TOP ACTIVATIONS
MAX = 0.458

She helped Tom use the wrench to adjust the nut .
Can you hand me the wrench ?" Lily picked up the
?" Lily picked up the wrench and gave it to the
, " This is a wrench . I use it to
, you can hold this wrench and this hammer . And
angry when she used a hammer . <|endoftext|> John and his
. She is using a wrench . " Mom
man started to use his hammer to fix his house .
bottom of the hill , bruised and bleeding . He was
you help me find the wrench ?". Sarah happily
them a jack and a wrench . They helped him lift
a screw driver , a wrench , and a flashlight .
it was a very special hammer . Thomas noticed that the
hit the bird with his hammer . The bird could not
home , but he was bruised and battered . His mummy
and kept playing with the hammer . Suddenly , he hit
, " I miss my hammer . Can you help me
car . He had a wrench and a jack . He
, she found a shiny wrench hidden among the plants .
She twists the wrench . She turns the tap

MIN ACTIVATIONS
MIN = -0.480

From then on , Julia would take her orange telephone
telephone as a gift . Julia gigg led with joy and
s that ? â Julia asked her cat curiously .
waved as they left . Julia felt happy to have borrowed
little girl called Julia . Julia was a happy child and
outside . One day , Julia found an orange telephone in
lived a little girl called Julia . Julia was a happy
was a little girl named Julia . She was only three
ne ared the top , Julia stopped in her tracks to
. Suddenly , Julia heard a mysterious voice .
strong ! Suddenly Julia heard a sn ort behind
lend him her boat . Julia said yes ; with a
Sam waved goodbye and thanked Julia for being so clever .
the leaves began to move Julia smiled . She felt brave
me about the crystal !" Julia went closer and saw that
couldn 't see anyone . Julia was scared , so she
, a little girl named Julia went exploring in the woods
the creek . He asked Julia if she would lend him
Together , Julia and her friends worked hard
to the town ," pleaded Julia . Her

INTERVAL 0.224 - 0.458
CONTAINS 0.858%

had to be brave and help the animal . They used
went to a store to buy some things . The store
if he could borrow her comb to comb his fur and
cat named Mitt ens . Mitt ens liked to play with
down and looked in his sack . The toy marriage made

INTERVAL -0.011 - 0.224
CONTAINS 63.018%

The rabbit hopped away , but it turned its head
shore and he drowned in the sea . The
and faster . Do you want to buy some ?"
want to hug the bear . And I want to hug
the clown jugg led balls , spun plates and rode a

INTERVAL -0.245 - -0.011
CONTAINS 35.671%

One day a pilot flew through the sky in a big
a rocket or a bird ." " OK !"
and from then on she was never sad again . <|endoftext|>
called Lisa . Lisa had a diary that she kept all
fish sw am away and Daisy felt happy and proud .

INTERVAL -0.480 - -0.245
CONTAINS 0.453%

decided to build a sand castle at the beach . They
was a young bear called Flash . He had very soft
castle so much . The mer maid said , " I
liked . Tom picked a movie about cars and trucks .
of the day , the superhero brought the little girl safely
TokenShe
Token ID3347
Feature activation+3.15e-02

Pos logit contributions
+0.009
 wrench+0.009
 swollen+0.009
 bruises+0.009
 hammer+0.008
Neg logit contributions
 Julia-0.012
 Mae-0.011
 Amber-0.011
 dolphins-0.011
La-0.011
Token helped
Token ID4193
Feature activation+1.47e-01

Pos logit contributions
 hammer+0.045
 bruised+0.045
 wrench+0.045
+0.045
 bruises+0.044
Neg logit contributions
 Julia-0.058
La-0.055
 Mae-0.053
 Amber-0.052
 mate-0.051
Token Tom
Token ID4186
Feature activation-4.71e-02

Pos logit contributions
 bruised+0.262
+0.260
 wrench+0.256
 swollen+0.255
 bruises+0.249
Neg logit contributions
 Julia-0.241
 Mae-0.216
La-0.213
 Amber-0.211
 Jeff-0.201
Token use
Token ID779
Feature activation+1.06e-01

Pos logit contributions
 Julia+0.077
 Mae+0.071
 Amber+0.068
La+0.068
 Lou+0.066
Neg logit contributions
 swollen-0.080
-0.080
 bruised-0.079
 bruises-0.077
 wrench-0.076
Token the
Token ID262
Feature activation+5.46e-02

Pos logit contributions
+0.235
leaning+0.226
pling+0.226
uc+0.221
 bruised+0.220
Neg logit contributions
 Julia-0.133
 Mae-0.114
 Jeff-0.107
 Amber-0.107
 Lou-0.104
Token wrench
Token ID34561
Feature activation+4.58e-01

Pos logit contributions
+0.078
leaning+0.073
pling+0.073
 bruised+0.070
uc+0.069
Neg logit contributions
 Julia-0.100
 Amber-0.092
 Mae-0.092
 dolphins-0.092
 dives-0.091
Token to
Token ID284
Feature activation+4.91e-03

Pos logit contributions
+0.871
 swollen+0.868
 bruised+0.861
 bruises+0.846
 wrench+0.816
Neg logit contributions
 Julia-0.661
 Mae-0.576
 Amber-0.554
La-0.551
 Jeff-0.550
Token adjust
Token ID4532
Feature activation+8.64e-02

Pos logit contributions
 bruised+0.009
 swollen+0.009
pling+0.009
+0.008
uc+0.008
Neg logit contributions
 dive-0.006
 Julia-0.006
 swim-0.006
 perform-0.006
 dance-0.006
Token the
Token ID262
Feature activation+3.78e-02

Pos logit contributions
+0.186
 swollen+0.172
leaning+0.172
 bruises+0.170
 bruised+0.170
Neg logit contributions
 Julia-0.111
 Zac-0.092
 Mae-0.091
 Amber-0.086
 Eric-0.085
Token nut
Token ID6701
Feature activation+1.38e-01

Pos logit contributions
+0.054
leaning+0.049
uc+0.048
pling+0.047
 swollen+0.046
Neg logit contributions
 Julia-0.064
 mate-0.063
 dolphins-0.062
 aliens-0.060
La-0.060
Token.
Token ID13
Feature activation+6.48e-03

Pos logit contributions
+0.241
 bruised+0.238
 swollen+0.233
 bruises+0.232
 wrench+0.232
Neg logit contributions
 Julia-0.221
 Mae-0.200
La-0.195
 Amber-0.195
 Zac-0.189
Token Can
Token ID1680
Feature activation+2.98e-02

Pos logit contributions
 bruised+0.029
 swollen+0.028
 wrench+0.028
+0.028
 bruises+0.028
Neg logit contributions
 Julia-0.029
La-0.026
 Mae-0.026
 Amber-0.025
 Zac-0.025
Token you
Token ID345
Feature activation-2.34e-02

Pos logit contributions
 bruised+0.061
 wrench+0.061
+0.060
 swollen+0.060
 bruises+0.059
Neg logit contributions
 Julia-0.039
La-0.034
 Mae-0.034
 Amber-0.033
 Zac-0.032
Token hand
Token ID1021
Feature activation+1.26e-01

Pos logit contributions
 Julia+0.048
La+0.044
 Mae+0.044
 Amber+0.043
 Lou+0.042
Neg logit contributions
 bruised-0.031
 swollen-0.030
 bruises-0.029
 wrench-0.029
-0.028
Token me
Token ID502
Feature activation-1.16e-01

Pos logit contributions
+0.234
 bruised+0.232
 wrench+0.228
 swollen+0.227
 bruises+0.224
Neg logit contributions
 Julia-0.188
La-0.171
 Mae-0.168
 Amber-0.165
 Zac-0.157
Token the
Token ID262
Feature activation+6.59e-02

Pos logit contributions
 Julia+0.169
La+0.153
 Mae+0.151
 Amber+0.149
 Zac+0.141
Neg logit contributions
 swollen-0.213
 bruised-0.212
 bruises-0.210
-0.209
 wrench-0.207
Token wrench
Token ID34561
Feature activation+4.52e-01

Pos logit contributions
+0.089
leaning+0.084
uc+0.080
ised+0.078
 swollen+0.077
Neg logit contributions
 princess-0.117
 mate-0.115
 knight-0.114
 actor-0.113
 pirate-0.112
Token?"
Token ID1701
Feature activation-1.13e-01

Pos logit contributions
 bruised+0.850
+0.847
 wrench+0.839
 swollen+0.833
 bruises+0.827
Neg logit contributions
 Julia-0.655
La-0.593
 Mae-0.581
 Amber-0.570
 Zac-0.544
Token Lily
Token ID20037
Feature activation-2.63e-02

Pos logit contributions
 Julia+0.208
La+0.195
 Mae+0.190
 Amber+0.186
 Zac+0.181
Neg logit contributions
-0.167
 bruised-0.166
 wrench-0.164
 hammer-0.160
 swollen-0.160
Token picked
Token ID6497
Feature activation+1.63e-01

Pos logit contributions
 Julia+0.045
La+0.042
 Mae+0.040
 Amber+0.040
 Zac+0.038
Neg logit contributions
-0.043
 bruised-0.042
 wrench-0.042
 hammer-0.042
 swollen-0.042
Token up
Token ID510
Feature activation-1.08e-01

Pos logit contributions
+0.379
 bruised+0.375
 bruises+0.375
 swollen+0.374
 wrench+0.359
Neg logit contributions
 Julia-0.183
 Mae-0.157
 Amber-0.155
La-0.148
 Zac-0.146
Token the
Token ID262
Feature activation+6.24e-02

Pos logit contributions
 Julia+0.150
 Amber+0.131
 Alicia+0.123
 Mae+0.122
La+0.121
Neg logit contributions
 bruised-0.227
 wrench-0.223
-0.221
 supervision-0.221
 bruises-0.220
Token?"
Token ID1701
Feature activation-1.13e-01

Pos logit contributions
 bruised+0.850
+0.847
 wrench+0.839
 swollen+0.833
 bruises+0.827
Neg logit contributions
 Julia-0.655
La-0.593
 Mae-0.581
 Amber-0.570
 Zac-0.544
Token Lily
Token ID20037
Feature activation-2.63e-02

Pos logit contributions
 Julia+0.208
La+0.195
 Mae+0.190
 Amber+0.186
 Zac+0.181
Neg logit contributions
-0.167
 bruised-0.166
 wrench-0.164
 hammer-0.160
 swollen-0.160
Token picked
Token ID6497
Feature activation+1.63e-01

Pos logit contributions
 Julia+0.045
La+0.042
 Mae+0.040
 Amber+0.040
 Zac+0.038
Neg logit contributions
-0.043
 bruised-0.042
 wrench-0.042
 hammer-0.042
 swollen-0.042
Token up
Token ID510
Feature activation-1.08e-01

Pos logit contributions
+0.379
 bruised+0.375
 bruises+0.375
 swollen+0.374
 wrench+0.359
Neg logit contributions
 Julia-0.183
 Mae-0.157
 Amber-0.155
La-0.148
 Zac-0.146
Token the
Token ID262
Feature activation+6.24e-02

Pos logit contributions
 Julia+0.150
 Amber+0.131
 Alicia+0.123
 Mae+0.122
La+0.121
Neg logit contributions
 bruised-0.227
 wrench-0.223
-0.221
 supervision-0.221
 bruises-0.220
Token wrench
Token ID34561
Feature activation+4.50e-01

Pos logit contributions
+0.083
leaning+0.077
uc+0.071
ised+0.071
anked+0.069
Neg logit contributions
 knight-0.111
 pirate-0.106
 princess-0.104
 mate-0.101
 king-0.098
Token and
Token ID290
Feature activation+2.38e-02

Pos logit contributions
 bruises+0.901
+0.892
 swollen+0.891
 bruised+0.883
 mercury+0.850
Neg logit contributions
 Julia-0.606
 Mae-0.531
 Amber-0.512
 Zac-0.494
 Jeff-0.490
Token gave
Token ID2921
Feature activation+7.44e-02

Pos logit contributions
+0.043
 bruised+0.042
 swollen+0.040
 bruises+0.040
 wrench+0.040
Neg logit contributions
 Julia-0.038
 Mae-0.034
 Amber-0.033
La-0.033
 Zac-0.032
Token it
Token ID340
Feature activation+1.94e-02

Pos logit contributions
+0.154
 bruised+0.146
 swollen+0.145
 wrench+0.144
 bruises+0.142
Neg logit contributions
 Julia-0.110
 Mae-0.096
 Amber-0.093
La-0.090
 Jeff-0.089
Token to
Token ID284
Feature activation+1.32e-02

Pos logit contributions
+0.034
 bruised+0.034
 wrench+0.033
 swollen+0.033
 bruises+0.032
Neg logit contributions
 Julia-0.031
La-0.028
 Mae-0.028
 Amber-0.027
 Lou-0.026
Token the
Token ID262
Feature activation+5.96e-02

Pos logit contributions
 bruised+0.023
+0.023
 wrench+0.022
 swollen+0.022
 bruises+0.022
Neg logit contributions
 Julia-0.022
 Mae-0.020
La-0.020
 Amber-0.020
 April-0.019
Token,
Token ID11
Feature activation+3.83e-03

Pos logit contributions
 wrench+0.009
 hammer+0.009
 bruised+0.009
 swollen+0.009
+0.009
Neg logit contributions
 Julia-0.013
La-0.013
 Mae-0.012
 Amber-0.012
 April-0.012
Token "
Token ID366
Feature activation+4.50e-02

Pos logit contributions
 bruised+0.006
 bruises+0.006
+0.006
 swollen+0.006
 wrench+0.006
Neg logit contributions
 Julia-0.006
 Mae-0.006
La-0.006
 Amber-0.006
 April-0.006
TokenThis
Token ID1212
Feature activation+1.34e-01

Pos logit contributions
+0.055
 bruised+0.054
 wrench+0.054
 bruises+0.052
 swollen+0.052
Neg logit contributions
 Julia-0.095
La-0.088
 Mae-0.087
 Amber-0.087
 April-0.084
Token is
Token ID318
Feature activation-4.35e-02

Pos logit contributions
 wrench+0.224
 hammer+0.224
+0.220
 bruised+0.217
 bruises+0.209
Neg logit contributions
 Julia-0.229
La-0.217
 Amber-0.206
 Mae-0.206
 April-0.197
Token a
Token ID257
Feature activation-5.88e-02

Pos logit contributions
 Julia+0.066
 Mae+0.059
La+0.058
 Amber+0.058
 Jeff+0.057
Neg logit contributions
-0.082
 bruised-0.079
 swollen-0.077
 wrench-0.077
 bruises-0.076
Token wrench
Token ID34561
Feature activation+4.49e-01

Pos logit contributions
 Julia+0.101
La+0.093
 Mae+0.091
 Amber+0.090
 Zac+0.087
Neg logit contributions
 bruised-0.096
-0.095
 wrench-0.095
 hammer-0.093
 swollen-0.093
Token.
Token ID13
Feature activation+1.98e-02

Pos logit contributions
+0.792
 bruised+0.787
 swollen+0.785
 bruises+0.773
 wrench+0.759
Neg logit contributions
 Julia-0.691
 Mae-0.620
La-0.612
 Amber-0.600
 Zac-0.590
Token I
Token ID314
Feature activation-1.41e-01

Pos logit contributions
 swollen+0.035
 bruised+0.035
+0.035
 wrench+0.034
 bruises+0.033
Neg logit contributions
 Julia-0.031
La-0.028
 Mae-0.028
 Amber-0.027
 Zac-0.027
Token use
Token ID779
Feature activation+1.14e-01

Pos logit contributions
 Julia+0.205
 mate+0.195
 dives+0.192
 actor+0.192
 Mae+0.186
Neg logit contributions
 swollen-0.250
 bruised-0.242
 bruises-0.235
ados-0.234
 supervision-0.229
Token it
Token ID340
Feature activation+2.16e-02

Pos logit contributions
+0.254
 bruised+0.239
 swollen+0.238
cut+0.229
pling+0.227
Neg logit contributions
 Julia-0.110
 mate-0.096
 dolphins-0.091
 dives-0.090
 dances-0.088
Token to
Token ID284
Feature activation+1.33e-02

Pos logit contributions
+0.036
 bruised+0.036
 wrench+0.035
 swollen+0.035
 bruises+0.034
Neg logit contributions
 Julia-0.036
La-0.033
 Mae-0.033
 Amber-0.032
 Zac-0.031
Token,
Token ID11
Feature activation+5.75e-03

Pos logit contributions
 bruised+0.176
+0.175
 wrench+0.174
 swollen+0.171
 hammer+0.171
Neg logit contributions
 Julia-0.178
La-0.165
 Mae-0.161
 Amber-0.157
 Zac-0.152
Token you
Token ID345
Feature activation-2.68e-02

Pos logit contributions
+0.010
 bruised+0.010
 wrench+0.010
 swollen+0.010
 bruises+0.010
Neg logit contributions
 Julia-0.009
La-0.008
 Mae-0.008
 Amber-0.008
 Zac-0.008
Token can
Token ID460
Feature activation+1.77e-01

Pos logit contributions
 Julia+0.044
 mate+0.043
 dives+0.041
La+0.040
 actor+0.040
Neg logit contributions
 swollen-0.043
 bruised-0.041
-0.040
 bruises-0.040
pling-0.039
Token hold
Token ID1745
Feature activation+8.68e-02

Pos logit contributions
 bruised+0.259
 swollen+0.251
+0.251
 wrench+0.251
 bruises+0.248
Neg logit contributions
 Julia-0.331
La-0.306
 Mae-0.302
 Amber-0.297
 Zac-0.289
Token this
Token ID428
Feature activation+4.86e-02

Pos logit contributions
+0.194
 swollen+0.187
 bruised+0.183
 scrape+0.178
cut+0.174
Neg logit contributions
 Julia-0.098
 Amber-0.085
 Mae-0.083
 dolphins-0.079
 April-0.076
Token wrench
Token ID34561
Feature activation+4.49e-01

Pos logit contributions
+0.077
 swollen+0.069
uc+0.068
 bruises+0.068
 bruised+0.065
Neg logit contributions
 mate-0.080
 Julia-0.078
 princess-0.074
 royal-0.074
 aliens-0.073
Token and
Token ID290
Feature activation+1.86e-02

Pos logit contributions
+0.817
 bruised+0.803
 swollen+0.802
 bruises+0.789
 wrench+0.783
Neg logit contributions
 Julia-0.676
 Mae-0.608
La-0.603
 Amber-0.593
 Zac-0.572
Token this
Token ID428
Feature activation+4.66e-02

Pos logit contributions
+0.035
 swollen+0.033
 bruised+0.032
pling+0.032
uc+0.032
Neg logit contributions
 Julia-0.024
 mate-0.023
 dives-0.023
La-0.022
 astronauts-0.022
Token hammer
Token ID15554
Feature activation+4.37e-01

Pos logit contributions
+0.069
 swollen+0.069
 bruised+0.066
uc+0.065
 bruises+0.065
Neg logit contributions
 mate-0.075
 Julia-0.073
 dives-0.069
 pirate-0.069
 actor-0.068
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
+0.776
 bruised+0.759
 swollen+0.749
 wrench+0.746
 bruises+0.736
Neg logit contributions
 Julia-0.678
La-0.613
 Mae-0.605
 Amber-0.588
 Zac-0.576
Token And
Token ID843
Feature activation+9.28e-03

Pos logit contributions
 swollen+0.017
 bruised+0.017
+0.017
 wrench+0.017
 hammer+0.017
Neg logit contributions
 Julia-0.017
La-0.016
 Mae-0.016
 Amber-0.015
 April-0.015
Token angry
Token ID7954
Feature activation+8.33e-02

Pos logit contributions
 Julia+0.019
La+0.017
 Mae+0.017
 Amber+0.017
 Zac+0.017
Neg logit contributions
 bruised-0.014
 swollen-0.014
 hammer-0.013
 bruises-0.013
 wrench-0.013
Token when
Token ID618
Feature activation+1.67e-02

Pos logit contributions
 bruised+0.126
 hammer+0.118
+0.117
 bruises+0.114
 wrench+0.113
Neg logit contributions
 Julia-0.147
La-0.146
 Mae-0.140
 Amber-0.138
 April-0.138
Token she
Token ID673
Feature activation-3.48e-02

Pos logit contributions
 bruised+0.025
+0.024
 swollen+0.024
 hammer+0.024
 bruises+0.024
Neg logit contributions
 Julia-0.030
La-0.029
 Mae-0.027
 Zac-0.027
 mate-0.027
Token used
Token ID973
Feature activation+8.09e-02

Pos logit contributions
 Julia+0.061
 Mae+0.056
La+0.056
 Amber+0.055
 Zac+0.054
Neg logit contributions
 bruised-0.056
-0.055
 hammer-0.053
 bruises-0.053
 swollen-0.052
Token a
Token ID257
Feature activation-6.63e-02

Pos logit contributions
+0.150
 wrench+0.146
 bruised+0.143
pling+0.142
 swollen+0.139
Neg logit contributions
 Julia-0.127
 Mae-0.110
 Amber-0.109
La-0.109
 Jeff-0.106
Token hammer
Token ID15554
Feature activation+4.47e-01

Pos logit contributions
 Julia+0.115
La+0.105
 Mae+0.103
 Amber+0.103
 Lou+0.101
Neg logit contributions
-0.102
 bruised-0.099
leaning-0.098
 bruises-0.097
 swollen-0.095
Token.
Token ID13
Feature activation+1.01e-02

Pos logit contributions
+0.812
 bruised+0.801
 swollen+0.797
 wrench+0.792
 bruises+0.790
Neg logit contributions
 Julia-0.682
La-0.607
 Mae-0.599
 Amber-0.586
 Lou-0.562
Token<|endoftext|>
Token ID50256
Feature activation+4.90e-02

Pos logit contributions
+0.016
 bruised+0.016
 wrench+0.016
 swollen+0.016
 hammer+0.016
Neg logit contributions
 Julia-0.018
La-0.016
 Mae-0.016
 Amber-0.016
 Zac-0.015
TokenJohn
Token ID7554
Feature activation-2.72e-02

Pos logit contributions
+0.071
 bruised+0.063
 hammer+0.062
 wrench+0.062
pling+0.062
Neg logit contributions
 Julia-0.095
La-0.089
 Mae-0.087
 actor-0.086
 Amber-0.086
Token and
Token ID290
Feature activation+1.82e-02

Pos logit contributions
 Julia+0.048
La+0.044
 Mae+0.043
 Amber+0.042
 Zac+0.041
Neg logit contributions
 bruised-0.043
-0.043
 wrench-0.042
 hammer-0.042
 swollen-0.041
Token his
Token ID465
Feature activation-4.81e-02

Pos logit contributions
 bruised+0.024
 hammer+0.024
 wrench+0.023
pling+0.023
 rag+0.022
Neg logit contributions
 Julia-0.035
La-0.034
 Mae-0.032
ae-0.032
 pir-0.031
Token.
Token ID13
Feature activation+2.21e-02

Pos logit contributions
 bruised+0.097
+0.097
 wrench+0.095
 swollen+0.095
 bruises+0.094
Neg logit contributions
 Julia-0.098
La-0.089
 Mae-0.088
 Amber-0.086
 Zac-0.083
Token She
Token ID1375
Feature activation+7.43e-03

Pos logit contributions
+0.034
 bruised+0.034
 wrench+0.034
 swollen+0.033
 hammer+0.033
Neg logit contributions
 Julia-0.039
La-0.037
 Mae-0.036
 Amber-0.035
 Zac-0.034
Token is
Token ID318
Feature activation-3.99e-02

Pos logit contributions
 bruised+0.013
 bruises+0.013
 wrench+0.012
 swollen+0.012
 hammer+0.012
Neg logit contributions
 Julia-0.012
La-0.011
 Mae-0.011
 Amber-0.011
 Lou-0.010
Token using
Token ID1262
Feature activation+1.07e-01

Pos logit contributions
 Julia+0.062
La+0.057
 Mae+0.055
 Amber+0.053
 Jeff+0.051
Neg logit contributions
-0.075
 wrench-0.070
pling-0.068
leaning-0.067
 bruised-0.066
Token a
Token ID257
Feature activation-6.54e-02

Pos logit contributions
+0.230
pling+0.218
leaning+0.206
uc+0.205
cut+0.203
Neg logit contributions
 Julia-0.138
 Mae-0.117
 dolphins-0.112
 Amber-0.111
 dances-0.109
Token wrench
Token ID34561
Feature activation+4.46e-01

Pos logit contributions
 Julia+0.132
La+0.122
 Mae+0.122
 dives+0.121
 Amber+0.119
Neg logit contributions
-0.081
 bruised-0.079
 bruises-0.077
leaning-0.076
pling-0.074
Token.
Token ID13
Feature activation+2.16e-02

Pos logit contributions
+0.933
 swollen+0.907
 bruised+0.885
 bruises+0.872
leaning+0.850
Neg logit contributions
 Julia-0.566
 Mae-0.482
La-0.471
 Zac-0.460
 Jeff-0.457
Token
Token ID198
Feature activation-2.31e-03

Pos logit contributions
 bruised+0.035
+0.035
 wrench+0.035
 swollen+0.034
 hammer+0.034
Neg logit contributions
 Julia-0.037
La-0.035
 Mae-0.034
 Amber-0.033
 Zac-0.032
Token
Token ID198
Feature activation-1.47e-03

Pos logit contributions
 Julia+0.004
La+0.003
 Mae+0.003
 Amber+0.003
 Zac+0.003
Neg logit contributions
 bruised-0.004
-0.004
 wrench-0.004
 swollen-0.004
 hammer-0.004
Token"
Token ID1
Feature activation+1.57e-02

Pos logit contributions
 Julia+0.002
La+0.002
 Mae+0.002
 Amber+0.002
 Zac+0.002
Neg logit contributions
 bruised-0.003
-0.003
 wrench-0.003
 hammer-0.002
 swollen-0.002
TokenMom
Token ID29252
Feature activation+9.73e-02

Pos logit contributions
 bruised+0.027
+0.026
 wrench+0.025
 hammer+0.025
pling+0.025
Neg logit contributions
La-0.028
 Julia-0.027
 Mae-0.025
 Amber-0.024
 Zac-0.023
Token man
Token ID582
Feature activation+1.04e-02

Pos logit contributions
+0.130
 bruised+0.129
 wrench+0.128
 swollen+0.125
 bruises+0.124
Neg logit contributions
 Julia-0.115
La-0.105
 Mae-0.103
 Amber-0.101
 Zac-0.096
Token started
Token ID2067
Feature activation-6.58e-03

Pos logit contributions
 bruised+0.015
 hammer+0.015
 bruises+0.014
 wrench+0.014
+0.014
Neg logit contributions
 Julia-0.019
La-0.019
 Mae-0.018
 Amber-0.018
 April-0.017
Token to
Token ID284
Feature activation+1.56e-02

Pos logit contributions
 Julia+0.012
La+0.011
 Mae+0.011
 Amber+0.010
 Zac+0.010
Neg logit contributions
 bruised-0.010
 wrench-0.010
-0.010
 hammer-0.010
 swollen-0.010
Token use
Token ID779
Feature activation+1.17e-01

Pos logit contributions
 bruised+0.025
 hammer+0.025
 wrench+0.025
 swollen+0.025
+0.024
Neg logit contributions
 Julia-0.027
La-0.025
 Mae-0.024
 Amber-0.024
 Zac-0.023
Token his
Token ID465
Feature activation-4.06e-02

Pos logit contributions
+0.241
pling+0.226
leaning+0.226
uc+0.221
 bruised+0.220
Neg logit contributions
 Julia-0.163
 Jeff-0.139
 Lou-0.138
 Mae-0.138
 Amber-0.134
Token hammer
Token ID15554
Feature activation+4.45e-01

Pos logit contributions
 Julia+0.074
La+0.069
 Mae+0.068
 Amber+0.067
 Lou+0.065
Neg logit contributions
-0.060
 bruised-0.059
 wrench-0.057
pling-0.057
 swollen-0.056
Token to
Token ID284
Feature activation+1.26e-02

Pos logit contributions
+0.939
 swollen+0.873
 bruised+0.873
 bruises+0.855
leaning+0.855
Neg logit contributions
 Julia-0.567
 Lou-0.480
La-0.469
 Jeff-0.461
 Mae-0.460
Token fix
Token ID4259
Feature activation+2.60e-01

Pos logit contributions
+0.022
pling+0.021
 bruised+0.021
leaning+0.021
 swollen+0.021
Neg logit contributions
 Julia-0.020
 Lou-0.018
 dive-0.017
 Mae-0.017
La-0.017
Token his
Token ID465
Feature activation-4.34e-02

Pos logit contributions
+0.514
 bruised+0.478
 swollen+0.474
leaning+0.467
 bruises+0.466
Neg logit contributions
 Julia-0.388
 Mae-0.333
 Jeff-0.329
 Zac-0.324
 Lou-0.320
Token house
Token ID2156
Feature activation+4.41e-02

Pos logit contributions
 Julia+0.074
ia+0.071
 mate+0.071
 royal+0.070
 pirate+0.070
Neg logit contributions
-0.062
leaning-0.056
Here-0.055
pling-0.055
 bruised-0.053
Token.
Token ID13
Feature activation+1.85e-02

Pos logit contributions
+0.093
 swollen+0.091
leaning+0.089
 bruises+0.089
 bruised+0.088
Neg logit contributions
 Julia-0.053
 Mae-0.044
 Jeff-0.043
 Lou-0.042
 Zac-0.041
Token bottom
Token ID4220
Feature activation-5.41e-02

Pos logit contributions
 bruised+0.094
+0.093
 swollen+0.092
 bruises+0.091
 wrench+0.091
Neg logit contributions
 Julia-0.072
La-0.065
 Mae-0.065
 Amber-0.063
 Zac-0.062
Token of
Token ID286
Feature activation-5.17e-02

Pos logit contributions
 Julia+0.095
La+0.092
 Amber+0.087
 Mae+0.086
 Zac+0.085
Neg logit contributions
-0.084
 hammer-0.084
 wrench-0.083
 bruised-0.083
 bruises-0.080
Token the
Token ID262
Feature activation+4.92e-02

Pos logit contributions
 Julia+0.088
 Mae+0.079
 Amber+0.078
La+0.076
 Jeff+0.075
Neg logit contributions
 bruised-0.089
-0.088
 wrench-0.087
pling-0.086
 swollen-0.085
Token hill
Token ID12788
Feature activation+7.78e-02

Pos logit contributions
 bruised+0.085
 swollen+0.083
uc+0.082
 bruises+0.081
+0.081
Neg logit contributions
 Julia-0.081
 Mae-0.073
La-0.073
 Zac-0.071
 Jeff-0.070
Token,
Token ID11
Feature activation+4.24e-03

Pos logit contributions
 bruised+0.129
+0.128
 wrench+0.128
 swollen+0.126
 bruises+0.124
Neg logit contributions
 Julia-0.131
La-0.119
 Mae-0.118
 Amber-0.116
 Zac-0.112
Token bruised
Token ID44379
Feature activation+4.45e-01

Pos logit contributions
+0.007
 wrench+0.007
 swollen+0.007
 bruised+0.007
pling+0.007
Neg logit contributions
 Julia-0.007
 Mae-0.007
 Amber-0.006
La-0.006
 Zac-0.006
Token and
Token ID290
Feature activation+1.69e-02

Pos logit contributions
+0.805
 wrench+0.785
 bruised+0.782
 swollen+0.778
 hammer+0.771
Neg logit contributions
 Julia-0.693
 Mae-0.625
La-0.610
 Amber-0.609
 Zac-0.598
Token bleeding
Token ID16832
Feature activation+3.20e-01

Pos logit contributions
+0.027
 wrench+0.024
leaning+0.024
pling+0.024
issors+0.024
Neg logit contributions
 Julia-0.030
 Mae-0.028
La-0.027
 Amber-0.027
 Jeff-0.027
Token.
Token ID13
Feature activation+1.06e-02

Pos logit contributions
 bruised+0.510
+0.500
 wrench+0.498
 swollen+0.490
 hammer+0.486
Neg logit contributions
 Julia-0.570
La-0.528
 Mae-0.516
 Amber-0.510
 Zac-0.488
Token He
Token ID679
Feature activation-1.92e-02

Pos logit contributions
+0.017
 bruised+0.016
 wrench+0.016
 bruises+0.016
 swollen+0.015
Neg logit contributions
 Julia-0.018
La-0.017
 actor-0.016
 Mae-0.016
 oper-0.016
Token was
Token ID373
Feature activation-4.00e-02

Pos logit contributions
 Julia+0.033
La+0.030
 Mae+0.030
 Amber+0.029
 Zac+0.028
Neg logit contributions
 bruised-0.032
 wrench-0.031
-0.031
 swollen-0.031
 bruises-0.031
Token you
Token ID345
Feature activation-3.04e-02

Pos logit contributions
+0.046
 bruised+0.045
 wrench+0.045
 swollen+0.044
 hammer+0.044
Neg logit contributions
 Julia-0.035
La-0.032
 Mae-0.031
 Amber-0.030
 Zac-0.029
Token help
Token ID1037
Feature activation+2.47e-01

Pos logit contributions
 Julia+0.058
La+0.054
 Lou+0.052
 mate+0.052
 Mae+0.052
Neg logit contributions
 bruised-0.042
 swollen-0.041
 bruises-0.041
-0.040
 mercury-0.040
Token me
Token ID502
Feature activation-1.25e-01

Pos logit contributions
 hammer+0.480
 wrench+0.477
 bruised+0.475
+0.473
 swollen+0.466
Neg logit contributions
 Julia-0.342
La-0.310
 Mae-0.301
 Amber-0.294
 Zac-0.290
Token find
Token ID1064
Feature activation-7.00e-02

Pos logit contributions
 Julia+0.198
La+0.183
 Mae+0.177
 Amber+0.175
 Lou+0.170
Neg logit contributions
 bruised-0.208
-0.208
 swollen-0.206
 bruises-0.205
 wrench-0.200
Token the
Token ID262
Feature activation+5.80e-02

Pos logit contributions
 Julia+0.110
La+0.101
 Mae+0.099
 Amber+0.097
 Zac+0.093
Neg logit contributions
-0.124
 bruised-0.124
 wrench-0.122
 swollen-0.121
 hammer-0.119
Token wrench
Token ID34561
Feature activation+4.45e-01

Pos logit contributions
+0.105
 bruised+0.102
 wrench+0.101
 bruises+0.100
 swollen+0.100
Neg logit contributions
 Julia-0.087
La-0.081
 Mae-0.078
 Amber-0.077
 April-0.076
Token?".
Token ID43634
Feature activation-1.11e-01

Pos logit contributions
 bruised+0.861
+0.858
 swollen+0.848
 wrench+0.845
 bruises+0.842
Neg logit contributions
 Julia-0.620
La-0.556
 Mae-0.548
 Amber-0.539
 Zac-0.511
Token
Token ID198
Feature activation-2.85e-03

Pos logit contributions
 Julia+0.179
La+0.153
 Amber+0.151
 Mae+0.150
 Lou+0.150
Neg logit contributions
 bruised-0.213
 swollen-0.210
 wrench-0.203
 bruises-0.201
 mercury-0.200
Token
Token ID198
Feature activation-2.36e-03

Pos logit contributions
 Julia+0.005
La+0.004
 Mae+0.004
 Amber+0.004
 Zac+0.004
Neg logit contributions
-0.005
 bruised-0.005
 wrench-0.005
 swollen-0.005
 hammer-0.005
TokenSarah
Token ID29284
Feature activation-9.62e-03

Pos logit contributions
 Julia+0.004
La+0.004
 Mae+0.004
 Amber+0.004
 Lou+0.004
Neg logit contributions
 bruised-0.004
-0.004
 wrench-0.004
 swollen-0.004
 bruises-0.004
Token happily
Token ID18177
Feature activation-1.51e-01

Pos logit contributions
 Julia+0.016
La+0.014
 Mae+0.014
 Amber+0.014
 Zac+0.013
Neg logit contributions
-0.016
 bruised-0.016
 wrench-0.016
 swollen-0.016
 hammer-0.016
Token them
Token ID606
Feature activation-7.39e-02

Pos logit contributions
 bruised+0.113
 wrench+0.113
 hammer+0.112
 oil+0.110
 bruises+0.109
Neg logit contributions
 Julia-0.113
La-0.110
 Mae-0.105
 Amber-0.102
ae-0.101
Token a
Token ID257
Feature activation-6.62e-02

Pos logit contributions
 Julia+0.110
La+0.099
 Mae+0.098
 Amber+0.097
 dolphins+0.092
Neg logit contributions
-0.141
 bruised-0.138
 swollen-0.136
 wrench-0.135
 hammer-0.130
Token jack
Token ID14509
Feature activation-7.74e-03

Pos logit contributions
 Julia+0.093
La+0.088
 Mae+0.083
 dives+0.083
 Amber+0.083
Neg logit contributions
-0.120
leaning-0.116
 bruised-0.115
 supervision-0.113
 bruises-0.111
Token and
Token ID290
Feature activation+1.45e-02

Pos logit contributions
 Julia+0.012
La+0.011
 Mae+0.011
 Amber+0.010
 mate+0.010
Neg logit contributions
-0.014
 bruised-0.014
 swollen-0.014
 wrench-0.013
 bruises-0.013
Token a
Token ID257
Feature activation-6.59e-02

Pos logit contributions
+0.026
 bruised+0.025
leaning+0.024
 mercury+0.023
 swollen+0.023
Neg logit contributions
 Julia-0.022
La-0.020
 Amber-0.020
 Mae-0.020
 April-0.019
Token wrench
Token ID34561
Feature activation+4.45e-01

Pos logit contributions
 Julia+0.095
 mate+0.092
La+0.091
 royal+0.090
 princess+0.089
Neg logit contributions
leaning-0.113
-0.108
ised-0.106
issors-0.104
mons-0.103
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
+0.894
 swollen+0.877
 bruised+0.863
leaning+0.845
 bruises+0.838
Neg logit contributions
 Julia-0.563
 Zac-0.488
 Mae-0.483
 mate-0.481
 Jeff-0.476
Token They
Token ID1119
Feature activation-2.52e-02

Pos logit contributions
 bruised+0.019
+0.019
 wrench+0.019
 swollen+0.019
 hammer+0.018
Neg logit contributions
 Julia-0.017
La-0.016
 Mae-0.015
 Amber-0.015
 Zac-0.015
Token helped
Token ID4193
Feature activation+1.44e-01

Pos logit contributions
 Julia+0.036
 Mae+0.032
La+0.031
 mate+0.031
 Ro+0.031
Neg logit contributions
-0.046
 bruised-0.046
 swollen-0.045
 supervision-0.044
 mercury-0.043
Token him
Token ID683
Feature activation+4.22e-02

Pos logit contributions
 bruised+0.263
+0.262
 wrench+0.261
 swollen+0.256
 hammer+0.255
Neg logit contributions
 Julia-0.219
La-0.200
 Mae-0.195
 Amber-0.192
 Zac-0.183
Token lift
Token ID10303
Feature activation+1.03e-01

Pos logit contributions
 swollen+0.067
+0.064
 bruised+0.064
leaning+0.062
 bruises+0.061
Neg logit contributions
 Julia-0.067
 dives-0.061
 dive-0.060
La-0.059
 dance-0.059
Token a
Token ID257
Feature activation-6.50e-02

Pos logit contributions
+0.234
leaning+0.217
mons+0.210
pling+0.209
uc+0.208
Neg logit contributions
 Julia-0.140
 Mae-0.117
 Amber-0.114
 April-0.112
 Jeff-0.109
Token screw
Token ID9580
Feature activation+2.97e-01

Pos logit contributions
 Julia+0.108
 mate+0.103
 Mae+0.100
 royal+0.100
 pir+0.099
Neg logit contributions
leaning-0.101
 bruises-0.101
-0.100
mons-0.098
 bruised-0.097
Tokendriver
Token ID26230
Feature activation+2.74e-01

Pos logit contributions
+0.458
 swollen+0.441
leaning+0.412
 bruised+0.405
 bruises+0.400
Neg logit contributions
 Julia-0.532
 mate-0.482
 Mae-0.468
 Zac-0.463
 Lou-0.459
Token,
Token ID11
Feature activation+1.51e-03

Pos logit contributions
+0.665
 swollen+0.633
leaning+0.620
uc+0.611
 bruises+0.611
Neg logit contributions
 Julia-0.237
 dragons-0.193
 Mae-0.192
 April-0.182
 mate-0.180
Token a
Token ID257
Feature activation-6.51e-02

Pos logit contributions
+0.003
uc+0.003
pling+0.003
leaning+0.003
 wrench+0.003
Neg logit contributions
 Julia-0.002
 Mae-0.002
La-0.002
 Amber-0.002
 Lou-0.002
Token wrench
Token ID34561
Feature activation+4.45e-01

Pos logit contributions
 Julia+0.130
 Mae+0.120
 mate+0.118
La+0.118
 pir+0.118
Neg logit contributions
-0.083
leaning-0.083
mons-0.076
 bruises-0.075
 bruised-0.074
Token,
Token ID11
Feature activation+1.39e-03

Pos logit contributions
+1.031
 swollen+1.011
leaning+0.992
 bruises+0.980
 moderation+0.973
Neg logit contributions
 Julia-0.384
 Ro-0.318
 Zac-0.316
 mate-0.309
 Jeff-0.298
Token and
Token ID290
Feature activation+1.51e-02

Pos logit contributions
+0.003
leaning+0.003
uc+0.003
pling+0.003
 swollen+0.003
Neg logit contributions
 Julia-0.001
 Mae-0.001
La-0.001
 April-0.001
 Lou-0.001
Token a
Token ID257
Feature activation-6.50e-02

Pos logit contributions
+0.029
 bruised+0.028
 wrench+0.027
 swollen+0.027
pling+0.027
Neg logit contributions
 Julia-0.022
 Mae-0.020
La-0.020
 Amber-0.019
 Lou-0.018
Token flashlight
Token ID38371
Feature activation+4.76e-02

Pos logit contributions
 Julia+0.138
 mate+0.129
 pirate+0.128
 princess+0.127
 pir+0.126
Neg logit contributions
leaning-0.075
-0.074
mons-0.067
issors-0.063
uc-0.063
Token.
Token ID13
Feature activation+1.07e-02

Pos logit contributions
+0.088
 swollen+0.086
 bruised+0.085
 bruises+0.084
pling+0.083
Neg logit contributions
 Julia-0.069
 Mae-0.060
La-0.060
 Zac-0.058
 Jeff-0.058
Token it
Token ID340
Feature activation+1.95e-02

Pos logit contributions
+0.018
 bruised+0.018
 swollen+0.018
 wrench+0.017
 bruises+0.017
Neg logit contributions
 Julia-0.016
 Mae-0.015
La-0.015
 Amber-0.014
 Jeff-0.014
Token was
Token ID373
Feature activation-4.17e-02

Pos logit contributions
 bruised+0.035
 wrench+0.035
+0.035
 hammer+0.034
 swollen+0.034
Neg logit contributions
 Julia-0.031
La-0.028
 Mae-0.028
 Amber-0.027
 Zac-0.026
Token a
Token ID257
Feature activation-5.36e-02

Pos logit contributions
 Julia+0.050
 Mae+0.044
 Amber+0.042
 mate+0.042
La+0.042
Neg logit contributions
-0.089
leaning-0.083
 bruises-0.082
 bruised-0.082
pling-0.081
Token very
Token ID845
Feature activation-2.74e-02

Pos logit contributions
 Julia+0.088
La+0.081
 actor+0.080
 Mae+0.080
 mate+0.078
Neg logit contributions
-0.086
 bruised-0.084
 bruises-0.084
pling-0.081
 swollen-0.081
Token special
Token ID2041
Feature activation-1.18e-02

Pos logit contributions
 Julia+0.046
La+0.041
 Mae+0.041
 Amber+0.040
 actor+0.039
Neg logit contributions
-0.046
 bruised-0.046
 bruises-0.044
 swollen-0.044
 wrench-0.044
Token hammer
Token ID15554
Feature activation+4.45e-01

Pos logit contributions
 Julia+0.024
 actor+0.023
 mate+0.022
 pirate+0.022
 Lou+0.022
Neg logit contributions
-0.013
 swollen-0.013
 bruised-0.012
 bruises-0.012
arl-0.012
Token.
Token ID13
Feature activation+1.49e-02

Pos logit contributions
+0.846
 swollen+0.818
 bruises+0.816
 bruised+0.812
leaning+0.781
Neg logit contributions
 Julia-0.613
 Mae-0.539
La-0.523
 actor-0.519
 Jeff-0.516
Token Thomas
Token ID5658
Feature activation-1.11e-01

Pos logit contributions
+0.023
 wrench+0.022
 bruised+0.022
 hammer+0.021
 swollen+0.021
Neg logit contributions
 Julia-0.027
La-0.026
 Mae-0.025
 Amber-0.024
 dolphins-0.024
Token noticed
Token ID6810
Feature activation-7.14e-02

Pos logit contributions
 Julia+0.184
La+0.171
 Mae+0.165
 Amber+0.164
 Zac+0.157
Neg logit contributions
-0.184
 bruised-0.184
 wrench-0.183
 hammer-0.180
 swollen-0.179
Token that
Token ID326
Feature activation+1.42e-01

Pos logit contributions
 Julia+0.120
La+0.107
 Mae+0.106
 Amber+0.105
 dolphins+0.102
Neg logit contributions
-0.123
 bruised-0.121
 swollen-0.117
 wrench-0.117
 bruises-0.115
Token the
Token ID262
Feature activation+5.12e-02

Pos logit contributions
+0.278
 bruised+0.268
 swollen+0.263
 bruises+0.262
 wrench+0.259
Neg logit contributions
 Julia-0.209
 Mae-0.185
La-0.182
 Amber-0.180
 Zac-0.173
Token hit
Token ID2277
Feature activation+8.59e-02

Pos logit contributions
+0.096
 bruised+0.092
 wrench+0.090
 swollen+0.090
 bruises+0.087
Neg logit contributions
 Julia-0.132
La-0.123
 Mae-0.121
 Amber-0.117
 Lou-0.115
Token the
Token ID262
Feature activation+5.55e-02

Pos logit contributions
+0.160
 wrench+0.152
 bruised+0.150
 swollen+0.149
pling+0.148
Neg logit contributions
 Julia-0.136
 Mae-0.120
La-0.120
 Amber-0.116
 Jeff-0.116
Token bird
Token ID6512
Feature activation-4.00e-02

Pos logit contributions
+0.096
leaning+0.091
 bruised+0.087
 swollen+0.087
uc+0.087
Neg logit contributions
 Julia-0.084
 dolphins-0.077
 princess-0.077
La-0.076
 Mae-0.076
Token with
Token ID351
Feature activation-2.75e-02

Pos logit contributions
 Julia+0.061
La+0.055
 Mae+0.054
 Amber+0.053
 Lou+0.051
Neg logit contributions
-0.073
 bruised-0.072
 wrench-0.072
 swollen-0.070
 bruises-0.070
Token his
Token ID465
Feature activation-4.43e-02

Pos logit contributions
 Julia+0.026
 Lou+0.023
 Jeff+0.022
 Mae+0.022
 Leo+0.021
Neg logit contributions
-0.067
leaning-0.064
pling-0.062
uc-0.062
mons-0.061
Token hammer
Token ID15554
Feature activation+4.44e-01

Pos logit contributions
 mate+0.083
 Julia+0.081
ia+0.081
 Lou+0.080
 royal+0.080
Neg logit contributions
leaning-0.052
Here-0.046
-0.045
 checkout-0.042
pling-0.040
Token.
Token ID13
Feature activation+1.61e-02

Pos logit contributions
+0.921
leaning+0.888
 bruises+0.865
 wrench+0.859
 bruised+0.855
Neg logit contributions
 Julia-0.522
La-0.457
 Mae-0.441
 Lou-0.435
 Jeff-0.428
Token The
Token ID383
Feature activation+8.90e-02

Pos logit contributions
 bruised+0.026
 swollen+0.026
 wrench+0.026
 bruises+0.025
pling+0.025
Neg logit contributions
 Julia-0.028
 Mae-0.026
La-0.025
 Amber-0.025
 Jeff-0.025
Token bird
Token ID6512
Feature activation-4.06e-02

Pos logit contributions
+0.184
 bruised+0.170
leaning+0.170
 wrench+0.169
issors+0.169
Neg logit contributions
 Julia-0.113
La-0.100
 dolphins-0.098
 princess-0.097
 Mae-0.097
Token could
Token ID714
Feature activation+1.61e-02

Pos logit contributions
 Julia+0.060
La+0.055
 Mae+0.054
 Lou+0.052
 Amber+0.051
Neg logit contributions
-0.076
 wrench-0.074
 bruised-0.072
 bruises-0.070
 swollen-0.070
Token not
Token ID407
Feature activation-1.85e-02

Pos logit contributions
 bruised+0.026
+0.026
 wrench+0.026
pling+0.025
 swollen+0.025
Neg logit contributions
 Julia-0.027
La-0.025
 Mae-0.024
 Amber-0.024
 Jeff-0.023
Token home
Token ID1363
Feature activation+1.36e-02

Pos logit contributions
 Julia+0.140
La+0.124
 Mae+0.123
 Amber+0.123
 Zac+0.118
Neg logit contributions
-0.147
 bruised-0.145
 wrench-0.145
 swollen-0.144
 bruises-0.141
Token,
Token ID11
Feature activation+2.25e-03

Pos logit contributions
+0.023
 bruised+0.023
 wrench+0.023
 swollen+0.022
 hammer+0.022
Neg logit contributions
 Julia-0.023
 Mae-0.020
La-0.020
 Amber-0.020
 Zac-0.019
Token but
Token ID475
Feature activation+7.69e-02

Pos logit contributions
+0.004
 bruised+0.004
 wrench+0.004
 swollen+0.004
 bruises+0.004
Neg logit contributions
 Julia-0.004
La-0.003
 Mae-0.003
 Amber-0.003
 Zac-0.003
Token he
Token ID339
Feature activation-6.12e-02

Pos logit contributions
 bruised+0.130
 wrench+0.128
+0.128
 swollen+0.126
 hammer+0.125
Neg logit contributions
 Julia-0.126
La-0.119
 Mae-0.115
 Amber-0.112
 April-0.107
Token was
Token ID373
Feature activation-4.52e-02

Pos logit contributions
 Julia+0.112
La+0.109
 Mae+0.103
 Amber+0.102
 April+0.098
Neg logit contributions
 bruised-0.092
-0.089
 swollen-0.088
 hammer-0.088
 wrench-0.088
Token bruised
Token ID44379
Feature activation+4.44e-01

Pos logit contributions
 Julia+0.085
La+0.080
 Mae+0.077
 Amber+0.076
 Lou+0.074
Neg logit contributions
 bruised-0.069
 swollen-0.066
-0.065
 wrench-0.065
 bruises-0.064
Token and
Token ID290
Feature activation+1.53e-02

Pos logit contributions
+0.807
 wrench+0.783
 hammer+0.774
 bruises+0.772
 swollen+0.768
Neg logit contributions
 Julia-0.677
 Mae-0.604
La-0.595
 Amber-0.593
 Zac-0.570
Token battered
Token ID33613
Feature activation+2.56e-01

Pos logit contributions
+0.027
leaning+0.026
issors+0.025
 wrench+0.025
mons+0.025
Neg logit contributions
 Julia-0.024
 Mae-0.022
 Jeff-0.021
 Lou-0.021
 Amber-0.021
Token.
Token ID13
Feature activation+1.41e-02

Pos logit contributions
+0.467
 wrench+0.450
 swollen+0.448
 bruised+0.443
 hammer+0.441
Neg logit contributions
 Julia-0.413
 Mae-0.365
 Amber-0.360
La-0.360
 Jeff-0.347
Token His
Token ID2399
Feature activation-2.57e-02

Pos logit contributions
 bruised+0.025
 swollen+0.024
 wrench+0.024
+0.024
 bruises+0.024
Neg logit contributions
 Julia-0.023
La-0.020
 Mae-0.020
 Amber-0.020
 Zac-0.019
Token mummy
Token ID38567
Feature activation-1.93e-02

Pos logit contributions
 Julia+0.043
La+0.039
 Mae+0.038
 Amber+0.038
 dolphins+0.037
Neg logit contributions
-0.042
 bruised-0.042
 wrench-0.041
pling-0.040
 swollen-0.040
Token and
Token ID290
Feature activation+2.37e-02

Pos logit contributions
 Julia+0.067
La+0.062
 Mae+0.060
 Amber+0.060
 Zac+0.057
Neg logit contributions
 bruised-0.059
-0.059
 wrench-0.059
 swollen-0.058
 hammer-0.057
Token kept
Token ID4030
Feature activation+1.02e-01

Pos logit contributions
 bruised+0.039
 wrench+0.038
 hammer+0.038
 swollen+0.038
 bruises+0.038
Neg logit contributions
 Julia-0.041
La-0.039
 Mae-0.037
 Amber-0.036
 April-0.035
Token playing
Token ID2712
Feature activation+4.82e-03

Pos logit contributions
+0.191
 wrench+0.179
 bruised+0.178
leaning+0.177
pling+0.176
Neg logit contributions
 Julia-0.154
La-0.138
 Mae-0.135
 Amber-0.133
 Jeff-0.132
Token with
Token ID351
Feature activation-2.78e-02

Pos logit contributions
+0.009
 bruised+0.009
 swollen+0.008
 wrench+0.008
 bruises+0.008
Neg logit contributions
 Julia-0.007
La-0.007
 Mae-0.007
 Amber-0.006
 Jeff-0.006
Token the
Token ID262
Feature activation+5.81e-02

Pos logit contributions
 Julia+0.049
La+0.046
 Mae+0.044
 Amber+0.044
 Zac+0.042
Neg logit contributions
 bruised-0.044
-0.044
 wrench-0.044
 swollen-0.043
 hammer-0.043
Token hammer
Token ID15554
Feature activation+4.44e-01

Pos logit contributions
+0.103
 bruised+0.095
pling+0.094
 swollen+0.093
 wrench+0.093
Neg logit contributions
 Julia-0.091
 dolphins-0.083
La-0.083
 actor-0.082
 Mae-0.081
Token.
Token ID13
Feature activation+1.78e-02

Pos logit contributions
+0.784
 bruised+0.765
 swollen+0.754
 wrench+0.754
 bruises+0.751
Neg logit contributions
 Julia-0.699
La-0.624
 Mae-0.623
 Amber-0.607
 Zac-0.592
Token Suddenly
Token ID24975
Feature activation-4.84e-02

Pos logit contributions
+0.026
 bruised+0.025
 wrench+0.025
 hammer+0.024
 bruises+0.024
Neg logit contributions
 Julia-0.033
La-0.032
 Mae-0.030
 Amber-0.030
 dolphins-0.030
Token,
Token ID11
Feature activation+7.87e-03

Pos logit contributions
La+0.096
 Julia+0.095
 Mae+0.088
 actor+0.087
 Amber+0.087
Neg logit contributions
 wrench-0.065
 hammer-0.065
 bruised-0.063
 bruises-0.060
 nails-0.060
Token he
Token ID339
Feature activation-5.84e-02

Pos logit contributions
 wrench+0.013
 bruised+0.013
 hammer+0.012
+0.012
 bruises+0.012
Neg logit contributions
 Julia-0.013
La-0.013
 Mae-0.012
 Amber-0.012
 actor-0.011
Token hit
Token ID2277
Feature activation+8.13e-02

Pos logit contributions
 Julia+0.111
 Mae+0.104
La+0.103
 Amber+0.102
 oper+0.101
Neg logit contributions
 hammer-0.086
 swollen-0.084
 bruised-0.084
 wrench-0.083
 bruises-0.082
Token,
Token ID11
Feature activation+7.89e-03

Pos logit contributions
 Julia+0.151
La+0.146
 Mae+0.138
 Amber+0.136
 Zac+0.131
Neg logit contributions
 bruised-0.126
 wrench-0.125
 hammer-0.123
-0.122
 swollen-0.120
Token "
Token ID366
Feature activation+4.98e-02

Pos logit contributions
 bruised+0.011
 wrench+0.011
+0.011
 hammer+0.011
 swollen+0.011
Neg logit contributions
 Julia-0.015
La-0.014
 Mae-0.014
 Amber-0.014
 Zac-0.013
TokenI
Token ID40
Feature activation+6.05e-02

Pos logit contributions
+0.074
 bruised+0.074
 wrench+0.073
 swollen+0.071
 hammer+0.071
Neg logit contributions
 Julia-0.094
La-0.088
 Mae-0.086
 Amber-0.084
 Zac-0.081
Token miss
Token ID2051
Feature activation-4.92e-03

Pos logit contributions
+0.090
 bruised+0.089
 wrench+0.089
 swollen+0.088
 bruises+0.087
Neg logit contributions
 Julia-0.112
La-0.104
 Mae-0.102
 Amber-0.100
 Zac-0.097
Token my
Token ID616
Feature activation-7.85e-02

Pos logit contributions
 Julia+0.007
La+0.006
 Mae+0.006
 Amber+0.006
 Jeff+0.006
Neg logit contributions
-0.009
 bruised-0.009
 wrench-0.009
 swollen-0.009
 bruises-0.009
Token hammer
Token ID15554
Feature activation+4.44e-01

Pos logit contributions
 princess+0.157
 mate+0.149
 pirate+0.147
 king+0.146
 royal+0.145
Neg logit contributions
-0.086
leaning-0.081
arl-0.077
 swollen-0.076
 bruised-0.076
Token.
Token ID13
Feature activation+2.01e-02

Pos logit contributions
 bruised+0.736
+0.728
 wrench+0.726
 swollen+0.712
 hammer+0.711
Neg logit contributions
 Julia-0.754
La-0.696
 Mae-0.681
 Amber-0.671
 Zac-0.643
Token Can
Token ID1680
Feature activation+3.15e-02

Pos logit contributions
 bruised+0.035
+0.034
 wrench+0.034
 hammer+0.034
 swollen+0.033
Neg logit contributions
 Julia-0.033
La-0.030
 Mae-0.029
 Amber-0.029
 Zac-0.028
Token you
Token ID345
Feature activation-2.36e-02

Pos logit contributions
 bruised+0.063
+0.063
 wrench+0.063
 swollen+0.062
 hammer+0.062
Neg logit contributions
 Julia-0.042
La-0.038
 Mae-0.037
 Amber-0.036
 Zac-0.034
Token help
Token ID1037
Feature activation+2.55e-01

Pos logit contributions
 Julia+0.047
La+0.045
 Mae+0.043
 Amber+0.043
 actor+0.042
Neg logit contributions
 bruised-0.031
-0.030
 swollen-0.030
 bruises-0.029
 wrench-0.029
Token me
Token ID502
Feature activation-1.17e-01

Pos logit contributions
 hammer+0.474
 wrench+0.469
 bruised+0.465
 swollen+0.455
 bruises+0.451
Neg logit contributions
 Julia-0.384
La-0.349
 Mae-0.340
 Amber-0.335
 Zac-0.325
Token car
Token ID1097
Feature activation+6.43e-03

Pos logit contributions
+0.085
uc+0.081
pling+0.080
 swollen+0.079
 bruised+0.077
Neg logit contributions
 Julia-0.064
La-0.060
 dolphins-0.058
 pirate-0.056
 actor-0.056
Token.
Token ID13
Feature activation+8.41e-03

Pos logit contributions
 bruised+0.010
 wrench+0.010
+0.010
 hammer+0.009
 bruises+0.009
Neg logit contributions
 Julia-0.012
La-0.011
 Mae-0.011
 Amber-0.011
 Zac-0.010
Token He
Token ID679
Feature activation-2.52e-02

Pos logit contributions
 bruised+0.016
 swollen+0.015
 wrench+0.015
+0.015
pling+0.015
Neg logit contributions
 Julia-0.013
 Mae-0.011
La-0.011
 Amber-0.011
 Jeff-0.011
Token had
Token ID550
Feature activation+4.29e-03

Pos logit contributions
 Julia+0.045
La+0.042
 Mae+0.041
 Amber+0.041
 Lou+0.039
Neg logit contributions
 bruised-0.040
 wrench-0.038
 bruises-0.038
 swollen-0.038
 hammer-0.038
Token a
Token ID257
Feature activation-6.77e-02

Pos logit contributions
 bruised+0.007
 wrench+0.007
 bruises+0.007
 swollen+0.007
 hammer+0.007
Neg logit contributions
 Julia-0.007
La-0.007
 Mae-0.006
 Amber-0.006
 Zac-0.006
Token wrench
Token ID34561
Feature activation+4.44e-01

Pos logit contributions
 Julia+0.106
La+0.101
 mate+0.096
 Mae+0.093
 Amber+0.093
Neg logit contributions
-0.114
pling-0.111
leaning-0.109
 bruised-0.109
uc-0.108
Token and
Token ID290
Feature activation+1.26e-02

Pos logit contributions
+0.860
 swollen+0.820
 bruised+0.804
leaning+0.798
 bruises+0.797
Neg logit contributions
 Julia-0.599
La-0.538
 Zac-0.529
 Ro-0.518
 Jeff-0.513
Token a
Token ID257
Feature activation-6.77e-02

Pos logit contributions
+0.024
 bruised+0.024
 wrench+0.023
 swollen+0.023
 hammer+0.023
Neg logit contributions
 Julia-0.018
La-0.017
 Mae-0.016
 Amber-0.016
 Zac-0.015
Token jack
Token ID14509
Feature activation-8.62e-03

Pos logit contributions
 Julia+0.120
 mate+0.120
La+0.117
 pirate+0.117
 princess+0.111
Neg logit contributions
-0.095
leaning-0.094
pling-0.088
mons-0.087
uc-0.086
Token.
Token ID13
Feature activation+9.20e-03

Pos logit contributions
 Julia+0.015
La+0.014
 Mae+0.013
 Amber+0.013
 Zac+0.013
Neg logit contributions
 bruised-0.014
 wrench-0.014
-0.014
 hammer-0.014
 swollen-0.014
Token He
Token ID679
Feature activation-2.54e-02

Pos logit contributions
 swollen+0.019
 bruised+0.019
 bruises+0.018
 wrench+0.018
pling+0.018
Neg logit contributions
 Julia-0.013
 Mae-0.011
 Jeff-0.011
 Amber-0.010
 Lou-0.010
Token,
Token ID11
Feature activation+7.88e-04

Pos logit contributions
 bruised+0.243
 swollen+0.238
 wrench+0.237
+0.237
 bruises+0.233
Neg logit contributions
 Julia-0.219
 Mae-0.192
La-0.191
 Amber-0.190
 Lou-0.183
Token she
Token ID673
Feature activation-4.40e-02

Pos logit contributions
 bruised+0.002
 swollen+0.002
 wrench+0.002
 mercury+0.001
+0.001
Neg logit contributions
 Julia-0.001
 Amber-0.001
 Mae-0.001
 Lou-0.001
La-0.001
Token found
Token ID1043
Feature activation-4.77e-02

Pos logit contributions
 Julia+0.078
La+0.075
 Mae+0.072
 Amber+0.071
 Zac+0.069
Neg logit contributions
 hammer-0.069
 wrench-0.068
-0.068
 bruised-0.068
 swollen-0.066
Token a
Token ID257
Feature activation-6.70e-02

Pos logit contributions
 Julia+0.078
 Mae+0.069
 Amber+0.068
La+0.067
 Lou+0.066
Neg logit contributions
-0.086
 bruised-0.084
 wrench-0.084
 swollen-0.082
 bruises-0.081
Token shiny
Token ID22441
Feature activation+6.23e-02

Pos logit contributions
 Julia+0.112
La+0.103
 Mae+0.101
 Amber+0.099
 April+0.097
Neg logit contributions
-0.110
 bruised-0.107
 bruises-0.106
leaning-0.103
 swollen-0.101
Token wrench
Token ID34561
Feature activation+4.44e-01

Pos logit contributions
+0.101
 bruises+0.100
 supervision+0.098
 bruised+0.095
uc+0.094
Neg logit contributions
 Julia-0.101
 princess-0.095
 actor-0.093
 mate-0.093
 pirate-0.092
Token hidden
Token ID7104
Feature activation-1.71e-02

Pos logit contributions
+0.808
 swollen+0.803
 bruises+0.798
 bruised+0.795
leaning+0.771
Neg logit contributions
 Julia-0.663
 Mae-0.584
 Amber-0.566
 Jeff-0.561
La-0.558
Token among
Token ID1871
Feature activation-6.57e-02

Pos logit contributions
 Julia+0.025
 pirate+0.022
 mate+0.022
 actor+0.022
 princess+0.021
Neg logit contributions
-0.035
 swollen-0.031
uc-0.030
 bruised-0.029
leaning-0.029
Token the
Token ID262
Feature activation+5.56e-02

Pos logit contributions
 Julia+0.116
La+0.104
 Mae+0.103
 Amber+0.103
 Jeff+0.099
Neg logit contributions
-0.109
 bruised-0.108
 wrench-0.107
 swollen-0.103
 hammer-0.103
Token plants
Token ID6134
Feature activation-2.29e-02

Pos logit contributions
+0.098
 bruised+0.096
 wrench+0.095
 swollen+0.093
 bruises+0.092
Neg logit contributions
 Julia-0.089
La-0.082
 Mae-0.080
 Amber-0.079
 April-0.076
Token.
Token ID13
Feature activation+7.27e-03

Pos logit contributions
 Julia+0.035
 Mae+0.031
 Amber+0.031
 Jeff+0.030
La+0.029
Neg logit contributions
 bruised-0.041
-0.041
 wrench-0.041
 swollen-0.041
 bruises-0.040
Token
Token ID198
Feature activation-2.31e-03

Pos logit contributions
 bruised+0.032
+0.032
 wrench+0.031
 swollen+0.031
 hammer+0.031
Neg logit contributions
 Julia-0.031
La-0.029
 Mae-0.028
 Amber-0.028
 Zac-0.026
Token
Token ID198
Feature activation-2.12e-03

Pos logit contributions
 Julia+0.003
La+0.003
 Mae+0.003
 Amber+0.003
 Lou+0.003
Neg logit contributions
 bruised-0.004
 wrench-0.004
-0.004
 hammer-0.004
 swollen-0.004
TokenShe
Token ID3347
Feature activation+2.14e-02

Pos logit contributions
 Julia+0.003
La+0.003
 Amber+0.003
 Mae+0.003
ia+0.003
Neg logit contributions
-0.004
 bruised-0.004
 wrench-0.004
 hammer-0.004
pling-0.004
Token twists
Token ID30953
Feature activation+3.23e-02

Pos logit contributions
 bruised+0.032
 wrench+0.032
 bruises+0.032
+0.031
 swollen+0.031
Neg logit contributions
 Julia-0.039
La-0.036
 Mae-0.036
 Amber-0.035
 Lou-0.034
Token the
Token ID262
Feature activation+5.01e-02

Pos logit contributions
+0.072
pling+0.068
uc+0.068
 wrench+0.066
 supervision+0.065
Neg logit contributions
 Julia-0.041
La-0.035
 Amber-0.033
 Jeff-0.033
 Mae-0.033
Token wrench
Token ID34561
Feature activation+4.44e-01

Pos logit contributions
 bruised+0.096
+0.096
 swollen+0.094
 bruises+0.092
 supervision+0.089
Neg logit contributions
 dives-0.068
 Julia-0.067
 dolphins-0.065
 dances-0.064
La-0.062
Token.
Token ID13
Feature activation+1.77e-02

Pos logit contributions
+0.815
 bruised+0.800
 swollen+0.792
 bruises+0.786
 wrench+0.785
Neg logit contributions
 Julia-0.686
La-0.615
 Mae-0.605
 Amber-0.587
 Zac-0.575
Token She
Token ID1375
Feature activation+5.40e-03

Pos logit contributions
 bruised+0.032
 wrench+0.032
pling+0.032
 swollen+0.032
 hammer+0.031
Neg logit contributions
 Julia-0.029
La-0.025
 Mae-0.025
 Amber-0.025
 April-0.023
Token turns
Token ID4962
Feature activation+1.96e-02

Pos logit contributions
 bruised+0.009
 bruises+0.009
 wrench+0.009
 swollen+0.008
 hammer+0.008
Neg logit contributions
 Julia-0.009
 Mae-0.009
 Amber-0.009
La-0.009
 Jeff-0.008
Token the
Token ID262
Feature activation+4.83e-02

Pos logit contributions
+0.036
 bruised+0.035
 swollen+0.035
 wrench+0.034
pling+0.034
Neg logit contributions
 Julia-0.031
La-0.028
 Mae-0.027
 Amber-0.026
 Zac-0.026
Token tap
Token ID9814
Feature activation+5.45e-02

Pos logit contributions
+0.089
 bruised+0.086
 swollen+0.085
pling+0.084
 bruises+0.081
Neg logit contributions
 Julia-0.069
 dolphins-0.068
 dives-0.064
 diver-0.064
 dances-0.063
Token
Token ID198
Feature activation-5.34e-03

Pos logit contributions
 actor+0.012
 mate+0.012
 diver+0.012
 Mae+0.012
 Julia+0.012
Neg logit contributions
-0.005
 hammer-0.004
 rag-0.004
 bruises-0.004
pling-0.004
TokenFrom
Token ID4863
Feature activation+4.18e-03

Pos logit contributions
 mate+0.011
 dolphins+0.010
 dives+0.010
 Dragon+0.010
 pirate+0.010
Neg logit contributions
leaning-0.005
issors-0.005
Here-0.004
ised-0.004
-0.004
Token then
Token ID788
Feature activation-1.95e-03

Pos logit contributions
+0.006
 bruised+0.006
 wrench+0.006
 swollen+0.005
 bruises+0.005
Neg logit contributions
 Julia-0.008
La-0.008
 Mae-0.008
 Amber-0.008
 Zac-0.007
Token on
Token ID319
Feature activation+5.67e-02

Pos logit contributions
 Julia+0.003
La+0.002
 Mae+0.002
 Ro+0.002
 dances+0.002
Neg logit contributions
 bruised-0.004
 swollen-0.004
 supervision-0.004
 bruises-0.004
 wrench-0.004
Token,
Token ID11
Feature activation-4.15e-03

Pos logit contributions
+0.070
 hammer+0.066
 bruised+0.065
 swollen+0.064
 wrench+0.064
Neg logit contributions
La-0.111
 Julia-0.107
 actor-0.107
 mate-0.104
 dolphins-0.103
Token Julia
Token ID22300
Feature activation-4.80e-01

Pos logit contributions
 mate+0.005
 camel+0.005
 awoke+0.005
land+0.005
ull+0.005
Neg logit contributions
-0.006
 hammer-0.006
 file-0.006
 don-0.006
free-0.006
Token would
Token ID561
Feature activation-2.73e-02

Pos logit contributions
 Julia+0.859
La+0.780
 Mae+0.769
 Amber+0.763
 Zac+0.760
Neg logit contributions
 bruised-0.734
 hammer-0.733
-0.727
 wrench-0.721
 swollen-0.707
Token take
Token ID1011
Feature activation+2.71e-02

Pos logit contributions
 Julia+0.044
La+0.041
 Mae+0.040
 Amber+0.039
 Zac+0.039
Neg logit contributions
 hammer-0.046
-0.046
 wrench-0.046
 bruised-0.046
 swollen-0.044
Token her
Token ID607
Feature activation-9.22e-03

Pos logit contributions
 hammer+0.036
 bruised+0.036
 wrench+0.036
+0.035
 bruises+0.035
Neg logit contributions
 Julia-0.051
 Mae-0.048
La-0.047
 Zac-0.046
 Amber-0.046
Token orange
Token ID10912
Feature activation+1.21e-02

Pos logit contributions
 Julia+0.015
 April+0.014
 royal+0.013
 Lou+0.013
La+0.013
Neg logit contributions
-0.016
 bruised-0.015
 bruises-0.015
 swollen-0.015
 mercury-0.015
Token telephone
Token ID11426
Feature activation+1.31e-01

Pos logit contributions
 bruised+0.020
 bruises+0.019
 swollen+0.019
 mercury+0.019
 supervision+0.019
Neg logit contributions
 Julia-0.019
 Lou-0.017
 April-0.017
La-0.017
 Boo-0.017
Token telephone
Token ID11426
Feature activation+1.30e-01

Pos logit contributions
 Julia+0.032
 Lou+0.027
La+0.027
 April+0.027
 Mae+0.027
Neg logit contributions
-0.038
 bruised-0.037
 bruises-0.036
 swollen-0.036
 wrench-0.035
Token as
Token ID355
Feature activation+2.60e-02

Pos logit contributions
 bruised+0.196
 wrench+0.193
 hammer+0.191
+0.191
 swollen+0.187
Neg logit contributions
 Julia-0.238
La-0.226
 Mae-0.220
 Amber-0.216
 Zac-0.210
Token a
Token ID257
Feature activation-6.85e-02

Pos logit contributions
 hammer+0.033
 bruised+0.032
 sugar+0.031
 nails+0.030
 wrench+0.030
Neg logit contributions
La-0.048
 Julia-0.047
 Mae-0.046
 Zac-0.045
 dives-0.044
Token gift
Token ID6979
Feature activation+1.07e-01

Pos logit contributions
 Julia+0.112
La+0.102
 Mae+0.100
 Amber+0.099
 April+0.095
Neg logit contributions
-0.117
 bruised-0.116
 wrench-0.113
 swollen-0.113
 bruises-0.112
Token.
Token ID13
Feature activation+5.13e-03

Pos logit contributions
 bruised+0.148
 wrench+0.147
 hammer+0.146
+0.143
 rag+0.139
Neg logit contributions
 Julia-0.206
La-0.203
 Mae-0.194
 Amber-0.190
 Zac-0.186
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
+0.003
leaning+0.003
Here+0.003
 +0.002
 hammer+0.002
Neg logit contributions
 actor-0.012
 diver-0.011
 mate-0.011
 dolphins-0.011
 royal-0.011
Token gigg
Token ID30442
Feature activation-1.14e-01

Pos logit contributions
 Julia+0.885
La+0.865
 Mae+0.822
 Zac+0.816
 actor+0.812
Neg logit contributions
 hammer-0.666
 swollen-0.629
 bruised-0.626
-0.620
 bruises-0.611
Tokenled
Token ID992
Feature activation+2.47e-02

Pos logit contributions
 Julia+0.209
La+0.194
 Mae+0.191
 Amber+0.187
 Zac+0.181
Neg logit contributions
 bruised-0.173
-0.172
 wrench-0.170
 swollen-0.168
 hammer-0.166
Token with
Token ID351
Feature activation-3.82e-02

Pos logit contributions
 bruised+0.028
 hammer+0.026
 swollen+0.026
 wrench+0.026
+0.025
Neg logit contributions
 Julia-0.052
La-0.052
 Mae-0.049
 Zac-0.049
ae-0.048
Token joy
Token ID8716
Feature activation-2.50e-02

Pos logit contributions
 Julia+0.062
La+0.057
 Mae+0.056
 Amber+0.055
 Zac+0.052
Neg logit contributions
 bruised-0.066
-0.066
 wrench-0.066
 swollen-0.064
 hammer-0.064
Token and
Token ID290
Feature activation+1.11e-02

Pos logit contributions
 Julia+0.040
 Mae+0.036
 Amber+0.036
La+0.036
 April+0.034
Neg logit contributions
-0.044
 wrench-0.044
 bruised-0.044
 bruises-0.043
 swollen-0.043
Tokens
Token ID82
Feature activation-2.47e-02

Pos logit contributions
 Julia+0.216
La+0.199
 Mae+0.196
 Amber+0.193
 Zac+0.186
Neg logit contributions
 bruised-0.179
-0.178
 wrench-0.177
 swollen-0.174
 bruises-0.173
Token that
Token ID326
Feature activation+1.34e-01

Pos logit contributions
 Julia+0.050
La+0.047
 Mae+0.046
 Amber+0.045
 April+0.044
Neg logit contributions
 bruised-0.033
-0.033
 wrench-0.033
 swollen-0.032
 bruises-0.032
Token?
Token ID30
Feature activation+2.21e-03

Pos logit contributions
 hammer+0.222
 bruised+0.221
 wrench+0.220
 swollen+0.214
+0.212
Neg logit contributions
 Julia-0.231
La-0.215
 Mae-0.209
 Amber-0.205
 Zac-0.197
Tokenâ
Token ID22940
Feature activation+3.27e-02

Pos logit contributions
+0.003
 bruised+0.003
 wrench+0.003
 hammer+0.003
 swollen+0.003
Neg logit contributions
 Julia-0.004
La-0.004
 Mae-0.004
 Amber-0.004
 dolphins-0.004
Token
Token ID26391
Feature activation-1.41e-01

Pos logit contributions
 bruised+0.072
 wrench+0.071
 swollen+0.070
 bruises+0.070
+0.070
Neg logit contributions
 Julia-0.039
La-0.035
 Mae-0.033
 Amber-0.033
 Lou-0.030
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
 Julia+0.251
La+0.231
 Mae+0.229
 Amber+0.225
 Zac+0.221
Neg logit contributions
-0.221
 bruised-0.211
 wrench-0.209
 hammer-0.206
 swollen-0.205
Token asked
Token ID1965
Feature activation-3.17e-02

Pos logit contributions
 Julia+0.899
La+0.843
 Mae+0.823
 Amber+0.815
 Zac+0.794
Neg logit contributions
 swollen-0.688
 bruised-0.688
 hammer-0.681
-0.678
 wrench-0.663
Token her
Token ID607
Feature activation-9.28e-03

Pos logit contributions
 Julia+0.056
La+0.052
 Mae+0.051
 Amber+0.050
 Zac+0.048
Neg logit contributions
 bruised-0.050
 wrench-0.049
-0.049
 swollen-0.048
 hammer-0.048
Token cat
Token ID3797
Feature activation+5.55e-02

Pos logit contributions
 Julia+0.017
La+0.017
 Mae+0.016
 Amber+0.016
 Zac+0.015
Neg logit contributions
 hammer-0.014
 wrench-0.013
 bruised-0.013
pling-0.013
 swollen-0.012
Token curiously
Token ID43211
Feature activation-6.39e-02

Pos logit contributions
 hammer+0.075
 bruised+0.074
 wrench+0.071
 swollen+0.067
pling+0.066
Neg logit contributions
La-0.108
 Julia-0.107
 Mae-0.103
 actor-0.099
 Amber-0.099
Token.
Token ID13
Feature activation+7.23e-03

Pos logit contributions
 Julia+0.118
La+0.113
 Mae+0.108
 Amber+0.106
 Zac+0.104
Neg logit contributions
 bruised-0.094
 wrench-0.091
 hammer-0.090
-0.089
 swollen-0.088
Token waved
Token ID26834
Feature activation-1.97e-01

Pos logit contributions
 bruised+0.017
 bruises+0.017
 swollen+0.016
 rag+0.015
 scrape+0.015
Neg logit contributions
La-0.022
 Julia-0.022
 mate-0.021
 actor-0.021
adies-0.021
Token as
Token ID355
Feature activation+2.60e-02

Pos logit contributions
 Julia+0.318
La+0.301
 Mae+0.292
 Amber+0.279
 actor+0.278
Neg logit contributions
 bruised-0.336
 swollen-0.320
 wrench-0.320
-0.314
 bruises-0.311
Token they
Token ID484
Feature activation-9.07e-02

Pos logit contributions
 bruised+0.036
 hammer+0.035
 bruises+0.034
 wrench+0.034
+0.033
Neg logit contributions
La-0.044
 mate-0.044
 Julia-0.042
 actor-0.041
 echo-0.041
Token left
Token ID1364
Feature activation+3.19e-02

Pos logit contributions
 Julia+0.143
La+0.131
 Mae+0.128
 Amber+0.126
 Zac+0.120
Neg logit contributions
 bruised-0.161
-0.160
 wrench-0.159
 swollen-0.156
 hammer-0.155
Token.
Token ID13
Feature activation+9.79e-03

Pos logit contributions
 bruised+0.054
 wrench+0.053
+0.053
 swollen+0.053
 hammer+0.052
Neg logit contributions
 Julia-0.054
 Mae-0.049
La-0.048
 Amber-0.048
 Zac-0.046
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
+0.014
 bruises+0.011
 bruised+0.011
 hammer+0.010
 wrench+0.010
Neg logit contributions
 mate-0.018
 dolphins-0.018
 actor-0.018
 diver-0.017
 aliens-0.017
Token felt
Token ID2936
Feature activation-3.19e-02

Pos logit contributions
 Julia+0.887
La+0.838
 mate+0.823
 Mae+0.815
adies+0.815
Neg logit contributions
 bruised-0.625
 swollen-0.621
-0.610
 bruises-0.604
pling-0.586
Token happy
Token ID3772
Feature activation-7.34e-02

Pos logit contributions
 Julia+0.061
La+0.056
 Mae+0.056
 Lou+0.055
 Jeff+0.053
Neg logit contributions
 bruised-0.048
 swollen-0.045
 bruises-0.043
-0.042
free-0.040
Token to
Token ID284
Feature activation+1.81e-03

Pos logit contributions
 Julia+0.143
La+0.139
 Mae+0.133
 Amber+0.130
 Zac+0.128
Neg logit contributions
 bruised-0.103
-0.100
 bruises-0.097
 swollen-0.097
 wrench-0.097
Token have
Token ID423
Feature activation+8.67e-04

Pos logit contributions
+0.003
 hammer+0.003
 bruised+0.002
 wrench+0.002
 swollen+0.002
Neg logit contributions
 Julia-0.003
 Mae-0.003
 Amber-0.003
 dolphins-0.003
 Zac-0.003
Token borrowed
Token ID22546
Feature activation+1.43e-01

Pos logit contributions
 bruises+0.001
+0.001
 bruised+0.001
 swollen+0.001
pling+0.001
Neg logit contributions
 Julia-0.002
 Zac-0.002
 Mae-0.002
 actor-0.002
La-0.002
Token little
Token ID1310
Feature activation+1.86e-02

Pos logit contributions
 Julia+0.130
 Mae+0.118
La+0.118
 Amber+0.115
 Zac+0.115
Neg logit contributions
 hammer-0.095
 wrench-0.094
 bruised-0.093
-0.092
 swollen-0.091
Token girl
Token ID2576
Feature activation-4.97e-03

Pos logit contributions
 bruised+0.030
+0.030
 wrench+0.030
 swollen+0.029
 hammer+0.029
Neg logit contributions
 Julia-0.032
La-0.030
 Mae-0.029
 Amber-0.029
 Zac-0.028
Token called
Token ID1444
Feature activation-6.02e-02

Pos logit contributions
 Julia+0.009
La+0.008
 Mae+0.008
 Amber+0.008
 Zac+0.007
Neg logit contributions
 bruised-0.008
-0.008
 wrench-0.008
 swollen-0.008
 hammer-0.008
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
La+0.097
 actor+0.095
 Julia+0.093
 dolphins+0.092
 Zac+0.090
Neg logit contributions
-0.096
 bruised-0.092
 hammer-0.091
 rag-0.088
 bruises-0.088
Token.
Token ID13
Feature activation+4.34e-03

Pos logit contributions
 Julia+1.061
 Zac+1.026
La+1.000
 Jeff+0.991
 dives+0.990
Neg logit contributions
 hammer-0.460
 bruised-0.453
 rag-0.448
pling-0.435
 therm-0.423
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
+0.003
leaning+0.003
issors+0.003
 file+0.003
 hammer+0.003
Neg logit contributions
 dolphins-0.009
 diver-0.009
 mate-0.009
 actor-0.009
orse-0.009
Token was
Token ID373
Feature activation-4.69e-02

Pos logit contributions
 Julia+0.858
La+0.789
 Mae+0.776
 Amber+0.766
 Zac+0.757
Neg logit contributions
 bruised-0.740
-0.736
 hammer-0.729
 swollen-0.719
 wrench-0.717
Token a
Token ID257
Feature activation-6.68e-02

Pos logit contributions
 Julia+0.077
La+0.071
 Mae+0.069
 Amber+0.068
 Zac+0.066
Neg logit contributions
 bruised-0.082
-0.080
 wrench-0.079
 hammer-0.079
 swollen-0.078
Token happy
Token ID3772
Feature activation-7.07e-02

Pos logit contributions
 Julia+0.115
La+0.105
 Mae+0.103
 Amber+0.102
 Zac+0.099
Neg logit contributions
 bruised-0.109
-0.109
 wrench-0.108
 hammer-0.106
 swollen-0.106
Token child
Token ID1200
Feature activation+2.02e-02

Pos logit contributions
 Julia+0.140
La+0.135
 Zac+0.130
 Mae+0.127
 Jeff+0.125
Neg logit contributions
 hammer-0.095
 bruised-0.094
 rag-0.094
 wrench-0.093
-0.093
Token and
Token ID290
Feature activation+1.26e-02

Pos logit contributions
 bruised+0.032
+0.032
 wrench+0.032
 swollen+0.031
 hammer+0.031
Neg logit contributions
 Julia-0.035
La-0.033
 Mae-0.032
 Amber-0.031
 Zac-0.030
Token outside
Token ID2354
Feature activation-6.18e-02

Pos logit contributions
 Julia+0.046
 Mae+0.043
La+0.041
 Amber+0.041
 Zac+0.041
Neg logit contributions
-0.036
 hammer-0.035
 wrench-0.034
 bruised-0.033
 rag-0.033
Token.
Token ID13
Feature activation+4.03e-03

Pos logit contributions
 Julia+0.122
La+0.116
 Amber+0.112
 Mae+0.112
 Zac+0.110
Neg logit contributions
-0.085
 bruised-0.083
 hammer-0.082
 wrench-0.079
 swollen-0.077
Token One
Token ID1881
Feature activation-3.23e-02

Pos logit contributions
+0.004
leaning+0.003
 rag+0.003
 hammer+0.003
issors+0.003
Neg logit contributions
 diver-0.009
 dolphins-0.009
 actor-0.008
 divers-0.008
 mate-0.008
Token day
Token ID1110
Feature activation+7.45e-03

Pos logit contributions
 Julia+0.065
La+0.060
 Amber+0.059
 Mae+0.059
 Zac+0.058
Neg logit contributions
 bruised-0.043
 hammer-0.043
 rag-0.042
-0.042
 wrench-0.041
Token,
Token ID11
Feature activation-2.24e-03

Pos logit contributions
+0.005
 hammer+0.005
uc+0.005
 wrench+0.004
mons+0.004
Neg logit contributions
La-0.017
nes-0.017
 actor-0.017
 aliens-0.017
 echo-0.017
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
 mate+0.004
rina+0.004
La+0.004
 aliens+0.004
 awoke+0.004
Neg logit contributions
 hammer-0.002
-0.002
 lipstick-0.002
mons-0.002
 don-0.002
Token found
Token ID1043
Feature activation-5.19e-02

Pos logit contributions
 Julia+0.943
 Zac+0.890
La+0.877
 Jeff+0.857
 Mae+0.841
Neg logit contributions
 hammer-0.660
 rag-0.638
 bruised-0.621
 oil-0.618
pling-0.613
Token an
Token ID281
Feature activation-1.30e-03

Pos logit contributions
 Julia+0.074
La+0.066
 Mae+0.065
 Amber+0.064
 Lou+0.061
Neg logit contributions
-0.101
 bruised-0.100
 wrench-0.100
 swollen-0.098
 hammer-0.097
Token orange
Token ID10912
Feature activation+1.39e-02

Pos logit contributions
 Julia+0.002
La+0.002
 Mae+0.002
 Amber+0.002
 Zac+0.002
Neg logit contributions
 bruised-0.002
-0.002
 wrench-0.002
 swollen-0.002
 hammer-0.002
Token telephone
Token ID11426
Feature activation+1.31e-01

Pos logit contributions
 bruised+0.022
 supervision+0.022
 bruises+0.021
 swollen+0.021
pling+0.020
Neg logit contributions
 Julia-0.023
 Mae-0.021
 Amber-0.021
 Jeff-0.021
La-0.020
Token in
Token ID287
Feature activation+7.37e-03

Pos logit contributions
 swollen+0.230
 bruised+0.229
+0.227
 wrench+0.226
 bruises+0.225
Neg logit contributions
 Julia-0.201
La-0.181
 Mae-0.177
 Amber-0.173
 April-0.171
Token lived
Token ID5615
Feature activation-4.21e-02

Pos logit contributions
+0.070
 bruised+0.069
 hammer+0.068
 wrench+0.067
 swollen+0.066
Neg logit contributions
 Julia-0.069
La-0.066
 Amber-0.063
 Mae-0.063
 April-0.060
Token a
Token ID257
Feature activation-6.66e-02

Pos logit contributions
 Julia+0.069
La+0.062
 Mae+0.062
 Amber+0.060
 Zac+0.059
Neg logit contributions
-0.072
 bruised-0.072
 hammer-0.070
 bruises-0.070
 wrench-0.070
Token little
Token ID1310
Feature activation+1.86e-02

Pos logit contributions
 Julia+0.130
 Mae+0.118
La+0.118
 Amber+0.115
 Zac+0.115
Neg logit contributions
 hammer-0.095
 wrench-0.094
 bruised-0.093
-0.092
 swollen-0.091
Token girl
Token ID2576
Feature activation-4.97e-03

Pos logit contributions
 bruised+0.030
+0.030
 wrench+0.030
 swollen+0.029
 hammer+0.029
Neg logit contributions
 Julia-0.032
La-0.030
 Mae-0.029
 Amber-0.029
 Zac-0.028
Token called
Token ID1444
Feature activation-6.02e-02

Pos logit contributions
 Julia+0.009
La+0.008
 Mae+0.008
 Amber+0.008
 Zac+0.007
Neg logit contributions
 bruised-0.008
-0.008
 wrench-0.008
 swollen-0.008
 hammer-0.008
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
La+0.097
 actor+0.095
 Julia+0.093
 dolphins+0.092
 Zac+0.090
Neg logit contributions
-0.096
 bruised-0.092
 hammer-0.091
 rag-0.088
 bruises-0.088
Token.
Token ID13
Feature activation+4.34e-03

Pos logit contributions
 Julia+1.061
 Zac+1.026
La+1.000
 Jeff+0.991
 dives+0.990
Neg logit contributions
 hammer-0.460
 bruised-0.453
 rag-0.448
pling-0.435
 therm-0.423
Token Julia
Token ID22300
Feature activation-4.79e-01

Pos logit contributions
+0.003
leaning+0.003
issors+0.003
 file+0.003
 hammer+0.003
Neg logit contributions
 dolphins-0.009
 diver-0.009
 mate-0.009
 actor-0.009
orse-0.009
Token was
Token ID373
Feature activation-4.69e-02

Pos logit contributions
 Julia+0.858
La+0.789
 Mae+0.776
 Amber+0.766
 Zac+0.757
Neg logit contributions
 bruised-0.740
-0.736
 hammer-0.729
 swollen-0.719
 wrench-0.717
Token a
Token ID257
Feature activation-6.68e-02

Pos logit contributions
 Julia+0.077
La+0.071
 Mae+0.069
 Amber+0.068
 Zac+0.066
Neg logit contributions
 bruised-0.082
-0.080
 wrench-0.079
 hammer-0.079
 swollen-0.078
Token happy
Token ID3772
Feature activation-7.07e-02

Pos logit contributions
 Julia+0.115
La+0.105
 Mae+0.103
 Amber+0.102
 Zac+0.099
Neg logit contributions
 bruised-0.109
-0.109
 wrench-0.108
 hammer-0.106
 swollen-0.106
Token was
Token ID373
Feature activation-4.92e-02

Pos logit contributions
+0.065
 bruised+0.065
 wrench+0.063
 hammer+0.063
 swollen+0.063
Neg logit contributions
 Julia-0.061
La-0.058
 Mae-0.055
 Amber-0.055
 April-0.052
Token a
Token ID257
Feature activation-7.07e-02

Pos logit contributions
 Julia+0.090
La+0.082
 Mae+0.081
 Amber+0.079
 Zac+0.079
Neg logit contributions
-0.074
 bruised-0.073
 bruises-0.073
 hammer-0.072
 rag-0.071
Token little
Token ID1310
Feature activation+1.47e-02

Pos logit contributions
 Julia+0.128
La+0.117
 Mae+0.116
 Amber+0.114
 Zac+0.111
Neg logit contributions
-0.109
 bruised-0.108
 wrench-0.108
 hammer-0.106
 swollen-0.105
Token girl
Token ID2576
Feature activation-8.30e-03

Pos logit contributions
+0.025
 bruised+0.025
 bruises+0.025
 wrench+0.025
 swollen+0.025
Neg logit contributions
 Julia-0.023
La-0.021
 Mae-0.021
 Amber-0.020
 dolphins-0.020
Token named
Token ID3706
Feature activation+5.93e-03

Pos logit contributions
 Julia+0.012
La+0.011
 Mae+0.011
 Amber+0.010
 April+0.010
Neg logit contributions
 wrench-0.016
-0.016
 bruised-0.015
 swollen-0.015
pling-0.015
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 wrench+0.011
 bruised+0.011
+0.011
 swollen+0.011
 bruises+0.011
Neg logit contributions
 Julia-0.009
La-0.008
 Mae-0.008
 Amber-0.008
 April-0.007
Token.
Token ID13
Feature activation+4.76e-03

Pos logit contributions
 Julia+0.931
La+0.857
 Mae+0.856
 Zac+0.849
 Amber+0.840
Neg logit contributions
 bruised-0.661
 hammer-0.639
-0.637
 swollen-0.627
 wrench-0.619
Token She
Token ID1375
Feature activation+9.26e-03

Pos logit contributions
+0.004
leaning+0.003
 hammer+0.003
issors+0.003
 file+0.003
Neg logit contributions
 actor-0.010
 mate-0.010
 dolphins-0.010
 oper-0.010
 diver-0.010
Token was
Token ID373
Feature activation-4.84e-02

Pos logit contributions
 bruised+0.012
+0.012
 hammer+0.012
 swollen+0.011
 bruises+0.011
Neg logit contributions
 Julia-0.019
 Mae-0.018
 Amber-0.018
 Zac-0.017
La-0.017
Token only
Token ID691
Feature activation+5.03e-02

Pos logit contributions
 Julia+0.078
La+0.071
 Mae+0.070
 Amber+0.068
 Zac+0.066
Neg logit contributions
 bruised-0.085
-0.084
 wrench-0.084
 swollen-0.082
 hammer-0.082
Token three
Token ID1115
Feature activation-9.59e-02

Pos logit contributions
 bruised+0.091
 wrench+0.088
 hammer+0.088
+0.087
 bruises+0.086
Neg logit contributions
 Julia-0.077
La-0.072
 Mae-0.069
 Amber-0.068
 Zac-0.066
Token ne
Token ID497
Feature activation-6.41e-02

Pos logit contributions
 Julia+0.067
La+0.061
 Mae+0.060
 Amber+0.058
 Zac+0.056
Neg logit contributions
 bruised-0.080
-0.079
 wrench-0.079
 swollen-0.078
 hammer-0.077
Tokenared
Token ID1144
Feature activation+1.66e-01

Pos logit contributions
 Julia+0.124
La+0.115
 Mae+0.111
 Amber+0.111
 Zac+0.108
Neg logit contributions
 bruised-0.093
 wrench-0.092
-0.092
 bruises-0.091
 hammer-0.089
Token the
Token ID262
Feature activation+5.04e-02

Pos logit contributions
 bruised+0.240
 hammer+0.231
 swollen+0.225
 wrench+0.224
 bruises+0.222
Neg logit contributions
La-0.303
 Julia-0.303
 actor-0.290
 Mae-0.283
 Amber-0.279
Token top
Token ID1353
Feature activation+5.27e-03

Pos logit contributions
+0.092
uc+0.088
leaning+0.087
 bruised+0.086
pling+0.086
Neg logit contributions
 Julia-0.074
 dolphins-0.066
La-0.066
 Lou-0.066
 Jeff-0.065
Token,
Token ID11
Feature activation-1.32e-03

Pos logit contributions
+0.008
 wrench+0.008
 hammer+0.008
 bruised+0.008
 swollen+0.007
Neg logit contributions
 Julia-0.010
La-0.009
 Mae-0.009
 Amber-0.009
 actor-0.009
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 mate+0.003
nes+0.003
 actor+0.003
ozo+0.003
 Birthday+0.003
Neg logit contributions
 hammer-0.001
 sugar-0.001
 nails-0.001
 knives-0.001
 bruises-0.001
Token stopped
Token ID5025
Feature activation-7.32e-02

Pos logit contributions
 Julia+0.911
La+0.826
 Mae+0.824
 Amber+0.824
 Zac+0.812
Neg logit contributions
 hammer-0.671
 bruised-0.646
 wrench-0.644
 swollen-0.640
-0.637
Token in
Token ID287
Feature activation+6.70e-03

Pos logit contributions
 Julia+0.117
La+0.107
 Mae+0.103
 Amber+0.101
 Lou+0.099
Neg logit contributions
-0.130
 bruised-0.128
 wrench-0.127
 swollen-0.124
 bruises-0.123
Token her
Token ID607
Feature activation-7.99e-03

Pos logit contributions
 wrench+0.014
uc+0.014
pling+0.014
issors+0.014
mons+0.014
Neg logit contributions
 Julia-0.010
 Mae-0.009
 Amber-0.008
La-0.008
 April-0.008
Token tracks
Token ID8339
Feature activation-1.10e-01

Pos logit contributions
 Julia+0.016
La+0.014
 Mae+0.014
 Amber+0.014
 Zac+0.014
Neg logit contributions
 bruised-0.011
-0.011
 wrench-0.011
 swollen-0.011
pling-0.011
Token to
Token ID284
Feature activation+2.51e-03

Pos logit contributions
 Julia+0.196
La+0.182
 Mae+0.178
 Amber+0.175
 Zac+0.169
Neg logit contributions
 bruised-0.171
-0.171
 wrench-0.169
 hammer-0.166
 swollen-0.165
Token.
Token ID13
Feature activation+6.90e-03

Pos logit contributions
 Julia+0.066
La+0.061
 Mae+0.060
 Amber+0.059
 Zac+0.057
Neg logit contributions
 bruised-0.059
-0.058
 wrench-0.058
 hammer-0.057
 swollen-0.057
Token
Token ID198
Feature activation-3.22e-03

Pos logit contributions
+0.006
Here+0.004
leaning+0.004
 hammer+0.004
issors+0.004
Neg logit contributions
 actor-0.015
 mate-0.015
 dolphins-0.015
 diver-0.015
 aliens-0.014
Token
Token ID198
Feature activation-3.03e-03

Pos logit contributions
 actor+0.008
 mate+0.007
 diver+0.007
 dolphins+0.007
 echo+0.007
Neg logit contributions
-0.002
 hammer-0.002
Here-0.001
leaning-0.001
itt-0.001
TokenSuddenly
Token ID38582
Feature activation-6.12e-02

Pos logit contributions
 Dragon+0.006
 mate+0.006
 aliens+0.005
 actor+0.005
 diver+0.005
Neg logit contributions
leaning-0.003
ised-0.002
Here-0.002
issors-0.002
If-0.002
Token,
Token ID11
Feature activation+1.53e-04

Pos logit contributions
 actor+0.159
La+0.154
nes+0.152
 echo+0.144
April+0.143
Neg logit contributions
 hammer-0.032
 nails-0.028
 bruised-0.026
 sugar-0.026
 knives-0.025
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 hammer+0.000
 nails+0.000
 sugar+0.000
mons+0.000
 bruises+0.000
Neg logit contributions
 mate-0.000
 actor-0.000
nes-0.000
La-0.000
anna-0.000
Token heard
Token ID2982
Feature activation-1.22e-01

Pos logit contributions
 Julia+1.008
 Zac+0.957
 Jeff+0.941
 oper+0.919
 actor+0.915
Neg logit contributions
 hammer-0.589
pling-0.545
 bruised-0.533
 wrench-0.525
 swollen-0.518
Token a
Token ID257
Feature activation-6.60e-02

Pos logit contributions
 Zac+0.249
ozo+0.247
 diver+0.246
 Julia+0.243
 Mae+0.241
Neg logit contributions
 bruises-0.128
 nails-0.125
 bruised-0.119
 hammer-0.119
 oil-0.118
Token mysterious
Token ID11428
Feature activation+2.76e-02

Pos logit contributions
 Julia+0.106
 Mae+0.096
La+0.096
 Amber+0.095
 Zac+0.091
Neg logit contributions
 swollen-0.115
 bruised-0.115
 wrench-0.114
 hammer-0.114
-0.111
Token voice
Token ID3809
Feature activation-1.00e-01

Pos logit contributions
 hammer+0.048
 bruised+0.047
 wrench+0.047
 swollen+0.046
+0.046
Neg logit contributions
 Julia-0.045
La-0.041
 Mae-0.041
 Amber-0.040
 Zac-0.038
Token.
Token ID13
Feature activation+7.03e-03

Pos logit contributions
 Julia+0.205
 Mae+0.195
La+0.191
 Amber+0.190
 Zac+0.188
Neg logit contributions
 hammer-0.128
 bruised-0.126
 therm-0.120
 wrench-0.119
cut-0.116
Token strong
Token ID1913
Feature activation+4.97e-02

Pos logit contributions
 bruised+0.019
 swollen+0.019
 bruises+0.019
 hammer+0.019
 wrench+0.018
Neg logit contributions
 Julia-0.023
La-0.022
 Mae-0.021
 Amber-0.021
 Zac-0.021
Token!
Token ID0
Feature activation-1.76e-02

Pos logit contributions
+0.085
 wrench+0.084
 bruised+0.084
 swollen+0.083
 bruises+0.083
Neg logit contributions
 Julia-0.081
La-0.073
 Mae-0.072
 Amber-0.071
 Zac-0.069
Token
Token ID198
Feature activation-4.43e-03

Pos logit contributions
 actor+0.035
 mate+0.035
 dolphins+0.035
La+0.033
 diver+0.033
Neg logit contributions
-0.023
 bruises-0.018
 bruised-0.018
 hammer-0.018
 wrench-0.018
Token
Token ID198
Feature activation-3.87e-03

Pos logit contributions
 actor+0.009
 mate+0.009
 Julia+0.009
 dolphins+0.009
 diver+0.009
Neg logit contributions
-0.005
 bruises-0.005
 hammer-0.004
 swollen-0.004
pling-0.004
TokenSuddenly
Token ID38582
Feature activation-6.13e-02

Pos logit contributions
 mate+0.007
 actor+0.007
 aliens+0.007
 pirate+0.007
 dolphins+0.007
Neg logit contributions
-0.005
 bruises-0.004
leaning-0.004
Here-0.004
 swollen-0.004
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 actor+0.137
La+0.135
 mate+0.128
 villains+0.127
ozo+0.127
Neg logit contributions
 hammer-0.043
 nails-0.042
 sugar-0.041
-0.040
 bruises-0.039
Token heard
Token ID2982
Feature activation-1.23e-01

Pos logit contributions
 Julia+0.844
 actor+0.792
 Zac+0.786
La+0.786
 oper+0.780
Neg logit contributions
 hammer-0.691
-0.681
 swollen-0.674
pling-0.664
 bruised-0.662
Token a
Token ID257
Feature activation-7.04e-02

Pos logit contributions
 diver+0.236
 dives+0.234
ozo+0.234
nes+0.231
 actor+0.229
Neg logit contributions
 bruises-0.126
 nails-0.123
 hammer-0.120
 sugar-0.119
 oil-0.118
Token sn
Token ID3013
Feature activation-1.64e-01

Pos logit contributions
 Julia+0.108
La+0.099
 Mae+0.097
 Amber+0.096
 Zac+0.091
Neg logit contributions
 bruised-0.127
 wrench-0.126
 swollen-0.126
-0.125
 hammer-0.125
Tokenort
Token ID419
Feature activation-3.20e-02

Pos logit contributions
 Julia+0.266
La+0.242
 Mae+0.235
 Amber+0.234
ia+0.226
Neg logit contributions
 bruised-0.285
 swollen-0.284
 wrench-0.283
 bruises-0.281
 hammer-0.274
Token behind
Token ID2157
Feature activation-8.96e-02

Pos logit contributions
 Julia+0.060
La+0.058
 Mae+0.055
 Amber+0.054
 Zac+0.052
Neg logit contributions
 bruised-0.045
 hammer-0.045
 wrench-0.045
-0.044
 swollen-0.042
Token lend
Token ID22096
Feature activation+1.41e-01

Pos logit contributions
 Julia+0.042
La+0.041
 Mae+0.038
 Amber+0.038
 Zac+0.037
Neg logit contributions
-0.039
 hammer-0.039
 wrench-0.038
 bruised-0.038
 bruises-0.036
Token him
Token ID683
Feature activation+4.20e-02

Pos logit contributions
 gum+0.153
 bruises+0.146
 hammer+0.146
 bruised+0.141
 handcuffs+0.139
Neg logit contributions
 Julia-0.288
La-0.285
 actor-0.278
 diver-0.276
ae-0.274
Token her
Token ID607
Feature activation-7.28e-03

Pos logit contributions
 bruised+0.068
 wrench+0.067
 hammer+0.066
+0.066
 bruises+0.065
Neg logit contributions
 Julia-0.072
La-0.067
 Mae-0.065
 Amber-0.064
 April-0.062
Token boat
Token ID8848
Feature activation-9.41e-02

Pos logit contributions
 Julia+0.013
La+0.012
 Mae+0.011
 Amber+0.011
 Zac+0.011
Neg logit contributions
 bruised-0.012
-0.012
 wrench-0.012
 swollen-0.011
 hammer-0.011
Token.
Token ID13
Feature activation+3.04e-03

Pos logit contributions
 Julia+0.159
La+0.146
 Mae+0.144
 Amber+0.142
 Zac+0.136
Neg logit contributions
 bruised-0.155
-0.154
 wrench-0.153
 swollen-0.151
 bruises-0.149
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
+0.006
 hammer+0.006
 wrench+0.006
 rag+0.005
issors+0.005
Neg logit contributions
 dolphins-0.004
La-0.003
 diver-0.003
 royal-0.003
 mate-0.003
Token said
Token ID531
Feature activation+3.46e-03

Pos logit contributions
La+0.919
 Julia+0.910
 Mae+0.838
 Amber+0.828
 April+0.817
Neg logit contributions
 hammer-0.617
-0.607
 swollen-0.596
 bruises-0.595
 bruised-0.593
Token yes
Token ID3763
Feature activation-4.27e-02

Pos logit contributions
+0.005
 bruised+0.005
 wrench+0.005
 hammer+0.005
 swollen+0.005
Neg logit contributions
 Julia-0.006
La-0.006
 Mae-0.006
 Amber-0.005
 actor-0.005
Token;
Token ID26
Feature activation+3.88e-02

Pos logit contributions
 Julia+0.086
La+0.084
 Mae+0.080
 Amber+0.079
 dolphins+0.077
Neg logit contributions
-0.054
 wrench-0.054
 bruised-0.054
 hammer-0.054
pling-0.051
Token with
Token ID351
Feature activation-3.60e-02

Pos logit contributions
 wrench+0.065
 bruised+0.064
 hammer+0.064
+0.064
 swollen+0.062
Neg logit contributions
 Julia-0.064
La-0.060
 Mae-0.058
 Amber-0.057
 actor-0.055
Token a
Token ID257
Feature activation-6.56e-02

Pos logit contributions
 Julia+0.056
La+0.050
 Mae+0.050
 Amber+0.049
 Zac+0.047
Neg logit contributions
-0.066
 bruised-0.066
 wrench-0.065
 swollen-0.064
 hammer-0.063
Token Sam
Token ID3409
Feature activation-8.33e-02

Pos logit contributions
 hammer+0.013
 bruises+0.013
 supervision+0.013
+0.013
 bruised+0.013
Neg logit contributions
La-0.008
 actor-0.008
 royal-0.008
 mate-0.008
ae-0.008
Token waved
Token ID26834
Feature activation-1.94e-01

Pos logit contributions
 Julia+0.151
La+0.139
 Mae+0.138
 Amber+0.136
 Lou+0.129
Neg logit contributions
 bruised-0.128
-0.125
 hammer-0.123
 wrench-0.123
 bruises-0.122
Token goodbye
Token ID24829
Feature activation-1.45e-01

Pos logit contributions
 Julia+0.299
La+0.290
 Mae+0.282
 actor+0.265
 Zac+0.263
Neg logit contributions
 bruised-0.335
 wrench-0.318
 swollen-0.316
 bruises-0.316
-0.313
Token and
Token ID290
Feature activation+6.88e-03

Pos logit contributions
 Julia+0.265
La+0.249
 Mae+0.242
 Amber+0.237
 Zac+0.231
Neg logit contributions
 bruised-0.218
-0.218
 wrench-0.213
 hammer-0.210
 swollen-0.209
Token thanked
Token ID26280
Feature activation+1.50e-02

Pos logit contributions
 bruises+0.010
 bruised+0.010
 swollen+0.009
 hammer+0.009
 rag+0.009
Neg logit contributions
La-0.011
 actor-0.011
adies-0.011
 oper-0.011
 Julia-0.011
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
+0.020
 hammer+0.019
 bruises+0.019
 rag+0.018
free+0.018
Neg logit contributions
 dives-0.022
 mate-0.022
 royal-0.021
 awoke-0.021
 diver-0.021
Token for
Token ID329
Feature activation+4.78e-02

Pos logit contributions
La+0.960
 Julia+0.952
 Mae+0.895
 actor+0.891
 diver+0.862
Neg logit contributions
 bruised-0.542
-0.509
pling-0.506
 wrench-0.505
 hammer-0.504
Token being
Token ID852
Feature activation+4.03e-03

Pos logit contributions
+0.087
 wrench+0.084
 bruised+0.082
pling+0.082
 hammer+0.081
Neg logit contributions
 Julia-0.078
La-0.068
 Mae-0.068
 Jeff-0.067
 Amber-0.066
Token so
Token ID523
Feature activation-6.63e-02

Pos logit contributions
 bruised+0.006
 swollen+0.005
 bruises+0.005
 oil+0.005
 hammer+0.005
Neg logit contributions
 Julia-0.007
La-0.007
 dances-0.007
 dives-0.007
 Mae-0.007
Token clever
Token ID14169
Feature activation+2.28e-02

Pos logit contributions
 Julia+0.120
La+0.112
 Mae+0.111
 Amber+0.109
 April+0.104
Neg logit contributions
 bruised-0.103
 swollen-0.101
 bruises-0.100
-0.100
 wrench-0.099
Token.
Token ID13
Feature activation-1.55e-03

Pos logit contributions
 bruised+0.029
+0.028
 wrench+0.028
 swollen+0.027
 bruises+0.027
Neg logit contributions
La-0.044
 Julia-0.043
 Mae-0.041
 Amber-0.040
 Zac-0.040
Token the
Token ID262
Feature activation+4.32e-02

Pos logit contributions
 mate+0.109
 actor+0.108
La+0.106
adies+0.103
 Zac+0.103
Neg logit contributions
 bruises-0.088
-0.087
 sugar-0.086
 hammer-0.085
 swollen-0.085
Token leaves
Token ID5667
Feature activation+1.35e-01

Pos logit contributions
 bruised+0.057
pling+0.054
 supervision+0.053
 swollen+0.052
cut+0.052
Neg logit contributions
 Julia-0.083
 Lou-0.077
ia-0.077
 Alicia-0.075
 Mae-0.075
Token began
Token ID2540
Feature activation-3.90e-02

Pos logit contributions
 bruised+0.261
 wrench+0.260
 supervision+0.256
leaning+0.255
 bruises+0.254
Neg logit contributions
 Julia-0.175
La-0.152
 Mae-0.150
 dives-0.146
 Amber-0.145
Token to
Token ID284
Feature activation+1.93e-03

Pos logit contributions
 Julia+0.057
 Mae+0.048
La+0.048
 Amber+0.048
 Lou+0.047
Neg logit contributions
-0.071
leaning-0.071
 bruised-0.071
 wrench-0.070
pling-0.070
Token move
Token ID1445
Feature activation-1.01e-01

Pos logit contributions
+0.004
 bruises+0.004
 bruised+0.004
 wrench+0.004
pling+0.004
Neg logit contributions
 Julia-0.003
La-0.002
 Mae-0.002
 Amber-0.002
 Zac-0.002
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 Julia+0.172
La+0.158
 Mae+0.155
 Amber+0.152
 Zac+0.146
Neg logit contributions
 bruised-0.165
-0.164
 wrench-0.163
 swollen-0.160
 hammer-0.159
Token smiled
Token ID13541
Feature activation-5.78e-02

Pos logit contributions
 Julia+0.684
La+0.640
 Mae+0.605
 Amber+0.600
 actor+0.589
Neg logit contributions
 bruised-0.909
-0.890
 hammer-0.888
 wrench-0.881
 swollen-0.875
Token.
Token ID13
Feature activation+9.06e-03

Pos logit contributions
 Julia+0.114
La+0.109
 Mae+0.106
 Amber+0.102
 Zac+0.101
Neg logit contributions
 bruised-0.081
 wrench-0.078
-0.077
 swollen-0.077
 hammer-0.076
Token She
Token ID1375
Feature activation+7.20e-03

Pos logit contributions
+0.012
 hammer+0.010
 wrench+0.009
 bruised+0.009
 bruises+0.009
Neg logit contributions
 actor-0.018
 mate-0.018
 dolphins-0.018
La-0.017
 diver-0.017
Token felt
Token ID2936
Feature activation-3.09e-02

Pos logit contributions
 swollen+0.010
 hammer+0.010
 bruised+0.010
 bruises+0.010
+0.010
Neg logit contributions
 Julia-0.013
La-0.013
 Amber-0.013
 Mae-0.012
adies-0.012
Token brave
Token ID14802
Feature activation-6.73e-02

Pos logit contributions
 Julia+0.048
La+0.044
 Mae+0.043
 Amber+0.043
 Zac+0.040
Neg logit contributions
-0.056
 wrench-0.055
 bruised-0.055
 hammer-0.054
 swollen-0.054
Token me
Token ID502
Feature activation-1.25e-01

Pos logit contributions
 diver+0.013
 royal+0.013
 actor+0.012
 pirate+0.012
La+0.012
Neg logit contributions
 sugar-0.008
 supervision-0.008
 knives-0.007
 oil-0.007
issors-0.007
Token about
Token ID546
Feature activation-1.83e-01

Pos logit contributions
 Julia+0.246
La+0.245
 Mae+0.229
 Zac+0.224
 Amber+0.221
Neg logit contributions
 hammer-0.163
 bruised-0.162
 wrench-0.160
 bruises-0.148
pling-0.148
Token the
Token ID262
Feature activation+5.56e-02

Pos logit contributions
 Julia+0.339
La+0.319
 Mae+0.309
 Amber+0.301
 Zac+0.299
Neg logit contributions
 bruised-0.270
-0.263
 wrench-0.261
 hammer-0.261
 swollen-0.259
Token crystal
Token ID15121
Feature activation+8.87e-02

Pos logit contributions
+0.097
 wrench+0.093
 bruised+0.090
 bruises+0.089
uc+0.089
Neg logit contributions
 Julia-0.090
 Amber-0.082
 Mae-0.081
La-0.080
 actor-0.079
Token!"
Token ID2474
Feature activation-1.48e-01

Pos logit contributions
 hammer+0.121
 bruised+0.120
+0.116
pling+0.115
 wrench+0.113
Neg logit contributions
La-0.177
 Julia-0.176
 Zac-0.159
 Mae-0.159
 Amber-0.155
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 mate+0.157
 actor+0.143
 diver+0.137
 dolphins+0.136
La+0.135
Neg logit contributions
-0.281
 nails-0.258
 bruises-0.258
 powd-0.256
 hammer-0.255
Token went
Token ID1816
Feature activation-5.80e-02

Pos logit contributions
 Julia+0.884
La+0.870
 Amber+0.813
 Mae+0.812
 actor+0.807
Neg logit contributions
 hammer-0.624
-0.591
 bruised-0.589
uc-0.588
 swollen-0.576
Token closer
Token ID5699
Feature activation-1.27e-01

Pos logit contributions
 Julia+0.104
La+0.100
 Mae+0.098
 Amber+0.094
ae+0.093
Neg logit contributions
 hammer-0.086
 wrench-0.086
 bruised-0.085
-0.082
 bruises-0.079
Token and
Token ID290
Feature activation+1.44e-02

Pos logit contributions
La+0.251
 Julia+0.251
 actor+0.237
 Mae+0.233
 Amber+0.231
Neg logit contributions
 hammer-0.153
 bruised-0.152
-0.151
 wrench-0.147
pling-0.142
Token saw
Token ID2497
Feature activation-6.08e-03

Pos logit contributions
 bruised+0.020
 hammer+0.019
pling+0.019
 wrench+0.018
+0.018
Neg logit contributions
La-0.026
 Julia-0.026
ia-0.024
 actor-0.024
 oper-0.024
Token that
Token ID326
Feature activation+1.39e-01

Pos logit contributions
 Julia+0.011
La+0.011
 Mae+0.010
 Amber+0.010
 Zac+0.010
Neg logit contributions
 bruised-0.009
 hammer-0.009
 swollen-0.009
 bruises-0.008
 wrench-0.008
Token couldn
Token ID3521
Feature activation+6.98e-02

Pos logit contributions
 Julia+0.092
La+0.091
 Mae+0.087
 oper+0.086
 Zac+0.086
Neg logit contributions
-0.057
 bruised-0.056
 hammer-0.056
 swollen-0.055
pling-0.055
Token't
Token ID470
Feature activation+2.93e-02

Pos logit contributions
+0.089
 hammer+0.066
 swollen+0.063
 wrench+0.063
 therm+0.061
Neg logit contributions
 Julia-0.162
 mate-0.155
 actor-0.154
 dolphins-0.153
La-0.153
Token see
Token ID766
Feature activation-1.27e-01

Pos logit contributions
 hammer+0.046
+0.044
 wrench+0.042
 bruised+0.041
pling+0.040
Neg logit contributions
 Julia-0.049
La-0.048
 Zac-0.047
 Amber-0.044
 dances-0.044
Token anyone
Token ID2687
Feature activation-1.68e-01

Pos logit contributions
 Julia+0.209
La+0.199
 Mae+0.190
 Amber+0.188
 Zac+0.185
Neg logit contributions
 bruised-0.213
 swollen-0.201
 hammer-0.200
 bruises-0.199
-0.197
Token.
Token ID13
Feature activation+8.92e-03

Pos logit contributions
 Julia+0.326
La+0.317
 Mae+0.307
 Amber+0.301
 Zac+0.296
Neg logit contributions
 bruised-0.239
 hammer-0.229
 swollen-0.219
 wrench-0.218
-0.216
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
+0.010
 nails+0.008
 hammer+0.008
 bruises+0.008
issors+0.008
Neg logit contributions
 actor-0.018
 dolphins-0.017
 diver-0.017
 mate-0.017
La-0.017
Token was
Token ID373
Feature activation-4.39e-02

Pos logit contributions
 Julia+0.942
La+0.906
 Mae+0.869
 Zac+0.863
 Amber+0.855
Neg logit contributions
 hammer-0.599
 bruised-0.579
-0.570
 swollen-0.570
pling-0.563
Token scared
Token ID12008
Feature activation+2.84e-04

Pos logit contributions
La+0.093
 Julia+0.092
 Mae+0.087
 actor+0.086
 Amber+0.085
Neg logit contributions
 bruised-0.055
 swollen-0.049
 struggling-0.047
 bruises-0.047
 hammer-0.045
Token,
Token ID11
Feature activation+4.45e-04

Pos logit contributions
 bruised+0.000
 wrench+0.000
 hammer+0.000
 swollen+0.000
+0.000
Neg logit contributions
 Julia-0.001
La-0.001
 Mae-0.001
 Zac-0.001
 Amber-0.001
Token so
Token ID523
Feature activation-6.01e-02

Pos logit contributions
 bruised+0.001
+0.001
 wrench+0.001
 swollen+0.001
 hammer+0.001
Neg logit contributions
 Julia-0.001
La-0.001
 Mae-0.001
 Amber-0.001
 Zac-0.001
Token she
Token ID673
Feature activation-4.55e-02

Pos logit contributions
 mate+0.129
 actor+0.125
La+0.122
 oper+0.118
 aliens+0.118
Neg logit contributions
 bruises-0.053
 bruised-0.051
 swollen-0.049
 hammer-0.048
-0.047
Token,
Token ID11
Feature activation+8.15e-04

Pos logit contributions
+0.009
 hammer+0.008
 wrench+0.008
 bruised+0.007
uc+0.007
Neg logit contributions
 Julia-0.018
La-0.018
 actor-0.017
 oper-0.017
 echo-0.017
Token a
Token ID257
Feature activation-6.38e-02

Pos logit contributions
+0.001
 bruised+0.001
 wrench+0.001
 hammer+0.001
 swollen+0.001
Neg logit contributions
 Julia-0.001
La-0.001
 Mae-0.001
 Amber-0.001
 actor-0.001
Token little
Token ID1310
Feature activation+2.45e-02

Pos logit contributions
 Julia+0.109
La+0.101
 Mae+0.099
 Amber+0.097
 Zac+0.093
Neg logit contributions
 bruised-0.104
-0.103
 wrench-0.102
 swollen-0.101
 bruises-0.100
Token girl
Token ID2576
Feature activation-4.25e-03

Pos logit contributions
+0.043
 bruised+0.043
 bruises+0.042
 swollen+0.042
 wrench+0.042
Neg logit contributions
 Julia-0.038
La-0.035
 Mae-0.034
 Amber-0.034
 dolphins-0.033
Token named
Token ID3706
Feature activation+9.90e-03

Pos logit contributions
 Julia+0.008
La+0.007
 Mae+0.007
 Amber+0.007
 Zac+0.007
Neg logit contributions
 bruised-0.007
-0.006
 hammer-0.006
 wrench-0.006
 bruises-0.006
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
 bruised+0.017
+0.017
 wrench+0.016
 hammer+0.016
 swollen+0.016
Neg logit contributions
 Julia-0.016
La-0.015
 Mae-0.015
 Amber-0.014
 Zac-0.014
Token went
Token ID1816
Feature activation-5.76e-02

Pos logit contributions
 Julia+0.833
La+0.761
 Zac+0.756
 Mae+0.751
 Amber+0.738
Neg logit contributions
 bruised-0.753
 hammer-0.749
-0.741
 bruises-0.721
pling-0.718
Token exploring
Token ID13504
Feature activation-5.71e-02

Pos logit contributions
 Julia+0.112
La+0.103
 Mae+0.103
 Amber+0.100
 Zac+0.098
Neg logit contributions
 hammer-0.083
 wrench-0.083
 bruised-0.082
-0.080
 bruises-0.078
Token in
Token ID287
Feature activation+1.08e-02

Pos logit contributions
 Julia+0.091
La+0.082
 Mae+0.081
 Amber+0.080
 dolphins+0.076
Neg logit contributions
 bruised-0.102
-0.102
 wrench-0.101
 swollen-0.100
 bruises-0.098
Token the
Token ID262
Feature activation+5.03e-02

Pos logit contributions
+0.018
 bruised+0.018
 wrench+0.018
 swollen+0.017
 hammer+0.017
Neg logit contributions
 Julia-0.019
La-0.017
 Mae-0.017
 Amber-0.017
 April-0.016
Token woods
Token ID16479
Feature activation-7.06e-02

Pos logit contributions
 rag+0.069
 hammer+0.067
 oil+0.065
 bruised+0.064
 therm+0.063
Neg logit contributions
 Julia-0.101
 Zac-0.096
 Mae-0.095
 Jeff-0.091
ae-0.091
Token the
Token ID262
Feature activation+5.49e-02

Pos logit contributions
 Julia+0.283
La+0.268
 Mae+0.263
 Amber+0.257
 Zac+0.250
Neg logit contributions
 bruised-0.182
 hammer-0.181
 wrench-0.180
 bruises-0.175
-0.174
Token creek
Token ID39459
Feature activation-9.60e-02

Pos logit contributions
 hammer+0.089
 wrench+0.087
 bruised+0.086
+0.086
 oil+0.085
Neg logit contributions
 Julia-0.097
La-0.088
 Mae-0.088
 Amber-0.085
 Lou-0.082
Token.
Token ID13
Feature activation+3.04e-03

Pos logit contributions
 Julia+0.185
La+0.180
 Mae+0.168
 Amber+0.166
 April+0.163
Neg logit contributions
-0.136
 bruised-0.135
 hammer-0.135
 wrench-0.133
 bruises-0.126
Token He
Token ID679
Feature activation-2.16e-02

Pos logit contributions
 wrench+0.004
 hammer+0.004
 nails+0.003
 rag+0.003
 screws+0.003
Neg logit contributions
La-0.006
 dolphins-0.006
 Amber-0.005
 oper-0.005
 diver-0.005
Token asked
Token ID1965
Feature activation-2.96e-02

Pos logit contributions
 Julia+0.045
La+0.043
 oper+0.041
 Amber+0.041
 Mae+0.041
Neg logit contributions
 hammer-0.026
 wrench-0.025
 rag-0.025
 oil-0.025
pling-0.024
Token Julia
Token ID22300
Feature activation-4.78e-01

Pos logit contributions
La+0.043
 Amber+0.040
 diver+0.039
 royal+0.039
ae+0.038
Neg logit contributions
 hammer-0.047
 wrench-0.045
 tools-0.045
 screws-0.043
 nails-0.042
Token if
Token ID611
Feature activation+3.66e-02

Pos logit contributions
La+0.996
 Julia+0.990
 Mae+0.905
 actor+0.899
 Amber+0.884
Neg logit contributions
 hammer-0.568
 wrench-0.553
 bruised-0.544
pling-0.525
 supervision-0.507
Token she
Token ID673
Feature activation-4.29e-02

Pos logit contributions
 hammer+0.035
 wrench+0.032
 knives+0.032
 tools+0.030
 bruises+0.030
Neg logit contributions
La-0.076
 echo-0.073
ack-0.072
 pir-0.071
 oper-0.071
Token would
Token ID561
Feature activation-2.43e-02

Pos logit contributions
La+0.089
 Julia+0.088
 oper+0.082
 Amber+0.081
 Mae+0.081
Neg logit contributions
 hammer-0.054
 bruised-0.052
-0.051
 wrench-0.050
pling-0.050
Token lend
Token ID22096
Feature activation+1.41e-01

Pos logit contributions
 Julia+0.042
La+0.041
 Mae+0.038
 Amber+0.038
 Zac+0.037
Neg logit contributions
-0.039
 hammer-0.039
 wrench-0.038
 bruised-0.038
 bruises-0.036
Token him
Token ID683
Feature activation+4.20e-02

Pos logit contributions
 gum+0.153
 bruises+0.146
 hammer+0.146
 bruised+0.141
 handcuffs+0.139
Neg logit contributions
 Julia-0.288
La-0.285
 actor-0.278
 diver-0.276
ae-0.274
Token
Token ID220
Feature activation+1.04e-01

Pos logit contributions
+0.011
issors+0.009
leaning+0.008
 bruises+0.008
 hammer+0.008
Neg logit contributions
 actor-0.020
La-0.019
 diver-0.019
 dolphins-0.019
 oper-0.018
Token
Token ID198
Feature activation-2.51e-03

Pos logit contributions
+0.141
 wrench+0.126
 hammer+0.125
 bruises+0.122
 bruised+0.122
Neg logit contributions
 actor-0.197
La-0.196
 dolphins-0.195
 Julia-0.194
 oper-0.187
Token
Token ID198
Feature activation-2.54e-03

Pos logit contributions
 actor+0.005
 aliens+0.005
 diver+0.005
 mate+0.005
 dolphins+0.005
Neg logit contributions
-0.002
pling-0.002
leaning-0.002
Here-0.002
 hammer-0.002
TokenTogether
Token ID41631
Feature activation+2.72e-02

Pos logit contributions
 aliens+0.005
 Dragon+0.005
 actor+0.005
 mate+0.005
 dolphins+0.005
Neg logit contributions
leaning-0.002
-0.002
issors-0.002
Here-0.002
ised-0.002
Token,
Token ID11
Feature activation-1.41e-03

Pos logit contributions
+0.036
 wrench+0.036
 hammer+0.035
 bruised+0.034
 bruises+0.034
Neg logit contributions
 Julia-0.050
La-0.050
 Mae-0.046
 actor-0.046
 Amber-0.046
Token Julia
Token ID22300
Feature activation-4.77e-01

Pos logit contributions
 mate+0.002
La+0.002
 aliens+0.002
 villains+0.002
 actor+0.002
Neg logit contributions
-0.002
uc-0.002
mons-0.002
ised-0.002
 hammer-0.002
Token and
Token ID290
Feature activation+1.25e-02

Pos logit contributions
 Julia+0.865
La+0.812
 Mae+0.780
 Amber+0.778
 Zac+0.756
Neg logit contributions
 hammer-0.736
 wrench-0.729
-0.721
 bruised-0.702
pling-0.700
Token her
Token ID607
Feature activation-9.85e-03

Pos logit contributions
 hammer+0.016
 wrench+0.016
 bruised+0.015
 swollen+0.015
 bruises+0.015
Neg logit contributions
 Julia-0.023
La-0.023
 Mae-0.022
ae-0.022
 Zac-0.022
Token friends
Token ID2460
Feature activation-5.88e-02

Pos logit contributions
 Julia+0.018
La+0.017
 Mae+0.017
 Amber+0.016
 Zac+0.016
Neg logit contributions
 hammer-0.015
-0.015
 wrench-0.015
 bruised-0.015
 bruises-0.014
Token worked
Token ID3111
Feature activation+1.55e-01

Pos logit contributions
 Julia+0.092
 Mae+0.083
La+0.082
 Amber+0.080
 Jeff+0.079
Neg logit contributions
 bruised-0.106
 swollen-0.105
-0.104
 bruises-0.102
 wrench-0.100
Token hard
Token ID1327
Feature activation+1.19e-01

Pos logit contributions
 bruised+0.296
 swollen+0.287
 bruises+0.281
+0.281
 wrench+0.278
Neg logit contributions
 Julia-0.238
La-0.209
 Mae-0.207
 Amber-0.205
 Jeff-0.201
Token to
Token ID284
Feature activation+4.70e-03

Pos logit contributions
 Julia+0.031
La+0.029
 Mae+0.028
 Amber+0.028
 Zac+0.027
Neg logit contributions
 bruised-0.027
-0.027
 wrench-0.027
 hammer-0.026
 bruises-0.026
Token the
Token ID262
Feature activation+4.26e-02

Pos logit contributions
 wrench+0.007
 hammer+0.007
 bruised+0.007
+0.007
 swollen+0.007
Neg logit contributions
 Julia-0.008
La-0.008
 Mae-0.008
 Amber-0.007
 Zac-0.007
Token town
Token ID3240
Feature activation+1.08e-02

Pos logit contributions
 bruised+0.073
+0.073
 wrench+0.073
 hammer+0.072
 swollen+0.071
Neg logit contributions
 Julia-0.070
La-0.064
 Mae-0.063
 Amber-0.062
 Zac-0.059
Token,"
Token ID553
Feature activation+1.42e-02

Pos logit contributions
 hammer+0.015
 bruised+0.015
 wrench+0.015
+0.014
 bruises+0.014
Neg logit contributions
 Julia-0.021
La-0.020
 Mae-0.019
 Amber-0.019
 Zac-0.018
Token pleaded
Token ID16407
Feature activation+1.34e-01

Pos logit contributions
 bruises+0.017
 hammer+0.016
 swollen+0.016
 wrench+0.016
+0.016
Neg logit contributions
La-0.027
 Julia-0.026
 actor-0.025
 Mae-0.025
 mate-0.025
Token Julia
Token ID22300
Feature activation-4.77e-01

Pos logit contributions
 hammer+0.181
 wrench+0.162
uc+0.161
 bruises+0.157
 nails+0.157
Neg logit contributions
La-0.223
 awoke-0.216
 actor-0.213
 royal-0.211
 mate-0.207
Token.
Token ID13
Feature activation+8.78e-03

Pos logit contributions
La+0.725
 Julia+0.716
 dolphins+0.637
 Amber+0.634
 Mae+0.627
Neg logit contributions
 bruised-0.851
 hammer-0.838
 wrench-0.827
 bruises-0.818
 swollen-0.807
Token
Token ID220
Feature activation+1.04e-01

Pos logit contributions
+0.011
 hammer+0.010
 wrench+0.010
 bruised+0.010
 bruises+0.010
Neg logit contributions
 Julia-0.018
La-0.018
 actor-0.017
 Mae-0.017
 dolphins-0.017
Token
Token ID198
Feature activation-2.43e-03

Pos logit contributions
+0.141
 wrench+0.130
 hammer+0.129
 bruised+0.128
 bruises+0.128
Neg logit contributions
La-0.196
 Julia-0.196
 actor-0.194
 dolphins-0.189
 Mae-0.185
Token
Token ID198
Feature activation-2.05e-03

Pos logit contributions
 actor+0.005
 dolphins+0.005
 diver+0.005
 Julia+0.005
 mate+0.005
Neg logit contributions
-0.003
 hammer-0.002
leaning-0.002
 bruises-0.002
issors-0.002
TokenHer
Token ID9360
Feature activation-4.50e-02

Pos logit contributions
 dolphins+0.004
 Julia+0.004
 Zac+0.004
 actor+0.004
 Dragon+0.004
Neg logit contributions
leaning-0.002
-0.002
issors-0.002
ised-0.002
 bruises-0.002
Token had
Token ID550
Feature activation+3.00e-03

Pos logit contributions
 Julia+0.113
 Mae+0.099
 Jeff+0.098
 Lou+0.098
 mate+0.095
Neg logit contributions
-0.188
 wrench-0.179
 swollen-0.178
 bruised-0.174
 bruises-0.172
Token to
Token ID284
Feature activation+1.55e-03

Pos logit contributions
+0.006
 wrench+0.006
pling+0.006
 bruised+0.006
 swollen+0.006
Neg logit contributions
 Julia-0.004
 Mae-0.003
La-0.003
 mate-0.003
 dolphins-0.003
Token be
Token ID307
Feature activation-8.33e-02

Pos logit contributions
 bruised+0.002
 swollen+0.002
+0.002
pling+0.002
 wrench+0.002
Neg logit contributions
 Julia-0.003
La-0.002
 mate-0.002
 Mae-0.002
 Lou-0.002
Token brave
Token ID14802
Feature activation-6.82e-02

Pos logit contributions
 Julia+0.135
La+0.125
 mate+0.120
 Mae+0.119
 Amber+0.119
Neg logit contributions
-0.146
 wrench-0.132
pling-0.131
 bruised-0.130
 swollen-0.129
Token and
Token ID290
Feature activation+1.11e-02

Pos logit contributions
 Julia+0.126
La+0.119
 Mae+0.115
 Amber+0.113
 Zac+0.109
Neg logit contributions
 bruised-0.104
 wrench-0.101
-0.099
 swollen-0.098
 bruises-0.098
Token help
Token ID1037
Feature activation+2.44e-01

Pos logit contributions
+0.016
 bruised+0.015
 wrench+0.015
 swollen+0.015
pling+0.014
Neg logit contributions
 Julia-0.022
La-0.020
 Mae-0.020
 Amber-0.019
 Lou-0.019
Token the
Token ID262
Feature activation+4.59e-02

Pos logit contributions
 hammer+0.333
 bruised+0.331
 wrench+0.331
 bruises+0.315
 oil+0.309
Neg logit contributions
 Julia-0.470
La-0.463
 Amber-0.436
 Mae-0.434
 actor-0.421
Token animal
Token ID5044
Feature activation-7.75e-02

Pos logit contributions
+0.088
 bruised+0.088
 wrench+0.088
 swollen+0.086
 bruises+0.085
Neg logit contributions
 Julia-0.066
La-0.060
 Mae-0.058
 Amber-0.057
 Lou-0.055
Token.
Token ID13
Feature activation+2.07e-03

Pos logit contributions
 Julia+0.130
La+0.122
 Mae+0.118
 Amber+0.116
 Zac+0.111
Neg logit contributions
 bruised-0.130
-0.129
 wrench-0.128
 hammer-0.127
 swollen-0.126
Token They
Token ID1119
Feature activation-2.74e-02

Pos logit contributions
+0.003
 bruises+0.002
 bruised+0.002
 wrench+0.002
 oil+0.002
Neg logit contributions
La-0.004
 actor-0.004
 Julia-0.004
 oper-0.004
 royal-0.004
Token used
Token ID973
Feature activation+7.56e-02

Pos logit contributions
 Julia+0.038
 Mae+0.033
 Lou+0.033
 mate+0.033
 Amber+0.033
Neg logit contributions
-0.054
 bruised-0.052
 wrench-0.051
 swollen-0.051
 supervision-0.050
Token went
Token ID1816
Feature activation-4.90e-02

Pos logit contributions
 wrench+0.265
 bruised+0.264
+0.262
 hammer+0.262
 bruises+0.253
Neg logit contributions
 Julia-0.258
La-0.240
 Mae-0.234
 Amber-0.231
 Zac-0.218
Token to
Token ID284
Feature activation+1.43e-02

Pos logit contributions
 Julia+0.084
La+0.077
 Mae+0.076
 Amber+0.075
 Zac+0.072
Neg logit contributions
 bruised-0.081
-0.080
 wrench-0.079
 swollen-0.078
 bruises-0.078
Token a
Token ID257
Feature activation-5.52e-02

Pos logit contributions
 hammer+0.021
 wrench+0.021
+0.020
 bruised+0.020
 bruises+0.020
Neg logit contributions
 Julia-0.027
La-0.026
 Mae-0.025
 Amber-0.025
 Zac-0.023
Token store
Token ID3650
Feature activation+1.53e-01

Pos logit contributions
 Julia+0.099
La+0.090
 Mae+0.090
 Amber+0.088
 Zac+0.085
Neg logit contributions
 wrench-0.089
 hammer-0.088
-0.086
 bruised-0.086
 swollen-0.082
Token to
Token ID284
Feature activation+1.48e-02

Pos logit contributions
+0.276
 bruised+0.275
 swollen+0.274
 wrench+0.269
 bruises+0.265
Neg logit contributions
 Julia-0.238
 Mae-0.210
La-0.209
 Amber-0.207
 Jeff-0.198
Token buy
Token ID2822
Feature activation+2.58e-01

Pos logit contributions
 bruised+0.018
 swollen+0.018
+0.018
 wrench+0.017
 bruises+0.017
Neg logit contributions
 Julia-0.031
La-0.029
 Mae-0.029
 Amber-0.029
 actor-0.028
Token some
Token ID617
Feature activation+3.19e-02

Pos logit contributions
+0.501
 bruised+0.494
 swollen+0.490
 wrench+0.474
 bruises+0.469
Neg logit contributions
 Julia-0.369
La-0.327
 Mae-0.322
 Amber-0.318
 dolphins-0.314
Token things
Token ID1243
Feature activation+1.68e-01

Pos logit contributions
+0.050
 swollen+0.049
 bruised+0.048
pling+0.045
 bruises+0.044
Neg logit contributions
 Julia-0.053
 dolphins-0.051
 actor-0.050
La-0.049
 oper-0.049
Token.
Token ID13
Feature activation+1.97e-02

Pos logit contributions
 bruised+0.290
+0.289
 swollen+0.284
 wrench+0.284
 bruises+0.280
Neg logit contributions
 Julia-0.271
La-0.244
 Mae-0.242
 Amber-0.239
 Zac-0.230
Token The
Token ID383
Feature activation+8.58e-02

Pos logit contributions
+0.029
 wrench+0.029
 bruised+0.028
 hammer+0.028
 bruises+0.027
Neg logit contributions
 Julia-0.037
La-0.035
 Mae-0.034
 Amber-0.033
 Zac-0.032
Token store
Token ID3650
Feature activation+1.53e-01

Pos logit contributions
 swollen+0.175
 bruised+0.167
uc+0.161
pling+0.161
+0.160
Neg logit contributions
 pirate-0.106
 mate-0.105
 princess-0.102
 actor-0.102
 theater-0.101
Token if
Token ID611
Feature activation+3.90e-02

Pos logit contributions
 Julia+0.065
La+0.060
 Mae+0.059
 Amber+0.058
 Zac+0.056
Neg logit contributions
 bruised-0.056
-0.055
 wrench-0.055
 swollen-0.054
 hammer-0.053
Token he
Token ID339
Feature activation-6.10e-02

Pos logit contributions
 bruised+0.058
 swollen+0.057
 wrench+0.057
 hammer+0.056
 bruises+0.055
Neg logit contributions
 Julia-0.069
La-0.067
 Mae-0.064
 actor-0.061
ae-0.061
Token could
Token ID714
Feature activation+9.76e-03

Pos logit contributions
 Julia+0.118
La+0.113
 Mae+0.108
 Amber+0.107
 April+0.105
Neg logit contributions
 bruised-0.088
 hammer-0.085
-0.085
 wrench-0.084
 bruises-0.082
Token borrow
Token ID8804
Feature activation+9.23e-02

Pos logit contributions
 bruised+0.017
 wrench+0.017
 swollen+0.017
+0.016
 bruises+0.016
Neg logit contributions
 Julia-0.015
La-0.014
 Mae-0.014
 Amber-0.014
 Zac-0.013
Token her
Token ID607
Feature activation-6.86e-03

Pos logit contributions
+0.176
 wrench+0.170
 bruised+0.168
pling+0.166
 hammer+0.165
Neg logit contributions
 Julia-0.137
La-0.124
 Amber-0.122
 Mae-0.120
 Zac-0.118
Token comb
Token ID1974
Feature activation+2.46e-01

Pos logit contributions
 Julia+0.012
La+0.011
 Mae+0.010
 Amber+0.010
 Zac+0.010
Neg logit contributions
-0.011
 wrench-0.011
 bruised-0.011
 bruises-0.011
 swollen-0.011
Token to
Token ID284
Feature activation+3.65e-03

Pos logit contributions
+0.413
 bruised+0.410
 wrench+0.406
 swollen+0.402
 bruises+0.399
Neg logit contributions
 Julia-0.413
La-0.375
 Mae-0.370
 Amber-0.366
 Zac-0.351
Token comb
Token ID1974
Feature activation+2.46e-01

Pos logit contributions
+0.006
 bruised+0.006
 wrench+0.006
 swollen+0.006
pling+0.006
Neg logit contributions
 Julia-0.006
La-0.005
 Mae-0.005
 Amber-0.005
 Zac-0.005
Token his
Token ID465
Feature activation-5.14e-02

Pos logit contributions
+0.441
 bruised+0.432
 wrench+0.432
 bruises+0.424
 swollen+0.421
Neg logit contributions
 Julia-0.393
La-0.355
 Mae-0.351
 Amber-0.345
 Zac-0.333
Token fur
Token ID9230
Feature activation+4.48e-02

Pos logit contributions
 Julia+0.081
La+0.076
 Mae+0.073
 Amber+0.072
 April+0.071
Neg logit contributions
-0.085
 wrench-0.082
 bruised-0.082
pling-0.081
leaning-0.081
Token and
Token ID290
Feature activation+1.55e-02

Pos logit contributions
+0.079
 bruises+0.079
 bruised+0.079
 swollen+0.079
 wrench+0.078
Neg logit contributions
 Julia-0.068
 Mae-0.061
 Amber-0.060
La-0.060
 Zac-0.057
Token cat
Token ID3797
Feature activation+5.54e-02

Pos logit contributions
 wrench+0.212
 hammer+0.212
 bruised+0.207
+0.203
 rag+0.197
Neg logit contributions
 Julia-0.281
La-0.265
 Mae-0.254
 Amber-0.251
 Zac-0.244
Token named
Token ID3706
Feature activation+1.07e-02

Pos logit contributions
 bruised+0.077
 hammer+0.076
 wrench+0.074
+0.074
 swollen+0.073
Neg logit contributions
 Julia-0.109
La-0.102
 Mae-0.100
 Amber-0.099
 Zac-0.097
Token Mitt
Token ID16627
Feature activation+2.54e-01

Pos logit contributions
 bruised+0.015
+0.015
 wrench+0.014
 swollen+0.014
 hammer+0.014
Neg logit contributions
 Julia-0.021
La-0.020
 Mae-0.019
 Amber-0.019
 Zac-0.019
Tokenens
Token ID641
Feature activation+2.53e-01

Pos logit contributions
 bruised+0.238
 wrench+0.237
 swollen+0.231
+0.230
 bruises+0.229
Neg logit contributions
 Julia-0.591
La-0.572
 Mae-0.564
 Amber-0.557
 Alicia-0.545
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
 bruised+0.416
+0.416
 wrench+0.412
 swollen+0.405
 hammer+0.401
Neg logit contributions
 Julia-0.431
La-0.396
 Mae-0.390
 Amber-0.383
 Zac-0.368
Token Mitt
Token ID16627
Feature activation+2.54e-01

Pos logit contributions
cut+0.020
isher+0.020
 supervision+0.020
pling+0.018
plug+0.018
Neg logit contributions
 Ro-0.024
 Julia-0.023
 Mae-0.023
La-0.023
 Leo-0.022
Tokenens
Token ID641
Feature activation+2.52e-01

Pos logit contributions
 wrench+0.276
 bruised+0.262
 hammer+0.262
 swollen+0.254
 bruises+0.253
Neg logit contributions
 Julia-0.556
La-0.541
 Mae-0.533
ia-0.525
 Amber-0.521
Token liked
Token ID8288
Feature activation+1.01e-02

Pos logit contributions
 bruised+0.390
+0.383
 hammer+0.382
 swollen+0.381
 wrench+0.378
Neg logit contributions
 Julia-0.453
La-0.415
 Mae-0.406
 Amber-0.404
 Zac-0.389
Token to
Token ID284
Feature activation+4.27e-03

Pos logit contributions
+0.019
 bruised+0.018
 wrench+0.018
 swollen+0.018
pling+0.018
Neg logit contributions
 Julia-0.015
La-0.014
 Mae-0.014
 Amber-0.013
 dolphins-0.013
Token play
Token ID711
Feature activation-2.51e-02

Pos logit contributions
 hammer+0.007
 bruised+0.007
+0.007
 wrench+0.007
 bruises+0.007
Neg logit contributions
 Julia-0.007
La-0.006
 Amber-0.006
 Mae-0.006
 Zac-0.006
Token with
Token ID351
Feature activation-3.60e-02

Pos logit contributions
 Julia+0.037
La+0.035
 Mae+0.033
 Amber+0.033
 Jeff+0.032
Neg logit contributions
 bruised-0.047
-0.046
 wrench-0.046
 swollen-0.046
 bruises-0.045
Token down
Token ID866
Feature activation+5.19e-02

Pos logit contributions
 Julia+0.156
La+0.139
 Amber+0.137
 Mae+0.133
 dolphins+0.131
Neg logit contributions
-0.225
 supervision-0.220
 wrench-0.212
 rag-0.209
 bruises-0.209
Token and
Token ID290
Feature activation+1.72e-02

Pos logit contributions
 bruised+0.095
+0.094
 wrench+0.094
 swollen+0.094
 bruises+0.092
Neg logit contributions
 Julia-0.078
 Mae-0.070
La-0.069
 Amber-0.068
 Zac-0.064
Token looked
Token ID3114
Feature activation-3.39e-02

Pos logit contributions
+0.031
 wrench+0.030
 bruised+0.030
 hammer+0.029
 swollen+0.029
Neg logit contributions
 Julia-0.027
La-0.025
 Mae-0.024
 Amber-0.024
 Zac-0.023
Token in
Token ID287
Feature activation+1.42e-02

Pos logit contributions
 Julia+0.054
La+0.049
 Mae+0.049
 Amber+0.048
 dolphins+0.046
Neg logit contributions
-0.061
 wrench-0.058
 bruised-0.058
 bruises-0.058
 swollen-0.058
Token his
Token ID465
Feature activation-4.70e-02

Pos logit contributions
+0.031
pling+0.031
uc+0.030
mons+0.030
 wrench+0.030
Neg logit contributions
 Julia-0.020
 Mae-0.016
 Amber-0.016
 Jeff-0.016
La-0.015
Token sack
Token ID23704
Feature activation+2.95e-01

Pos logit contributions
 Julia+0.083
 Amber+0.078
 royal+0.073
La+0.073
 April+0.072
Neg logit contributions
pling-0.071
-0.070
 bruised-0.069
 swollen-0.069
free-0.068
Token.
Token ID13
Feature activation+1.40e-02

Pos logit contributions
+0.557
 bruised+0.549
 swollen+0.548
 bruises+0.542
 wrench+0.540
Neg logit contributions
 Julia-0.422
 Mae-0.369
 Amber-0.365
La-0.355
 Jeff-0.352
Token The
Token ID383
Feature activation+8.22e-02

Pos logit contributions
+0.024
 bruised+0.024
 wrench+0.024
 hammer+0.023
 swollen+0.023
Neg logit contributions
 Julia-0.023
La-0.021
 Mae-0.021
 Amber-0.020
 Zac-0.019
Token toy
Token ID13373
Feature activation+6.85e-02

Pos logit contributions
pling+0.130
+0.130
 bruised+0.129
 wrench+0.126
uc+0.124
Neg logit contributions
 Julia-0.135
 Amber-0.128
 dolphins-0.127
 princess-0.126
 mate-0.124
Token marriage
Token ID4845
Feature activation-1.62e-01

Pos logit contributions
 bruised+0.097
+0.097
 wrench+0.096
 swollen+0.094
 hammer+0.093
Neg logit contributions
 Julia-0.132
La-0.123
 Mae-0.121
 Amber-0.119
 Zac-0.115
Token made
Token ID925
Feature activation+7.27e-02

Pos logit contributions
 Julia+0.267
La+0.252
 Mae+0.244
 Amber+0.239
 Zac+0.232
Neg logit contributions
 bruised-0.271
-0.269
 wrench-0.267
 hammer-0.264
 swollen-0.262
Token
Token ID198
Feature activation-4.86e-03

Pos logit contributions
 Julia+0.009
La+0.008
 Mae+0.008
 Amber+0.008
 actor+0.008
Neg logit contributions
-0.007
 wrench-0.007
 bruised-0.007
 hammer-0.007
 swollen-0.007
TokenThe
Token ID464
Feature activation+5.81e-02

Pos logit contributions
 Julia+0.009
La+0.008
 Mae+0.008
 Amber+0.008
 Zac+0.008
Neg logit contributions
 bruised-0.007
-0.007
 wrench-0.007
 swollen-0.007
 hammer-0.007
Token rabbit
Token ID22746
Feature activation-7.60e-04

Pos logit contributions
+0.110
 bruised+0.110
 wrench+0.108
 bruises+0.107
pling+0.103
Neg logit contributions
 Julia-0.082
La-0.075
 Amber-0.075
 Mae-0.074
 Lou-0.073
Token hopped
Token ID45230
Feature activation-1.75e-01

Pos logit contributions
 Julia+0.001
La+0.001
 actor+0.001
 Mae+0.001
 Amber+0.001
Neg logit contributions
 swollen-0.001
 bruised-0.001
-0.001
 hammer-0.001
 bruises-0.001
Token away
Token ID1497
Feature activation+1.48e-01

Pos logit contributions
 Julia+0.355
La+0.337
 Mae+0.327
 Amber+0.317
 actor+0.316
Neg logit contributions
 bruised-0.224
-0.223
 wrench-0.217
 hammer-0.213
 swollen-0.212
Token,
Token ID11
Feature activation-2.05e-03

Pos logit contributions
 bruised+0.212
+0.204
 swollen+0.203
 hammer+0.197
 wrench+0.195
Neg logit contributions
 Julia-0.279
La-0.278
 Mae-0.259
 actor-0.253
 Amber-0.253
Token but
Token ID475
Feature activation+7.44e-02

Pos logit contributions
 Julia+0.004
La+0.003
 Mae+0.003
 Amber+0.003
 Zac+0.003
Neg logit contributions
 bruised-0.003
-0.003
 wrench-0.003
 swollen-0.003
 hammer-0.003
Token it
Token ID340
Feature activation+9.96e-03

Pos logit contributions
 bruised+0.119
+0.115
 swollen+0.113
 bruises+0.111
 hammer+0.110
Neg logit contributions
 Julia-0.128
La-0.128
 Mae-0.115
 Amber-0.114
 actor-0.114
Token turned
Token ID2900
Feature activation-1.17e-02

Pos logit contributions
 bruised+0.015
 swollen+0.015
+0.014
 bruises+0.014
 hammer+0.013
Neg logit contributions
 Julia-0.019
La-0.018
 Amber-0.017
 Mae-0.017
 actor-0.017
Token its
Token ID663
Feature activation-6.14e-02

Pos logit contributions
 Julia+0.018
La+0.016
 Mae+0.016
 Amber+0.016
 Zac+0.015
Neg logit contributions
-0.021
 wrench-0.021
 bruised-0.020
 bruises-0.020
 hammer-0.020
Token head
Token ID1182
Feature activation-2.59e-02

Pos logit contributions
 Julia+0.108
La+0.100
 Mae+0.098
 Amber+0.096
 Zac+0.093
Neg logit contributions
 bruised-0.099
-0.097
 wrench-0.097
 swollen-0.096
 hammer-0.095
Token shore
Token ID15191
Feature activation-1.78e-01

Pos logit contributions
 bruised+0.004
+0.004
 bruises+0.004
 hammer+0.004
 wrench+0.004
Neg logit contributions
 Julia-0.006
La-0.006
 Mae-0.005
 Amber-0.005
 April-0.005
Token and
Token ID290
Feature activation+1.62e-02

Pos logit contributions
 Julia+0.308
La+0.292
 Mae+0.281
 Amber+0.276
 April+0.266
Neg logit contributions
 bruised-0.289
 wrench-0.283
-0.281
 hammer-0.277
 swollen-0.274
Token he
Token ID339
Feature activation-6.21e-02

Pos logit contributions
 bruised+0.027
 hammer+0.026
 wrench+0.026
 bruises+0.026
 swollen+0.025
Neg logit contributions
 Julia-0.028
La-0.027
 Mae-0.026
 Amber-0.026
 Zac-0.024
Token drowned
Token ID30859
Feature activation+7.09e-02

Pos logit contributions
 Julia+0.126
La+0.119
 Mae+0.117
 Amber+0.117
 April+0.111
Neg logit contributions
 bruised-0.087
 hammer-0.084
 swollen-0.083
 wrench-0.082
 bruises-0.081
Token in
Token ID287
Feature activation+7.91e-03

Pos logit contributions
 bruised+0.108
 wrench+0.103
+0.101
 hammer+0.101
 bruises+0.100
Neg logit contributions
 Julia-0.135
La-0.129
 Mae-0.124
 Amber-0.123
 April-0.118
Token the
Token ID262
Feature activation+5.14e-02

Pos logit contributions
+0.015
pling+0.015
 wrench+0.015
uc+0.015
mons+0.015
Neg logit contributions
 Julia-0.012
 Mae-0.010
 Amber-0.010
La-0.010
 dolphins-0.010
Token sea
Token ID5417
Feature activation-2.00e-01

Pos logit contributions
 bruised+0.108
+0.108
 wrench+0.107
 swollen+0.106
 hammer+0.105
Neg logit contributions
 Julia-0.065
La-0.058
 Mae-0.056
 Amber-0.055
 Zac-0.052
Token.
Token ID13
Feature activation+1.44e-02

Pos logit contributions
 Julia+0.368
La+0.348
 Mae+0.335
 Amber+0.331
 Zac+0.320
Neg logit contributions
 bruised-0.306
-0.301
 wrench-0.300
 hammer-0.295
 swollen-0.290
Token
Token ID198
Feature activation+5.20e-06

Pos logit contributions
 wrench+0.029
 swollen+0.029
 bruised+0.029
 supervision+0.029
pling+0.028
Neg logit contributions
 Julia-0.019
 Jeff-0.016
 Boo-0.016
 Lou-0.015
 Ro-0.015
Token
Token ID198
Feature activation+2.80e-05

Feature not active on prev token;
no predictions were affected.
TokenThe
Token ID464
Feature activation+6.02e-02

Pos logit contributions
 supervision+0.000
 bruised+0.000
 wrench+0.000
+0.000
pling+0.000
Neg logit contributions
La-0.000
 Julia-0.000
ia-0.000
idi-0.000
 Ro-0.000
Token and
Token ID290
Feature activation+2.06e-02

Pos logit contributions
 Julia+0.020
La+0.019
 Mae+0.018
 Amber+0.018
 Zac+0.017
Neg logit contributions
 bruised-0.018
-0.018
 wrench-0.018
 swollen-0.017
 hammer-0.017
Token faster
Token ID5443
Feature activation-2.29e-01

Pos logit contributions
 bruised+0.034
 swollen+0.033
 bruises+0.033
 wrench+0.033
 hammer+0.033
Neg logit contributions
 Julia-0.036
La-0.033
 Mae-0.033
 Amber-0.032
 April-0.031
Token.
Token ID13
Feature activation+1.49e-02

Pos logit contributions
 Julia+0.386
La+0.354
 Mae+0.348
 Amber+0.343
 Zac+0.329
Neg logit contributions
-0.382
 bruised-0.381
 wrench-0.377
 swollen-0.371
 hammer-0.368
Token Do
Token ID2141
Feature activation+4.83e-03

Pos logit contributions
+0.022
 bruised+0.022
 wrench+0.022
 bruises+0.021
 hammer+0.021
Neg logit contributions
 Julia-0.027
La-0.025
 Mae-0.024
 Amber-0.024
 dolphins-0.024
Token you
Token ID345
Feature activation-2.96e-02

Pos logit contributions
 bruises+0.006
+0.006
 hammer+0.006
 wrench+0.006
 swollen+0.006
Neg logit contributions
 Julia-0.009
 Mae-0.009
La-0.009
 actor-0.009
 Amber-0.008
Token want
Token ID765
Feature activation-8.53e-03

Pos logit contributions
 Julia+0.053
La+0.049
 Mae+0.048
 Amber+0.047
 Zac+0.046
Neg logit contributions
 bruised-0.046
 wrench-0.046
 hammer-0.045
-0.045
 bruises-0.044
Token to
Token ID284
Feature activation+1.38e-02

Pos logit contributions
 Julia+0.017
La+0.016
 Mae+0.016
 Amber+0.015
 Zac+0.015
Neg logit contributions
 hammer-0.012
 bruises-0.011
 wrench-0.011
 oil-0.011
 bruised-0.011
Token buy
Token ID2822
Feature activation+2.55e-01

Pos logit contributions
+0.022
 bruised+0.022
 wrench+0.022
 hammer+0.021
 swollen+0.021
Neg logit contributions
 Julia-0.024
La-0.022
 Mae-0.022
 Amber-0.022
 Zac-0.021
Token some
Token ID617
Feature activation+2.87e-02

Pos logit contributions
 oil+0.367
 hammer+0.367
 wrench+0.365
 bruised+0.362
 bruises+0.357
Neg logit contributions
 Julia-0.471
La-0.447
 Mae-0.429
 Amber-0.412
 April-0.406
Token?"
Token ID1701
Feature activation-1.17e-01

Pos logit contributions
 bruised+0.045
+0.045
 wrench+0.044
 swollen+0.044
 bruises+0.043
Neg logit contributions
 Julia-0.051
La-0.047
 Mae-0.046
 Amber-0.045
 Zac-0.044
Token
Token ID198
Feature activation-1.71e-03

Pos logit contributions
 Julia+0.190
La+0.171
 Mae+0.170
 Amber+0.167
 Zac+0.160
Neg logit contributions
 bruised-0.205
 wrench-0.201
-0.201
 swollen-0.200
 bruises-0.197
Token want
Token ID765
Feature activation-8.15e-04

Pos logit contributions
 Julia+0.266
La+0.247
 Mae+0.243
 Amber+0.240
 Zac+0.232
Neg logit contributions
 bruised-0.191
-0.190
 wrench-0.188
 hammer-0.184
 swollen-0.183
Token to
Token ID284
Feature activation+1.90e-02

Pos logit contributions
 Julia+0.002
La+0.002
 Zac+0.002
 Amber+0.002
ia+0.002
Neg logit contributions
 oil-0.001
 hammer-0.001
 bruises-0.001
 gum-0.001
 sugar-0.001
Token hug
Token ID16225
Feature activation+1.09e-01

Pos logit contributions
 hammer+0.029
 wrench+0.025
 scrape+0.025
 bruised+0.025
 rag+0.025
Neg logit contributions
 Julia-0.035
 Mae-0.033
 Zac-0.032
 Amber-0.032
La-0.031
Token the
Token ID262
Feature activation+6.39e-02

Pos logit contributions
 hammer+0.123
 gum+0.122
 bruised+0.120
 oil+0.120
 wrench+0.116
Neg logit contributions
 Julia-0.210
 dives-0.199
 diver-0.199
ae-0.194
La-0.194
Token bear
Token ID6842
Feature activation+3.10e-02

Pos logit contributions
 bruised+0.120
+0.120
 wrench+0.119
 swollen+0.117
 hammer+0.116
Neg logit contributions
 Julia-0.093
La-0.085
 Mae-0.083
 Amber-0.081
 Zac-0.077
Token.
Token ID13
Feature activation+1.95e-02

Pos logit contributions
 bruised+0.039
 hammer+0.038
pling+0.037
cut+0.037
 wrench+0.037
Neg logit contributions
 Julia-0.062
La-0.061
 Amber-0.055
 Mae-0.055
 dives-0.054
Token And
Token ID843
Feature activation+1.20e-02

Pos logit contributions
 bruised+0.027
 wrench+0.026
+0.026
 hammer+0.025
 bruises+0.025
Neg logit contributions
 Julia-0.036
 dolphins-0.034
La-0.034
 Amber-0.034
 actor-0.033
Token I
Token ID314
Feature activation-1.37e-01

Pos logit contributions
 bruised+0.020
 hammer+0.019
 wrench+0.019
 bruises+0.019
+0.018
Neg logit contributions
La-0.019
 Julia-0.019
 Mae-0.017
 Amber-0.017
 actor-0.017
Token want
Token ID765
Feature activation-1.24e-03

Pos logit contributions
 Julia+0.270
La+0.251
 Mae+0.247
 Amber+0.244
 Zac+0.235
Neg logit contributions
 bruised-0.190
-0.189
 wrench-0.187
 swollen-0.183
 hammer-0.182
Token to
Token ID284
Feature activation+1.75e-02

Pos logit contributions
 Julia+0.003
La+0.003
 Amber+0.003
 Zac+0.003
 Mae+0.003
Neg logit contributions
 oil-0.001
 hammer-0.001
 bruises-0.001
 gum-0.001
 bruised-0.001
Token hug
Token ID16225
Feature activation+1.09e-01

Pos logit contributions
 hammer+0.026
 bruised+0.024
 scrape+0.024
 rag+0.024
 bruises+0.023
Neg logit contributions
 Julia-0.032
 Mae-0.031
 Zac-0.030
 Amber-0.029
La-0.029
Token the
Token ID262
Feature activation+4.97e-02

Pos logit contributions
 swollen+0.039
 hammer+0.039
 bruised+0.039
+0.038
 bruises+0.037
Neg logit contributions
La-0.057
 mate-0.057
 Julia-0.055
 dives-0.054
 actor-0.053
Token clown
Token ID25573
Feature activation-2.89e-01

Pos logit contributions
 bruises+0.079
 bruised+0.079
 therm+0.078
 swollen+0.077
 rag+0.076
Neg logit contributions
 Julia-0.071
La-0.069
 Mae-0.069
 Lou-0.068
 Amber-0.067
Token jugg
Token ID24873
Feature activation-2.55e-01

Pos logit contributions
 Julia+0.513
La+0.491
 Mae+0.486
 Amber+0.476
ae+0.453
Neg logit contributions
-0.419
 hammer-0.413
 wrench-0.402
 bruised-0.402
uc-0.400
Tokenled
Token ID992
Feature activation+3.22e-02

Pos logit contributions
 Julia+0.583
 April+0.554
La+0.552
 Mae+0.551
 throne+0.547
Neg logit contributions
 tant-0.251
ised-0.233
uc-0.221
anked-0.199
 bruised-0.193
Token balls
Token ID11333
Feature activation+9.51e-02

Pos logit contributions
 bruised+0.043
 hammer+0.042
 wrench+0.041
 swollen+0.041
 bruises+0.041
Neg logit contributions
 Julia-0.058
La-0.058
ae-0.056
 Mae-0.056
 Amber-0.054
Token,
Token ID11
Feature activation+1.88e-03

Pos logit contributions
 hammer+0.131
 wrench+0.129
+0.128
 bruised+0.127
 swollen+0.119
Neg logit contributions
 Julia-0.189
La-0.182
 Mae-0.175
 Amber-0.171
 April-0.168
Token spun
Token ID26843
Feature activation-1.48e-01

Pos logit contributions
 bruised+0.003
 wrench+0.003
 hammer+0.003
 swollen+0.003
+0.003
Neg logit contributions
 Julia-0.003
La-0.003
 Mae-0.003
 Amber-0.003
ae-0.003
Token plates
Token ID13375
Feature activation+4.30e-02

Pos logit contributions
 Julia+0.263
ae+0.262
 Mae+0.252
La+0.246
ia+0.243
Neg logit contributions
 oil-0.192
 wrench-0.191
 hammer-0.191
 swollen-0.191
 tant-0.189
Token and
Token ID290
Feature activation+1.66e-02

Pos logit contributions
 hammer+0.058
 wrench+0.058
+0.056
 bruised+0.056
pling+0.052
Neg logit contributions
 Julia-0.085
La-0.083
 Mae-0.077
 Amber-0.076
 April-0.075
Token rode
Token ID22075
Feature activation-1.12e-02

Pos logit contributions
 swollen+0.026
 hammer+0.026
 bruises+0.026
 bruised+0.025
 wrench+0.025
Neg logit contributions
 Julia-0.029
ae-0.026
La-0.026
ia-0.026
 actor-0.026
Token a
Token ID257
Feature activation-6.49e-02

Pos logit contributions
 Julia+0.021
ae+0.020
La+0.020
 Mae+0.020
 actor+0.020
Neg logit contributions
free-0.014
 hammer-0.013
 tant-0.013
cut-0.013
 wrench-0.013
TokenOne
Token ID3198
Feature activation-4.81e-02

Pos logit contributions
+0.058
 bruised+0.051
pling+0.050
 wrench+0.050
 hammer+0.050
Neg logit contributions
 Julia-0.088
La-0.084
 Mae-0.082
 actor-0.081
 Amber-0.080
Token day
Token ID1110
Feature activation+1.18e-02

Pos logit contributions
 Julia+0.104
La+0.102
 Amber+0.098
 Mae+0.096
 aliens+0.096
Neg logit contributions
 rag-0.049
 hammer-0.049
 bruised-0.049
-0.047
 therm-0.046
Token a
Token ID257
Feature activation-6.46e-02

Pos logit contributions
+0.014
 hammer+0.012
 wrench+0.012
 bruised+0.012
 swollen+0.011
Neg logit contributions
La-0.025
 Julia-0.025
 actor-0.025
 aliens-0.023
 oper-0.023
Token pilot
Token ID8022
Feature activation-2.18e-01

Pos logit contributions
 Julia+0.127
 Mae+0.116
La+0.114
 Amber+0.113
 Zac+0.112
Neg logit contributions
-0.092
 wrench-0.089
 hammer-0.089
 rag-0.086
 therm-0.084
Token flew
Token ID13112
Feature activation-1.32e-01

Pos logit contributions
 Julia+0.411
La+0.380
 Mae+0.378
 Amber+0.371
ae+0.355
Neg logit contributions
 hammer-0.329
 bruised-0.327
cut-0.320
 wrench-0.318
pling-0.316
Token through
Token ID832
Feature activation-4.47e-02

Pos logit contributions
 Julia+0.264
 Mae+0.248
La+0.243
 Amber+0.234
ae+0.233
Neg logit contributions
 bruised-0.176
 hammer-0.174
 wrench-0.171
 swollen-0.168
cut-0.164
Token the
Token ID262
Feature activation+5.42e-02

Pos logit contributions
 Julia+0.089
La+0.084
 Mae+0.083
 Amber+0.081
 Zac+0.079
Neg logit contributions
 hammer-0.058
 bruised-0.058
-0.057
 wrench-0.057
 swollen-0.057
Token sky
Token ID6766
Feature activation-6.63e-02

Pos logit contributions
 hammer+0.079
 oil+0.077
 rag+0.076
 therm+0.075
 bruised+0.075
Neg logit contributions
 Julia-0.104
 Mae-0.098
 Amber-0.093
 Zac-0.092
ae-0.091
Token in
Token ID287
Feature activation+1.12e-02

Pos logit contributions
 Julia+0.147
La+0.147
 Mae+0.137
 Amber+0.137
 April+0.135
Neg logit contributions
 hammer-0.070
 bruised-0.067
-0.066
 oil-0.062
cut-0.062
Token a
Token ID257
Feature activation-6.49e-02

Pos logit contributions
 bruised+0.015
 hammer+0.014
 bruises+0.014
 wrench+0.014
+0.014
Neg logit contributions
 Julia-0.022
La-0.021
 Mae-0.020
 Amber-0.020
 Zac-0.020
Token big
Token ID1263
Feature activation-4.90e-02

Pos logit contributions
 Julia+0.126
 Mae+0.119
 Zac+0.116
 Amber+0.114
 Jeff+0.111
Neg logit contributions
 hammer-0.091
 swollen-0.088
 rag-0.086
 wrench-0.086
 therm-0.084
Token a
Token ID257
Feature activation-6.52e-02

Pos logit contributions
 Julia+0.088
 Mae+0.081
La+0.080
ae+0.077
 Amber+0.076
Neg logit contributions
 hammer-0.059
 oil-0.059
 bruised-0.058
 wrench-0.057
 rag-0.055
Token rocket
Token ID10701
Feature activation-1.65e-01

Pos logit contributions
 Julia+0.132
 Mae+0.121
La+0.115
 Amber+0.114
 Alicia+0.112
Neg logit contributions
 hammer-0.096
 wrench-0.094
 rag-0.092
-0.091
 swollen-0.091
Token or
Token ID393
Feature activation-4.72e-02

Pos logit contributions
 Julia+0.327
La+0.315
 Mae+0.299
 Amber+0.299
Jul+0.287
Neg logit contributions
 hammer-0.235
 wrench-0.226
cut-0.217
 bruised-0.211
-0.210
Token a
Token ID257
Feature activation-6.48e-02

Pos logit contributions
 Julia+0.091
 Mae+0.085
La+0.084
ae+0.083
ia+0.083
Neg logit contributions
 hammer-0.069
 bruises-0.066
 wrench-0.066
 bruised-0.065
 nails-0.064
Token bird
Token ID6512
Feature activation-5.19e-02

Pos logit contributions
 Julia+0.115
 Mae+0.104
La+0.102
 Amber+0.099
 Alicia+0.097
Neg logit contributions
 hammer-0.109
 wrench-0.108
 bruised-0.103
 swollen-0.102
 rag-0.101
Token."
Token ID526
Feature activation-9.06e-02

Pos logit contributions
 Julia+0.102
 Mae+0.092
 Amber+0.091
La+0.090
ae+0.087
Neg logit contributions
 hammer-0.076
 bruised-0.074
 wrench-0.072
cut-0.071
 oil-0.070
Token
Token ID198
Feature activation-1.04e-03

Pos logit contributions
La+0.150
 actor+0.140
 oper+0.139
 royal+0.136
 April+0.134
Neg logit contributions
-0.150
 hammer-0.135
 oil-0.134
 wrench-0.134
 bruises-0.132
Token
Token ID198
Feature activation-6.79e-04

Pos logit contributions
 actor+0.002
 Julia+0.002
La+0.002
 Amber+0.002
 oper+0.002
Neg logit contributions
-0.001
 hammer-0.001
 oil-0.001
 wrench-0.001
 bruises-0.001
Token"
Token ID1
Feature activation+1.83e-02

Pos logit contributions
 Julia+0.001
 April+0.001
 oper+0.001
 actor+0.001
 Amber+0.001
Neg logit contributions
leaning-0.001
 wrench-0.001
 oil-0.001
 hammer-0.001
-0.001
TokenOK
Token ID11380
Feature activation+1.62e-01

Pos logit contributions
 bruised+0.023
+0.023
 wrench+0.023
 swollen+0.022
 hammer+0.022
Neg logit contributions
 Julia-0.039
La-0.037
 Mae-0.036
 Amber-0.035
 Zac-0.034
Token!"
Token ID2474
Feature activation-1.47e-01

Pos logit contributions
 wrench+0.227
 hammer+0.226
 bruised+0.224
+0.220
 swollen+0.212
Neg logit contributions
 Julia-0.305
La-0.304
 Mae-0.281
 Amber-0.279
 actor-0.272
Token and
Token ID290
Feature activation+1.34e-02

Pos logit contributions
 Julia+0.020
La+0.019
 Mae+0.019
 Amber+0.018
 Zac+0.018
Neg logit contributions
 bruised-0.016
-0.016
 wrench-0.016
 swollen-0.015
 hammer-0.015
Token from
Token ID422
Feature activation-5.14e-02

Pos logit contributions
 bruised+0.020
 swollen+0.020
 bruises+0.020
 hammer+0.019
+0.019
Neg logit contributions
 Julia-0.023
La-0.023
adies-0.021
 Mae-0.021
 actor-0.021
Token then
Token ID788
Feature activation+7.66e-04

Pos logit contributions
 Julia+0.089
La+0.085
 Mae+0.081
 Amber+0.080
 Zac+0.078
Neg logit contributions
 bruised-0.081
 swollen-0.078
-0.078
 wrench-0.078
 bruises-0.077
Token on
Token ID319
Feature activation+5.78e-02

Pos logit contributions
 bruised+0.001
+0.001
 wrench+0.001
 hammer+0.001
 swollen+0.001
Neg logit contributions
 Julia-0.002
La-0.001
 Mae-0.001
 Amber-0.001
 Zac-0.001
Token she
Token ID673
Feature activation-4.64e-02

Pos logit contributions
+0.094
 bruised+0.093
 wrench+0.092
 swollen+0.091
 hammer+0.091
Neg logit contributions
 Julia-0.098
La-0.093
 Mae-0.089
 Amber-0.088
 Zac-0.086
Token was
Token ID373
Feature activation-4.60e-02

Pos logit contributions
 Julia+0.085
La+0.080
 Zac+0.078
 Mae+0.078
 mate+0.077
Neg logit contributions
 bruised-0.066
-0.065
 hammer-0.064
 swollen-0.063
pling-0.062
Token never
Token ID1239
Feature activation-1.01e-02

Pos logit contributions
 Julia+0.086
La+0.081
 Zac+0.078
 dives+0.077
 Amber+0.077
Neg logit contributions
 bruised-0.070
 swollen-0.065
 bruises-0.062
 hammer-0.060
-0.059
Token sad
Token ID6507
Feature activation+9.53e-02

Pos logit contributions
 Julia+0.019
La+0.018
 Mae+0.018
 Zac+0.018
 Amber+0.017
Neg logit contributions
 bruised-0.015
 swollen-0.014
 bruises-0.014
 wrench-0.014
 hammer-0.014
Token again
Token ID757
Feature activation-1.52e-01

Pos logit contributions
 bruised+0.157
+0.145
 swollen+0.141
 wrench+0.141
 hammer+0.140
Neg logit contributions
 Julia-0.162
La-0.160
 Mae-0.148
 Zac-0.147
 April-0.146
Token.
Token ID13
Feature activation+7.25e-03

Pos logit contributions
 Julia+0.271
La+0.252
 Mae+0.246
 Amber+0.242
 Zac+0.236
Neg logit contributions
 bruised-0.239
-0.238
 wrench-0.233
 swollen-0.231
 bruises-0.228
Token<|endoftext|>
Token ID50256
Feature activation+3.91e-02

Pos logit contributions
+0.011
 bruised+0.011
 wrench+0.010
 swollen+0.010
 hammer+0.010
Neg logit contributions
 Julia-0.013
La-0.012
 actor-0.012
 dolphins-0.012
 Amber-0.012
Token called
Token ID1444
Feature activation-6.17e-02

Pos logit contributions
 Julia+0.010
La+0.009
 Mae+0.009
 Amber+0.009
 April+0.008
Neg logit contributions
-0.011
 wrench-0.011
 bruised-0.011
 swollen-0.011
 bruises-0.011
Token Lisa
Token ID15378
Feature activation-3.36e-01

Pos logit contributions
La+0.092
 Julia+0.089
 actor+0.087
 dolphins+0.086
 dives+0.085
Neg logit contributions
-0.106
 bruised-0.105
 hammer-0.102
 bruises-0.099
 wrench-0.098
Token.
Token ID13
Feature activation+8.39e-03

Pos logit contributions
 Julia+0.640
La+0.591
 Mae+0.583
 Amber+0.577
 Zac+0.569
Neg logit contributions
 bruised-0.482
 hammer-0.463
 swollen-0.462
 wrench-0.462
-0.461
Token Lisa
Token ID15378
Feature activation-3.36e-01

Pos logit contributions
+0.015
 hammer+0.014
 bruises+0.014
 bruised+0.014
issors+0.014
Neg logit contributions
 dolphins-0.011
 diver-0.011
 mate-0.011
 Julia-0.010
 Zac-0.010
Token had
Token ID550
Feature activation+4.08e-03

Pos logit contributions
 Julia+0.617
La+0.563
 Mae+0.556
 Amber+0.553
 Zac+0.545
Neg logit contributions
 bruised-0.508
 hammer-0.500
-0.497
 swollen-0.494
 wrench-0.488
Token a
Token ID257
Feature activation-6.78e-02

Pos logit contributions
 bruises+0.006
 bruised+0.006
 hammer+0.006
+0.006
 rag+0.006
Neg logit contributions
 Julia-0.008
 Zac-0.007
 dives-0.007
 Mae-0.007
 actor-0.007
Token diary
Token ID26339
Feature activation+5.07e-02

Pos logit contributions
 Julia+0.110
La+0.101
 Mae+0.099
 Amber+0.098
 April+0.093
Neg logit contributions
 bruised-0.116
-0.114
 wrench-0.113
 bruises-0.113
 swollen-0.111
Token that
Token ID326
Feature activation+1.34e-01

Pos logit contributions
 bruised+0.075
+0.075
 wrench+0.074
 swollen+0.073
 hammer+0.072
Neg logit contributions
 Julia-0.094
La-0.087
 Mae-0.086
 Amber-0.085
 Zac-0.082
Token she
Token ID673
Feature activation-4.43e-02

Pos logit contributions
 bruised+0.219
+0.215
 hammer+0.215
 wrench+0.211
 swollen+0.211
Neg logit contributions
 Julia-0.223
La-0.213
 Mae-0.202
 Amber-0.200
 Zac-0.198
Token kept
Token ID4030
Feature activation+8.91e-02

Pos logit contributions
 Julia+0.078
La+0.072
 Mae+0.070
 Amber+0.069
 Zac+0.067
Neg logit contributions
 bruised-0.071
-0.070
 wrench-0.070
 swollen-0.069
 hammer-0.068
Token all
Token ID477
Feature activation+3.98e-02

Pos logit contributions
+0.182
 wrench+0.168
 bruised+0.164
 bruises+0.164
 swollen+0.162
Neg logit contributions
 Julia-0.129
 April-0.114
La-0.113
 Amber-0.113
 Mae-0.111
Token fish
Token ID5916
Feature activation-1.30e-01

Pos logit contributions
+0.164
 bruised+0.163
 swollen+0.160
 wrench+0.160
 bruises+0.157
Neg logit contributions
 Julia-0.101
La-0.090
 Mae-0.088
 Amber-0.088
 dolphins-0.087
Token sw
Token ID1509
Feature activation-8.83e-02

Pos logit contributions
 Julia+0.224
La+0.216
 Mae+0.204
 Amber+0.202
 April+0.196
Neg logit contributions
 bruised-0.210
 hammer-0.209
-0.205
 swollen-0.204
 wrench-0.203
Tokenam
Token ID321
Feature activation-1.31e-01

Pos logit contributions
 Julia+0.123
La+0.113
 Mae+0.111
 Amber+0.107
 April+0.103
Neg logit contributions
 bruised-0.174
 wrench-0.172
-0.171
 bruises-0.169
 hammer-0.168
Token away
Token ID1497
Feature activation+1.47e-01

Pos logit contributions
 Julia+0.266
La+0.263
 Mae+0.252
 Amber+0.243
 April+0.241
Neg logit contributions
 bruised-0.165
 wrench-0.160
 hammer-0.160
 swollen-0.153
-0.151
Token and
Token ID290
Feature activation+1.30e-02

Pos logit contributions
 bruised+0.212
+0.205
 swollen+0.200
 hammer+0.196
 wrench+0.193
Neg logit contributions
 Julia-0.282
La-0.282
 Mae-0.262
 Amber-0.256
 actor-0.253
Token Daisy
Token ID40355
Feature activation-1.06e-01

Pos logit contributions
 bruised+0.023
 swollen+0.022
 bruises+0.022
+0.021
 hammer+0.021
Neg logit contributions
 Julia-0.021
La-0.021
 Mae-0.019
 actor-0.019
 Zac-0.019
Token felt
Token ID2936
Feature activation-3.02e-02

Pos logit contributions
 Julia+0.201
La+0.194
 Amber+0.184
 Mae+0.182
adies+0.179
Neg logit contributions
 bruised-0.149
-0.142
 hammer-0.141
 swollen-0.141
 bruises-0.139
Token happy
Token ID3772
Feature activation-7.14e-02

Pos logit contributions
 Julia+0.047
La+0.043
 Mae+0.042
 Amber+0.042
 Zac+0.040
Neg logit contributions
-0.054
 wrench-0.053
 bruised-0.053
 hammer-0.052
 swollen-0.052
Token and
Token ID290
Feature activation+1.22e-02

Pos logit contributions
 Julia+0.129
La+0.127
 Mae+0.120
 Amber+0.117
 Zac+0.116
Neg logit contributions
 bruised-0.109
-0.106
 bruises-0.103
 swollen-0.103
 wrench-0.102
Token proud
Token ID6613
Feature activation+8.33e-02

Pos logit contributions
 bruised+0.018
 swollen+0.017
 bruises+0.017
+0.017
 hammer+0.016
Neg logit contributions
 Julia-0.023
La-0.022
 Mae-0.021
 Zac-0.021
 Amber-0.021
Token.
Token ID13
Feature activation+3.18e-03

Pos logit contributions
 bruised+0.115
+0.112
 bruises+0.109
 wrench+0.109
 hammer+0.108
Neg logit contributions
 Julia-0.159
La-0.158
 Mae-0.149
 Amber-0.146
 Zac-0.145
Token decided
Token ID3066
Feature activation-7.55e-02

Pos logit contributions
 Julia+0.387
La+0.367
 Mae+0.354
 Zac+0.351
 Amber+0.345
Neg logit contributions
 hammer-0.278
-0.276
 bruised-0.275
 wrench-0.266
 bruises-0.261
Token to
Token ID284
Feature activation+4.58e-03

Pos logit contributions
 Julia+0.155
La+0.149
 Mae+0.145
 Amber+0.140
 Zac+0.139
Neg logit contributions
 hammer-0.098
 wrench-0.097
 bruised-0.095
 bruises-0.092
 swollen-0.092
Token build
Token ID1382
Feature activation+5.37e-02

Pos logit contributions
 hammer+0.007
+0.007
 wrench+0.007
 bruised+0.007
 rag+0.007
Neg logit contributions
 Julia-0.008
La-0.008
 Mae-0.007
 Amber-0.007
 Zac-0.007
Token a
Token ID257
Feature activation-6.77e-02

Pos logit contributions
 hammer+0.085
 wrench+0.083
 bruises+0.083
 bruised+0.081
 oil+0.080
Neg logit contributions
 Julia-0.094
La-0.093
 Mae-0.089
 Amber-0.087
 April-0.085
Token sand
Token ID6450
Feature activation-7.96e-02

Pos logit contributions
 Julia+0.126
La+0.120
 Mae+0.116
 Amber+0.114
 Zac+0.113
Neg logit contributions
 hammer-0.113
 wrench-0.109
 rag-0.102
 oil-0.101
 bruised-0.098
Token castle
Token ID16669
Feature activation-2.49e-01

Pos logit contributions
La+0.126
 Julia+0.123
 April+0.118
 Amber+0.117
 Mae+0.116
Neg logit contributions
 hammer-0.153
 wrench-0.148
-0.137
 nails-0.136
 oil-0.135
Token at
Token ID379
Feature activation-1.01e-03

Pos logit contributions
 Julia+0.503
La+0.497
 Mae+0.473
 Amber+0.471
 April+0.457
Neg logit contributions
 hammer-0.340
 wrench-0.332
-0.322
 bruised-0.319
 bruises-0.305
Token the
Token ID262
Feature activation+4.90e-02

Pos logit contributions
 Julia+0.002
La+0.002
 Mae+0.002
 Amber+0.002
 Zac+0.002
Neg logit contributions
-0.001
 hammer-0.001
 bruised-0.001
 wrench-0.001
 bruises-0.001
Token beach
Token ID10481
Feature activation-2.25e-01

Pos logit contributions
 hammer+0.077
 oil+0.073
+0.071
 file+0.071
 wrench+0.070
Neg logit contributions
 Julia-0.088
La-0.086
 April-0.080
 Zac-0.078
 Lou-0.078
Token.
Token ID13
Feature activation+7.63e-03

Pos logit contributions
 Julia+0.424
La+0.407
 Mae+0.386
 Amber+0.383
 Zac+0.374
Neg logit contributions
-0.323
 bruised-0.321
 wrench-0.320
 hammer-0.320
 bruises-0.306
Token They
Token ID1119
Feature activation-2.34e-02

Pos logit contributions
+0.012
 wrench+0.011
 hammer+0.011
 bruised+0.010
 rag+0.010
Neg logit contributions
La-0.014
 Julia-0.013
 Zac-0.012
 Mae-0.012
 oper-0.012
Token was
Token ID373
Feature activation-4.20e-02

Pos logit contributions
 bruised+0.084
+0.084
 wrench+0.083
 swollen+0.082
 bruises+0.081
Neg logit contributions
 Julia-0.065
La-0.059
 Mae-0.058
 Amber-0.056
 Zac-0.054
Token a
Token ID257
Feature activation-6.49e-02

Pos logit contributions
 Julia+0.075
La+0.069
 Mae+0.066
 Amber+0.065
 April+0.064
Neg logit contributions
-0.066
 bruised-0.065
 bruises-0.064
 hammer-0.064
 rag-0.063
Token young
Token ID1862
Feature activation+1.44e-02

Pos logit contributions
 Julia+0.118
La+0.109
 Mae+0.107
 Amber+0.106
 Zac+0.102
Neg logit contributions
 bruised-0.099
-0.099
 wrench-0.099
 hammer-0.097
 swollen-0.096
Token bear
Token ID6842
Feature activation+2.23e-02

Pos logit contributions
 bruised+0.023
 wrench+0.023
+0.023
 hammer+0.022
 swollen+0.022
Neg logit contributions
 Julia-0.025
La-0.024
 Mae-0.023
 Amber-0.023
 Zac-0.022
Token called
Token ID1444
Feature activation-5.65e-02

Pos logit contributions
 bruised+0.036
+0.036
 wrench+0.036
 swollen+0.035
 hammer+0.035
Neg logit contributions
 Julia-0.038
La-0.035
 Mae-0.035
 Amber-0.034
 Zac-0.033
Token Flash
Token ID9973
Feature activation-3.06e-01

Pos logit contributions
 Julia+0.105
La+0.105
 actor+0.094
 dances+0.094
 Mae+0.094
Neg logit contributions
 wrench-0.079
 hammer-0.079
 bruised-0.079
-0.078
pling-0.074
Token.
Token ID13
Feature activation+1.14e-02

Pos logit contributions
 Julia+0.614
La+0.576
 Amber+0.554
 Mae+0.554
 Alicia+0.533
Neg logit contributions
 bruised-0.420
 hammer-0.391
 wrench-0.390
-0.387
 swollen-0.386
Token He
Token ID679
Feature activation-1.78e-02

Pos logit contributions
 bruised+0.018
+0.018
 hammer+0.018
cut+0.018
 therm+0.018
Neg logit contributions
 Julia-0.017
 actor-0.016
 Amber-0.016
La-0.016
 diver-0.015
Token had
Token ID550
Feature activation+8.71e-03

Pos logit contributions
 Julia+0.035
 Mae+0.031
 Amber+0.031
La+0.031
 Alicia+0.029
Neg logit contributions
 bruised-0.026
 hammer-0.025
-0.025
 swollen-0.025
 wrench-0.025
Token very
Token ID845
Feature activation-3.60e-02

Pos logit contributions
 bruised+0.014
 bruises+0.014
 hammer+0.014
 swollen+0.014
 wrench+0.014
Neg logit contributions
 Julia-0.016
 Mae-0.014
 Amber-0.014
La-0.014
 Alicia-0.013
Token soft
Token ID2705
Feature activation+1.92e-02

Pos logit contributions
 Julia+0.064
La+0.059
 Mae+0.058
 Amber+0.057
 Zac+0.055
Neg logit contributions
 bruised-0.057
-0.057
 wrench-0.056
 swollen-0.056
 hammer-0.055
Token castle
Token ID16669
Feature activation-2.49e-01

Pos logit contributions
 Julia+0.099
La+0.092
 Mae+0.090
 Alicia+0.088
 Amber+0.087
Neg logit contributions
 bruised-0.081
 hammer-0.080
 wrench-0.078
 bruises-0.077
-0.076
Token so
Token ID523
Feature activation-6.11e-02

Pos logit contributions
 Julia+0.431
La+0.397
 Mae+0.390
 Amber+0.384
 Zac+0.369
Neg logit contributions
 bruised-0.403
-0.401
 wrench-0.398
 swollen-0.391
 hammer-0.389
Token much
Token ID881
Feature activation-6.80e-02

Pos logit contributions
 Julia+0.113
La+0.108
 Mae+0.105
 Amber+0.102
 Zac+0.100
Neg logit contributions
 swollen-0.093
 bruised-0.092
 hammer-0.086
 wrench-0.085
-0.084
Token.
Token ID13
Feature activation+9.12e-03

Pos logit contributions
 Julia+0.132
La+0.125
 Mae+0.122
 Amber+0.119
 Zac+0.117
Neg logit contributions
 bruised-0.097
 wrench-0.095
-0.093
 swollen-0.093
 hammer-0.093
Token The
Token ID383
Feature activation+7.81e-02

Pos logit contributions
+0.012
 hammer+0.011
 wrench+0.011
 bruised+0.010
 swollen+0.010
Neg logit contributions
La-0.018
 Julia-0.018
 dolphins-0.017
 Mae-0.017
 Zac-0.017
Token mer
Token ID4017
Feature activation-3.02e-01

Pos logit contributions
 hammer+0.140
 iron+0.138
 cloth+0.136
 swollen+0.136
 jug+0.134
Neg logit contributions
La-0.090
 Julia-0.088
 Zac-0.088
ae-0.084
 Mae-0.083
Tokenmaid
Token ID23151
Feature activation-2.60e-01

Pos logit contributions
La+0.442
 Julia+0.395
 Mae+0.393
 Ro+0.389
 Alicia+0.371
Neg logit contributions
uc-0.551
 tant-0.512
isher-0.498
 bruised-0.490
ised-0.488
Token said
Token ID531
Feature activation+3.46e-03

Pos logit contributions
 Julia+0.515
La+0.499
 Mae+0.482
 Amber+0.468
 Zac+0.457
Neg logit contributions
 swollen-0.352
 bruised-0.344
 hammer-0.341
 wrench-0.329
-0.328
Token,
Token ID11
Feature activation-4.63e-04

Pos logit contributions
+0.005
 bruised+0.005
 wrench+0.005
 swollen+0.005
 bruises+0.005
Neg logit contributions
 Julia-0.006
La-0.006
 Mae-0.006
 Amber-0.006
 Jeff-0.005
Token "
Token ID366
Feature activation+4.31e-02

Pos logit contributions
 Julia+0.001
 Mae+0.001
 Amber+0.001
La+0.001
 April+0.001
Neg logit contributions
 bruised-0.001
 bruises-0.001
-0.001
 swollen-0.001
 wrench-0.001
TokenI
Token ID40
Feature activation+4.95e-02

Pos logit contributions
+0.052
 swollen+0.051
 bruised+0.051
 wrench+0.050
 hammer+0.050
Neg logit contributions
 Julia-0.091
 Mae-0.084
La-0.083
 Amber-0.083
 Zac-0.081
Token liked
Token ID8288
Feature activation+1.20e-02

Pos logit contributions
 Julia+0.144
La+0.132
 Mae+0.131
 Amber+0.129
 Zac+0.124
Neg logit contributions
 bruised-0.132
-0.131
 wrench-0.130
 swollen-0.128
 hammer-0.128
Token.
Token ID13
Feature activation+1.38e-02

Pos logit contributions
 bruised+0.015
 hammer+0.015
 wrench+0.015
 oil+0.015
 bruises+0.015
Neg logit contributions
 Julia-0.024
La-0.023
 Amber-0.022
 Mae-0.022
 Zac-0.021
Token Tom
Token ID4186
Feature activation-4.62e-02

Pos logit contributions
 bruised+0.025
 wrench+0.025
+0.025
 swollen+0.025
 hammer+0.024
Neg logit contributions
 Julia-0.022
La-0.020
 Mae-0.019
 Amber-0.019
 Zac-0.018
Token picked
Token ID6497
Feature activation+1.56e-01

Pos logit contributions
 Julia+0.096
La+0.090
 Amber+0.089
 oper+0.087
 Mae+0.085
Neg logit contributions
 bruised-0.056
 hammer-0.054
 bruises-0.053
 wrench-0.052
 swollen-0.051
Token a
Token ID257
Feature activation-6.16e-02

Pos logit contributions
 oil+0.193
 gum+0.185
 hammer+0.180
 bruised+0.179
 sugar+0.178
Neg logit contributions
La-0.300
 Julia-0.284
nes-0.277
 mate-0.270
ae-0.268
Token movie
Token ID3807
Feature activation-2.96e-01

Pos logit contributions
 Julia+0.112
 Amber+0.100
 Mae+0.100
La+0.099
 Lou+0.096
Neg logit contributions
 wrench-0.097
 hammer-0.097
 bruised-0.094
 therm-0.093
 swollen-0.091
Token about
Token ID546
Feature activation-1.80e-01

Pos logit contributions
 Julia+0.567
La+0.555
 Amber+0.527
 Mae+0.523
 April+0.507
Neg logit contributions
 wrench-0.406
 bruised-0.402
-0.401
 hammer-0.396
 bruises-0.385
Token cars
Token ID5006
Feature activation+7.08e-03

Pos logit contributions
 Julia+0.328
La+0.314
 Amber+0.303
 Mae+0.301
ae+0.287
Neg logit contributions
 bruised-0.261
 bruises-0.259
 hammer-0.258
 swollen-0.258
 oil-0.254
Token and
Token ID290
Feature activation+1.51e-02

Pos logit contributions
 bruised+0.012
+0.012
 wrench+0.012
 swollen+0.012
 bruises+0.011
Neg logit contributions
 Julia-0.012
La-0.011
 Mae-0.011
 Amber-0.010
 Zac-0.010
Token trucks
Token ID13960
Feature activation+1.11e-01

Pos logit contributions
 bruises+0.021
 bruised+0.020
 swollen+0.020
 wrench+0.020
 hammer+0.020
Neg logit contributions
 Julia-0.030
 Amber-0.028
La-0.028
ae-0.027
 Mae-0.027
Token.
Token ID13
Feature activation+1.36e-02

Pos logit contributions
 bruised+0.178
+0.177
 wrench+0.175
 swollen+0.172
 hammer+0.171
Neg logit contributions
 Julia-0.195
La-0.180
 Mae-0.176
 Amber-0.174
 Zac-0.167
Token of
Token ID286
Feature activation-5.51e-02

Pos logit contributions
 hammer+0.044
 bruised+0.043
 rag+0.041
 wrench+0.041
 swollen+0.040
Neg logit contributions
La-0.078
 Mae-0.074
 Julia-0.074
 Amber-0.074
 actor-0.073
Token the
Token ID262
Feature activation+5.64e-02

Pos logit contributions
 Julia+0.102
La+0.094
 Mae+0.093
 Amber+0.091
 Zac+0.088
Neg logit contributions
 bruised-0.084
-0.084
 wrench-0.084
 swollen-0.081
 hammer-0.081
Token day
Token ID1110
Feature activation+9.05e-03

Pos logit contributions
+0.082
 wrench+0.077
 hammer+0.077
 bruised+0.077
 swollen+0.076
Neg logit contributions
 Julia-0.114
 Mae-0.103
La-0.103
 Amber-0.102
 Lou-0.099
Token,
Token ID11
Feature activation+1.83e-03

Pos logit contributions
+0.010
 sugar+0.008
 hammer+0.008
 bruised+0.008
 oil+0.008
Neg logit contributions
La-0.021
 mate-0.020
 Julia-0.019
 Zac-0.019
 actor-0.019
Token the
Token ID262
Feature activation+5.59e-02

Pos logit contributions
 wrench+0.003
 bruised+0.003
+0.003
 bruises+0.003
 swollen+0.003
Neg logit contributions
 Julia-0.003
 Amber-0.003
 Mae-0.003
La-0.003
 April-0.003
Token superhero
Token ID17343
Feature activation-2.56e-01

Pos logit contributions
 bruised+0.073
+0.072
 wrench+0.071
 hammer+0.071
 swollen+0.069
Neg logit contributions
 Julia-0.116
 Mae-0.107
La-0.106
 Amber-0.103
 Jeff-0.103
Token brought
Token ID3181
Feature activation+4.49e-02

Pos logit contributions
 Julia+0.488
La+0.477
 Mae+0.447
 Amber+0.435
 Zac+0.433
Neg logit contributions
pling-0.349
 hammer-0.349
 bruised-0.345
 wrench-0.344
-0.337
Token the
Token ID262
Feature activation+5.53e-02

Pos logit contributions
 bruised+0.061
 bruises+0.058
 oil+0.058
 groceries+0.058
 hammer+0.057
Neg logit contributions
 Julia-0.088
La-0.086
 actor-0.081
 Zac-0.080
 Mae-0.080
Token little
Token ID1310
Feature activation+1.95e-02

Pos logit contributions
 bruised+0.082
 wrench+0.081
 hammer+0.080
+0.080
 bruises+0.078
Neg logit contributions
 Julia-0.106
La-0.099
 Mae-0.097
 Amber-0.095
 Lou-0.093
Token girl
Token ID2576
Feature activation-4.14e-03

Pos logit contributions
 bruised+0.022
 wrench+0.021
 hammer+0.020
 rag+0.020
 oil+0.019
Neg logit contributions
 Julia-0.043
La-0.042
 Jeff-0.041
 Mae-0.041
 Lou-0.040
Token safely
Token ID11512
Feature activation-4.34e-02

Pos logit contributions
 Julia+0.007
La+0.007
 Mae+0.006
 Amber+0.006
 Zac+0.006
Neg logit contributions
 bruised-0.007
 hammer-0.007
 bruises-0.007
-0.007
 wrench-0.007