TOP ACTIVATIONS
MAX = 11.788

disob edient <|endoftext|> T essa was
something strange the k itty was
hung rily she had seen milk
it was ugly â they loved it just
high rat much bigger than the
something tasty a mineral . But
and surprises Amy had managed to
his food only to find it
something important the doll had a
and smiled it was a special
was too late â the o at had
big surprise there was a big
favorite cereal she ate it with
, pink scarf â she kept it safe
final destination Lilly 's favorite place
't believe it â this one feather seemed
he liked a big soft swing
special thing wings . Every day
forest today a small fairy who
other family a dad , a

MIN ACTIVATIONS
MIN = -11.480

pieces to try and check mate your opponent ."
s King in check mate . The little
. " Check mate !" declared Elsa . Her
enough to find your soul mate , don 't be afraid
popular that everyone wanted a pair of the same .
I could use an extra pair of hands ." The boy
and make it go v room v room . When they
, v room , v room ! They were
makes v room - v room sounds . "
Lily back to the play room . She made Ben look
Now he had a great pair of pants to show off
sure to keep the extra pair of pants , just in
explore . She had a pair of sunglasses with her to
girl saw a bright orange pair of shoes that she liked
around . They see a pair of scissors on the table
a shirt , then a pair of pants , then a
sque ak and saw a pair of bright eyes staring at
wardrobe and she found a pair of sunglasses . She had
day , Alice designed a pair of wings for Bob .
They were quite a charming pair and they always made everyone

INTERVAL 5.971 - 11.788
CONTAINS 0.695%

joy ed . He had conquered something dangerous and he was
One day , Tim my 's mom made him a y
, " Yes ! Let 's play before the sunrise !"
um ma kissed him good night and urged him to go
and talk about why it 's important to catch healthy fish

INTERVAL 0.154 - 5.971
CONTAINS 47.204%

. One day , while she was playing in the park
a magic stone . The human did as he was told
bread and cakes , and the love of their mother and
John . John loved to explore the outdoors and he often
field , eating grass and feeling the wind in his fur

INTERVAL -5.663 - 0.154
CONTAINS 51.593%

in circles , cl apping his fins together .
He was always pushing the other fishes away when they tried
day , her mom gave her an ancient card that was
. They hugged her and told her what happened .
getting ready to make a cup of hot chocolate . The

INTERVAL -11.480 - -5.663
CONTAINS 0.507%

watched as everyone took a few . When it
having fun . She saw them in their old house and
lost my way home and now I don 't own a
measured the ingredients and blended them all together . Finally ,
Pat ches got an exciting idea . She decided to put
Token disob
Token ID30163
Feature activation+3.64e+00

Pos logit contributions
 them+0.217
 wreck+0.201
 polar+0.194
mate+0.183
inn+0.181
Neg logit contributions
-0.213
cious-0.209
's-0.204
ience-0.201
ages-0.201
Tokenedient
Token ID35279
Feature activation+2.79e+00

Pos logit contributions
+0.189
's+0.176
cious+0.176
ience+0.176
ages+0.175
Neg logit contributions
 them-0.283
 wreck-0.262
mate-0.259
 polar-0.253
inn-0.247
Token
Token ID6184
Feature activation+2.01e-01

Pos logit contributions
+0.213
cious+0.208
ience+0.203
ages+0.203
's+0.202
Neg logit contributions
 them-0.149
 wreck-0.129
mate-0.128
 polar-0.123
inn-0.117
Token
Token ID95
Feature activation-1.68e-01

Pos logit contributions
+0.012
's+0.012
ages+0.012
cious+0.012
ience+0.011
Neg logit contributions
 them-0.013
 wreck-0.013
mate-0.012
 polar-0.012
inn-0.012
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.011
 wreck+0.011
 them+0.011
inn+0.011
 ramp+0.010
Neg logit contributions
ience-0.009
ages-0.009
-0.009
cious-0.009
's-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.200
 persever+0.191
 decisions+0.190
 dedication+0.186
aul+0.186
Neg logit contributions
 them-0.212
 wreck-0.193
 polar-0.184
mate-0.178
 ramp-0.176
Token
Token ID250
Feature activation-1.63e-01

Pos logit contributions
cious+0.832
+0.829
's+0.805
ages+0.800
ita+0.775
Neg logit contributions
 them-0.654
 wreck-0.626
mate-0.585
 polar-0.558
 tow-0.535
Token<|endoftext|>
Token ID50256
Feature activation+9.28e-01

Pos logit contributions
 them+0.012
 wreck+0.011
mate+0.011
 polar+0.011
inn+0.010
Neg logit contributions
-0.009
's-0.009
ages-0.008
cious-0.008
ience-0.008
TokenT
Token ID51
Feature activation+2.13e+00

Pos logit contributions
ages+0.065
ience+0.064
cious+0.063
+0.062
's+0.060
Neg logit contributions
 them-0.056
mate-0.048
 wreck-0.047
 polar-0.047
inn-0.045
Tokenessa
Token ID21411
Feature activation+3.35e+00

Pos logit contributions
+0.122
cious+0.115
's+0.114
ages+0.114
ience+0.113
Neg logit contributions
 them-0.155
 wreck-0.141
mate-0.139
 polar-0.139
inn-0.135
Token was
Token ID373
Feature activation-4.77e-01

Pos logit contributions
cious+0.260
ages+0.255
+0.253
ience+0.251
 Michelle+0.249
Neg logit contributions
 them-0.161
 polar-0.140
 wreck-0.139
mate-0.127
 rock-0.123
Token something
Token ID1223
Feature activation+4.36e+00

Pos logit contributions
 them+0.040
 polar+0.031
 wreck+0.031
mate+0.031
 his+0.029
Neg logit contributions
ages-0.050
-0.049
cious-0.049
ience-0.048
's-0.047
Token strange
Token ID6283
Feature activation+1.33e+00

Pos logit contributions
+0.307
's+0.291
cious+0.288
ages+0.284
ience+0.283
Neg logit contributions
 them-0.259
 wreck-0.235
mate-0.233
 polar-0.227
inn-0.218
Token
Token ID6184
Feature activation+2.01e-01

Pos logit contributions
+0.103
ages+0.102
cious+0.101
ience+0.101
's+0.098
Neg logit contributions
 them-0.071
 wreck-0.059
mate-0.058
 polar-0.055
inn-0.052
Token
Token ID95
Feature activation-1.61e-01

Pos logit contributions
ages+0.014
's+0.013
+0.012
 pads+0.012
ita+0.012
Neg logit contributions
 them-0.012
 wreck-0.011
mate-0.011
 ramp-0.011
inn-0.010
Token
Token ID26391
Feature activation+3.42e+00

Pos logit contributions
 them+0.010
mate+0.010
 wreck+0.010
inn+0.009
 tow+0.009
Neg logit contributions
ages-0.009
ience-0.009
 thoughts-0.008
bey-0.008
aul-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.228
 decisions+0.220
 dedication+0.215
 persever+0.213
aul+0.211
Neg logit contributions
 them-0.193
 row-0.156
Mike-0.153
room-0.153
 block-0.151
Token
Token ID250
Feature activation-1.70e-01

Pos logit contributions
ages+0.974
cious+0.935
 thoughts+0.923
 decisions+0.900
ita+0.891
Neg logit contributions
 them-0.539
 wreck-0.480
mate-0.426
 ramp-0.416
inn-0.411
Token the
Token ID262
Feature activation+3.53e-01

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
 polar+0.010
inn+0.009
Neg logit contributions
-0.011
ages-0.011
's-0.010
ience-0.010
cious-0.010
Token k
Token ID479
Feature activation+4.28e-01

Pos logit contributions
+0.025
ages+0.025
cious+0.024
ience+0.024
's+0.024
Neg logit contributions
 them-0.020
 wreck-0.018
mate-0.018
 polar-0.018
 tow-0.016
Tokenitty
Token ID9760
Feature activation-8.50e-01

Pos logit contributions
+0.029
's+0.028
cious+0.028
ages+0.028
ience+0.027
Neg logit contributions
 them-0.026
 wreck-0.024
mate-0.024
 polar-0.023
inn-0.022
Token was
Token ID373
Feature activation-4.88e-01

Pos logit contributions
 them+0.054
 wreck+0.049
mate+0.048
 polar+0.047
 rock+0.046
Neg logit contributions
ages-0.051
 List-0.051
aul-0.049
cious-0.049
assy-0.048
Token hung
Token ID9174
Feature activation+7.37e-01

Pos logit contributions
 them+0.067
 wreck+0.050
 his+0.050
 rock+0.048
mate+0.047
Neg logit contributions
ages-0.121
cious-0.120
ience-0.114
grand-0.114
aul-0.112
Tokenrily
Token ID28904
Feature activation-7.77e+00

Pos logit contributions
+0.054
's+0.051
cious+0.051
ages+0.050
ience+0.050
Neg logit contributions
 them-0.041
 wreck-0.038
mate-0.037
 polar-0.036
inn-0.034
Token
Token ID6184
Feature activation+2.01e-01

Pos logit contributions
 them+0.404
mate+0.345
 wreck+0.341
 polar+0.321
inn+0.303
Neg logit contributions
-0.602
cious-0.584
ages-0.584
ience-0.576
's-0.563
Token
Token ID95
Feature activation-1.61e-01

Pos logit contributions
ages+0.015
's+0.015
+0.015
ience+0.014
 listens+0.014
Neg logit contributions
 them-0.010
mate-0.009
 wreck-0.009
inn-0.009
 tow-0.008
Token
Token ID26391
Feature activation+3.42e+00

Pos logit contributions
inn+0.010
mate+0.010
 them+0.010
 wreck+0.010
 load+0.010
Neg logit contributions
ience-0.009
ages-0.008
 praying-0.008
rises-0.008
's-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.223
 decisions+0.220
 persever+0.214
 dedication+0.212
 List+0.211
Neg logit contributions
 them-0.181
 block-0.158
room-0.157
 row-0.153
 ramp-0.145
Token
Token ID250
Feature activation-1.70e-01

Pos logit contributions
ages+0.887
cious+0.887
+0.861
's+0.857
ita+0.839
Neg logit contributions
 them-0.581
 wreck-0.539
mate-0.505
inn-0.472
 polar-0.471
Token she
Token ID673
Feature activation+3.65e+00

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
 polar+0.010
inn+0.010
Neg logit contributions
-0.011
's-0.010
cious-0.010
ages-0.010
ience-0.010
Token had
Token ID550
Feature activation+1.72e+00

Pos logit contributions
cious+0.176
 List+0.165
rises+0.164
 Michelle+0.163
andra+0.161
Neg logit contributions
 them-0.261
 match-0.251
 wreck-0.243
 rock-0.234
 load-0.231
Token seen
Token ID1775
Feature activation-1.95e+00

Pos logit contributions
ages+0.121
ience+0.121
cious+0.121
+0.120
aul+0.117
Neg logit contributions
 them-0.101
 wreck-0.079
 polar-0.077
mate-0.074
 his-0.073
Token milk
Token ID7545
Feature activation+4.21e+00

Pos logit contributions
 them+0.114
 wreck+0.088
mate+0.087
 polar+0.087
 his+0.081
Neg logit contributions
ages-0.144
cious-0.144
-0.141
ience-0.140
 listens-0.134
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
 them+0.013
 wreck+0.011
mate+0.010
 his+0.009
 polar+0.009
Neg logit contributions
cious-0.019
ages-0.019
ience-0.019
aul-0.018
-0.018
Token was
Token ID373
Feature activation-4.84e-01

Pos logit contributions
 them+0.084
 wreck+0.075
mate+0.067
 match+0.067
 rock+0.066
Neg logit contributions
cious-0.104
ages-0.102
aul-0.101
 Michelle-0.098
 Denise-0.095
Token ugly
Token ID13400
Feature activation+2.49e-01

Pos logit contributions
 them+0.030
 wreck+0.027
mate+0.027
 polar+0.026
inn+0.025
Neg logit contributions
-0.033
's-0.031
cious-0.031
ages-0.031
ience-0.031
Tokenâ
Token ID22940
Feature activation+9.67e-01

Pos logit contributions
+0.019
cious+0.018
's+0.018
ages+0.018
ience+0.018
Neg logit contributions
 them-0.013
 wreck-0.012
mate-0.012
 polar-0.011
inn-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
ience+0.060
cious+0.059
ages+0.058
 praying+0.058
 Louise+0.057
Neg logit contributions
inn-0.058
mate-0.058
 them-0.057
 match-0.052
room-0.052
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.167
aul+0.163
 Michelle+0.162
 persever+0.158
 List+0.156
Neg logit contributions
 them-0.262
room-0.220
 wreck-0.219
mate-0.215
 ramp-0.214
Token
Token ID251
Feature activation+6.97e-01

Pos logit contributions
cious+0.896
+0.884
ages+0.879
's+0.851
ience+0.846
Neg logit contributions
 them-0.608
 wreck-0.544
mate-0.529
inn-0.485
 ramp-0.485
Tokenthey
Token ID9930
Feature activation-1.06e-01

Pos logit contributions
cious+0.054
ience+0.054
ages+0.054
+0.052
 Michelle+0.051
Neg logit contributions
 them-0.038
 wreck-0.029
mate-0.028
 polar-0.027
 his-0.027
Token loved
Token ID6151
Feature activation+1.14e+00

Pos logit contributions
 them+0.007
 wreck+0.006
mate+0.006
 polar+0.006
inn+0.006
Neg logit contributions
-0.007
's-0.007
cious-0.007
ience-0.006
ages-0.006
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
+0.087
's+0.084
cious+0.082
ience+0.080
ages+0.080
Neg logit contributions
 them-0.055
 wreck-0.054
mate-0.054
inn-0.053
 polar-0.052
Token just
Token ID655
Feature activation-3.57e-01

Pos logit contributions
 them+0.071
 wreck+0.061
mate+0.060
 polar+0.057
inn+0.054
Neg logit contributions
cious-0.113
ages-0.112
-0.111
ience-0.111
aul-0.108
Token high
Token ID1029
Feature activation+6.18e-01

Pos logit contributions
+0.088
's+0.081
cious+0.080
ience+0.078
ages+0.076
Neg logit contributions
 them-0.081
inn-0.070
mate-0.070
 wreck-0.066
 polar-0.063
Token rat
Token ID4227
Feature activation-3.77e-01

Pos logit contributions
+0.046
's+0.044
ience+0.042
cious+0.041
ages+0.040
Neg logit contributions
 them-0.036
mate-0.030
 polar-0.028
inn-0.028
 wreck-0.028
Token
Token ID6184
Feature activation+1.97e-01

Pos logit contributions
 them+0.020
 wreck+0.018
mate+0.018
 polar+0.017
inn+0.016
Neg logit contributions
-0.028
cious-0.027
ages-0.027
's-0.027
ience-0.027
Token
Token ID95
Feature activation-1.57e-01

Pos logit contributions
ages+0.012
's+0.012
+0.012
ience+0.011
 pays+0.011
Neg logit contributions
mate-0.012
 them-0.012
 wreck-0.012
inn-0.011
 polar-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.010
 them+0.009
 wreck+0.009
 ramp+0.009
inn+0.009
Neg logit contributions
ience-0.009
-0.008
ages-0.008
rises-0.008
bey-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.221
 decisions+0.214
 persever+0.210
 List+0.209
 dedication+0.208
Neg logit contributions
 them-0.182
 block-0.170
 snowball-0.170
 ramp-0.168
room-0.167
Token
Token ID250
Feature activation-1.63e-01

Pos logit contributions
ages+0.863
cious+0.858
+0.825
ita+0.824
's+0.824
Neg logit contributions
 them-0.602
 wreck-0.570
mate-0.526
 polar-0.507
 ramp-0.506
Token much
Token ID881
Feature activation+2.72e+00

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
 polar+0.010
inn+0.009
Neg logit contributions
-0.010
cious-0.010
's-0.010
ages-0.009
ience-0.009
Token bigger
Token ID5749
Feature activation-1.33e+00

Pos logit contributions
cious+0.189
+0.188
ages+0.185
's+0.181
ience+0.179
Neg logit contributions
 them-0.161
 wreck-0.147
mate-0.142
 polar-0.140
 tow-0.130
Token than
Token ID621
Feature activation+2.76e+00

Pos logit contributions
 them+0.071
 wreck+0.063
mate+0.063
 polar+0.061
inn+0.057
Neg logit contributions
-0.101
cious-0.097
's-0.096
ages-0.095
ience-0.095
Token the
Token ID262
Feature activation+3.57e-01

Pos logit contributions
+0.168
's+0.157
cious+0.156
ages+0.154
ience+0.153
Neg logit contributions
 them-0.187
 wreck-0.174
mate-0.170
 polar-0.169
inn-0.163
Token something
Token ID1223
Feature activation+4.36e+00

Pos logit contributions
 them+0.108
 wreck+0.093
mate+0.091
 polar+0.090
 tow+0.083
Neg logit contributions
-0.144
cious-0.139
ages-0.138
ience-0.137
's-0.136
Token tasty
Token ID25103
Feature activation+1.43e+00

Pos logit contributions
+0.300
's+0.285
cious+0.277
ages+0.270
ience+0.269
Neg logit contributions
 them-0.267
mate-0.246
 wreck-0.246
 polar-0.238
inn-0.231
Token
Token ID6184
Feature activation+1.97e-01

Pos logit contributions
+0.104
's+0.100
cious+0.098
ages+0.097
ience+0.095
Neg logit contributions
 them-0.080
 wreck-0.074
mate-0.072
 polar-0.072
inn-0.068
Token
Token ID95
Feature activation-1.57e-01

Pos logit contributions
ages+0.016
's+0.015
+0.015
ience+0.015
ita+0.015
Neg logit contributions
 wreck-0.009
mate-0.008
 them-0.008
inn-0.008
 tow-0.008
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.010
 wreck+0.010
 them+0.010
 tow+0.009
inn+0.009
Neg logit contributions
ience-0.009
ages-0.009
-0.009
 listens-0.008
cious-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.228
 dedication+0.213
 decisions+0.212
 persever+0.210
 masse+0.209
Neg logit contributions
 them-0.184
 block-0.170
 wreck-0.170
room-0.164
 row-0.162
Token
Token ID250
Feature activation-1.63e-01

Pos logit contributions
ages+0.831
+0.822
cious+0.815
ita+0.792
's+0.788
Neg logit contributions
 them-0.639
 wreck-0.596
mate-0.543
 polar-0.532
 ramp-0.519
Token a
Token ID257
Feature activation+1.26e+00

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
 polar+0.009
inn+0.009
Neg logit contributions
-0.010
's-0.010
cious-0.010
ages-0.010
ience-0.010
Token mineral
Token ID18352
Feature activation+4.09e-02

Pos logit contributions
+0.088
cious+0.083
's+0.083
ience+0.081
ages+0.080
Neg logit contributions
 them-0.079
mate-0.069
 wreck-0.068
inn-0.067
 polar-0.065
Token.
Token ID13
Feature activation+1.33e-01

Pos logit contributions
+0.003
cious+0.003
ages+0.003
's+0.003
ience+0.003
Neg logit contributions
 them-0.002
 wreck-0.002
mate-0.002
 polar-0.002
inn-0.002
Token But
Token ID887
Feature activation+1.06e+00

Pos logit contributions
ages+0.006
cious+0.005
ience+0.005
ita+0.005
's+0.005
Neg logit contributions
 them-0.011
mate-0.010
 tow-0.010
 polar-0.010
inn-0.010
Token and
Token ID290
Feature activation+5.27e-01

Pos logit contributions
+0.196
's+0.186
cious+0.184
ages+0.183
ience+0.179
Neg logit contributions
 them-0.156
 wreck-0.145
mate-0.142
 polar-0.139
inn-0.134
Token surprises
Token ID24072
Feature activation+3.37e+00

Pos logit contributions
+0.037
cious+0.037
ience+0.037
ages+0.036
aul+0.035
Neg logit contributions
 them-0.031
 wreck-0.026
 polar-0.025
mate-0.025
 tow-0.024
Token
Token ID6184
Feature activation+1.97e-01

Pos logit contributions
+0.244
's+0.233
ages+0.225
cious+0.224
ience+0.218
Neg logit contributions
 them-0.184
 wreck-0.180
 polar-0.175
mate-0.172
inn-0.168
Token
Token ID95
Feature activation-1.64e-01

Pos logit contributions
's+0.011
ages+0.011
+0.011
ience+0.011
ita+0.010
Neg logit contributions
 them-0.013
mate-0.012
 wreck-0.012
inn-0.012
 tow-0.012
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
 them+0.011
mate+0.011
 tow+0.010
inn+0.010
room+0.010
Neg logit contributions
ience-0.009
ages-0.009
aul-0.008
 praying-0.008
rises-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.197
 decisions+0.186
 persever+0.182
 masse+0.179
 Tooth+0.176
Neg logit contributions
 them-0.224
room-0.208
 block-0.194
 tow-0.190
 row-0.189
Token
Token ID250
Feature activation-1.66e-01

Pos logit contributions
+0.861
ages+0.852
cious+0.849
's+0.846
ience+0.828
Neg logit contributions
 them-0.620
 wreck-0.553
mate-0.541
 polar-0.510
inn-0.507
Token Amy
Token ID14235
Feature activation+2.53e+00

Pos logit contributions
 them+0.012
 wreck+0.011
 polar+0.011
mate+0.011
 ramp+0.010
Neg logit contributions
-0.010
cious-0.009
's-0.009
ages-0.009
ience-0.009
Token had
Token ID550
Feature activation+1.72e+00

Pos logit contributions
cious+0.185
ience+0.184
+0.184
ages+0.182
 List+0.178
Neg logit contributions
 them-0.128
mate-0.109
 wreck-0.104
 polar-0.104
 tow-0.095
Token managed
Token ID5257
Feature activation-2.67e-01

Pos logit contributions
+0.117
cious+0.112
ages+0.111
ience+0.111
's+0.111
Neg logit contributions
 them-0.107
 wreck-0.094
mate-0.092
 polar-0.091
inn-0.085
Token to
Token ID284
Feature activation+6.17e-01

Pos logit contributions
 them+0.015
 wreck+0.013
mate+0.012
 polar+0.012
inn+0.011
Neg logit contributions
-0.019
cious-0.019
ages-0.019
ience-0.019
's-0.018
Token his
Token ID465
Feature activation-4.00e+00

Pos logit contributions
 them+0.147
mate+0.133
 wreck+0.132
 polar+0.126
inn+0.122
Neg logit contributions
-0.252
's-0.242
cious-0.238
ages-0.236
ience-0.235
Token food
Token ID2057
Feature activation+3.13e+00

Pos logit contributions
 them+0.230
 wreck+0.208
 polar+0.204
mate+0.190
 match+0.183
Neg logit contributions
-0.294
ience-0.272
cious-0.271
ages-0.266
's-0.263
Token
Token ID6184
Feature activation+2.04e-01

Pos logit contributions
+0.228
's+0.219
ages+0.211
cious+0.208
ience+0.204
Neg logit contributions
 them-0.178
 wreck-0.162
mate-0.160
 polar-0.159
inn-0.153
Token
Token ID95
Feature activation-1.62e-01

Pos logit contributions
+0.014
's+0.013
ages+0.013
cious+0.013
ience+0.013
Neg logit contributions
 them-0.012
 wreck-0.011
mate-0.011
inn-0.011
 polar-0.010
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
 wreck+0.011
mate+0.011
 them+0.011
inn+0.011
 tow+0.010
Neg logit contributions
ience-0.009
cious-0.008
-0.008
ages-0.008
's-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.203
 decisions+0.185
ita+0.184
 Michelle+0.182
 persever+0.182
Neg logit contributions
 them-0.212
 wreck-0.207
room-0.198
 block-0.196
 ramp-0.194
Token
Token ID251
Feature activation+7.01e-01

Pos logit contributions
cious+0.866
ages+0.848
+0.823
's+0.816
ita+0.806
Neg logit contributions
 them-0.607
 wreck-0.579
mate-0.533
 snowball-0.513
inn-0.511
Token only
Token ID691
Feature activation-1.79e+00

Pos logit contributions
cious+0.048
ience+0.047
ages+0.046
+0.046
's+0.044
Neg logit contributions
 them-0.043
 wreck-0.037
 polar-0.036
mate-0.036
 his-0.035
Token to
Token ID284
Feature activation+6.05e-01

Pos logit contributions
 them+0.105
 wreck+0.096
 polar+0.095
mate+0.094
inn+0.089
Neg logit contributions
ages-0.118
-0.116
cious-0.114
aul-0.113
ience-0.111
Token find
Token ID1064
Feature activation-1.00e+00

Pos logit contributions
+0.047
's+0.045
cious+0.045
ages+0.044
ience+0.044
Neg logit contributions
 them-0.032
 wreck-0.028
mate-0.028
 polar-0.027
inn-0.026
Token it
Token ID340
Feature activation-1.45e+00

Pos logit contributions
 them+0.064
 wreck+0.059
mate+0.058
 polar+0.057
inn+0.055
Neg logit contributions
-0.066
's-0.062
cious-0.062
ages-0.061
ience-0.061
Token something
Token ID1223
Feature activation+4.36e+00

Pos logit contributions
 them+0.042
 wreck+0.034
mate+0.034
 polar+0.032
 his+0.031
Neg logit contributions
ages-0.047
cious-0.047
-0.047
ience-0.046
's-0.044
Token important
Token ID1593
Feature activation+4.18e+00

Pos logit contributions
+0.302
's+0.284
cious+0.273
ience+0.267
ages+0.263
Neg logit contributions
 them-0.265
mate-0.247
 wreck-0.247
 polar-0.244
inn-0.232
Token
Token ID6184
Feature activation+2.05e-01

Pos logit contributions
+0.308
ages+0.301
ience+0.300
cious+0.299
's+0.292
Neg logit contributions
 them-0.236
mate-0.205
 wreck-0.197
inn-0.184
 polar-0.182
Token
Token ID95
Feature activation-1.58e-01

Pos logit contributions
ages+0.013
's+0.013
+0.012
 pads+0.012
ita+0.012
Neg logit contributions
 them-0.012
 wreck-0.012
mate-0.012
inn-0.011
 ramp-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.010
 them+0.010
room+0.009
inn+0.009
 ramp+0.009
Neg logit contributions
ience-0.009
ages-0.009
bey-0.008
 decisions-0.008
aul-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.198
 decisions+0.192
 List+0.183
 dedication+0.182
 persever+0.181
Neg logit contributions
 them-0.214
room-0.195
 block-0.184
 row-0.178
 ramp-0.176
Token
Token ID250
Feature activation-1.67e-01

Pos logit contributions
ages+1.018
cious+1.003
 thoughts+0.987
 decisions+0.958
ita+0.957
Neg logit contributions
 them-0.473
 wreck-0.411
mate-0.359
 ramp-0.349
inn-0.344
Token the
Token ID262
Feature activation+3.52e-01

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
 polar+0.010
inn+0.010
Neg logit contributions
-0.010
's-0.010
cious-0.010
ages-0.010
ience-0.010
Token doll
Token ID3654
Feature activation-3.72e+00

Pos logit contributions
+0.026
's+0.025
cious+0.025
ages+0.024
ience+0.024
Neg logit contributions
 them-0.020
 wreck-0.018
mate-0.017
 polar-0.017
inn-0.017
Token had
Token ID550
Feature activation+1.73e+00

Pos logit contributions
 them+0.240
 wreck+0.223
mate+0.218
 polar+0.204
 match+0.201
Neg logit contributions
cious-0.219
ages-0.216
 List-0.209
aul-0.206
-0.204
Token a
Token ID257
Feature activation+1.26e+00

Pos logit contributions
+0.117
ages+0.116
cious+0.115
ience+0.114
aul+0.112
Neg logit contributions
 them-0.107
 wreck-0.092
mate-0.087
 polar-0.087
 tow-0.083
Token and
Token ID290
Feature activation+5.40e-01

Pos logit contributions
ages+0.362
ience+0.353
cious+0.351
+0.342
aul+0.340
Neg logit contributions
 them-0.189
 wreck-0.137
mate-0.127
 polar-0.127
 match-0.127
Token smiled
Token ID13541
Feature activation+2.61e+00

Pos logit contributions
ages+0.045
cious+0.042
aul+0.041
+0.040
ience+0.040
Neg logit contributions
 them-0.030
 wreck-0.023
 his-0.023
 rock-0.021
 polar-0.020
Token
Token ID6184
Feature activation+2.05e-01

Pos logit contributions
+0.206
ages+0.203
cious+0.201
ience+0.201
's+0.195
Neg logit contributions
 them-0.135
 wreck-0.110
mate-0.110
 polar-0.109
inn-0.100
Token
Token ID95
Feature activation-1.58e-01

Pos logit contributions
ages+0.012
's+0.012
 Michelle+0.011
+0.011
cious+0.011
Neg logit contributions
 them-0.013
 wreck-0.013
mate-0.012
inn-0.012
 ramp-0.012
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
 them+0.010
 wreck+0.010
mate+0.010
 tow+0.010
room+0.010
Neg logit contributions
ages-0.009
ience-0.009
cious-0.008
 pads-0.008
bey-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.247
 masse+0.237
 decisions+0.235
 Tooth+0.230
 persever+0.228
Neg logit contributions
 them-0.169
room-0.154
 block-0.151
 wreck-0.143
 ramp-0.137
Token
Token ID251
Feature activation+6.93e-01

Pos logit contributions
ages+0.890
cious+0.875
+0.841
's+0.837
ita+0.833
Neg logit contributions
 them-0.600
 wreck-0.540
mate-0.512
 tow-0.477
 ramp-0.476
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
ages+0.059
cious+0.058
ience+0.058
ccoli+0.055
ita+0.054
Neg logit contributions
 them-0.035
 his-0.026
mate-0.024
 tow-0.022
inn-0.021
Token was
Token ID373
Feature activation-4.88e-01

Pos logit contributions
 them+0.079
 wreck+0.066
 polar+0.065
mate+0.064
 his+0.062
Neg logit contributions
aul-0.104
ages-0.104
cious-0.103
 Michelle-0.100
 decisions-0.096
Token a
Token ID257
Feature activation+1.26e+00

Pos logit contributions
 them+0.028
 polar+0.023
 wreck+0.022
mate+0.022
 tow+0.022
Neg logit contributions
ages-0.037
cious-0.036
ience-0.036
-0.034
's-0.034
Token special
Token ID2041
Feature activation+2.45e+00

Pos logit contributions
+0.085
cious+0.081
's+0.081
ages+0.081
ience+0.080
Neg logit contributions
 them-0.077
 wreck-0.071
mate-0.069
 polar-0.068
inn-0.065
Token was
Token ID373
Feature activation-4.82e-01

Pos logit contributions
 them+0.081
 wreck+0.073
mate+0.071
 polar+0.070
inn+0.065
Neg logit contributions
cious-0.106
ages-0.102
aul-0.101
ience-0.099
-0.099
Token too
Token ID1165
Feature activation-3.00e+00

Pos logit contributions
 them+0.025
mate+0.023
 wreck+0.023
 polar+0.022
inn+0.021
Neg logit contributions
-0.038
's-0.036
ages-0.035
ience-0.035
cious-0.034
Token late
Token ID2739
Feature activation+3.96e+00

Pos logit contributions
 them+0.220
 wreck+0.206
 polar+0.199
inn+0.195
mate+0.192
Neg logit contributions
cious-0.166
-0.163
's-0.158
ages-0.156
ience-0.155
Tokenâ
Token ID22940
Feature activation+9.65e-01

Pos logit contributions
+0.321
cious+0.320
ages+0.318
ience+0.316
's+0.308
Neg logit contributions
 them-0.195
mate-0.158
 wreck-0.155
 polar-0.142
inn-0.141
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
cious+0.055
ience+0.053
pill+0.051
+0.050
ups+0.050
Neg logit contributions
inn-0.065
mate-0.064
 wreck-0.061
 them-0.060
 ramp-0.058
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.177
aul+0.162
ita+0.161
 Michelle+0.159
 Louise+0.156
Neg logit contributions
 them-0.247
 wreck-0.232
 ramp-0.229
mate-0.221
room-0.216
Token
Token ID251
Feature activation+6.99e-01

Pos logit contributions
cious+0.952
+0.935
ages+0.914
's+0.914
ita+0.899
Neg logit contributions
 them-0.550
 wreck-0.497
mate-0.482
 ramp-0.443
inn-0.441
Tokenthe
Token ID1169
Feature activation+1.42e+00

Pos logit contributions
cious+0.047
ience+0.046
+0.046
ages+0.044
's+0.044
Neg logit contributions
 them-0.044
 wreck-0.037
mate-0.037
 polar-0.036
inn-0.035
Token o
Token ID267
Feature activation-1.51e+00

Pos logit contributions
+0.108
cious+0.104
's+0.103
ages+0.102
ience+0.101
Neg logit contributions
 them-0.076
 wreck-0.069
mate-0.067
 polar-0.066
inn-0.062
Tokenat
Token ID265
Feature activation+3.87e+00

Pos logit contributions
mate+0.086
 wreck+0.085
 them+0.084
 polar+0.079
inn+0.079
Neg logit contributions
-0.100
ience-0.097
cious-0.096
 Michelle-0.094
's-0.092
Token had
Token ID550
Feature activation+1.73e+00

Pos logit contributions
+0.245
cious+0.238
ience+0.231
ages+0.228
's+0.227
Neg logit contributions
 them-0.250
 wreck-0.232
mate-0.227
 polar-0.222
inn-0.210
Token big
Token ID1263
Feature activation-1.36e+00

Pos logit contributions
+0.093
's+0.087
cious+0.084
ience+0.082
ages+0.080
Neg logit contributions
 them-0.075
inn-0.066
mate-0.065
 wreck-0.060
 polar-0.059
Token surprise
Token ID5975
Feature activation-3.91e+00

Pos logit contributions
 them+0.080
mate+0.072
inn+0.072
 polar+0.064
 wreck+0.062
Neg logit contributions
-0.099
's-0.094
ience-0.088
cious-0.086
 Michelle-0.084
Token
Token ID6184
Feature activation+2.01e-01

Pos logit contributions
 them+0.240
 tant+0.233
 wreck+0.231
inn+0.226
 polar+0.223
Neg logit contributions
-0.259
's-0.243
cious-0.216
 Michelle-0.207
ages-0.201
Token
Token ID95
Feature activation-1.60e-01

Pos logit contributions
ages+0.013
's+0.013
+0.013
ience+0.012
cious+0.012
Neg logit contributions
 wreck-0.012
 them-0.012
mate-0.011
 ramp-0.011
inn-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.010
 them+0.010
 wreck+0.010
 ramp+0.010
inn+0.010
Neg logit contributions
ience-0.009
ages-0.009
cious-0.008
bey-0.008
ita-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.218
 persever+0.204
 decisions+0.202
 dedication+0.199
aul+0.199
Neg logit contributions
 them-0.203
room-0.184
 wreck-0.180
 ramp-0.179
 block-0.174
Token
Token ID250
Feature activation-1.62e-01

Pos logit contributions
ages+0.800
cious+0.793
+0.793
's+0.763
ience+0.761
Neg logit contributions
 them-0.691
 wreck-0.634
mate-0.597
 polar-0.573
 ramp-0.571
Token there
Token ID612
Feature activation+7.52e-02

Pos logit contributions
 them+0.011
 wreck+0.011
mate+0.010
 polar+0.010
inn+0.010
Neg logit contributions
-0.010
cious-0.009
's-0.009
ages-0.009
ience-0.008
Token was
Token ID373
Feature activation-4.74e-01

Pos logit contributions
ages+0.005
cious+0.005
ience+0.005
+0.005
aul+0.005
Neg logit contributions
 them-0.005
 wreck-0.004
mate-0.004
 polar-0.004
 tow-0.004
Token a
Token ID257
Feature activation+1.25e+00

Pos logit contributions
 them+0.027
 polar+0.021
 wreck+0.021
mate+0.020
 his+0.020
Neg logit contributions
ages-0.036
cious-0.035
ience-0.035
aul-0.034
-0.034
Token big
Token ID1263
Feature activation-1.35e+00

Pos logit contributions
+0.092
cious+0.085
's+0.085
ience+0.082
ages+0.080
Neg logit contributions
 them-0.075
inn-0.064
mate-0.064
 wreck-0.062
 polar-0.060
Token favorite
Token ID4004
Feature activation-1.07e+00

Pos logit contributions
 them+0.055
 wreck+0.050
mate+0.049
 polar+0.048
inn+0.046
Neg logit contributions
-0.067
's-0.063
cious-0.063
ages-0.062
ience-0.062
Token cereal
Token ID33158
Feature activation-5.05e-02

Pos logit contributions
 them+0.065
mate+0.065
inn+0.064
 polar+0.062
 wreck+0.061
Neg logit contributions
's-0.073
-0.071
ience-0.058
 thoughts-0.058
aul-0.056
Token
Token ID6184
Feature activation+1.94e-01

Pos logit contributions
 them+0.003
 wreck+0.003
mate+0.003
 polar+0.003
inn+0.002
Neg logit contributions
-0.004
's-0.004
ages-0.003
cious-0.003
ience-0.003
Token
Token ID95
Feature activation-1.62e-01

Pos logit contributions
ages+0.014
's+0.014
+0.013
cious+0.013
ience+0.013
Neg logit contributions
 them-0.010
mate-0.010
 wreck-0.010
inn-0.009
 tow-0.009
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
mate+0.011
room+0.010
 them+0.010
inn+0.010
 tow+0.010
Neg logit contributions
ience-0.009
ages-0.009
 decisions-0.008
cious-0.008
 pads-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.226
 decisions+0.223
 persever+0.211
 dedication+0.207
 List+0.204
Neg logit contributions
 them-0.187
room-0.173
 block-0.164
 row-0.159
 tow-0.155
Token
Token ID251
Feature activation+6.88e-01

Pos logit contributions
ages+0.869
cious+0.867
+0.855
's+0.836
ience+0.833
Neg logit contributions
 them-0.608
 wreck-0.545
mate-0.537
 polar-0.494
inn-0.491
Token she
Token ID673
Feature activation+3.65e+00

Pos logit contributions
cious+0.051
ages+0.051
ience+0.051
ccoli+0.048
aul+0.047
Neg logit contributions
 them-0.039
 his-0.031
mate-0.030
 polar-0.029
 tow-0.029
Token ate
Token ID15063
Feature activation+1.99e+00

Pos logit contributions
cious+0.242
+0.241
ages+0.239
ience+0.235
aul+0.228
Neg logit contributions
 them-0.218
 wreck-0.195
 polar-0.192
mate-0.189
inn-0.178
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
+0.144
's+0.138
aul+0.124
ience+0.123
ages+0.122
Neg logit contributions
 wreck-0.112
 polar-0.106
mate-0.105
 ramp-0.100
 tow-0.099
Token with
Token ID351
Feature activation+1.35e+00

Pos logit contributions
 them+0.069
mate+0.057
 wreck+0.057
inn+0.053
 polar+0.052
Neg logit contributions
cious-0.117
-0.115
ience-0.115
ages-0.114
 Michelle-0.109
Token,
Token ID11
Feature activation+4.07e-01

Pos logit contributions
 them+0.055
 wreck+0.050
mate+0.049
 polar+0.047
inn+0.046
Neg logit contributions
-0.074
's-0.071
cious-0.070
ages-0.070
ience-0.069
Token pink
Token ID11398
Feature activation-2.28e+00

Pos logit contributions
ages+0.033
cious+0.032
ience+0.031
+0.031
aul+0.031
Neg logit contributions
 them-0.019
 polar-0.017
 snowball-0.016
 his-0.016
 wreck-0.016
Token scarf
Token ID36953
Feature activation-1.57e+00

Pos logit contributions
 them+0.135
 wreck+0.128
mate+0.124
inn+0.121
 polar+0.118
Neg logit contributions
-0.156
's-0.155
ience-0.147
ages-0.147
cious-0.147
Tokenâ
Token ID22940
Feature activation+9.63e-01

Pos logit contributions
 them+0.088
 wreck+0.079
mate+0.078
 polar+0.076
inn+0.072
Neg logit contributions
-0.115
cious-0.109
's-0.109
ages-0.109
ience-0.108
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
cious+0.051
ience+0.050
ages+0.049
+0.047
 Louise+0.046
Neg logit contributions
mate-0.067
inn-0.067
 them-0.065
room-0.061
 wreck-0.061
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.185
 decisions+0.177
 persever+0.172
+0.171
 List+0.170
Neg logit contributions
 them-0.236
room-0.207
 ramp-0.207
 snowball-0.205
mate-0.202
Token
Token ID251
Feature activation+6.88e-01

Pos logit contributions
+0.872
cious+0.869
ages+0.851
's+0.834
ience+0.825
Neg logit contributions
 them-0.621
mate-0.554
 wreck-0.550
 polar-0.509
inn-0.504
Tokenshe
Token ID7091
Feature activation-5.27e-01

Pos logit contributions
ience+0.058
cious+0.058
ages+0.057
ccoli+0.054
ita+0.053
Neg logit contributions
 them-0.033
 his-0.027
mate-0.025
 wreck-0.024
 tow-0.023
Token kept
Token ID4030
Feature activation-1.61e-01

Pos logit contributions
 them+0.035
 wreck+0.032
mate+0.032
 polar+0.031
inn+0.030
Neg logit contributions
-0.033
cious-0.031
ages-0.031
's-0.030
ience-0.030
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
 them+0.008
 wreck+0.007
mate+0.007
 polar+0.007
inn+0.007
Neg logit contributions
-0.013
's-0.012
cious-0.012
ages-0.012
ience-0.012
Token safe
Token ID3338
Feature activation+2.76e+00

Pos logit contributions
 them+0.075
 wreck+0.066
mate+0.065
 polar+0.064
inn+0.060
Neg logit contributions
-0.108
cious-0.107
ience-0.107
ages-0.106
aul-0.102
Token final
Token ID2457
Feature activation-1.97e-01

Pos logit contributions
 them+0.077
 wreck+0.069
mate+0.067
 polar+0.066
inn+0.062
Neg logit contributions
-0.112
cious-0.107
ages-0.107
's-0.107
ience-0.105
Token destination
Token ID10965
Feature activation-4.50e+00

Pos logit contributions
 them+0.011
mate+0.010
 wreck+0.010
 polar+0.010
inn+0.009
Neg logit contributions
-0.015
's-0.014
cious-0.014
ience-0.013
ages-0.013
Token
Token ID6184
Feature activation+1.94e-01

Pos logit contributions
 them+0.236
 wreck+0.207
mate+0.204
 polar+0.197
inn+0.187
Neg logit contributions
-0.347
cious-0.332
ages-0.330
's-0.330
ience-0.330
Token
Token ID95
Feature activation-1.62e-01

Pos logit contributions
+0.012
ages+0.012
's+0.012
cious+0.012
ience+0.011
Neg logit contributions
 them-0.013
 wreck-0.012
mate-0.012
 polar-0.011
inn-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
 them+0.011
mate+0.010
 wreck+0.010
 tow+0.010
inn+0.010
Neg logit contributions
-0.009
cious-0.009
ages-0.009
ience-0.009
aul-0.009
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.195
aul+0.190
 decisions+0.182
 Michelle+0.181
+0.180
Neg logit contributions
 them-0.228
 wreck-0.196
 tow-0.195
room-0.194
 ramp-0.194
Token
Token ID250
Feature activation-1.72e-01

Pos logit contributions
cious+0.792
+0.791
ages+0.787
's+0.769
ience+0.755
Neg logit contributions
 them-0.694
 wreck-0.633
mate-0.617
 polar-0.594
 tow-0.575
Token Lilly
Token ID35134
Feature activation-8.00e-02

Pos logit contributions
 them+0.012
mate+0.011
 wreck+0.011
 polar+0.010
inn+0.010
Neg logit contributions
-0.010
's-0.010
ages-0.010
cious-0.010
ience-0.010
Token's
Token ID338
Feature activation+9.30e+00

Pos logit contributions
 them+0.003
 wreck+0.003
mate+0.003
 polar+0.003
 rock+0.002
Neg logit contributions
cious-0.007
ages-0.006
 List-0.006
aul-0.006
 Michelle-0.006
Token favorite
Token ID4004
Feature activation-1.08e+00

Pos logit contributions
+0.650
ages+0.642
cious+0.639
ience+0.625
's+0.622
Neg logit contributions
 them-0.548
 wreck-0.496
 polar-0.476
mate-0.472
 tow-0.447
Token place
Token ID1295
Feature activation-2.34e+00

Pos logit contributions
 them+0.058
inn+0.054
 wreck+0.052
mate+0.052
 polar+0.051
Neg logit contributions
-0.081
's-0.078
ience-0.074
cious-0.074
ages-0.073
Token't
Token ID470
Feature activation-4.18e+00

Pos logit contributions
cious+0.033
ages+0.033
ience+0.032
aul+0.031
+0.031
Neg logit contributions
 them-0.019
 wreck-0.015
mate-0.015
inn-0.014
 tow-0.014
Token believe
Token ID1975
Feature activation+3.60e+00

Pos logit contributions
 them+0.316
 polar+0.302
mate+0.294
 wreck+0.294
 snowball+0.294
Neg logit contributions
-0.234
's-0.207
cious-0.197
ience-0.184
ages-0.180
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
+0.246
's+0.234
cious+0.228
ience+0.225
ages+0.217
Neg logit contributions
 wreck-0.206
 them-0.205
 polar-0.204
mate-0.194
inn-0.190
Tokenâ
Token ID22940
Feature activation+9.62e-01

Pos logit contributions
 them+0.077
mate+0.065
 wreck+0.058
inn+0.054
 match+0.054
Neg logit contributions
ages-0.113
cious-0.109
aul-0.108
ience-0.106
-0.105
Token
Token ID26391
Feature activation+3.42e+00

Pos logit contributions
ages+0.052
cious+0.051
ience+0.049
 praying+0.048
aul+0.047
Neg logit contributions
inn-0.066
mate-0.065
 them-0.064
room-0.059
 match-0.058
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.182
 List+0.177
 decisions+0.176
aul+0.173
 persever+0.166
Neg logit contributions
 them-0.235
room-0.216
mate-0.198
 ramp-0.192
 tow-0.184
Token
Token ID251
Feature activation+6.97e-01

Pos logit contributions
cious+0.958
ages+0.947
+0.935
's+0.925
ience+0.917
Neg logit contributions
 them-0.534
mate-0.464
 wreck-0.441
inn-0.427
 ramp-0.407
Tokenthis
Token ID5661
Feature activation+2.06e+00

Pos logit contributions
ience+0.060
ages+0.058
cious+0.058
ccoli+0.058
aul+0.054
Neg logit contributions
 them-0.035
 his-0.024
mate-0.023
 tow-0.020
 control-0.020
Token one
Token ID530
Feature activation+1.04e+00

Pos logit contributions
ages+0.153
cious+0.153
+0.150
aul+0.144
's+0.142
Neg logit contributions
 them-0.113
mate-0.097
 wreck-0.097
 polar-0.091
inn-0.090
Token feather
Token ID17972
Feature activation+1.43e+00

Pos logit contributions
ages+0.078
cious+0.077
+0.074
ience+0.072
aul+0.072
Neg logit contributions
 them-0.055
 wreck-0.050
mate-0.049
 polar-0.046
inn-0.045
Token seemed
Token ID3947
Feature activation+4.31e-01

Pos logit contributions
+0.097
cious+0.093
ages+0.093
ience+0.091
's+0.090
Neg logit contributions
 them-0.088
mate-0.077
 wreck-0.077
 polar-0.075
inn-0.071
Token he
Token ID339
Feature activation+5.21e-01

Pos logit contributions
+0.334
's+0.315
cious+0.300
ages+0.290
ience+0.287
Neg logit contributions
 them-0.237
mate-0.226
 wreck-0.213
 polar-0.209
inn-0.205
Token liked
Token ID8288
Feature activation-1.16e+00

Pos logit contributions
+0.032
's+0.030
cious+0.028
ience+0.027
ages+0.026
Neg logit contributions
 them-0.037
mate-0.036
 wreck-0.035
inn-0.034
 polar-0.033
Token
Token ID6184
Feature activation+1.91e-01

Pos logit contributions
 them+0.064
 wreck+0.058
mate+0.057
 polar+0.056
inn+0.054
Neg logit contributions
-0.086
's-0.082
cious-0.081
ages-0.080
ience-0.079
Token
Token ID95
Feature activation-1.64e-01

Pos logit contributions
ages+0.012
+0.012
's+0.012
ience+0.011
cious+0.011
Neg logit contributions
 them-0.012
 wreck-0.011
mate-0.011
inn-0.011
 polar-0.011
Token
Token ID26391
Feature activation+3.41e+00

Pos logit contributions
 them+0.011
 wreck+0.011
mate+0.011
inn+0.011
 ramp+0.010
Neg logit contributions
ience-0.009
ages-0.009
-0.008
's-0.008
cious-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.205
+0.194
 decisions+0.188
 Michelle+0.188
aul+0.187
Neg logit contributions
 them-0.224
 wreck-0.204
 polar-0.194
 ramp-0.194
mate-0.193
Token
Token ID250
Feature activation-1.75e-01

Pos logit contributions
ages+0.763
+0.763
cious+0.750
ience+0.731
's+0.729
Neg logit contributions
 them-0.724
 wreck-0.673
mate-0.623
 polar-0.622
inn-0.608
Token a
Token ID257
Feature activation+1.25e+00

Pos logit contributions
 them+0.012
 wreck+0.011
mate+0.011
 polar+0.011
 ramp+0.010
Neg logit contributions
-0.011
's-0.010
cious-0.010
ages-0.010
ience-0.009
Token big
Token ID1263
Feature activation-1.36e+00

Pos logit contributions
+0.096
cious+0.091
's+0.091
ience+0.089
ages+0.088
Neg logit contributions
 them-0.069
mate-0.060
 wreck-0.059
inn-0.057
 polar-0.056
Token soft
Token ID2705
Feature activation+1.15e+00

Pos logit contributions
 them+0.076
mate+0.067
inn+0.065
 wreck+0.064
 polar+0.061
Neg logit contributions
-0.102
's-0.098
cious-0.095
ience-0.094
ages-0.091
Token swing
Token ID9628
Feature activation+8.45e-02

Pos logit contributions
+0.084
's+0.081
cious+0.080
ages+0.078
ience+0.078
Neg logit contributions
 them-0.065
 wreck-0.058
mate-0.057
inn-0.054
 polar-0.054
Token special
Token ID2041
Feature activation+2.44e+00

Pos logit contributions
 them+0.153
 wreck+0.140
 polar+0.136
mate+0.134
inn+0.129
Neg logit contributions
-0.188
's-0.176
cious-0.175
ience-0.175
ages-0.174
Token thing
Token ID1517
Feature activation-1.63e-01

Pos logit contributions
+0.173
's+0.166
cious+0.164
ages+0.163
ience+0.160
Neg logit contributions
 them-0.143
 wreck-0.126
mate-0.125
 polar-0.122
inn-0.120
Token
Token ID6184
Feature activation+1.89e-01

Pos logit contributions
 them+0.010
 wreck+0.009
mate+0.009
 polar+0.009
inn+0.008
Neg logit contributions
-0.011
's-0.011
cious-0.011
ages-0.011
ience-0.010
Token
Token ID95
Feature activation-1.64e-01

Pos logit contributions
+0.013
ages+0.013
's+0.013
ience+0.012
cious+0.012
Neg logit contributions
 them-0.011
 wreck-0.010
mate-0.010
inn-0.009
 polar-0.009
Token
Token ID26391
Feature activation+3.40e+00

Pos logit contributions
 them+0.010
 wreck+0.010
mate+0.010
inn+0.010
 tow+0.009
Neg logit contributions
ience-0.010
ages-0.010
-0.010
's-0.009
cious-0.009
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.213
 decisions+0.209
+0.203
 dedication+0.201
 List+0.194
Neg logit contributions
 them-0.204
 wreck-0.183
room-0.181
 block-0.180
 tow-0.177
Token
Token ID250
Feature activation-1.76e-01

Pos logit contributions
+0.686
ages+0.656
cious+0.652
's+0.644
ience+0.640
Neg logit contributions
 them-0.815
 wreck-0.763
mate-0.735
 polar-0.733
inn-0.700
Token wings
Token ID12098
Feature activation+4.50e+00

Pos logit contributions
 them+0.012
 wreck+0.011
mate+0.011
 polar+0.011
 ramp+0.010
Neg logit contributions
-0.011
cious-0.010
's-0.010
ages-0.010
aul-0.009
Token.
Token ID13
Feature activation+1.23e-01

Pos logit contributions
+0.318
's+0.303
ience+0.272
ages+0.270
cious+0.265
Neg logit contributions
 them-0.281
mate-0.259
 polar-0.254
inn-0.252
 wreck-0.247
Token Every
Token ID3887
Feature activation-1.14e+00

Pos logit contributions
ages+0.008
cious+0.007
ience+0.007
's+0.007
+0.007
Neg logit contributions
 them-0.008
 polar-0.007
mate-0.007
 wreck-0.007
 tow-0.007
Token day
Token ID1110
Feature activation+2.28e+00

Pos logit contributions
 them+0.074
mate+0.067
 wreck+0.067
 polar+0.064
 rock+0.061
Neg logit contributions
cious-0.072
-0.072
ages-0.071
ience-0.070
aul-0.067
Token forest
Token ID8222
Feature activation+1.93e+00

Pos logit contributions
ages+0.030
cious+0.029
ience+0.028
+0.027
aul+0.026
Neg logit contributions
 wreck-0.017
 them-0.016
 polar-0.015
 rock-0.015
 tow-0.015
Token today
Token ID1909
Feature activation-6.32e-02

Pos logit contributions
+0.148
cious+0.143
ages+0.143
ience+0.143
's+0.140
Neg logit contributions
 them-0.100
 wreck-0.086
mate-0.084
 polar-0.081
 tow-0.076
Token
Token ID6184
Feature activation+1.89e-01

Pos logit contributions
 them+0.003
mate+0.003
 wreck+0.003
 polar+0.002
 tow+0.002
Neg logit contributions
ages-0.005
-0.005
ience-0.005
cious-0.005
's-0.005
Token
Token ID95
Feature activation-1.64e-01

Pos logit contributions
ages+0.012
's+0.011
 pads+0.011
ience+0.011
cious+0.011
Neg logit contributions
 them-0.012
mate-0.011
 wreck-0.011
 tow-0.011
inn-0.011
Token
Token ID26391
Feature activation+3.40e+00

Pos logit contributions
mate+0.011
 them+0.011
room+0.010
 tow+0.010
 wreck+0.010
Neg logit contributions
ience-0.009
ages-0.009
cious-0.009
aul-0.008
bey-0.008
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.248
 decisions+0.240
aul+0.233
 persever+0.226
night+0.226
Neg logit contributions
 them-0.166
room-0.145
 tow-0.143
mate-0.137
 block-0.136
Token
Token ID250
Feature activation-1.76e-01

Pos logit contributions
ages+0.791
cious+0.788
+0.783
's+0.758
ience+0.749
Neg logit contributions
 them-0.688
 wreck-0.637
mate-0.622
 polar-0.588
 tow-0.574
Token a
Token ID257
Feature activation+1.25e+00

Pos logit contributions
 them+0.012
mate+0.011
 wreck+0.011
 polar+0.011
inn+0.010
Neg logit contributions
-0.011
's-0.010
cious-0.010
ages-0.010
ience-0.010
Token small
Token ID1402
Feature activation+5.43e-02

Pos logit contributions
+0.082
ages+0.081
cious+0.080
's+0.079
ience+0.078
Neg logit contributions
 them-0.077
 wreck-0.071
mate-0.070
 polar-0.070
 tow-0.065
Token fairy
Token ID25607
Feature activation+3.96e-01

Pos logit contributions
+0.004
ages+0.004
cious+0.004
's+0.004
ience+0.004
Neg logit contributions
 them-0.003
 wreck-0.003
mate-0.003
 polar-0.003
 tow-0.003
Token who
Token ID508
Feature activation+6.62e-02

Pos logit contributions
ages+0.028
cious+0.028
+0.027
aul+0.026
ience+0.026
Neg logit contributions
 them-0.022
mate-0.020
 polar-0.019
 wreck-0.019
 tow-0.018
Token other
Token ID584
Feature activation-1.64e+00

Pos logit contributions
 them+0.079
 wreck+0.071
 polar+0.069
mate+0.068
 tow+0.064
Neg logit contributions
-0.103
cious-0.099
ages-0.098
ience-0.097
's-0.096
Token family
Token ID1641
Feature activation+2.05e+00

Pos logit contributions
 them+0.088
 polar+0.072
 wreck+0.072
 tow+0.062
 cubes+0.062
Neg logit contributions
-0.134
ages-0.129
ience-0.129
cious-0.127
aul-0.124
Token
Token ID6184
Feature activation+1.89e-01

Pos logit contributions
ience+0.161
+0.157
cious+0.156
ages+0.155
 Michelle+0.151
Neg logit contributions
 them-0.107
 wreck-0.089
mate-0.083
 polar-0.083
 tow-0.079
Token
Token ID95
Feature activation-1.64e-01

Pos logit contributions
+0.012
ages+0.012
's+0.011
ience+0.011
cious+0.011
Neg logit contributions
 them-0.012
 wreck-0.011
mate-0.011
inn-0.011
 polar-0.010
Token
Token ID26391
Feature activation+3.40e+00

Pos logit contributions
 them+0.011
 wreck+0.010
mate+0.010
inn+0.010
 tow+0.010
Neg logit contributions
ience-0.010
-0.009
ages-0.009
's-0.009
 Louise-0.009
Token
Token ID447
Feature activation+1.18e+01

Pos logit contributions
ience+0.190
+0.182
 decisions+0.178
 Michelle+0.175
 persever+0.171
Neg logit contributions
 them-0.236
room-0.205
 wreck-0.203
 tow-0.200
 row-0.199
Token
Token ID250
Feature activation-1.76e-01

Pos logit contributions
+0.627
cious+0.583
's+0.582
ages+0.581
ience+0.574
Neg logit contributions
 them-0.898
 wreck-0.833
mate-0.816
 polar-0.806
inn-0.779
Token a
Token ID257
Feature activation+1.25e+00

Pos logit contributions
 them+0.012
 wreck+0.011
mate+0.011
 polar+0.011
 ramp+0.010
Neg logit contributions
-0.011
cious-0.010
's-0.010
ages-0.009
ience-0.009
Token dad
Token ID9955
Feature activation+1.55e+00

Pos logit contributions
+0.086
's+0.082
cious+0.081
ages+0.079
ience+0.079
Neg logit contributions
 them-0.077
mate-0.069
 wreck-0.068
inn-0.067
 polar-0.064
Token,
Token ID11
Feature activation+3.84e-01

Pos logit contributions
+0.117
cious+0.116
ages+0.115
ience+0.114
's+0.110
Neg logit contributions
 them-0.080
 wreck-0.071
mate-0.069
 polar-0.068
 tow-0.063
Token a
Token ID257
Feature activation+1.25e+00

Pos logit contributions
+0.026
cious+0.025
ages+0.025
's+0.025
ience+0.025
Neg logit contributions
 them-0.023
 wreck-0.021
mate-0.021
 polar-0.020
inn-0.019
Token pieces
Token ID5207
Feature activation-8.72e-01

Pos logit contributions
+0.011
's+0.010
ience+0.010
cious+0.010
ages+0.010
Neg logit contributions
 them-0.009
 wreck-0.008
mate-0.008
 polar-0.008
inn-0.007
Token to
Token ID284
Feature activation+3.98e-01

Pos logit contributions
 them+0.048
 wreck+0.043
mate+0.042
 polar+0.042
inn+0.040
Neg logit contributions
-0.065
's-0.062
cious-0.061
ages-0.061
ience-0.061
Token try
Token ID1949
Feature activation-2.54e+00

Pos logit contributions
+0.029
's+0.026
cious+0.022
ience+0.021
 dedication+0.021
Neg logit contributions
 polar-0.023
inn-0.023
mate-0.022
 snowball-0.022
 tant-0.021
Token and
Token ID290
Feature activation+3.27e-01

Pos logit contributions
 them+0.120
 wreck+0.114
mate+0.112
 polar+0.112
inn+0.105
Neg logit contributions
-0.203
's-0.197
cious-0.190
ages-0.189
ience-0.188
Token check
Token ID2198
Feature activation-1.05e+00

Pos logit contributions
+0.027
cious+0.026
's+0.026
ages+0.026
ience+0.025
Neg logit contributions
 them-0.015
 wreck-0.014
mate-0.013
 polar-0.013
inn-0.012
Tokenmate
Token ID9830
Feature activation-1.15e+01

Pos logit contributions
 wreck+0.043
 polar+0.042
 them+0.042
mate+0.039
inn+0.037
Neg logit contributions
-0.090
's-0.087
ience-0.085
ages-0.084
cious-0.084
Token your
Token ID534
Feature activation-1.01e+00

Pos logit contributions
 them+0.620
mate+0.530
 wreck+0.523
 polar+0.494
inn+0.474
Neg logit contributions
-0.879
cious-0.842
ages-0.833
's-0.826
ience-0.825
Token opponent
Token ID6125
Feature activation+2.91e+00

Pos logit contributions
 them+0.066
 wreck+0.061
mate+0.059
 polar+0.058
inn+0.055
Neg logit contributions
-0.065
cious-0.062
ages-0.061
ience-0.061
's-0.060
Token."
Token ID526
Feature activation-2.37e+00

Pos logit contributions
+0.220
cious+0.219
ience+0.212
ages+0.210
 Michelle+0.208
Neg logit contributions
 them-0.155
 wreck-0.130
mate-0.128
 polar-0.115
 match-0.115
Token
Token ID198
Feature activation+6.97e-02

Pos logit contributions
 them+0.138
 wreck+0.121
mate+0.119
 polar+0.116
inn+0.113
Neg logit contributions
-0.170
cious-0.165
ience-0.164
ages-0.162
's-0.161
Token
Token ID198
Feature activation+5.18e-02

Pos logit contributions
+0.004
's+0.004
ages+0.004
cious+0.004
aul+0.003
Neg logit contributions
 them-0.005
 wreck-0.005
 polar-0.005
mate-0.005
 captain-0.004
Token
Token ID8151
Feature activation+5.71e+00

Pos logit contributions
 Michelle+0.134
ience+0.133
aul+0.133
 Denise+0.119
 Louise+0.119
Neg logit contributions
 them-0.273
 wreck-0.253
 ramp-0.253
mate-0.252
room-0.247
Tokens
Token ID82
Feature activation+2.26e-01

Pos logit contributions
cious+0.395
ience+0.390
 listens+0.384
ages+0.383
+0.382
Neg logit contributions
 them-0.334
mate-0.294
 wreck-0.288
inn-0.282
 polar-0.264
Token King
Token ID2677
Feature activation+3.08e+00

Pos logit contributions
+0.016
cious+0.015
ages+0.015
ience+0.015
's+0.015
Neg logit contributions
 them-0.014
 wreck-0.012
mate-0.012
 polar-0.011
inn-0.011
Token in
Token ID287
Feature activation+3.35e-01

Pos logit contributions
cious+0.133
ience+0.131
 Michelle+0.128
 List+0.128
aul+0.127
Neg logit contributions
 them-0.254
mate-0.237
 wreck-0.234
 polar-0.219
 match-0.218
Token check
Token ID2198
Feature activation-9.68e-01

Pos logit contributions
+0.024
's+0.022
 pays+0.020
ience+0.019
ages+0.019
Neg logit contributions
 polar-0.019
 them-0.019
 wreck-0.018
inn-0.018
mate-0.018
Tokenmate
Token ID9830
Feature activation-1.14e+01

Pos logit contributions
 them+0.051
 wreck+0.047
mate+0.046
 polar+0.045
inn+0.043
Neg logit contributions
-0.074
's-0.070
cious-0.069
ages-0.069
ience-0.069
Token.
Token ID13
Feature activation+1.75e-03

Pos logit contributions
 them+0.633
 wreck+0.563
mate+0.558
 polar+0.540
inn+0.517
Neg logit contributions
-0.840
cious-0.799
's-0.798
ages-0.793
ience-0.787
Token
Token ID198
Feature activation+1.62e-01

Pos logit contributions
+0.000
's+0.000
ages+0.000
cious+0.000
ience+0.000
Neg logit contributions
 them-0.000
mate-0.000
 wreck-0.000
 polar-0.000
inn-0.000
Token
Token ID198
Feature activation+1.58e-01

Pos logit contributions
+0.008
's+0.007
ages+0.006
cious+0.005
aul+0.005
Neg logit contributions
 wreck-0.014
 captain-0.013
 polar-0.013
mate-0.013
 them-0.013
TokenThe
Token ID464
Feature activation+9.59e-01

Pos logit contributions
+0.013
's+0.011
ages+0.010
cious+0.009
appointed+0.009
Neg logit contributions
 them-0.009
 polar-0.009
mate-0.008
 wreck-0.008
 detector-0.008
Token little
Token ID1310
Feature activation-2.26e-01

Pos logit contributions
cious+0.071
ages+0.071
+0.070
ience+0.068
aul+0.068
Neg logit contributions
 them-0.051
 wreck-0.049
mate-0.046
 polar-0.046
 tow-0.042
Token.
Token ID13
Feature activation+1.42e-02

Pos logit contributions
ience+0.076
+0.076
cious+0.075
ages+0.075
aul+0.071
Neg logit contributions
 them-0.048
mate-0.040
 wreck-0.039
 polar-0.039
inn-0.036
Token
Token ID198
Feature activation+1.70e-01

Pos logit contributions
ience+0.001
cious+0.001
ages+0.001
+0.001
aul+0.001
Neg logit contributions
 them-0.001
inn-0.001
mate-0.001
 his-0.001
 tow-0.001
Token
Token ID198
Feature activation+1.68e-01

Pos logit contributions
+0.011
's+0.010
cious+0.010
ages+0.010
ience+0.010
Neg logit contributions
 them-0.011
 wreck-0.010
 polar-0.010
mate-0.010
inn-0.009
Token"
Token ID1
Feature activation-8.86e-01

Pos logit contributions
ages+0.012
cious+0.012
ience+0.011
aul+0.011
 listens+0.011
Neg logit contributions
 them-0.010
mate-0.008
inn-0.008
 wreck-0.007
 tow-0.007
TokenCheck
Token ID9787
Feature activation-1.16e+00

Pos logit contributions
 them+0.046
mate+0.040
 wreck+0.039
 polar+0.038
inn+0.037
Neg logit contributions
-0.069
ages-0.068
ience-0.067
cious-0.067
's-0.067
Tokenmate
Token ID9830
Feature activation-1.14e+01

Pos logit contributions
 them+0.062
 wreck+0.060
 polar+0.058
 tow+0.055
inn+0.053
Neg logit contributions
-0.087
's-0.083
ages-0.083
cious-0.082
ience-0.081
Token!"
Token ID2474
Feature activation-4.19e+00

Pos logit contributions
 them+0.462
mate+0.394
 wreck+0.362
 polar+0.322
 control+0.305
Neg logit contributions
-0.977
cious-0.963
aul-0.941
ages-0.941
ience-0.941
Token declared
Token ID6875
Feature activation-2.36e+00

Pos logit contributions
 them+0.295
 his+0.233
 match+0.216
inn+0.215
mate+0.211
Neg logit contributions
cious-0.257
aul-0.251
ages-0.248
lette-0.244
 decisions-0.241
Token Elsa
Token ID19226
Feature activation+2.63e+00

Pos logit contributions
 them+0.142
 his+0.114
mate+0.109
 polar+0.105
inn+0.101
Neg logit contributions
cious-0.172
ages-0.172
ience-0.171
-0.169
aul-0.165
Token.
Token ID13
Feature activation-5.66e-03

Pos logit contributions
ience+0.231
aul+0.223
 Michelle+0.221
 listens+0.219
 List+0.218
Neg logit contributions
 them-0.104
 wreck-0.076
 polar-0.075
mate-0.075
 his-0.075
Token Her
Token ID2332
Feature activation+9.60e-01

Pos logit contributions
 them+0.000
mate+0.000
inn+0.000
 tow+0.000
 his+0.000
Neg logit contributions
ience-0.000
cious-0.000
ages-0.000
-0.000
 listens-0.000
Token enough
Token ID1576
Feature activation-5.99e-01

Pos logit contributions
cious+0.290
ages+0.282
+0.277
ience+0.274
aul+0.266
Neg logit contributions
 them-0.186
mate-0.147
 wreck-0.146
 polar-0.135
 control-0.129
Token to
Token ID284
Feature activation+5.95e-01

Pos logit contributions
 them+0.032
 wreck+0.027
mate+0.026
 polar+0.025
 tow+0.024
Neg logit contributions
-0.047
cious-0.047
ages-0.046
's-0.045
ience-0.045
Token find
Token ID1064
Feature activation-1.00e+00

Pos logit contributions
+0.046
's+0.044
cious+0.044
ages+0.043
ience+0.043
Neg logit contributions
 them-0.031
 wreck-0.028
mate-0.027
 polar-0.026
inn-0.025
Token your
Token ID534
Feature activation-7.92e-01

Pos logit contributions
 them+0.066
mate+0.052
 wreck+0.052
 polar+0.050
 his+0.048
Neg logit contributions
-0.070
cious-0.070
ages-0.069
ience-0.067
aul-0.065
Token soul
Token ID5848
Feature activation+6.05e+00

Pos logit contributions
 them+0.048
 wreck+0.044
 polar+0.041
mate+0.041
 rock+0.039
Neg logit contributions
-0.055
cious-0.054
ages-0.053
ience-0.052
's-0.051
Tokenmate
Token ID9830
Feature activation-1.13e+01

Pos logit contributions
cious+0.468
+0.462
ages+0.458
ience+0.452
aul+0.445
Neg logit contributions
 them-0.319
mate-0.258
 wreck-0.254
 polar-0.234
 control-0.230
Token,
Token ID11
Feature activation+3.84e-01

Pos logit contributions
 them+0.636
mate+0.538
 wreck+0.477
 his+0.457
 control+0.453
Neg logit contributions
cious-0.808
ages-0.799
-0.794
aul-0.791
ience-0.771
Token don
Token ID836
Feature activation+2.78e+00

Pos logit contributions
cious+0.027
ages+0.026
ience+0.025
aul+0.024
assy+0.024
Neg logit contributions
 them-0.026
mate-0.021
 wreck-0.021
 his-0.020
 match-0.020
Token't
Token ID470
Feature activation-4.19e+00

Pos logit contributions
cious+0.235
ages+0.230
+0.229
ience+0.226
's+0.223
Neg logit contributions
 them-0.127
 wreck-0.104
mate-0.102
 polar-0.093
inn-0.091
Token be
Token ID307
Feature activation-1.43e+00

Pos logit contributions
 them+0.232
 wreck+0.209
mate+0.202
 polar+0.198
inn+0.186
Neg logit contributions
-0.309
cious-0.299
ages-0.298
ience-0.295
's-0.294
Token afraid
Token ID7787
Feature activation+2.28e+00

Pos logit contributions
 them+0.097
 wreck+0.088
 polar+0.085
mate+0.084
inn+0.080
Neg logit contributions
-0.089
cious-0.085
ages-0.084
ience-0.084
's-0.083
Token popular
Token ID2968
Feature activation-1.66e+00

Pos logit contributions
+0.149
cious+0.141
ages+0.140
ience+0.139
's+0.137
Neg logit contributions
 them-0.088
 wreck-0.081
 polar-0.075
mate-0.072
 tow-0.069
Token that
Token ID326
Feature activation-1.12e+00

Pos logit contributions
 them+0.093
 wreck+0.083
mate+0.082
 polar+0.081
inn+0.080
Neg logit contributions
-0.120
's-0.115
cious-0.113
ages-0.112
ience-0.111
Token everyone
Token ID2506
Feature activation+2.26e+00

Pos logit contributions
 them+0.070
 wreck+0.061
mate+0.060
 polar+0.059
 tow+0.055
Neg logit contributions
cious-0.075
-0.074
ages-0.074
ience-0.073
's-0.070
Token wanted
Token ID2227
Feature activation-7.88e-01

Pos logit contributions
cious+0.147
ages+0.145
+0.144
ience+0.139
aul+0.136
Neg logit contributions
 them-0.143
 wreck-0.128
mate-0.121
 polar-0.118
 match-0.110
Token a
Token ID257
Feature activation+7.44e-01

Pos logit contributions
 them+0.044
mate+0.033
 wreck+0.033
 his+0.031
 polar+0.030
Neg logit contributions
cious-0.062
ience-0.062
ages-0.061
-0.061
aul-0.058
Token pair
Token ID5166
Feature activation-9.35e+00

Pos logit contributions
+0.055
's+0.052
cious+0.051
ience+0.050
ages+0.050
Neg logit contributions
 them-0.043
mate-0.038
 wreck-0.038
 polar-0.037
inn-0.036
Token of
Token ID286
Feature activation-3.46e+00

Pos logit contributions
 polar+0.516
 them+0.506
 wreck+0.495
 ramp+0.489
 snowball+0.488
Neg logit contributions
-0.740
's-0.682
ages-0.612
cious-0.607
ience-0.578
Token the
Token ID262
Feature activation+3.22e-02

Pos logit contributions
 them+0.205
 wreck+0.201
mate+0.199
inn+0.196
 polar+0.192
Neg logit contributions
-0.234
's-0.219
cious-0.209
ience-0.209
ages-0.208
Token same
Token ID976
Feature activation-2.01e+00

Pos logit contributions
+0.002
ages+0.002
cious+0.002
's+0.002
ience+0.002
Neg logit contributions
 them-0.002
 wreck-0.002
mate-0.002
 polar-0.002
 tow-0.002
Token.
Token ID13
Feature activation-1.04e-01

Pos logit contributions
 them+0.113
 wreck+0.103
 polar+0.101
mate+0.101
inn+0.100
Neg logit contributions
-0.148
's-0.141
cious-0.135
ience-0.133
ages-0.133
Token
Token ID220
Feature activation+1.16e+00

Pos logit contributions
 them+0.007
 wreck+0.006
mate+0.006
 polar+0.006
inn+0.006
Neg logit contributions
ages-0.007
cious-0.007
ience-0.006
-0.006
's-0.006
Token I
Token ID314
Feature activation+2.11e+00

Pos logit contributions
 them+0.044
 wreck+0.038
mate+0.038
 polar+0.038
inn+0.036
Neg logit contributions
cious-0.045
ages-0.045
ience-0.045
-0.044
's-0.043
Token could
Token ID714
Feature activation-1.51e+00

Pos logit contributions
+0.148
's+0.135
ience+0.128
ages+0.122
cious+0.122
Neg logit contributions
 them-0.134
mate-0.125
 wreck-0.123
 polar-0.121
inn-0.117
Token use
Token ID779
Feature activation-3.45e+00

Pos logit contributions
 them+0.079
mate+0.076
inn+0.074
 polar+0.073
 wreck+0.072
Neg logit contributions
-0.120
's-0.111
ages-0.101
cious-0.100
ience-0.100
Token an
Token ID281
Feature activation+1.77e+00

Pos logit contributions
mate+0.186
 them+0.185
 polar+0.184
 wreck+0.183
inn+0.175
Neg logit contributions
-0.252
's-0.237
ience-0.225
ages-0.218
cious-0.214
Token extra
Token ID3131
Feature activation-2.54e+00

Pos logit contributions
+0.122
's+0.112
ience+0.108
ages+0.104
cious+0.102
Neg logit contributions
 them-0.114
mate-0.103
 polar-0.101
 wreck-0.100
inn-0.095
Token pair
Token ID5166
Feature activation-9.28e+00

Pos logit contributions
 them+0.143
 wreck+0.128
mate+0.126
 polar+0.123
inn+0.118
Neg logit contributions
-0.187
's-0.177
cious-0.177
ages-0.176
ience-0.174
Token of
Token ID286
Feature activation-3.42e+00

Pos logit contributions
 polar+0.494
 them+0.471
 wreck+0.467
 snowball+0.448
 ramp+0.431
Neg logit contributions
-0.770
's-0.710
ages-0.646
cious-0.631
ience-0.629
Token hands
Token ID2832
Feature activation+2.77e+00

Pos logit contributions
 wreck+0.195
mate+0.190
 them+0.184
inn+0.181
 polar+0.177
Neg logit contributions
-0.247
's-0.233
ience-0.220
cious-0.219
ages-0.216
Token."
Token ID526
Feature activation-2.38e+00

Pos logit contributions
+0.204
cious+0.195
's+0.194
ages+0.193
ience+0.192
Neg logit contributions
 them-0.153
 wreck-0.136
mate-0.134
 polar-0.131
inn-0.126
Token The
Token ID383
Feature activation+1.89e+00

Pos logit contributions
 them+0.137
mate+0.110
 polar+0.109
inn+0.108
 wreck+0.107
Neg logit contributions
cious-0.174
ience-0.173
ages-0.170
-0.163
 listens-0.160
Token boy
Token ID2933
Feature activation-2.61e+00

Pos logit contributions
+0.138
's+0.131
cious+0.131
ages+0.130
ience+0.129
Neg logit contributions
 them-0.107
 wreck-0.097
mate-0.095
 polar-0.093
inn-0.088
Token and
Token ID290
Feature activation-2.86e-01

Pos logit contributions
ience+0.183
cious+0.183
+0.182
ages+0.180
aul+0.178
Neg logit contributions
 them-0.121
mate-0.102
 control-0.084
 wreck-0.083
 tow-0.080
Token make
Token ID787
Feature activation-3.48e+00

Pos logit contributions
 them+0.017
 wreck+0.015
 polar+0.014
mate+0.014
 tow+0.013
Neg logit contributions
-0.020
cious-0.020
ages-0.019
ience-0.019
's-0.019
Token it
Token ID340
Feature activation-2.19e+00

Pos logit contributions
 them+0.188
 wreck+0.170
mate+0.169
 polar+0.164
inn+0.158
Neg logit contributions
-0.260
's-0.248
cious-0.246
ages-0.245
ience-0.242
Token go
Token ID467
Feature activation+7.53e-01

Pos logit contributions
 them+0.119
mate+0.108
 wreck+0.106
 polar+0.104
inn+0.101
Neg logit contributions
-0.166
's-0.159
ages-0.153
cious-0.153
ience-0.151
Token v
Token ID410
Feature activation-4.88e+00

Pos logit contributions
+0.056
's+0.053
ience+0.049
ages+0.048
cious+0.047
Neg logit contributions
mate-0.040
 polar-0.039
 them-0.038
 wreck-0.037
 tow-0.036
Tokenroom
Token ID3823
Feature activation-9.24e+00

Pos logit contributions
 them+0.186
 polar+0.161
 wreck+0.152
mate+0.138
 tow+0.137
Neg logit contributions
ages-0.449
-0.446
's-0.437
cious-0.418
aul-0.414
Token v
Token ID410
Feature activation-4.72e+00

Pos logit contributions
 them+0.503
 wreck+0.455
mate+0.449
 polar+0.442
inn+0.423
Neg logit contributions
-0.697
's-0.667
ages-0.653
cious-0.649
ience-0.644
Tokenroom
Token ID3823
Feature activation-9.13e+00

Pos logit contributions
 them+0.208
 polar+0.187
 wreck+0.173
 tow+0.163
mate+0.159
Neg logit contributions
ages-0.409
-0.400
's-0.397
cious-0.370
aul-0.367
Token.
Token ID13
Feature activation-1.37e-01

Pos logit contributions
 them+0.502
 wreck+0.458
 polar+0.451
mate+0.451
inn+0.429
Neg logit contributions
-0.686
's-0.661
ages-0.641
cious-0.629
ience-0.626
Token When
Token ID1649
Feature activation+1.23e+00

Pos logit contributions
 them+0.009
 wreck+0.008
mate+0.008
 polar+0.007
inn+0.007
Neg logit contributions
-0.009
's-0.009
cious-0.009
ages-0.009
ience-0.009
Token they
Token ID484
Feature activation+1.62e+00

Pos logit contributions
+0.091
cious+0.087
's+0.087
ages+0.086
ience+0.086
Neg logit contributions
 them-0.069
 wreck-0.062
mate-0.061
 polar-0.059
inn-0.056
Token,
Token ID11
Feature activation-4.22e-01

Pos logit contributions
cious+0.018
ience+0.018
ages+0.017
+0.017
ita+0.017
Neg logit contributions
 them-0.012
 his-0.010
 wreck-0.009
 splash-0.009
 polar-0.009
Token v
Token ID410
Feature activation-5.25e+00

Pos logit contributions
 them+0.025
 wreck+0.022
mate+0.022
 polar+0.021
inn+0.020
Neg logit contributions
-0.030
cious-0.029
's-0.029
ages-0.029
ience-0.028
Tokenroom
Token ID3823
Feature activation-9.56e+00

Pos logit contributions
 them+0.286
 polar+0.261
 wreck+0.250
mate+0.237
 tow+0.237
Neg logit contributions
ages-0.399
's-0.386
-0.381
cious-0.367
aul-0.361
Token,
Token ID11
Feature activation-2.34e-01

Pos logit contributions
 them+0.542
mate+0.502
 wreck+0.500
 polar+0.488
inn+0.481
Neg logit contributions
-0.695
's-0.682
ages-0.651
cious-0.641
ience-0.636
Token v
Token ID410
Feature activation-4.88e+00

Pos logit contributions
 them+0.011
mate+0.011
inn+0.010
 wreck+0.010
 polar+0.010
Neg logit contributions
-0.019
's-0.018
cious-0.017
ages-0.017
ience-0.017
Tokenroom
Token ID3823
Feature activation-9.24e+00

Pos logit contributions
 them+0.205
 polar+0.194
 wreck+0.174
 tow+0.170
 paddle+0.164
Neg logit contributions
ages-0.436
's-0.416
-0.397
cious-0.392
aul-0.387
Token!
Token ID0
Feature activation-1.51e+00

Pos logit contributions
 them+0.533
 wreck+0.508
mate+0.507
 polar+0.494
inn+0.488
Neg logit contributions
-0.657
's-0.653
ages-0.617
cious-0.599
ience-0.589
Token
Token ID198
Feature activation-2.48e-02

Pos logit contributions
 them+0.102
 wreck+0.095
mate+0.093
 polar+0.091
inn+0.088
Neg logit contributions
-0.096
's-0.089
cious-0.085
ages-0.085
ience-0.083
Token
Token ID198
Feature activation+2.95e-02

Pos logit contributions
 wreck+0.002
mate+0.002
 polar+0.002
 them+0.002
 captain+0.002
Neg logit contributions
-0.001
's-0.001
 Louise-0.001
 List-0.001
cious-0.001
TokenThey
Token ID2990
Feature activation-1.40e+00

Pos logit contributions
+0.002
's+0.002
cious+0.002
ages+0.002
ience+0.002
Neg logit contributions
 them-0.002
 wreck-0.002
 polar-0.002
mate-0.002
inn-0.002
Token were
Token ID547
Feature activation-3.55e+00

Pos logit contributions
 them+0.085
mate+0.084
inn+0.080
 polar+0.079
 wreck+0.079
Neg logit contributions
-0.099
's-0.093
ience-0.085
cious-0.084
 listens-0.081
Token makes
Token ID1838
Feature activation+2.60e+00

Pos logit contributions
cious+0.025
ages+0.024
+0.023
ience+0.022
 Michelle+0.022
Neg logit contributions
 them-0.031
 wreck-0.027
 polar-0.026
mate-0.025
 tow-0.025
Token v
Token ID410
Feature activation-5.25e+00

Pos logit contributions
+0.202
cious+0.198
ages+0.193
ience+0.193
's+0.190
Neg logit contributions
 them-0.140
 wreck-0.119
mate-0.115
 polar-0.112
inn-0.103
Tokenroom
Token ID3823
Feature activation-9.56e+00

Pos logit contributions
 them+0.237
 polar+0.215
 wreck+0.209
mate+0.200
 tow+0.191
Neg logit contributions
ages-0.439
-0.435
's-0.429
cious-0.410
aul-0.398
Token-
Token ID12
Feature activation+3.72e-01

Pos logit contributions
 them+0.502
 wreck+0.451
mate+0.446
 polar+0.435
inn+0.416
Neg logit contributions
-0.736
's-0.706
ages-0.696
cious-0.694
ience-0.685
Tokenv
Token ID85
Feature activation-5.12e+00

Pos logit contributions
+0.026
ages+0.025
's+0.025
cious+0.024
ience+0.023
Neg logit contributions
 them-0.022
mate-0.020
 polar-0.020
 wreck-0.020
inn-0.019
Tokenroom
Token ID3823
Feature activation-9.24e+00

Pos logit contributions
 polar+0.187
 them+0.176
 wreck+0.174
 paddle+0.166
 tow+0.162
Neg logit contributions
's-0.470
ages-0.459
-0.432
ups-0.409
 thoughts-0.395
Token sounds
Token ID5238
Feature activation+5.42e+00

Pos logit contributions
 them+0.479
 wreck+0.430
mate+0.428
 polar+0.416
inn+0.400
Neg logit contributions
-0.718
's-0.691
ages-0.680
cious-0.676
ience-0.665
Token.
Token ID13
Feature activation-1.91e-01

Pos logit contributions
+0.390
's+0.373
ages+0.355
cious+0.350
ience+0.339
Neg logit contributions
 them-0.308
mate-0.294
 wreck-0.291
 polar-0.289
 tow-0.275
Token
Token ID198
Feature activation+2.95e-02

Pos logit contributions
 them+0.012
 wreck+0.011
mate+0.011
 polar+0.011
inn+0.010
Neg logit contributions
-0.013
cious-0.012
's-0.012
ages-0.012
ience-0.012
Token
Token ID198
Feature activation+6.27e-02

Pos logit contributions
+0.002
's+0.001
 List+0.001
ages+0.001
 Louise+0.001
Neg logit contributions
 wreck-0.002
mate-0.002
 polar-0.002
 them-0.002
 captain-0.002
Token"
Token ID1
Feature activation-9.64e-01

Pos logit contributions
+0.005
's+0.004
cious+0.004
ages+0.004
ience+0.004
Neg logit contributions
 them-0.004
 wreck-0.003
 polar-0.003
mate-0.003
inn-0.003
Token Lily
Token ID20037
Feature activation-2.93e+00

Pos logit contributions
mate+0.178
inn+0.178
 them+0.171
 polar+0.162
Mike+0.152
Neg logit contributions
's-0.306
-0.277
 thoughts-0.232
 listens-0.229
aul-0.221
Token back
Token ID736
Feature activation-1.08e+00

Pos logit contributions
 them+0.175
mate+0.148
 wreck+0.147
 polar+0.146
inn+0.136
Neg logit contributions
cious-0.197
-0.197
ience-0.192
ages-0.191
aul-0.185
Token to
Token ID284
Feature activation-1.37e-01

Pos logit contributions
 them+0.059
 wreck+0.054
mate+0.053
 polar+0.052
inn+0.050
Neg logit contributions
-0.080
's-0.076
cious-0.075
ages-0.075
ience-0.074
Token the
Token ID262
Feature activation-2.85e-01

Pos logit contributions
 them+0.006
 wreck+0.006
mate+0.006
 polar+0.005
inn+0.005
Neg logit contributions
-0.011
's-0.011
cious-0.011
ages-0.010
ience-0.010
Token play
Token ID711
Feature activation-3.26e+00

Pos logit contributions
 them+0.016
mate+0.014
 wreck+0.014
inn+0.014
 polar+0.014
Neg logit contributions
-0.021
's-0.020
cious-0.020
ages-0.019
ience-0.019
Tokenroom
Token ID3823
Feature activation-9.24e+00

Pos logit contributions
 wreck+0.183
 polar+0.182
 them+0.166
 tow+0.161
 paddle+0.160
Neg logit contributions
's-0.244
-0.235
ages-0.191
cious-0.187
ience-0.181
Token.
Token ID13
Feature activation-2.30e-01

Pos logit contributions
 them+0.516
 wreck+0.470
mate+0.465
 polar+0.457
inn+0.444
Neg logit contributions
-0.681
's-0.659
ages-0.631
ience-0.628
cious-0.626
Token She
Token ID1375
Feature activation+4.14e-01

Pos logit contributions
 them+0.016
 wreck+0.014
mate+0.014
 polar+0.014
inn+0.013
Neg logit contributions
-0.014
cious-0.013
ages-0.013
ience-0.013
's-0.013
Token made
Token ID925
Feature activation-7.67e-01

Pos logit contributions
+0.026
's+0.024
cious+0.024
ages+0.024
ience+0.024
Neg logit contributions
 them-0.028
 wreck-0.026
mate-0.025
 polar-0.025
inn-0.024
Token Ben
Token ID3932
Feature activation+9.02e-01

Pos logit contributions
mate+0.046
 wreck+0.045
 them+0.044
 polar+0.043
inn+0.043
Neg logit contributions
-0.051
's-0.050
ages-0.047
cious-0.045
ience-0.044
Token look
Token ID804
Feature activation+2.01e-01

Pos logit contributions
+0.062
cious+0.059
ience+0.059
's+0.059
ages+0.059
Neg logit contributions
 them-0.054
 wreck-0.049
mate-0.048
 polar-0.048
inn-0.045
Token Now
Token ID2735
Feature activation-9.83e-01

Pos logit contributions
 them+0.005
 wreck+0.004
mate+0.004
 tow+0.004
 polar+0.004
Neg logit contributions
cious-0.012
ience-0.012
-0.012
ages-0.012
's-0.012
Token he
Token ID339
Feature activation+2.53e-01

Pos logit contributions
 them+0.050
 wreck+0.044
mate+0.043
 polar+0.042
inn+0.039
Neg logit contributions
-0.078
cious-0.075
ages-0.075
ience-0.075
's-0.074
Token had
Token ID550
Feature activation+1.46e+00

Pos logit contributions
+0.016
cious+0.015
ages+0.015
's+0.015
ience+0.015
Neg logit contributions
 them-0.017
 wreck-0.015
mate-0.015
 polar-0.015
inn-0.014
Token a
Token ID257
Feature activation+9.86e-01

Pos logit contributions
+0.108
cious+0.103
ages+0.103
ience+0.102
's+0.102
Neg logit contributions
 them-0.083
 wreck-0.072
mate-0.070
 polar-0.070
inn-0.066
Token great
Token ID1049
Feature activation-1.61e+00

Pos logit contributions
+0.067
's+0.063
ience+0.057
cious+0.056
 pays+0.053
Neg logit contributions
 them-0.065
mate-0.060
inn-0.057
 wreck-0.056
 polar-0.055
Token pair
Token ID5166
Feature activation-9.20e+00

Pos logit contributions
 them+0.108
mate+0.099
 wreck+0.097
 polar+0.094
inn+0.090
Neg logit contributions
-0.105
's-0.100
ages-0.094
cious-0.093
ience-0.089
Token of
Token ID286
Feature activation-3.38e+00

Pos logit contributions
 them+0.494
 polar+0.469
 wreck+0.467
 snowball+0.447
 ramp+0.442
Neg logit contributions
-0.741
's-0.698
cious-0.627
ages-0.625
ience-0.592
Token pants
Token ID12581
Feature activation-1.07e+00

Pos logit contributions
mate+0.218
 wreck+0.206
 them+0.195
inn+0.195
 ramp+0.193
Neg logit contributions
-0.218
's-0.208
ages-0.185
 thoughts-0.184
ience-0.176
Token to
Token ID284
Feature activation+3.18e-01

Pos logit contributions
 them+0.062
 wreck+0.057
 polar+0.056
mate+0.056
inn+0.053
Neg logit contributions
-0.077
's-0.073
ages-0.072
cious-0.071
ience-0.069
Token show
Token ID905
Feature activation-3.75e+00

Pos logit contributions
+0.022
's+0.020
cious+0.017
ages+0.016
 Michelle+0.016
Neg logit contributions
 snowball-0.018
 wreck-0.018
inn-0.018
 tant-0.018
mate-0.018
Token off
Token ID572
Feature activation+8.09e-01

Pos logit contributions
 them+0.129
 wreck+0.118
mate+0.112
 polar+0.111
inn+0.105
Neg logit contributions
-0.349
's-0.338
ages-0.332
cious-0.331
ience-0.327
Token sure
Token ID1654
Feature activation+2.22e+00

Pos logit contributions
 them+0.040
 wreck+0.031
 polar+0.031
mate+0.031
inn+0.028
Neg logit contributions
-0.062
cious-0.061
ience-0.060
ages-0.060
's-0.058
Token to
Token ID284
Feature activation+3.40e-01

Pos logit contributions
+0.164
cious+0.163
ages+0.159
ience+0.159
's+0.154
Neg logit contributions
 them-0.129
 wreck-0.110
mate-0.107
 polar-0.102
 tow-0.099
Token keep
Token ID1394
Feature activation-1.62e+00

Pos logit contributions
+0.025
's+0.022
 Michelle+0.021
cious+0.020
ages+0.020
Neg logit contributions
 them-0.019
inn-0.018
mate-0.018
 wreck-0.017
 polar-0.017
Token the
Token ID262
Feature activation+7.70e-02

Pos logit contributions
 them+0.074
 wreck+0.060
mate+0.058
 polar+0.057
inn+0.053
Neg logit contributions
-0.140
cious-0.135
ages-0.135
ience-0.134
's-0.134
Token extra
Token ID3131
Feature activation-2.39e+00

Pos logit contributions
+0.006
cious+0.006
ages+0.006
's+0.006
ience+0.006
Neg logit contributions
 them-0.004
 wreck-0.004
mate-0.003
 polar-0.003
inn-0.003
Token pair
Token ID5166
Feature activation-9.20e+00

Pos logit contributions
 them+0.130
 wreck+0.119
 polar+0.113
mate+0.112
inn+0.105
Neg logit contributions
-0.178
cious-0.172
ience-0.171
's-0.169
ages-0.168
Token of
Token ID286
Feature activation-3.38e+00

Pos logit contributions
 them+0.444
 wreck+0.388
mate+0.384
 polar+0.368
inn+0.351
Neg logit contributions
-0.744
cious-0.715
's-0.711
ages-0.711
ience-0.706
Token pants
Token ID12581
Feature activation-1.07e+00

Pos logit contributions
mate+0.192
 wreck+0.182
inn+0.182
 them+0.182
 polar+0.170
Neg logit contributions
-0.235
's-0.221
ience-0.216
ages-0.210
cious-0.209
Token,
Token ID11
Feature activation+1.09e-01

Pos logit contributions
 them+0.059
 wreck+0.052
mate+0.051
 polar+0.049
inn+0.047
Neg logit contributions
-0.079
cious-0.076
ience-0.075
ages-0.075
's-0.075
Token just
Token ID655
Feature activation-6.54e-01

Pos logit contributions
cious+0.008
+0.008
ages+0.008
ience+0.008
aul+0.008
Neg logit contributions
 them-0.006
 wreck-0.005
 polar-0.005
mate-0.005
 tow-0.004
Token in
Token ID287
Feature activation+1.57e-01

Pos logit contributions
 them+0.032
 wreck+0.025
 his+0.024
 polar+0.024
 control+0.023
Neg logit contributions
ages-0.055
cious-0.054
ience-0.053
-0.053
aul-0.052
Token explore
Token ID7301
Feature activation+3.31e+00

Pos logit contributions
+0.029
's+0.027
cious+0.027
ages+0.027
ience+0.025
Neg logit contributions
 them-0.016
mate-0.014
 wreck-0.014
inn-0.013
 polar-0.013
Token.
Token ID13
Feature activation-1.47e-01

Pos logit contributions
+0.252
's+0.241
cious+0.236
ages+0.234
ience+0.233
Neg logit contributions
 them-0.176
 wreck-0.160
mate-0.159
 polar-0.155
inn-0.149
Token She
Token ID1375
Feature activation+4.58e-01

Pos logit contributions
 them+0.008
 wreck+0.007
mate+0.007
 tow+0.007
inn+0.007
Neg logit contributions
ience-0.011
ages-0.011
cious-0.011
-0.011
's-0.010
Token had
Token ID550
Feature activation+1.45e+00

Pos logit contributions
+0.031
ages+0.030
cious+0.030
ience+0.029
's+0.029
Neg logit contributions
 them-0.028
 wreck-0.025
 polar-0.025
mate-0.024
inn-0.023
Token a
Token ID257
Feature activation+9.82e-01

Pos logit contributions
ience+0.144
aul+0.129
ccoli+0.128
ages+0.127
ica+0.127
Neg logit contributions
 them-0.064
 his-0.047
 block-0.039
 snowball-0.037
 lot-0.037
Token pair
Token ID5166
Feature activation-9.20e+00

Pos logit contributions
+0.076
's+0.073
cious+0.072
ages+0.070
ience+0.070
Neg logit contributions
 them-0.053
mate-0.046
 wreck-0.045
 polar-0.043
inn-0.043
Token of
Token ID286
Feature activation-3.39e+00

Pos logit contributions
 them+0.457
 polar+0.441
 wreck+0.435
 ramp+0.406
 snowball+0.403
Neg logit contributions
-0.768
's-0.734
cious-0.681
ages-0.678
ience-0.627
Token sunglasses
Token ID36251
Feature activation+2.63e+00

Pos logit contributions
 wreck+0.214
 them+0.207
mate+0.207
 polar+0.198
 ramp+0.193
Neg logit contributions
-0.213
's-0.210
ages-0.198
cious-0.196
 thoughts-0.186
Token with
Token ID351
Feature activation+1.06e+00

Pos logit contributions
+0.202
cious+0.197
ience+0.197
ages+0.193
's+0.189
Neg logit contributions
 them-0.138
 wreck-0.119
mate-0.118
 polar-0.110
inn-0.109
Token her
Token ID607
Feature activation-1.23e+00

Pos logit contributions
+0.072
's+0.069
cious+0.065
ages+0.062
ience+0.058
Neg logit contributions
 wreck-0.066
mate-0.063
 ramp-0.063
 polar-0.060
inn-0.059
Token to
Token ID284
Feature activation+2.99e-01

Pos logit contributions
 them+0.077
 wreck+0.072
mate+0.070
 polar+0.070
inn+0.067
Neg logit contributions
-0.083
's-0.081
cious-0.078
ages-0.078
ience-0.075
Token girl
Token ID2576
Feature activation-1.54e+00

Pos logit contributions
+0.005
's+0.005
cious+0.005
ages+0.005
ience+0.005
Neg logit contributions
 them-0.005
 wreck-0.005
mate-0.005
 polar-0.005
inn-0.005
Token saw
Token ID2497
Feature activation-3.82e+00

Pos logit contributions
 them+0.042
 row+0.031
 block+0.028
 load+0.025
 stuffed+0.025
Neg logit contributions
 decisions-0.158
ience-0.148
ages-0.147
 List-0.146
 prayers-0.145
Token a
Token ID257
Feature activation+9.85e-01

Pos logit contributions
 them+0.240
 wreck+0.227
mate+0.227
 polar+0.218
inn+0.216
Neg logit contributions
-0.251
's-0.240
cious-0.234
ages-0.233
ience-0.229
Token bright
Token ID6016
Feature activation+1.36e+00

Pos logit contributions
+0.069
's+0.065
cious+0.061
ience+0.059
 poem+0.056
Neg logit contributions
 them-0.063
inn-0.056
mate-0.054
 control-0.050
 polar-0.049
Token orange
Token ID10912
Feature activation-3.21e-01

Pos logit contributions
+0.093
's+0.091
ience+0.079
 Michelle+0.078
cious+0.076
Neg logit contributions
 them-0.082
mate-0.078
inn-0.072
 wreck-0.067
 polar-0.067
Token pair
Token ID5166
Feature activation-9.20e+00

Pos logit contributions
 them+0.019
inn+0.016
 polar+0.016
mate+0.016
 wreck+0.016
Neg logit contributions
-0.023
's-0.022
cious-0.020
ages-0.020
ience-0.019
Token of
Token ID286
Feature activation-3.39e+00

Pos logit contributions
 them+0.498
 polar+0.492
 snowball+0.470
 wreck+0.468
 ramp+0.455
Neg logit contributions
-0.759
's-0.711
cious-0.622
ages-0.581
ience-0.568
Token shoes
Token ID10012
Feature activation-1.45e+00

Pos logit contributions
mate+0.233
 wreck+0.221
inn+0.215
 ramp+0.208
 them+0.203
Neg logit contributions
-0.209
's-0.200
cious-0.181
ience-0.177
 thoughts-0.166
Token that
Token ID326
Feature activation-5.38e-01

Pos logit contributions
 them+0.086
mate+0.084
 polar+0.083
 wreck+0.081
 snowball+0.078
Neg logit contributions
-0.103
's-0.099
cious-0.091
ages-0.090
ience-0.088
Token she
Token ID673
Feature activation+3.36e+00

Pos logit contributions
 them+0.033
 wreck+0.026
mate+0.026
 his+0.026
 match+0.026
Neg logit contributions
ience-0.039
ages-0.037
cious-0.037
-0.037
 decisions-0.036
Token liked
Token ID8288
Feature activation-1.47e+00

Pos logit contributions
+0.203
's+0.192
cious+0.184
ience+0.180
ages+0.179
Neg logit contributions
 them-0.241
mate-0.225
 wreck-0.225
 polar-0.220
 tow-0.208
Token around
Token ID1088
Feature activation+2.66e+00

Pos logit contributions
+0.009
's+0.009
cious+0.009
ages+0.008
ience+0.008
Neg logit contributions
 them-0.009
 wreck-0.008
mate-0.008
 polar-0.008
inn-0.008
Token.
Token ID13
Feature activation-1.38e-01

Pos logit contributions
+0.146
's+0.141
 Louise+0.118
 thoughts+0.116
 Michelle+0.116
Neg logit contributions
 wreck-0.189
 polar-0.188
mate-0.182
 ramp-0.180
 tow-0.173
Token They
Token ID1119
Feature activation-5.03e-01

Pos logit contributions
 them+0.008
 wreck+0.007
mate+0.007
 polar+0.007
inn+0.007
Neg logit contributions
-0.010
's-0.009
cious-0.009
ages-0.009
ience-0.009
Token see
Token ID766
Feature activation-3.38e+00

Pos logit contributions
 them+0.026
mate+0.024
 polar+0.023
 wreck+0.023
inn+0.022
Neg logit contributions
-0.040
's-0.038
cious-0.037
ience-0.036
ages-0.036
Token a
Token ID257
Feature activation+9.86e-01

Pos logit contributions
 them+0.212
mate+0.199
 wreck+0.197
 polar+0.192
inn+0.191
Neg logit contributions
-0.223
's-0.212
cious-0.205
ages-0.205
ience-0.202
Token pair
Token ID5166
Feature activation-9.19e+00

Pos logit contributions
+0.063
's+0.062
cious+0.049
 poem+0.048
 praying+0.048
Neg logit contributions
inn-0.069
 them-0.064
mate-0.059
ases-0.059
ir-0.056
Token of
Token ID286
Feature activation-3.38e+00

Pos logit contributions
 polar+0.601
 snowball+0.534
 wreck+0.514
 ramp+0.505
Ar+0.505
Neg logit contributions
-0.753
's-0.683
cious-0.501
ience-0.487
 listens-0.486
Token scissors
Token ID41786
Feature activation-3.45e+00

Pos logit contributions
mate+0.201
inn+0.197
 wreck+0.170
 them+0.167
 polar+0.163
Neg logit contributions
's-0.225
-0.224
ages-0.196
ience-0.196
 thoughts-0.192
Token on
Token ID319
Feature activation-5.49e-01

Pos logit contributions
 polar+0.206
 them+0.205
 wreck+0.199
mate+0.199
inn+0.189
Neg logit contributions
-0.237
's-0.231
cious-0.207
ages-0.206
ience-0.201
Token the
Token ID262
Feature activation+5.83e-02

Pos logit contributions
 polar+0.037
 wreck+0.036
mate+0.035
inn+0.033
 splash+0.031
Neg logit contributions
-0.033
's-0.031
 Michelle-0.028
 Denise-0.026
 Louise-0.026
Token table
Token ID3084
Feature activation-3.35e+00

Pos logit contributions
+0.004
's+0.004
cious+0.004
ages+0.003
ience+0.003
Neg logit contributions
 them-0.004
inn-0.003
mate-0.003
 polar-0.003
 wreck-0.003
Token a
Token ID257
Feature activation+9.98e-01

Pos logit contributions
 wreck+0.053
 polar+0.046
mate+0.043
 tow+0.037
 captain+0.036
Neg logit contributions
-0.139
's-0.137
ience-0.123
cious-0.122
 pays-0.122
Token shirt
Token ID10147
Feature activation-4.28e+00

Pos logit contributions
+0.075
's+0.073
cious+0.065
ages+0.063
ience+0.062
Neg logit contributions
 them-0.058
mate-0.054
 wreck-0.053
 polar-0.052
inn-0.050
Token,
Token ID11
Feature activation+1.36e-01

Pos logit contributions
 them+0.253
 wreck+0.241
 polar+0.237
mate+0.230
 tow+0.214
Neg logit contributions
-0.304
's-0.292
ages-0.276
cious-0.273
ience-0.265
Token then
Token ID788
Feature activation-1.70e+00

Pos logit contributions
ience+0.011
+0.010
ages+0.010
cious+0.010
aul+0.010
Neg logit contributions
 them-0.007
 polar-0.006
 his-0.005
 match-0.005
 wreck-0.005
Token a
Token ID257
Feature activation+9.94e-01

Pos logit contributions
 them+0.088
 wreck+0.071
mate+0.068
 polar+0.066
 match+0.066
Neg logit contributions
cious-0.133
ages-0.131
ience-0.130
-0.130
aul-0.128
Token pair
Token ID5166
Feature activation-9.19e+00

Pos logit contributions
+0.077
's+0.072
cious+0.067
ience+0.065
ages+0.064
Neg logit contributions
 them-0.058
inn-0.049
mate-0.049
 wreck-0.048
 polar-0.046
Token of
Token ID286
Feature activation-3.37e+00

Pos logit contributions
 polar+0.612
 wreck+0.578
 snowball+0.538
 them+0.534
 ramp+0.513
Neg logit contributions
-0.739
's-0.648
cious-0.499
ience-0.489
ages-0.484
Token pants
Token ID12581
Feature activation-1.07e+00

Pos logit contributions
mate+0.191
 them+0.186
 wreck+0.186
inn+0.177
 polar+0.171
Neg logit contributions
-0.242
's-0.228
ages-0.213
ience-0.213
cious-0.213
Token,
Token ID11
Feature activation+1.23e-01

Pos logit contributions
 them+0.062
 wreck+0.057
 polar+0.056
mate+0.056
inn+0.052
Neg logit contributions
-0.077
's-0.074
ages-0.071
cious-0.071
ience-0.070
Token then
Token ID788
Feature activation-1.71e+00

Pos logit contributions
cious+0.010
ages+0.010
ience+0.010
+0.010
aul+0.009
Neg logit contributions
 them-0.007
 polar-0.005
 wreck-0.005
mate-0.005
 his-0.005
Token a
Token ID257
Feature activation+9.67e-01

Pos logit contributions
 them+0.093
 wreck+0.079
mate+0.077
 polar+0.075
 tow+0.072
Neg logit contributions
cious-0.127
-0.127
ages-0.124
ience-0.124
aul-0.121
Token sque
Token ID35776
Feature activation-1.79e+00

Pos logit contributions
+0.067
's+0.064
cious+0.058
 listens+0.055
ience+0.055
Neg logit contributions
inn-0.064
 them-0.062
mate-0.061
 cubes-0.055
 wreck-0.054
Tokenak
Token ID461
Feature activation-1.96e-01

Pos logit contributions
 them+0.129
 wreck+0.117
 polar+0.117
mate+0.117
 tow+0.114
Neg logit contributions
-0.101
's-0.098
ages-0.091
ita-0.088
 listens-0.088
Token and
Token ID290
Feature activation+2.73e-01

Pos logit contributions
 them+0.012
mate+0.012
inn+0.011
 wreck+0.011
 polar+0.011
Neg logit contributions
-0.013
's-0.013
cious-0.011
ages-0.011
 listens-0.011
Token saw
Token ID2497
Feature activation-3.82e+00

Pos logit contributions
+0.019
's+0.018
cious+0.018
ages+0.018
ience+0.017
Neg logit contributions
 them-0.016
mate-0.015
 wreck-0.015
inn-0.014
 polar-0.014
Token a
Token ID257
Feature activation+9.94e-01

Pos logit contributions
 them+0.220
mate+0.214
 wreck+0.211
inn+0.207
 polar+0.202
Neg logit contributions
-0.258
's-0.253
cious-0.245
ages-0.243
ience-0.239
Token pair
Token ID5166
Feature activation-9.19e+00

Pos logit contributions
+0.065
's+0.062
cious+0.058
 poem+0.054
 praying+0.054
Neg logit contributions
 them-0.065
inn-0.065
mate-0.056
ases-0.054
room-0.050
Token of
Token ID286
Feature activation-3.37e+00

Pos logit contributions
 polar+0.566
Ar+0.558
 tant+0.548
 them+0.546
 snowball+0.545
Neg logit contributions
-0.692
's-0.634
cious-0.474
adow-0.405
ience-0.401
Token bright
Token ID6016
Feature activation+1.36e+00

Pos logit contributions
mate+0.199
inn+0.197
 them+0.192
 wreck+0.182
 tow+0.171
Neg logit contributions
-0.224
's-0.213
cious-0.197
ages-0.196
ience-0.196
Token eyes
Token ID2951
Feature activation-2.47e-01

Pos logit contributions
's+0.083
+0.079
 thoughts+0.071
 Michelle+0.066
 decisions+0.065
Neg logit contributions
mate-0.089
inn-0.082
 them-0.077
 wreck-0.071
Ar-0.069
Token staring
Token ID16143
Feature activation+2.20e+00

Pos logit contributions
 them+0.013
mate+0.012
 wreck+0.012
 polar+0.012
inn+0.011
Neg logit contributions
-0.019
's-0.018
cious-0.017
ages-0.017
ience-0.017
Token at
Token ID379
Feature activation+9.72e-01

Pos logit contributions
+0.152
ages+0.144
cious+0.144
ience+0.143
's+0.143
Neg logit contributions
 them-0.134
 wreck-0.119
mate-0.116
 polar-0.116
inn-0.111
Token wardrobe
Token ID35593
Feature activation-9.87e-01

Pos logit contributions
+0.671
ience+0.643
cious+0.635
ages+0.633
's+0.614
Neg logit contributions
 them-0.513
 wreck-0.469
 polar-0.447
mate-0.439
 tow-0.416
Token and
Token ID290
Feature activation+2.73e-01

Pos logit contributions
 them+0.047
 wreck+0.042
mate+0.041
 load+0.037
 polar+0.037
Neg logit contributions
-0.080
ience-0.079
cious-0.078
ages-0.075
 Michelle-0.073
Token she
Token ID673
Feature activation+3.39e+00

Pos logit contributions
ience+0.017
+0.017
cious+0.017
ages+0.017
aul+0.016
Neg logit contributions
 them-0.019
 wreck-0.017
 polar-0.016
 his-0.016
mate-0.016
Token found
Token ID1043
Feature activation-2.16e+00

Pos logit contributions
+0.246
cious+0.242
ages+0.241
ience+0.239
aul+0.232
Neg logit contributions
 them-0.190
 wreck-0.169
 polar-0.163
mate-0.159
 tow-0.148
Token a
Token ID257
Feature activation+9.97e-01

Pos logit contributions
 them+0.135
 wreck+0.121
mate+0.120
 polar+0.118
inn+0.112
Neg logit contributions
-0.148
cious-0.140
's-0.139
ages-0.139
ience-0.138
Token pair
Token ID5166
Feature activation-9.18e+00

Pos logit contributions
+0.071
's+0.067
cious+0.063
ience+0.060
ages+0.060
Neg logit contributions
 them-0.062
inn-0.055
mate-0.053
 polar-0.049
 wreck-0.049
Token of
Token ID286
Feature activation-3.37e+00

Pos logit contributions
 them+0.499
 polar+0.492
 snowball+0.482
 wreck+0.472
 ramp+0.470
Neg logit contributions
-0.755
's-0.699
cious-0.608
ages-0.584
ience-0.549
Token sunglasses
Token ID36251
Feature activation+2.66e+00

Pos logit contributions
 wreck+0.208
mate+0.201
 them+0.196
inn+0.190
 tow+0.186
Neg logit contributions
-0.227
's-0.219
ience-0.198
cious-0.196
ages-0.192
Token.
Token ID13
Feature activation-1.47e-01

Pos logit contributions
+0.198
's+0.189
cious+0.185
ages+0.185
ience+0.182
Neg logit contributions
 them-0.147
 wreck-0.133
mate-0.131
 polar-0.129
inn-0.122
Token She
Token ID1375
Feature activation+4.58e-01

Pos logit contributions
 them+0.010
 polar+0.009
 wreck+0.009
mate+0.009
inn+0.009
Neg logit contributions
-0.009
ience-0.009
cious-0.008
ages-0.008
's-0.008
Token had
Token ID550
Feature activation+1.45e+00

Pos logit contributions
+0.029
's+0.028
cious+0.026
ages+0.026
ience+0.026
Neg logit contributions
 them-0.031
mate-0.029
 wreck-0.029
 polar-0.028
inn-0.027
Token day
Token ID1110
Feature activation+1.66e+00

Pos logit contributions
 them+0.076
 wreck+0.072
 polar+0.069
 rock+0.066
 snowball+0.063
Neg logit contributions
ages-0.093
cious-0.093
ience-0.092
-0.088
 Michelle-0.087
Token,
Token ID11
Feature activation-1.14e-01

Pos logit contributions
+0.127
ience+0.124
cious+0.124
ages+0.124
's+0.122
Neg logit contributions
 them-0.088
mate-0.075
 wreck-0.074
 polar-0.073
inn-0.069
Token Alice
Token ID14862
Feature activation+2.26e+00

Pos logit contributions
 them+0.007
 wreck+0.007
mate+0.007
 polar+0.006
inn+0.006
Neg logit contributions
-0.008
's-0.007
cious-0.007
ages-0.007
ience-0.007
Token designed
Token ID3562
Feature activation-2.64e+00

Pos logit contributions
ages+0.174
cious+0.169
 List+0.168
 decisions+0.167
ience+0.166
Neg logit contributions
 them-0.116
 polar-0.096
 wreck-0.093
 rock-0.086
 row-0.085
Token a
Token ID257
Feature activation+9.41e-01

Pos logit contributions
 them+0.166
 wreck+0.144
mate+0.141
 polar+0.138
inn+0.133
Neg logit contributions
-0.181
cious-0.175
ages-0.174
ience-0.174
's-0.171
Token pair
Token ID5166
Feature activation-9.18e+00

Pos logit contributions
+0.066
's+0.063
cious+0.058
ience+0.057
ages+0.056
Neg logit contributions
 them-0.059
inn-0.052
mate-0.052
 polar-0.049
 wreck-0.048
Token of
Token ID286
Feature activation-3.33e+00

Pos logit contributions
 polar+0.594
 snowball+0.556
 them+0.552
 ramp+0.549
 wreck+0.528
Neg logit contributions
-0.718
's-0.651
ages-0.520
cious-0.506
grand-0.468
Token wings
Token ID12098
Feature activation+4.31e+00

Pos logit contributions
 them+0.188
 wreck+0.186
mate+0.182
inn+0.178
 polar+0.173
Neg logit contributions
-0.233
's-0.223
cious-0.213
ages-0.211
ience-0.206
Token for
Token ID329
Feature activation-2.62e+00

Pos logit contributions
+0.311
's+0.298
ages+0.288
cious+0.285
ience+0.281
Neg logit contributions
 them-0.246
 wreck-0.229
mate-0.225
 polar-0.223
inn-0.210
Token Bob
Token ID5811
Feature activation+2.86e+00

Pos logit contributions
 them+0.150
 wreck+0.144
mate+0.141
 polar+0.138
inn+0.134
Neg logit contributions
-0.182
's-0.173
cious-0.173
ages-0.170
ience-0.168
Token.
Token ID13
Feature activation-1.76e-03

Pos logit contributions
ience+0.200
+0.196
 List+0.194
cious+0.191
 listens+0.190
Neg logit contributions
 them-0.169
 polar-0.140
 wreck-0.137
mate-0.137
 his-0.128
Token They
Token ID1119
Feature activation-4.77e-01

Pos logit contributions
 them+0.008
 wreck+0.006
mate+0.006
 tow+0.006
inn+0.006
Neg logit contributions
cious-0.008
ience-0.008
ages-0.008
-0.008
's-0.008
Token were
Token ID547
Feature activation-3.60e+00

Pos logit contributions
 them+0.030
 wreck+0.027
mate+0.027
 polar+0.026
inn+0.025
Neg logit contributions
-0.032
's-0.031
cious-0.030
ages-0.030
ience-0.030
Token quite
Token ID2407
Feature activation-2.73e+00

Pos logit contributions
 them+0.207
 polar+0.176
 wreck+0.174
mate+0.170
 tow+0.162
Neg logit contributions
-0.262
cious-0.256
ience-0.256
ages-0.256
's-0.246
Token a
Token ID257
Feature activation+9.95e-01

Pos logit contributions
 them+0.178
 polar+0.161
 wreck+0.157
 tow+0.146
 his+0.145
Neg logit contributions
ages-0.173
ience-0.172
-0.171
cious-0.169
aul-0.168
Token charming
Token ID23332
Feature activation-6.89e-01

Pos logit contributions
+0.070
ages+0.067
cious+0.067
's+0.066
ience+0.065
Neg logit contributions
 them-0.058
 wreck-0.053
mate-0.052
 polar-0.051
inn-0.048
Token pair
Token ID5166
Feature activation-9.18e+00

Pos logit contributions
 them+0.034
mate+0.030
 wreck+0.029
inn+0.028
 polar+0.027
Neg logit contributions
-0.055
's-0.053
cious-0.052
ience-0.051
ages-0.050
Token and
Token ID290
Feature activation+2.75e-01

Pos logit contributions
 them+0.398
 wreck+0.352
 polar+0.333
mate+0.330
inn+0.310
Neg logit contributions
-0.807
's-0.773
cious-0.756
ages-0.749
ience-0.740
Token they
Token ID484
Feature activation+1.56e+00

Pos logit contributions
+0.018
ience+0.018
aul+0.018
ages+0.018
cious+0.018
Neg logit contributions
 them-0.017
 polar-0.014
mate-0.014
 wreck-0.014
 his-0.013
Token always
Token ID1464
Feature activation-1.07e+00

Pos logit contributions
+0.102
's+0.097
ience+0.092
cious+0.092
ages+0.091
Neg logit contributions
 them-0.100
 wreck-0.094
mate-0.093
 polar-0.090
inn-0.090
Token made
Token ID925
Feature activation-7.69e-01

Pos logit contributions
 them+0.070
 wreck+0.064
 polar+0.063
mate+0.061
 tow+0.057
Neg logit contributions
-0.067
ages-0.063
ience-0.063
cious-0.063
aul-0.062
Token everyone
Token ID2506
Feature activation+2.73e+00

Pos logit contributions
 them+0.050
 wreck+0.038
mate+0.038
 polar+0.038
 his+0.037
Neg logit contributions
-0.055
cious-0.053
ience-0.053
ages-0.051
aul-0.051
Tokenjoy
Token ID2633
Feature activation-5.90e-01

Pos logit contributions
+0.211
's+0.198
cious+0.191
ience+0.188
ages+0.188
Neg logit contributions
 them-0.193
 wreck-0.184
mate-0.180
 polar-0.177
inn-0.171
Tokened
Token ID276
Feature activation+3.05e+00

Pos logit contributions
 them+0.041
mate+0.038
inn+0.036
 his+0.035
 control+0.033
Neg logit contributions
ience-0.033
cious-0.032
 Michelle-0.032
-0.030
 Louise-0.030
Token.
Token ID13
Feature activation+1.24e-01

Pos logit contributions
cious+0.241
+0.238
ages+0.235
ience+0.235
's+0.229
Neg logit contributions
 them-0.155
mate-0.127
 wreck-0.124
 polar-0.121
inn-0.115
Token He
Token ID679
Feature activation-2.00e+00

Pos logit contributions
cious+0.011
+0.011
ages+0.010
ience+0.010
's+0.010
Neg logit contributions
 them-0.005
 wreck-0.004
 polar-0.004
mate-0.004
inn-0.004
Token had
Token ID550
Feature activation+1.72e+00

Pos logit contributions
 them+0.126
 wreck+0.113
 polar+0.111
mate+0.109
inn+0.105
Neg logit contributions
cious-0.126
-0.125
ages-0.124
ience-0.119
's-0.119
Token conquered
Token ID29346
Feature activation+7.77e+00

Pos logit contributions
+0.108
cious+0.108
ages+0.107
ience+0.107
aul+0.103
Neg logit contributions
 them-0.113
 wreck-0.100
mate-0.096
 polar-0.094
 tow-0.091
Token something
Token ID1223
Feature activation+4.36e+00

Pos logit contributions
+0.683
's+0.653
cious+0.625
ience+0.622
ages+0.619
Neg logit contributions
 wreck-0.294
 them-0.288
mate-0.278
 polar-0.271
inn-0.258
Token dangerous
Token ID4923
Feature activation+2.54e+00

Pos logit contributions
+0.307
cious+0.299
ages+0.298
's+0.293
ience+0.292
Neg logit contributions
 them-0.255
 wreck-0.228
mate-0.221
 polar-0.216
inn-0.207
Token and
Token ID290
Feature activation+5.28e-01

Pos logit contributions
+0.193
cious+0.189
ience+0.187
ages+0.186
's+0.183
Neg logit contributions
 them-0.136
 wreck-0.117
mate-0.117
 polar-0.109
inn-0.104
Token he
Token ID339
Feature activation+5.18e-01

Pos logit contributions
cious+0.041
aul+0.040
ages+0.040
ience+0.039
ita+0.039
Neg logit contributions
 them-0.028
 wreck-0.024
 his-0.023
 polar-0.022
mate-0.021
Token was
Token ID373
Feature activation-4.90e-01

Pos logit contributions
cious+0.033
+0.032
ages+0.032
aul+0.031
 Michelle+0.031
Neg logit contributions
 them-0.033
 wreck-0.030
 polar-0.028
mate-0.028
 tow-0.026
Token One
Token ID1881
Feature activation-5.95e-01

Pos logit contributions
+0.006
's+0.006
aul+0.005
 List+0.005
ages+0.005
Neg logit contributions
mate-0.007
 them-0.007
 wreck-0.006
 polar-0.006
 control-0.006
Token day
Token ID1110
Feature activation+2.26e+00

Pos logit contributions
 them+0.041
 wreck+0.038
 polar+0.037
mate+0.037
inn+0.035
Neg logit contributions
-0.036
cious-0.035
ience-0.035
ages-0.034
's-0.033
Token,
Token ID11
Feature activation+3.59e-01

Pos logit contributions
+0.149
's+0.145
ages+0.137
cious+0.135
ience+0.131
Neg logit contributions
 them-0.142
mate-0.131
 wreck-0.131
 polar-0.128
inn-0.122
Token Tim
Token ID5045
Feature activation-1.97e+00

Pos logit contributions
's+0.023
+0.023
 Michelle+0.019
ages+0.019
 thoughts+0.019
Neg logit contributions
mate-0.021
inn-0.020
 detector-0.019
 polar-0.019
 them-0.019
Tokenmy
Token ID1820
Feature activation+3.30e+00

Pos logit contributions
 them+0.147
 wreck+0.136
mate+0.135
 polar+0.132
inn+0.128
Neg logit contributions
-0.109
's-0.103
cious-0.101
ages-0.101
ience-0.099
Token's
Token ID338
Feature activation+9.26e+00

Pos logit contributions
cious+0.156
ience+0.152
+0.145
aul+0.142
ages+0.142
Neg logit contributions
 them-0.264
 wreck-0.256
 polar-0.251
 rock-0.233
 his-0.231
Token mom
Token ID1995
Feature activation+4.63e-01

Pos logit contributions
's+0.570
ages+0.504
+0.485
 thoughts+0.457
ama+0.455
Neg logit contributions
 them-0.612
mate-0.595
inn-0.582
Nick-0.524
Mike-0.523
Token made
Token ID925
Feature activation-5.46e-01

Pos logit contributions
+0.030
cious+0.028
's+0.028
ages+0.028
ience+0.027
Neg logit contributions
 them-0.031
 wreck-0.028
mate-0.028
 polar-0.027
inn-0.026
Token him
Token ID683
Feature activation-4.23e+00

Pos logit contributions
mate+0.029
 wreck+0.027
 polar+0.027
 them+0.026
 ramp+0.026
Neg logit contributions
-0.041
's-0.041
ages-0.039
aul-0.035
cious-0.035
Token a
Token ID257
Feature activation+1.21e+00

Pos logit contributions
 them+0.241
 wreck+0.219
 polar+0.209
mate+0.206
inn+0.195
Neg logit contributions
-0.305
cious-0.295
ience-0.294
ages-0.286
's-0.286
Token y
Token ID331
Feature activation+1.24e+00

Pos logit contributions
's+0.076
+0.074
ience+0.062
 poem+0.062
cious+0.061
Neg logit contributions
 them-0.084
mate-0.077
inn-0.075
Mike-0.068
 control-0.068
Token,
Token ID11
Feature activation+3.99e-01

Pos logit contributions
 them+0.016
mate+0.014
 wreck+0.014
 polar+0.013
inn+0.013
Neg logit contributions
-0.023
ages-0.023
cious-0.022
ience-0.022
's-0.022
Token "
Token ID366
Feature activation-1.44e+00

Pos logit contributions
ience+0.030
+0.030
ages+0.030
cious+0.030
's+0.029
Neg logit contributions
 them-0.021
 wreck-0.017
 polar-0.017
mate-0.017
inn-0.017
TokenYes
Token ID5297
Feature activation-2.13e+00

Pos logit contributions
 them+0.081
 wreck+0.074
mate+0.072
 polar+0.071
inn+0.067
Neg logit contributions
-0.107
's-0.100
cious-0.100
ages-0.099
ience-0.098
Token!
Token ID0
Feature activation-1.16e+00

Pos logit contributions
 wreck+0.141
 them+0.137
 polar+0.133
inn+0.129
 ramp+0.127
Neg logit contributions
-0.139
's-0.130
ience-0.120
cious-0.112
 Louise-0.105
Token Let
Token ID3914
Feature activation+1.77e+00

Pos logit contributions
 them+0.077
mate+0.075
 polar+0.072
 tow+0.070
inn+0.069
Neg logit contributions
cious-0.067
ages-0.065
ience-0.064
aul-0.062
's-0.061
Token's
Token ID338
Feature activation+9.30e+00

Pos logit contributions
ages+0.059
cious+0.055
aul+0.051
ccoli+0.048
 List+0.048
Neg logit contributions
 them-0.194
mate-0.156
 his-0.154
 row-0.153
 rock-0.152
Token play
Token ID711
Feature activation-2.74e+00

Pos logit contributions
+0.784
's+0.739
ages+0.633
cious+0.627
ience+0.607
Neg logit contributions
 them-0.445
inn-0.429
mate-0.423
 polar-0.402
 tant-0.396
Token before
Token ID878
Feature activation+4.21e+00

Pos logit contributions
 tow+0.157
 polar+0.153
inn+0.153
 wreck+0.152
 them+0.147
Neg logit contributions
-0.201
's-0.198
ience-0.167
ages-0.166
cious-0.165
Token the
Token ID262
Feature activation+3.49e-01

Pos logit contributions
+0.246
cious+0.242
ages+0.241
ience+0.235
's+0.232
Neg logit contributions
 them-0.304
 wreck-0.271
mate-0.264
 polar-0.263
 tow-0.247
Token sunrise
Token ID40048
Feature activation+3.61e+00

Pos logit contributions
+0.026
's+0.025
cious+0.024
ience+0.024
ages+0.024
Neg logit contributions
 them-0.019
mate-0.017
inn-0.017
 wreck-0.017
 polar-0.016
Token!"
Token ID2474
Feature activation-4.06e+00

Pos logit contributions
cious+0.296
+0.296
ience+0.289
ages+0.288
 Michelle+0.281
Neg logit contributions
 them-0.167
 wreck-0.139
mate-0.137
 polar-0.129
 match-0.122
Tokenum
Token ID388
Feature activation+2.31e+00

Pos logit contributions
 them+0.219
 wreck+0.200
mate+0.198
 polar+0.193
inn+0.188
Neg logit contributions
-0.223
's-0.210
cious-0.209
ages-0.207
ience-0.205
Tokenma
Token ID2611
Feature activation-4.46e-01

Pos logit contributions
+0.178
cious+0.178
ages+0.173
ience+0.171
 Michelle+0.165
Neg logit contributions
 them-0.122
 polar-0.095
 wreck-0.094
mate-0.091
inn-0.088
Token kissed
Token ID27946
Feature activation+3.47e+00

Pos logit contributions
 them+0.027
 polar+0.022
 wreck+0.022
mate+0.021
 tow+0.020
Neg logit contributions
cious-0.030
ages-0.030
-0.030
ience-0.029
 List-0.028
Token him
Token ID683
Feature activation-4.82e+00

Pos logit contributions
+0.264
's+0.251
cious+0.248
ages+0.247
ience+0.243
Neg logit contributions
 them-0.182
 wreck-0.169
mate-0.164
 polar-0.160
inn-0.153
Token good
Token ID922
Feature activation-2.40e-01

Pos logit contributions
 them+0.227
 his+0.178
 polar+0.178
 wreck+0.170
mate+0.167
Neg logit contributions
cious-0.394
ience-0.383
aul-0.377
ages-0.371
-0.369
Tokennight
Token ID3847
Feature activation+6.54e+00

Pos logit contributions
 them+0.016
 wreck+0.014
mate+0.014
 polar+0.014
inn+0.013
Neg logit contributions
-0.016
cious-0.015
's-0.015
ience-0.014
ages-0.014
Token and
Token ID290
Feature activation+1.81e-01

Pos logit contributions
+0.519
ience+0.502
cious+0.502
ages+0.489
's+0.483
Neg logit contributions
 them-0.335
 polar-0.273
mate-0.273
 wreck-0.270
inn-0.248
Token urged
Token ID11643
Feature activation-2.04e+00

Pos logit contributions
cious+0.013
ages+0.012
aul+0.012
ience+0.012
+0.012
Neg logit contributions
 them-0.011
 his-0.009
 polar-0.009
 wreck-0.009
mate-0.008
Token him
Token ID683
Feature activation-4.44e+00

Pos logit contributions
 them+0.112
mate+0.093
 wreck+0.092
 polar+0.090
inn+0.086
Neg logit contributions
-0.156
cious-0.151
ience-0.149
ages-0.149
's-0.148
Token to
Token ID284
Feature activation+3.79e-01

Pos logit contributions
 them+0.207
 his+0.166
mate+0.155
 polar+0.134
 load+0.133
Neg logit contributions
ience-0.372
aul-0.368
cious-0.365
 Michelle-0.359
ita-0.357
Token go
Token ID467
Feature activation+1.19e+00

Pos logit contributions
+0.029
cious+0.029
ience+0.028
ages+0.028
's+0.028
Neg logit contributions
 them-0.020
 wreck-0.017
mate-0.017
 polar-0.017
 tow-0.015
Token and
Token ID290
Feature activation+2.82e-01

Pos logit contributions
cious+0.195
+0.193
ages+0.192
ience+0.192
aul+0.185
Neg logit contributions
 them-0.120
mate-0.099
 wreck-0.096
inn-0.093
 polar-0.091
Token talk
Token ID1561
Feature activation+5.18e-01

Pos logit contributions
cious+0.021
ages+0.021
ience+0.020
+0.020
ita+0.020
Neg logit contributions
 them-0.016
 wreck-0.014
 polar-0.013
mate-0.013
 his-0.012
Token about
Token ID546
Feature activation+1.99e+00

Pos logit contributions
+0.037
cious+0.036
ages+0.036
's+0.035
ience+0.035
Neg logit contributions
 them-0.030
mate-0.026
 wreck-0.026
 polar-0.025
inn-0.024
Token why
Token ID1521
Feature activation+3.87e+00

Pos logit contributions
+0.146
's+0.138
cious+0.138
ages+0.137
ience+0.136
Neg logit contributions
 them-0.112
 wreck-0.101
mate-0.099
 polar-0.097
inn-0.093
Token it
Token ID340
Feature activation-1.70e+00

Pos logit contributions
cious+0.294
+0.289
ages+0.284
ita+0.273
ience+0.272
Neg logit contributions
 them-0.220
 wreck-0.180
mate-0.166
 polar-0.162
 his-0.162
Token's
Token ID338
Feature activation+9.04e+00

Pos logit contributions
 them+0.137
 wreck+0.130
 rock+0.125
 match+0.123
 tank+0.123
Neg logit contributions
cious-0.082
enna-0.077
 Michelle-0.075
Nik-0.072
Jennifer-0.071
Token important
Token ID1593
Feature activation+3.91e+00

Pos logit contributions
+0.628
cious+0.613
ience+0.599
ages+0.594
's+0.588
Neg logit contributions
 them-0.561
 wreck-0.502
 polar-0.480
mate-0.471
inn-0.446
Token to
Token ID284
Feature activation+3.50e-01

Pos logit contributions
+0.307
cious+0.298
ience+0.293
's+0.284
ages+0.281
Neg logit contributions
 them-0.206
mate-0.174
 wreck-0.169
 polar-0.153
inn-0.148
Token catch
Token ID4929
Feature activation-7.77e-01

Pos logit contributions
+0.028
's+0.026
cious+0.026
ages+0.026
ience+0.025
Neg logit contributions
 them-0.017
mate-0.016
 wreck-0.016
 polar-0.015
inn-0.015
Token healthy
Token ID5448
Feature activation+3.28e+00

Pos logit contributions
 them+0.045
 wreck+0.036
mate+0.035
 polar+0.034
 his+0.032
Neg logit contributions
cious-0.059
-0.057
ages-0.057
ience-0.056
 Michelle-0.054
Token fish
Token ID5916
Feature activation-2.13e+00

Pos logit contributions
+0.251
cious+0.243
ience+0.240
ages+0.230
 Michelle+0.230
Neg logit contributions
 them-0.183
 wreck-0.153
mate-0.149
 polar-0.149
 snowball-0.136
Token.
Token ID13
Feature activation+1.17e-03

Pos logit contributions
+0.037
's+0.036
cious+0.032
ages+0.032
ience+0.031
Neg logit contributions
 them-0.030
 wreck-0.029
 polar-0.028
mate-0.028
 tow-0.027
Token One
Token ID1881
Feature activation-6.90e-01

Pos logit contributions
+0.000
cious+0.000
ages+0.000
's+0.000
ience+0.000
Neg logit contributions
 them-0.000
 wreck-0.000
mate-0.000
 polar-0.000
inn-0.000
Token day
Token ID1110
Feature activation+2.17e+00

Pos logit contributions
 them+0.034
 polar+0.032
 his+0.031
 wreck+0.031
 snowball+0.031
Neg logit contributions
ience-0.056
ages-0.053
cious-0.052
-0.050
aries-0.050
Token,
Token ID11
Feature activation+2.78e-01

Pos logit contributions
+0.160
's+0.152
cious+0.151
ages+0.150
ience+0.149
Neg logit contributions
 them-0.122
 wreck-0.109
mate-0.108
 polar-0.105
inn-0.101
Token while
Token ID981
Feature activation-2.43e+00

Pos logit contributions
's+0.010
+0.010
 Michelle+0.008
cious+0.008
ages+0.008
Neg logit contributions
 wreck-0.025
 polar-0.024
 them-0.024
mate-0.024
 ramp-0.023
Token she
Token ID673
Feature activation+3.52e+00

Pos logit contributions
 them+0.166
 wreck+0.141
 polar+0.141
mate+0.140
 tow+0.133
Neg logit contributions
-0.155
ience-0.154
ages-0.154
cious-0.151
aul-0.146
Token was
Token ID373
Feature activation-6.15e-01

Pos logit contributions
+0.240
's+0.229
cious+0.223
ages+0.222
ience+0.219
Neg logit contributions
 them-0.220
 wreck-0.198
mate-0.197
 polar-0.192
inn-0.185
Token playing
Token ID2712
Feature activation-1.22e+00

Pos logit contributions
 them+0.033
 wreck+0.028
mate+0.026
 polar+0.026
inn+0.025
Neg logit contributions
-0.049
ience-0.049
cious-0.047
ages-0.047
aul-0.046
Token in
Token ID287
Feature activation+3.11e-01

Pos logit contributions
 polar+0.078
 them+0.078
 wreck+0.078
 tow+0.075
inn+0.074
Neg logit contributions
's-0.074
-0.071
ages-0.062
cious-0.060
 Louise-0.059
Token the
Token ID262
Feature activation+2.13e-01

Pos logit contributions
+0.019
's+0.018
 Louise+0.016
 Michelle+0.016
 Denise+0.015
Neg logit contributions
mate-0.021
inn-0.020
 them-0.020
 wreck-0.019
 captain-0.019
Token park
Token ID3952
Feature activation-1.14e+00

Pos logit contributions
ience+0.017
+0.017
ages+0.017
cious+0.017
's+0.016
Neg logit contributions
 them-0.011
 wreck-0.010
 polar-0.009
mate-0.009
 snowball-0.008
Token a
Token ID257
Feature activation+1.11e+00

Pos logit contributions
 them+0.165
 wreck+0.149
mate+0.147
 polar+0.144
inn+0.138
Neg logit contributions
-0.221
's-0.210
cious-0.207
ages-0.207
ience-0.205
Token magic
Token ID5536
Feature activation-6.10e-01

Pos logit contributions
+0.085
's+0.081
cious+0.080
ages+0.080
ience+0.079
Neg logit contributions
 them-0.059
 wreck-0.052
mate-0.052
 polar-0.051
inn-0.049
Token stone
Token ID7815
Feature activation+1.43e+00

Pos logit contributions
 them+0.036
 wreck+0.032
 polar+0.032
mate+0.031
inn+0.031
Neg logit contributions
-0.043
's-0.041
ages-0.040
cious-0.039
ience-0.039
Token.
Token ID13
Feature activation-3.29e-02

Pos logit contributions
+0.108
cious+0.104
ience+0.103
ages+0.102
's+0.101
Neg logit contributions
 them-0.077
 wreck-0.067
mate-0.067
 polar-0.063
inn-0.061
Token The
Token ID383
Feature activation+1.92e+00

Pos logit contributions
 them+0.002
 his+0.002
 load+0.002
 wreck+0.002
 tow+0.002
Neg logit contributions
cious-0.002
ages-0.002
ience-0.002
-0.002
ita-0.002
Token human
Token ID1692
Feature activation+1.69e+00

Pos logit contributions
+0.151
ages+0.149
ience+0.142
aul+0.142
cious+0.142
Neg logit contributions
 wreck-0.088
 polar-0.088
 them-0.087
 rock-0.086
 captain-0.086
Token did
Token ID750
Feature activation-1.72e+00

Pos logit contributions
cious+0.121
ages+0.113
aul+0.113
 Michelle+0.113
+0.111
Neg logit contributions
 them-0.096
 wreck-0.086
 polar-0.082
 captain-0.081
 rock-0.081
Token as
Token ID355
Feature activation-5.87e-01

Pos logit contributions
 them+0.120
 wreck+0.110
mate+0.109
 polar+0.107
inn+0.103
Neg logit contributions
-0.104
's-0.098
cious-0.097
ages-0.096
ience-0.096
Token he
Token ID339
Feature activation+3.40e-01

Pos logit contributions
 them+0.027
 his+0.022
 wreck+0.021
 polar+0.021
mate+0.020
Neg logit contributions
ages-0.050
cious-0.050
-0.049
ience-0.047
's-0.047
Token was
Token ID373
Feature activation-6.59e-01

Pos logit contributions
+0.021
's+0.020
ience+0.020
ages+0.019
cious+0.019
Neg logit contributions
 them-0.024
mate-0.021
 wreck-0.021
 polar-0.021
inn-0.020
Token told
Token ID1297
Feature activation-7.34e-02

Pos logit contributions
 them+0.043
 wreck+0.039
mate+0.038
 polar+0.037
inn+0.036
Neg logit contributions
-0.042
cious-0.041
ience-0.040
ages-0.040
's-0.039
Token bread
Token ID8509
Feature activation+8.06e-01

Pos logit contributions
+0.022
ages+0.022
cious+0.021
ience+0.020
's+0.020
Neg logit contributions
 them-0.018
 wreck-0.016
 polar-0.015
 match-0.015
 rock-0.014
Token and
Token ID290
Feature activation+4.82e-01

Pos logit contributions
+0.064
cious+0.062
ience+0.062
ages+0.061
's+0.059
Neg logit contributions
 them-0.042
 wreck-0.034
 polar-0.033
mate-0.033
 match-0.030
Token cakes
Token ID30849
Feature activation+1.07e+00

Pos logit contributions
+0.043
ience+0.042
cious+0.042
ages+0.041
ita+0.040
Neg logit contributions
 them-0.023
 his-0.017
 polar-0.016
 wreck-0.016
 rock-0.015
Token,
Token ID11
Feature activation+3.28e-01

Pos logit contributions
+0.084
cious+0.080
ience+0.080
ages+0.079
's+0.079
Neg logit contributions
 them-0.056
 wreck-0.048
mate-0.047
 polar-0.046
inn-0.043
Token and
Token ID290
Feature activation+4.61e-01

Pos logit contributions
+0.028
cious+0.028
ience+0.027
ages+0.027
's+0.026
Neg logit contributions
 them-0.017
 polar-0.012
 wreck-0.012
 his-0.011
mate-0.011
Token the
Token ID262
Feature activation+2.77e-01

Pos logit contributions
cious+0.037
ages+0.036
+0.036
ience+0.036
ita+0.034
Neg logit contributions
 them-0.025
 wreck-0.020
 polar-0.020
 his-0.019
 match-0.018
Token love
Token ID1842
Feature activation+3.20e-01

Pos logit contributions
+0.020
ages+0.020
cious+0.019
ience+0.019
's+0.019
Neg logit contributions
 them-0.016
 wreck-0.015
 polar-0.014
mate-0.013
 tow-0.013
Token of
Token ID286
Feature activation-3.18e+00

Pos logit contributions
+0.025
cious+0.025
ages+0.024
ience+0.024
's+0.024
Neg logit contributions
 them-0.017
mate-0.014
 wreck-0.014
 polar-0.013
inn-0.012
Token their
Token ID511
Feature activation-3.21e+00

Pos logit contributions
 them+0.155
 wreck+0.128
mate+0.124
 polar+0.122
 tow+0.112
Neg logit contributions
-0.267
cious-0.257
ages-0.257
's-0.255
ience-0.254
Token mother
Token ID2802
Feature activation+2.71e+00

Pos logit contributions
 them+0.205
 wreck+0.176
 polar+0.170
mate+0.161
 match+0.159
Neg logit contributions
-0.225
ages-0.211
cious-0.209
ience-0.203
's-0.201
Token and
Token ID290
Feature activation+4.56e-01

Pos logit contributions
cious+0.229
ages+0.228
ience+0.224
 List+0.220
+0.220
Neg logit contributions
 them-0.123
 polar-0.091
 wreck-0.089
mate-0.084
 his-0.081
Token John
Token ID1757
Feature activation+2.13e+00

Pos logit contributions
 them+0.134
 wreck+0.117
mate+0.114
 polar+0.112
inn+0.105
Neg logit contributions
-0.246
's-0.235
cious-0.235
ages-0.234
ience-0.233
Token.
Token ID13
Feature activation-4.78e-02

Pos logit contributions
's+0.142
+0.139
ages+0.122
cious+0.122
aul+0.116
Neg logit contributions
mate-0.138
 them-0.136
 ramp-0.131
 wreck-0.129
inn-0.128
Token John
Token ID1757
Feature activation+2.12e+00

Pos logit contributions
 them+0.003
 wreck+0.003
mate+0.003
 polar+0.002
inn+0.002
Neg logit contributions
-0.003
cious-0.003
's-0.003
ages-0.003
ience-0.003
Token loved
Token ID6151
Feature activation+9.49e-01

Pos logit contributions
+0.136
cious+0.132
ience+0.132
ages+0.130
's+0.126
Neg logit contributions
 them-0.135
 wreck-0.124
 polar-0.121
mate-0.116
inn-0.112
Token to
Token ID284
Feature activation+4.31e-01

Pos logit contributions
+0.066
's+0.063
cious+0.063
ages+0.062
ience+0.062
Neg logit contributions
 them-0.057
 wreck-0.051
mate-0.051
 polar-0.050
inn-0.048
Token explore
Token ID7301
Feature activation+3.40e+00

Pos logit contributions
+0.035
's+0.033
cious+0.032
ages+0.032
ience+0.030
Neg logit contributions
 them-0.021
mate-0.019
 wreck-0.019
inn-0.019
 polar-0.018
Token the
Token ID262
Feature activation+1.60e-01

Pos logit contributions
+0.252
's+0.249
cious+0.225
ience+0.224
ages+0.222
Neg logit contributions
mate-0.176
 them-0.171
 wreck-0.168
inn-0.166
 polar-0.163
Token outdoors
Token ID24349
Feature activation+1.44e+00

Pos logit contributions
+0.011
ience+0.011
ages+0.011
cious+0.010
's+0.010
Neg logit contributions
 them-0.010
 wreck-0.009
 polar-0.009
mate-0.009
 rock-0.008
Token and
Token ID290
Feature activation+3.50e-01

Pos logit contributions
+0.116
ience+0.115
cious+0.114
ages+0.112
's+0.108
Neg logit contributions
 them-0.072
 wreck-0.062
mate-0.059
 polar-0.058
inn-0.052
Token he
Token ID339
Feature activation+3.33e-01

Pos logit contributions
ience+0.028
+0.028
ages+0.028
cious+0.028
aul+0.027
Neg logit contributions
 them-0.018
 wreck-0.015
mate-0.014
 polar-0.014
 rock-0.013
Token often
Token ID1690
Feature activation-2.68e+00

Pos logit contributions
+0.021
cious+0.021
ages+0.020
ience+0.020
's+0.020
Neg logit contributions
 them-0.022
 wreck-0.019
mate-0.019
 polar-0.019
inn-0.018
Token field
Token ID2214
Feature activation-6.02e-02

Pos logit contributions
+0.022
cious+0.021
's+0.021
ages+0.020
ience+0.020
Neg logit contributions
 them-0.016
 wreck-0.015
mate-0.014
 polar-0.014
inn-0.013
Token,
Token ID11
Feature activation+3.60e-01

Pos logit contributions
 them+0.003
 wreck+0.003
mate+0.003
 polar+0.003
inn+0.003
Neg logit contributions
-0.005
's-0.004
cious-0.004
ages-0.004
ience-0.004
Token eating
Token ID6600
Feature activation+1.43e+00

Pos logit contributions
ages+0.026
+0.025
ience+0.025
cious+0.025
aul+0.025
Neg logit contributions
 them-0.021
 wreck-0.018
 polar-0.018
mate-0.017
 ramp-0.016
Token grass
Token ID8701
Feature activation+5.00e+00

Pos logit contributions
+0.101
's+0.097
cious+0.093
ages+0.092
aul+0.090
Neg logit contributions
 wreck-0.078
 them-0.078
mate-0.078
inn-0.075
 polar-0.073
Token and
Token ID290
Feature activation+5.11e-01

Pos logit contributions
+0.385
cious+0.369
ience+0.367
ages+0.365
's+0.363
Neg logit contributions
 them-0.266
 wreck-0.229
mate-0.225
 polar-0.220
inn-0.209
Token feeling
Token ID4203
Feature activation+7.20e-01

Pos logit contributions
ience+0.037
+0.036
cious+0.036
ages+0.036
aul+0.035
Neg logit contributions
 them-0.030
 polar-0.027
 wreck-0.026
 his-0.025
 rock-0.025
Token the
Token ID262
Feature activation+3.35e-01

Pos logit contributions
+0.050
cious+0.049
ages+0.048
ience+0.048
's+0.047
Neg logit contributions
 them-0.044
 wreck-0.038
 polar-0.038
mate-0.037
 tow-0.034
Token wind
Token ID2344
Feature activation+4.35e+00

Pos logit contributions
ages+0.025
ience+0.024
cious+0.024
+0.023
 Michelle+0.023
Neg logit contributions
 them-0.018
 polar-0.017
 wreck-0.017
 rock-0.017
 snowball-0.016
Token in
Token ID287
Feature activation+4.50e-01

Pos logit contributions
+0.331
cious+0.323
ience+0.320
ages+0.319
's+0.312
Neg logit contributions
 them-0.230
 wreck-0.202
mate-0.196
 polar-0.192
inn-0.181
Token his
Token ID465
Feature activation-4.01e+00

Pos logit contributions
+0.036
's+0.032
 Michelle+0.032
 Louise+0.030
 Denise+0.030
Neg logit contributions
 ramp-0.022
 wreck-0.021
mate-0.020
inn-0.019
 cubes-0.019
Token fur
Token ID9230
Feature activation+7.59e-01

Pos logit contributions
 them+0.229
 wreck+0.207
 polar+0.205
 rock+0.201
 his+0.198
Neg logit contributions
-0.282
ience-0.280
 Michelle-0.274
 List-0.257
cious-0.256
Token in
Token ID287
Feature activation+2.75e-01

Pos logit contributions
+0.181
's+0.175
cious+0.151
ience+0.148
 Louise+0.146
Neg logit contributions
 polar-0.160
 wreck-0.158
 them-0.157
mate-0.156
inn-0.156
Token circles
Token ID13332
Feature activation-8.38e-01

Pos logit contributions
+0.017
's+0.015
 decisions+0.013
 pays+0.013
 dedication+0.012
Neg logit contributions
inn-0.019
mate-0.018
 polar-0.017
 ramp-0.017
 wreck-0.017
Token,
Token ID11
Feature activation+2.24e-01

Pos logit contributions
 them+0.045
 wreck+0.041
mate+0.040
 polar+0.039
inn+0.037
Neg logit contributions
-0.063
's-0.060
cious-0.060
ages-0.059
ience-0.059
Token cl
Token ID537
Feature activation-2.95e+00

Pos logit contributions
+0.016
cious+0.015
ages+0.015
ience+0.015
's+0.015
Neg logit contributions
 them-0.014
 wreck-0.012
 polar-0.012
mate-0.012
inn-0.011
Tokenapping
Token ID5912
Feature activation-1.59e-01

Pos logit contributions
 them+0.218
 wreck+0.200
mate+0.198
 polar+0.196
inn+0.192
Neg logit contributions
-0.167
cious-0.156
's-0.155
ience-0.155
ages-0.154
Token his
Token ID465
Feature activation-4.20e+00

Pos logit contributions
 wreck+0.006
 them+0.006
mate+0.005
 polar+0.005
inn+0.005
Neg logit contributions
-0.014
's-0.014
ages-0.013
cious-0.013
ience-0.013
Token fins
Token ID42080
Feature activation+2.47e+00

Pos logit contributions
 them+0.269
 wreck+0.251
 polar+0.242
 row+0.237
 match+0.236
Neg logit contributions
-0.263
 Michelle-0.262
assy-0.255
ience-0.254
aul-0.251
Token together
Token ID1978
Feature activation-3.00e+00

Pos logit contributions
+0.184
's+0.176
cious+0.173
ages+0.172
ience+0.170
Neg logit contributions
 them-0.136
 wreck-0.122
mate-0.122
 polar-0.118
inn-0.112
Token.
Token ID13
Feature activation-7.05e-02

Pos logit contributions
 them+0.166
 wreck+0.153
mate+0.150
 polar+0.149
inn+0.142
Neg logit contributions
-0.222
's-0.212
cious-0.207
ages-0.205
ience-0.203
Token
Token ID198
Feature activation+9.73e-02

Pos logit contributions
 them+0.004
inn+0.003
 polar+0.003
 tow+0.003
 wreck+0.003
Neg logit contributions
cious-0.005
ience-0.005
ages-0.005
-0.005
ita-0.005
Token
Token ID198
Feature activation+9.03e-02

Pos logit contributions
+0.006
's+0.005
ages+0.004
cious+0.004
aul+0.004
Neg logit contributions
 wreck-0.008
mate-0.007
 them-0.007
 polar-0.007
 captain-0.007
Token He
Token ID679
Feature activation-2.16e+00

Pos logit contributions
 them+0.002
 wreck+0.001
mate+0.001
 polar+0.001
inn+0.001
Neg logit contributions
-0.002
's-0.002
cious-0.002
ages-0.002
ience-0.002
Token was
Token ID373
Feature activation-6.32e-01

Pos logit contributions
 them+0.154
mate+0.148
 wreck+0.142
 polar+0.141
 tow+0.140
Neg logit contributions
-0.125
's-0.123
ience-0.115
cious-0.112
ages-0.112
Token always
Token ID1464
Feature activation-9.53e-01

Pos logit contributions
 them+0.042
mate+0.042
 wreck+0.040
 polar+0.039
 ramp+0.038
Neg logit contributions
-0.038
's-0.037
cious-0.034
ages-0.034
ience-0.033
Token pushing
Token ID7796
Feature activation-2.95e+00

Pos logit contributions
 them+0.064
 wreck+0.059
mate+0.059
 polar+0.057
inn+0.055
Neg logit contributions
-0.059
's-0.056
cious-0.055
ages-0.055
ience-0.054
Token the
Token ID262
Feature activation+1.91e-01

Pos logit contributions
 wreck+0.110
 polar+0.109
 tow+0.107
mate+0.106
 ramp+0.101
Neg logit contributions
-0.262
's-0.259
cious-0.234
ience-0.234
ages-0.234
Token other
Token ID584
Feature activation-1.81e+00

Pos logit contributions
's+0.015
+0.015
cious+0.014
ages+0.013
ience+0.013
Neg logit contributions
 them-0.010
mate-0.010
 tow-0.009
inn-0.009
 wreck-0.009
Token fishes
Token ID49765
Feature activation+3.36e-01

Pos logit contributions
 them+0.101
 wreck+0.092
mate+0.091
 polar+0.087
inn+0.085
Neg logit contributions
-0.134
's-0.131
cious-0.125
ages-0.124
ience-0.121
Token away
Token ID1497
Feature activation-3.25e+00

Pos logit contributions
+0.023
's+0.023
ages+0.020
cious+0.020
ience+0.020
Neg logit contributions
 them-0.020
 polar-0.020
 tow-0.019
mate-0.019
 wreck-0.019
Token when
Token ID618
Feature activation+4.45e+00

Pos logit contributions
 them+0.196
 polar+0.185
 wreck+0.184
mate+0.180
 tow+0.174
Neg logit contributions
-0.220
's-0.214
ages-0.201
cious-0.199
ience-0.194
Token they
Token ID484
Feature activation+1.64e+00

Pos logit contributions
+0.353
's+0.337
cious+0.335
ages+0.333
ience+0.330
Neg logit contributions
 them-0.224
 wreck-0.199
mate-0.196
 polar-0.191
inn-0.182
Token tried
Token ID3088
Feature activation-2.89e+00

Pos logit contributions
+0.107
's+0.107
ience+0.097
cious+0.093
ages+0.092
Neg logit contributions
 them-0.100
mate-0.100
inn-0.096
 polar-0.095
 snowball-0.093
Token day
Token ID1110
Feature activation+2.13e+00

Pos logit contributions
 them+0.036
 wreck+0.034
 polar+0.033
 his+0.032
inn+0.032
Neg logit contributions
ience-0.058
ages-0.055
cious-0.055
-0.052
 Michelle-0.052
Token,
Token ID11
Feature activation+2.49e-01

Pos logit contributions
+0.168
cious+0.163
ience+0.162
ages+0.162
's+0.157
Neg logit contributions
 them-0.110
mate-0.094
 wreck-0.092
 polar-0.088
inn-0.087
Token her
Token ID607
Feature activation-1.10e+00

Pos logit contributions
's+0.010
+0.010
 Michelle+0.009
cious+0.008
ages+0.008
Neg logit contributions
 wreck-0.022
 polar-0.021
 them-0.021
mate-0.020
 ramp-0.020
Token mom
Token ID1995
Feature activation+3.27e-01

Pos logit contributions
 them+0.055
 wreck+0.049
mate+0.047
 polar+0.047
 tow+0.044
Neg logit contributions
-0.089
ience-0.086
cious-0.085
ages-0.084
's-0.081
Token gave
Token ID2921
Feature activation+2.84e+00

Pos logit contributions
+0.021
cious+0.020
ages+0.020
ience+0.019
's+0.019
Neg logit contributions
 them-0.022
 wreck-0.020
mate-0.019
 polar-0.019
inn-0.018
Token her
Token ID607
Feature activation-1.12e+00

Pos logit contributions
+0.192
's+0.185
ages+0.175
cious+0.174
ience+0.174
Neg logit contributions
 them-0.172
 wreck-0.166
mate-0.162
 polar-0.161
 tow-0.152
Token an
Token ID281
Feature activation+2.12e+00

Pos logit contributions
 them+0.062
 wreck+0.050
mate+0.049
 polar+0.049
 his+0.047
Neg logit contributions
-0.085
cious-0.084
ience-0.082
ages-0.082
's-0.078
Token ancient
Token ID6156
Feature activation+7.11e-01

Pos logit contributions
+0.121
's+0.115
ience+0.104
ages+0.100
cious+0.100
Neg logit contributions
 them-0.157
mate-0.146
 wreck-0.137
 polar-0.136
inn-0.136
Token card
Token ID2657
Feature activation-2.14e+00

Pos logit contributions
+0.048
's+0.047
cious+0.045
ience+0.044
ages+0.044
Neg logit contributions
 them-0.045
mate-0.039
inn-0.039
 wreck-0.038
 polar-0.037
Token that
Token ID326
Feature activation-4.20e-01

Pos logit contributions
 them+0.111
mate+0.100
 wreck+0.096
 polar+0.093
inn+0.089
Neg logit contributions
-0.163
ience-0.160
cious-0.157
's-0.151
ages-0.151
Token was
Token ID373
Feature activation-6.59e-01

Pos logit contributions
 them+0.024
mate+0.019
 match+0.019
 wreck+0.019
 his+0.018
Neg logit contributions
cious-0.032
ience-0.032
ages-0.032
ita-0.031
-0.030
Token.
Token ID13
Feature activation-5.20e-02

Pos logit contributions
 them+0.098
mate+0.093
 polar+0.091
 wreck+0.089
inn+0.088
Neg logit contributions
's-0.106
-0.105
cious-0.096
ages-0.093
ience-0.090
Token They
Token ID1119
Feature activation-4.10e-01

Pos logit contributions
 them+0.003
 wreck+0.003
mate+0.003
 polar+0.003
inn+0.003
Neg logit contributions
-0.003
's-0.003
cious-0.003
ages-0.003
ience-0.003
Token hugged
Token ID41130
Feature activation+1.87e+00

Pos logit contributions
mate+0.030
inn+0.029
 them+0.027
 wreck+0.027
 ramp+0.026
Neg logit contributions
's-0.025
-0.024
cious-0.021
ience-0.020
 listens-0.019
Token her
Token ID607
Feature activation-1.12e+00

Pos logit contributions
's+0.129
+0.127
 Michelle+0.114
ages+0.114
 listens+0.110
Neg logit contributions
 wreck-0.109
mate-0.106
inn-0.102
 polar-0.100
 ramp-0.100
Token and
Token ID290
Feature activation+3.50e-01

Pos logit contributions
 them+0.060
 wreck+0.053
mate+0.052
 polar+0.051
inn+0.048
Neg logit contributions
-0.087
cious-0.083
ages-0.082
ience-0.082
's-0.082
Token told
Token ID1297
Feature activation-8.62e-02

Pos logit contributions
+0.024
cious+0.022
ience+0.022
ages+0.022
aul+0.021
Neg logit contributions
 them-0.023
 wreck-0.020
 polar-0.020
 his-0.019
mate-0.019
Token her
Token ID607
Feature activation-1.13e+00

Pos logit contributions
 wreck+0.006
mate+0.006
 ramp+0.005
inn+0.005
 polar+0.005
Neg logit contributions
-0.005
's-0.005
ages-0.005
 Michelle-0.005
 listens-0.005
Token what
Token ID644
Feature activation+1.74e+00

Pos logit contributions
 them+0.066
 wreck+0.054
 polar+0.052
mate+0.050
 his+0.050
Neg logit contributions
-0.085
cious-0.083
ience-0.082
ages-0.081
aul-0.078
Token happened
Token ID3022
Feature activation+2.97e+00

Pos logit contributions
ience+0.100
ages+0.096
cious+0.094
+0.093
ita+0.089
Neg logit contributions
 them-0.133
 wreck-0.117
 polar-0.113
mate-0.113
 his-0.110
Token.
Token ID13
Feature activation-7.62e-02

Pos logit contributions
+0.248
ience+0.248
cious+0.245
ages+0.243
aul+0.239
Neg logit contributions
 them-0.141
 wreck-0.112
mate-0.108
 polar-0.101
inn-0.096
Token
Token ID198
Feature activation+9.10e-02

Pos logit contributions
 them+0.004
 wreck+0.004
mate+0.004
 polar+0.004
inn+0.004
Neg logit contributions
-0.006
cious-0.005
's-0.005
ience-0.005
ages-0.005
Token getting
Token ID1972
Feature activation-3.41e-01

Pos logit contributions
 them+0.213
 wreck+0.187
 polar+0.184
mate+0.179
inn+0.170
Neg logit contributions
-0.244
ience-0.238
cious-0.236
ages-0.233
's-0.227
Token ready
Token ID3492
Feature activation+1.37e+00

Pos logit contributions
 them+0.024
 wreck+0.021
 polar+0.021
mate+0.020
inn+0.020
Neg logit contributions
-0.021
cious-0.021
ience-0.020
ages-0.020
's-0.019
Token to
Token ID284
Feature activation+4.66e-01

Pos logit contributions
+0.103
cious+0.098
's+0.098
ages+0.097
ience+0.097
Neg logit contributions
 them-0.076
 wreck-0.067
mate-0.066
 polar-0.064
inn-0.062
Token make
Token ID787
Feature activation-2.73e+00

Pos logit contributions
+0.036
's+0.035
cious+0.034
ages+0.034
ience+0.033
Neg logit contributions
 them-0.024
mate-0.022
 wreck-0.022
 polar-0.021
inn-0.020
Token a
Token ID257
Feature activation+1.12e+00

Pos logit contributions
 them+0.167
 wreck+0.129
 polar+0.126
 his+0.120
 snowball+0.118
Neg logit contributions
ience-0.205
-0.205
cious-0.203
ages-0.192
 listens-0.188
Token cup
Token ID6508
Feature activation-2.06e+00

Pos logit contributions
+0.080
's+0.077
cious+0.072
ages+0.072
ience+0.071
Neg logit contributions
 them-0.066
mate-0.061
inn-0.059
 wreck-0.056
 polar-0.055
Token of
Token ID286
Feature activation-3.26e+00

Pos logit contributions
 them+0.106
 wreck+0.095
mate+0.094
 polar+0.093
inn+0.088
Neg logit contributions
-0.163
's-0.158
ages-0.153
cious-0.150
ience-0.146
Token hot
Token ID3024
Feature activation+2.93e+00

Pos logit contributions
mate+0.208
 them+0.196
 wreck+0.194
inn+0.189
 polar+0.177
Neg logit contributions
-0.215
's-0.207
ages-0.190
cious-0.187
aul-0.182
Token chocolate
Token ID11311
Feature activation-2.75e-01

Pos logit contributions
+0.202
cious+0.193
ience+0.193
ages+0.189
's+0.188
Neg logit contributions
 them-0.181
 wreck-0.161
 polar-0.157
mate-0.156
inn-0.148
Token.
Token ID13
Feature activation-2.50e-02

Pos logit contributions
 them+0.015
mate+0.014
 wreck+0.014
 polar+0.014
inn+0.013
Neg logit contributions
-0.020
's-0.020
ages-0.019
cious-0.019
ience-0.018
Token The
Token ID383
Feature activation+1.93e+00

Pos logit contributions
 them+0.002
 tow+0.001
 wreck+0.001
 polar+0.001
inn+0.001
Neg logit contributions
ience-0.002
cious-0.002
ages-0.002
-0.002
aul-0.002
Token watched
Token ID7342
Feature activation+3.31e+00

Pos logit contributions
+0.229
cious+0.224
ages+0.219
ience+0.216
's+0.213
Neg logit contributions
 them-0.241
 wreck-0.217
mate-0.211
 polar-0.210
inn-0.198
Token as
Token ID355
Feature activation-4.04e-01

Pos logit contributions
+0.271
's+0.258
cious+0.254
ages+0.253
ience+0.251
Neg logit contributions
 them-0.154
 wreck-0.142
mate-0.139
 polar-0.135
inn-0.128
Token everyone
Token ID2506
Feature activation+3.00e+00

Pos logit contributions
 them+0.021
 wreck+0.016
mate+0.016
 polar+0.016
 tow+0.015
Neg logit contributions
-0.033
ages-0.032
cious-0.032
's-0.032
ience-0.032
Token took
Token ID1718
Feature activation+2.10e+00

Pos logit contributions
+0.205
cious+0.201
ages+0.200
ience+0.195
's+0.193
Neg logit contributions
 them-0.183
 wreck-0.162
mate-0.156
 polar-0.155
inn-0.147
Token a
Token ID257
Feature activation+1.26e+00

Pos logit contributions
cious+0.180
+0.180
ience+0.178
ages+0.177
's+0.171
Neg logit contributions
 them-0.101
mate-0.071
 wreck-0.069
 polar-0.066
 tow-0.063
Token few
Token ID1178
Feature activation-5.89e+00

Pos logit contributions
+0.103
's+0.097
cious+0.095
ience+0.094
ages+0.093
Neg logit contributions
 them-0.064
mate-0.056
 wreck-0.055
 polar-0.054
inn-0.051
Token.
Token ID13
Feature activation+1.17e-01

Pos logit contributions
 them+0.361
 wreck+0.334
mate+0.332
 polar+0.324
inn+0.310
Neg logit contributions
-0.401
's-0.378
ience-0.373
cious-0.372
ages-0.371
Token
Token ID198
Feature activation+2.88e-01

Pos logit contributions
ience+0.008
ages+0.008
cious+0.008
's+0.008
+0.007
Neg logit contributions
 them-0.007
inn-0.006
 tow-0.006
mate-0.006
 polar-0.006
Token
Token ID198
Feature activation+2.84e-01

Pos logit contributions
+0.018
's+0.016
cious+0.015
ages+0.015
aul+0.014
Neg logit contributions
 wreck-0.020
 them-0.019
 polar-0.019
mate-0.019
 ramp-0.018
TokenWhen
Token ID2215
Feature activation-1.15e+00

Pos logit contributions
+0.017
's+0.015
cious+0.015
ages+0.015
ience+0.015
Neg logit contributions
 them-0.021
 wreck-0.019
 polar-0.019
mate-0.019
inn-0.018
Token it
Token ID340
Feature activation-1.46e+00

Pos logit contributions
 them+0.064
mate+0.048
 wreck+0.046
 his+0.045
 polar+0.045
Neg logit contributions
ages-0.095
cious-0.092
ience-0.091
-0.090
aul-0.088
Token having
Token ID1719
Feature activation-1.94e+00

Pos logit contributions
+0.019
cious+0.017
's+0.017
ages+0.017
ience+0.017
Neg logit contributions
 them-0.021
 wreck-0.019
mate-0.018
 polar-0.018
inn-0.017
Token fun
Token ID1257
Feature activation-3.69e-01

Pos logit contributions
 them+0.114
 wreck+0.104
mate+0.103
 polar+0.100
inn+0.097
Neg logit contributions
-0.138
's-0.130
cious-0.130
ages-0.129
ience-0.127
Token.
Token ID13
Feature activation-1.27e-01

Pos logit contributions
 them+0.020
 wreck+0.019
mate+0.019
 polar+0.018
inn+0.018
Neg logit contributions
-0.027
's-0.026
ages-0.025
cious-0.025
ience-0.025
Token She
Token ID1375
Feature activation+4.75e-01

Pos logit contributions
 them+0.009
 polar+0.007
 wreck+0.007
 his+0.007
mate+0.007
Neg logit contributions
ience-0.008
ages-0.008
cious-0.008
-0.008
's-0.007
Token saw
Token ID2497
Feature activation-3.81e+00

Pos logit contributions
+0.029
's+0.027
cious+0.027
ages+0.026
ience+0.026
Neg logit contributions
 them-0.033
 wreck-0.031
mate-0.030
 polar-0.030
inn-0.029
Token them
Token ID606
Feature activation-7.18e+00

Pos logit contributions
 them+0.205
 wreck+0.197
mate+0.194
 polar+0.188
inn+0.183
Neg logit contributions
-0.279
's-0.269
cious-0.260
ience-0.259
ages-0.258
Token in
Token ID287
Feature activation+1.97e-01

Pos logit contributions
 them+0.427
 wreck+0.381
mate+0.370
 polar+0.365
inn+0.345
Neg logit contributions
-0.505
cious-0.487
ages-0.484
ience-0.479
's-0.476
Token their
Token ID511
Feature activation-3.40e+00

Pos logit contributions
+0.012
's+0.011
 Michelle+0.011
 Louise+0.010
 Denise+0.010
Neg logit contributions
inn-0.013
 wreck-0.013
mate-0.013
 polar-0.012
 ramp-0.012
Token old
Token ID1468
Feature activation+1.01e+00

Pos logit contributions
 them+0.191
 wreck+0.167
 polar+0.158
mate+0.154
 tow+0.151
Neg logit contributions
-0.259
ages-0.246
cious-0.243
ience-0.241
aul-0.233
Token house
Token ID2156
Feature activation-1.64e+00

Pos logit contributions
+0.074
cious+0.069
ience+0.069
ages+0.068
aul+0.066
Neg logit contributions
 them-0.059
 wreck-0.053
mate-0.051
 polar-0.050
 ramp-0.047
Token and
Token ID290
Feature activation+2.72e-01

Pos logit contributions
 them+0.090
 wreck+0.080
mate+0.078
 polar+0.076
inn+0.072
Neg logit contributions
-0.123
cious-0.117
's-0.117
ages-0.116
ience-0.116
Token lost
Token ID2626
Feature activation-1.12e+00

Pos logit contributions
 them+0.002
 wreck+0.002
mate+0.002
 polar+0.002
inn+0.002
Neg logit contributions
-0.002
's-0.002
cious-0.002
ages-0.002
ience-0.002
Token my
Token ID616
Feature activation-2.10e+00

Pos logit contributions
 them+0.063
 wreck+0.060
inn+0.060
mate+0.060
 polar+0.058
Neg logit contributions
-0.078
's-0.075
cious-0.070
ience-0.070
ages-0.070
Token way
Token ID835
Feature activation+2.86e+00

Pos logit contributions
 them+0.115
mate+0.101
inn+0.101
 wreck+0.098
 polar+0.094
Neg logit contributions
-0.157
's-0.152
ages-0.146
cious-0.146
ience-0.144
Token home
Token ID1363
Feature activation-2.64e+00

Pos logit contributions
+0.201
's+0.187
cious+0.159
ience+0.157
ages+0.154
Neg logit contributions
 them-0.172
 wreck-0.166
mate-0.164
 polar-0.164
inn-0.160
Token and
Token ID290
Feature activation+4.91e-01

Pos logit contributions
 them+0.125
mate+0.107
 wreck+0.106
 polar+0.098
 control+0.092
Neg logit contributions
cious-0.211
-0.210
ience-0.208
ages-0.206
's-0.199
Token now
Token ID783
Feature activation-6.74e+00

Pos logit contributions
cious+0.031
ages+0.030
+0.029
aul+0.029
ience+0.029
Neg logit contributions
 them-0.034
 wreck-0.031
mate-0.029
 polar-0.029
 his-0.027
Token I
Token ID314
Feature activation+2.96e+00

Pos logit contributions
 them+0.431
 wreck+0.364
 his+0.349
 rock+0.345
mate+0.343
Neg logit contributions
ages-0.478
cious-0.475
aul-0.463
ita-0.444
assy-0.439
Token don
Token ID836
Feature activation+2.76e+00

Pos logit contributions
+0.212
's+0.199
ience+0.191
ages+0.188
cious+0.187
Neg logit contributions
 them-0.181
mate-0.164
 wreck-0.161
 polar-0.157
inn-0.155
Token't
Token ID470
Feature activation-4.20e+00

Pos logit contributions
+0.222
's+0.210
cious+0.205
ages+0.205
ience+0.204
Neg logit contributions
 them-0.139
 wreck-0.125
mate-0.122
 polar-0.121
inn-0.113
Token own
Token ID898
Feature activation-2.32e+00

Pos logit contributions
 them+0.256
mate+0.234
 wreck+0.229
 polar+0.225
inn+0.222
Neg logit contributions
-0.302
's-0.281
cious-0.266
ience-0.262
ages-0.262
Token a
Token ID257
Feature activation+1.24e+00

Pos logit contributions
 them+0.128
 wreck+0.109
 polar+0.101
mate+0.097
 his+0.096
Neg logit contributions
ages-0.184
cious-0.180
aul-0.180
-0.179
ience-0.179
Token measured
Token ID8630
Feature activation-4.75e-01

Pos logit contributions
 them+0.015
 wreck+0.013
mate+0.013
 polar+0.012
inn+0.012
Neg logit contributions
-0.022
cious-0.021
ages-0.021
's-0.021
ience-0.020
Token the
Token ID262
Feature activation+3.48e-01

Pos logit contributions
 wreck+0.027
 polar+0.026
 them+0.026
mate+0.025
 tow+0.023
Neg logit contributions
-0.033
's-0.033
ages-0.030
cious-0.030
aul-0.029
Token ingredients
Token ID9391
Feature activation+4.41e+00

Pos logit contributions
ages+0.029
cious+0.029
ience+0.028
+0.028
 Michelle+0.027
Neg logit contributions
 them-0.015
 wreck-0.015
mate-0.013
 ramp-0.013
 rock-0.013
Token and
Token ID290
Feature activation+5.35e-01

Pos logit contributions
+0.324
's+0.308
cious+0.297
ages+0.297
ience+0.289
Neg logit contributions
 them-0.245
 wreck-0.231
 polar-0.228
mate-0.227
inn-0.209
Token blended
Token ID32063
Feature activation-3.27e+00

Pos logit contributions
+0.039
cious+0.039
ages+0.038
ience+0.038
's+0.036
Neg logit contributions
 them-0.031
 wreck-0.026
mate-0.025
 polar-0.024
inn-0.023
Token them
Token ID606
Feature activation-6.92e+00

Pos logit contributions
 wreck+0.132
mate+0.131
 polar+0.131
 tow+0.115
 them+0.115
Neg logit contributions
-0.284
's-0.275
ages-0.263
cious-0.258
aul-0.256
Token all
Token ID477
Feature activation-2.35e+00

Pos logit contributions
 them+0.353
inn+0.260
 wreck+0.260
mate+0.257
 his+0.245
Neg logit contributions
ience-0.561
cious-0.547
-0.544
 Michelle-0.538
ages-0.532
Token together
Token ID1978
Feature activation-2.81e+00

Pos logit contributions
 them+0.135
 wreck+0.117
mate+0.115
 polar+0.111
inn+0.111
Neg logit contributions
-0.172
cious-0.168
ages-0.166
ience-0.165
's-0.162
Token.
Token ID13
Feature activation+1.25e-01

Pos logit contributions
 them+0.148
 wreck+0.131
mate+0.129
 polar+0.125
inn+0.120
Neg logit contributions
-0.217
's-0.206
cious-0.206
ages-0.205
ience-0.203
Token Finally
Token ID9461
Feature activation-9.28e-01

Pos logit contributions
ience+0.008
cious+0.008
ages+0.008
's+0.008
+0.007
Neg logit contributions
 them-0.009
inn-0.007
 tow-0.007
mate-0.007
 his-0.006
Token,
Token ID11
Feature activation+3.97e-01

Pos logit contributions
 them+0.055
 wreck+0.051
mate+0.050
 polar+0.049
inn+0.047
Neg logit contributions
-0.066
's-0.062
cious-0.061
ages-0.061
ience-0.060
Token Pat
Token ID3208
Feature activation+1.97e+00

Pos logit contributions
+0.018
's+0.017
cious+0.017
ages+0.016
ience+0.016
Neg logit contributions
 them-0.019
 wreck-0.018
mate-0.017
 polar-0.017
inn-0.017
Tokenches
Token ID2052
Feature activation+9.62e-01

Pos logit contributions
+0.114
cious+0.107
ages+0.107
's+0.106
ience+0.104
Neg logit contributions
 them-0.142
 wreck-0.129
mate-0.129
 polar-0.127
inn-0.122
Token got
Token ID1392
Feature activation+4.50e-01

Pos logit contributions
+0.065
ages+0.063
cious+0.063
ience+0.063
's+0.061
Neg logit contributions
 them-0.059
 wreck-0.052
mate-0.051
 polar-0.050
inn-0.047
Token an
Token ID281
Feature activation+2.16e+00

Pos logit contributions
cious+0.034
ience+0.033
ages+0.031
+0.031
aul+0.031
Neg logit contributions
 them-0.027
 wreck-0.023
 polar-0.022
 his-0.022
inn-0.021
Token exciting
Token ID7895
Feature activation+1.90e+00

Pos logit contributions
+0.168
's+0.160
ages+0.155
ience+0.153
cious+0.152
Neg logit contributions
 them-0.116
mate-0.103
 wreck-0.097
 polar-0.094
 tow-0.093
Token idea
Token ID2126
Feature activation-5.88e+00

Pos logit contributions
+0.147
's+0.143
cious+0.139
ages+0.138
ience+0.136
Neg logit contributions
 them-0.099
mate-0.086
 wreck-0.084
 polar-0.083
inn-0.081
Token.
Token ID13
Feature activation+3.56e-03

Pos logit contributions
 them+0.336
 wreck+0.312
 polar+0.302
mate+0.297
inn+0.286
Neg logit contributions
-0.429
's-0.412
cious-0.396
ages-0.395
ience-0.381
Token She
Token ID1375
Feature activation+6.07e-01

Pos logit contributions
ience+0.000
+0.000
cious+0.000
ages+0.000
's+0.000
Neg logit contributions
 them-0.000
 polar-0.000
mate-0.000
 wreck-0.000
inn-0.000
Token decided
Token ID3066
Feature activation+1.22e+00

Pos logit contributions
+0.041
's+0.039
cious+0.039
ages+0.038
ience+0.038
Neg logit contributions
 them-0.038
 wreck-0.034
mate-0.034
 polar-0.033
inn-0.032
Token to
Token ID284
Feature activation+4.75e-01

Pos logit contributions
+0.106
ages+0.106
ience+0.106
cious+0.105
aul+0.105
Neg logit contributions
 them-0.059
 his-0.042
 wreck-0.041
mate-0.039
 rock-0.037
Token put
Token ID1234
Feature activation-1.84e+00

Pos logit contributions
+0.039
's+0.037
cious+0.034
ages+0.034
ience+0.034
Neg logit contributions
 them-0.023
 polar-0.021
mate-0.021
 wreck-0.021
inn-0.020