TOP ACTIVATIONS
MAX = 0.094

left this shark in the bath tub ?" Sara and Ben
As they buried the ham ster , Tim my felt
One sunny day , the oy ster saw a shiny pearl
little oy ster . This oy ster was very happy .
they needed to put the ham ster in a special place
understood . He opened the oy ster carefully and looked inside
One day , the oy ster met a little fish
interesting happening on the books he lf near the river .
One day , the graceful oy ster had a special treat
he decided to remove the seaw eed from the water so
, he noticed that his toy car was missing a wheel
them to go . The careless jelly fish said goodbye to
They decide to leave the oy ster alone and play with
play and saw Tim 's broken sp ade . Tom said
One day , the oy ster went swimming and it
and her mom left the oy ster at the beach ,
my told him that the ham ster would always be with
he felt better . The green oct opus and his friends
to get out of the bath tub now . The water
with ." The white oct opus felt bad for

MIN ACTIVATIONS
MIN = -0.089

it before it was gone . She chased it and chased
for ways to help others . One day
was too dangerous for her . K aren
a big and beautiful nation . Every day she would go
little girl named Mad eline . One day , she was
catch it in her hands . When she
lunch ," her mom shouted . Alice took
she heard a loud noise . Mad eline panicked and looked
to bake cookies and cakes . One day while baking ,
she went on a search . She walked around the garden
mum wouldn 't permit it . She said the garden was
eline who was very thoughtful . She loved to do nice
was a girl named Karen . She wore a cozy scarf
a voice said " Mine !". A small girl with curly
was time to go home . So Karen and her mom
big fan of fire men . She often asked her daddy
veil on the kitchen table . She asked her mom ,
very happy and playful girl . She loved to play with
the hallway seemed less scary . She t ipt
Lucy went for a walk . She saw lots of interesting

INTERVAL 0.049 - 0.094
CONTAINS 0.208%

back to his family . While he was swimming , he
ocean . One day , Nem o saw a plastic bag
ery became one of her favourite snacks .
careful not to touch the hot stove . They
big sea . In the sea , there was a small

INTERVAL 0.003 - 0.049
CONTAINS 42.941%

and smiles . <|endoftext|> S ara liked to play with her
- ray that he shone on the tree . With the
Sam and Mia are friends . They like to play outside
oven and took out the cake . She checked if Sara
told his mom about the monster he found . His mom

INTERVAL -0.043 - 0.003
CONTAINS 56.464%

. Anna heard him cry and came to see
green and brown , like a leaf ," Lily said ,
cat a toy mouse . The cat seemed to like it
again tomorrow . <|endoftext|> Once upon a time , there was
very bright star that was cheerful and shiny .

INTERVAL -0.089 - -0.043
CONTAINS 0.387%

." Ben and Sam looked at Bob o .
and picked up her steak . She took it to the
the vendor and asked for a red balloon . The vendor
that was filled with toys . As Lucy walked through the
wanted to get some of the cher ries but she couldn
Token left
Token ID1364
Feature activation+3.32e-02

Pos logit contributions
ble+0.001
bled+0.000
ster+0.000
balls+0.000
 fin+0.000
Neg logit contributions
rell-0.000
 Karen-0.000
aren-0.000
ird-0.000
Ell-0.000
Token this
Token ID428
Feature activation+6.12e-02

Pos logit contributions
ble+0.070
bled+0.064
balls+0.061
ster+0.058
 fin+0.057
Neg logit contributions
ird-0.060
 Karen-0.055
aren-0.055
rell-0.054
 Stella-0.052
Token shark
Token ID21027
Feature activation+6.21e-02

Pos logit contributions
bled+0.091
ble+0.085
balls+0.077
ster+0.069
 Miles+0.067
Neg logit contributions
ird-0.130
hell-0.121
rell-0.119
he-0.117
chair-0.114
Token in
Token ID287
Feature activation+2.84e-02

Pos logit contributions
ble+0.143
bled+0.143
balls+0.131
 fin+0.119
ster+0.117
Neg logit contributions
ird-0.095
he-0.075
old-0.074
rell-0.071
aren-0.070
Token the
Token ID262
Feature activation+4.22e-02

Pos logit contributions
ble+0.063
bled+0.059
balls+0.056
ster+0.053
 fin+0.050
Neg logit contributions
ird-0.050
 Karen-0.045
rell-0.045
aren-0.044
 Ellen-0.043
Token bath
Token ID7837
Feature activation+9.42e-02

Pos logit contributions
ble+0.110
bled+0.107
balls+0.097
ster+0.093
 Miles+0.088
Neg logit contributions
rell-0.052
ird-0.052
 Eve-0.046
chair-0.040
he-0.039
Tokentub
Token ID37995
Feature activation+5.23e-02

Pos logit contributions
ble+0.135
bled+0.130
balls+0.115
 fin+0.106
ster+0.105
Neg logit contributions
ird-0.220
rell-0.199
he-0.191
 Eve-0.185
ives-0.181
Token?"
Token ID1701
Feature activation+4.89e-02

Pos logit contributions
ble+0.118
bled+0.115
balls+0.108
ster+0.101
 fin+0.099
Neg logit contributions
ird-0.086
rell-0.072
aren-0.070
 Karen-0.069
old-0.066
Token Sara
Token ID24799
Feature activation-1.90e-04

Pos logit contributions
bled+0.139
ble+0.137
balls+0.127
ster+0.121
bing+0.114
Neg logit contributions
ird-0.049
 Karen-0.047
 Stella-0.042
 Eve-0.041
aren-0.040
Token and
Token ID290
Feature activation+4.39e-02

Pos logit contributions
ird+0.000
rell+0.000
aren+0.000
 Karen+0.000
 Ellen+0.000
Neg logit contributions
ble-0.000
bled-0.000
balls-0.000
ster-0.000
 fin-0.000
Token Ben
Token ID3932
Feature activation-2.47e-02

Pos logit contributions
ble+0.089
bled+0.085
balls+0.078
ster+0.072
 fin+0.069
Neg logit contributions
ird-0.079
rell-0.076
 Karen-0.076
aren-0.073
 Kathy-0.072
Token
Token ID198
Feature activation+5.34e-03

Pos logit contributions
ird+0.027
rell+0.025
aren+0.024
 Karen+0.023
 Ellen+0.023
Neg logit contributions
ble-0.028
bled-0.024
balls-0.023
emo-0.022
 fin-0.021
TokenAs
Token ID1722
Feature activation+1.90e-02

Pos logit contributions
ble+0.013
bled+0.011
ster+0.011
balls+0.011
 fin+0.011
Neg logit contributions
ird-0.009
 Karen-0.008
rell-0.008
aren-0.008
Ell-0.007
Token they
Token ID484
Feature activation+1.05e-02

Pos logit contributions
ble+0.042
bled+0.038
ster+0.037
balls+0.036
 fin+0.035
Neg logit contributions
 Karen-0.033
ird-0.032
aren-0.031
rell-0.030
 Stella-0.029
Token buried
Token ID11694
Feature activation+2.96e-02

Pos logit contributions
ble+0.029
bled+0.026
balls+0.025
ster+0.025
 fin+0.024
Neg logit contributions
ird-0.013
rell-0.012
 Karen-0.011
aren-0.011
hell-0.010
Token the
Token ID262
Feature activation+4.41e-02

Pos logit contributions
ble+0.072
bled+0.065
ster+0.065
balls+0.063
 fin+0.060
Neg logit contributions
 Karen-0.041
 Ellie-0.037
rell-0.037
 Stella-0.037
 Ellen-0.036
Token ham
Token ID8891
Feature activation+9.08e-02

Pos logit contributions
bled+0.092
ble+0.091
balls+0.078
 Miles+0.072
 fin+0.072
Neg logit contributions
hell-0.070
ird-0.065
rell-0.065
 aback-0.061
 Stella-0.061
Tokenster
Token ID1706
Feature activation+4.18e-02

Pos logit contributions
bled+0.082
ble+0.078
 surfing+0.073
 fin+0.065
 skating+0.064
Neg logit contributions
hell-0.213
ives-0.203
ird-0.200
 bars-0.195
hel-0.186
Token,
Token ID11
Feature activation+8.13e-03

Pos logit contributions
ble+0.105
bled+0.100
balls+0.093
 fin+0.090
emo+0.090
Neg logit contributions
ird-0.051
rell-0.047
hell-0.043
 Karen-0.042
chair-0.040
Token Tim
Token ID5045
Feature activation+1.11e-02

Pos logit contributions
ble+0.020
bled+0.018
ster+0.018
balls+0.017
 fin+0.017
Neg logit contributions
 Karen-0.012
ird-0.012
aren-0.011
rell-0.011
 Stella-0.011
Tokenmy
Token ID1820
Feature activation+4.23e-02

Pos logit contributions
ble+0.032
bled+0.029
ster+0.029
balls+0.028
 fin+0.028
Neg logit contributions
ird-0.012
 Karen-0.011
aren-0.010
rell-0.010
 Stella-0.009
Token felt
Token ID2936
Feature activation+1.50e-02

Pos logit contributions
ble+0.108
bled+0.101
balls+0.096
ster+0.088
 fin+0.087
Neg logit contributions
ird-0.058
rell-0.048
old-0.046
 Karen-0.046
aren-0.045
TokenOne
Token ID3198
Feature activation+3.61e-02

Pos logit contributions
ble+0.017
bled+0.015
balls+0.015
ster+0.014
 fin+0.014
Neg logit contributions
ird-0.009
rell-0.008
aren-0.008
 Karen-0.008
 Ellen-0.007
Token sunny
Token ID27737
Feature activation+6.45e-02

Pos logit contributions
bled+0.103
balls+0.098
ble+0.097
ster+0.086
bing+0.085
Neg logit contributions
old-0.033
ird-0.030
he-0.029
chair-0.029
 Karen-0.029
Token day
Token ID1110
Feature activation+2.72e-02

Pos logit contributions
bled+0.180
balls+0.179
ble+0.173
ster+0.155
 fin+0.149
Neg logit contributions
aren-0.051
rell-0.048
ird-0.046
chair-0.045
 Karen-0.045
Token,
Token ID11
Feature activation+1.28e-02

Pos logit contributions
ble+0.080
bled+0.075
balls+0.071
ster+0.071
 fin+0.069
Neg logit contributions
ird-0.026
 Karen-0.023
rell-0.022
aren-0.020
 Stella-0.020
Token the
Token ID262
Feature activation+5.44e-02

Pos logit contributions
ble+0.034
bled+0.032
ster+0.031
balls+0.030
 fin+0.029
Neg logit contributions
 Karen-0.015
 Stella-0.014
ird-0.014
 Alice-0.013
 Ellie-0.013
Token oy
Token ID35104
Feature activation+9.01e-02

Pos logit contributions
bled+0.121
balls+0.118
ble+0.112
bing+0.100
zes+0.097
Neg logit contributions
hell-0.065
 Eve-0.061
aren-0.059
 Mad-0.059
 Stella-0.059
Tokenster
Token ID1706
Feature activation+6.07e-02

Pos logit contributions
bled+0.096
balls+0.080
bing+0.075
sing+0.075
ble+0.075
Neg logit contributions
ird-0.198
hell-0.188
old-0.183
ives-0.181
mie-0.178
Token saw
Token ID2497
Feature activation+3.40e-02

Pos logit contributions
ble+0.166
bled+0.159
balls+0.150
 fin+0.137
ster+0.135
Neg logit contributions
ird-0.073
old-0.059
rell-0.059
aren-0.056
 Karen-0.054
Token a
Token ID257
Feature activation+2.94e-02

Pos logit contributions
ble+0.087
balls+0.083
bled+0.081
ster+0.079
 fin+0.075
Neg logit contributions
 Karen-0.034
ird-0.034
rell-0.032
old-0.032
 Stella-0.030
Token shiny
Token ID22441
Feature activation+3.87e-02

Pos logit contributions
bled+0.079
ble+0.074
balls+0.071
bing+0.065
ster+0.062
Neg logit contributions
ird-0.031
 Karen-0.029
hell-0.028
aren-0.028
ha-0.027
Token pearl
Token ID43836
Feature activation+4.69e-02

Pos logit contributions
bled+0.077
balls+0.076
ble+0.068
 Miles+0.062
bing+0.058
Neg logit contributions
hell-0.063
ird-0.063
aren-0.059
rell-0.059
 Karen-0.059
Token little
Token ID1310
Feature activation+3.38e-02

Pos logit contributions
bled+0.074
ble+0.069
balls+0.066
bing+0.060
urg+0.057
Neg logit contributions
ird-0.026
 Karen-0.024
 Stella-0.023
aren-0.022
ha-0.021
Token oy
Token ID35104
Feature activation+8.05e-02

Pos logit contributions
bled+0.065
ble+0.064
balls+0.057
ster+0.048
bing+0.047
Neg logit contributions
ird-0.062
aren-0.060
 Karen-0.060
 Stella-0.057
rell-0.057
Tokenster
Token ID1706
Feature activation+3.33e-02

Pos logit contributions
bled+0.091
ble+0.065
bing+0.059
balls+0.057
zes+0.049
Neg logit contributions
ird-0.195
hell-0.187
aren-0.170
old-0.167
ha-0.165
Token.
Token ID13
Feature activation-3.73e-03

Pos logit contributions
bled+0.099
ble+0.094
balls+0.090
 fin+0.080
 Miles+0.080
Neg logit contributions
ird-0.030
hell-0.022
old-0.022
chair-0.021
 Karen-0.020
Token This
Token ID770
Feature activation+7.36e-02

Pos logit contributions
ird+0.006
 Karen+0.005
rell+0.005
aren+0.005
 Ellen+0.005
Neg logit contributions
ble-0.009
bled-0.008
balls-0.008
ster-0.008
 fin-0.007
Token oy
Token ID35104
Feature activation+8.96e-02

Pos logit contributions
bled+0.138
balls+0.132
ble+0.129
bing+0.113
bling+0.102
Neg logit contributions
hell-0.126
ird-0.123
ha-0.123
aren-0.119
rell-0.118
Tokenster
Token ID1706
Feature activation+4.24e-02

Pos logit contributions
bled+0.087
ble+0.068
balls+0.065
bing+0.060
 fin+0.059
Neg logit contributions
ird-0.217
hell-0.216
aren-0.203
old-0.200
mie-0.195
Token was
Token ID373
Feature activation+2.49e-02

Pos logit contributions
ble+0.105
bled+0.104
balls+0.096
 fin+0.089
 diving+0.088
Neg logit contributions
ird-0.055
hell-0.049
chair-0.047
old-0.047
 Karen-0.047
Token very
Token ID845
Feature activation+4.64e-02

Pos logit contributions
ble+0.064
balls+0.063
ster+0.063
bled+0.062
bing+0.059
Neg logit contributions
 Ellie-0.021
 seen-0.020
 Karen-0.020
 Stella-0.019
rell-0.017
Token happy
Token ID3772
Feature activation+3.44e-02

Pos logit contributions
bled+0.121
ble+0.120
balls+0.115
bing+0.110
ster+0.109
Neg logit contributions
hell-0.039
ird-0.039
old-0.037
 know-0.035
 Karen-0.035
Token.
Token ID13
Feature activation-3.27e-03

Pos logit contributions
bled+0.103
ble+0.101
balls+0.090
ster+0.088
 fin+0.085
Neg logit contributions
ird-0.024
 Karen-0.022
chair-0.020
 Ellie-0.020
rell-0.019
Token they
Token ID484
Feature activation-1.45e-02

Pos logit contributions
ble+0.051
bled+0.048
ster+0.044
balls+0.043
 fin+0.042
Neg logit contributions
ird-0.034
 Karen-0.034
rell-0.032
 Ellie-0.032
 Ellen-0.031
Token needed
Token ID2622
Feature activation+3.50e-03

Pos logit contributions
ird+0.026
rell+0.024
aren+0.024
 Karen+0.024
 Ellen+0.022
Neg logit contributions
ble-0.032
bled-0.029
balls-0.027
ster-0.026
 fin-0.026
Token to
Token ID284
Feature activation+1.70e-02

Pos logit contributions
ble+0.008
bled+0.007
ster+0.007
balls+0.007
 fin+0.006
Neg logit contributions
ird-0.006
 Karen-0.006
rell-0.005
aren-0.005
 Stella-0.005
Token put
Token ID1234
Feature activation+1.13e-02

Pos logit contributions
ble+0.036
bled+0.034
balls+0.033
ster+0.031
 fin+0.029
Neg logit contributions
ird-0.031
 Karen-0.027
rell-0.027
aren-0.026
 Stella-0.025
Token the
Token ID262
Feature activation+5.05e-02

Pos logit contributions
ble+0.024
bled+0.022
balls+0.021
ster+0.020
 fin+0.020
Neg logit contributions
ird-0.020
rell-0.019
 Karen-0.019
aren-0.018
 Stella-0.017
Token ham
Token ID8891
Feature activation+8.89e-02

Pos logit contributions
ble+0.103
bled+0.102
balls+0.087
bing+0.079
 diving+0.077
Neg logit contributions
hell-0.081
ird-0.078
rell-0.074
 Stella-0.073
 aback-0.073
Tokenster
Token ID1706
Feature activation+4.31e-02

Pos logit contributions
bled+0.080
ble+0.077
 diving+0.065
 surfing+0.059
 Miles+0.057
Neg logit contributions
hell-0.224
ird-0.221
ives-0.211
 bars-0.195
he-0.194
Token in
Token ID287
Feature activation-2.31e-04

Pos logit contributions
ble+0.084
bled+0.077
balls+0.072
 fin+0.068
ster+0.067
Neg logit contributions
ird-0.081
rell-0.073
hell-0.072
 Karen-0.068
aren-0.068
Token a
Token ID257
Feature activation-4.08e-03

Pos logit contributions
ird+0.000
rell+0.000
aren+0.000
 Karen+0.000
 Ellen+0.000
Neg logit contributions
ble-0.000
bled-0.000
 fin-0.000
balls-0.000
ster-0.000
Token special
Token ID2041
Feature activation+1.50e-02

Pos logit contributions
ird+0.008
rell+0.007
 Karen+0.007
aren+0.007
 Ellen+0.007
Neg logit contributions
ble-0.008
bled-0.008
balls-0.007
ster-0.007
 fin-0.007
Token place
Token ID1295
Feature activation+4.17e-03

Pos logit contributions
ble+0.032
bled+0.029
balls+0.027
ster+0.026
 fin+0.026
Neg logit contributions
ird-0.028
aren-0.025
rell-0.025
 Karen-0.025
 Ellen-0.024
Token understood
Token ID7247
Feature activation+1.36e-02

Pos logit contributions
ble+0.008
bled+0.007
balls+0.007
ster+0.007
 fin+0.007
Neg logit contributions
ird-0.006
rell-0.006
 Karen-0.006
aren-0.006
 Ellen-0.005
Token.
Token ID13
Feature activation+5.73e-03

Pos logit contributions
ble+0.030
bled+0.027
balls+0.026
ster+0.025
 fin+0.025
Neg logit contributions
ird-0.024
 Karen-0.022
aren-0.022
rell-0.022
 Ellen-0.020
Token He
Token ID679
Feature activation+1.21e-02

Pos logit contributions
ble+0.011
bled+0.010
balls+0.009
ster+0.009
 fin+0.009
Neg logit contributions
ird-0.012
rell-0.011
 Karen-0.011
aren-0.011
 Ellen-0.010
Token opened
Token ID4721
Feature activation+7.12e-03

Pos logit contributions
ble+0.028
bled+0.025
balls+0.024
ster+0.023
 fin+0.023
Neg logit contributions
ird-0.021
rell-0.019
 Karen-0.019
aren-0.019
 Ellen-0.017
Token the
Token ID262
Feature activation+3.56e-02

Pos logit contributions
ble+0.013
bled+0.012
balls+0.011
ster+0.011
 fin+0.010
Neg logit contributions
ird-0.015
rell-0.014
 Karen-0.014
aren-0.014
 Ellen-0.013
Token oy
Token ID35104
Feature activation+8.78e-02

Pos logit contributions
ble+0.057
bled+0.056
balls+0.054
emo+0.042
bing+0.042
Neg logit contributions
ird-0.073
rell-0.072
aren-0.071
hell-0.070
 Karen-0.069
Tokenster
Token ID1706
Feature activation+3.44e-02

Pos logit contributions
bled+0.083
ble+0.071
balls+0.071
 spitting+0.056
 diving+0.055
Neg logit contributions
ird-0.206
hell-0.199
land-0.193
ives-0.191
mie-0.188
Token carefully
Token ID7773
Feature activation+2.65e-03

Pos logit contributions
ble+0.083
bled+0.079
balls+0.074
 fin+0.068
emo+0.068
Neg logit contributions
ird-0.051
rell-0.044
aren-0.043
hell-0.043
 Karen-0.042
Token and
Token ID290
Feature activation+4.89e-03

Pos logit contributions
ble+0.006
bled+0.006
balls+0.005
ster+0.005
 fin+0.005
Neg logit contributions
ird-0.004
rell-0.004
aren-0.004
 Karen-0.004
 Ellen-0.004
Token looked
Token ID3114
Feature activation-3.48e-03

Pos logit contributions
ble+0.012
bled+0.010
balls+0.010
 fin+0.010
ster+0.010
Neg logit contributions
ird-0.008
 Karen-0.007
aren-0.007
rell-0.007
 Stella-0.006
Token inside
Token ID2641
Feature activation+2.93e-02

Pos logit contributions
ird+0.007
rell+0.007
 Karen+0.007
aren+0.007
 Ellen+0.006
Neg logit contributions
ble-0.007
bled-0.006
balls-0.006
ster-0.006
 fin-0.005
Token
Token ID198
Feature activation+6.42e-03

Pos logit contributions
ird+0.036
rell+0.034
aren+0.032
 Karen+0.031
 Ellen+0.030
Neg logit contributions
ble-0.030
balls-0.024
bled-0.024
emo-0.023
 diving-0.022
TokenOne
Token ID3198
Feature activation+3.31e-02

Pos logit contributions
ble+0.017
bled+0.016
balls+0.015
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.008
rell-0.008
 Karen-0.008
aren-0.008
 Ellen-0.007
Token day
Token ID1110
Feature activation+2.31e-02

Pos logit contributions
bled+0.095
balls+0.090
ble+0.089
ster+0.078
bing+0.075
Neg logit contributions
old-0.031
 Karen-0.029
aren-0.029
ird-0.028
chair-0.028
Token,
Token ID11
Feature activation+4.40e-03

Pos logit contributions
ble+0.062
bled+0.057
balls+0.055
ster+0.054
 fin+0.053
Neg logit contributions
ird-0.029
 Karen-0.027
rell-0.026
aren-0.025
 Stella-0.024
Token the
Token ID262
Feature activation+5.07e-02

Pos logit contributions
ble+0.011
bled+0.010
ster+0.010
balls+0.010
 fin+0.009
Neg logit contributions
 Karen-0.006
ird-0.006
 Stella-0.005
 Ellen-0.005
 Ellie-0.005
Token oy
Token ID35104
Feature activation+8.70e-02

Pos logit contributions
bled+0.099
balls+0.099
ble+0.091
bing+0.080
zes+0.077
Neg logit contributions
hell-0.074
 Stella-0.072
 Mad-0.071
aren-0.070
ird-0.069
Tokenster
Token ID1706
Feature activation+5.46e-02

Pos logit contributions
bled+0.086
balls+0.077
ble+0.069
 diving+0.067
sing+0.065
Neg logit contributions
ird-0.194
hell-0.190
mie-0.183
old-0.183
aren-0.182
Token met
Token ID1138
Feature activation+2.20e-02

Pos logit contributions
ble+0.145
bled+0.139
balls+0.133
 fin+0.120
ster+0.119
Neg logit contributions
ird-0.068
rell-0.057
old-0.056
aren-0.055
 Karen-0.055
Token a
Token ID257
Feature activation+1.05e-02

Pos logit contributions
ble+0.049
bled+0.045
balls+0.043
ster+0.041
 fin+0.041
Neg logit contributions
ird-0.038
 Karen-0.035
rell-0.035
aren-0.035
 Ellen-0.032
Token little
Token ID1310
Feature activation+1.94e-02

Pos logit contributions
ble+0.026
bled+0.026
balls+0.023
ster+0.022
 fin+0.021
Neg logit contributions
ird-0.014
 Karen-0.014
aren-0.013
 Stella-0.013
 Ellen-0.012
Token fish
Token ID5916
Feature activation+1.60e-02

Pos logit contributions
bled+0.046
ble+0.045
balls+0.040
ster+0.034
emo+0.034
Neg logit contributions
ird-0.028
 Karen-0.028
aren-0.028
 Stella-0.027
old-0.026
Token interesting
Token ID3499
Feature activation+1.23e-02

Pos logit contributions
 Karen+0.009
ird+0.009
hell+0.009
rell+0.008
aren+0.008
Neg logit contributions
ble-0.018
bled-0.018
ster-0.017
balls-0.016
 fin-0.015
Token happening
Token ID5836
Feature activation+1.91e-02

Pos logit contributions
ble+0.029
bled+0.028
balls+0.026
ster+0.026
 fin+0.025
Neg logit contributions
ird-0.016
 Karen-0.016
hell-0.015
rell-0.015
aren-0.015
Token on
Token ID319
Feature activation+4.32e-02

Pos logit contributions
ble+0.042
bled+0.040
balls+0.040
ster+0.038
 fin+0.036
Neg logit contributions
 Karen-0.027
aren-0.026
 Stella-0.025
rell-0.025
 aback-0.024
Token the
Token ID262
Feature activation+2.29e-02

Pos logit contributions
ble+0.103
bled+0.102
balls+0.099
ster+0.095
 fin+0.086
Neg logit contributions
 Karen-0.058
rell-0.054
 Stella-0.053
ird-0.052
aren-0.052
Token books
Token ID3835
Feature activation+6.03e-03

Pos logit contributions
bled+0.044
ble+0.044
balls+0.038
ster+0.037
emo+0.034
Neg logit contributions
rell-0.042
ird-0.038
hell-0.037
 Eve-0.036
 Karen-0.036
Tokenhe
Token ID258
Feature activation+8.62e-02

Pos logit contributions
ble+0.020
bled+0.018
balls+0.017
ster+0.017
 fin+0.017
Neg logit contributions
ird-0.004
 Karen-0.004
aren-0.004
rell-0.004
 Ellen-0.003
Tokenlf
Token ID1652
Feature activation+2.76e-02

Pos logit contributions
bled+0.182
ble+0.179
ster+0.164
balls+0.162
 hyd+0.158
Neg logit contributions
chair-0.110
ird-0.110
he-0.102
hell-0.102
rell-0.095
Token near
Token ID1474
Feature activation+2.11e-02

Pos logit contributions
ble+0.070
bled+0.064
balls+0.061
ster+0.060
 fin+0.060
Neg logit contributions
ird-0.039
rell-0.035
 Karen-0.035
aren-0.035
 Ellen-0.031
Token the
Token ID262
Feature activation+6.46e-03

Pos logit contributions
ble+0.044
bled+0.041
balls+0.040
ster+0.038
 fin+0.035
Neg logit contributions
ird-0.038
 Karen-0.036
aren-0.035
rell-0.035
 Stella-0.033
Token river
Token ID7850
Feature activation+2.38e-02

Pos logit contributions
ble+0.013
bled+0.012
balls+0.011
ster+0.011
 fin+0.010
Neg logit contributions
ird-0.012
rell-0.012
 Karen-0.011
aren-0.011
hell-0.011
Token.
Token ID13
Feature activation+3.18e-02

Pos logit contributions
ble+0.060
bled+0.055
balls+0.052
ster+0.051
 fin+0.050
Neg logit contributions
ird-0.034
rell-0.031
 Karen-0.031
aren-0.031
 Ellen-0.028
TokenOne
Token ID3198
Feature activation+4.05e-02

Pos logit contributions
ble+0.071
bled+0.065
balls+0.061
ster+0.060
 fin+0.059
Neg logit contributions
ird-0.053
aren-0.048
 Karen-0.048
rell-0.046
Ell-0.044
Token day
Token ID1110
Feature activation+3.32e-02

Pos logit contributions
bled+0.112
ble+0.110
balls+0.106
ster+0.098
bing+0.094
Neg logit contributions
old-0.039
ird-0.036
hell-0.036
aren-0.036
he-0.034
Token,
Token ID11
Feature activation+2.36e-02

Pos logit contributions
ble+0.088
bled+0.081
balls+0.076
ster+0.076
 fin+0.074
Neg logit contributions
ird-0.045
 Karen-0.041
aren-0.039
rell-0.039
 Stella-0.038
Token the
Token ID262
Feature activation+4.81e-02

Pos logit contributions
ble+0.055
bled+0.051
ster+0.046
balls+0.046
 fin+0.045
Neg logit contributions
ird-0.038
 Karen-0.038
aren-0.035
 Stella-0.035
 Ellen-0.034
Token graceful
Token ID44363
Feature activation+6.19e-02

Pos logit contributions
ble+0.091
bled+0.086
balls+0.073
bing+0.071
ster+0.068
Neg logit contributions
ird-0.082
 Stella-0.081
aren-0.079
hell-0.078
 Karen-0.077
Token oy
Token ID35104
Feature activation+8.61e-02

Pos logit contributions
ble+0.110
bled+0.101
balls+0.088
ster+0.078
 fin+0.078
Neg logit contributions
 Stella-0.119
aren-0.119
rell-0.118
ird-0.118
 Karen-0.115
Tokenster
Token ID1706
Feature activation+6.38e-02

Pos logit contributions
bled+0.083
ble+0.078
bing+0.069
 fin+0.063
sing+0.060
Neg logit contributions
ird-0.198
ives-0.190
hell-0.188
old-0.184
aren-0.182
Token had
Token ID550
Feature activation+1.14e-02

Pos logit contributions
ble+0.157
bled+0.149
balls+0.137
 fin+0.125
ster+0.125
Neg logit contributions
ird-0.088
old-0.077
rell-0.077
aren-0.074
hell-0.072
Token a
Token ID257
Feature activation+7.57e-03

Pos logit contributions
ble+0.026
bled+0.024
balls+0.022
ster+0.022
 fin+0.021
Neg logit contributions
ird-0.019
 Karen-0.018
aren-0.017
rell-0.017
 Stella-0.016
Token special
Token ID2041
Feature activation+3.31e-02

Pos logit contributions
ble+0.019
bled+0.019
balls+0.017
ster+0.016
bing+0.016
Neg logit contributions
ird-0.010
 Karen-0.009
aren-0.009
 Stella-0.009
hell-0.008
Token treat
Token ID2190
Feature activation+3.38e-02

Pos logit contributions
ble+0.070
bled+0.068
balls+0.060
ster+0.057
bing+0.054
Neg logit contributions
ird-0.057
aren-0.052
 Karen-0.052
rell-0.050
 Stella-0.050
Token he
Token ID339
Feature activation+4.02e-02

Pos logit contributions
ble+0.045
ster+0.041
bled+0.040
 fin+0.038
iflower+0.035
Neg logit contributions
 Karen-0.034
 Ellie-0.032
ird-0.031
 Stella-0.029
rell-0.029
Token decided
Token ID3066
Feature activation+1.12e-02

Pos logit contributions
ble+0.104
bled+0.094
balls+0.089
ster+0.086
 fin+0.081
Neg logit contributions
ird-0.055
ha-0.046
 Karen-0.044
rell-0.044
aren-0.042
Token to
Token ID284
Feature activation+2.76e-02

Pos logit contributions
ble+0.028
bled+0.026
balls+0.024
 fin+0.023
ster+0.023
Neg logit contributions
 Karen-0.016
ird-0.015
 Stella-0.015
aren-0.015
 Ellen-0.014
Token remove
Token ID4781
Feature activation+2.01e-02

Pos logit contributions
ble+0.059
bled+0.055
balls+0.051
ster+0.051
 fin+0.048
Neg logit contributions
ird-0.051
 Karen-0.046
rell-0.046
aren-0.045
 Ellen-0.042
Token the
Token ID262
Feature activation+5.40e-02

Pos logit contributions
ble+0.042
bled+0.038
balls+0.036
ster+0.035
 fin+0.034
Neg logit contributions
ird-0.037
rell-0.035
 Karen-0.035
aren-0.035
 Ellen-0.032
Token seaw
Token ID32085
Feature activation+8.59e-02

Pos logit contributions
bled+0.103
ble+0.100
balls+0.091
ster+0.080
bing+0.078
Neg logit contributions
hell-0.092
rell-0.087
ird-0.085
ha-0.082
aren-0.080
Tokeneed
Token ID2308
Feature activation+6.36e-02

Pos logit contributions
bled+0.127
ble+0.122
balls+0.102
 shark+0.095
 digging+0.092
Neg logit contributions
hell-0.185
ird-0.180
ives-0.171
rell-0.168
ha-0.157
Token from
Token ID422
Feature activation+1.89e-02

Pos logit contributions
bled+0.146
ble+0.143
balls+0.130
emo+0.127
 fin+0.124
Neg logit contributions
hell-0.079
ird-0.076
rell-0.073
 aback-0.068
he-0.068
Token the
Token ID262
Feature activation+5.16e-02

Pos logit contributions
ble+0.040
ster+0.037
bled+0.036
balls+0.035
 fin+0.032
Neg logit contributions
 Karen-0.032
ird-0.031
rell-0.030
aren-0.029
 Stella-0.028
Token water
Token ID1660
Feature activation+5.66e-02

Pos logit contributions
bled+0.104
ble+0.100
balls+0.087
bing+0.081
emo+0.080
Neg logit contributions
hell-0.074
ird-0.072
rell-0.072
 bars-0.068
hy-0.068
Token so
Token ID523
Feature activation+3.31e-02

Pos logit contributions
ble+0.152
bled+0.147
balls+0.134
ster+0.130
emo+0.128
Neg logit contributions
ird-0.064
rell-0.058
aren-0.050
hell-0.050
 Karen-0.049
Token,
Token ID11
Feature activation+2.93e-02

Pos logit contributions
ble+0.168
bled+0.155
ster+0.154
 fin+0.145
ich+0.145
Neg logit contributions
he-0.072
rell-0.072
ird-0.071
 Ellie-0.069
hell-0.068
Token he
Token ID339
Feature activation+5.79e-02

Pos logit contributions
ble+0.071
ster+0.067
bled+0.064
 fin+0.061
balls+0.061
Neg logit contributions
 Karen-0.046
 Ellie-0.041
 Stella-0.038
ird-0.038
 Ellen-0.038
Token noticed
Token ID6810
Feature activation+3.09e-02

Pos logit contributions
ble+0.158
bled+0.146
balls+0.142
ster+0.130
 fin+0.128
Neg logit contributions
ird-0.072
rell-0.055
old-0.054
 Karen-0.052
aren-0.052
Token that
Token ID326
Feature activation+4.56e-02

Pos logit contributions
ble+0.067
bled+0.062
balls+0.060
ster+0.058
 fin+0.056
Neg logit contributions
ird-0.052
 Karen-0.050
aren-0.047
rell-0.047
old-0.045
Token his
Token ID465
Feature activation+8.14e-02

Pos logit contributions
ble+0.089
bled+0.083
ster+0.078
balls+0.076
 fin+0.068
Neg logit contributions
 Karen-0.088
ird-0.086
 Ellie-0.082
 Stella-0.081
 Alice-0.081
Token toy
Token ID13373
Feature activation+8.56e-02

Pos logit contributions
ble+0.187
bled+0.180
bling+0.149
ster+0.145
bing+0.144
Neg logit contributions
ird-0.092
hell-0.092
 keys-0.092
rell-0.087
chair-0.086
Token car
Token ID1097
Feature activation+2.57e-02

Pos logit contributions
ble+0.194
bled+0.182
 diving+0.170
emo+0.164
 surfing+0.161
Neg logit contributions
chairs-0.093
ird-0.092
chair-0.091
rell-0.089
 Ellie-0.086
Token was
Token ID373
Feature activation+2.11e-02

Pos logit contributions
ble+0.057
bled+0.051
balls+0.049
 fin+0.049
ster+0.048
Neg logit contributions
ird-0.044
rell-0.041
 Karen-0.040
aren-0.039
hell-0.037
Token missing
Token ID4814
Feature activation+2.20e-02

Pos logit contributions
ble+0.046
ster+0.043
balls+0.042
bled+0.041
 fin+0.040
Neg logit contributions
ird-0.032
rell-0.030
 Karen-0.030
 Stella-0.027
hell-0.027
Token a
Token ID257
Feature activation+1.02e-02

Pos logit contributions
ble+0.058
bled+0.054
balls+0.052
ster+0.051
 fin+0.050
Neg logit contributions
ird-0.028
 Karen-0.024
rell-0.024
aren-0.023
hell-0.022
Token wheel
Token ID7825
Feature activation+4.04e-04

Pos logit contributions
ble+0.023
bled+0.022
balls+0.019
ster+0.019
 fin+0.018
Neg logit contributions
ird-0.018
 Karen-0.016
rell-0.015
aren-0.015
hell-0.015
Token them
Token ID606
Feature activation-9.09e-03

Pos logit contributions
ble+0.045
bled+0.041
balls+0.038
ster+0.037
 fin+0.037
Neg logit contributions
ird-0.045
rell-0.042
aren-0.042
 Karen-0.042
 Ellen-0.039
Token to
Token ID284
Feature activation+1.59e-02

Pos logit contributions
ird+0.019
rell+0.018
aren+0.017
 Karen+0.017
 Ellen+0.016
Neg logit contributions
ble-0.017
bled-0.015
balls-0.015
ster-0.014
 fin-0.014
Token go
Token ID467
Feature activation+1.14e-02

Pos logit contributions
ble+0.035
bled+0.031
balls+0.029
ster+0.029
 fin+0.028
Neg logit contributions
ird-0.029
aren-0.027
rell-0.027
 Karen-0.027
 Ellen-0.025
Token.
Token ID13
Feature activation-2.60e-03

Pos logit contributions
ble+0.026
bled+0.024
balls+0.022
ster+0.022
 fin+0.021
Neg logit contributions
ird-0.019
 Karen-0.018
rell-0.018
aren-0.018
 Ellen-0.016
Token The
Token ID383
Feature activation+5.67e-02

Pos logit contributions
ird+0.004
rell+0.004
 Karen+0.004
aren+0.004
 Ellen+0.004
Neg logit contributions
ble-0.006
bled-0.006
balls-0.005
ster-0.005
 fin-0.005
Token careless
Token ID36138
Feature activation+8.43e-02

Pos logit contributions
bled+0.107
ble+0.101
balls+0.090
bing+0.083
urg+0.079
Neg logit contributions
hell-0.088
ird-0.088
 Eve-0.080
ha-0.079
he-0.078
Token jelly
Token ID27644
Feature activation+5.51e-02

Pos logit contributions
bled+0.144
ble+0.133
balls+0.126
bing+0.101
 fin+0.097
Neg logit contributions
hell-0.161
ird-0.150
ha-0.150
 Eve-0.142
rell-0.140
Tokenfish
Token ID11084
Feature activation+1.56e-02

Pos logit contributions
ble+0.060
bled+0.054
balls+0.043
 fin+0.039
ster+0.034
Neg logit contributions
ird-0.137
hell-0.134
ha-0.130
aren-0.127
rell-0.126
Token said
Token ID531
Feature activation+1.42e-02

Pos logit contributions
ble+0.034
bled+0.031
balls+0.029
ster+0.028
 fin+0.028
Neg logit contributions
ird-0.028
 Karen-0.025
rell-0.025
aren-0.025
 Ellen-0.024
Token goodbye
Token ID24829
Feature activation+1.94e-02

Pos logit contributions
ble+0.038
bled+0.034
balls+0.032
 fin+0.032
ster+0.032
Neg logit contributions
ird-0.017
 Karen-0.016
 Stella-0.015
 Ellen-0.015
 Ellie-0.014
Token to
Token ID284
Feature activation+1.97e-02

Pos logit contributions
ble+0.044
bled+0.040
balls+0.037
ster+0.036
 fin+0.036
Neg logit contributions
ird-0.033
 Karen-0.031
rell-0.031
aren-0.030
 Ellen-0.028
TokenThey
Token ID2990
Feature activation-8.85e-03

Pos logit contributions
ble+0.038
bled+0.035
balls+0.033
ster+0.033
 fin+0.032
Neg logit contributions
ird-0.032
 Karen-0.031
aren-0.031
rell-0.029
Ell-0.028
Token decide
Token ID5409
Feature activation-7.97e-04

Pos logit contributions
ird+0.017
rell+0.017
aren+0.016
 Karen+0.016
chairs+0.015
Neg logit contributions
ble-0.017
bled-0.015
 fin-0.015
ster-0.014
balls-0.014
Token to
Token ID284
Feature activation+1.07e-02

Pos logit contributions
ird+0.002
 Karen+0.002
aren+0.001
rell+0.001
 Ellen+0.001
Neg logit contributions
ble-0.002
bled-0.001
balls-0.001
ster-0.001
 fin-0.001
Token leave
Token ID2666
Feature activation+4.05e-03

Pos logit contributions
ble+0.022
bled+0.020
balls+0.019
ster+0.018
 fin+0.018
Neg logit contributions
ird-0.020
rell-0.019
aren-0.019
 Karen-0.019
 Ellen-0.017
Token the
Token ID262
Feature activation+4.71e-02

Pos logit contributions
ble+0.007
bled+0.007
balls+0.006
 fin+0.006
ster+0.006
Neg logit contributions
ird-0.009
rell-0.008
aren-0.008
 Karen-0.008
 Ellen-0.008
Token oy
Token ID35104
Feature activation+8.39e-02

Pos logit contributions
bled+0.089
balls+0.085
ble+0.081
 Miles+0.067
bing+0.066
Neg logit contributions
aren-0.075
 Mad-0.071
hell-0.071
ha-0.070
 Karen-0.069
Tokenster
Token ID1706
Feature activation+3.51e-02

Pos logit contributions
bled+0.085
balls+0.076
ble+0.060
 Finn+0.057
 diving+0.056
Neg logit contributions
ird-0.190
hell-0.179
mie-0.178
 seen-0.170
land-0.166
Token alone
Token ID3436
Feature activation+1.98e-02

Pos logit contributions
ble+0.076
bled+0.070
balls+0.066
ster+0.062
 fin+0.061
Neg logit contributions
ird-0.062
rell-0.055
aren-0.055
 Karen-0.054
hell-0.052
Token and
Token ID290
Feature activation+1.78e-02

Pos logit contributions
ble+0.045
bled+0.041
balls+0.039
ster+0.038
 fin+0.038
Neg logit contributions
ird-0.034
rell-0.031
 Karen-0.031
aren-0.031
 Ellen-0.029
Token play
Token ID711
Feature activation+3.56e-02

Pos logit contributions
ble+0.043
bled+0.040
balls+0.037
ster+0.036
 fin+0.035
Neg logit contributions
ird-0.028
 Karen-0.026
aren-0.025
rell-0.024
 Stella-0.024
Token with
Token ID351
Feature activation+7.66e-03

Pos logit contributions
ble+0.073
bled+0.069
balls+0.064
ster+0.061
 fin+0.058
Neg logit contributions
ird-0.059
hell-0.054
 Stella-0.049
 aback-0.049
 Karen-0.049
Token play
Token ID711
Feature activation+3.21e-02

Pos logit contributions
ird+0.000
rell+0.000
aren+0.000
 Karen+0.000
 Ellen+0.000
Neg logit contributions
ble-0.000
bled-0.000
balls-0.000
ster-0.000
 fin-0.000
Token and
Token ID290
Feature activation+1.12e-02

Pos logit contributions
ble+0.085
bled+0.082
ster+0.075
balls+0.075
 fin+0.072
Neg logit contributions
ird-0.040
hell-0.032
rell-0.031
he-0.030
 Karen-0.029
Token saw
Token ID2497
Feature activation+1.26e-02

Pos logit contributions
ble+0.027
bled+0.025
balls+0.023
ster+0.023
 fin+0.023
Neg logit contributions
ird-0.018
 Karen-0.016
rell-0.016
aren-0.016
 Ellen-0.015
Token Tim
Token ID5045
Feature activation-4.10e-03

Pos logit contributions
ble+0.026
bled+0.023
balls+0.022
ster+0.021
 fin+0.021
Neg logit contributions
ird-0.025
 Karen-0.023
rell-0.023
aren-0.023
 Ellen-0.021
Token's
Token ID338
Feature activation+6.78e-02

Pos logit contributions
ird+0.007
rell+0.006
 Karen+0.006
aren+0.006
hell+0.006
Neg logit contributions
ble-0.009
bled-0.009
balls-0.008
ster-0.008
 fin-0.008
Token broken
Token ID5445
Feature activation+8.39e-02

Pos logit contributions
bled+0.134
ble+0.128
ster+0.112
bing+0.110
balls+0.109
Neg logit contributions
hell-0.101
ird-0.097
rell-0.096
 umbrella-0.092
 Karen-0.086
Token sp
Token ID599
Feature activation+6.34e-02

Pos logit contributions
bled+0.165
balls+0.150
 Shark+0.148
ble+0.148
ster+0.142
Neg logit contributions
rell-0.118
hell-0.112
ird-0.111
 umbrella-0.105
 keys-0.100
Tokenade
Token ID671
Feature activation+2.17e-02

Pos logit contributions
ble+0.163
bled+0.159
 fin+0.158
ster+0.155
 Shark+0.153
Neg logit contributions
hell-0.068
ird-0.067
rell-0.062
he-0.054
ives-0.053
Token.
Token ID13
Feature activation-8.85e-03

Pos logit contributions
ble+0.056
bled+0.052
balls+0.049
ster+0.049
 fin+0.049
Neg logit contributions
ird-0.029
rell-0.026
 Karen-0.025
aren-0.025
hell-0.023
Token Tom
Token ID4186
Feature activation-4.20e-04

Pos logit contributions
ird+0.014
 Karen+0.014
rell+0.013
aren+0.013
 Ellen+0.012
Neg logit contributions
ble-0.022
bled-0.019
ster-0.018
 fin-0.018
balls-0.018
Token said
Token ID531
Feature activation+5.32e-03

Pos logit contributions
ird+0.001
rell+0.001
 Karen+0.001
aren+0.001
 Ellen+0.000
Neg logit contributions
ble-0.001
bled-0.001
balls-0.001
ster-0.001
 fin-0.001
Token
Token ID198
Feature activation+9.24e-03

Pos logit contributions
ird+0.027
rell+0.026
aren+0.024
 Karen+0.024
 Ellen+0.022
Neg logit contributions
ble-0.027
bled-0.022
balls-0.022
 fin-0.021
emo-0.021
TokenOne
Token ID3198
Feature activation+3.84e-02

Pos logit contributions
ble+0.024
bled+0.022
balls+0.021
ster+0.021
 fin+0.021
Neg logit contributions
ird-0.013
rell-0.012
aren-0.011
 Karen-0.011
 Ellen-0.010
Token day
Token ID1110
Feature activation+2.73e-02

Pos logit contributions
bled+0.108
balls+0.104
ble+0.103
ster+0.091
bing+0.088
Neg logit contributions
old-0.035
 Karen-0.032
aren-0.032
ird-0.031
hell-0.031
Token,
Token ID11
Feature activation+1.21e-02

Pos logit contributions
ble+0.077
bled+0.072
balls+0.068
ster+0.068
 fin+0.066
Neg logit contributions
ird-0.030
 Karen-0.027
rell-0.026
aren-0.024
 Stella-0.023
Token the
Token ID262
Feature activation+5.01e-02

Pos logit contributions
ble+0.031
bled+0.029
ster+0.028
balls+0.027
 fin+0.026
Neg logit contributions
 Karen-0.016
ird-0.015
 Stella-0.014
 Ellie-0.014
 Ellen-0.014
Token oy
Token ID35104
Feature activation+8.34e-02

Pos logit contributions
bled+0.097
balls+0.096
ble+0.088
bing+0.079
zes+0.075
Neg logit contributions
aren-0.074
hell-0.073
 Stella-0.073
 Karen-0.072
 Mad-0.072
Tokenster
Token ID1706
Feature activation+5.85e-02

Pos logit contributions
bled+0.080
balls+0.068
ble+0.065
 fin+0.059
 diving+0.058
Neg logit contributions
ird-0.198
hell-0.190
aren-0.186
old-0.186
mie-0.182
Token went
Token ID1816
Feature activation+3.15e-02

Pos logit contributions
ble+0.157
bled+0.151
balls+0.142
 fin+0.130
ster+0.128
Neg logit contributions
ird-0.073
old-0.060
rell-0.060
aren-0.058
 Karen-0.057
Token swimming
Token ID14899
Feature activation+4.90e-02

Pos logit contributions
ble+0.082
bled+0.079
ster+0.070
balls+0.067
 fin+0.067
Neg logit contributions
rell-0.037
 Ellie-0.032
 aback-0.031
 Ellen-0.029
Ell-0.028
Token and
Token ID290
Feature activation+2.82e-02

Pos logit contributions
ble+0.111
bled+0.104
ster+0.096
balls+0.095
 fin+0.088
Neg logit contributions
rell-0.068
ird-0.067
hell-0.064
 aback-0.064
 Karen-0.062
Token it
Token ID340
Feature activation+4.14e-02

Pos logit contributions
ble+0.077
bled+0.069
ster+0.066
iflower+0.062
balls+0.062
Neg logit contributions
rell-0.028
 Karen-0.028
 Stella-0.027
 Ellie-0.027
 seen-0.025
Token and
Token ID290
Feature activation+1.13e-02

Pos logit contributions
ird+0.025
rell+0.024
 Karen+0.024
aren+0.023
 Ellen+0.022
Neg logit contributions
ble-0.029
bled-0.027
ster-0.025
balls-0.025
 fin-0.024
Token her
Token ID607
Feature activation+1.50e-02

Pos logit contributions
ble+0.026
bled+0.023
balls+0.022
ster+0.022
 fin+0.021
Neg logit contributions
ird-0.019
rell-0.018
aren-0.018
 Karen-0.018
 Ellen-0.016
Token mom
Token ID1995
Feature activation-4.46e-02

Pos logit contributions
ble+0.024
bled+0.021
balls+0.019
ster+0.019
 fin+0.018
Neg logit contributions
ird-0.036
rell-0.034
 Karen-0.034
aren-0.034
 Ellen-0.032
Token left
Token ID1364
Feature activation-8.05e-03

Pos logit contributions
rell+0.093
 Ellen+0.087
ird+0.087
 Karen+0.085
aren+0.085
Neg logit contributions
ble-0.071
bled-0.070
ster-0.062
balls-0.060
 fin-0.057
Token the
Token ID262
Feature activation+8.79e-03

Pos logit contributions
ird+0.018
rell+0.017
aren+0.016
hell+0.016
 Karen+0.016
Neg logit contributions
ble-0.014
bled-0.013
balls-0.012
 fin-0.011
ster-0.011
Token oy
Token ID35104
Feature activation+8.33e-02

Pos logit contributions
ble+0.020
bled+0.018
balls+0.017
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.015
 Karen-0.014
aren-0.014
rell-0.014
 Stella-0.013
Tokenster
Token ID1706
Feature activation+3.67e-02

Pos logit contributions
bled+0.077
balls+0.067
ble+0.066
 fin+0.052
 spitting+0.050
Neg logit contributions
ird-0.206
mie-0.189
hell-0.189
aren-0.182
 seen-0.179
Token at
Token ID379
Feature activation+7.99e-03

Pos logit contributions
ble+0.078
bled+0.072
balls+0.067
 fin+0.062
ster+0.062
Neg logit contributions
ird-0.066
rell-0.058
aren-0.057
hell-0.057
 Karen-0.056
Token the
Token ID262
Feature activation+3.10e-02

Pos logit contributions
ble+0.018
bled+0.016
balls+0.016
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.013
 Karen-0.013
aren-0.013
rell-0.012
 Stella-0.012
Token beach
Token ID10481
Feature activation+2.64e-02

Pos logit contributions
bled+0.062
ble+0.061
balls+0.055
bing+0.048
emo+0.047
Neg logit contributions
 Stella-0.048
ird-0.048
aren-0.047
 Karen-0.047
rell-0.046
Token,
Token ID11
Feature activation+1.02e-02

Pos logit contributions
ble+0.066
bled+0.061
balls+0.057
ster+0.056
 fin+0.056
Neg logit contributions
ird-0.039
rell-0.035
 Karen-0.035
aren-0.035
 Ellen-0.032
Tokenmy
Token ID1820
Feature activation-1.44e-02

Pos logit contributions
ird+0.033
rell+0.032
aren+0.030
 Karen+0.029
 Stella+0.028
Neg logit contributions
ble-0.069
bled-0.064
ster-0.062
balls-0.061
 fin-0.058
Token told
Token ID1297
Feature activation-2.78e-02

Pos logit contributions
ird+0.025
rell+0.024
aren+0.023
 Karen+0.022
hell+0.021
Neg logit contributions
ble-0.032
bled-0.029
ster-0.028
balls-0.028
 fin-0.027
Token him
Token ID683
Feature activation+4.08e-03

Pos logit contributions
ird+0.067
aren+0.061
rell+0.059
Aud+0.057
 Karen+0.057
Neg logit contributions
ble-0.045
bled-0.040
balls-0.035
 fin-0.032
ster-0.032
Token that
Token ID326
Feature activation+2.24e-02

Pos logit contributions
ble+0.009
bled+0.008
ster+0.007
 fin+0.007
balls+0.007
Neg logit contributions
ird-0.007
 Karen-0.007
 Ellen-0.006
rell-0.006
aren-0.006
Token the
Token ID262
Feature activation+4.02e-02

Pos logit contributions
ble+0.048
bled+0.046
ster+0.042
balls+0.041
 fin+0.039
Neg logit contributions
 Karen-0.037
ird-0.036
rell-0.034
 Ellen-0.033
 Ellie-0.033
Token ham
Token ID8891
Feature activation+8.28e-02

Pos logit contributions
bled+0.090
ble+0.087
balls+0.074
bing+0.071
 fin+0.068
Neg logit contributions
ird-0.058
hell-0.057
 Stella-0.055
rell-0.055
 aback-0.052
Tokenster
Token ID1706
Feature activation+2.22e-02

Pos logit contributions
ble+0.105
bled+0.103
 diving+0.091
 surfing+0.090
bing+0.088
Neg logit contributions
hell-0.179
ird-0.173
ives-0.167
 bars-0.151
aren-0.151
Token would
Token ID561
Feature activation+9.62e-03

Pos logit contributions
ble+0.049
bled+0.045
balls+0.041
ster+0.040
 fin+0.039
Neg logit contributions
ird-0.039
rell-0.036
 Karen-0.036
aren-0.035
old-0.033
Token always
Token ID1464
Feature activation-4.46e-03

Pos logit contributions
ble+0.018
bled+0.017
balls+0.016
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.020
rell-0.018
 Karen-0.018
aren-0.018
 Ellen-0.017
Token be
Token ID307
Feature activation+9.72e-03

Pos logit contributions
ird+0.009
rell+0.008
 Karen+0.008
aren+0.008
 Ellen+0.008
Neg logit contributions
ble-0.009
bled-0.008
balls-0.008
ster-0.007
 fin-0.007
Token with
Token ID351
Feature activation-5.43e-03

Pos logit contributions
ble+0.022
bled+0.020
balls+0.019
ster+0.018
 fin+0.018
Neg logit contributions
ird-0.017
rell-0.016
aren-0.016
 Karen-0.016
 Ellen-0.014
Token he
Token ID339
Feature activation+4.45e-02

Pos logit contributions
ble+0.068
bled+0.061
balls+0.058
ster+0.056
 fin+0.055
Neg logit contributions
ird-0.061
rell-0.057
 Karen-0.057
aren-0.056
 Ellen-0.052
Token felt
Token ID2936
Feature activation+9.47e-03

Pos logit contributions
ble+0.101
bled+0.092
balls+0.087
ster+0.083
 fin+0.082
Neg logit contributions
ird-0.072
rell-0.067
 Karen-0.062
hell-0.062
aren-0.061
Token better
Token ID1365
Feature activation+2.73e-02

Pos logit contributions
ble+0.024
bled+0.022
balls+0.021
ster+0.021
 fin+0.020
Neg logit contributions
ird-0.014
rell-0.012
 Karen-0.012
aren-0.012
hell-0.011
Token.
Token ID13
Feature activation+6.75e-03

Pos logit contributions
ble+0.075
bled+0.072
balls+0.069
 fin+0.065
ster+0.065
Neg logit contributions
ird-0.031
aren-0.024
 Karen-0.023
rell-0.023
hell-0.022
Token The
Token ID383
Feature activation+5.13e-02

Pos logit contributions
ble+0.015
bled+0.014
balls+0.013
ster+0.013
 fin+0.013
Neg logit contributions
ird-0.012
rell-0.011
 Karen-0.011
aren-0.011
 Ellen-0.010
Token green
Token ID4077
Feature activation+8.26e-02

Pos logit contributions
bled+0.131
ble+0.128
balls+0.114
ster+0.105
bing+0.104
Neg logit contributions
rell-0.053
hell-0.053
ird-0.051
 Many-0.045
 officer-0.044
Token oct
Token ID19318
Feature activation+5.91e-02

Pos logit contributions
bled+0.167
ble+0.141
bling+0.133
balls+0.133
ross+0.119
Neg logit contributions
rell-0.122
hell-0.116
les-0.106
est-0.103
ha-0.096
Tokenopus
Token ID25790
Feature activation+2.62e-02

Pos logit contributions
bled+0.100
ble+0.095
balls+0.078
 fin+0.078
 hyd+0.073
Neg logit contributions
ird-0.118
hell-0.118
ives-0.110
aren-0.105
he-0.103
Token and
Token ID290
Feature activation+5.37e-02

Pos logit contributions
ble+0.062
bled+0.057
balls+0.054
ster+0.051
 fin+0.050
Neg logit contributions
ird-0.041
rell-0.037
 Karen-0.035
aren-0.035
hell-0.033
Token his
Token ID465
Feature activation+5.68e-02

Pos logit contributions
bled+0.121
ble+0.121
balls+0.112
ster+0.107
ider+0.104
Neg logit contributions
 Ellie-0.072
rell-0.070
 Karen-0.067
hell-0.065
 Ellen-0.063
Token friends
Token ID2460
Feature activation-2.48e-02

Pos logit contributions
ble+0.129
bled+0.123
ster+0.107
balls+0.103
bing+0.099
Neg logit contributions
rell-0.080
ird-0.080
 Karen-0.076
hell-0.075
aren-0.071
Token to
Token ID284
Feature activation+2.29e-02

Pos logit contributions
ble+0.021
bled+0.019
balls+0.017
ster+0.017
 fin+0.017
Neg logit contributions
ird-0.019
 Karen-0.018
rell-0.018
aren-0.018
 Stella-0.016
Token get
Token ID651
Feature activation+3.68e-02

Pos logit contributions
ble+0.048
bled+0.043
balls+0.040
ster+0.040
 fin+0.039
Neg logit contributions
ird-0.044
rell-0.040
 Karen-0.040
aren-0.040
 Ellen-0.037
Token out
Token ID503
Feature activation+3.08e-02

Pos logit contributions
ble+0.083
ster+0.077
bled+0.073
balls+0.071
 fin+0.070
Neg logit contributions
ird-0.057
rell-0.052
 Karen-0.050
 Eve-0.049
old-0.047
Token of
Token ID286
Feature activation+1.59e-02

Pos logit contributions
ble+0.078
bled+0.072
balls+0.068
ster+0.067
 fin+0.066
Neg logit contributions
ird-0.045
rell-0.040
 Karen-0.039
aren-0.039
old-0.036
Token the
Token ID262
Feature activation+4.53e-02

Pos logit contributions
ble+0.035
bled+0.031
balls+0.030
ster+0.030
 fin+0.028
Neg logit contributions
ird-0.028
 Karen-0.027
rell-0.026
aren-0.026
 Ellen-0.025
Token bath
Token ID7837
Feature activation+8.25e-02

Pos logit contributions
ble+0.106
bled+0.103
ster+0.095
balls+0.093
 Miles+0.087
Neg logit contributions
ird-0.061
rell-0.060
 Eve-0.056
 Karen-0.051
 Mad-0.050
Tokentub
Token ID37995
Feature activation+1.97e-02

Pos logit contributions
ble+0.121
bled+0.108
 fin+0.095
ster+0.094
balls+0.090
Neg logit contributions
ird-0.197
rell-0.187
 Eve-0.172
aren-0.169
ives-0.168
Token now
Token ID783
Feature activation+1.98e-02

Pos logit contributions
ble+0.048
bled+0.044
balls+0.042
ster+0.041
 fin+0.040
Neg logit contributions
ird-0.030
rell-0.028
 Karen-0.028
aren-0.027
 Ellen-0.025
Token.
Token ID13
Feature activation+1.11e-02

Pos logit contributions
ble+0.050
bled+0.046
balls+0.043
ster+0.043
 fin+0.042
Neg logit contributions
ird-0.029
rell-0.026
 Karen-0.026
aren-0.026
 Ellen-0.024
Token The
Token ID383
Feature activation+4.69e-02

Pos logit contributions
ble+0.025
bled+0.023
balls+0.022
ster+0.021
 fin+0.021
Neg logit contributions
ird-0.019
 Karen-0.018
rell-0.017
aren-0.017
 Ellen-0.016
Token water
Token ID1660
Feature activation+4.80e-02

Pos logit contributions
ble+0.102
bled+0.098
ster+0.090
balls+0.087
iflower+0.081
Neg logit contributions
ird-0.072
rell-0.068
 Karen-0.064
 Eve-0.063
 Stella-0.061
Token with
Token ID351
Feature activation+7.61e-03

Pos logit contributions
ble+0.017
bled+0.016
balls+0.014
ster+0.014
 fin+0.013
Neg logit contributions
ird-0.020
rell-0.019
 Karen-0.019
aren-0.018
hell-0.018
Token."
Token ID526
Feature activation+4.01e-03

Pos logit contributions
ble+0.018
bled+0.016
balls+0.015
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.012
aren-0.011
rell-0.011
 Karen-0.011
 Ellen-0.010
Token
Token ID198
Feature activation+4.74e-03

Pos logit contributions
ble+0.010
bled+0.009
balls+0.009
ster+0.009
 fin+0.008
Neg logit contributions
ird-0.006
rell-0.005
 Karen-0.005
aren-0.005
 Ellen-0.005
Token
Token ID198
Feature activation+1.68e-02

Pos logit contributions
ble+0.011
bled+0.010
balls+0.010
ster+0.009
 fin+0.009
Neg logit contributions
ird-0.008
rell-0.007
aren-0.007
 Karen-0.007
 Ellen-0.006
TokenThe
Token ID464
Feature activation+4.90e-02

Pos logit contributions
ble+0.042
bled+0.040
balls+0.038
ster+0.036
 fin+0.036
Neg logit contributions
ird-0.024
Ell-0.022
aren-0.022
 Karen-0.022
Alice-0.021
Token white
Token ID2330
Feature activation+8.25e-02

Pos logit contributions
ble+0.128
bled+0.127
balls+0.116
ster+0.109
bing+0.106
Neg logit contributions
hell-0.044
rell-0.041
ird-0.040
 Karen-0.036
 Stella-0.036
Token oct
Token ID19318
Feature activation+5.89e-02

Pos logit contributions
bled+0.142
ble+0.125
balls+0.118
bling+0.109
bing+0.100
Neg logit contributions
hell-0.137
rell-0.136
ha-0.124
aren-0.120
aff-0.120
Tokenopus
Token ID25790
Feature activation+1.05e-02

Pos logit contributions
bled+0.096
ble+0.092
balls+0.076
 fin+0.074
bing+0.072
Neg logit contributions
ird-0.122
hell-0.119
ives-0.113
aren-0.112
old-0.110
Token felt
Token ID2936
Feature activation-7.01e-03

Pos logit contributions
ble+0.026
bled+0.023
balls+0.022
ster+0.021
 fin+0.021
Neg logit contributions
ird-0.016
rell-0.014
 Karen-0.014
aren-0.014
 Ellen-0.013
Token bad
Token ID2089
Feature activation+1.65e-03

Pos logit contributions
ird+0.012
rell+0.011
aren+0.011
 Karen+0.011
 Ellen+0.010
Neg logit contributions
ble-0.016
bled-0.014
balls-0.013
ster-0.013
 fin-0.013
Token for
Token ID329
Feature activation+8.42e-03

Pos logit contributions
ble+0.004
bled+0.004
balls+0.003
ster+0.003
 fin+0.003
Neg logit contributions
ird-0.002
rell-0.002
 Karen-0.002
aren-0.002
hell-0.002
Token it
Token ID340
Feature activation-1.64e-02

Pos logit contributions
ird+0.036
aren+0.034
rell+0.033
 Karen+0.032
Ell+0.030
Neg logit contributions
ble-0.024
bled-0.019
 fin-0.018
ster-0.018
balls-0.018
Token before
Token ID878
Feature activation-1.70e-02

Pos logit contributions
ird+0.030
aren+0.029
rell+0.028
 Karen+0.027
old+0.026
Neg logit contributions
ble-0.035
bled-0.030
balls-0.030
ster-0.029
 fin-0.028
Token it
Token ID340
Feature activation+1.02e-02

Pos logit contributions
ird+0.033
aren+0.031
rell+0.031
old+0.029
 Karen+0.029
Neg logit contributions
ble-0.035
bled-0.030
balls-0.029
 fin-0.027
ster-0.027
Token was
Token ID373
Feature activation+1.91e-03

Pos logit contributions
ble+0.022
bled+0.020
balls+0.018
ster+0.018
 fin+0.018
Neg logit contributions
ird-0.019
rell-0.017
 Karen-0.017
aren-0.017
 Ellen-0.016
Token gone
Token ID3750
Feature activation+4.71e-03

Pos logit contributions
ble+0.004
bled+0.004
balls+0.003
 fin+0.003
ster+0.003
Neg logit contributions
ird-0.004
aren-0.003
rell-0.003
 Karen-0.003
 Ellen-0.003
Token.
Token ID13
Feature activation-8.87e-02

Pos logit contributions
ble+0.011
bled+0.010
balls+0.009
ster+0.009
 fin+0.009
Neg logit contributions
ird-0.008
aren-0.007
rell-0.007
 Karen-0.007
old-0.007
Token She
Token ID1375
Feature activation+3.73e-03

Pos logit contributions
hell+0.170
ird+0.159
ock+0.155
ives+0.153
Aud+0.152
Neg logit contributions
ble-0.145
bled-0.135
 next-0.127
balls-0.118
ster-0.115
Token chased
Token ID26172
Feature activation-5.08e-02

Pos logit contributions
ble+0.009
bled+0.008
balls+0.007
ster+0.007
 fin+0.007
Neg logit contributions
ird-0.006
 Karen-0.006
aren-0.006
rell-0.006
 Ellen-0.005
Token it
Token ID340
Feature activation-1.53e-02

Pos logit contributions
ird+0.138
rell+0.127
chairs+0.117
aren+0.116
hell+0.115
Neg logit contributions
ble-0.064
bled-0.051
 next-0.046
bing-0.043
ster-0.043
Token and
Token ID290
Feature activation-2.87e-02

Pos logit contributions
ird+0.031
aren+0.028
 Karen+0.027
rell+0.027
old+0.025
Neg logit contributions
ble-0.031
bled-0.027
balls-0.027
 diving-0.025
 fin-0.025
Token chased
Token ID26172
Feature activation-4.82e-02

Pos logit contributions
ird+0.056
rell+0.053
aren+0.052
 Karen+0.050
hell+0.049
Neg logit contributions
ble-0.058
bled-0.053
balls-0.050
ster-0.048
 fin-0.047
Token for
Token ID329
Feature activation-2.43e-02

Pos logit contributions
ird+0.012
aren+0.012
rell+0.012
 Karen+0.011
 Ellen+0.011
Neg logit contributions
ble-0.009
bled-0.008
balls-0.007
ster-0.007
 fin-0.007
Token ways
Token ID2842
Feature activation-3.86e-02

Pos logit contributions
rell+0.053
ird+0.052
aren+0.052
ives+0.048
Ell+0.048
Neg logit contributions
ble-0.038
 diving-0.033
 fin-0.032
bled-0.031
balls-0.031
Token to
Token ID284
Feature activation-9.80e-03

Pos logit contributions
ird+0.094
rell+0.089
aren+0.087
 Karen+0.085
hell+0.082
Neg logit contributions
ble-0.059
bled-0.049
balls-0.049
ster-0.046
 diving-0.044
Token help
Token ID1037
Feature activation-3.34e-02

Pos logit contributions
ird+0.020
aren+0.020
rell+0.019
 Karen+0.019
chairs+0.018
Neg logit contributions
ble-0.019
bled-0.016
balls-0.015
 fin-0.015
ster-0.015
Token others
Token ID1854
Feature activation-9.70e-03

Pos logit contributions
aren+0.064
ird+0.063
rell+0.061
 Karen+0.057
Aud+0.057
Neg logit contributions
ble-0.068
bled-0.054
balls-0.052
ster-0.052
 fin-0.050
Token.
Token ID13
Feature activation-8.60e-02

Pos logit contributions
ird+0.015
rell+0.014
aren+0.014
 Karen+0.013
 Ellen+0.012
Neg logit contributions
ble-0.024
bled-0.022
balls-0.021
ster-0.020
 fin-0.020
Token
Token ID220
Feature activation-4.42e-02

Pos logit contributions
hell+0.186
ives+0.184
ird+0.182
ited+0.179
aney+0.176
Neg logit contributions
ble-0.109
 next-0.096
bled-0.083
bing-0.072
balls-0.071
Token
Token ID198
Feature activation-6.82e-02

Pos logit contributions
rell+0.102
ird+0.099
ited+0.095
aren+0.094
ha+0.092
Neg logit contributions
ble-0.071
bled-0.057
balls-0.054
emo-0.046
ster-0.043
Token
Token ID198
Feature activation-6.30e-02

Pos logit contributions
ird+0.166
rell+0.162
ives+0.159
vious+0.157
hell+0.151
Neg logit contributions
ble-0.089
 next-0.067
bled-0.062
balls-0.057
 diving-0.050
TokenOne
Token ID3198
Feature activation+5.12e-03

Pos logit contributions
ives+0.125
ird+0.123
 aback+0.119
aren+0.118
aney+0.117
Neg logit contributions
ble-0.102
 next-0.090
bled-0.085
bing-0.079
ster-0.068
Token day
Token ID1110
Feature activation-4.13e-02

Pos logit contributions
ble+0.012
bled+0.012
balls+0.011
ster+0.010
 fin+0.010
Neg logit contributions
ird-0.008
aren-0.008
 Karen-0.008
old-0.007
chair-0.007
Token was
Token ID373
Feature activation+8.94e-03

Pos logit contributions
ble+0.024
bled+0.021
balls+0.020
ster+0.020
 fin+0.019
Neg logit contributions
ird-0.024
rell-0.022
aren-0.022
 Karen-0.022
 Ellen-0.021
Token too
Token ID1165
Feature activation+7.54e-03

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.016
 Karen-0.015
rell-0.015
aren-0.015
 Ellen-0.014
Token dangerous
Token ID4923
Feature activation-3.41e-02

Pos logit contributions
ble+0.015
bled+0.013
balls+0.012
ster+0.012
 fin+0.012
Neg logit contributions
ird-0.015
rell-0.014
aren-0.014
 Karen-0.014
 Ellen-0.013
Token for
Token ID329
Feature activation-4.52e-02

Pos logit contributions
ird+0.073
aren+0.068
rell+0.065
 Karen+0.063
Ell+0.062
Neg logit contributions
ble-0.061
balls-0.055
bled-0.053
 diving-0.049
ster-0.049
Token her
Token ID607
Feature activation+8.40e-03

Pos logit contributions
ives+0.089
ird+0.088
hell+0.086
 aback+0.082
old+0.081
Neg logit contributions
ble-0.085
bled-0.075
balls-0.069
 diving-0.066
 next-0.062
Token.
Token ID13
Feature activation-8.48e-02

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.014
rell-0.013
aren-0.013
 Karen-0.013
 Ellen-0.012
Token
Token ID220
Feature activation-4.35e-02

Pos logit contributions
hell+0.136
ives+0.131
ird+0.130
ock+0.127
ited+0.121
Neg logit contributions
ble-0.161
balls-0.152
bled-0.146
 next-0.141
bing-0.129
Token
Token ID198
Feature activation-6.48e-02

Pos logit contributions
rell+0.098
ird+0.097
ited+0.094
hy+0.092
aren+0.092
Neg logit contributions
ble-0.069
bled-0.055
balls-0.053
emo-0.043
 Miles-0.039
Token
Token ID198
Feature activation-5.94e-02

Pos logit contributions
ird+0.161
rell+0.161
vious+0.159
ives+0.150
hell+0.148
Neg logit contributions
ble-0.077
balls-0.052
bled-0.049
 next-0.049
emo-0.042
TokenK
Token ID42
Feature activation-5.00e-02

Pos logit contributions
ird+0.089
ives+0.088
rell+0.085
hell+0.083
 aback+0.075
Neg logit contributions
ble-0.129
bled-0.112
 next-0.108
bing-0.103
balls-0.102
Tokenaren
Token ID5757
Feature activation-6.23e-03

Pos logit contributions
ird+0.042
hell+0.040
rell+0.039
Ell+0.037
 Ellen+0.037
Neg logit contributions
ble-0.127
ster-0.117
bled-0.113
 next-0.107
ich-0.107
Token a
Token ID257
Feature activation-1.81e-02

Pos logit contributions
ird+0.055
rell+0.053
aren+0.052
hell+0.050
 Karen+0.049
Neg logit contributions
ble-0.044
bled-0.039
 fin-0.036
balls-0.035
 diving-0.035
Token big
Token ID1263
Feature activation-3.02e-02

Pos logit contributions
ird+0.032
 Karen+0.029
aren+0.029
rell+0.029
 Ellen+0.027
Neg logit contributions
ble-0.041
bled-0.037
balls-0.035
ster-0.034
 fin-0.033
Token and
Token ID290
Feature activation+5.60e-03

Pos logit contributions
ird+0.061
rell+0.056
 Karen+0.055
aren+0.055
hell+0.052
Neg logit contributions
ble-0.057
bled-0.049
ster-0.049
 fin-0.049
balls-0.047
Token beautiful
Token ID4950
Feature activation-5.43e-03

Pos logit contributions
ble+0.012
bled+0.011
balls+0.010
ster+0.010
 fin+0.010
Neg logit contributions
ird-0.010
rell-0.010
aren-0.010
 Karen-0.009
 Ellen-0.009
Token nation
Token ID3277
Feature activation-1.51e-02

Pos logit contributions
ird+0.011
rell+0.010
aren+0.010
 Karen+0.010
 Ellen+0.009
Neg logit contributions
ble-0.011
bled-0.010
balls-0.009
ster-0.009
 fin-0.009
Token.
Token ID13
Feature activation-8.44e-02

Pos logit contributions
ird+0.026
rell+0.025
 Karen+0.025
aren+0.024
 Ellen+0.023
Neg logit contributions
ble-0.033
bled-0.029
balls-0.028
ster-0.028
 fin-0.027
Token Every
Token ID3887
Feature activation+1.05e-03

Pos logit contributions
ird+0.163
hell+0.149
ha+0.148
ives+0.142
rell+0.140
Neg logit contributions
 next-0.121
ble-0.099
 Then-0.098
balls-0.095
bled-0.094
Token day
Token ID1110
Feature activation-4.30e-02

Pos logit contributions
ble+0.002
bled+0.002
balls+0.002
ster+0.002
 fin+0.002
Neg logit contributions
ird-0.002
rell-0.002
 Karen-0.002
aren-0.002
 Ellen-0.002
Token she
Token ID673
Feature activation+2.79e-02

Pos logit contributions
rell+0.094
ird+0.090
aren+0.086
hell+0.086
Ell+0.085
Neg logit contributions
ble-0.057
bled-0.055
balls-0.052
 next-0.050
 diving-0.048
Token would
Token ID561
Feature activation-1.20e-02

Pos logit contributions
ble+0.064
bled+0.059
balls+0.055
ster+0.054
 fin+0.053
Neg logit contributions
ird-0.047
rell-0.043
aren-0.042
 Karen-0.042
 Ellen-0.039
Token go
Token ID467
Feature activation-1.11e-02

Pos logit contributions
ird+0.022
rell+0.021
aren+0.021
 Karen+0.020
hell+0.019
Neg logit contributions
ble-0.025
bled-0.022
balls-0.021
ster-0.021
 fin-0.021
Token little
Token ID1310
Feature activation-1.02e-03

Pos logit contributions
ird+0.012
rell+0.011
aren+0.011
 Karen+0.011
Ell+0.010
Neg logit contributions
ble-0.013
ster-0.011
bled-0.011
 fin-0.011
 shark-0.011
Token girl
Token ID2576
Feature activation-1.01e-02

Pos logit contributions
ird+0.002
 Karen+0.002
aren+0.002
rell+0.002
 Ellen+0.001
Neg logit contributions
ble-0.002
bled-0.002
balls-0.002
ster-0.002
 fin-0.002
Token named
Token ID3706
Feature activation-2.58e-02

Pos logit contributions
ird+0.017
rell+0.016
aren+0.016
 Karen+0.015
 Ellen+0.014
Neg logit contributions
ble-0.023
bled-0.020
ster-0.020
balls-0.019
 fin-0.019
Token Mad
Token ID4627
Feature activation-5.98e-02

Pos logit contributions
ird+0.057
rell+0.057
hell+0.055
 stret+0.055
chairs+0.054
Neg logit contributions
ble-0.030
balls-0.029
 next-0.027
bled-0.026
 diving-0.025
Tokeneline
Token ID4470
Feature activation-9.59e-03

Pos logit contributions
ird+0.075
rell+0.072
Aud+0.068
ives+0.067
hell+0.064
Neg logit contributions
ble-0.145
ster-0.135
bled-0.132
bing-0.117
 next-0.115
Token.
Token ID13
Feature activation-8.42e-02

Pos logit contributions
ird+0.017
rell+0.017
aren+0.016
Aud+0.016
 Karen+0.016
Neg logit contributions
ble-0.020
ster-0.018
bled-0.018
balls-0.016
 fin-0.016
Token One
Token ID1881
Feature activation+4.05e-03

Pos logit contributions
ird+0.150
old+0.148
ives+0.142
hell+0.138
anne+0.129
Neg logit contributions
ble-0.150
bled-0.140
 next-0.134
bing-0.116
balls-0.109
Token day
Token ID1110
Feature activation-4.57e-02

Pos logit contributions
ble+0.009
bled+0.008
balls+0.008
ster+0.007
 fin+0.007
Neg logit contributions
ird-0.007
 Karen-0.007
aren-0.007
rell-0.007
chair-0.006
Token,
Token ID11
Feature activation-6.51e-02

Pos logit contributions
ives+0.097
rell+0.095
Aud+0.094
old+0.093
hell+0.093
Neg logit contributions
ble-0.069
bled-0.064
balls-0.052
ster-0.051
 next-0.047
Token she
Token ID673
Feature activation+2.90e-02

Pos logit contributions
ives+0.128
rell+0.121
hell+0.119
old+0.117
ird+0.112
Neg logit contributions
ble-0.121
bled-0.118
 next-0.102
ster-0.095
balls-0.094
Token was
Token ID373
Feature activation-5.22e-04

Pos logit contributions
ble+0.062
bled+0.055
balls+0.052
ster+0.051
 fin+0.050
Neg logit contributions
ird-0.054
rell-0.050
aren-0.050
 Karen-0.050
 Ellen-0.046
Token catch
Token ID4929
Feature activation-6.55e-03

Pos logit contributions
ird+0.030
aren+0.029
rell+0.028
 Karen+0.028
chairs+0.027
Neg logit contributions
ble-0.024
bled-0.020
balls-0.019
 fin-0.019
ster-0.018
Token it
Token ID340
Feature activation-9.70e-03

Pos logit contributions
ird+0.016
aren+0.015
rell+0.015
 Karen+0.014
Ell+0.013
Neg logit contributions
ble-0.011
bled-0.009
balls-0.009
 fin-0.009
ster-0.008
Token in
Token ID287
Feature activation-2.32e-02

Pos logit contributions
ird+0.017
aren+0.016
 Karen+0.015
rell+0.015
 Ellen+0.014
Neg logit contributions
ble-0.022
bled-0.019
balls-0.019
ster-0.018
 fin-0.018
Token her
Token ID607
Feature activation+1.31e-02

Pos logit contributions
rell+0.056
ird+0.056
aren+0.052
ives+0.052
hell+0.051
Neg logit contributions
ble-0.034
 fin-0.029
 diving-0.028
bled-0.027
 digging-0.026
Token hands
Token ID2832
Feature activation-1.30e-02

Pos logit contributions
ble+0.029
bled+0.027
balls+0.025
ster+0.023
emo+0.022
Neg logit contributions
ird-0.023
aren-0.020
 Karen-0.020
rell-0.020
 Ellen-0.019
Token.
Token ID13
Feature activation-8.40e-02

Pos logit contributions
ird+0.022
aren+0.021
rell+0.021
 Karen+0.021
 Ellen+0.019
Neg logit contributions
ble-0.030
bled-0.026
balls-0.025
ster-0.025
 fin-0.024
Token
Token ID220
Feature activation-4.69e-02

Pos logit contributions
hell+0.163
ird+0.155
ock+0.151
ited+0.150
Aud+0.149
Neg logit contributions
ble-0.136
bled-0.126
 next-0.113
balls-0.108
bing-0.101
Token
Token ID198
Feature activation-6.60e-02

Pos logit contributions
rell+0.100
ird+0.099
ited+0.098
aren+0.094
hell+0.093
Neg logit contributions
ble-0.082
bled-0.070
balls-0.065
emo-0.053
ster-0.050
Token
Token ID198
Feature activation-5.45e-02

Pos logit contributions
ird+0.159
rell+0.159
vious+0.155
ited+0.151
hell+0.149
Neg logit contributions
ble-0.088
bled-0.066
 next-0.059
balls-0.058
emo-0.049
TokenWhen
Token ID2215
Feature activation-4.44e-02

Pos logit contributions
ird+0.108
rell+0.108
aren+0.104
hell+0.102
ives+0.101
Neg logit contributions
ble-0.097
bled-0.085
bing-0.072
 next-0.071
ster-0.069
Token she
Token ID673
Feature activation+7.21e-03

Pos logit contributions
ird+0.072
rell+0.070
hell+0.070
ives+0.069
old+0.067
Neg logit contributions
ble-0.096
bled-0.089
ster-0.078
balls-0.076
 next-0.075
Token lunch
Token ID9965
Feature activation-1.61e-02

Pos logit contributions
ird+0.009
aren+0.008
rell+0.008
 Karen+0.008
hell+0.008
Neg logit contributions
ble-0.007
bled-0.006
balls-0.006
 fin-0.006
ster-0.006
Token,"
Token ID553
Feature activation-4.28e-02

Pos logit contributions
rell+0.026
aren+0.025
 Karen+0.025
ird+0.024
 Ellen+0.023
Neg logit contributions
ble-0.036
balls-0.032
bled-0.031
ster-0.030
 diving-0.029
Token her
Token ID607
Feature activation-8.16e-03

Pos logit contributions
rell+0.083
ird+0.077
ives+0.075
hell+0.074
aren+0.073
Neg logit contributions
ble-0.086
bled-0.076
balls-0.071
ster-0.070
 fin-0.067
Token mom
Token ID1995
Feature activation-3.95e-02

Pos logit contributions
rell+0.022
ird+0.022
aren+0.021
hell+0.021
ives+0.021
Neg logit contributions
ble-0.010
bled-0.008
balls-0.008
ster-0.007
 fin-0.007
Token shouted
Token ID17293
Feature activation-2.59e-02

Pos logit contributions
rell+0.069
aren+0.065
hell+0.064
Aud+0.062
 Karen+0.061
Neg logit contributions
ble-0.080
bled-0.074
balls-0.071
ster-0.069
 fin-0.064
Token.
Token ID13
Feature activation-8.38e-02

Pos logit contributions
ird+0.045
rell+0.043
aren+0.043
 Karen+0.041
hell+0.038
Neg logit contributions
ble-0.057
bled-0.052
balls-0.050
ster-0.048
 fin-0.047
Token
Token ID220
Feature activation-5.97e-02

Pos logit contributions
hell+0.164
ird+0.159
rell+0.155
ock+0.154
vious+0.142
Neg logit contributions
balls-0.130
ble-0.125
bled-0.121
 next-0.119
 Mitch-0.100
Token
Token ID198
Feature activation-6.74e-02

Pos logit contributions
ird+0.128
rell+0.123
hell+0.121
aren+0.117
ited+0.117
Neg logit contributions
ble-0.103
bled-0.089
balls-0.087
emo-0.066
 next-0.064
Token
Token ID198
Feature activation-5.59e-02

Pos logit contributions
rell+0.153
ird+0.140
vious+0.139
hell+0.133
chairs+0.132
Neg logit contributions
ble-0.095
 next-0.083
bled-0.082
balls-0.078
But-0.063
TokenAlice
Token ID44484
Feature activation+8.57e-03

Pos logit contributions
rell+0.081
ird+0.065
hell+0.063
chairs+0.060
ives+0.058
Neg logit contributions
ble-0.126
bled-0.122
 next-0.118
balls-0.112
ster-0.109
Token took
Token ID1718
Feature activation-4.90e-02

Pos logit contributions
ble+0.020
bled+0.018
balls+0.017
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.015
 Karen-0.013
aren-0.013
rell-0.013
 Ellen-0.012
Token she
Token ID673
Feature activation+5.51e-02

Pos logit contributions
ives+0.018
aren+0.018
ird+0.017
Aud+0.017
hell+0.017
Neg logit contributions
ble-0.014
bled-0.012
ster-0.011
balls-0.011
 fin-0.010
Token heard
Token ID2982
Feature activation-3.33e-02

Pos logit contributions
ble+0.134
bled+0.122
balls+0.119
ster+0.114
 fin+0.110
Neg logit contributions
ird-0.082
rell-0.068
aren-0.064
 Karen-0.064
hell-0.062
Token a
Token ID257
Feature activation-1.41e-02

Pos logit contributions
ird+0.088
aren+0.087
rell+0.083
irst+0.083
ives+0.082
Neg logit contributions
ble-0.036
bled-0.023
 next-0.021
ster-0.017
 his-0.016
Token loud
Token ID7812
Feature activation-9.40e-03

Pos logit contributions
ird+0.022
 Karen+0.021
rell+0.020
aren+0.020
 Ellen+0.019
Neg logit contributions
ble-0.033
bled-0.031
balls-0.029
ster-0.028
 fin-0.027
Token noise
Token ID7838
Feature activation-2.71e-03

Pos logit contributions
ird+0.015
rell+0.014
 Karen+0.014
aren+0.014
 Ellen+0.013
Neg logit contributions
ble-0.022
bled-0.020
balls-0.019
ster-0.018
 fin-0.018
Token.
Token ID13
Feature activation-8.38e-02

Pos logit contributions
ird+0.004
aren+0.004
rell+0.004
 Karen+0.004
 Ellen+0.004
Neg logit contributions
ble-0.006
bled-0.006
balls-0.006
ster-0.005
 fin-0.005
Token Mad
Token ID4627
Feature activation-5.60e-02

Pos logit contributions
ird+0.166
hell+0.160
old+0.153
ited+0.152
ock+0.150
Neg logit contributions
ble-0.143
bled-0.123
balls-0.102
 next-0.100
ster-0.100
Tokeneline
Token ID4470
Feature activation+8.45e-03

Pos logit contributions
ird+0.064
rell+0.062
hell+0.058
Aud+0.055
ives+0.053
Neg logit contributions
ble-0.145
bled-0.137
ster-0.133
bing-0.119
ich-0.113
Token panicked
Token ID38225
Feature activation-8.58e-03

Pos logit contributions
ble+0.019
bled+0.018
balls+0.017
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.014
 Karen-0.013
aren-0.013
rell-0.013
 Ellen-0.012
Token and
Token ID290
Feature activation-1.17e-02

Pos logit contributions
ird+0.016
aren+0.015
rell+0.015
 Karen+0.014
old+0.013
Neg logit contributions
ble-0.019
bled-0.016
balls-0.015
ster-0.015
 fin-0.015
Token looked
Token ID3114
Feature activation-2.40e-02

Pos logit contributions
ird+0.020
 Karen+0.019
rell+0.019
aren+0.019
 Ellen+0.018
Neg logit contributions
ble-0.026
bled-0.024
balls-0.022
ster-0.022
 fin-0.022
Token to
Token ID284
Feature activation+5.75e-03

Pos logit contributions
 Karen+0.002
ird+0.002
rell+0.002
aren+0.002
 Ellen+0.002
Neg logit contributions
ble-0.003
bled-0.003
ster-0.003
 fin-0.002
balls-0.002
Token bake
Token ID28450
Feature activation-2.06e-03

Pos logit contributions
ble+0.013
bled+0.012
balls+0.011
ster+0.011
 fin+0.011
Neg logit contributions
ird-0.010
rell-0.009
 Karen-0.009
aren-0.009
 Ellen-0.008
Token cookies
Token ID14746
Feature activation-1.06e-02

Pos logit contributions
ird+0.004
aren+0.004
rell+0.004
 Karen+0.003
 Ellen+0.003
Neg logit contributions
ble-0.004
bled-0.004
balls-0.004
ster-0.004
 diving-0.003
Token and
Token ID290
Feature activation-1.50e-02

Pos logit contributions
ird+0.021
aren+0.020
rell+0.020
 Karen+0.020
 Ellen+0.019
Neg logit contributions
ble-0.021
bled-0.019
balls-0.018
ster-0.017
 fin-0.016
Token cakes
Token ID30849
Feature activation-9.29e-03

Pos logit contributions
ird+0.029
aren+0.028
rell+0.028
 Karen+0.027
 Ellen+0.026
Neg logit contributions
ble-0.030
bled-0.027
balls-0.026
ster-0.024
 fin-0.024
Token.
Token ID13
Feature activation-8.33e-02

Pos logit contributions
ird+0.017
aren+0.016
rell+0.016
 Karen+0.016
 Ellen+0.015
Neg logit contributions
ble-0.020
bled-0.018
balls-0.017
ster-0.016
 fin-0.015
Token One
Token ID1881
Feature activation-8.63e-04

Pos logit contributions
ird+0.159
ives+0.155
hell+0.152
ited+0.151
 aback+0.145
Neg logit contributions
ble-0.122
 next-0.109
balls-0.097
bled-0.088
 diving-0.088
Token day
Token ID1110
Feature activation-4.22e-02

Pos logit contributions
ird+0.002
 Karen+0.002
aren+0.002
rell+0.002
 Ellen+0.001
Neg logit contributions
ble-0.002
bled-0.002
balls-0.002
ster-0.001
 fin-0.001
Token while
Token ID981
Feature activation-1.93e-02

Pos logit contributions
ird+0.091
ives+0.086
hell+0.085
rell+0.085
old+0.084
Neg logit contributions
ble-0.069
bled-0.061
balls-0.058
 diving-0.052
ster-0.049
Token baking
Token ID16871
Feature activation+1.05e-02

Pos logit contributions
ird+0.041
rell+0.039
hell+0.038
aren+0.038
 aback+0.037
Neg logit contributions
ble-0.035
bled-0.031
balls-0.029
 diving-0.027
ster-0.027
Token,
Token ID11
Feature activation-5.38e-02

Pos logit contributions
ble+0.023
bled+0.020
balls+0.019
ster+0.019
 fin+0.018
Neg logit contributions
ird-0.019
aren-0.018
rell-0.017
 Karen-0.017
 Ellen-0.016
Token she
Token ID673
Feature activation+1.61e-02

Pos logit contributions
ird+0.020
rell+0.019
aren+0.018
 Karen+0.017
ives+0.017
Neg logit contributions
ble-0.026
bled-0.023
balls-0.022
ster-0.021
 fin-0.021
Token went
Token ID1816
Feature activation-2.36e-03

Pos logit contributions
ble+0.036
bled+0.033
balls+0.031
ster+0.031
 fin+0.030
Neg logit contributions
ird-0.028
 Karen-0.025
rell-0.025
aren-0.025
 Ellen-0.023
Token on
Token ID319
Feature activation-1.42e-02

Pos logit contributions
ird+0.004
rell+0.004
 Karen+0.004
aren+0.004
 Ellen+0.004
Neg logit contributions
ble-0.005
bled-0.005
ster-0.004
balls-0.004
 fin-0.004
Token a
Token ID257
Feature activation-3.00e-02

Pos logit contributions
ird+0.034
rell+0.031
aren+0.030
ives+0.030
hell+0.029
Neg logit contributions
ble-0.023
 diving-0.020
bled-0.019
 fin-0.018
 digging-0.018
Token search
Token ID2989
Feature activation-1.73e-02

Pos logit contributions
ird+0.051
rell+0.047
aren+0.046
 Karen+0.046
 Ellen+0.042
Neg logit contributions
ble-0.068
bled-0.060
balls-0.059
ster-0.058
 fin-0.057
Token.
Token ID13
Feature activation-8.27e-02

Pos logit contributions
ird+0.037
aren+0.035
rell+0.035
 Karen+0.033
Aud+0.033
Neg logit contributions
ble-0.031
bled-0.028
balls-0.026
ster-0.024
 diving-0.024
Token She
Token ID1375
Feature activation+1.61e-02

Pos logit contributions
hell+0.167
ird+0.165
ives+0.158
vious+0.144
rell+0.141
Neg logit contributions
ble-0.136
 next-0.104
bled-0.103
balls-0.099
ster-0.097
Token walked
Token ID6807
Feature activation-1.77e-02

Pos logit contributions
ble+0.038
bled+0.034
balls+0.032
ster+0.032
 fin+0.031
Neg logit contributions
ird-0.026
rell-0.024
 Karen-0.024
aren-0.024
 Ellen-0.022
Token around
Token ID1088
Feature activation-8.38e-03

Pos logit contributions
ird+0.030
 Karen+0.028
rell+0.028
aren+0.028
 Ellen+0.026
Neg logit contributions
ble-0.040
bled-0.036
balls-0.034
ster-0.034
 fin-0.033
Token the
Token ID262
Feature activation-6.98e-03

Pos logit contributions
ird+0.014
 Karen+0.013
aren+0.013
rell+0.013
 Ellen+0.012
Neg logit contributions
ble-0.019
bled-0.017
balls-0.016
ster-0.016
 fin-0.016
Token garden
Token ID11376
Feature activation-2.66e-03

Pos logit contributions
ird+0.010
aren+0.010
 Karen+0.010
rell+0.010
old+0.009
Neg logit contributions
ble-0.017
bled-0.016
balls-0.015
 fin-0.014
ster-0.014
Token mum
Token ID25682
Feature activation-1.68e-02

Pos logit contributions
ble+0.018
bled+0.015
balls+0.014
ster+0.014
 fin+0.014
Neg logit contributions
ird-0.026
rell-0.025
aren-0.024
 Karen-0.024
hell-0.023
Token wouldn
Token ID3636
Feature activation-2.06e-02

Pos logit contributions
rell+0.030
ird+0.029
hell+0.028
aren+0.027
 Karen+0.027
Neg logit contributions
ble-0.034
bled-0.031
ster-0.030
balls-0.029
 fin-0.027
Token't
Token ID470
Feature activation-1.36e-02

Pos logit contributions
ird+0.038
rell+0.035
 Karen+0.035
aren+0.035
 Ellen+0.032
Neg logit contributions
ble-0.044
bled-0.040
balls-0.037
ster-0.037
 fin-0.036
Token permit
Token ID8749
Feature activation-5.05e-02

Pos logit contributions
ird+0.026
aren+0.025
rell+0.024
 Karen+0.024
chairs+0.023
Neg logit contributions
ble-0.029
bled-0.024
balls-0.023
ster-0.023
 fin-0.023
Token it
Token ID340
Feature activation-2.50e-02

Pos logit contributions
ird+0.103
ives+0.093
hell+0.090
old+0.087
rell+0.087
Neg logit contributions
ble-0.092
bled-0.082
balls-0.081
 diving-0.073
 next-0.072
Token.
Token ID13
Feature activation-8.21e-02

Pos logit contributions
ird+0.050
aren+0.048
rell+0.045
 Karen+0.044
old+0.044
Neg logit contributions
ble-0.048
bled-0.042
balls-0.041
ster-0.037
 diving-0.036
Token She
Token ID1375
Feature activation-4.45e-03

Pos logit contributions
ird+0.157
hell+0.154
ock+0.148
ives+0.144
isi+0.137
Neg logit contributions
ble-0.134
balls-0.129
bled-0.118
 next-0.109
 digging-0.102
Token said
Token ID531
Feature activation-1.88e-02

Pos logit contributions
ird+0.007
rell+0.007
aren+0.007
 Karen+0.007
hell+0.006
Neg logit contributions
ble-0.010
bled-0.009
balls-0.009
ster-0.009
 fin-0.009
Token the
Token ID262
Feature activation+9.34e-03

Pos logit contributions
ird+0.030
 Karen+0.028
rell+0.027
aren+0.027
 Ellen+0.025
Neg logit contributions
ble-0.045
bled-0.041
balls-0.039
ster-0.038
 fin-0.038
Token garden
Token ID11376
Feature activation+1.21e-02

Pos logit contributions
ble+0.018
bled+0.016
balls+0.015
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.019
rell-0.018
aren-0.018
 Karen-0.018
 Ellen-0.017
Token was
Token ID373
Feature activation+8.94e-03

Pos logit contributions
ble+0.024
bled+0.021
balls+0.020
ster+0.020
 fin+0.019
Neg logit contributions
ird-0.024
rell-0.022
aren-0.022
 Karen-0.022
 Ellen-0.021
Tokeneline
Token ID4470
Feature activation-1.56e-02

Pos logit contributions
rell+0.079
ird+0.078
ives+0.075
Aud+0.074
hell+0.070
Neg logit contributions
ble-0.157
ster-0.145
bled-0.142
bing-0.129
 next-0.125
Token who
Token ID508
Feature activation-3.32e-03

Pos logit contributions
rell+0.029
ird+0.029
aren+0.027
Aud+0.027
 Karen+0.026
Neg logit contributions
ble-0.032
ster-0.028
bled-0.027
balls-0.026
 fin-0.025
Token was
Token ID373
Feature activation-9.21e-03

Pos logit contributions
ird+0.006
rell+0.006
aren+0.006
 Karen+0.006
 Ellen+0.005
Neg logit contributions
ble-0.007
bled-0.006
ster-0.005
balls-0.005
 fin-0.005
Token very
Token ID845
Feature activation+9.33e-03

Pos logit contributions
ird+0.016
 Karen+0.015
aren+0.014
rell+0.014
 Ellen+0.014
Neg logit contributions
ble-0.021
bled-0.019
balls-0.018
ster-0.018
 fin-0.018
Token thoughtful
Token ID22677
Feature activation-3.00e-02

Pos logit contributions
ble+0.021
bled+0.020
balls+0.019
ster+0.018
 fin+0.018
Neg logit contributions
ird-0.016
 Karen-0.014
aren-0.014
rell-0.014
 Stella-0.013
Token.
Token ID13
Feature activation-8.18e-02

Pos logit contributions
ird+0.058
rell+0.057
aren+0.055
Aud+0.054
Ell+0.053
Neg logit contributions
ble-0.057
ster-0.050
balls-0.049
bled-0.047
 fin-0.045
Token She
Token ID1375
Feature activation+1.49e-02

Pos logit contributions
ird+0.153
old+0.149
ives+0.147
hell+0.146
Aud+0.137
Neg logit contributions
ble-0.131
 next-0.116
bled-0.116
bing-0.092
balls-0.091
Token loved
Token ID6151
Feature activation-6.56e-03

Pos logit contributions
ble+0.037
bled+0.033
balls+0.032
ster+0.031
 fin+0.031
Neg logit contributions
ird-0.023
 Karen-0.021
aren-0.021
rell-0.021
 Ellen-0.019
Token to
Token ID284
Feature activation-7.06e-03

Pos logit contributions
 Karen+0.011
ird+0.011
rell+0.010
aren+0.010
 Ellen+0.010
Neg logit contributions
ble-0.015
bled-0.014
ster-0.013
 fin-0.012
balls-0.012
Token do
Token ID466
Feature activation-9.40e-03

Pos logit contributions
ird+0.013
rell+0.012
 Karen+0.012
aren+0.012
 Ellen+0.011
Neg logit contributions
ble-0.016
bled-0.014
balls-0.013
ster-0.013
 fin-0.013
Token nice
Token ID3621
Feature activation+4.09e-03

Pos logit contributions
ird+0.019
aren+0.018
rell+0.018
 Karen+0.018
 Stella+0.016
Neg logit contributions
ble-0.018
bled-0.016
balls-0.015
ster-0.015
 fin-0.014
Token was
Token ID373
Feature activation-1.00e-02

Pos logit contributions
ird+0.048
rell+0.045
aren+0.044
Aud+0.043
ives+0.043
Neg logit contributions
ble-0.032
bled-0.027
balls-0.024
ster-0.023
 fin-0.023
Token a
Token ID257
Feature activation+3.45e-03

Pos logit contributions
ird+0.022
rell+0.021
aren+0.021
 Karen+0.020
chairs+0.019
Neg logit contributions
ble-0.018
bled-0.016
 fin-0.014
balls-0.014
 diving-0.013
Token girl
Token ID2576
Feature activation-1.30e-02

Pos logit contributions
ble+0.009
bled+0.009
balls+0.008
ster+0.007
 fin+0.007
Neg logit contributions
ird-0.005
 Karen-0.004
aren-0.004
 Stella-0.004
 Ellen-0.004
Token named
Token ID3706
Feature activation-2.26e-02

Pos logit contributions
ird+0.021
rell+0.020
aren+0.019
 Karen+0.019
 Ellen+0.018
Neg logit contributions
ble-0.030
bled-0.027
ster-0.026
balls-0.026
 fin-0.026
Token Karen
Token ID18678
Feature activation-3.41e-02

Pos logit contributions
rell+0.050
ird+0.049
hell+0.048
chairs+0.046
 aback+0.046
Neg logit contributions
ble-0.033
bled-0.029
balls-0.028
 next-0.026
 digging-0.025
Token.
Token ID13
Feature activation-8.18e-02

Pos logit contributions
rell+0.080
ird+0.076
Aud+0.074
ives+0.073
 aback+0.072
Neg logit contributions
ble-0.051
ster-0.043
 next-0.038
bled-0.037
 fin-0.034
Token She
Token ID1375
Feature activation+2.95e-02

Pos logit contributions
ird+0.183
old+0.156
ives+0.155
hell+0.151
lier+0.143
Neg logit contributions
ble-0.114
 next-0.100
bled-0.092
balls-0.087
 diving-0.078
Token wore
Token ID12408
Feature activation-2.11e-02

Pos logit contributions
ble+0.078
bled+0.073
balls+0.069
ster+0.066
 fin+0.066
Neg logit contributions
ird-0.039
 Karen-0.035
aren-0.034
rell-0.034
Ell-0.032
Token a
Token ID257
Feature activation-2.59e-02

Pos logit contributions
ird+0.049
aren+0.047
old+0.043
rell+0.043
 Karen+0.043
Neg logit contributions
ble-0.036
bled-0.033
balls-0.029
 fin-0.027
ster-0.027
Token cozy
Token ID37438
Feature activation+3.84e-03

Pos logit contributions
ird+0.045
rell+0.042
aren+0.042
 Karen+0.042
 Ellen+0.038
Neg logit contributions
ble-0.058
bled-0.052
balls-0.049
ster-0.049
 fin-0.048
Token scarf
Token ID36953
Feature activation-1.26e-02

Pos logit contributions
ble+0.009
bled+0.008
balls+0.008
ster+0.007
 fin+0.007
Neg logit contributions
ird-0.006
rell-0.006
aren-0.006
 Karen-0.006
 Ellen-0.005
Token a
Token ID257
Feature activation-9.95e-03

Pos logit contributions
ird+0.038
rell+0.035
aren+0.033
hell+0.032
 Karen+0.032
Neg logit contributions
ble-0.048
bled-0.043
balls-0.041
ster-0.039
 fin-0.039
Token voice
Token ID3809
Feature activation-4.61e-02

Pos logit contributions
ird+0.018
rell+0.017
aren+0.016
 Karen+0.016
 Ellen+0.015
Neg logit contributions
ble-0.022
bled-0.019
balls-0.019
ster-0.018
 fin-0.018
Token said
Token ID531
Feature activation-1.28e-02

Pos logit contributions
ird+0.081
rell+0.080
aren+0.074
hell+0.074
Aud+0.071
Neg logit contributions
ble-0.098
bled-0.086
ster-0.081
 fin-0.078
 shark-0.076
Token "
Token ID366
Feature activation-2.79e-02

Pos logit contributions
ird+0.023
rell+0.021
aren+0.021
 Karen+0.020
 Ellen+0.019
Neg logit contributions
ble-0.029
bled-0.026
balls-0.024
ster-0.024
 fin-0.024
TokenMine
Token ID24461
Feature activation+3.80e-03

Pos logit contributions
ird+0.050
rell+0.045
aren+0.043
 Karen+0.043
hell+0.042
Neg logit contributions
ble-0.060
bled-0.052
ster-0.049
balls-0.049
 fin-0.046
Token!".
Token ID48220
Feature activation-8.13e-02

Pos logit contributions
ble+0.010
bled+0.010
balls+0.009
 fin+0.009
ster+0.009
Neg logit contributions
ird-0.005
 Karen-0.004
rell-0.004
aren-0.004
 Ellen-0.004
Token A
Token ID317
Feature activation-1.97e-02

Pos logit contributions
ird+0.150
hell+0.135
rell+0.134
irst+0.128
ock+0.126
Neg logit contributions
ble-0.158
bled-0.132
balls-0.130
 fin-0.121
 next-0.121
Token small
Token ID1402
Feature activation-2.22e-02

Pos logit contributions
ird+0.040
rell+0.039
aren+0.037
 Karen+0.036
hell+0.035
Neg logit contributions
ble-0.038
 fin-0.032
ster-0.032
bled-0.032
balls-0.031
Token girl
Token ID2576
Feature activation-3.49e-02

Pos logit contributions
ird+0.045
rell+0.044
aren+0.041
 Karen+0.041
hell+0.039
Neg logit contributions
ble-0.042
 fin-0.036
ster-0.036
bled-0.035
 shark-0.035
Token with
Token ID351
Feature activation-3.22e-02

Pos logit contributions
rell+0.071
ird+0.067
aren+0.064
hell+0.063
ives+0.062
Neg logit contributions
ble-0.065
bled-0.056
ster-0.055
 fin-0.055
balls-0.052
Token curly
Token ID45731
Feature activation-3.97e-02

Pos logit contributions
ird+0.073
rell+0.070
aren+0.070
chairs+0.065
 Karen+0.065
Neg logit contributions
ble-0.052
 fin-0.047
bled-0.047
 shark-0.040
ster-0.039
Token was
Token ID373
Feature activation+1.02e-02

Pos logit contributions
ird+0.038
aren+0.036
rell+0.036
 Karen+0.035
old+0.033
Neg logit contributions
ble-0.033
bled-0.029
balls-0.028
ster-0.027
 fin-0.026
Token time
Token ID640
Feature activation-8.04e-03

Pos logit contributions
ble+0.023
bled+0.020
balls+0.019
ster+0.019
 fin+0.019
Neg logit contributions
ird-0.018
rell-0.017
aren-0.017
 Karen-0.017
 Ellen-0.015
Token to
Token ID284
Feature activation-9.10e-03

Pos logit contributions
ird+0.016
 Karen+0.015
aren+0.015
rell+0.015
 Ellen+0.014
Neg logit contributions
ble-0.016
bled-0.014
balls-0.013
ster-0.013
 fin-0.013
Token go
Token ID467
Feature activation-6.82e-03

Pos logit contributions
ird+0.018
rell+0.017
aren+0.017
 Karen+0.016
chairs+0.016
Neg logit contributions
ble-0.019
bled-0.016
 fin-0.015
ster-0.015
balls-0.015
Token home
Token ID1363
Feature activation-2.10e-04

Pos logit contributions
ird+0.013
rell+0.012
 Karen+0.012
aren+0.012
 Ellen+0.011
Neg logit contributions
ble-0.015
bled-0.013
balls-0.012
ster-0.012
 fin-0.012
Token.
Token ID13
Feature activation-8.12e-02

Pos logit contributions
ird+0.000
rell+0.000
aren+0.000
 Karen+0.000
 Ellen+0.000
Neg logit contributions
ble-0.000
bled-0.000
balls-0.000
ster-0.000
 fin-0.000
Token So
Token ID1406
Feature activation-5.68e-02

Pos logit contributions
hell+0.141
ird+0.126
ock+0.124
ives+0.119
ha+0.109
Neg logit contributions
ble-0.145
balls-0.142
bled-0.135
 next-0.131
bing-0.123
Token Karen
Token ID18678
Feature activation-5.15e-03

Pos logit contributions
ives+0.106
hell+0.105
ird+0.103
rell+0.101
old+0.097
Neg logit contributions
ble-0.108
bled-0.098
balls-0.090
 next-0.087
 diving-0.082
Token and
Token ID290
Feature activation-1.19e-02

Pos logit contributions
ird+0.009
rell+0.009
 Karen+0.008
aren+0.008
 Ellen+0.008
Neg logit contributions
ble-0.011
bled-0.010
balls-0.010
ster-0.010
 fin-0.009
Token her
Token ID607
Feature activation+1.30e-02

Pos logit contributions
ird+0.020
rell+0.019
aren+0.018
hell+0.018
ives+0.017
Neg logit contributions
ble-0.027
bled-0.024
balls-0.023
ster-0.022
 fin-0.022
Token mom
Token ID1995
Feature activation-4.29e-02

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
ster+0.015
 fin+0.014
Neg logit contributions
ird-0.032
rell-0.031
aren-0.031
 Karen-0.031
 Ellen-0.029
Token big
Token ID1263
Feature activation-2.93e-02

Pos logit contributions
ird+0.029
 Karen+0.026
aren+0.026
rell+0.025
 Ellen+0.024
Neg logit contributions
ble-0.038
bled-0.035
balls-0.033
ster-0.032
 fin-0.031
Token fan
Token ID4336
Feature activation-9.45e-03

Pos logit contributions
ird+0.058
rell+0.055
aren+0.053
 Karen+0.053
hell+0.051
Neg logit contributions
ble-0.056
 fin-0.048
ster-0.048
bled-0.048
 shark-0.047
Token of
Token ID286
Feature activation-2.47e-02

Pos logit contributions
ird+0.014
 Karen+0.013
rell+0.013
aren+0.012
 Ellen+0.011
Neg logit contributions
ble-0.024
bled-0.022
balls-0.020
ster-0.020
 fin-0.020
Token fire
Token ID2046
Feature activation+1.22e-02

Pos logit contributions
ird+0.044
rell+0.041
aren+0.041
 Karen+0.040
hell+0.037
Neg logit contributions
ble-0.055
bled-0.050
balls-0.047
ster-0.045
 fin-0.045
Tokenmen
Token ID3653
Feature activation-1.76e-02

Pos logit contributions
ble+0.033
bled+0.031
balls+0.028
ster+0.028
 fin+0.028
Neg logit contributions
ird-0.015
rell-0.014
 Karen-0.014
aren-0.013
chairs-0.012
Token.
Token ID13
Feature activation-8.11e-02

Pos logit contributions
ird+0.028
rell+0.026
aren+0.026
 Karen+0.025
 Ellen+0.023
Neg logit contributions
ble-0.041
bled-0.037
balls-0.036
ster-0.035
 fin-0.035
Token She
Token ID1375
Feature activation+1.66e-02

Pos logit contributions
ird+0.185
hell+0.171
ives+0.166
ited+0.156
ha+0.153
Neg logit contributions
 next-0.091
ble-0.085
balls-0.078
 Then-0.072
 He-0.064
Token often
Token ID1690
Feature activation-2.73e-03

Pos logit contributions
ble+0.038
bled+0.034
balls+0.032
ster+0.032
 fin+0.031
Neg logit contributions
ird-0.029
rell-0.026
 Karen-0.026
aren-0.026
 Ellen-0.024
Token asked
Token ID1965
Feature activation-3.27e-02

Pos logit contributions
ird+0.005
rell+0.005
aren+0.005
 Karen+0.005
 Ellen+0.004
Neg logit contributions
ble-0.006
bled-0.005
balls-0.005
ster-0.005
 fin-0.005
Token her
Token ID607
Feature activation-1.37e-04

Pos logit contributions
ird+0.067
rell+0.065
aren+0.064
hell+0.063
Aud+0.059
Neg logit contributions
ble-0.054
bled-0.051
balls-0.047
ster-0.042
 next-0.042
Token daddy
Token ID35822
Feature activation-1.62e-02

Pos logit contributions
ird+0.000
rell+0.000
aren+0.000
 Karen+0.000
hell+0.000
Neg logit contributions
ble-0.000
bled-0.000
balls-0.000
ster-0.000
 fin-0.000
Token veil
Token ID30615
Feature activation+6.21e-03

Pos logit contributions
ble+0.007
bled+0.006
balls+0.006
ster+0.006
 fin+0.006
Neg logit contributions
ird-0.005
rell-0.004
 Karen-0.004
aren-0.004
 Ellen-0.004
Token on
Token ID319
Feature activation-1.72e-02

Pos logit contributions
ble+0.012
bled+0.011
balls+0.010
 fin+0.010
ster+0.010
Neg logit contributions
ird-0.013
rell-0.012
 Karen-0.012
aren-0.011
 Ellen-0.011
Token the
Token ID262
Feature activation-1.94e-02

Pos logit contributions
ird+0.038
aren+0.035
rell+0.034
old+0.033
 Karen+0.033
Neg logit contributions
ble-0.032
bled-0.027
 fin-0.025
balls-0.024
ster-0.024
Token kitchen
Token ID9592
Feature activation+8.92e-03

Pos logit contributions
ird+0.041
aren+0.038
 Karen+0.038
rell+0.038
 Ellen+0.036
Neg logit contributions
ble-0.037
bled-0.032
balls-0.030
ster-0.030
 fin-0.029
Token table
Token ID3084
Feature activation+1.94e-02

Pos logit contributions
ble+0.020
bled+0.019
balls+0.017
ster+0.017
 fin+0.017
Neg logit contributions
ird-0.015
rell-0.014
 Karen-0.014
aren-0.014
 Ellen-0.012
Token.
Token ID13
Feature activation-8.00e-02

Pos logit contributions
ble+0.051
bled+0.047
balls+0.045
ster+0.044
 fin+0.044
Neg logit contributions
ird-0.026
rell-0.023
aren-0.022
 Karen-0.022
hell-0.020
Token She
Token ID1375
Feature activation+5.56e-03

Pos logit contributions
ird+0.168
ives+0.149
hell+0.148
old+0.143
ited+0.142
Neg logit contributions
ble-0.133
bled-0.106
balls-0.102
 next-0.093
 diving-0.086
Token asked
Token ID1965
Feature activation-4.07e-02

Pos logit contributions
ble+0.013
bled+0.012
balls+0.012
ster+0.011
 fin+0.011
Neg logit contributions
ird-0.009
 Karen-0.008
aren-0.008
rell-0.008
 Ellen-0.007
Token her
Token ID607
Feature activation+9.28e-03

Pos logit contributions
ives+0.096
ird+0.095
rell+0.092
hell+0.090
ited+0.088
Neg logit contributions
ble-0.059
bled-0.054
 next-0.042
 his-0.041
balls-0.038
Token mom
Token ID1995
Feature activation-3.23e-02

Pos logit contributions
ble+0.012
ster+0.010
bled+0.010
balls+0.009
 fin+0.008
Neg logit contributions
ird-0.025
rell-0.025
aren-0.024
 Karen-0.023
chairs-0.022
Token,
Token ID11
Feature activation+1.20e-03

Pos logit contributions
ird+0.050
rell+0.045
aren+0.045
 Karen+0.044
 Ellen+0.041
Neg logit contributions
ble-0.078
bled-0.071
balls-0.068
ster-0.068
 fin-0.065
Token very
Token ID845
Feature activation+2.68e-02

Pos logit contributions
ird+0.014
 Karen+0.012
aren+0.012
rell+0.012
 Ellen+0.011
Neg logit contributions
ble-0.020
bled-0.019
balls-0.018
ster-0.017
 fin-0.016
Token happy
Token ID3772
Feature activation+1.69e-02

Pos logit contributions
ble+0.066
bled+0.061
balls+0.057
ster+0.055
 fin+0.055
Neg logit contributions
ird-0.041
aren-0.037
 Karen-0.037
 Stella-0.035
old-0.034
Token and
Token ID290
Feature activation+5.83e-03

Pos logit contributions
ble+0.039
bled+0.036
balls+0.034
ster+0.032
 fin+0.032
Neg logit contributions
ird-0.028
 Karen-0.026
aren-0.026
rell-0.026
 Ellen-0.024
Token playful
Token ID34264
Feature activation-1.42e-02

Pos logit contributions
ble+0.014
bled+0.013
balls+0.012
ster+0.012
 fin+0.012
Neg logit contributions
ird-0.009
 Karen-0.008
rell-0.008
aren-0.008
 Ellen-0.008
Token girl
Token ID2576
Feature activation-1.79e-02

Pos logit contributions
ird+0.026
rell+0.024
aren+0.024
 Karen+0.023
hell+0.022
Neg logit contributions
ble-0.029
bled-0.025
balls-0.025
ster-0.025
 fin-0.025
Token.
Token ID13
Feature activation-7.96e-02

Pos logit contributions
ird+0.029
rell+0.027
aren+0.027
 Karen+0.026
hell+0.025
Neg logit contributions
ble-0.041
bled-0.037
balls-0.035
ster-0.035
 fin-0.035
Token She
Token ID1375
Feature activation+1.63e-02

Pos logit contributions
ird+0.161
hell+0.144
ives+0.139
ha+0.139
rell+0.134
Neg logit contributions
 next-0.115
ble-0.104
balls-0.097
bled-0.095
 diving-0.090
Token loved
Token ID6151
Feature activation+6.46e-03

Pos logit contributions
ble+0.037
bled+0.033
balls+0.031
ster+0.031
 fin+0.030
Neg logit contributions
ird-0.028
rell-0.026
 Karen-0.026
aren-0.026
 Ellen-0.024
Token to
Token ID284
Feature activation-7.91e-03

Pos logit contributions
ble+0.016
bled+0.014
ster+0.013
 fin+0.012
balls+0.012
Neg logit contributions
 Karen-0.010
rell-0.010
ird-0.010
 Ellen-0.009
 Stella-0.009
Token play
Token ID711
Feature activation+2.83e-03

Pos logit contributions
ird+0.014
rell+0.013
aren+0.013
 Karen+0.013
 Ellen+0.012
Neg logit contributions
ble-0.017
bled-0.015
balls-0.015
ster-0.014
 fin-0.014
Token with
Token ID351
Feature activation-1.90e-02

Pos logit contributions
ble+0.006
bled+0.005
ster+0.005
balls+0.005
 fin+0.005
Neg logit contributions
ird-0.005
rell-0.005
 Karen-0.005
aren-0.005
 Ellen-0.005
Token the
Token ID262
Feature activation-1.25e-03

Pos logit contributions
ird+0.110
rell+0.103
hell+0.099
ives+0.099
aren+0.095
Neg logit contributions
ble-0.099
bled-0.091
balls-0.084
 diving-0.078
ster-0.077
Token hallway
Token ID23959
Feature activation-3.32e-03

Pos logit contributions
ird+0.002
aren+0.002
rell+0.002
 Karen+0.002
 Stella+0.002
Neg logit contributions
ble-0.003
bled-0.002
balls-0.002
 fin-0.002
ster-0.002
Token seemed
Token ID3947
Feature activation-1.41e-02

Pos logit contributions
ird+0.007
aren+0.006
rell+0.006
 Karen+0.006
 Ellen+0.006
Neg logit contributions
ble-0.006
bled-0.006
balls-0.005
ster-0.005
 fin-0.005
Token less
Token ID1342
Feature activation-1.13e-02

Pos logit contributions
ird+0.026
rell+0.024
aren+0.024
 Karen+0.024
 Ellen+0.022
Neg logit contributions
ble-0.030
bled-0.027
balls-0.025
ster-0.025
 fin-0.025
Token scary
Token ID14343
Feature activation-1.19e-02

Pos logit contributions
ird+0.020
aren+0.019
 Karen+0.019
rell+0.019
 Ellen+0.017
Neg logit contributions
ble-0.025
bled-0.022
 fin-0.021
balls-0.021
ster-0.020
Token.
Token ID13
Feature activation-7.89e-02

Pos logit contributions
ird+0.019
aren+0.017
rell+0.017
 Karen+0.017
 Ellen+0.016
Neg logit contributions
ble-0.028
bled-0.026
balls-0.025
ster-0.024
 fin-0.024
Token
Token ID198
Feature activation-6.08e-02

Pos logit contributions
ird+0.159
hell+0.156
ives+0.153
ock+0.147
ited+0.145
Neg logit contributions
ble-0.113
balls-0.098
bled-0.096
 Then-0.095
 next-0.095
Token
Token ID198
Feature activation-5.56e-02

Pos logit contributions
rell+0.147
ird+0.144
ives+0.142
vious+0.138
hell+0.138
Neg logit contributions
ble-0.081
bled-0.059
 next-0.056
balls-0.052
 diving-0.044
TokenShe
Token ID3347
Feature activation+1.18e-02

Pos logit contributions
ives+0.095
ird+0.094
hell+0.092
rell+0.092
forts+0.081
Neg logit contributions
ble-0.111
bled-0.098
 next-0.091
balls-0.088
bing-0.088
Token t
Token ID256
Feature activation-1.24e-02

Pos logit contributions
ble+0.026
bled+0.024
balls+0.023
ster+0.022
 fin+0.022
Neg logit contributions
ird-0.020
rell-0.019
aren-0.019
 Karen-0.019
 Ellen-0.017
Tokenipt
Token ID10257
Feature activation-2.03e-02

Pos logit contributions
ird+0.034
rell+0.032
 Karen+0.032
aren+0.032
 Ellen+0.031
Neg logit contributions
ble-0.016
bled-0.013
balls-0.011
ster-0.011
 fin-0.010
Token Lucy
Token ID22162
Feature activation-3.69e-03

Pos logit contributions
rell+0.027
hell+0.024
ird+0.024
chairs+0.023
ives+0.023
Neg logit contributions
ble-0.031
bled-0.029
 Mitch-0.027
balls-0.027
 Shark-0.026
Token went
Token ID1816
Feature activation-1.01e-02

Pos logit contributions
ird+0.008
rell+0.008
aren+0.007
 Karen+0.007
hell+0.007
Neg logit contributions
ble-0.006
bled-0.005
ster-0.005
 fin-0.005
balls-0.005
Token for
Token ID329
Feature activation-3.79e-02

Pos logit contributions
ird+0.017
rell+0.016
 Karen+0.016
 Ellen+0.015
aren+0.015
Neg logit contributions
ble-0.023
bled-0.022
ster-0.020
balls-0.020
 fin-0.019
Token a
Token ID257
Feature activation-4.43e-02

Pos logit contributions
ird+0.096
rell+0.094
hell+0.093
ives+0.090
aren+0.090
Neg logit contributions
ble-0.044
 diving-0.038
bled-0.036
 digging-0.035
balls-0.034
Token walk
Token ID2513
Feature activation-2.25e-02

Pos logit contributions
ird+0.065
rell+0.061
aren+0.059
 Karen+0.058
 Ellen+0.053
Neg logit contributions
ble-0.108
bled-0.098
balls-0.096
ster-0.094
 fin-0.093
Token.
Token ID13
Feature activation-7.89e-02

Pos logit contributions
ird+0.046
rell+0.043
aren+0.043
 Karen+0.041
hell+0.040
Neg logit contributions
ble-0.043
bled-0.037
balls-0.035
ster-0.034
 fin-0.034
Token She
Token ID1375
Feature activation+5.66e-03

Pos logit contributions
ird+0.166
hell+0.150
ha+0.147
Aud+0.138
ock+0.138
Neg logit contributions
ble-0.113
balls-0.109
bled-0.103
 next-0.095
 digging-0.094
Token saw
Token ID2497
Feature activation-1.96e-02

Pos logit contributions
ble+0.012
bled+0.011
balls+0.010
ster+0.010
 fin+0.010
Neg logit contributions
ird-0.010
rell-0.009
aren-0.009
 Karen-0.009
hell-0.009
Token lots
Token ID6041
Feature activation-3.48e-02

Pos logit contributions
ird+0.047
rell+0.045
aren+0.043
hell+0.040
ives+0.040
Neg logit contributions
ble-0.032
bled-0.028
balls-0.024
ster-0.024
 fin-0.023
Token of
Token ID286
Feature activation-4.66e-02

Pos logit contributions
ird+0.068
aren+0.059
 Karen+0.057
rell+0.057
Ell+0.056
Neg logit contributions
ble-0.069
 diving-0.058
bled-0.057
 shark-0.056
balls-0.056
Token interesting
Token ID3499
Feature activation-2.39e-02

Pos logit contributions
ird+0.107
rell+0.097
aren+0.095
Aud+0.092
ives+0.090
Neg logit contributions
ble-0.077
 shark-0.064
 diving-0.062
 fin-0.060
 digging-0.058
Token back
Token ID736
Feature activation+5.07e-03

Pos logit contributions
ble+0.007
bled+0.006
balls+0.006
ster+0.006
 fin+0.006
Neg logit contributions
ird-0.006
rell-0.006
 Karen-0.006
aren-0.006
 Ellen-0.005
Token to
Token ID284
Feature activation+1.57e-02

Pos logit contributions
ble+0.010
bled+0.008
balls+0.008
ster+0.008
 fin+0.008
Neg logit contributions
ird-0.011
rell-0.010
aren-0.010
 Karen-0.010
hell-0.009
Token his
Token ID465
Feature activation+3.66e-02

Pos logit contributions
ble+0.033
bled+0.032
balls+0.029
ster+0.028
 fin+0.025
Neg logit contributions
 Karen-0.028
ird-0.027
rell-0.026
 Stella-0.026
 Ellen-0.026
Token family
Token ID1641
Feature activation-1.42e-02

Pos logit contributions
bled+0.084
ble+0.083
balls+0.072
ster+0.067
bling+0.065
Neg logit contributions
ird-0.049
 Karen-0.048
aren-0.046
 Eve-0.046
rell-0.045
Token.
Token ID13
Feature activation+1.65e-02

Pos logit contributions
ird+0.025
rell+0.024
aren+0.024
 Karen+0.023
 Ellen+0.021
Neg logit contributions
ble-0.031
bled-0.027
balls-0.027
ster-0.026
 fin-0.026
Token While
Token ID2893
Feature activation+6.32e-02

Pos logit contributions
ble+0.050
bled+0.046
iflower+0.041
fish+0.041
ster+0.041
Neg logit contributions
rell-0.015
ird-0.014
 Karen-0.014
 aback-0.014
 Ellen-0.014
Token he
Token ID339
Feature activation+3.27e-02

Pos logit contributions
ble+0.156
bled+0.134
ster+0.132
castle+0.132
urg+0.127
Neg logit contributions
 Ellie-0.079
ird-0.074
 Ellen-0.073
 Stella-0.072
 Karen-0.070
Token was
Token ID373
Feature activation-6.37e-03

Pos logit contributions
ble+0.077
bled+0.069
balls+0.068
ster+0.064
 fin+0.061
Neg logit contributions
ird-0.051
 Karen-0.045
aren-0.044
rell-0.044
ha-0.040
Token swimming
Token ID14899
Feature activation+1.31e-02

Pos logit contributions
ird+0.013
aren+0.012
rell+0.012
 Karen+0.012
 Ellen+0.012
Neg logit contributions
ble-0.012
bled-0.011
balls-0.010
ster-0.010
 fin-0.010
Token,
Token ID11
Feature activation+5.97e-02

Pos logit contributions
ble+0.031
bled+0.029
balls+0.027
ster+0.027
 fin+0.026
Neg logit contributions
ird-0.021
rell-0.019
 Karen-0.019
aren-0.019
 Ellen-0.017
Token he
Token ID339
Feature activation+4.57e-02

Pos logit contributions
ble+0.136
ster+0.125
castle+0.125
bled+0.121
iflower+0.118
Neg logit contributions
 Ellie-0.074
 Ellen-0.068
ird-0.067
 Beat-0.067
 Stella-0.067
Token ocean
Token ID9151
Feature activation+5.32e-02

Pos logit contributions
ble+0.092
bled+0.091
balls+0.081
ster+0.071
 fin+0.065
Neg logit contributions
ird-0.093
rell-0.093
aren-0.082
 Eve-0.080
hell-0.080
Token.
Token ID13
Feature activation+2.71e-02

Pos logit contributions
ble+0.137
bled+0.135
ster+0.117
balls+0.116
 fin+0.112
Neg logit contributions
ird-0.060
rell-0.056
hell-0.054
 Eve-0.050
he-0.048
Token One
Token ID1881
Feature activation+4.28e-02

Pos logit contributions
ble+0.075
bled+0.071
ster+0.066
balls+0.066
 fin+0.063
Neg logit contributions
rell-0.032
ird-0.031
 Karen-0.029
aren-0.027
 Stella-0.026
Token day
Token ID1110
Feature activation+3.46e-02

Pos logit contributions
bled+0.113
ble+0.112
balls+0.107
ster+0.100
bing+0.091
Neg logit contributions
rell-0.044
ird-0.043
 Karen-0.042
aren-0.042
old-0.041
Token,
Token ID11
Feature activation+3.35e-02

Pos logit contributions
ble+0.095
bled+0.087
ster+0.083
balls+0.082
 fin+0.081
Neg logit contributions
ird-0.041
rell-0.038
 Karen-0.036
aren-0.034
 Stella-0.034
Token Nem
Token ID22547
Feature activation+5.93e-02

Pos logit contributions
ble+0.072
bled+0.065
ster+0.064
balls+0.063
 fin+0.061
Neg logit contributions
 Stella-0.053
 Karen-0.051
 Mad-0.048
rell-0.047
 Ellie-0.047
Tokeno
Token ID78
Feature activation+5.80e-02

Pos logit contributions
ble+0.132
bled+0.131
balls+0.115
 fin+0.105
 diving+0.103
Neg logit contributions
ird-0.091
aren-0.086
old-0.084
rell-0.083
ives-0.081
Token saw
Token ID2497
Feature activation+3.93e-02

Pos logit contributions
ble+0.156
bled+0.146
balls+0.137
ster+0.129
 fin+0.125
Neg logit contributions
ird-0.074
rell-0.065
aren-0.062
 Karen-0.061
old-0.059
Token a
Token ID257
Feature activation+3.16e-02

Pos logit contributions
ble+0.092
bled+0.086
balls+0.085
ster+0.082
 fin+0.076
Neg logit contributions
 Karen-0.051
rell-0.050
ird-0.049
old-0.047
 Stella-0.045
Token plastic
Token ID7309
Feature activation+6.34e-02

Pos logit contributions
bled+0.085
ble+0.078
balls+0.075
bing+0.069
ster+0.066
Neg logit contributions
ird-0.030
hell-0.030
ha-0.030
 Karen-0.029
aren-0.027
Token bag
Token ID6131
Feature activation+2.40e-02

Pos logit contributions
bled+0.122
ble+0.103
bing+0.095
 Miles+0.093
balls+0.093
Neg logit contributions
ird-0.101
 keys-0.095
hell-0.094
 bars-0.092
 umbrella-0.092
Tokenery
Token ID1924
Feature activation+5.48e-02

Pos logit contributions
ble+0.063
bled+0.056
 fin+0.055
balls+0.054
ster+0.053
Neg logit contributions
ird-0.070
rell-0.070
hell-0.068
aren-0.064
ives-0.064
Token became
Token ID2627
Feature activation+1.07e-02

Pos logit contributions
ble+0.127
bled+0.118
balls+0.109
 fin+0.106
ster+0.105
Neg logit contributions
rell-0.077
ird-0.076
old-0.067
hell-0.067
ives-0.067
Token one
Token ID530
Feature activation-8.63e-03

Pos logit contributions
ble+0.024
bled+0.022
balls+0.021
ster+0.020
 fin+0.020
Neg logit contributions
ird-0.019
 Karen-0.017
rell-0.017
aren-0.017
 Ellen-0.016
Token of
Token ID286
Feature activation+2.34e-03

Pos logit contributions
ird+0.009
aren+0.008
 Karen+0.008
rell+0.008
old+0.007
Neg logit contributions
ble-0.025
bled-0.024
balls-0.023
ster-0.022
 fin-0.022
Token her
Token ID607
Feature activation+3.19e-02

Pos logit contributions
ble+0.005
bled+0.005
balls+0.005
ster+0.005
 fin+0.005
Neg logit contributions
ird-0.004
 Karen-0.004
rell-0.004
aren-0.004
 Ellen-0.003
Token favourite
Token ID12507
Feature activation+5.00e-02

Pos logit contributions
ble+0.069
bled+0.064
balls+0.057
 fin+0.054
ster+0.054
Neg logit contributions
ird-0.056
aren-0.051
 Karen-0.051
rell-0.050
old-0.048
Token snacks
Token ID28487
Feature activation+5.52e-03

Pos logit contributions
ble+0.104
bled+0.100
balls+0.087
ster+0.081
 fin+0.077
Neg logit contributions
ird-0.090
rell-0.085
hell-0.085
 Karen-0.080
aren-0.079
Token.
Token ID13
Feature activation-2.65e-02

Pos logit contributions
ble+0.013
bled+0.011
balls+0.011
ster+0.011
 fin+0.010
Neg logit contributions
ird-0.009
rell-0.009
 Karen-0.009
aren-0.009
 Ellen-0.008
Token
Token ID220
Feature activation-1.23e-02

Pos logit contributions
ird+0.053
 keys+0.048
rell+0.048
Aud+0.046
hell+0.046
Neg logit contributions
ble-0.041
balls-0.037
 Miles-0.036
emo-0.036
 Mitch-0.034
Token
Token ID198
Feature activation-1.60e-02

Pos logit contributions
ird+0.022
aren+0.021
rell+0.021
 Karen+0.021
 Ellen+0.020
Neg logit contributions
ble-0.026
bled-0.023
balls-0.022
ster-0.020
emo-0.020
Token
Token ID198
Feature activation-1.05e-02

Pos logit contributions
ird+0.032
rell+0.030
aren+0.028
 Karen+0.028
 Ellen+0.027
Neg logit contributions
ble-0.031
balls-0.027
bled-0.027
emo-0.025
ster-0.024
Token careful
Token ID8161
Feature activation+1.49e-02

Pos logit contributions
ble+0.064
bled+0.060
ster+0.058
 fin+0.054
 shark+0.052
Neg logit contributions
ird-0.032
 Karen-0.031
 Alice-0.030
rell-0.028
 Stella-0.028
Token not
Token ID407
Feature activation+2.67e-02

Pos logit contributions
ble+0.031
bled+0.029
ster+0.027
 fin+0.026
 shark+0.025
Neg logit contributions
ird-0.025
rell-0.025
 Karen-0.024
hell-0.023
aren-0.023
Token to
Token ID284
Feature activation+3.07e-02

Pos logit contributions
ble+0.066
bled+0.065
ster+0.057
 fin+0.054
balls+0.054
Neg logit contributions
ird-0.034
rell-0.032
 Karen-0.031
hell-0.030
old-0.030
Token touch
Token ID3638
Feature activation+2.11e-02

Pos logit contributions
ble+0.065
bled+0.065
ster+0.058
balls+0.058
 fin+0.054
Neg logit contributions
ird-0.054
rell-0.049
 Karen-0.049
aren-0.047
 Stella-0.045
Token the
Token ID262
Feature activation+3.63e-02

Pos logit contributions
ble+0.047
bled+0.043
balls+0.040
ster+0.040
 fin+0.039
Neg logit contributions
ird-0.037
rell-0.034
 Karen-0.034
aren-0.034
 Ellen-0.031
Token hot
Token ID3024
Feature activation+4.92e-02

Pos logit contributions
ble+0.069
bled+0.064
balls+0.056
 fin+0.055
ster+0.055
Neg logit contributions
ird-0.072
rell-0.069
 Karen-0.066
aren-0.066
hell-0.064
Token stove
Token ID27635
Feature activation+2.75e-02

Pos logit contributions
ble+0.101
bled+0.095
balls+0.085
ster+0.082
 fin+0.082
Neg logit contributions
ird-0.092
rell-0.089
 Karen-0.084
aren-0.084
hell-0.079
Token.
Token ID13
Feature activation+7.75e-03

Pos logit contributions
ble+0.075
bled+0.072
balls+0.065
ster+0.065
 fin+0.064
Neg logit contributions
ird-0.032
rell-0.028
 Karen-0.027
hell-0.026
aren-0.025
Token
Token ID198
Feature activation+1.13e-02

Pos logit contributions
ble+0.018
bled+0.017
balls+0.016
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.013
 Karen-0.012
rell-0.012
aren-0.012
 Ellen-0.011
Token
Token ID198
Feature activation+2.97e-02

Pos logit contributions
ble+0.026
bled+0.023
balls+0.022
 fin+0.021
ster+0.021
Neg logit contributions
ird-0.020
rell-0.019
aren-0.018
 Karen-0.018
 Ellen-0.017
TokenThey
Token ID2990
Feature activation+1.35e-02

Pos logit contributions
ble+0.066
bled+0.062
ster+0.057
 fin+0.056
balls+0.056
Neg logit contributions
ird-0.051
aren-0.047
 Karen-0.047
Alice-0.046
Ell-0.045
Token big
Token ID1263
Feature activation+2.68e-03

Pos logit contributions
ble+0.029
bled+0.027
balls+0.025
ster+0.024
 fin+0.024
Neg logit contributions
ird-0.024
 Karen-0.022
aren-0.022
rell-0.021
 Ellen-0.020
Token sea
Token ID5417
Feature activation+4.70e-02

Pos logit contributions
ble+0.006
bled+0.006
balls+0.005
ster+0.005
 fin+0.005
Neg logit contributions
ird-0.004
 Karen-0.004
rell-0.004
aren-0.004
 Ellen-0.004
Token.
Token ID13
Feature activation-1.47e-03

Pos logit contributions
bled+0.127
ble+0.127
balls+0.109
 fin+0.104
ster+0.103
Neg logit contributions
ird-0.051
hell-0.044
chair-0.044
aren-0.043
rell-0.042
Token In
Token ID554
Feature activation+4.30e-02

Pos logit contributions
ird+0.003
aren+0.002
rell+0.002
 Karen+0.002
old+0.002
Neg logit contributions
ble-0.003
bled-0.003
balls-0.003
 fin-0.003
ster-0.003
Token the
Token ID262
Feature activation+3.77e-02

Pos logit contributions
balls+0.118
bled+0.113
ble+0.111
ster+0.105
ider+0.101
Neg logit contributions
rell-0.037
 Karen-0.034
 Stella-0.030
 Alice-0.030
 course-0.030
Token sea
Token ID5417
Feature activation+5.24e-02

Pos logit contributions
bled+0.077
balls+0.077
ble+0.076
ster+0.066
bing+0.064
Neg logit contributions
rell-0.055
he-0.053
ird-0.050
hell-0.050
aren-0.047
Token,
Token ID11
Feature activation+3.41e-02

Pos logit contributions
bled+0.157
ble+0.154
balls+0.147
 fin+0.136
ster+0.135
Neg logit contributions
ird-0.037
aren-0.029
old-0.029
rell-0.028
hell-0.026
Token there
Token ID612
Feature activation+1.05e-02

Pos logit contributions
ble+0.084
balls+0.083
ster+0.080
bled+0.080
 fin+0.079
Neg logit contributions
 Karen-0.038
ird-0.037
 Stella-0.033
 Ellie-0.032
 Ellen-0.032
Token was
Token ID373
Feature activation+3.11e-02

Pos logit contributions
ble+0.024
bled+0.023
balls+0.022
ster+0.021
 fin+0.021
Neg logit contributions
ird-0.017
 Karen-0.015
rell-0.015
aren-0.015
hell-0.014
Token a
Token ID257
Feature activation+4.14e-02

Pos logit contributions
ble+0.085
bled+0.080
ster+0.079
balls+0.077
bing+0.074
Neg logit contributions
 Karen-0.030
rell-0.029
ird-0.028
 Stella-0.026
aren-0.026
Token small
Token ID1402
Feature activation+3.83e-02

Pos logit contributions
bled+0.108
ble+0.101
balls+0.096
bing+0.092
ster+0.085
Neg logit contributions
ird-0.043
ha-0.038
 Karen-0.038
hell-0.037
aren-0.036
Token and
Token ID290
Feature activation+6.04e-03

Pos logit contributions
ble+0.078
bled+0.075
balls+0.069
ster+0.066
 fin+0.065
Neg logit contributions
ird-0.038
 Karen-0.034
aren-0.033
rell-0.033
 Ellen-0.030
Token smiles
Token ID21845
Feature activation+1.29e-02

Pos logit contributions
ble+0.016
bled+0.015
balls+0.013
 fin+0.013
emo+0.013
Neg logit contributions
ird-0.007
rell-0.007
old-0.007
 Karen-0.006
 Alice-0.006
Token.
Token ID13
Feature activation+3.70e-02

Pos logit contributions
ble+0.035
bled+0.033
balls+0.031
 fin+0.030
ster+0.030
Neg logit contributions
ird-0.014
rell-0.013
 Karen-0.013
 Ellen-0.012
aren-0.012
Token<|endoftext|>
Token ID50256
Feature activation+3.49e-02

Pos logit contributions
ble+0.093
bled+0.087
balls+0.080
ster+0.079
 fin+0.076
Neg logit contributions
ird-0.053
 Karen-0.050
aren-0.048
rell-0.048
 Ellen-0.045
TokenS
Token ID50
Feature activation-2.59e-03

Pos logit contributions
ble+0.077
bled+0.071
balls+0.066
ster+0.065
 fin+0.064
Neg logit contributions
ird-0.061
rell-0.057
aren-0.056
 Karen-0.056
 Ellen-0.052
Tokenara
Token ID3301
Feature activation+3.26e-02

Pos logit contributions
ird+0.005
rell+0.004
 Karen+0.004
aren+0.004
 Ellen+0.004
Neg logit contributions
ble-0.005
bled-0.005
ster-0.005
balls-0.005
 fin-0.004
Token liked
Token ID8288
Feature activation+2.19e-02

Pos logit contributions
ble+0.084
bled+0.078
balls+0.074
 fin+0.070
ster+0.070
Neg logit contributions
ird-0.045
aren-0.041
rell-0.041
 Karen-0.040
old-0.039
Token to
Token ID284
Feature activation+1.90e-02

Pos logit contributions
ble+0.057
bled+0.054
ster+0.050
 fin+0.047
balls+0.046
Neg logit contributions
 Karen-0.029
rell-0.026
 Stella-0.026
 Mad-0.026
 Ellie-0.025
Token play
Token ID711
Feature activation+2.35e-02

Pos logit contributions
ble+0.044
bled+0.042
balls+0.038
ster+0.038
 fin+0.036
Neg logit contributions
ird-0.032
rell-0.029
 Karen-0.029
aren-0.027
 Ellen-0.027
Token with
Token ID351
Feature activation+1.32e-02

Pos logit contributions
ble+0.049
bled+0.046
ster+0.042
balls+0.041
 fin+0.041
Neg logit contributions
ird-0.040
rell-0.035
he-0.035
 Ellie-0.034
hell-0.034
Token her
Token ID607
Feature activation+2.54e-02

Pos logit contributions
ble+0.036
bled+0.033
ster+0.032
balls+0.031
 fin+0.030
Neg logit contributions
ird-0.016
 Karen-0.015
rell-0.015
 Ellen-0.014
aren-0.014
Token-
Token ID12
Feature activation-2.45e-02

Pos logit contributions
ble+0.001
bled+0.001
balls+0.001
ster+0.001
 fin+0.001
Neg logit contributions
ird-0.001
rell-0.001
aren-0.001
 Karen-0.001
 Ellen-0.001
Tokenray
Token ID2433
Feature activation+1.81e-02

Pos logit contributions
ird+0.047
rell+0.043
aren+0.043
 Karen+0.043
 Stella+0.040
Neg logit contributions
ble-0.050
bled-0.045
ster-0.043
balls-0.042
 fin-0.041
Token that
Token ID326
Feature activation+7.52e-03

Pos logit contributions
ble+0.041
bled+0.038
balls+0.035
ster+0.033
 fin+0.033
Neg logit contributions
ird-0.032
rell-0.028
 Karen-0.027
aren-0.027
old-0.026
Token he
Token ID339
Feature activation+1.69e-02

Pos logit contributions
ble+0.016
bled+0.015
balls+0.014
ster+0.014
 fin+0.013
Neg logit contributions
ird-0.013
rell-0.012
 Karen-0.012
aren-0.012
 Ellen-0.011
Token shone
Token ID44193
Feature activation+9.02e-04

Pos logit contributions
ble+0.036
bled+0.033
balls+0.031
ster+0.030
 fin+0.030
Neg logit contributions
ird-0.031
rell-0.029
 Karen-0.029
aren-0.029
 Ellen-0.027
Token on
Token ID319
Feature activation+9.26e-03

Pos logit contributions
ble+0.002
bled+0.002
balls+0.001
ster+0.001
 fin+0.001
Neg logit contributions
ird-0.002
 Karen-0.002
rell-0.002
aren-0.002
 Ellen-0.002
Token the
Token ID262
Feature activation-8.08e-03

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
ster+0.016
 fin+0.015
Neg logit contributions
ird-0.018
 Karen-0.017
rell-0.017
aren-0.016
 Ellen-0.016
Token tree
Token ID5509
Feature activation+1.10e-02

Pos logit contributions
ird+0.016
rell+0.015
 Karen+0.015
aren+0.015
 Ellen+0.014
Neg logit contributions
ble-0.017
bled-0.015
balls-0.014
ster-0.014
 fin-0.013
Token.
Token ID13
Feature activation-7.80e-03

Pos logit contributions
ble+0.026
bled+0.024
balls+0.023
ster+0.022
 fin+0.022
Neg logit contributions
ird-0.017
rell-0.016
 Karen-0.016
aren-0.016
 Ellen-0.014
Token With
Token ID2080
Feature activation-4.12e-03

Pos logit contributions
ird+0.016
aren+0.015
rell+0.015
 Karen+0.014
hell+0.014
Neg logit contributions
ble-0.014
bled-0.013
balls-0.013
 fin-0.012
ster-0.012
Token the
Token ID262
Feature activation-1.67e-02

Pos logit contributions
ird+0.008
aren+0.008
rell+0.008
 Karen+0.008
 Ellen+0.007
Neg logit contributions
ble-0.008
bled-0.007
balls-0.007
ster-0.006
 fin-0.006
TokenSam
Token ID16305
Feature activation-5.87e-03

Pos logit contributions
ble+0.063
bled+0.056
balls+0.053
ster+0.052
 fin+0.051
Neg logit contributions
ird-0.056
 Karen-0.052
rell-0.052
aren-0.052
 Ellen-0.049
Token and
Token ID290
Feature activation+6.44e-03

Pos logit contributions
ird+0.011
 Karen+0.010
rell+0.010
aren+0.010
 Ellen+0.010
Neg logit contributions
ble-0.012
bled-0.011
ster-0.010
 fin-0.010
balls-0.010
Token Mia
Token ID32189
Feature activation-2.35e-02

Pos logit contributions
ble+0.014
bled+0.013
balls+0.012
ster+0.012
 fin+0.011
Neg logit contributions
ird-0.012
 Karen-0.011
aren-0.011
rell-0.011
 Ellen-0.010
Token are
Token ID389
Feature activation-8.02e-03

Pos logit contributions
ird+0.043
 Karen+0.041
rell+0.041
 Ellen+0.040
aren+0.040
Neg logit contributions
ble-0.046
bled-0.040
ster-0.039
 fin-0.037
balls-0.036
Token friends
Token ID2460
Feature activation-3.24e-02

Pos logit contributions
ird+0.016
 Karen+0.015
aren+0.015
rell+0.015
 Ellen+0.014
Neg logit contributions
ble-0.016
bled-0.014
balls-0.013
ster-0.013
 fin-0.013
Token.
Token ID13
Feature activation+4.39e-03

Pos logit contributions
ird+0.060
rell+0.058
aren+0.055
 Karen+0.055
 Ellen+0.052
Neg logit contributions
ble-0.069
bled-0.060
ster-0.057
balls-0.057
 fin-0.057
Token They
Token ID1119
Feature activation+1.77e-03

Pos logit contributions
ble+0.007
balls+0.006
bled+0.006
 fin+0.006
ster+0.006
Neg logit contributions
ird-0.010
rell-0.009
aren-0.008
hell-0.008
 Karen-0.008
Token like
Token ID588
Feature activation+8.32e-03

Pos logit contributions
ble+0.004
bled+0.004
balls+0.004
ster+0.003
 fin+0.003
Neg logit contributions
ird-0.003
aren-0.003
 Karen-0.003
rell-0.003
chairs-0.002
Token to
Token ID284
Feature activation+8.65e-03

Pos logit contributions
ble+0.020
bled+0.018
ster+0.016
balls+0.016
 fin+0.016
Neg logit contributions
ird-0.013
 Karen-0.013
rell-0.012
aren-0.012
 Stella-0.012
Token play
Token ID711
Feature activation+1.50e-02

Pos logit contributions
ble+0.018
bled+0.017
balls+0.016
ster+0.015
 fin+0.015
Neg logit contributions
ird-0.016
rell-0.015
aren-0.015
 Karen-0.015
 Ellen-0.014
Token outside
Token ID2354
Feature activation+1.20e-02

Pos logit contributions
ble+0.031
bled+0.028
ster+0.026
 fin+0.026
balls+0.025
Neg logit contributions
ird-0.027
 Karen-0.023
rell-0.023
 Ellen-0.023
 Ellie-0.022
Token oven
Token ID14361
Feature activation+6.29e-03

Pos logit contributions
ble+0.016
bled+0.015
balls+0.014
ster+0.014
 fin+0.013
Neg logit contributions
ird-0.014
 Karen-0.013
rell-0.013
aren-0.013
 Ellen-0.012
Token and
Token ID290
Feature activation-1.13e-02

Pos logit contributions
ble+0.014
bled+0.013
balls+0.012
ster+0.012
 fin+0.012
Neg logit contributions
ird-0.011
aren-0.010
rell-0.010
 Karen-0.010
 Ellen-0.009
Token took
Token ID1718
Feature activation-2.33e-02

Pos logit contributions
ird+0.019
rell+0.018
aren+0.018
 Karen+0.018
hell+0.017
Neg logit contributions
ble-0.025
bled-0.023
balls-0.022
ster-0.022
 fin-0.021
Token out
Token ID503
Feature activation+3.93e-03

Pos logit contributions
ird+0.048
aren+0.048
rell+0.047
hell+0.044
 Karen+0.044
Neg logit contributions
ble-0.040
bled-0.038
balls-0.036
ster-0.031
 fin-0.031
Token the
Token ID262
Feature activation-1.24e-02

Pos logit contributions
ble+0.006
bled+0.005
balls+0.005
 fin+0.005
 diving+0.004
Neg logit contributions
ird-0.009
rell-0.009
aren-0.009
hell-0.008
 Karen-0.008
Token cake
Token ID12187
Feature activation+1.90e-02

Pos logit contributions
ird+0.024
rell+0.023
aren+0.023
 Karen+0.023
 Ellen+0.021
Neg logit contributions
ble-0.025
bled-0.022
balls-0.022
ster-0.021
 fin-0.020
Token.
Token ID13
Feature activation-2.98e-03

Pos logit contributions
ble+0.044
bled+0.040
balls+0.039
ster+0.038
 fin+0.037
Neg logit contributions
ird-0.031
aren-0.029
 Karen-0.029
rell-0.029
 Ellen-0.027
Token She
Token ID1375
Feature activation-2.43e-03

Pos logit contributions
ird+0.005
rell+0.004
aren+0.004
 Karen+0.004
hell+0.004
Neg logit contributions
ble-0.007
bled-0.006
balls-0.006
 fin-0.006
ster-0.006
Token checked
Token ID10667
Feature activation-1.12e-02

Pos logit contributions
ird+0.004
rell+0.004
aren+0.004
 Karen+0.004
hell+0.003
Neg logit contributions
ble-0.005
bled-0.005
balls-0.005
ster-0.005
 fin-0.005
Token if
Token ID611
Feature activation+1.34e-02

Pos logit contributions
ird+0.023
rell+0.022
aren+0.021
 Karen+0.020
Ell+0.020
Neg logit contributions
ble-0.022
bled-0.020
balls-0.018
ster-0.018
 fin-0.017
Token Sara
Token ID24799
Feature activation+3.81e-03

Pos logit contributions
ble+0.031
bled+0.028
balls+0.026
ster+0.026
 fin+0.026
Neg logit contributions
ird-0.022
 Karen-0.021
aren-0.021
rell-0.020
 Ellen-0.019
Token told
Token ID1297
Feature activation+2.57e-02

Pos logit contributions
ble+0.069
bled+0.063
balls+0.060
ster+0.057
 fin+0.056
Neg logit contributions
ird-0.041
 Karen-0.037
aren-0.037
rell-0.036
 Ellen-0.033
Token his
Token ID465
Feature activation+3.73e-02

Pos logit contributions
ble+0.036
bled+0.030
balls+0.027
ster+0.026
 fin+0.025
Neg logit contributions
ird-0.067
rell-0.064
 Karen-0.064
aren-0.064
 Ellen-0.061
Token mom
Token ID1995
Feature activation-1.92e-02

Pos logit contributions
ble+0.072
bled+0.068
balls+0.059
ster+0.056
bing+0.055
Neg logit contributions
ird-0.071
 Karen-0.070
aren-0.067
old-0.066
rell-0.065
Token about
Token ID546
Feature activation+1.99e-02

Pos logit contributions
ird+0.030
rell+0.027
aren+0.027
 Karen+0.026
 Ellen+0.024
Neg logit contributions
ble-0.047
bled-0.043
balls-0.041
ster-0.040
 fin-0.039
Token the
Token ID262
Feature activation+1.65e-02

Pos logit contributions
ble+0.041
bled+0.038
balls+0.035
ster+0.035
 fin+0.033
Neg logit contributions
 Karen-0.037
ird-0.035
aren-0.034
rell-0.034
 Ellen-0.033
Token monster
Token ID9234
Feature activation+3.38e-03

Pos logit contributions
ble+0.035
bled+0.033
balls+0.030
ster+0.029
 fin+0.028
Neg logit contributions
ird-0.029
rell-0.028
 Karen-0.027
aren-0.027
hell-0.026
Token he
Token ID339
Feature activation+3.02e-03

Pos logit contributions
ble+0.007
bled+0.007
balls+0.006
ster+0.006
 fin+0.006
Neg logit contributions
ird-0.006
rell-0.006
aren-0.005
 Karen-0.005
 Ellen-0.005
Token found
Token ID1043
Feature activation+1.23e-03

Pos logit contributions
ble+0.007
bled+0.006
ster+0.006
balls+0.006
 fin+0.006
Neg logit contributions
ird-0.005
rell-0.005
 Karen-0.005
aren-0.005
 Ellen-0.005
Token.
Token ID13
Feature activation+6.25e-03

Pos logit contributions
ble+0.003
bled+0.002
balls+0.002
ster+0.002
 fin+0.002
Neg logit contributions
ird-0.002
rell-0.002
aren-0.002
 Karen-0.002
 Ellen-0.002
Token His
Token ID2399
Feature activation+2.25e-02

Pos logit contributions
ble+0.015
bled+0.013
ster+0.012
balls+0.012
 fin+0.012
Neg logit contributions
ird-0.011
 Karen-0.010
rell-0.010
aren-0.010
 Stella-0.009
Token mom
Token ID1995
Feature activation-2.85e-02

Pos logit contributions
ble+0.042
bled+0.038
balls+0.034
ster+0.033
 fin+0.032
Neg logit contributions
ird-0.047
 Karen-0.045
aren-0.044
rell-0.043
 Ellen-0.041
Token.
Token ID13
Feature activation+1.89e-02

Pos logit contributions
ble+0.071
bled+0.068
balls+0.063
ster+0.063
 fin+0.061
Neg logit contributions
ird-0.029
rell-0.025
 Karen-0.025
hell-0.023
 Ellen-0.022
Token
Token ID198
Feature activation+3.03e-02

Pos logit contributions
ble+0.037
bled+0.034
balls+0.031
ster+0.031
 fin+0.029
Neg logit contributions
ird-0.039
 Karen-0.036
rell-0.036
aren-0.036
 Ellen-0.034
Token
Token ID198
Feature activation+3.75e-02

Pos logit contributions
ble+0.079
bled+0.073
balls+0.069
ster+0.069
 fin+0.067
Neg logit contributions
ird-0.042
 Karen-0.037
aren-0.037
rell-0.036
old-0.033
TokenAnna
Token ID31160
Feature activation-1.76e-03

Pos logit contributions
ble+0.087
bled+0.082
balls+0.076
ster+0.074
 fin+0.072
Neg logit contributions
ird-0.063
aren-0.057
 Karen-0.056
old-0.056
Alice-0.055
Token heard
Token ID2982
Feature activation+1.00e-02

Pos logit contributions
ird+0.003
rell+0.003
 Karen+0.003
aren+0.003
 Ellen+0.003
Neg logit contributions
ble-0.004
bled-0.003
ster-0.003
balls-0.003
 fin-0.003
Token him
Token ID683
Feature activation-7.43e-03

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
 fin+0.015
ster+0.015
Neg logit contributions
ird-0.021
aren-0.020
rell-0.020
 Karen-0.019
Ell-0.018
Token cry
Token ID3960
Feature activation+1.17e-02

Pos logit contributions
ird+0.011
rell+0.010
 Karen+0.010
aren+0.009
 Ellen+0.009
Neg logit contributions
ble-0.019
bled-0.017
balls-0.017
ster-0.016
 fin-0.016
Token and
Token ID290
Feature activation+5.74e-03

Pos logit contributions
ble+0.031
bled+0.029
balls+0.027
ster+0.027
 fin+0.026
Neg logit contributions
ird-0.015
rell-0.014
 Karen-0.013
aren-0.012
hell-0.012
Token came
Token ID1625
Feature activation+8.33e-03

Pos logit contributions
ble+0.014
bled+0.012
balls+0.012
ster+0.011
 fin+0.011
Neg logit contributions
ird-0.009
 Karen-0.009
rell-0.008
aren-0.008
 Ellen-0.008
Token to
Token ID284
Feature activation-3.80e-03

Pos logit contributions
ble+0.018
bled+0.016
ster+0.015
balls+0.015
 fin+0.015
Neg logit contributions
ird-0.014
 Karen-0.014
rell-0.014
aren-0.013
 Ellen-0.013
Token see
Token ID766
Feature activation+7.70e-04

Pos logit contributions
ird+0.007
rell+0.007
aren+0.006
 Karen+0.006
hell+0.006
Neg logit contributions
ble-0.008
bled-0.007
balls-0.007
ster-0.007
 fin-0.007
Token green
Token ID4077
Feature activation-1.82e-03

Pos logit contributions
ird+0.006
rell+0.006
 Karen+0.006
aren+0.006
 Stella+0.005
Neg logit contributions
ble-0.008
bled-0.007
balls-0.007
ster-0.007
 fin-0.007
Token and
Token ID290
Feature activation+4.70e-03

Pos logit contributions
ird+0.003
rell+0.002
 Karen+0.002
aren+0.002
 Ellen+0.002
Neg logit contributions
ble-0.005
bled-0.004
balls-0.004
ster-0.004
 fin-0.004
Token brown
Token ID7586
Feature activation-1.32e-02

Pos logit contributions
ble+0.011
bled+0.010
ster+0.009
balls+0.009
 fin+0.009
Neg logit contributions
ird-0.007
rell-0.007
 Karen-0.007
aren-0.007
 Stella-0.006
Token,
Token ID11
Feature activation-6.60e-03

Pos logit contributions
ird+0.018
aren+0.016
 Karen+0.016
rell+0.016
 Ellen+0.015
Neg logit contributions
ble-0.035
bled-0.032
balls-0.030
ster-0.030
 fin-0.029
Token like
Token ID588
Feature activation-8.97e-03

Pos logit contributions
ird+0.010
 Karen+0.010
rell+0.010
aren+0.009
 Stella+0.009
Neg logit contributions
ble-0.015
bled-0.014
balls-0.014
ster-0.013
 fin-0.013
Token a
Token ID257
Feature activation-1.20e-02

Pos logit contributions
ird+0.018
aren+0.017
rell+0.016
 Karen+0.016
old+0.015
Neg logit contributions
ble-0.018
bled-0.016
balls-0.015
 fin-0.015
ster-0.015
Token leaf
Token ID12835
Feature activation-1.34e-02

Pos logit contributions
ird+0.022
rell+0.020
 Karen+0.020
aren+0.020
 Ellen+0.019
Neg logit contributions
ble-0.025
bled-0.023
balls-0.022
ster-0.021
 fin-0.021
Token,"
Token ID553
Feature activation+1.61e-02

Pos logit contributions
ird+0.016
aren+0.014
 Karen+0.014
rell+0.014
 Ellen+0.013
Neg logit contributions
ble-0.038
bled-0.034
balls-0.034
ster-0.033
 fin-0.033
Token Lily
Token ID20037
Feature activation-2.29e-02

Pos logit contributions
ble+0.032
balls+0.029
bled+0.028
 fin+0.027
ster+0.027
Neg logit contributions
ird-0.031
rell-0.029
aren-0.029
 Karen-0.027
hell-0.026
Token said
Token ID531
Feature activation-1.80e-02

Pos logit contributions
aren+0.041
rell+0.040
ird+0.040
 Karen+0.040
irst+0.038
Neg logit contributions
ble-0.046
ster-0.040
 fin-0.040
balls-0.039
bled-0.038
Token,
Token ID11
Feature activation+1.01e-02

Pos logit contributions
ird+0.028
aren+0.026
rell+0.026
 Karen+0.026
 Ellen+0.024
Neg logit contributions
ble-0.043
bled-0.040
balls-0.038
ster-0.037
 fin-0.036
Token cat
Token ID3797
Feature activation-4.28e-03

Pos logit contributions
rell+0.013
ird+0.013
aren+0.012
hell+0.012
 Karen+0.012
Neg logit contributions
ble-0.007
ster-0.007
balls-0.006
 next-0.006
 rod-0.006
Token a
Token ID257
Feature activation-1.92e-02

Pos logit contributions
ird+0.009
aren+0.009
rell+0.008
 Karen+0.008
 Ellen+0.008
Neg logit contributions
ble-0.008
bled-0.007
balls-0.007
ster-0.006
 fin-0.006
Token toy
Token ID13373
Feature activation+4.62e-03

Pos logit contributions
ird+0.034
rell+0.032
aren+0.032
 Karen+0.032
 Ellen+0.029
Neg logit contributions
ble-0.042
bled-0.036
balls-0.036
 fin-0.036
ster-0.036
Token mouse
Token ID10211
Feature activation-8.98e-03

Pos logit contributions
ble+0.011
bled+0.010
balls+0.010
ster+0.009
 fin+0.009
Neg logit contributions
ird-0.007
rell-0.007
 Karen-0.007
aren-0.007
 Ellen-0.006
Token.
Token ID13
Feature activation-2.27e-02

Pos logit contributions
ird+0.016
aren+0.015
rell+0.015
 Karen+0.015
 Ellen+0.014
Neg logit contributions
ble-0.019
bled-0.017
balls-0.017
ster-0.017
 fin-0.016
Token The
Token ID383
Feature activation-2.68e-03

Pos logit contributions
ird+0.042
aren+0.039
rell+0.039
 Karen+0.038
hell+0.037
Neg logit contributions
ble-0.046
bled-0.041
balls-0.040
ster-0.038
 fin-0.038
Token cat
Token ID3797
Feature activation-9.60e-03

Pos logit contributions
ird+0.005
rell+0.005
aren+0.005
 Karen+0.005
hell+0.004
Neg logit contributions
ble-0.006
ster-0.005
bled-0.005
balls-0.005
 fin-0.005
Token seemed
Token ID3947
Feature activation-1.69e-02

Pos logit contributions
ird+0.019
rell+0.018
aren+0.018
 Karen+0.018
 Ellen+0.016
Neg logit contributions
ble-0.019
bled-0.017
balls-0.016
ster-0.016
 fin-0.016
Token to
Token ID284
Feature activation-8.36e-03

Pos logit contributions
ird+0.032
rell+0.030
aren+0.030
 Karen+0.030
 Ellen+0.028
Neg logit contributions
ble-0.035
bled-0.031
balls-0.029
ster-0.029
 fin-0.028
Token like
Token ID588
Feature activation-3.22e-03

Pos logit contributions
ird+0.016
aren+0.015
chairs+0.015
chair+0.014
rell+0.014
Neg logit contributions
ble-0.018
 fin-0.015
bled-0.014
ster-0.013
 diving-0.013
Token it
Token ID340
Feature activation-1.40e-02

Pos logit contributions
ird+0.006
rell+0.006
aren+0.006
 Karen+0.006
 Ellen+0.005
Neg logit contributions
ble-0.007
bled-0.006
balls-0.006
ster-0.005
 fin-0.005
Token again
Token ID757
Feature activation+1.55e-02

Pos logit contributions
ble+0.049
bled+0.045
balls+0.043
ster+0.042
 fin+0.041
Neg logit contributions
ird-0.030
rell-0.027
 Karen-0.027
aren-0.027
 Ellen-0.025
Token tomorrow
Token ID9439
Feature activation+1.38e-02

Pos logit contributions
ble+0.038
bled+0.035
balls+0.034
ster+0.033
 fin+0.032
Neg logit contributions
ird-0.023
 Karen-0.021
rell-0.021
aren-0.021
 Ellen-0.019
Token.
Token ID13
Feature activation-5.48e-04

Pos logit contributions
ble+0.034
bled+0.031
balls+0.029
ster+0.029
 fin+0.028
Neg logit contributions
ird-0.021
rell-0.020
aren-0.019
 Karen-0.019
 Ellen-0.018
Token<|endoftext|>
Token ID50256
Feature activation+6.71e-03

Pos logit contributions
ird+0.001
rell+0.001
aren+0.001
 Karen+0.001
hell+0.001
Neg logit contributions
ble-0.001
bled-0.001
balls-0.001
ster-0.001
 fin-0.001
TokenOnce
Token ID7454
Feature activation+3.35e-03

Pos logit contributions
ble+0.013
bled+0.011
balls+0.011
ster+0.011
 fin+0.010
Neg logit contributions
ird-0.013
rell-0.012
 Karen-0.012
aren-0.012
 Ellen-0.011
Token upon
Token ID2402
Feature activation-8.63e-03

Pos logit contributions
ble+0.006
bled+0.005
balls+0.004
ster+0.004
 fin+0.004
Neg logit contributions
ird-0.008
rell-0.008
 Karen-0.007
aren-0.007
 Ellen-0.007
Token a
Token ID257
Feature activation-1.95e-02

Pos logit contributions
ird+0.018
aren+0.018
rell+0.017
 Karen+0.017
 Ellen+0.016
Neg logit contributions
ble-0.016
bled-0.014
balls-0.013
 fin-0.013
ster-0.013
Token time
Token ID640
Feature activation-1.41e-02

Pos logit contributions
ird+0.030
 Karen+0.029
rell+0.029
aren+0.028
 Ellen+0.027
Neg logit contributions
ble-0.044
balls-0.039
bled-0.039
 fin-0.039
ster-0.038
Token,
Token ID11
Feature activation+4.99e-03

Pos logit contributions
ird+0.026
rell+0.024
aren+0.023
 Karen+0.023
chairs+0.022
Neg logit contributions
ble-0.031
bled-0.027
balls-0.026
ster-0.025
 fin-0.025
Token there
Token ID612
Feature activation+6.30e-04

Pos logit contributions
ble+0.010
bled+0.009
balls+0.009
ster+0.009
 fin+0.009
Neg logit contributions
ird-0.009
 Karen-0.008
aren-0.008
rell-0.008
 Ellen-0.008
Token was
Token ID373
Feature activation+5.99e-03

Pos logit contributions
ble+0.001
bled+0.001
balls+0.001
ster+0.001
 fin+0.001
Neg logit contributions
ird-0.001
rell-0.001
aren-0.001
 Karen-0.001
 Ellen-0.001
Token very
Token ID845
Feature activation+2.95e-02

Pos logit contributions
ble+0.013
bled+0.012
balls+0.011
ster+0.011
 fin+0.011
Neg logit contributions
ird-0.010
 Karen-0.009
aren-0.009
rell-0.009
 Ellen-0.008
Token bright
Token ID6016
Feature activation-9.84e-04

Pos logit contributions
ble+0.070
bled+0.066
balls+0.061
ster+0.060
 fin+0.058
Neg logit contributions
ird-0.044
 Karen-0.042
aren-0.042
rell-0.040
hell-0.038
Token star
Token ID3491
Feature activation+2.58e-02

Pos logit contributions
ird+0.002
rell+0.002
aren+0.002
 Karen+0.002
 Ellen+0.002
Neg logit contributions
ble-0.002
bled-0.002
balls-0.002
ster-0.002
 fin-0.002
Token that
Token ID326
Feature activation+1.61e-02

Pos logit contributions
bled+0.052
ble+0.052
balls+0.048
emo+0.047
 fin+0.044
Neg logit contributions
ird-0.044
chair-0.039
rell-0.038
hell-0.037
 Karen-0.037
Token was
Token ID373
Feature activation+3.69e-03

Pos logit contributions
ble+0.036
bled+0.032
balls+0.031
ster+0.030
 fin+0.029
Neg logit contributions
ird-0.028
 Karen-0.026
aren-0.026
rell-0.026
 Ellen-0.024
Token cheerful
Token ID37999
Feature activation+2.16e-03

Pos logit contributions
ble+0.007
bled+0.007
balls+0.007
ster+0.007
 fin+0.006
Neg logit contributions
ird-0.006
 Karen-0.006
rell-0.006
aren-0.006
 Ellen-0.006
Token and
Token ID290
Feature activation+4.27e-03

Pos logit contributions
ble+0.005
bled+0.005
balls+0.005
ster+0.005
 fin+0.005
Neg logit contributions
ird-0.003
rell-0.003
 Karen-0.003
aren-0.003
 Ellen-0.003
Token shiny
Token ID22441
Feature activation-3.70e-03

Pos logit contributions
ble+0.010
bled+0.010
ster+0.009
balls+0.009
 fin+0.009
Neg logit contributions
ird-0.006
 Karen-0.006
 Ellen-0.005
rell-0.005
 Stella-0.005
Token.
Token ID13
Feature activation+2.19e-03

Pos logit contributions
ird+0.005
rell+0.005
 Karen+0.005
aren+0.005
 Ellen+0.004
Neg logit contributions
ble-0.009
bled-0.009
balls-0.008
ster-0.008
 fin-0.008
Token
Token ID198
Feature activation+6.92e-03

Pos logit contributions
ble+0.004
bled+0.004
balls+0.004
 fin+0.004
ster+0.004
Neg logit contributions
ird-0.004
rell-0.004
aren-0.004
 Karen-0.004
hell-0.004
Token
Token ID198
Feature activation+1.99e-02

Pos logit contributions
ble+0.013
bled+0.011
balls+0.011
emo+0.011
 fin+0.011
Neg logit contributions
rell-0.014
ird-0.014
aren-0.012
 Karen-0.012
chairs-0.012
Token."
Token ID526
Feature activation+3.77e-02

Pos logit contributions
ird+0.005
aren+0.005
 Karen+0.005
rell+0.005
 Ellen+0.005
Neg logit contributions
ble-0.007
bled-0.007
balls-0.006
ster-0.006
 fin-0.006
Token
Token ID198
Feature activation+3.40e-02

Pos logit contributions
ble+0.101
bled+0.094
balls+0.088
ster+0.087
 fin+0.085
Neg logit contributions
ird-0.050
 Karen-0.045
aren-0.045
rell-0.044
 Ellen-0.040
Token
Token ID198
Feature activation+4.87e-02

Pos logit contributions
ble+0.086
bled+0.079
balls+0.076
ster+0.074
 fin+0.073
Neg logit contributions
ird-0.049
rell-0.045
 Karen-0.044
aren-0.044
 Ellen-0.040
TokenBen
Token ID11696
Feature activation-6.88e-03

Pos logit contributions
ble+0.106
bled+0.098
balls+0.091
ster+0.086
 fin+0.085
Neg logit contributions
ird-0.086
aren-0.081
old-0.077
 Karen-0.077
Aud-0.075
Token and
Token ID290
Feature activation+3.44e-02

Pos logit contributions
ird+0.011
rell+0.010
 Karen+0.010
aren+0.010
 Ellen+0.009
Neg logit contributions
ble-0.016
bled-0.015
balls-0.014
ster-0.014
 fin-0.014
Token Sam
Token ID3409
Feature activation-4.79e-02

Pos logit contributions
ble+0.066
bled+0.059
balls+0.054
ster+0.054
 fin+0.052
Neg logit contributions
ird-0.071
 Karen-0.067
rell-0.067
aren-0.067
 Ellen-0.063
Token looked
Token ID3114
Feature activation-1.72e-02

Pos logit contributions
rell+0.088
 Karen+0.087
ird+0.087
aren+0.086
chairs+0.085
Neg logit contributions
ble-0.095
bled-0.088
ster-0.082
 fin-0.079
balls-0.078
Token at
Token ID379
Feature activation+8.01e-03

Pos logit contributions
ird+0.042
 Karen+0.040
aren+0.040
rell+0.039
Ell+0.037
Neg logit contributions
ble-0.026
bled-0.024
balls-0.022
ster-0.022
 fin-0.021
Token Bob
Token ID5811
Feature activation+1.00e-02

Pos logit contributions
ble+0.015
bled+0.014
balls+0.013
 fin+0.012
ster+0.012
Neg logit contributions
ird-0.017
rell-0.016
aren-0.016
 Karen-0.015
 Ellen-0.014
Tokeno
Token ID78
Feature activation+8.19e-03

Pos logit contributions
ble+0.027
bled+0.025
balls+0.023
ster+0.023
 fin+0.023
Neg logit contributions
ird-0.013
rell-0.012
 Karen-0.011
aren-0.011
hell-0.011
Token.
Token ID13
Feature activation+1.28e-02

Pos logit contributions
ble+0.021
bled+0.020
balls+0.019
ster+0.018
 fin+0.018
Neg logit contributions
ird-0.011
rell-0.010
 Karen-0.009
aren-0.009
hell-0.009
Token and
Token ID290
Feature activation-8.44e-03

Pos logit contributions
ird+0.012
 Karen+0.011
rell+0.011
aren+0.011
 Ellen+0.010
Neg logit contributions
ble-0.016
bled-0.015
balls-0.014
ster-0.013
 fin-0.013
Token picked
Token ID6497
Feature activation-1.48e-02

Pos logit contributions
ird+0.014
 Karen+0.013
rell+0.013
aren+0.013
 Ellen+0.012
Neg logit contributions
ble-0.019
bled-0.018
balls-0.017
ster-0.016
 fin-0.016
Token up
Token ID510
Feature activation-8.55e-03

Pos logit contributions
ird+0.030
rell+0.027
aren+0.027
 Karen+0.027
 Ellen+0.025
Neg logit contributions
ble-0.030
bled-0.027
balls-0.025
ster-0.024
 fin-0.024
Token her
Token ID607
Feature activation+3.46e-02

Pos logit contributions
ird+0.018
rell+0.016
aren+0.016
Ell+0.015
 Karen+0.015
Neg logit contributions
ble-0.016
bled-0.014
balls-0.014
ster-0.013
 fin-0.013
Token steak
Token ID26320
Feature activation+3.04e-02

Pos logit contributions
ble+0.070
bled+0.065
balls+0.056
 fin+0.055
ster+0.055
Neg logit contributions
ird-0.061
rell-0.055
hell-0.055
 Alice-0.054
 Stella-0.054
Token.
Token ID13
Feature activation-4.49e-02

Pos logit contributions
ble+0.077
bled+0.072
 fin+0.068
balls+0.067
ster+0.067
Neg logit contributions
ird-0.039
 Karen-0.035
rell-0.034
aren-0.034
hell-0.033
Token She
Token ID1375
Feature activation+1.69e-02

Pos logit contributions
ird+0.091
hell+0.086
aren+0.080
rell+0.079
irst+0.079
Neg logit contributions
ble-0.074
bled-0.066
balls-0.066
 next-0.055
 diving-0.052
Token took
Token ID1718
Feature activation-4.04e-02

Pos logit contributions
ble+0.039
bled+0.035
balls+0.033
ster+0.033
 fin+0.032
Neg logit contributions
ird-0.029
 Karen-0.026
aren-0.026
rell-0.025
 Stella-0.024
Token it
Token ID340
Feature activation-2.17e-03

Pos logit contributions
ird+0.098
aren+0.096
rell+0.093
 Karen+0.090
Ell+0.087
Neg logit contributions
ble-0.062
bled-0.057
balls-0.050
ster-0.045
 fin-0.045
Token to
Token ID284
Feature activation-1.64e-02

Pos logit contributions
ird+0.004
aren+0.004
rell+0.004
 Karen+0.004
 Ellen+0.004
Neg logit contributions
ble-0.004
bled-0.004
balls-0.004
ster-0.004
 fin-0.004
Token the
Token ID262
Feature activation+3.32e-03

Pos logit contributions
ird+0.029
rell+0.028
aren+0.027
 Karen+0.026
hell+0.025
Neg logit contributions
ble-0.036
bled-0.032
balls-0.030
 fin-0.030
ster-0.030
Token the
Token ID262
Feature activation-3.11e-02

Pos logit contributions
ird+0.050
rell+0.048
aren+0.045
hell+0.045
chairs+0.045
Neg logit contributions
ble-0.035
bled-0.030
balls-0.028
 fin-0.027
 diving-0.026
Token vendor
Token ID18371
Feature activation-1.60e-02

Pos logit contributions
ird+0.053
rell+0.049
 Karen+0.048
aren+0.048
 Ellen+0.044
Neg logit contributions
ble-0.071
bled-0.065
balls-0.061
ster-0.060
 fin-0.059
Token and
Token ID290
Feature activation-1.79e-02

Pos logit contributions
ird+0.027
rell+0.025
 Karen+0.025
aren+0.025
 Ellen+0.023
Neg logit contributions
ble-0.036
bled-0.032
balls-0.031
ster-0.031
 fin-0.030
Token asked
Token ID1965
Feature activation-2.90e-03

Pos logit contributions
ird+0.031
rell+0.029
aren+0.028
 Karen+0.028
hell+0.027
Neg logit contributions
ble-0.039
bled-0.036
balls-0.034
ster-0.033
 fin-0.033
Token for
Token ID329
Feature activation-1.05e-02

Pos logit contributions
ird+0.006
rell+0.005
aren+0.005
 Karen+0.005
 Ellen+0.005
Neg logit contributions
ble-0.006
bled-0.005
balls-0.005
ster-0.005
 fin-0.005
Token a
Token ID257
Feature activation-4.37e-02

Pos logit contributions
ird+0.021
rell+0.020
aren+0.020
 Karen+0.019
 Ellen+0.018
Neg logit contributions
ble-0.020
bled-0.019
balls-0.018
 fin-0.017
ster-0.016
Token red
Token ID2266
Feature activation-2.79e-02

Pos logit contributions
ird+0.081
rell+0.080
 Karen+0.075
aren+0.075
chairs+0.074
Neg logit contributions
ble-0.087
balls-0.077
bled-0.075
ster-0.075
 fin-0.074
Token balloon
Token ID21190
Feature activation+3.83e-03

Pos logit contributions
ird+0.053
rell+0.049
 Karen+0.048
aren+0.048
 Ellen+0.046
Neg logit contributions
ble-0.058
bled-0.051
balls-0.049
ster-0.048
 fin-0.048
Token.
Token ID13
Feature activation+1.17e-02

Pos logit contributions
ble+0.009
bled+0.008
balls+0.008
ster+0.008
 fin+0.008
Neg logit contributions
ird-0.006
 Karen-0.005
rell-0.005
aren-0.005
 Ellen-0.005
Token The
Token ID383
Feature activation-2.76e-02

Pos logit contributions
ble+0.031
bled+0.028
ster+0.026
balls+0.026
 fin+0.025
Neg logit contributions
ird-0.016
 Karen-0.015
aren-0.015
rell-0.015
 Stella-0.014
Token vendor
Token ID18371
Feature activation-1.45e-02

Pos logit contributions
ird+0.054
rell+0.050
 Karen+0.046
aren+0.046
Ell+0.045
Neg logit contributions
ble-0.057
bled-0.051
balls-0.049
ster-0.048
 shark-0.046
Token that
Token ID326
Feature activation+2.46e-03

Pos logit contributions
ird+0.003
rell+0.003
aren+0.003
 Karen+0.003
 Ellen+0.003
Neg logit contributions
ble-0.003
bled-0.003
balls-0.003
ster-0.003
 fin-0.003
Token was
Token ID373
Feature activation-6.54e-03

Pos logit contributions
ble+0.006
bled+0.005
balls+0.005
ster+0.005
 fin+0.005
Neg logit contributions
ird-0.004
rell-0.004
aren-0.004
 Karen-0.004
 Ellen-0.003
Token filled
Token ID5901
Feature activation+9.34e-03

Pos logit contributions
ird+0.011
 Karen+0.010
rell+0.010
aren+0.010
 Ellen+0.010
Neg logit contributions
ble-0.015
bled-0.013
balls-0.013
ster-0.012
 fin-0.012
Token with
Token ID351
Feature activation-2.64e-02

Pos logit contributions
ble+0.018
bled+0.017
balls+0.016
ster+0.016
 fin+0.016
Neg logit contributions
ird-0.017
 Karen-0.016
 Stella-0.016
aren-0.016
hell-0.015
Token toys
Token ID14958
Feature activation-1.02e-02

Pos logit contributions
ird+0.055
aren+0.053
rell+0.050
 Karen+0.049
Alice+0.047
Neg logit contributions
ble-0.051
bled-0.044
 fin-0.043
balls-0.042
ster-0.041
Token.
Token ID13
Feature activation-6.74e-02

Pos logit contributions
ird+0.018
aren+0.017
 Karen+0.017
rell+0.016
 Ellen+0.015
Neg logit contributions
ble-0.023
bled-0.020
balls-0.020
ster-0.019
 fin-0.018
Token As
Token ID1081
Feature activation-1.51e-02

Pos logit contributions
ird+0.141
ives+0.130
hell+0.130
ited+0.126
ha+0.124
Neg logit contributions
ble-0.084
balls-0.077
 next-0.073
 diving-0.067
 Finn-0.067
Token Lucy
Token ID22162
Feature activation+1.18e-02

Pos logit contributions
ird+0.030
rell+0.029
ives+0.028
hell+0.027
aren+0.027
Neg logit contributions
ble-0.029
bled-0.026
balls-0.025
 diving-0.024
 shark-0.023
Token walked
Token ID6807
Feature activation-1.31e-02

Pos logit contributions
ble+0.025
bled+0.022
balls+0.021
ster+0.020
 fin+0.020
Neg logit contributions
ird-0.022
rell-0.021
 Karen-0.021
aren-0.021
 Ellen-0.019
Token through
Token ID832
Feature activation-1.42e-02

Pos logit contributions
ird+0.023
rell+0.021
 Karen+0.021
aren+0.021
 Ellen+0.019
Neg logit contributions
ble-0.030
bled-0.027
balls-0.025
ster-0.025
 fin-0.025
Token the
Token ID262
Feature activation-1.80e-02

Pos logit contributions
ird+0.034
rell+0.032
aren+0.031
ives+0.030
hell+0.030
Neg logit contributions
ble-0.022
bled-0.019
 diving-0.018
balls-0.017
ster-0.017
Token wanted
Token ID2227
Feature activation-3.38e-02

Pos logit contributions
ble+0.014
bled+0.013
balls+0.012
ster+0.012
 fin+0.012
Neg logit contributions
ird-0.010
 Karen-0.009
aren-0.009
rell-0.009
 Ellen-0.008
Token to
Token ID284
Feature activation-3.12e-02

Pos logit contributions
ird+0.086
rell+0.079
aren+0.077
 Karen+0.075
 Ellen+0.074
Neg logit contributions
ble-0.050
bled-0.042
balls-0.041
ster-0.038
 diving-0.037
Token get
Token ID651
Feature activation-2.17e-02

Pos logit contributions
ird+0.068
aren+0.065
rell+0.062
chairs+0.062
 Karen+0.061
Neg logit contributions
ble-0.056
bled-0.043
balls-0.041
 fin-0.041
ster-0.040
Token some
Token ID617
Feature activation-6.83e-02

Pos logit contributions
ird+0.044
rell+0.040
aren+0.040
 Karen+0.039
 Ellen+0.037
Neg logit contributions
ble-0.043
bled-0.039
balls-0.036
ster-0.035
 fin-0.035
Token of
Token ID286
Feature activation-2.95e-02

Pos logit contributions
 Ellen+0.129
ird+0.125
Ell+0.124
Aud+0.122
 Kathy+0.120
Neg logit contributions
ble-0.135
bled-0.106
 diving-0.103
balls-0.103
 next-0.103
Token the
Token ID262
Feature activation-5.96e-02

Pos logit contributions
ird+0.072
aren+0.063
rell+0.063
chair+0.061
chairs+0.061
Neg logit contributions
ble-0.049
bled-0.041
balls-0.037
 diving-0.036
 fin-0.036
Token cher
Token ID21382
Feature activation-1.82e-02

Pos logit contributions
ird+0.117
Aud+0.109
 Ellen+0.107
old+0.105
jo+0.101
Neg logit contributions
ble-0.118
 next-0.096
bled-0.093
balls-0.091
ster-0.084
Tokenries
Token ID1678
Feature activation-2.83e-02

Pos logit contributions
ird+0.037
 Karen+0.035
 Ellen+0.034
rell+0.034
aren+0.034
Neg logit contributions
ble-0.036
bled-0.031
ster-0.028
balls-0.028
 diving-0.025
Token but
Token ID475
Feature activation-2.24e-02

Pos logit contributions
ird+0.054
aren+0.051
rell+0.050
 Karen+0.050
 Ellen+0.048
Neg logit contributions
ble-0.060
bled-0.052
balls-0.049
ster-0.048
 fin-0.046
Token she
Token ID673
Feature activation+9.09e-03

Pos logit contributions
ird+0.040
rell+0.037
aren+0.036
 Karen+0.036
 Ellen+0.033
Neg logit contributions
ble-0.050
bled-0.045
balls-0.043
ster-0.042
 fin-0.041
Token couldn
Token ID3521
Feature activation-3.68e-02

Pos logit contributions
ble+0.019
bled+0.017
balls+0.016
 fin+0.016
ster+0.016
Neg logit contributions
ird-0.017
 Karen-0.016
aren-0.016
rell-0.015
 Stella-0.014