210
packs-0.184
oc-0.172
tr-0.172
 fort-0.171
Token
Token ID198
Feature activation+1.11e-02

Pos logit contributions
ter+0.110
ous+0.107
fully+0.105
izing+0.098
iful+0.097
Neg logit contributions
house-0.102
oc-0.093
 handcuffs-0.084
packs-0.079
 bars-0.079
Token
Token ID198
Feature activation+1.16e-02

Pos logit contributions
ter+0.026
ous+0.023
fully+0.023
iful+0.022
izing+0.022
Neg logit contributions
house-0.014
oc-0.013
ris-0.011
 handcuffs-0.010
packs-0.010
TokenHop
Token ID23483
Feature activation+3.11e-01

Pos logit contributions
ter+0.018
ous+0.017
fully+0.016
ancer+0.015
iddy+0.015
Neg logit contributions
house-0.023
oc-0.022
 handcuffs-0.019
ris-0.019
tr-0.018
Token was
Token ID373
Feature activation-3.02e-02

Pos logit contributions
ter+0.649
ancer+0.606
ous+0.590
izing+0.586
 Dolphin+0.569
Neg logit contributions
house-0.433
oc-0.377
packs-0.333
ris-0.308
tr-0.302
Token so
Token ID523
Feature activation+3.41e-03

Pos logit contributions
house+0.061
packs+0.054
tr+0.052
oc+0.051
 fort+0.050
Neg logit contributions
ter-0.047
fully-0.040
ous-0.039
ancer-0.038
ting-0.038
Token excited
Token ID6568
Feature activation+3.14e-02

Pos logit contributions
ter+0.005
fully+0.005
ting+0.004
ous+0.004
ancer+0.004
Neg logit contributions
house-0.007
packs-0.006
oc-0.006
tr-0.006
 fort-0.006
Token.
Token ID13
Feature activation+5.48e-03

Pos logit contributions
ter+0.065
ancer+0.062
 Dolphin+0.061
ternally+0.060
redible+0.059
Neg logit contributions
house-0.043
 fort-0.034
packs-0.032
oc-0.031
 bars-0.030
Token He
Token ID679
Feature activation-1.98e-02

Pos logit contributions
ter+0.012
fully+0.011
izing+0.011
ous+0.010
redible+0.010
Neg logit contributions
house-0.007
oc-0.006
 handcuffs-0.006
tr-0.006
packs-0.006
Token.
Token ID13
Feature activation-3.01e-03

Pos logit contributions
ter+0.173
fully+0.148
ous+0.145
ting+0.142
 Dolphin+0.142
Neg logit contributions
house-0.184
packs-0.165
tr-0.158
 fort-0.156
oc-0.152
Token He
Token ID679
Feature activation-2.56e-02

Pos logit contributions
house+0.005
packs+0.004
tr+0.004
 fort+0.004
oc+0.004
Neg logit contributions
ter-0.006
fully-0.005
ous-0.005
ting-0.005
izing-0.005
Token said
Token ID531
Feature activation-1.68e-02

Pos logit contributions
house+0.045
packs+0.041
tr+0.039
 fort+0.038
ringe+0.038
Neg logit contributions
ter-0.046
fully-0.039
ous-0.038
 Dolphin-0.038
ting-0.037
Token,
Token ID11
Feature activation+2.85e-02

Pos logit contributions
house+0.019
 fort+0.016
packs+0.015
 collection+0.014
oc+0.014
Neg logit contributions
ter-0.037
ancer-0.035
izing-0.035
ting-0.034
ivating-0.034
Token "
Token ID366
Feature activation+2.24e-02

Pos logit contributions
ter+0.054
iring+0.054
heart+0.054
izing+0.053
ternally+0.051
Neg logit contributions
house-0.037
tr-0.032
oc-0.031
 trailer-0.027
 fort-0.027
TokenHop
Token ID23483
Feature activation+3.10e-01

Pos logit contributions
ter+0.039
ous+0.035
fully+0.034
ting+0.032
izing+0.032
Neg logit contributions
house-0.040
tr-0.035
oc-0.035
packs-0.034
ringe-0.033
Token,
Token ID11
Feature activation+2.92e-02

Pos logit contributions
ter+0.495
fully+0.487
ous+0.486
izing+0.445
ancer+0.442
Neg logit contributions
house-0.540
oc-0.480
packs-0.462
 handcuffs-0.439
 fort-0.432
Token hop
Token ID1725
Feature activation+5.44e-02

Pos logit contributions
ter+0.047
izing+0.046
fully+0.045
ous+0.045
gging+0.045
Neg logit contributions
house-0.049
 fort-0.045
tr-0.043
oc-0.041
 bars-0.040
Token,
Token ID11
Feature activation+2.97e-02

Pos logit contributions
ter+0.098
ous+0.086
fully+0.083
ancer+0.083
 Dolphin+0.081
Neg logit contributions
house-0.095
packs-0.082
 fort-0.081
oc-0.081
tr-0.079
Token hop
Token ID1725
Feature activation+5.45e-02

Pos logit contributions
ter+0.050
izing+0.047
fully+0.047
ous+0.046
gging+0.046
Neg logit contributions
house-0.051
 fort-0.045
tr-0.044
oc-0.041
packs-0.041
Token,
Token ID11
Feature activation+3.04e-02

Pos logit contributions
ter+0.100
ous+0.088
ancer+0.085
fully+0.084
 Dolphin+0.083
Neg logit contributions
house-0.094
packs-0.080
 fort-0.080
oc-0.079
tr-0.077
Token love
Token ID1842
Feature activation+1.80e-02

Pos logit contributions
ter+0.404
fully+0.360
ous+0.348
 Dolphin+0.346
izing+0.345
Neg logit contributions
house-0.265
packs-0.237
oc-0.209
tr-0.208
 fort-0.201
Token.
Token ID13
Feature activation-5.72e-03

Pos logit contributions
ter+0.032
 Dolphin+0.029
ous+0.029
ivated+0.028
fully+0.028
Neg logit contributions
house-0.030
packs-0.028
oc-0.025
 fort-0.024
tr-0.023
Token
Token ID220
Feature activation+4.35e-03

Pos logit contributions
house+0.010
oc+0.008
packs+0.008
tr+0.007
 handcuffs+0.007
Neg logit contributions
ter-0.011
fully-0.010
ous-0.010
izing-0.010
ble-0.009
Token
Token ID198
Feature activation+4.22e-03

Pos logit contributions
ter+0.009
fully+0.008
ya+0.008
izing+0.008
ting+0.008
Neg logit contributions
house-0.007
packs-0.006
tr-0.005
oc-0.005
 handcuffs-0.005
Token
Token ID198
Feature activation+4.99e-03

Pos logit contributions
ter+0.009
ous+0.008
fully+0.008
ting+0.008
ancer+0.008
Neg logit contributions
house-0.006
oc-0.005
ris-0.004
packs-0.004
tr-0.004
TokenHop
Token ID23483
Feature activation+3.10e-01

Pos logit contributions
ter+0.007
ous+0.007
fully+0.006
ancer+0.006
 Dolphin+0.006
Neg logit contributions
house-0.010
oc-0.009
packs-0.009
tr-0.009
 fort-0.009
Token felt
Token ID2936
Feature activation+3.99e-02

Pos logit contributions
ter+0.604
ous+0.579
ancer+0.571
izing+0.550
fully+0.545
Neg logit contributions
house-0.461
oc-0.397
packs-0.348
ringe-0.318
tr-0.317
Token a
Token ID257
Feature activation-1.85e-02

Pos logit contributions
ter+0.071
ous+0.063
fully+0.061
ancer+0.060
 Dolphin+0.059
Neg logit contributions
house-0.072
oc-0.061
packs-0.060
 handcuffs-0.058
tr-0.057
Token little
Token ID1310
Feature activation+9.24e-02

Pos logit contributions
house+0.037
oc+0.033
tr+0.032
 fort+0.032
packs+0.030
Neg logit contributions
ter-0.029
ting-0.025
fully-0.024
ous-0.023
ancer-0.023
Token better
Token ID1365
Feature activation-1.71e-02

Pos logit contributions
ter+0.180
fully+0.160
ous+0.157
ting+0.154
 Dolphin+0.153
Neg logit contributions
house-0.149
packs-0.125
tr-0.124
 fort-0.123
oc-0.121
Token knowing
Token ID6970
Feature activation-3.02e-02

Pos logit contributions
house+0.036
packs+0.033
tr+0.032
 fort+0.031
ringe+0.031
Neg logit contributions
ter-0.026
fully-0.021
ous-0.021
ting-0.020
 Dolphin-0.020
Token to
Token ID284
Feature activation-2.60e-02

Pos logit contributions
ter+0.458
ous+0.386
fully+0.385
 Dolphin+0.380
ting+0.377
Neg logit contributions
house-0.495
packs-0.442
 fort-0.415
tr-0.412
oc-0.407
Token reach
Token ID3151
Feature activation+1.07e-02

Pos logit contributions
house+0.042
packs+0.038
tr+0.036
 fort+0.035
oc+0.034
Neg logit contributions
ter-0.051
fully-0.044
ous-0.043
ting-0.043
 Dolphin-0.042
Token.
Token ID13
Feature activation-5.78e-03

Pos logit contributions
ter+0.021
ous+0.019
 Dolphin+0.018
ancer+0.018
fully+0.018
Neg logit contributions
house-0.016
packs-0.015
 fort-0.014
 handcuffs-0.014
oc-0.013
Token
Token ID198
Feature activation+2.12e-04

Pos logit contributions
house+0.007
oc+0.005
 fort+0.004
ringe+0.004
 pockets+0.004
Neg logit contributions
ous-0.014
ter-0.013
iful-0.013
fully-0.013
heart-0.013
Token
Token ID198
Feature activation+1.18e-03

Pos logit contributions
ter+0.000
ting+0.000
ous+0.000
fully+0.000
izing+0.000
Neg logit contributions
house-0.000
packs-0.000
oc-0.000
ris-0.000
 fort-0.000
TokenHop
Token ID23483
Feature activation+3.08e-01

Pos logit contributions
ous+0.002
ter+0.002
fully+0.002
ancer+0.001
iful+0.001
Neg logit contributions
house-0.002
oc-0.002
ringe-0.002
 case-0.002
 fort-0.002
Token asked
Token ID1965
Feature activation-2.56e-02

Pos logit contributions
ter+0.628
ous+0.589
ancer+0.563
fully+0.556
izing+0.553
Neg logit contributions
house-0.463
packs-0.383
oc-0.379
 fort-0.341
tr-0.336
Token his
Token ID465
Feature activation-5.85e-02

Pos logit contributions
house+0.036
oc+0.032
packs+0.031
 fort+0.031
 handcuffs+0.030
Neg logit contributions
ter-0.049
izing-0.048
ous-0.048
ivating-0.047
ivated-0.046
Token friend
Token ID1545
Feature activation-7.05e-02

Pos logit contributions
house+0.099
packs+0.088
tr+0.084
oc+0.083
ringe+0.082
Neg logit contributions
ter-0.109
ous-0.093
fully-0.093
 Dolphin-0.092
 Hop-0.088
Token,
Token ID11
Feature activation+2.58e-02

Pos logit contributions
house+0.096
packs+0.081
 fort+0.076
oc+0.068
tr+0.068
Neg logit contributions
izing-0.149
ter-0.140
ivating-0.137
ivated-0.135
redible-0.132
Token a
Token ID257
Feature activation-2.06e-02

Pos logit contributions
izing+0.057
ping+0.052
ous+0.050
ly+0.050
gging+0.049
Neg logit contributions
house-0.031
 handcuffs-0.024
 bisc-0.024
 named-0.024
tr-0.024
Token,"
Token ID553
Feature activation+5.07e-02

Pos logit contributions
ter+0.030
 Dolphin+0.028
ous+0.028
ancer+0.028
gging+0.027
Neg logit contributions
house-0.024
packs-0.021
 fort-0.020
tr-0.019
oc-0.017
Token she
Token ID673
Feature activation-3.73e-02

Pos logit contributions
ter+0.094
ous+0.084
fully+0.083
ting+0.081
izing+0.080
Neg logit contributions
house-0.085
packs-0.074
tr-0.070
 fort-0.069
oc-0.069
Token said
Token ID531
Feature activation-1.69e-02

Pos logit contributions
house+0.056
packs+0.048
tr+0.047
 fort+0.046
oc+0.043
Neg logit contributions
ter-0.077
ting-0.067
ous-0.067
fully-0.067
izing-0.066
Token.
Token ID13
Feature activation-1.88e-02

Pos logit contributions
house+0.020
packs+0.017
tr+0.014
 fort+0.013
 handcuffs+0.013
Neg logit contributions
ter-0.038
redible-0.036
 Dolphin-0.036
izing-0.035
ancer-0.035
Token "
Token ID366
Feature activation+1.32e-02

Pos logit contributions
house+0.029
packs+0.025
tr+0.024
 handcuffs+0.022
oc+0.022
Neg logit contributions
ter-0.038
izing-0.034
fully-0.033
ous-0.033
iful-0.032
TokenHop
Token ID23483
Feature activation+3.08e-01

Pos logit contributions
ter+0.023
ous+0.020
fully+0.020
ting+0.019
 Dolphin+0.018
Neg logit contributions
house-0.024
tr-0.021
packs-0.021
oc-0.020
 fort-0.020
Token in
Token ID287
Feature activation+9.67e-02

Pos logit contributions
ter+0.512
fully+0.500
ous+0.497
izing+0.473
redible+0.463
Neg logit contributions
house-0.508
oc-0.455
packs-0.444
 handcuffs-0.419
ringe-0.395
Token."
Token ID526
Feature activation-3.10e-03

Pos logit contributions
ter+0.124
fully+0.097
 Dolphin+0.095
ting+0.095
ous+0.094
Neg logit contributions
house-0.217
tr-0.201
packs-0.199
 fort-0.192
ringe-0.190
Token
Token ID198
Feature activation+5.48e-03

Pos logit contributions
house+0.005
packs+0.004
tr+0.003
 fort+0.003
 handcuffs+0.003
Neg logit contributions
ter-0.006
izing-0.006
ous-0.006
iful-0.006
fully-0.006
Token
Token ID198
Feature activation+7.23e-03

Pos logit contributions
ter+0.011
ous+0.010
ting+0.010
fully+0.009
izing+0.009
Neg logit contributions
house-0.009
packs-0.008
oc-0.007
tr-0.007
 fort-0.007
TokenR
Token ID49
Feature activation+1.33e-01

Pos logit contributions
ter+0.015
ous+0.013
fully+0.013
ting+0.013
ancer+0.012
Neg logit contributions
house-0.011
packs-0.010
oc-0.009
tr-0.009
 fort-0.009
Token around
Token ID1088
Feature activation+6.42e-02

Pos logit contributions
ter+0.073
ous+0.065
ancer+0.061
 Dolphin+0.061
ting+0.061
Neg logit contributions
house-0.087
packs-0.079
 fort-0.076
tr-0.072
oc-0.071
Token looking
Token ID2045
Feature activation+1.13e-01

Pos logit contributions
ter+0.122
ous+0.115
redible+0.112
ancer+0.112
rus+0.109
Neg logit contributions
house-0.099
 fort-0.090
packs-0.088
oc-0.083
 collection-0.072
Token at
Token ID379
Feature activation-3.53e-02

Pos logit contributions
ter+0.276
redible+0.255
ous+0.252
 Dolphin+0.250
ner+0.249
Neg logit contributions
house-0.108
 fort-0.104
packs-0.101
oc-0.088
 pockets-0.075
Token the
Token ID262
Feature activation+4.42e-02

Pos logit contributions
house+0.078
packs+0.071
tr+0.069
 fort+0.068
oc+0.067
Neg logit contributions
ter-0.048
fully-0.040
ous-0.039
ting-0.037
 Dolphin-0.037
Token small
Token ID1402
Feature activation-2.82e-02

Pos logit contributions
fully+0.083
ter+0.079
ous+0.078
ner+0.074
gging+0.071
Neg logit contributions
 fort-0.060
house-0.056
packs-0.054
oc-0.054
 pile-0.052
Token fishes
Token ID49765
Feature activation+3.00e-01

Pos logit contributions
house+0.050
packs+0.048
oc+0.048
 fort+0.048
 pile+0.043
Neg logit contributions
ter-0.043
fully-0.043
ous-0.040
ancer-0.039
ivated-0.038
Token.
Token ID13
Feature activation+5.42e-03

Pos logit contributions
ter+0.590
redible+0.565
ous+0.542
ancer+0.541
mber+0.540
Neg logit contributions
house-0.443
packs-0.403
 fort-0.352
oc-0.346
 bars-0.314
Token And
Token ID843
Feature activation-2.93e-02

Pos logit contributions
ter+0.011
fully+0.010
ous+0.010
izing+0.009
ting+0.009
Neg logit contributions
house-0.009
packs-0.007
tr-0.007
oc-0.007
 fort-0.007
Token they
Token ID484
Feature activation-5.59e-02

Pos logit contributions
house+0.052
packs+0.045
oc+0.043
 fort+0.043
tr+0.042
Neg logit contributions
ter-0.051
ous-0.045
fully-0.044
ancer-0.043
redible-0.043
Token waved
Token ID26834
Feature activation+1.12e-01

Pos logit contributions
house+0.086
packs+0.074
tr+0.070
 fort+0.069
oc+0.068
Neg logit contributions
ter-0.112
fully-0.100
ous-0.100
izing-0.097
ting-0.097
Token to
Token ID284
Feature activation-1.84e-02

Pos logit contributions
ter+0.235
redible+0.216
ancer+0.212
izing+0.207
ner+0.207
Neg logit contributions
packs-0.146
house-0.145
 fort-0.130
oc-0.119
 handcuffs-0.114
Token climb
Token ID12080
Feature activation+4.82e-04

Pos logit contributions
house+0.046
packs+0.043
oc+0.042
tr+0.041
 fort+0.041
Neg logit contributions
ter-0.044
 Dolphin-0.039
fully-0.037
 Joey-0.036
 Hop-0.036
Token up
Token ID510
Feature activation+3.98e-02

Pos logit contributions
ter+0.001
fully+0.001
ous+0.001
 Dolphin+0.001
ancer+0.001
Neg logit contributions
house-0.001
packs-0.001
tr-0.001
 fort-0.001
oc-0.001
Token to
Token ID284
Feature activation-2.69e-02

Pos logit contributions
ter+0.052
fully+0.042
ous+0.041
 Dolphin+0.040
ting+0.040
Neg logit contributions
house-0.090
packs-0.083
tr-0.080
 fort-0.079
oc-0.077
Token the
Token ID262
Feature activation+4.41e-02

Pos logit contributions
house+0.064
packs+0.059
tr+0.058
 fort+0.057
oc+0.056
Neg logit contributions
ter-0.032
fully-0.025
ous-0.024
ting-0.024
 Dolphin-0.023
Token tree
Token ID5509
Feature activation-3.44e-02

Pos logit contributions
ter+0.096
fully+0.092
ous+0.085
izing+0.084
gging+0.084
Neg logit contributions
house-0.062
tr-0.056
 fort-0.054
packs-0.048
 bars-0.045
Tokenhouse
Token ID4803
Feature activation-3.07e-01

Pos logit contributions
house+0.068
packs+0.062
tr+0.060
 fort+0.059
ringe+0.059
Neg logit contributions
ter-0.054
fully-0.045
ous-0.044
ting-0.043
 Dolphin-0.042
Token in
Token ID287
Feature activation+9.62e-02

Pos logit contributions
house+0.575
packs+0.522
tr+0.508
oc+0.497
ringe+0.496
Neg logit contributions
ter-0.513
fully-0.440
ous-0.422
ting-0.417
 Dolphin-0.403
Token her
Token ID607
Feature activation-3.29e-02

Pos logit contributions
ter+0.055
 Fly+0.032
 young+0.030
 itself+0.028
ting+0.025
Neg logit contributions
tr-0.280
house-0.259
ringe-0.254
 fort-0.253
oc-0.252
Token backyard
Token ID24296
Feature activation-4.84e-02

Pos logit contributions
 handcuffs+0.054
ris+0.051
oc+0.050
ringe+0.047
house+0.045
Neg logit contributions
 young-0.049
ter-0.046
 little-0.043
 Fly-0.041
ya-0.041
Token,
Token ID11
Feature activation+2.53e-02

Pos logit contributions
house+0.092
packs+0.084
tr+0.083
ringe+0.081
oc+0.080
Neg logit contributions
ter-0.078
fully-0.066
ous-0.064
ting-0.063
 Dolphin-0.061
Token but
Token ID475
Feature activation+2.96e-02

Pos logit contributions
ter+0.044
ous+0.041
rus+0.039
 Dolphin+0.039
izing+0.039
Neg logit contributions
house-0.045
packs-0.040
 fort-0.037
 bars-0.036
tr-0.036
Token great
Token ID1049
Feature activation-5.67e-02

Pos logit contributions
packs+0.050
house+0.050
ringe+0.049
tr+0.047
 handcuffs+0.044
Neg logit contributions
ter-0.035
 Dolphin-0.034
 young-0.034
ya-0.032
 dolphin-0.032
Token time
Token ID640
Feature activation-2.27e-02

Pos logit contributions
house+0.110
packs+0.105
ringe+0.105
tr+0.101
 handcuffs+0.100
Neg logit contributions
ter-0.074
fully-0.072
ya-0.068
 Dolphin-0.068
 Joey-0.067
Token in
Token ID287
Feature activation+9.36e-02

Pos logit contributions
house+0.042
packs+0.038
tr+0.038
 fort+0.037
ringe+0.036
Neg logit contributions
ter-0.038
fully-0.033
ous-0.032
ting-0.032
 Dolphin-0.031
Token the
Token ID262
Feature activation+3.53e-02

Pos logit contributions
ter+0.062
 Henry+0.046
 Joey+0.045
 Dolphin+0.044
 Simon+0.042
Neg logit contributions
tr-0.259
house-0.253
oc-0.246
ringe-0.244
 fort-0.234
Token play
Token ID711
Feature activation-7.21e-02

Pos logit contributions
ter+0.061
 Dolphin+0.054
ting+0.051
ya+0.050
ous+0.049
Neg logit contributions
house-0.061
packs-0.058
tr-0.054
ringe-0.053
oc-0.053
Tokenhouse
Token ID4803
Feature activation-3.07e-01

Pos logit contributions
house+0.129
packs+0.126
tr+0.122
 handcuffs+0.118
ringe+0.118
Neg logit contributions
ter-0.122
fully-0.107
ting-0.103
 Dolphin-0.096
ing-0.095
Token.
Token ID13
Feature activation-1.31e-02

Pos logit contributions
house+0.587
packs+0.542
ringe+0.542
tr+0.537
oc+0.525
Neg logit contributions
ter-0.486
fully-0.420
ting-0.386
ous-0.386
izing-0.367
Token Tom
Token ID4186
Feature activation-7.41e-02

Pos logit contributions
house+0.027
packs+0.025
tr+0.024
 fort+0.024
ringe+0.024
Neg logit contributions
ter-0.019
fully-0.016
ous-0.015
 Dolphin-0.015
ting-0.015
Token and
Token ID290
Feature activation-6.09e-04

Pos logit contributions
house+0.122
packs+0.104
tr+0.100
 fort+0.100
oc+0.099
Neg logit contributions
ter-0.143
fully-0.125
ous-0.125
ting-0.121
 Dolphin-0.120
Token Sally
Token ID25737
Feature activation-1.54e-02

Pos logit contributions
house+0.001
packs+0.001
tr+0.001
 fort+0.001
oc+0.001
Neg logit contributions
ter-0.001
fully-0.001
ous-0.001
ting-0.001
izing-0.001
Token were
Token ID547
Feature activation+1.90e-02

Pos logit contributions
house+0.026
packs+0.024
tr+0.023
 fort+0.023
ringe+0.022
Neg logit contributions
ter-0.028
fully-0.025
ous-0.024
ting-0.024
 Dolphin-0.023
Token would
Token ID561
Feature activation-1.02e-02

Pos logit contributions
house+0.128
packs+0.118
 fort+0.114
tr+0.113
ringe+0.112
Neg logit contributions
ter-0.118
fully-0.103
ous-0.098
ting-0.094
ancer-0.094
Token find
Token ID1064
Feature activation-4.07e-02

Pos logit contributions
house+0.017
packs+0.016
tr+0.015
oc+0.015
ringe+0.015
Neg logit contributions
ter-0.019
fully-0.016
ous-0.016
 Dolphin-0.015
 Hop-0.015
Token a
Token ID257
Feature activation-2.24e-02

Pos logit contributions
house+0.089
tr+0.081
oc+0.081
packs+0.078
 fort+0.076
Neg logit contributions
ter-0.056
fully-0.047
ting-0.044
 Eddie-0.042
 Joey-0.042
Token new
Token ID649
Feature activation+1.85e-03

Pos logit contributions
house+0.046
packs+0.045
oc+0.043
ringe+0.042
tr+0.040
Neg logit contributions
ter-0.029
 Dolphin-0.024
 Hop-0.023
ous-0.023
 mer-0.023
Token doll
Token ID3654
Feature activation-1.48e-01

Pos logit contributions
ter+0.004
fully+0.003
 Hop+0.003
ous+0.003
 mer+0.003
Neg logit contributions
house-0.002
ringe-0.002
oc-0.002
 handcuffs-0.002
packs-0.002
Tokenhouse
Token ID4803
Feature activation-3.07e-01

Pos logit contributions
house+0.284
 handcuffs+0.283
packs+0.281
ringe+0.280
oc+0.276
Neg logit contributions
ter-0.217
fully-0.175
ous-0.161
ting-0.160
ancer-0.148
Token and
Token ID290
Feature activation+5.04e-03

Pos logit contributions
house+0.631
ringe+0.591
oc+0.580
packs+0.577
tr+0.568
Neg logit contributions
ter-0.444
fully-0.385
ous-0.338
ting-0.326
ing-0.303
Token that
Token ID326
Feature activation-4.82e-02

Pos logit contributions
ter+0.008
fully+0.007
 Hop+0.007
ous+0.007
 hello+0.006
Neg logit contributions
house-0.009
oc-0.009
ringe-0.008
packs-0.008
ris-0.008
Token they
Token ID484
Feature activation-6.92e-02

Pos logit contributions
ringe+0.099
oc+0.095
house+0.095
ris+0.095
 handcuffs+0.092
Neg logit contributions
ter-0.064
 Joey-0.050
fully-0.048
 Hop-0.048
ous-0.048
Token would
Token ID561
Feature activation-1.03e-02

Pos logit contributions
house+0.137
packs+0.133
 fort+0.128
ringe+0.127
tr+0.126
Neg logit contributions
ter-0.100
fully-0.084
ous-0.078
 Hop-0.077
ancer-0.073
Token never
Token ID1239
Feature activation-9.76e-02

Pos logit contributions
house+0.018
packs+0.018
oc+0.017
ringe+0.016
tr+0.016
Neg logit contributions
ter-0.017
fully-0.015
ous-0.014
 Hop-0.014
 Eddie-0.013
Token still
Token ID991
Feature activation-3.69e-02

Pos logit contributions
ter+0.034
fully+0.029
ous+0.029
 Dolphin+0.028
ting+0.028
Neg logit contributions
house-0.030
packs-0.028
tr-0.026
 fort-0.026
ringe-0.025
Token play
Token ID711
Feature activation-6.60e-02

Pos logit contributions
house+0.060
packs+0.056
tr+0.053
 fort+0.052
ringe+0.051
Neg logit contributions
ter-0.069
fully-0.060
ous-0.059
 Dolphin-0.056
ting-0.056
Token with
Token ID351
Feature activation-7.77e-02

Pos logit contributions
house+0.128
packs+0.118
tr+0.114
 fort+0.111
ringe+0.109
Neg logit contributions
ter-0.106
fully-0.090
ous-0.086
ting-0.086
 Dolphin-0.084
Token the
Token ID262
Feature activation+4.21e-02

Pos logit contributions
house+0.166
packs+0.152
 fort+0.145
tr+0.143
oc+0.141
Neg logit contributions
ter-0.111
fully-0.093
ous-0.092
 Dolphin-0.088
ting-0.087
Token doll
Token ID3654
Feature activation-1.44e-01

Pos logit contributions
ter+0.087
fully+0.077
ous+0.075
ting+0.074
 Dolphin+0.074
Neg logit contributions
house-0.063
packs-0.055
tr-0.052
 fort-0.051
oc-0.050
Tokenhouse
Token ID4803
Feature activation-3.06e-01

Pos logit contributions
house+0.263
 handcuffs+0.260
packs+0.257
ringe+0.253
tr+0.250
Neg logit contributions
ter-0.222
fully-0.182
ous-0.169
ting-0.167
ancer-0.157
Token with
Token ID351
Feature activation-7.91e-02

Pos logit contributions
house+0.591
packs+0.538
tr+0.534
ringe+0.531
oc+0.525
Neg logit contributions
ter-0.485
fully-0.418
ous-0.382
ting-0.374
izing-0.361
Token Lily
Token ID20037
Feature activation-1.30e-01

Pos logit contributions
house+0.182
packs+0.167
tr+0.163
 fort+0.161
oc+0.158
Neg logit contributions
ter-0.099
fully-0.080
ous-0.078
ting-0.075
 Dolphin-0.075
Token.
Token ID13
Feature activation-3.66e-03

Pos logit contributions
house+0.239
packs+0.214
tr+0.206
 fort+0.203
oc+0.198
Neg logit contributions
ter-0.222
fully-0.191
ous-0.188
ting-0.183
 Dolphin-0.182
Token She
Token ID1375
Feature activation-4.03e-02

Pos logit contributions
house+0.007
packs+0.006
 fort+0.006
tr+0.006
ringe+0.006
Neg logit contributions
ter-0.006
fully-0.005
ous-0.005
 Dolphin-0.005
ting-0.005
Token will
Token ID481
Feature activation-5.55e-02

Pos logit contributions
house+0.063
packs+0.060
tr+0.055
 fort+0.055
ringe+0.055
Neg logit contributions
ter-0.074
 Hop-0.062
ous-0.062
fully-0.061
ancer-0.060
Token and
Token ID290
Feature activation+1.79e-03

Pos logit contributions
house+0.114
packs+0.102
tr+0.098
 fort+0.098
oc+0.095
Neg logit contributions
ter-0.103
 Dolphin-0.085
ous-0.083
fully-0.083
ting-0.081
Token made
Token ID925
Feature activation-1.29e-01

Pos logit contributions
ter+0.004
fully+0.003
ous+0.003
ting+0.003
 Dolphin+0.003
Neg logit contributions
house-0.003
packs-0.002
tr-0.002
 fort-0.002
oc-0.002
Token a
Token ID257
Feature activation-2.40e-02

Pos logit contributions
house+0.266
packs+0.234
 fort+0.233
tr+0.225
oc+0.220
Neg logit contributions
ter-0.193
ous-0.162
 Dolphin-0.159
ancer-0.158
fully-0.157
Token beautiful
Token ID4950
Feature activation+3.44e-02

Pos logit contributions
house+0.043
packs+0.043
ringe+0.039
tr+0.038
 handcuffs+0.038
Neg logit contributions
ter-0.038
ous-0.032
fully-0.032
izing-0.031
 Dolphin-0.030
Token bird
Token ID6512
Feature activation+5.65e-02

Pos logit contributions
ter+0.071
fully+0.063
ous+0.062
 Dolphin+0.060
ting+0.060
Neg logit contributions
house-0.050
packs-0.045
tr-0.042
ringe-0.041
 fort-0.040
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
ter+0.093
fully+0.079
ous+0.078
ting+0.072
izing+0.070
Neg logit contributions
packs-0.098
house-0.098
ringe-0.093
tr-0.093
 handcuffs-0.093
Token with
Token ID351
Feature activation-8.77e-02

Pos logit contributions
house+0.593
packs+0.558
ringe+0.552
oc+0.541
tr+0.534
Neg logit contributions
ter-0.464
fully-0.401
ous-0.384
izing-0.354
ting-0.353
Token lots
Token ID6041
Feature activation-1.30e-01

Pos logit contributions
house+0.183
packs+0.166
tr+0.164
 fort+0.160
oc+0.157
Neg logit contributions
ter-0.130
fully-0.109
ous-0.105
ting-0.102
 Dolphin-0.100
Token of
Token ID286
Feature activation+3.01e-02

Pos logit contributions
house+0.234
packs+0.206
 fort+0.199
oc+0.198
tr+0.192
Neg logit contributions
ter-0.234
 Dolphin-0.200
fully-0.194
ancer-0.192
ous-0.190
Token seeds
Token ID11904
Feature activation+9.04e-02

Pos logit contributions
ter+0.058
fully+0.051
ous+0.050
 Dolphin+0.050
ting+0.049
Neg logit contributions
house-0.049
packs-0.043
tr-0.041
 fort-0.041
oc-0.040
Token and
Token ID290
Feature activation+2.19e-03

Pos logit contributions
ter+0.157
fully+0.135
ous+0.132
 Dolphin+0.130
ting+0.129
Neg logit contributions
house-0.165
packs-0.148
tr-0.142
 fort-0.140
oc-0.136
Token shot
Token ID2823
Feature activation-6.64e-03

Pos logit contributions
ter+0.072
fully+0.063
 Dolphin+0.062
ous+0.062
ancer+0.061
Neg logit contributions
house-0.062
packs-0.053
tr-0.052
 fort-0.051
oc-0.049
Token out
Token ID503
Feature activation-3.82e-03

Pos logit contributions
house+0.012
packs+0.011
tr+0.010
 fort+0.010
 handcuffs+0.010
Neg logit contributions
ter-0.011
ancer-0.010
 Dolphin-0.010
ya-0.009
ous-0.009
Token of
Token ID286
Feature activation+3.22e-02

Pos logit contributions
house+0.008
packs+0.007
tr+0.007
 fort+0.007
oc+0.006
Neg logit contributions
ter-0.006
fully-0.005
ous-0.005
 Dolphin-0.005
ting-0.005
Token the
Token ID262
Feature activation+4.38e-02

Pos logit contributions
ter+0.041
fully+0.033
ting+0.031
ous+0.031
 Dolphin+0.030
Neg logit contributions
house-0.074
packs-0.068
tr-0.067
oc-0.065
 fort-0.065
Token tree
Token ID5509
Feature activation-3.00e-02

Pos logit contributions
ter+0.084
ting+0.070
fully+0.070
ous+0.070
 Dolphin+0.068
Neg logit contributions
house-0.069
packs-0.065
tr-0.059
ringe-0.058
oc-0.058
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
house+0.059
packs+0.055
tr+0.053
ringe+0.052
 fort+0.052
Neg logit contributions
ter-0.045
fully-0.039
ous-0.038
ting-0.036
izing-0.035
Token window
Token ID4324
Feature activation+8.91e-03

Pos logit contributions
house+0.596
packs+0.541
tr+0.524
 fort+0.514
ringe+0.507
Neg logit contributions
ter-0.481
fully-0.410
ous-0.400
ting-0.388
 Dolphin-0.382
Token and
Token ID290
Feature activation+9.85e-03

Pos logit contributions
ter+0.015
fully+0.012
 Dolphin+0.012
ous+0.012
ting+0.012
Neg logit contributions
house-0.017
packs-0.015
tr-0.015
 fort-0.015
oc-0.014
Token flew
Token ID13112
Feature activation+2.22e-02

Pos logit contributions
ter+0.017
fully+0.014
ous+0.014
ting+0.014
 Dolphin+0.014
Neg logit contributions
house-0.018
packs-0.017
tr-0.016
 fort-0.016
oc-0.015
Token into
Token ID656
Feature activation+1.92e-02

Pos logit contributions
ter+0.046
ous+0.041
ancer+0.041
 Dolphin+0.040
fully+0.040
Neg logit contributions
house-0.033
packs-0.028
 fort-0.027
tr-0.026
 handcuffs-0.026
Token the
Token ID262
Feature activation+3.80e-02

Pos logit contributions
ter+0.020
fully+0.016
ting+0.015
 Dolphin+0.014
ous+0.014
Neg logit contributions
house-0.048
tr-0.045
packs-0.044
ringe-0.043
 fort-0.043
Token dolls
Token ID36062
Feature activation-5.93e-02

Pos logit contributions
ter+0.071
fully+0.062
ous+0.061
 Dolphin+0.060
ting+0.060
Neg logit contributions
house-0.040
packs-0.034
tr-0.032
oc-0.031
 fort-0.030
Token back
Token ID736
Feature activation-7.40e-02

Pos logit contributions
house+0.111
tr+0.098
packs+0.098
oc+0.097
 fort+0.095
Neg logit contributions
ter-0.100
fully-0.085
ting-0.081
ous-0.080
 Dolphin-0.078
Token in
Token ID287
Feature activation+9.71e-02

Pos logit contributions
house+0.154
packs+0.139
tr+0.136
 fort+0.135
oc+0.134
Neg logit contributions
ter-0.108
fully-0.091
ous-0.087
ting-0.085
 Dolphin-0.082
Token the
Token ID262
Feature activation+3.19e-02

Pos logit contributions
ter+0.051
 Nathan+0.040
 Eddie+0.039
 Henry+0.037
 young+0.037
Neg logit contributions
tr-0.270
house-0.264
ringe-0.246
oc-0.246
atters-0.241
Token doll
Token ID3654
Feature activation-1.45e-01

Pos logit contributions
ter+0.062
 Dolphin+0.053
 hello+0.051
 Eddie+0.051
ting+0.050
Neg logit contributions
house-0.045
oc-0.042
packs-0.041
ris-0.041
ringe-0.038
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
 handcuffs+0.265
ringe+0.254
 choice+0.253
omo+0.250
oc+0.249
Neg logit contributions
ter-0.197
fully-0.163
ting-0.162
ly-0.152
ous-0.151
Token.
Token ID13
Feature activation-2.77e-03

Pos logit contributions
house+0.658
ringe+0.610
oc+0.607
tr+0.596
packs+0.586
Neg logit contributions
ter-0.398
fully-0.347
ous-0.300
ting-0.292
ing-0.275
Token They
Token ID1119
Feature activation-9.01e-02

Pos logit contributions
house+0.005
packs+0.004
oc+0.004
ringe+0.004
tr+0.004
Neg logit contributions
ter-0.005
fully-0.004
 Dolphin-0.004
 Hop-0.004
ous-0.004
Token hugged
Token ID41130
Feature activation+1.67e-02

Pos logit contributions
house+0.193
oc+0.177
ringe+0.175
packs+0.174
tr+0.174
Neg logit contributions
ter-0.122
fully-0.095
 Nathan-0.092
 Eddie-0.092
 Eli-0.089
Token each
Token ID1123
Feature activation+1.57e-02

Pos logit contributions
ter+0.030
ous+0.028
ancer+0.026
 Dolphin+0.026
izing+0.026
Neg logit contributions
house-0.029
packs-0.026
 fort-0.024
oc-0.024
tr-0.024
Token other
Token ID584
Feature activation-1.67e-02

Pos logit contributions
ter+0.022
fully+0.020
ting+0.019
 Dolphin+0.018
ous+0.018
Neg logit contributions
house-0.033
tr-0.029
packs-0.029
 fort-0.028
ringe-0.028
Token they
Token ID484
Feature activation-7.00e-02

Pos logit contributions
house+0.125
packs+0.113
tr+0.108
 fort+0.106
oc+0.104
Neg logit contributions
ter-0.108
fully-0.092
ous-0.091
 Dolphin-0.088
ting-0.088
Token could
Token ID714
Feature activation+5.55e-02

Pos logit contributions
house+0.125
packs+0.114
tr+0.110
 fort+0.108
ringe+0.105
Neg logit contributions
ter-0.122
fully-0.105
ting-0.102
ous-0.101
 Dolphin-0.098
Token repair
Token ID9185
Feature activation+8.03e-02

Pos logit contributions
ter+0.090
fully+0.075
ble+0.075
 Hop+0.072
 Eddie+0.071
Neg logit contributions
house-0.097
packs-0.096
tr-0.092
ringe-0.092
oc-0.091
Token their
Token ID511
Feature activation-1.04e-01

Pos logit contributions
ter+0.127
fully+0.106
ting+0.103
 Dolphin+0.099
izing+0.099
Neg logit contributions
house-0.157
tr-0.142
packs-0.141
 fort-0.137
oc-0.133
Token doll
Token ID3654
Feature activation-1.44e-01

Pos logit contributions
house+0.157
tr+0.135
packs+0.131
ringe+0.126
oc+0.125
Neg logit contributions
ter-0.203
fully-0.173
 Dolphin-0.169
ous-0.169
izing-0.167
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
 handcuffs+0.118
ringe+0.114
packs+0.113
ris+0.112
omo+0.112
Neg logit contributions
ter-0.343
fully-0.304
ting-0.291
ous-0.287
izing-0.281
Token and
Token ID290
Feature activation+2.74e-03

Pos logit contributions
house+0.598
ringe+0.569
oc+0.557
packs+0.549
tr+0.547
Neg logit contributions
ter-0.440
fully-0.378
ous-0.326
ting-0.320
izing-0.314
Token make
Token ID787
Feature activation-2.84e-02

Pos logit contributions
ter+0.004
fully+0.004
ous+0.004
 Hop+0.003
 Joey+0.003
Neg logit contributions
house-0.005
oc-0.005
packs-0.004
ris-0.004
ringe-0.004
Token it
Token ID340
Feature activation+9.85e-03

Pos logit contributions
house+0.062
packs+0.057
tr+0.056
 fort+0.054
oc+0.054
Neg logit contributions
ter-0.039
fully-0.032
ous-0.030
ting-0.030
 Dolphin-0.029
Token look
Token ID804
Feature activation+2.08e-02

Pos logit contributions
ter+0.015
fully+0.013
ous+0.013
ble+0.012
izing+0.011
Neg logit contributions
house-0.018
oc-0.017
packs-0.017
ringe-0.017
ris-0.017
Token original
Token ID2656
Feature activation-3.91e-02

Pos logit contributions
ter+0.032
fully+0.028
ous+0.025
izing+0.025
ting+0.025
Neg logit contributions
house-0.041
packs-0.038
tr-0.037
ringe-0.036
oc-0.036
Token fell
Token ID3214
Feature activation-6.82e-02

Pos logit contributions
packs+0.057
house+0.054
ringe+0.051
 fort+0.049
 handcuffs+0.048
Neg logit contributions
ter-0.060
fully-0.052
ous-0.050
ting-0.048
izing-0.048
Token on
Token ID319
Feature activation-1.14e-01

Pos logit contributions
house+0.129
packs+0.115
tr+0.114
 fort+0.111
ringe+0.106
Neg logit contributions
ter-0.112
 Dolphin-0.096
ous-0.094
fully-0.093
ya-0.093
Token Tim
Token ID5045
Feature activation-7.63e-02

Pos logit contributions
house+0.246
packs+0.222
tr+0.216
 fort+0.214
oc+0.210
Neg logit contributions
ter-0.160
fully-0.133
ous-0.131
 Dolphin-0.126
ting-0.125
Token's
Token ID338
Feature activation-9.06e-03

Pos logit contributions
house+0.139
packs+0.123
tr+0.120
 fort+0.117
oc+0.115
Neg logit contributions
ter-0.133
fully-0.114
ous-0.113
 Dolphin-0.111
ting-0.109
Token play
Token ID711
Feature activation-7.46e-02

Pos logit contributions
house+0.015
packs+0.013
tr+0.012
oc+0.012
 fort+0.012
Neg logit contributions
ter-0.017
fully-0.015
ous-0.015
ting-0.014
 Dolphin-0.014
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
packs+0.135
house+0.134
ringe+0.125
tr+0.122
oc+0.120
Neg logit contributions
ter-0.124
fully-0.109
ting-0.100
ous-0.098
izing-0.094
Token and
Token ID290
Feature activation+2.35e-03

Pos logit contributions
house+0.624
packs+0.616
ringe+0.612
oc+0.585
tr+0.578
Neg logit contributions
ter-0.413
fully-0.365
ous-0.320
ting-0.318
izing-0.314
Token broke
Token ID6265
Feature activation-1.13e-01

Pos logit contributions
ter+0.004
fully+0.003
 Joey+0.003
ting+0.003
ous+0.003
Neg logit contributions
packs-0.005
house-0.005
ris-0.005
ringe-0.004
tr-0.004
Token it
Token ID340
Feature activation+8.90e-03

Pos logit contributions
house+0.244
packs+0.223
tr+0.217
 fort+0.213
oc+0.209
Neg logit contributions
ter-0.160
fully-0.133
ous-0.129
ting-0.125
 Dolphin-0.124
Token.
Token ID13
Feature activation-1.28e-02

Pos logit contributions
ter+0.013
fully+0.011
ous+0.010
izing+0.010
ting+0.010
Neg logit contributions
house-0.018
packs-0.018
ringe-0.017
oc-0.017
 handcuffs-0.017
Token Tim
Token ID5045
Feature activation-7.74e-02

Pos logit contributions
house+0.023
packs+0.020
tr+0.020
 fort+0.019
oc+0.019
Neg logit contributions
ter-0.023
fully-0.020
ous-0.019
ting-0.019
 Dolphin-0.019
Token He
Token ID679
Feature activation-2.62e-02

Pos logit contributions
house+0.011
packs+0.010
tr+0.010
 fort+0.010
ringe+0.009
Neg logit contributions
ter-0.008
fully-0.007
 Dolphin-0.007
ous-0.006
 Hop-0.006
Token ran
Token ID4966
Feature activation-4.80e-02

Pos logit contributions
house+0.047
packs+0.044
tr+0.042
ringe+0.041
 fort+0.041
Neg logit contributions
ter-0.045
fully-0.039
ya-0.037
ous-0.036
 Dolphin-0.036
Token to
Token ID284
Feature activation-2.75e-02

Pos logit contributions
house+0.106
packs+0.097
tr+0.095
oc+0.092
 fort+0.091
Neg logit contributions
ter-0.066
fully-0.057
ting-0.054
 Dolphin-0.050
ous-0.049
Token the
Token ID262
Feature activation+3.51e-02

Pos logit contributions
house+0.057
packs+0.051
tr+0.050
 fort+0.049
oc+0.048
Neg logit contributions
ter-0.042
fully-0.035
ous-0.034
ting-0.034
 Dolphin-0.034
Token play
Token ID711
Feature activation-6.86e-02

Pos logit contributions
ter+0.073
fully+0.066
ous+0.065
ting+0.063
 Dolphin+0.062
Neg logit contributions
house-0.052
packs-0.045
tr-0.043
 fort-0.043
oc-0.040
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
house+0.124
packs+0.113
tr+0.109
 fort+0.105
ringe+0.104
Neg logit contributions
ter-0.120
fully-0.104
ting-0.099
ous-0.099
 Dolphin-0.098
Token and
Token ID290
Feature activation+5.89e-03

Pos logit contributions
house+0.646
ringe+0.623
packs+0.612
tr+0.610
 handcuffs+0.593
Neg logit contributions
ter-0.400
fully-0.348
ting-0.299
ous-0.290
izing-0.277
Token jumped
Token ID11687
Feature activation+5.24e-02

Pos logit contributions
ter+0.011
fully+0.009
 Hop+0.009
ting+0.009
 Joey+0.009
Neg logit contributions
house-0.010
packs-0.009
tr-0.009
ringe-0.008
 fort-0.008
Token on
Token ID319
Feature activation-1.12e-01

Pos logit contributions
ter+0.089
ous+0.078
fully+0.076
 Dolphin+0.074
ancer+0.074
Neg logit contributions
house-0.097
packs-0.086
 fort-0.084
tr-0.083
oc-0.079
Token the
Token ID262
Feature activation+3.38e-02

Pos logit contributions
house+0.258
packs+0.239
tr+0.232
 fort+0.226
ringe+0.222
Neg logit contributions
ter-0.142
fully-0.115
ous-0.110
ting-0.109
 Dolphin-0.108
Token table
Token ID3084
Feature activation-7.43e-02

Pos logit contributions
ter+0.075
fully+0.067
ous+0.067
 Dolphin+0.066
ancer+0.066
Neg logit contributions
house-0.045
 fort-0.040
tr-0.038
packs-0.036
oc-0.033
Token together
Token ID1978
Feature activation-7.72e-02

Pos logit contributions
ter+0.014
fully+0.012
ous+0.011
ting+0.011
 Dolphin+0.011
Neg logit contributions
house-0.019
packs-0.018
tr-0.017
ringe-0.017
 fort-0.017
Token in
Token ID287
Feature activation+9.41e-02

Pos logit contributions
house+0.162
packs+0.149
tr+0.146
ringe+0.142
 fort+0.141
Neg logit contributions
ter-0.111
fully-0.095
ous-0.089
ting-0.089
 Dolphin-0.086
Token their
Token ID511
Feature activation-1.07e-01

Pos logit contributions
ter+0.028
 young+0.012
 Hop+0.009
 Fly+0.004
 the+0.004
Neg logit contributions
tr-0.283
ringe-0.272
oc-0.267
house-0.265
 bisc-0.262
Token old
Token ID1468
Feature activation+6.66e-02

Pos logit contributions
ris+0.185
ringe+0.185
house+0.185
 handcuffs+0.185
tr+0.184
Neg logit contributions
ter-0.130
 young-0.123
 Hop-0.118
ya-0.112
 Fly-0.108
Token tree
Token ID5509
Feature activation-3.43e-02

Pos logit contributions
ter+0.120
 Hop+0.104
 Eddie+0.100
ya+0.100
fully+0.100
Neg logit contributions
house-0.094
tr-0.093
ris-0.090
atters-0.086
 handcuffs-0.086
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
 handcuffs+0.047
ringe+0.045
 receipt+0.041
 bisc+0.040
tr+0.040
Neg logit contributions
ter-0.059
fully-0.058
ing-0.054
ous-0.052
 by-0.051
Token.
Token ID13
Feature activation-1.07e-02

Pos logit contributions
house+0.591
ringe+0.590
 handcuffs+0.574
packs+0.558
tr+0.558
Neg logit contributions
ter-0.417
fully-0.373
ting-0.326
ing-0.314
ous-0.308
Token<|endoftext|>
Token ID50256
Feature activation+3.53e-02

Pos logit contributions
house+0.024
packs+0.024
ringe+0.022
 handcuffs+0.022
 fort+0.022
Neg logit contributions
ter-0.012
 Dolphin-0.010
ous-0.010
fully-0.010
 Hop-0.010
TokenOnce
Token ID7454
Feature activation-9.27e-02

Pos logit contributions
ter+0.064
fully+0.055
ous+0.055
izing+0.053
ting+0.053
Neg logit contributions
house-0.063
packs-0.056
tr-0.054
 fort-0.053
oc-0.052
Token upon
Token ID2402
Feature activation+1.77e-01

Pos logit contributions
house+0.224
packs+0.208
tr+0.200
 fort+0.199
oc+0.194
Neg logit contributions
ter-0.106
fully-0.082
ous-0.080
ting-0.079
 Dolphin-0.077
Token a
Token ID257
Feature activation-2.08e-02

Pos logit contributions
ter+0.320
ous+0.284
ting+0.277
izing+0.274
 Dolphin+0.273
Neg logit contributions
house-0.320
packs-0.277
tr-0.266
 fort-0.264
oc-0.253
Token a
Token ID257
Feature activation-2.51e-02

Pos logit contributions
house+0.085
oc+0.082
packs+0.079
ringe+0.076
 fort+0.075
Neg logit contributions
ter-0.063
fully-0.052
ous-0.048
ting-0.047
 Hop-0.045
Token window
Token ID4324
Feature activation+7.30e-03

Pos logit contributions
house+0.041
packs+0.039
ringe+0.037
oc+0.036
tr+0.034
Neg logit contributions
ter-0.044
 Dolphin-0.037
ous-0.036
 young-0.036
 Hop-0.036
Token in
Token ID287
Feature activation+9.81e-02

Pos logit contributions
ter+0.010
fully+0.008
ous+0.008
ting+0.008
ing+0.007
Neg logit contributions
house-0.015
packs-0.014
ringe-0.014
oc-0.014
 fort-0.014
Token the
Token ID262
Feature activation+3.07e-02

Pos logit contributions
ter+0.071
fully+0.041
 Henry+0.041
 Fly+0.040
 Simon+0.039
Neg logit contributions
house-0.265
tr-0.263
oc-0.253
 fort-0.245
ringe-0.242
Token play
Token ID711
Feature activation-6.84e-02

Pos logit contributions
ter+0.058
fully+0.048
ting+0.048
ous+0.046
 hello+0.046
Neg logit contributions
house-0.044
packs-0.041
oc-0.039
ringe-0.039
ris-0.038
Tokenhouse
Token ID4803
Feature activation-3.04e-01

Pos logit contributions
packs+0.096
house+0.095
ringe+0.095
 handcuffs+0.093
tr+0.090
Neg logit contributions
ter-0.135
fully-0.119
ting-0.110
ous-0.108
izing-0.105
Token.
Token ID13
Feature activation-1.44e-02

Pos logit contributions
house+0.612
ringe+0.588
oc+0.573
packs+0.572
tr+0.569
Neg logit contributions
ter-0.431
fully-0.361
ting-0.327
ous-0.317
izing-0.307
Token She
Token ID1375
Feature activation-4.07e-02

Pos logit contributions
house+0.023
packs+0.020
tr+0.019
 fort+0.018
oc+0.018
Neg logit contributions
ter-0.030
fully-0.026
ous-0.025
ting-0.025
izing-0.025
Token looks
Token ID3073
Feature activation+2.30e-02

Pos logit contributions
house+0.068
packs+0.061
ringe+0.059
tr+0.059
oc+0.057
Neg logit contributions
ter-0.071
ous-0.060
 Fly-0.059
ting-0.059
fully-0.059
Token for
Token ID329
Feature activation-5.99e-02

Pos logit contributions
ter+0.037
fully+0.031
ous+0.031
 Dolphin+0.030
ting+0.030
Neg logit contributions
house-0.045
packs-0.041
tr-0.039
 fort-0.039
oc-0.038
Token the
Token ID262
Feature activation+3.04e-02

Pos logit contributions
house+0.128
packs+0.117
tr+0.114
 fort+0.111
oc+0.110
Neg logit contributions
ter-0.086
fully-0.072
ous-0.070
ting-0.068
 Dolphin-0.067
Token of
Token ID286
Feature activation+3.04e-02

Pos logit contributions
house+0.161
packs+0.145
tr+0.140
 fort+0.138
oc+0.135
Neg logit contributions
ter-0.134
fully-0.114
ous-0.112
ting-0.109
 Dolphin-0.109
Token water
Token ID1660
Feature activation+3.71e-02

Pos logit contributions
ter+0.053
fully+0.046
ous+0.046
ting+0.044
 Dolphin+0.044
Neg logit contributions
house-0.055
packs-0.050
 fort-0.047
tr-0.047
 handcuffs-0.046
Token in
Token ID287
Feature activation+9.57e-02

Pos logit contributions
ter+0.060
fully+0.053
ous+0.053
 Dolphin+0.052
izing+0.051
Neg logit contributions
house-0.069
packs-0.062
 fort-0.059
tr-0.059
 handcuffs-0.055
Token her
Token ID607
Feature activation-3.73e-02

Pos logit contributions
ter+0.122
fully+0.098
ting+0.090
 Dolphin+0.089
 Joey+0.089
Neg logit contributions
house-0.221
tr-0.207
packs-0.198
 fort-0.197
ringe-0.194
Token bird
Token ID6512
Feature activation+5.73e-02

Pos logit contributions
house+0.051
tr+0.045
packs+0.044
ris+0.043
oc+0.042
Neg logit contributions
ter-0.077
 Dolphin-0.067
fully-0.066
 Hop-0.065
ous-0.064
Tokenhouse
Token ID4803
Feature activation-3.03e-01

Pos logit contributions
ter+0.145
fully+0.129
ous+0.127
ting+0.125
 Dolphin+0.124
Neg logit contributions
house-0.053
packs-0.045
tr-0.045
 fort-0.042
 handcuffs-0.041
Token.
Token ID13
Feature activation-5.41e-03

Pos logit contributions
house+0.586
packs+0.531
tr+0.517
 fort+0.510
oc+0.507
Neg logit contributions
ter-0.484
fully-0.413
ous-0.395
ting-0.385
 Dolphin-0.375
Token "
Token ID366
Feature activation+1.48e-02

Pos logit contributions
house+0.010
packs+0.009
tr+0.008
 fort+0.008
oc+0.008
Neg logit contributions
ter-0.010
fully-0.008
ous-0.008
ting-0.008
 Dolphin-0.008
TokenI
Token ID40
Feature activation+8.57e-03

Pos logit contributions
ter+0.024
fully+0.020
ous+0.020
ting+0.019
 Dolphin+0.019
Neg logit contributions
house-0.029
packs-0.027
tr-0.025
 fort-0.025
oc-0.025
Token can
Token ID460
Feature activation+1.58e-02

Pos logit contributions
ter+0.013
 Joey+0.011
 Hop+0.011
ous+0.011
 Fly+0.011
Neg logit contributions
packs-0.015
house-0.015
ringe-0.014
tr-0.014
 handcuffs-0.014
Token help
Token ID1037
Feature activation+6.02e-02

Pos logit contributions
ter+0.030
fully+0.026
ous+0.025
ting+0.025
 Dolphin+0.025
Neg logit contributions
house-0.027
packs-0.022
tr-0.022
 fort-0.022
oc-0.022
Token She
Token ID1375
Feature activation-4.46e-02

Pos logit contributions
house+0.016
packs+0.014
tr+0.014
 fort+0.013
oc+0.013
Neg logit contributions
ter-0.020
fully-0.018
ous-0.018
izing-0.017
 Dolphin-0.017
Token had
Token ID550
Feature activation-7.80e-02

Pos logit contributions
house+0.077
packs+0.070
 handcuffs+0.067
ringe+0.067
 fort+0.067
Neg logit contributions
ter-0.078
fully-0.067
 Dolphin-0.065
ya-0.064
ous-0.064
Token a
Token ID257
Feature activation-1.99e-02

Pos logit contributions
house+0.179
oc+0.170
packs+0.162
tr+0.159
 fort+0.158
Neg logit contributions
ter-0.094
fully-0.077
ous-0.071
ting-0.071
 Dolphin-0.070
Token big
Token ID1263
Feature activation-1.17e-01

Pos logit contributions
house+0.036
packs+0.035
oc+0.035
ringe+0.033
 handcuffs+0.032
Neg logit contributions
ter-0.029
 Dolphin-0.026
 young-0.026
 dolphin-0.026
 Hop-0.025
Token doll
Token ID3654
Feature activation-1.45e-01

Pos logit contributions
house+0.164
packs+0.158
ris+0.147
oc+0.145
tr+0.145
Neg logit contributions
ter-0.228
 Dolphin-0.199
ya-0.192
 Hop-0.189
ous-0.188
Tokenhouse
Token ID4803
Feature activation-3.03e-01

Pos logit contributions
packs+0.336
 handcuffs+0.332
ushes+0.315
ws+0.313
ringe+0.310
Neg logit contributions
ter-0.162
ting-0.112
ous-0.112
fully-0.100
 Hop-0.093
Token with
Token ID351
Feature activation-8.26e-02

Pos logit contributions
oc+0.682
ringe+0.677
house+0.663
packs+0.655
 handcuffs+0.648
Neg logit contributions
ter-0.353
fully-0.295
ous-0.252
ing-0.223
ting-0.215
Token many
Token ID867
Feature activation-1.42e-01

Pos logit contributions
house+0.168
packs+0.152
 fort+0.148
oc+0.146
tr+0.145
Neg logit contributions
ter-0.126
fully-0.106
ting-0.104
ous-0.104
 Dolphin-0.098
Token rooms
Token ID9519
Feature activation-9.87e-02

Pos logit contributions
house+0.241
tr+0.223
packs+0.213
ringe+0.208
oc+0.205
Neg logit contributions
ter-0.241
fully-0.216
ous-0.208
izing-0.208
ting-0.207
Token and
Token ID290
Feature activation+7.33e-03

Pos logit contributions
house+0.203
oc+0.186
ringe+0.184
tr+0.184
 fort+0.181
Neg logit contributions
ter-0.139
fully-0.121
ous-0.119
ting-0.115
ing-0.108
Token furniture
Token ID13091
Feature activation-4.28e-02

Pos logit contributions
ter+0.012
fully+0.011
ous+0.010
ting+0.010
ly+0.010
Neg logit contributions
house-0.013
oc-0.012
packs-0.012
ringe-0.012
tr-0.012
Token and
Token ID290
Feature activation+8.06e-03

Pos logit contributions
ter+0.070
fully+0.059
ting+0.056
ous+0.056
 Dolphin+0.055
Neg logit contributions
house-0.087
packs-0.080
tr-0.077
 fort-0.076
ringe-0.074
Token went
Token ID1816
Feature activation-9.83e-02

Pos logit contributions
ter+0.013
fully+0.011
ting+0.011
 Hop+0.010
 Eddie+0.010
Neg logit contributions
house-0.015
packs-0.014
ris-0.014
tr-0.014
 fort-0.014
Token to
Token ID284
Feature activation-2.99e-02

Pos logit contributions
house+0.215
oc+0.204
tr+0.204
packs+0.199
ringe+0.191
Neg logit contributions
ter-0.123
fully-0.109
ting-0.106
 Hop-0.100
 Dolphin-0.095
Token the
Token ID262
Feature activation+3.67e-02

Pos logit contributions
house+0.058
packs+0.052
tr+0.050
 fort+0.050
oc+0.049
Neg logit contributions
ter-0.049
fully-0.042
ous-0.041
ting-0.040
 Dolphin-0.039
Token tree
Token ID5509
Feature activation-2.84e-02

Pos logit contributions
ter+0.075
 Dolphin+0.064
ting+0.064
ous+0.062
fully+0.062
Neg logit contributions
house-0.054
packs-0.052
oc-0.046
ringe-0.044
tr-0.044
Tokenhouse
Token ID4803
Feature activation-3.03e-01

Pos logit contributions
 handcuffs+0.054
ringe+0.053
house+0.052
tr+0.052
packs+0.052
Neg logit contributions
ter-0.038
fully-0.034
ous-0.029
ting-0.029
ing-0.028
Token.
Token ID13
Feature activation-6.38e-03

Pos logit contributions
house+0.595
packs+0.548
ringe+0.546
oc+0.541
tr+0.536
Neg logit contributions
ter-0.461
fully-0.404
ting-0.361
ous-0.360
ing-0.335
Token Rose
Token ID8049
Feature activation-1.84e-02

Pos logit contributions
house+0.012
packs+0.011
tr+0.011
 fort+0.010
ringe+0.010
Neg logit contributions
ter-0.010
fully-0.009
ous-0.009
 Dolphin-0.009
ting-0.009
Token said
Token ID531
Feature activation-2.01e-02

Pos logit contributions
house+0.030
packs+0.027
tr+0.026
 fort+0.025
ringe+0.024
Neg logit contributions
ter-0.035
fully-0.031
ous-0.030
ting-0.030
 Dolphin-0.030
Token sorry
Token ID7926
Feature activation+2.79e-02

Pos logit contributions
house+0.028
 fort+0.022
tr+0.022
packs+0.021
 collection+0.020
Neg logit contributions
ter-0.041
ancer-0.038
izing-0.037
ting-0.036
 Dolphin-0.036
Token to
Token ID284
Feature activation-3.00e-02

Pos logit contributions
ter+0.049
 Dolphin+0.041
ting+0.040
izing+0.040
ous+0.040
Neg logit contributions
house-0.051
tr-0.045
packs-0.045
 fort-0.045
oc-0.041
Token,
Token ID11
Feature activation+2.61e-02

Pos logit contributions
house+0.155
packs+0.143
tr+0.140
 fort+0.136
ringe+0.134
Neg logit contributions
ter-0.103
fully-0.085
ous-0.082
ting-0.081
 Dolphin-0.080
Token he
Token ID339
Feature activation-2.01e-04

Pos logit contributions
ter+0.040
fully+0.034
ous+0.034
izing+0.033
 Dolphin+0.032
Neg logit contributions
house-0.053
packs-0.047
tr-0.045
 fort-0.045
oc-0.044
Token made
Token ID925
Feature activation-1.27e-01

Pos logit contributions
house+0.000
packs+0.000
tr+0.000
 fort+0.000
ringe+0.000
Neg logit contributions
ter-0.000
fully-0.000
ous-0.000
 Dolphin-0.000
ting-0.000
Token a
Token ID257
Feature activation-2.22e-02

Pos logit contributions
house+0.255
packs+0.228
 fort+0.222
tr+0.217
oc+0.213
Neg logit contributions
ter-0.199
ous-0.164
fully-0.164
ting-0.160
 Dolphin-0.159
Token tree
Token ID5509
Feature activation-3.07e-02

Pos logit contributions
house+0.039
packs+0.039
ringe+0.035
tr+0.034
 handcuffs+0.034
Neg logit contributions
ter-0.036
 Dolphin-0.031
ous-0.031
fully-0.030
ya-0.029
Tokenhouse
Token ID4803
Feature activation-3.03e-01

Pos logit contributions
packs+0.050
house+0.048
ringe+0.048
 handcuffs+0.047
tr+0.046
Neg logit contributions
ter-0.052
ous-0.044
fully-0.044
ing-0.041
ble-0.041
Token out
Token ID503
Feature activation-5.55e-03

Pos logit contributions
house+0.620
packs+0.582
ringe+0.569
oc+0.559
tr+0.558
Neg logit contributions
ter-0.435
fully-0.370
ous-0.356
ting-0.328
izing-0.323
Token of
Token ID286
Feature activation+3.34e-02

Pos logit contributions
house+0.011
packs+0.010
tr+0.010
 fort+0.010
ringe+0.010
Neg logit contributions
ter-0.008
fully-0.007
ous-0.007
ting-0.007
 Dolphin-0.007
Token tw
Token ID665
Feature activation+1.28e-01

Pos logit contributions
ter+0.050
fully+0.041
ting+0.039
 Joey+0.038
 Dolphin+0.038
Neg logit contributions
house-0.069
tr-0.063
packs-0.062
oc-0.060
ringe-0.059
Tokenigs
Token ID9235
Feature activation+2.11e-02

Pos logit contributions
ter+0.239
fully+0.209
ous+0.205
izing+0.204
 Dolphin+0.200
Neg logit contributions
house-0.207
tr-0.187
packs-0.184
 fort-0.175
 handcuffs-0.172
Token just
Token ID655
Feature activation+2.39e-02

Pos logit contributions
ter+0.036
 Dolphin+0.032
ancer+0.031
ting+0.030
fully+0.030
Neg logit contributions
house-0.038
packs-0.034
 fort-0.032
tr-0.032
oc-0.030
Token who
Token ID508
Feature activation+8.28e-02

Pos logit contributions
house+0.010
packs+0.010
 fort+0.010
 handcuffs+0.010
tr+0.009
Neg logit contributions
ter-0.020
fully-0.017
ting-0.017
ous-0.016
 Dolphin-0.016
Token lived
Token ID5615
Feature activation+4.32e-02

Pos logit contributions
ter+0.106
fully+0.085
 Hop+0.084
ous+0.082
 Dolphin+0.080
Neg logit contributions
house-0.169
packs-0.168
 fort-0.158
ringe-0.158
 handcuffs-0.155
Token in
Token ID287
Feature activation+1.00e-01

Pos logit contributions
ous+0.066
ter+0.062
ya+0.059
fully+0.058
izing+0.057
Neg logit contributions
house-0.088
packs-0.077
 fort-0.073
tr-0.072
ringe-0.072
Token a
Token ID257
Feature activation-1.75e-02

Pos logit contributions
ter+0.127
 Dolphin+0.098
fully+0.097
ting+0.095
 Hop+0.093
Neg logit contributions
house-0.226
packs-0.212
tr-0.207
 fort-0.202
oc-0.197
Token tree
Token ID5509
Feature activation-2.98e-02

Pos logit contributions
house+0.034
packs+0.034
tr+0.032
 handcuffs+0.032
ringe+0.032
Neg logit contributions
ter-0.024
 Dolphin-0.021
 Hop-0.020
 young-0.020
ya-0.019
Tokenhouse
Token ID4803
Feature activation-3.02e-01

Pos logit contributions
house+0.055
packs+0.051
tr+0.049
 fort+0.049
ringe+0.048
Neg logit contributions
ter-0.049
fully-0.041
ous-0.040
ting-0.039
izing-0.038
Token.
Token ID13
Feature activation-3.51e-03

Pos logit contributions
house+0.577
packs+0.540
ringe+0.517
tr+0.516
 fort+0.514
Neg logit contributions
ter-0.484
fully-0.405
ous-0.384
ting-0.382
izing-0.368
Token One
Token ID1881
Feature activation-1.62e-01

Pos logit contributions
house+0.006
packs+0.005
tr+0.005
 fort+0.005
oc+0.005
Neg logit contributions
ter-0.007
fully-0.006
ous-0.006
ting-0.006
izing-0.006
Token day
Token ID1110
Feature activation-6.14e-02

Pos logit contributions
house+0.243
packs+0.217
tr+0.204
 fort+0.200
ringe+0.198
Neg logit contributions
ter-0.333
fully-0.293
ous-0.288
 Dolphin-0.281
ting-0.280
Token he
Token ID339
Feature activation+3.20e-03

Pos logit contributions
house+0.093
packs+0.077
oc+0.073
 fort+0.070
tr+0.068
Neg logit contributions
ter-0.125
ping-0.112
fully-0.112
izing-0.112
ous-0.112
Token had
Token ID550
Feature activation-7.63e-02

Pos logit contributions
ter+0.005
fully+0.004
 Dolphin+0.004
 Hop+0.004
 Fly+0.004
Neg logit contributions
house-0.005
ringe-0.005
 handcuffs-0.005
packs-0.005
tr-0.005
Token's
Token ID338
Feature activation-3.97e-03

Pos logit contributions
house+0.064
packs+0.057
tr+0.055
 fort+0.054
oc+0.053
Neg logit contributions
ter-0.065
fully-0.057
ous-0.056
ting-0.054
 Dolphin-0.054
Token sleep
Token ID3993
Feature activation+4.97e-02

Pos logit contributions
house+0.006
packs+0.006
tr+0.005
oc+0.005
 fort+0.005
Neg logit contributions
ter-0.008
fully-0.007
ous-0.006
 Dolphin-0.006
ting-0.006
Token in
Token ID287
Feature activation+1.00e-01

Pos logit contributions
ter+0.080
fully+0.068
ous+0.065
ting+0.064
 Dolphin+0.063
Neg logit contributions
house-0.097
packs-0.088
tr-0.086
 fort-0.084
oc-0.082
Token the
Token ID262
Feature activation+3.60e-02

Pos logit contributions
ter+0.078
fully+0.052
ting+0.050
 Henry+0.048
 Eddie+0.047
Neg logit contributions
tr-0.267
house-0.258
ringe-0.241
oc-0.238
packs-0.237
Token tree
Token ID5509
Feature activation-3.24e-02

Pos logit contributions
ter+0.066
ting+0.056
 Dolphin+0.053
fully+0.052
ous+0.051
Neg logit contributions
house-0.057
packs-0.056
ris-0.052
oc-0.049
ringe-0.048
Tokenhouse
Token ID4803
Feature activation-3.02e-01

Pos logit contributions
house+0.063
 handcuffs+0.060
tr+0.060
ringe+0.059
packs+0.059
Neg logit contributions
ter-0.046
fully-0.039
ous-0.036
ting-0.034
izing-0.033
Token.
Token ID13
Feature activation-2.75e-03

Pos logit contributions
house+0.608
packs+0.551
tr+0.547
ringe+0.539
oc+0.531
Neg logit contributions
ter-0.453
fully-0.387
ous-0.359
ting-0.348
izing-0.337
Token It
Token ID632
Feature activation+2.93e-02

Pos logit contributions
house+0.005
packs+0.004
tr+0.004
 fort+0.004
oc+0.004
Neg logit contributions
ter-0.005
fully-0.004
ous-0.004
ting-0.004
 Dolphin-0.004
Token's
Token ID338
Feature activation-2.73e-03

Pos logit contributions
ter+0.061
fully+0.053
ous+0.052
ting+0.051
 Dolphin+0.050
Neg logit contributions
house-0.041
packs-0.039
ringe-0.035
 handcuffs-0.035
tr-0.034
Token cozy
Token ID37438
Feature activation-4.30e-02

Pos logit contributions
house+0.005
packs+0.005
oc+0.005
tr+0.005
ringe+0.005
Neg logit contributions
ter-0.004
fully-0.004
ting-0.003
ous-0.003
izing-0.003
Token and
Token ID290
Feature activation+1.20e-02

Pos logit contributions
house+0.089
packs+0.081
tr+0.081
oc+0.080
 handcuffs+0.079
Neg logit contributions
ter-0.063
fully-0.052
ous-0.046
ting-0.045
 Dolphin-0.044
Token He
Token ID679
Feature activation-3.20e-02

Pos logit contributions
house+0.019
tr+0.016
packs+0.016
 fort+0.016
oc+0.016
Neg logit contributions
ter-0.021
fully-0.019
ous-0.018
ting-0.018
ancer-0.018
Token cried
Token ID16896
Feature activation+1.37e-03

Pos logit contributions
packs+0.065
house+0.064
 fort+0.056
tr+0.055
ringe+0.055
Neg logit contributions
ter-0.047
fully-0.041
ly-0.038
ting-0.038
 Hop-0.037
Token because
Token ID780
Feature activation-7.32e-02

Pos logit contributions
ter+0.002
 Dolphin+0.002
ous+0.002
ya+0.002
izing+0.002
Neg logit contributions
house-0.002
tr-0.002
 fort-0.002
packs-0.002
ringe-0.002
Token his
Token ID465
Feature activation-6.69e-02

Pos logit contributions
house+0.157
packs+0.147
tr+0.142
 fort+0.138
ringe+0.138
Neg logit contributions
ter-0.105
fully-0.087
ous-0.084
ting-0.080
 Dolphin-0.079
Token play
Token ID711
Feature activation-7.38e-02

Pos logit contributions
house+0.114
oc+0.110
packs+0.107
ris+0.104
ringe+0.100
Neg logit contributions
ter-0.112
fully-0.095
ous-0.094
ble-0.091
 eagle-0.088
Tokenhouse
Token ID4803
Feature activation-3.02e-01

Pos logit contributions
packs+0.101
ringe+0.090
ris+0.088
omo+0.080
 handcuffs+0.079
Neg logit contributions
ter-0.157
fully-0.146
ting-0.128
ing-0.125
ous-0.123
Token was
Token ID373
Feature activation-4.16e-02

Pos logit contributions
house+0.606
packs+0.600
ringe+0.587
oc+0.574
tr+0.554
Neg logit contributions
ter-0.435
fully-0.390
ous-0.334
ting-0.316
izing-0.311
Token broken
Token ID5445
Feature activation-6.13e-02

Pos logit contributions
house+0.066
packs+0.065
oc+0.065
ringe+0.058
ris+0.057
Neg logit contributions
ter-0.080
fully-0.070
ous-0.066
ting-0.063
ping-0.062
Token.
Token ID13
Feature activation-1.08e-02

Pos logit contributions
house+0.117
packs+0.107
tr+0.103
 fort+0.100
ringe+0.099
Neg logit contributions
ter-0.101
fully-0.087
ous-0.084
ting-0.082
izing-0.079
Token His
Token ID2399
Feature activation-2.03e-02

Pos logit contributions
house+0.020
packs+0.018
tr+0.018
 fort+0.017
oc+0.017
Neg logit contributions
ter-0.018
fully-0.016
ous-0.015
ting-0.015
 Dolphin-0.015
Token mom
Token ID1995
Feature activation-1.45e-01

Pos logit contributions
house+0.039
oc+0.038
packs+0.038
ris+0.037
ringe+0.035
Neg logit contributions
ter-0.028
 young-0.023
 eagle-0.022
ble-0.022
fully-0.022
Token then
Token ID788
Feature activation-4.89e-02

Pos logit contributions
ter+0.017
fully+0.015
ting+0.014
ous+0.014
 Dolphin+0.014
Neg logit contributions
house-0.016
packs-0.014
 fort-0.014
ringe-0.014
tr-0.014
Token played
Token ID2826
Feature activation-3.10e-02

Pos logit contributions
house+0.091
oc+0.082
ringe+0.079
tr+0.079
 fort+0.078
Neg logit contributions
ter-0.083
fully-0.072
ting-0.068
 Dolphin-0.068
ous-0.066
Token with
Token ID351
Feature activation-8.29e-02

Pos logit contributions
house+0.056
packs+0.050
tr+0.048
 fort+0.047
oc+0.046
Neg logit contributions
ter-0.055
fully-0.047
ous-0.046
ting-0.045
 Dolphin-0.045
Token the
Token ID262
Feature activation+3.49e-02

Pos logit contributions
house+0.173
tr+0.161
packs+0.157
oc+0.153
ringe+0.152
Neg logit contributions
ter-0.118
fully-0.098
ting-0.091
 Dolphin-0.088
ous-0.088
Token doll
Token ID3654
Feature activation-1.44e-01

Pos logit contributions
ter+0.079
fully+0.070
ous+0.069
ting+0.068
 Dolphin+0.068
Neg logit contributions
house-0.046
packs-0.039
tr-0.037
 fort-0.036
oc-0.035
Tokenhouse
Token ID4803
Feature activation-3.02e-01

Pos logit contributions
 handcuffs+0.276
ringe+0.276
omo+0.266
house+0.266
ris+0.264
Neg logit contributions
ter-0.191
fully-0.164
ting-0.152
ous-0.141
ing-0.137
Token.
Token ID13
Feature activation-7.90e-03

Pos logit contributions
house+0.611
ringe+0.584
oc+0.577
packs+0.556
tr+0.554
Neg logit contributions
ter-0.432
fully-0.382
ting-0.329
ous-0.325
ing-0.307
Token They
Token ID1119
Feature activation-8.83e-02

Pos logit contributions
house+0.014
packs+0.013
ringe+0.013
oc+0.013
 fort+0.012
Neg logit contributions
ter-0.012
 Hop-0.011
fully-0.011
 Dolphin-0.011
ous-0.011
Token had
Token ID550
Feature activation-7.81e-02

Pos logit contributions
house+0.176
packs+0.165
ringe+0.164
oc+0.157
 fort+0.155
Neg logit contributions
ter-0.133
fully-0.105
 Hop-0.097
 mer-0.096
 Dolphin-0.095
Token a
Token ID257
Feature activation-2.28e-02

Pos logit contributions
house+0.163
packs+0.149
oc+0.149
tr+0.147
ringe+0.143
Neg logit contributions
ter-0.109
fully-0.091
 Dolphin-0.086
ous-0.084
 Hop-0.084
Token lot
Token ID1256
Feature activation-1.70e-01

Pos logit contributions
packs+0.039
house+0.039
ringe+0.038
oc+0.034
tr+0.034
Neg logit contributions
ter-0.033
 young-0.032
 Hop-0.030
 mer-0.030
 Dolphin-0.030
TokenThe
Token ID464
Feature activation+2.32e-02

Pos logit contributions
ter+0.013
fully+0.011
ous+0.010
ting+0.010
 Dolphin+0.010
Neg logit contributions
house-0.021
packs-0.019
tr-0.019
 fort-0.019
ringe-0.018
Token end
Token ID886
Feature activation-3.80e-02

Pos logit contributions
ter+0.040
fully+0.034
ous+0.033
 Dolphin+0.033
ting+0.033
Neg logit contributions
house-0.043
packs-0.039
tr-0.038
oc-0.036
 fort-0.036
Token.
Token ID13
Feature activation+2.52e-03

Pos logit contributions
house+0.068
packs+0.061
tr+0.059
 fort+0.058
oc+0.056
Neg logit contributions
ter-0.068
fully-0.058
ous-0.057
ting-0.056
 Dolphin-0.056
Token<|endoftext|>
Token ID50256
Feature activation+4.68e-02

Pos logit contributions
ter+0.004
fully+0.003
ous+0.003
ting+0.003
 Dolphin+0.003
Neg logit contributions
house-0.005
packs-0.004
tr-0.004
 fort-0.004
oc-0.004
TokenOnce
Token ID7454
Feature activation-8.61e-02

Pos logit contributions
ter+0.094
izing+0.088
fully+0.087
ous+0.086
ya+0.085
Neg logit contributions
house-0.075
packs-0.066
tr-0.059
oc-0.059
 fort-0.057
Token upon
Token ID2402
Feature activation+1.88e-01

Pos logit contributions
house+0.196
packs+0.190
 fort+0.176
tr+0.169
oc+0.165
Neg logit contributions
ter-0.112
ting-0.091
izing-0.084
fully-0.084
 Dolphin-0.083
Token a
Token ID257
Feature activation-1.49e-02

Pos logit contributions
ter+0.348
ous+0.305
ting+0.302
izing+0.297
 Dolphin+0.294
Neg logit contributions
house-0.338
packs-0.296
tr-0.278
 fort-0.276
oc-0.263
Token time
Token ID640
Feature activation-1.00e-02

Pos logit contributions
house+0.026
packs+0.024
ringe+0.023
tr+0.022
 handcuffs+0.022
Neg logit contributions
ter-0.025
fully-0.023
ous-0.022
 Dolphin-0.021
ya-0.021
Token,
Token ID11
Feature activation+3.52e-02

Pos logit contributions
house+0.014
packs+0.012
 fort+0.010
oc+0.010
tr+0.010
Neg logit contributions
ter-0.022
ting-0.021
izing-0.020
ping-0.020
 Dolphin-0.020
Token there
Token ID612
Feature activation-5.72e-03

Pos logit contributions
ter+0.061
ting+0.057
ous+0.057
 Dolphin+0.056
izing+0.055
Neg logit contributions
house-0.067
packs-0.057
tr-0.052
 fort-0.051
oc-0.049
Token was
Token ID373
Feature activation-2.88e-02

Pos logit contributions
house+0.009
packs+0.008
tr+0.007
 fort+0.007
oc+0.007
Neg logit contributions
ter-0.011
fully-0.010
ous-0.010
ting-0.010
 Dolphin-0.010
Token.
Token ID13
Feature activation-1.04e-02

Pos logit contributions
ter+0.079
ous+0.069
fully+0.068
 Dolphin+0.068
izing+0.066
Neg logit contributions
house-0.075
packs-0.066
 fort-0.064
tr-0.063
ringe-0.059
Token
Token ID198
Feature activation+4.00e-03

Pos logit contributions
house+0.016
packs+0.014
tr+0.014
 fort+0.012
 handcuffs+0.012
Neg logit contributions
ter-0.021
fully-0.019
izing-0.019
ous-0.019
ting-0.018
Token
Token ID198
Feature activation+4.79e-03

Pos logit contributions
ter+0.007
fully+0.006
ous+0.006
ting+0.006
 Dolphin+0.006
Neg logit contributions
house-0.007
packs-0.006
tr-0.006
 fort-0.006
oc-0.006
Token"
Token ID1
Feature activation-5.71e-02

Pos logit contributions
ter+0.008
fully+0.007
ous+0.007
ting+0.007
 Dolphin+0.007
Neg logit contributions
house-0.009
packs-0.008
tr-0.008
 fort-0.007
oc-0.007
TokenThank
Token ID10449
Feature activation-4.48e-02

Pos logit contributions
house+0.085
tr+0.081
packs+0.076
ringe+0.071
 bisc+0.070
Neg logit contributions
ter-0.105
ous-0.103
fully-0.099
 Joey-0.093
 Dolphin-0.093
Token you
Token ID345
Feature activation+1.81e-01

Pos logit contributions
house+0.096
packs+0.085
oc+0.082
tr+0.080
tail+0.079
Neg logit contributions
ter-0.056
ous-0.053
 Dolphin-0.050
fully-0.050
 Nathan-0.047
Token,
Token ID11
Feature activation+2.37e-02

Pos logit contributions
ter+0.260
fully+0.215
ous+0.209
 Dolphin+0.198
ting+0.197
Neg logit contributions
house-0.386
packs-0.358
tr-0.346
ringe-0.334
 fort-0.333
Token Sara
Token ID24799
Feature activation-7.91e-02

Pos logit contributions
ter+0.044
fully+0.037
ous+0.037
 Dolphin+0.036
ting+0.036
Neg logit contributions
house-0.042
packs-0.037
tr-0.035
 fort-0.034
ringe-0.034
Token.
Token ID13
Feature activation-7.65e-03

Pos logit contributions
house+0.151
packs+0.137
tr+0.132
 fort+0.129
oc+0.126
Neg logit contributions
ter-0.131
fully-0.111
ous-0.109
ting-0.106
 Dolphin-0.105
Token You
Token ID921
Feature activation+9.61e-02

Pos logit contributions
house+0.017
packs+0.015
tr+0.015
 fort+0.014
oc+0.014
Neg logit contributions
ter-0.010
fully-0.009
ous-0.009
izing-0.009
ting-0.008
Token are
Token ID389
Feature activation+3.59e-02

Pos logit contributions
ter+0.196
fully+0.177
ous+0.174
ting+0.170
izing+0.169
Neg logit contributions
house-0.145
packs-0.124
tr-0.121
 fort-0.118
oc-0.113
Tokenmy
Token ID1820
Feature activation-8.76e-04

Pos logit contributions
house+0.206
packs+0.177
tr+0.169
 fort+0.166
oc+0.163
Neg logit contributions
ter-0.273
fully-0.244
ous-0.239
 Dolphin-0.237
ting-0.231
Token will
Token ID481
Feature activation-6.05e-02

Pos logit contributions
house+0.001
packs+0.001
tr+0.001
 fort+0.001
oc+0.001
Neg logit contributions
ter-0.002
fully-0.002
ous-0.002
 Dolphin-0.002
ting-0.001
Token be
Token ID307
Feature activation+1.71e-02

Pos logit contributions
house+0.116
packs+0.111
tr+0.103
 fort+0.102
oc+0.099
Neg logit contributions
ter-0.101
fully-0.084
ous-0.080
 Dolphin-0.079
 Hop-0.078
Token mad
Token ID8805
Feature activation+1.66e-02

Pos logit contributions
ter+0.025
fully+0.019
ting+0.019
bling+0.018
izing+0.018
Neg logit contributions
house-0.036
tr-0.035
packs-0.034
oc-0.033
 fort-0.032
Token if
Token ID611
Feature activation-2.17e-02

Pos logit contributions
ter+0.031
 Dolphin+0.028
izing+0.027
ya+0.027
ous+0.027
Neg logit contributions
house-0.026
packs-0.023
tr-0.022
 fort-0.022
oc-0.020
Token you
Token ID345
Feature activation+1.84e-01

Pos logit contributions
house+0.043
packs+0.040
 fort+0.038
tr+0.036
 handcuffs+0.035
Neg logit contributions
ter-0.031
fully-0.028
ous-0.028
ivated-0.027
ting-0.027
Token make
Token ID787
Feature activation-2.67e-02

Pos logit contributions
ter+0.356
fully+0.313
ous+0.309
ting+0.302
 Dolphin+0.302
Neg logit contributions
house-0.299
packs-0.263
tr-0.253
 fort-0.246
oc-0.242
Token a
Token ID257
Feature activation-2.03e-02

Pos logit contributions
house+0.050
packs+0.044
 fort+0.043
tr+0.042
oc+0.042
Neg logit contributions
ter-0.044
fully-0.039
ous-0.039
ancer-0.037
ting-0.037
Token mess
Token ID2085
Feature activation-1.19e-01

Pos logit contributions
house+0.032
packs+0.030
tr+0.027
ringe+0.026
ris+0.025
Neg logit contributions
ter-0.039
ble-0.034
 Dolphin-0.033
 dolphin-0.033
ya-0.033
Token,"
Token ID553
Feature activation+4.84e-02

Pos logit contributions
house+0.198
packs+0.170
 fort+0.162
tr+0.162
oc+0.159
Neg logit contributions
ter-0.217
 Dolphin-0.196
ous-0.192
fully-0.190
ya-0.185
Token Ben
Token ID3932
Feature activation+8.75e-03

Pos logit contributions
ter+0.098
ous+0.088
fully+0.087
ting+0.086
izing+0.084
Neg logit contributions
house-0.073
packs-0.061
tr-0.059
 fort-0.058
oc-0.058
Token.
Token ID13
Feature activation+1.79e-03

Pos logit contributions
ter+0.023
 Dolphin+0.020
ous+0.019
ting+0.019
fully+0.019
Neg logit contributions
house-0.023
packs-0.021
tr-0.019
 fort-0.019
 handcuffs-0.018
Token I
Token ID314
Feature activation+1.01e-01

Pos logit contributions
ter+0.003
fully+0.002
ous+0.002
izing+0.002
ting+0.002
Neg logit contributions
house-0.004
tr-0.003
oc-0.003
packs-0.003
ringe-0.003
Token was
Token ID373
Feature activation-3.52e-02

Pos logit contributions
ter+0.173
fully+0.147
ous+0.146
 Hop+0.144
ancer+0.142
Neg logit contributions
house-0.168
packs-0.162
tr-0.153
 fort-0.153
ringe-0.151
Token wondering
Token ID11263
Feature activation+9.08e-02

Pos logit contributions
house+0.067
packs+0.057
 fort+0.055
tr+0.054
 handcuffs+0.053
Neg logit contributions
ter-0.058
fully-0.052
ous-0.051
 Dolphin-0.049
izing-0.049
Token if
Token ID611
Feature activation-1.49e-02

Pos logit contributions
ter+0.166
 Dolphin+0.147
gging+0.142
rus+0.141
izing+0.141
Neg logit contributions
house-0.142
packs-0.133
 fort-0.122
oc-0.113
tr-0.113
Token you
Token ID345
Feature activation+1.88e-01

Pos logit contributions
house+0.031
packs+0.028
 fort+0.027
tr+0.025
 handcuffs+0.025
Neg logit contributions
ter-0.021
fully-0.018
gging-0.017
ous-0.017
ivated-0.017
Token could
Token ID714
Feature activation+6.35e-02

Pos logit contributions
ter+0.373
fully+0.329
ous+0.324
izing+0.323
ting+0.322
Neg logit contributions
house-0.295
packs-0.262
tr-0.243
oc-0.237
 fort-0.237
Token donate
Token ID16565
Feature activation-1.12e-01

Pos logit contributions
ter+0.126
fully+0.110
ous+0.108
ting+0.106
 Dolphin+0.106
Neg logit contributions
house-0.100
packs-0.089
tr-0.085
 fort-0.083
oc-0.081
Token anything
Token ID1997
Feature activation+7.44e-02

Pos logit contributions
house+0.203
packs+0.186
tr+0.173
 handcuffs+0.172
oc+0.170
Neg logit contributions
ter-0.192
ous-0.168
fully-0.163
 Dolphin-0.161
ancer-0.156
Token?"
Token ID1701
Feature activation+3.90e-02

Pos logit contributions
ter+0.134
 Dolphin+0.115
fully+0.113
ous+0.112
ting+0.111
Neg logit contributions
house-0.128
packs-0.118
tr-0.109
 fort-0.109
oc-0.105
Token The
Token ID383
Feature activation-6.49e-03

Pos logit contributions
ter+0.076
ous+0.066
fully+0.066
izing+0.065
ting+0.062
Neg logit contributions
house-0.066
oc-0.057
packs-0.054
tr-0.053
 handcuffs-0.052
Token easier
Token ID4577
Feature activation-4.43e-02

Pos logit contributions
house+0.119
packs+0.108
tr+0.101
 fort+0.100
oc+0.097
Neg logit contributions
ter-0.117
fully-0.101
 Dolphin-0.099
ting-0.098
ous-0.098
Token and
Token ID290
Feature activation+1.47e-02

Pos logit contributions
house+0.082
packs+0.074
tr+0.071
 fort+0.070
oc+0.068
Neg logit contributions
ter-0.076
fully-0.065
ous-0.064
 Dolphin-0.063
ting-0.062
Token faster
Token ID5443
Feature activation+4.69e-03

Pos logit contributions
ter+0.027
fully+0.023
ous+0.023
 Dolphin+0.022
ting+0.022
Neg logit contributions
house-0.026
packs-0.023
tr-0.022
 fort-0.022
oc-0.021
Token.
Token ID13
Feature activation-1.51e-03

Pos logit contributions
ter+0.008
fully+0.007
ous+0.007
ting+0.007
 Dolphin+0.007
Neg logit contributions
house-0.008
packs-0.007
tr-0.007
 fort-0.007
oc-0.007
Token Thank
Token ID6952
Feature activation+1.68e-02

Pos logit contributions
house+0.003
packs+0.003
tr+0.003
oc+0.003
 fort+0.003
Neg logit contributions
ter-0.002
fully-0.002
ous-0.002
izing-0.002
ting-0.002
Token you
Token ID345
Feature activation+1.88e-01

Pos logit contributions
ous+0.028
fully+0.026
ter+0.025
iful+0.024
ivating+0.023
Neg logit contributions
house-0.029
packs-0.025
ossible-0.025
oc-0.025
 handcuffs-0.023
Token very
Token ID845
Feature activation-3.50e-02

Pos logit contributions
ter+0.359
fully+0.312
izing+0.312
ting+0.307
ous+0.306
Neg logit contributions
house-0.309
packs-0.273
 fort-0.267
tr-0.257
oc-0.257
Token much
Token ID881
Feature activation-1.34e-01

Pos logit contributions
house+0.034
packs+0.028
oc+0.026
 fort+0.024
tr+0.022
Neg logit contributions
ter-0.091
ting-0.083
fully-0.081
 Dolphin-0.079
ous-0.077
Token.
Token ID13
Feature activation-9.87e-04

Pos logit contributions
house+0.228
packs+0.205
tr+0.196
 fort+0.195
oc+0.190
Neg logit contributions
ter-0.249
fully-0.213
ous-0.210
 Dolphin-0.209
ting-0.208
Token Here
Token ID3423
Feature activation+9.58e-02

Pos logit contributions
house+0.002
packs+0.002
tr+0.002
oc+0.002
ringe+0.002
Neg logit contributions
ter-0.001
fully-0.001
ous-0.001
izing-0.001
ting-0.001
Token,
Token ID11
Feature activation+3.16e-02

Pos logit contributions
ter+0.162
fully+0.146
ous+0.142
ting+0.136
 Dolphin+0.134
Neg logit contributions
house-0.180
packs-0.159
 fort-0.152
tr-0.148
oc-0.147
Token one
Token ID530
Feature activation-1.15e-01

Pos logit contributions
house+0.075
packs+0.068
tr+0.065
 fort+0.064
ringe+0.063
Neg logit contributions
ter-0.060
fully-0.052
ous-0.052
 Dolphin-0.050
izing-0.049
Token I
Token ID314
Feature activation+9.74e-02

Pos logit contributions
house+0.227
packs+0.206
tr+0.202
ringe+0.201
 fort+0.197
Neg logit contributions
ter-0.183
fully-0.153
ous-0.145
ting-0.140
 Dolphin-0.140
Token like
Token ID588
Feature activation+5.55e-03

Pos logit contributions
ter+0.169
fully+0.142
ous+0.140
ancer+0.136
ting+0.133
Neg logit contributions
house-0.170
packs-0.153
 fort-0.152
tr-0.149
ringe-0.148
Token."
Token ID526
Feature activation-2.96e-03

Pos logit contributions
ter+0.009
fully+0.008
ous+0.007
izing+0.007
ancer+0.007
Neg logit contributions
house-0.011
packs-0.010
 fort-0.009
tr-0.009
 handcuffs-0.009
Token Tim
Token ID5045
Feature activation-6.77e-02

Pos logit contributions
house+0.005
packs+0.004
tr+0.004
 fort+0.004
oc+0.004
Neg logit contributions
ter-0.006
fully-0.005
ous-0.005
 Dolphin-0.005
ting-0.005
Token thought
Token ID1807
Feature activation+3.75e-02

Pos logit contributions
house+0.117
packs+0.100
tr+0.099
 fort+0.096
oc+0.095
Neg logit contributions
ter-0.125
ous-0.110
fully-0.109
 Dolphin-0.107
ting-0.107
Token for
Token ID329
Feature activation-5.92e-02

Pos logit contributions
ter+0.080
 Dolphin+0.070
redible+0.066
ting+0.065
izing+0.065
Neg logit contributions
house-0.055
 fort-0.051
packs-0.046
 bars-0.043
 handcuffs-0.042
Token a
Token ID257
Feature activation-2.16e-02

Pos logit contributions
house+0.106
tr+0.088
 fort+0.086
packs+0.085
 handcuffs+0.084
Neg logit contributions
ter-0.108
ous-0.097
 Dolphin-0.096
ting-0.095
fully-0.094
Token moment
Token ID2589
Feature activation+6.22e-02

Pos logit contributions
house+0.048
packs+0.044
tr+0.042
ringe+0.041
 fort+0.041
Neg logit contributions
ter-0.028
fully-0.025
ous-0.023
 Dolphin-0.023
ancer-0.021
Token and
Token ID290
Feature activation+8.05e-03

Pos logit contributions
ter+0.111
 Dolphin+0.096
ous+0.094
ancer+0.092
fully+0.091
Neg logit contributions
house-0.111
packs-0.095
tr-0.092
oc-0.092
 fort-0.092
Token then
Token ID788
Feature activation-4.94e-02

Pos logit contributions
ter+0.016
ous+0.015
ancer+0.015
ting+0.015
fully+0.015
Neg logit contributions
house-0.012
packs-0.009
tr-0.009
 fort-0.009
oc-0.009
Token
Token ID198
Feature activation+4.34e-02

Pos logit contributions
ter+0.154
fully+0.137
ous+0.136
ancer+0.133
 Dolphin+0.130
Neg logit contributions
house-0.127
tr-0.101
packs-0.100
oc-0.099
 fort-0.096
Token
Token ID198
Feature activation+3.60e-02

Pos logit contributions
ter+0.069
fully+0.059
 Dolphin+0.057
ous+0.057
ting+0.055
Neg logit contributions
house-0.084
packs-0.076
tr-0.074
 fort-0.073
ringe-0.071
TokenHe
Token ID1544
Feature activation-6.91e-03

Pos logit contributions
ter+0.058
fully+0.050
 Dolphin+0.049
ous+0.047
ting+0.046
Neg logit contributions
house-0.068
packs-0.062
tr-0.061
 fort-0.060
ringe-0.058
Token said
Token ID531
Feature activation+1.94e-04

Pos logit contributions
house+0.012
packs+0.011
tr+0.011
ringe+0.010
 fort+0.010
Neg logit contributions
ter-0.012
fully-0.010
ya-0.010
 Dolphin-0.010
 Hop-0.010
Token,
Token ID11
Feature activation+4.17e-02

Pos logit contributions
ter+0.000
ancer+0.000
izing+0.000
ting+0.000
iful+0.000
Neg logit contributions
house-0.000
tr-0.000
oc-0.000
packs-0.000
 fort-0.000
Token "
Token ID366
Feature activation+2.81e-02

Pos logit contributions
ter+0.072
izing+0.068
heart+0.063
ancer+0.063
ternally+0.062
Neg logit contributions
house-0.071
tr-0.064
packs-0.058
 fort-0.058
oc-0.057
TokenI
Token ID40
Feature activation+2.18e-02

Pos logit contributions
ter+0.047
ous+0.041
fully+0.040
ting+0.038
izing+0.038
Neg logit contributions
house-0.053
packs-0.047
tr-0.046
oc-0.045
ringe-0.044
Token found
Token ID1043
Feature activation-6.33e-02

Pos logit contributions
ter+0.035
fully+0.030
ous+0.029
 Joey+0.029
 Hop+0.028
Neg logit contributions
house-0.039
packs-0.037
 fort-0.036
ringe-0.036
tr-0.035
Token a
Token ID257
Feature activation-1.78e-02

Pos logit contributions
house+0.120
packs+0.106
 fort+0.104
 handcuffs+0.102
oc+0.102
Neg logit contributions
ter-0.100
ous-0.091
ancer-0.089
fully-0.088
 Dolphin-0.086
Token new
Token ID649
Feature activation+6.95e-03

Pos logit contributions
house+0.029
packs+0.027
tr+0.025
ringe+0.025
oc+0.024
Neg logit contributions
ter-0.033
 Dolphin-0.028
ous-0.028
fully-0.028
ting-0.027
Token house
Token ID2156
Feature activation-1.06e-01

Pos logit contributions
ter+0.015
fully+0.013
ous+0.012
 Dolphin+0.012
ya+0.012
Neg logit contributions
house-0.010
packs-0.009
tr-0.008
ringe-0.008
oc-0.008
Token wanted
Token ID2227
Feature activation-9.64e-03

Pos logit contributions
house+0.066
packs+0.062
 handcuffs+0.059
ringe+0.058
oc+0.058
Neg logit contributions
ter-0.063
fully-0.057
ous-0.052
ya-0.052
ly-0.051
Token to
Token ID284
Feature activation-1.96e-02

Pos logit contributions
house+0.017
packs+0.016
 fort+0.015
tr+0.015
 handcuffs+0.014
Neg logit contributions
ter-0.017
ous-0.015
fully-0.015
izing-0.014
ancer-0.014
Token print
Token ID3601
Feature activation+2.34e-02

Pos logit contributions
house+0.034
packs+0.034
tr+0.033
 fort+0.031
 handcuffs+0.031
Neg logit contributions
ter-0.032
fully-0.028
 Joey-0.026
ble-0.026
 Dolphin-0.026
Token a
Token ID257
Feature activation-1.72e-02

Pos logit contributions
ter+0.042
ous+0.037
fully+0.036
 Dolphin+0.036
ancer+0.036
Neg logit contributions
house-0.041
packs-0.037
 fort-0.035
tr-0.035
 handcuffs-0.033
Token picture
Token ID4286
Feature activation-1.14e-01

Pos logit contributions
house+0.028
packs+0.025
tr+0.024
 fort+0.023
ringe+0.023
Neg logit contributions
ter-0.033
fully-0.029
ous-0.028
 Dolphin-0.028
ting-0.027
Token of
Token ID286
Feature activation+3.81e-02

Pos logit contributions
oc+0.242
house+0.240
ringe+0.238
packs+0.236
 handcuffs+0.230
Neg logit contributions
ter-0.148
fully-0.134
ous-0.118
ting-0.109
ing-0.108
Token her
Token ID607
Feature activation-2.92e-02

Pos logit contributions
ter+0.064
fully+0.057
ous+0.057
ancer+0.055
 Dolphin+0.055
Neg logit contributions
house-0.071
packs-0.062
 fort-0.059
tr-0.058
 handcuffs-0.057
Token big
Token ID1263
Feature activation-1.11e-01

Pos logit contributions
house+0.048
packs+0.043
tr+0.041
 fort+0.040
oc+0.039
Neg logit contributions
ter-0.056
fully-0.048
ous-0.048
 Dolphin-0.047
ting-0.046
Token stone
Token ID7815
Feature activation+1.06e-01

Pos logit contributions
house+0.198
packs+0.180
tr+0.179
ringe+0.169
oc+0.167
Neg logit contributions
ter-0.196
fully-0.163
 Dolphin-0.162
ous-0.160
ting-0.159
Token.
Token ID13
Feature activation+3.51e-03

Pos logit contributions
ter+0.156
fully+0.126
ous+0.116
ting+0.111
 Hop+0.107
Neg logit contributions
house-0.207
 handcuffs-0.205
packs-0.203
ringe-0.202
oc-0.201
Token She
Token ID1375
Feature activation-3.72e-02

Pos logit contributions
ter+0.008
 Dolphin+0.007
ancer+0.006
ble+0.006
ous+0.006
Neg logit contributions
house-0.006
tr-0.004
packs-0.004
 fort-0.004
 bars-0.004
Token his
Token ID465
Feature activation-5.72e-02

Pos logit contributions
ancer+0.258
ter+0.253
redible+0.238
ous+0.236
 Dolphin+0.233
Neg logit contributions
house-0.167
oc-0.136
packs-0.134
 fort-0.128
tr-0.120
Token head
Token ID1182
Feature activation+1.25e-02

Pos logit contributions
house+0.086
packs+0.074
 fort+0.074
tr+0.074
ringe+0.069
Neg logit contributions
ter-0.118
ous-0.104
fully-0.104
ting-0.103
 Dolphin-0.101
Token and
Token ID290
Feature activation+1.50e-02

Pos logit contributions
ter+0.022
 Dolphin+0.020
ancer+0.020
ous+0.019
ya+0.019
Neg logit contributions
house-0.022
tr-0.019
packs-0.018
oc-0.018
 fort-0.017
Token was
Token ID373
Feature activation-3.08e-02

Pos logit contributions
ter+0.026
fully+0.023
ous+0.022
ting+0.022
 Dolphin+0.022
Neg logit contributions
house-0.027
packs-0.025
tr-0.024
 fort-0.023
ringe-0.022
Token very
Token ID845
Feature activation-3.78e-02

Pos logit contributions
house+0.062
packs+0.056
tr+0.055
 fort+0.054
oc+0.053
Neg logit contributions
ter-0.048
fully-0.040
ous-0.039
ting-0.038
 Dolphin-0.038
Token careful
Token ID8161
Feature activation+5.11e-02

Pos logit contributions
house+0.074
packs+0.066
tr+0.062
 fort+0.062
oc+0.062
Neg logit contributions
ter-0.062
fully-0.053
ous-0.052
ting-0.052
 Dolphin-0.050
Token.
Token ID13
Feature activation+3.20e-04

Pos logit contributions
ter+0.114
 Dolphin+0.106
ancer+0.105
redible+0.101
ting+0.101
Neg logit contributions
house-0.068
packs-0.054
tr-0.048
 fort-0.046
oc-0.046
Token He
Token ID679
Feature activation-2.29e-02

Pos logit contributions
ter+0.001
fully+0.001
redible+0.001
izing+0.001
ous+0.001
Neg logit contributions
house-0.000
tr-0.000
oc-0.000
 fort-0.000
packs-0.000
Token stepped
Token ID10764
Feature activation+9.59e-02

Pos logit contributions
house+0.041
packs+0.040
 fort+0.036
tr+0.036
ringe+0.036
Neg logit contributions
ter-0.038
fully-0.033
ous-0.032
 Hop-0.031
ting-0.031
Token around
Token ID1088
Feature activation+6.31e-02

Pos logit contributions
ter+0.181
ous+0.167
izing+0.157
fully+0.157
 Dolphin+0.157
Neg logit contributions
house-0.156
packs-0.136
 fort-0.132
tr-0.130
 handcuffs-0.129
Token the
Token ID262
Feature activation+4.39e-02

Pos logit contributions
ter+0.108
ous+0.103
ancer+0.102
redible+0.098
 Dolphin+0.096
Neg logit contributions
house-0.111
oc-0.096
 fort-0.094
 handcuffs-0.088
packs-0.088
Token
Token ID198
Feature activation+1.29e-02

Pos logit contributions
ter+0.022
fully+0.019
ous+0.018
ting+0.018
 Dolphin+0.018
Neg logit contributions
house-0.027
packs-0.024
tr-0.023
 fort-0.023
oc-0.022
TokenAfter
Token ID3260
Feature activation-3.97e-02

Pos logit contributions
ter+0.021
fully+0.018
ous+0.018
 Dolphin+0.018
ting+0.017
Neg logit contributions
house-0.024
packs-0.022
tr-0.021
 fort-0.021
ringe-0.020
Token eating
Token ID6600
Feature activation-1.49e-02

Pos logit contributions
house+0.073
packs+0.065
tr+0.063
 fort+0.062
oc+0.060
Neg logit contributions
ter-0.069
fully-0.059
ous-0.058
ting-0.056
 Dolphin-0.056
Token,
Token ID11
Feature activation+2.94e-02

Pos logit contributions
house+0.028
packs+0.025
 fort+0.025
tr+0.023
oc+0.023
Neg logit contributions
ter-0.025
ous-0.022
 Dolphin-0.021
ancer-0.021
ting-0.021
Token Lily
Token ID20037
Feature activation-1.30e-01

Pos logit contributions
ter+0.046
fully+0.040
ous+0.040
izing+0.038
 Dolphin+0.038
Neg logit contributions
house-0.058
packs-0.051
tr-0.049
 fort-0.049
oc-0.047
Token felt
Token ID2936
Feature activation+3.09e-02

Pos logit contributions
house+0.216
packs+0.194
tr+0.185
 fort+0.181
ringe+0.176
Neg logit contributions
ter-0.245
fully-0.214
ous-0.210
ting-0.206
 Dolphin-0.205
Token very
Token ID845
Feature activation-4.13e-02

Pos logit contributions
ter+0.062
ous+0.056
fully+0.055
ancer+0.054
 Dolphin+0.053
Neg logit contributions
house-0.048
packs-0.039
 fort-0.038
 handcuffs-0.038
tr-0.037
Token sleepy
Token ID42023
Feature activation+2.97e-02

Pos logit contributions
house+0.066
packs+0.056
 fort+0.054
oc+0.053
tr+0.051
Neg logit contributions
ter-0.081
fully-0.071
ous-0.071
ting-0.070
 Dolphin-0.069
Token.
Token ID13
Feature activation-5.03e-03

Pos logit contributions
ter+0.052
ous+0.044
 Dolphin+0.044
fully+0.044
izing+0.043
Neg logit contributions
house-0.053
packs-0.046
tr-0.046
 fort-0.045
oc-0.043
Token She
Token ID1375
Feature activation-4.06e-02

Pos logit contributions
house+0.007
tr+0.006
packs+0.006
 fort+0.006
ringe+0.005
Neg logit contributions
ter-0.011
izing-0.009
fully-0.009
ting-0.009
ous-0.009
Token went
Token ID1816
Feature activation-1.03e-01

Pos logit contributions
house+0.070
packs+0.063
tr+0.061
 fort+0.059
ringe+0.058
Neg logit contributions
ter-0.073
fully-0.064
ous-0.062
 Dolphin-0.061
ting-0.061
Token cast
Token ID3350
Feature activation+7.04e-03

Pos logit contributions
house+0.113
packs+0.099
tr+0.088
oc+0.087
 fort+0.087
Neg logit contributions
ter-0.191
fully-0.170
ous-0.170
ting-0.163
 Dolphin-0.160
Token.
Token ID13
Feature activation-2.25e-03

Pos logit contributions
ter+0.013
ous+0.011
 Dolphin+0.011
fully+0.011
ting+0.011
Neg logit contributions
house-0.012
packs-0.011
tr-0.010
 fort-0.010
oc-0.010
Token<|endoftext|>
Token ID50256
Feature activation+4.39e-02

Pos logit contributions
house+0.004
packs+0.004
tr+0.004
 fort+0.004
oc+0.004
Neg logit contributions
ter-0.004
fully-0.003
ous-0.003
 Dolphin-0.003
ting-0.003
TokenTom
Token ID13787
Feature activation-9.04e-02

Pos logit contributions
ter+0.083
fully+0.075
ous+0.075
izing+0.073
ya+0.072
Neg logit contributions
house-0.075
packs-0.068
tr-0.062
 fort-0.062
oc-0.060
Token and
Token ID290
Feature activation+1.47e-02

Pos logit contributions
house+0.157
packs+0.140
tr+0.140
 fort+0.135
ringe+0.131
Neg logit contributions
ter-0.161
fully-0.138
ous-0.133
ting-0.131
 Dolphin-0.131
Token Sam
Token ID3409
Feature activation-4.20e-02

Pos logit contributions
ter+0.023
fully+0.020
 Dolphin+0.019
ous+0.019
ting+0.018
Neg logit contributions
house-0.029
packs-0.026
tr-0.026
 fort-0.026
ringe-0.025
Token are
Token ID389
Feature activation+4.20e-02

Pos logit contributions
house+0.063
packs+0.055
tr+0.054
 fort+0.052
ringe+0.051
Neg logit contributions
ter-0.086
fully-0.075
ous-0.073
ting-0.073
 Dolphin-0.072
Token friends
Token ID2460
Feature activation+4.36e-02

Pos logit contributions
ter+0.062
fully+0.050
 Dolphin+0.050
ous+0.049
ancer+0.048
Neg logit contributions
house-0.087
tr-0.080
oc-0.078
packs-0.077
ringe-0.077
Token.
Token ID13
Feature activation+1.55e-03

Pos logit contributions
ter+0.078
fully+0.067
ous+0.066
ting+0.064
 Dolphin+0.064
Neg logit contributions
house-0.078
packs-0.070
tr-0.067
 fort-0.066
oc-0.064
Token They
Token ID1119
Feature activation-8.15e-02

Pos logit contributions
ter+0.002
 Joey+0.002
 Dolphin+0.002
ous+0.002
 Nathan+0.002
Neg logit contributions
house-0.003
ringe-0.003
 fort-0.003
tr-0.003
oc-0.003
Token like
Token ID588
Feature activation+1.09e-02

Pos logit contributions
house+0.122
packs+0.109
tr+0.099
 fort+0.098
oc+0.094
Neg logit contributions
ter-0.166
izing-0.148
ous-0.147
fully-0.147
ting-0.144
Tokenmy
Token ID1820
Feature activation+1.90e-04

Pos logit contributions
house+0.221
packs+0.197
tr+0.187
 fort+0.185
oc+0.179
Neg logit contributions
ter-0.282
fully-0.248
ous-0.244
ting-0.239
 Dolphin-0.234
Token loved
Token ID6151
Feature activation+6.86e-02

Pos logit contributions
ter+0.000
fully+0.000
ous+0.000
ting+0.000
 Dolphin+0.000
Neg logit contributions
house-0.000
packs-0.000
tr-0.000
 fort-0.000
oc-0.000
Token it
Token ID340
Feature activation+1.53e-02

Pos logit contributions
ter+0.112
fully+0.093
ous+0.093
 Dolphin+0.091
ting+0.089
Neg logit contributions
house-0.132
packs-0.119
 fort-0.114
tr-0.113
ringe-0.108
Token very
Token ID845
Feature activation-4.17e-02

Pos logit contributions
ter+0.024
fully+0.021
ous+0.020
ting+0.019
izing+0.018
Neg logit contributions
house-0.030
packs-0.029
ringe-0.027
oc-0.027
tr-0.027
Token much
Token ID881
Feature activation-1.39e-01

Pos logit contributions
house+0.055
packs+0.049
tr+0.046
ringe+0.045
 fort+0.044
Neg logit contributions
ter-0.093
fully-0.084
ous-0.083
izing-0.081
 Dolphin-0.079
Token.
Token ID13
Feature activation-7.36e-03

Pos logit contributions
house+0.277
packs+0.261
tr+0.249
ringe+0.245
 handcuffs+0.237
Neg logit contributions
ter-0.218
fully-0.194
ous-0.183
ting-0.176
izing-0.174
Token From
Token ID3574
Feature activation+3.12e-03

Pos logit contributions
house+0.014
packs+0.013
tr+0.013
 fort+0.013
ringe+0.012
Neg logit contributions
ter-0.011
fully-0.010
ous-0.010
ting-0.009
 Dolphin-0.009
Token that
Token ID326
Feature activation-4.34e-02

Pos logit contributions
ter+0.006
ancer+0.005
ous+0.005
fully+0.005
 Dolphin+0.005
Neg logit contributions
house-0.005
packs-0.004
 fort-0.004
tr-0.004
ris-0.004
Token day
Token ID1110
Feature activation-6.09e-02

Pos logit contributions
house+0.071
packs+0.066
tr+0.060
ringe+0.059
 fort+0.059
Neg logit contributions
ter-0.082
fully-0.073
ous-0.071
 Dolphin-0.070
ting-0.069
Token on
Token ID319
Feature activation-1.08e-01

Pos logit contributions
house+0.119
packs+0.110
 fort+0.108
tr+0.105
oc+0.104
Neg logit contributions
ter-0.095
ting-0.084
fully-0.083
ing-0.082
ous-0.078
Token,
Token ID11
Feature activation+2.36e-02

Pos logit contributions
house+0.230
packs+0.212
tr+0.208
 fort+0.201
oc+0.200
Neg logit contributions
ter-0.155
fully-0.130
ous-0.125
ting-0.122
 Dolphin-0.121
Token
Token ID198
Feature activation+9.23e-03

Pos logit contributions
ter+0.018
ous+0.015
fully+0.015
ting+0.015
izing+0.015
Neg logit contributions
house-0.014
oc-0.012
packs-0.012
tr-0.012
 handcuffs-0.011
TokenThe
Token ID464
Feature activation+2.67e-02

Pos logit contributions
ter+0.015
fully+0.013
ous+0.013
ting+0.012
 Dolphin+0.012
Neg logit contributions
house-0.018
packs-0.016
tr-0.016
oc-0.015
 fort-0.015
Token little
Token ID1310
Feature activation+9.31e-02

Pos logit contributions
ter+0.038
fully+0.037
ous+0.037
ting+0.034
izing+0.033
Neg logit contributions
house-0.055
 fort-0.051
tr-0.048
oc-0.046
 handcuffs-0.044
Token boy
Token ID2933
Feature activation-6.48e-03

Pos logit contributions
ter+0.180
ous+0.178
fully+0.177
ting+0.165
izing+0.165
Neg logit contributions
house-0.148
 fort-0.125
packs-0.121
oc-0.115
tr-0.112
Token decided
Token ID3066
Feature activation-4.49e-02

Pos logit contributions
house+0.010
packs+0.009
tr+0.009
 fort+0.009
ringe+0.009
Neg logit contributions
ter-0.013
fully-0.011
ous-0.011
ting-0.011
 Dolphin-0.010
Token to
Token ID284
Feature activation-2.34e-02

Pos logit contributions
house+0.076
 fort+0.064
packs+0.063
tr+0.063
 handcuffs+0.063
Neg logit contributions
ter-0.082
ancer-0.073
ous-0.071
 Dolphin-0.071
fully-0.070
Token explore
Token ID7301
Feature activation+7.63e-02

Pos logit contributions
house+0.038
packs+0.033
tr+0.032
 fort+0.031
oc+0.030
Neg logit contributions
ter-0.045
fully-0.040
ous-0.039
ting-0.038
 Dolphin-0.038
Token some
Token ID617
Feature activation-1.20e-01

Pos logit contributions
ter+0.134
ous+0.128
ancer+0.123
 Dolphin+0.120
iddy+0.112
Neg logit contributions
house-0.134
 fort-0.119
packs-0.117
oc-0.114
 handcuffs-0.109
Token other
Token ID584
Feature activation-8.52e-03

Pos logit contributions
house+0.180
packs+0.155
 fort+0.149
tr+0.145
oc+0.142
Neg logit contributions
ter-0.250
fully-0.222
ous-0.221
 Dolphin-0.214
ting-0.211
Token places
Token ID4113
Feature activation-1.49e-02

Pos logit contributions
house+0.016
 fort+0.015
packs+0.014
oc+0.014
 bars+0.013
Neg logit contributions
ter-0.014
ous-0.013
fully-0.013
 Dolphin-0.012
rus-0.012
Token.
Token ID13
Feature activation+1.99e-03

Pos logit contributions
house+0.028
packs+0.025
 fort+0.023
tr+0.023
oc+0.022
Neg logit contributions
ter-0.025
ous-0.022
 Dolphin-0.021
fully-0.021
izing-0.021
Token showed
Token ID3751
Feature activation-4.20e-02

Pos logit contributions
ter+0.106
fully+0.093
ous+0.092
ting+0.091
 Dolphin+0.090
Neg logit contributions
house-0.079
packs-0.073
tr-0.068
 fort-0.066
ringe-0.065
Token the
Token ID262
Feature activation+3.00e-02

Pos logit contributions
house+0.086
packs+0.078
tr+0.076
 fort+0.074
oc+0.074
Neg logit contributions
ter-0.064
fully-0.054
ous-0.052
ting-0.051
 Dolphin-0.051
Token wine
Token ID8237
Feature activation+5.14e-02

Pos logit contributions
ter+0.056
fully+0.049
ous+0.048
ting+0.047
 Dolphin+0.047
Neg logit contributions
house-0.051
packs-0.045
tr-0.043
 fort-0.042
oc-0.041
Token to
Token ID284
Feature activation-3.12e-02

Pos logit contributions
ter+0.081
fully+0.069
ous+0.068
 Dolphin+0.066
ting+0.066
Neg logit contributions
house-0.102
packs-0.090
tr-0.088
 fort-0.088
oc-0.083
Token his
Token ID465
Feature activation-6.66e-02

Pos logit contributions
house+0.059
packs+0.053
tr+0.051
 fort+0.050
oc+0.049
Neg logit contributions
ter-0.053
fully-0.046
ous-0.044
ting-0.043
 Dolphin-0.043
Token mom
Token ID1995
Feature activation-1.41e-01

Pos logit contributions
packs+0.126
ringe+0.125
house+0.125
tr+0.121
atters+0.121
Neg logit contributions
ter-0.091
 young-0.076
 Dolphin-0.072
 Joey-0.070
 Hop-0.069
Token,
Token ID11
Feature activation+2.23e-02

Pos logit contributions
house+0.288
packs+0.262
tr+0.254
 fort+0.250
oc+0.245
Neg logit contributions
ter-0.214
fully-0.180
ous-0.176
ting-0.171
 Dolphin-0.170
Token she
Token ID673
Feature activation-4.00e-02

Pos logit contributions
ter+0.034
fully+0.029
ous+0.027
ting+0.027
 Dolphin+0.027
Neg logit contributions
house-0.045
packs-0.042
tr-0.041
oc-0.040
 fort-0.040
Token was
Token ID373
Feature activation-3.92e-02

Pos logit contributions
house+0.070
packs+0.062
tr+0.060
 fort+0.059
oc+0.057
Neg logit contributions
ter-0.073
fully-0.063
ous-0.062
ting-0.060
 Dolphin-0.060
Token very
Token ID845
Feature activation-4.24e-02

Pos logit contributions
house+0.071
packs+0.059
tr+0.058
 fort+0.056
 handcuffs+0.056
Neg logit contributions
ter-0.069
fully-0.063
ous-0.061
izing-0.060
ting-0.060
Token happy
Token ID3772
Feature activation+8.28e-03

Pos logit contributions
house+0.082
packs+0.073
 fort+0.069
tr+0.068
oc+0.068
Neg logit contributions
ter-0.070
fully-0.061
ting-0.059
ous-0.059
 Dolphin-0.057
Token the
Token ID262
Feature activation+4.14e-02

Pos logit contributions
ter+0.016
fully+0.013
ous+0.013
 Dolphin+0.013
ancer+0.013
Neg logit contributions
house-0.018
packs-0.016
 fort-0.015
tr-0.015
 handcuffs-0.015
Token new
Token ID649
Feature activation+1.09e-02

Pos logit contributions
ter+0.064
fully+0.053
ous+0.052
ting+0.052
 Dolphin+0.050
Neg logit contributions
house-0.082
packs-0.077
oc-0.071
tr-0.071
ringe-0.069
Token man
Token ID582
Feature activation+1.68e-02

Pos logit contributions
ter+0.017
fully+0.016
ya+0.015
ly+0.015
ous+0.015
Neg logit contributions
house-0.019
tr-0.018
oc-0.017
packs-0.017
ris-0.017
Token.
Token ID13
Feature activation+1.71e-03

Pos logit contributions
ter+0.028
fully+0.024
ous+0.024
ting+0.023
 Dolphin+0.023
Neg logit contributions
house-0.032
packs-0.029
tr-0.028
 fort-0.027
ringe-0.027
Token
Token ID198
Feature activation+7.10e-03

Pos logit contributions
ter+0.003
fully+0.003
ous+0.002
ting+0.002
 Dolphin+0.002
Neg logit contributions
house-0.003
packs-0.003
tr-0.003
 fort-0.003
oc-0.003
Token
Token ID198
Feature activation+7.59e-03

Pos logit contributions
ter+0.012
fully+0.010
ous+0.010
ting+0.009
 Dolphin+0.009
Neg logit contributions
house-0.014
packs-0.012
tr-0.012
 fort-0.012
oc-0.012
TokenHer
Token ID9360
Feature activation-4.24e-02

Pos logit contributions
ter+0.014
ous+0.012
fully+0.012
 Dolphin+0.012
ting+0.012
Neg logit contributions
house-0.013
packs-0.011
tr-0.011
 fort-0.011
oc-0.011
Token mom
Token ID1995
Feature activation-1.36e-01

Pos logit contributions
house+0.090
oc+0.088
packs+0.086
ringe+0.083
atters+0.081
Neg logit contributions
 young-0.046
ter-0.045
ous-0.040
 little-0.039
 older-0.038
Token said
Token ID531
Feature activation-1.60e-02

Pos logit contributions
house+0.242
packs+0.227
tr+0.212
 fort+0.211
ringe+0.206
Neg logit contributions
ter-0.240
fully-0.206
ous-0.201
ting-0.198
 Dolphin-0.192
Token,
Token ID11
Feature activation+3.17e-02

Pos logit contributions
house+0.023
 fort+0.017
packs+0.017
tr+0.016
 bars+0.016
Neg logit contributions
ter-0.034
 Dolphin-0.031
ancer-0.031
ting-0.030
izing-0.030
Token "
Token ID366
Feature activation+2.59e-02

Pos logit contributions
ter+0.059
 plunge+0.056
 Dolphin+0.055
ancer+0.055
heart+0.055
Neg logit contributions
house-0.050
tr-0.045
 handcuffs-0.036
 fort-0.036
oc-0.035
Token loud
Token ID7812
Feature activation-5.21e-02

Pos logit contributions
house+0.032
packs+0.031
ringe+0.028
tr+0.028
 handcuffs+0.027
Neg logit contributions
ter-0.028
ous-0.024
fully-0.024
izing-0.023
 Dolphin-0.023
Token noise
Token ID7838
Feature activation+9.60e-02

Pos logit contributions
house+0.104
oc+0.091
 fort+0.090
tr+0.089
packs+0.082
Neg logit contributions
ter-0.084
ancer-0.074
 Dolphin-0.072
fully-0.069
ting-0.068
Token.
Token ID13
Feature activation-4.65e-04

Pos logit contributions
ter+0.170
fully+0.147
ous+0.144
 Dolphin+0.141
ting+0.141
Neg logit contributions
house-0.172
packs-0.153
tr-0.148
 fort-0.145
oc-0.142
Token
Token ID198
Feature activation+5.38e-03

Pos logit contributions
house+0.001
packs+0.001
tr+0.001
oc+0.001
 handcuffs+0.001
Neg logit contributions
ter-0.001
fully-0.001
ous-0.001
izing-0.001
ting-0.001
Token
Token ID198
Feature activation+5.95e-03

Pos logit contributions
ter+0.008
fully+0.007
ous+0.007
 Dolphin+0.007
ting+0.007
Neg logit contributions
house-0.011
packs-0.010
tr-0.009
 fort-0.009
oc-0.009
TokenS
Token ID50
Feature activation-1.50e-01

Pos logit contributions
ter+0.012
fully+0.010
ous+0.010
ting+0.010
 Dolphin+0.010
Neg logit contributions
house-0.010
packs-0.008
tr-0.008
 fort-0.008
ringe-0.008
Tokenara
Token ID3301
Feature activation-4.39e-02

Pos logit contributions
house+0.181
packs+0.152
 fort+0.140
tr+0.139
oc+0.132
Neg logit contributions
ter-0.360
fully-0.328
ous-0.322
 Dolphin-0.316
ting-0.313
Token came
Token ID1625
Feature activation+2.82e-02

Pos logit contributions
house+0.080
packs+0.070
tr+0.068
 fort+0.067
oc+0.066
Neg logit contributions
ter-0.078
fully-0.067
ous-0.066
 Dolphin-0.064
ting-0.064
Token back
Token ID736
Feature activation-6.48e-02

Pos logit contributions
ter+0.053
izing+0.050
ous+0.048
redible+0.046
ancer+0.046
Neg logit contributions
house-0.046
packs-0.044
 fort-0.038
tr-0.036
 handcuffs-0.035
Token and
Token ID290
Feature activation+1.37e-02

Pos logit contributions
house+0.133
packs+0.121
tr+0.119
 fort+0.116
ringe+0.115
Neg logit contributions
ter-0.097
fully-0.082
ous-0.080
ting-0.078
 Dolphin-0.076
Token saw
Token ID2497
Feature activation-4.56e-02

Pos logit contributions
ter+0.027
fully+0.024
ous+0.023
ting+0.023
 Dolphin+0.023
Neg logit contributions
house-0.022
packs-0.019
tr-0.018
 fort-0.018
oc-0.017
Token bag
Token ID6131
Feature activation-1.53e-01

Pos logit contributions
ter+0.086
 plunge+0.070
 young+0.070
ous+0.070
redible+0.070
Neg logit contributions
house-0.062
packs-0.061
oc-0.051
ringe-0.051
ris-0.049
Token and
Token ID290
Feature activation+5.78e-03

Pos logit contributions
house+0.329
packs+0.308
ringe+0.305
tr+0.303
oc+0.302
Neg logit contributions
ter-0.201
fully-0.180
ous-0.161
ing-0.157
ting-0.155
Token the
Token ID262
Feature activation+4.26e-02

Pos logit contributions
ter+0.011
fully+0.009
ting+0.009
 Dolphin+0.009
ous+0.009
Neg logit contributions
house-0.010
packs-0.009
tr-0.008
 fort-0.008
oc-0.008
Token pear
Token ID25286
Feature activation-3.31e-02

Pos logit contributions
ter+0.084
fully+0.067
ous+0.066
 Dolphin+0.066
 Hop+0.065
Neg logit contributions
house-0.067
packs-0.065
ringe-0.057
oc-0.056
tr-0.056
Token was
Token ID373
Feature activation-3.61e-02

Pos logit contributions
ringe+0.055
packs+0.055
house+0.055
oc+0.053
ris+0.052
Neg logit contributions
ter-0.057
fully-0.049
ing-0.047
izing-0.045
ly-0.045
Token now
Token ID783
Feature activation-1.45e-01

Pos logit contributions
house+0.071
oc+0.068
packs+0.067
tr+0.065
ringe+0.064
Neg logit contributions
ter-0.056
fully-0.047
ous-0.044
ble-0.043
bling-0.042
Token ripe
Token ID29036
Feature activation+2.78e-02

Pos logit contributions
house+0.271
packs+0.245
tr+0.242
oc+0.242
ringe+0.233
Neg logit contributions
ter-0.250
fully-0.212
ous-0.203
 Dolphin-0.195
izing-0.194
Token and
Token ID290
Feature activation+6.54e-03

Pos logit contributions
ter+0.044
fully+0.037
ous+0.036
ting+0.036
 Dolphin+0.035
Neg logit contributions
house-0.055
packs-0.050
tr-0.049
 fort-0.047
ringe-0.047
Token delicious
Token ID12625
Feature activation+8.23e-04

Pos logit contributions
ter+0.011
fully+0.009
ous+0.009
izing+0.009
ting+0.009
Neg logit contributions
house-0.012
packs-0.012
tr-0.011
oc-0.011
ringe-0.011
Token.
Token ID13
Feature activation-6.49e-03

Pos logit contributions
ter+0.001
fully+0.001
ous+0.001
ting+0.001
izing+0.001
Neg logit contributions
house-0.002
packs-0.001
tr-0.001
ringe-0.001
oc-0.001
Token Lily
Token ID20037
Feature activation-1.25e-01

Pos logit contributions
house+0.012
packs+0.011
 fort+0.011
ringe+0.011
tr+0.011
Neg logit contributions
ter-0.010
 Dolphin-0.009
fully-0.009
ous-0.009
 Hop-0.009
Token kind
Token ID1611
Feature activation+5.20e-02

Pos logit contributions
house+0.036
 fort+0.033
oc+0.032
tr+0.031
packs+0.030
Neg logit contributions
ter-0.022
fully-0.020
ous-0.019
ting-0.019
izing-0.017
Token and
Token ID290
Feature activation+1.21e-02

Pos logit contributions
ter+0.092
izing+0.087
fully+0.085
ting+0.085
ivated+0.084
Neg logit contributions
house-0.087
 fort-0.077
oc-0.075
tr-0.072
packs-0.068
Token compassionate
Token ID32533
Feature activation+7.90e-02

Pos logit contributions
ter+0.021
fully+0.019
ting+0.019
ous+0.018
izing+0.017
Neg logit contributions
house-0.021
 fort-0.018
packs-0.018
tr-0.018
oc-0.017
Token lady
Token ID10846
Feature activation-5.78e-02

Pos logit contributions
ter+0.136
izing+0.129
fully+0.126
ivated+0.126
ting+0.126
Neg logit contributions
house-0.134
 fort-0.120
tr-0.115
oc-0.111
packs-0.109
Token.
Token ID13
Feature activation+1.36e-03

Pos logit contributions
house+0.092
 fort+0.085
 handcuffs+0.084
tr+0.084
ringe+0.083
Neg logit contributions
ter-0.110
fully-0.093
ous-0.090
ting-0.088
ing-0.086
Token One
Token ID1881
Feature activation-1.62e-01

Pos logit contributions
ter+0.002
fully+0.002
ous+0.002
 Dolphin+0.002
ting+0.002
Neg logit contributions
house-0.002
packs-0.002
tr-0.002
 fort-0.002
oc-0.002
Token day
Token ID1110
Feature activation-6.20e-02

Pos logit contributions
house+0.221
tr+0.180
packs+0.179
 fort+0.173
oc+0.168
Neg logit contributions
ter-0.361
fully-0.325
ous-0.321
ting-0.318
izing-0.315
Token she
Token ID673
Feature activation-3.84e-02

Pos logit contributions
house+0.097
packs+0.081
tr+0.074
 fort+0.073
oc+0.072
Neg logit contributions
ter-0.123
fully-0.111
ous-0.109
izing-0.109
ping-0.109
Token got
Token ID1392
Feature activation-8.04e-02

Pos logit contributions
house+0.060
packs+0.055
tr+0.053
ringe+0.052
 fort+0.052
Neg logit contributions
ter-0.075
fully-0.064
 Dolphin-0.062
ous-0.061
ting-0.061
Token a
Token ID257
Feature activation-1.60e-02

Pos logit contributions
house+0.116
packs+0.103
 handcuffs+0.092
ringe+0.091
tr+0.090
Neg logit contributions
ous-0.160
ter-0.159
fully-0.152
izing-0.150
 Dolphin-0.146
Token strange
Token ID6283
Feature activation+3.77e-02

Pos logit contributions
house+0.026
packs+0.024
tr+0.022
 fort+0.022
oc+0.022
Neg logit contributions
ter-0.030
fully-0.026
ous-0.026
 Dolphin-0.025
ting-0.025
Token forever
Token ID8097
Feature activation-3.32e-02

Pos logit contributions
house+0.017
packs+0.015
 fort+0.014
tr+0.014
oc+0.014
Neg logit contributions
ter-0.019
ous-0.016
 Dolphin-0.016
ting-0.016
fully-0.016
Token and
Token ID290
Feature activation+1.06e-02

Pos logit contributions
house+0.056
packs+0.050
 fort+0.046
oc+0.045
tr+0.045
Neg logit contributions
ter-0.062
fully-0.054
 Dolphin-0.054
ous-0.054
ancer-0.053
Token ever
Token ID1683
Feature activation+6.50e-02

Pos logit contributions
ter+0.021
fully+0.018
ous+0.018
ting+0.018
 Dolphin+0.017
Neg logit contributions
house-0.017
packs-0.015
tr-0.014
 fort-0.014
ringe-0.014
Token.
Token ID13
Feature activation-2.17e-03

Pos logit contributions
ter+0.116
fully+0.103
 Dolphin+0.101
ous+0.100
izing+0.097
Neg logit contributions
house-0.114
packs-0.098
 fort-0.095
oc-0.094
tr-0.093
Token<|endoftext|>
Token ID50256
Feature activation+4.23e-02

Pos logit contributions
house+0.004
packs+0.004
tr+0.003
oc+0.003
 fort+0.003
Neg logit contributions
ter-0.004
fully-0.003
ous-0.003
ting-0.003
izing-0.003
TokenOne
Token ID3198
Feature activation-1.69e-01

Pos logit contributions
ter+0.080
fully+0.072
ous+0.071
izing+0.070
ya+0.067
Neg logit contributions
house-0.073
packs-0.065
tr-0.060
oc-0.059
 fort-0.059
Token day
Token ID1110
Feature activation-6.06e-02

Pos logit contributions
house+0.226
packs+0.191
 fort+0.177
tr+0.175
oc+0.164
Neg logit contributions
ter-0.389
fully-0.345
ting-0.342
ous-0.338
izing-0.333
Token there
Token ID612
Feature activation-9.17e-03

Pos logit contributions
house+0.072
packs+0.066
 fort+0.050
ossible+0.047
tr+0.046
Neg logit contributions
ter-0.140
ting-0.129
ping-0.126
izing-0.125
ous-0.123
Token was
Token ID373
Feature activation-3.32e-02

Pos logit contributions
house+0.016
packs+0.014
tr+0.014
 fort+0.013
ringe+0.013
Neg logit contributions
ter-0.017
fully-0.014
ous-0.014
 Dolphin-0.014
ting-0.014
Token a
Token ID257
Feature activation-1.73e-02

Pos logit contributions
house+0.060
packs+0.052
tr+0.049
 fort+0.048
 handcuffs+0.045
Neg logit contributions
ter-0.060
ous-0.055
ting-0.052
fully-0.052
izing-0.052
Token polite
Token ID23507
Feature activation+6.92e-02

Pos logit contributions
house+0.038
 fort+0.033
tr+0.033
packs+0.033
oc+0.032
Neg logit contributions
ter-0.024
fully-0.021
ous-0.020
ting-0.020
izing-0.018
Token't
Token ID470
Feature activation+5.28e-02

Pos logit contributions
ter+0.145
fully+0.120
ous+0.116
ting+0.115
 Dolphin+0.114
Neg logit contributions
house-0.208
packs-0.188
tr-0.184
 fort-0.179
oc-0.175
Token have
Token ID423
Feature activation-7.54e-02

Pos logit contributions
ter+0.102
fully+0.090
ous+0.085
ancer+0.084
ting+0.083
Neg logit contributions
house-0.082
packs-0.074
 fort-0.073
tr-0.072
oc-0.071
Token any
Token ID597
Feature activation-3.56e-02

Pos logit contributions
house+0.157
packs+0.141
oc+0.140
tr+0.138
 fort+0.136
Neg logit contributions
ter-0.110
fully-0.091
ous-0.090
ting-0.086
 Dolphin-0.085
Token friends
Token ID2460
Feature activation+3.82e-02

Pos logit contributions
house+0.063
tr+0.056
oc+0.056
packs+0.054
ringe+0.053
Neg logit contributions
ter-0.062
fully-0.053
ous-0.051
ting-0.050
ancer-0.048
Token.
Token ID13
Feature activation-4.64e-03

Pos logit contributions
ter+0.058
fully+0.049
ous+0.048
ting+0.047
 Dolphin+0.045
Neg logit contributions
house-0.077
packs-0.070
tr-0.069
 fort-0.068
oc-0.067
Token One
Token ID1881
Feature activation-1.64e-01

Pos logit contributions
house+0.008
packs+0.007
tr+0.007
 fort+0.006
oc+0.006
Neg logit contributions
ter-0.009
fully-0.008
ous-0.008
ting-0.007
izing-0.007
Token day
Token ID1110
Feature activation-6.16e-02

Pos logit contributions
house+0.243
packs+0.213
tr+0.203
 fort+0.199
oc+0.193
Neg logit contributions
ter-0.343
fully-0.302
ous-0.298
 Dolphin-0.291
ting-0.291
Token she
Token ID673
Feature activation-3.83e-02

Pos logit contributions
house+0.106
packs+0.092
tr+0.087
 fort+0.083
oc+0.080
Neg logit contributions
ter-0.116
fully-0.101
 Dolphin-0.100
ous-0.099
izing-0.098
Token decided
Token ID3066
Feature activation-4.94e-02

Pos logit contributions
house+0.065
tr+0.060
packs+0.059
ringe+0.059
 handcuffs+0.058
Neg logit contributions
ter-0.068
 Dolphin-0.056
fully-0.056
ting-0.055
 Nathan-0.054
Token to
Token ID284
Feature activation-2.75e-02

Pos logit contributions
house+0.060
packs+0.046
tr+0.045
 fort+0.043
 bars+0.040
Neg logit contributions
ter-0.112
ancer-0.107
iful-0.105
 Dolphin-0.103
bling-0.101
Token find
Token ID1064
Feature activation-3.80e-02

Pos logit contributions
house+0.042
packs+0.037
tr+0.036
 fort+0.035
oc+0.035
Neg logit contributions
ter-0.055
fully-0.048
ous-0.047
 Dolphin-0.047
ting-0.046