TOP ACTIVATIONS
MAX = 0.283

make even the most ordinary things special . <|endoftext|> Once upon
some of the most beautiful things in the world . <|endoftext|>
I can find the most things ." So they started counting
it was the most lovely things they ever built . They
creative were the most important things in life .
always creating the most wonderful things with it . <|endoftext|> Once
one of the most beautiful things little Mark had ever seen
loved to imagine the coolest things . One day
love are the most important things in life - always remember
even find the most wonderful things along the way ! <|endoftext|>
day finding the prett iest things to put around the house
to eat the most disgusting things . So one day he
perfect in life are the things you can 't buy ".
sometimes even the most disgusting things can be made better .
full of the most wonderful things . The sack was so
and bird had a special something to make the garden shine
confronted with the most amazing things she 'd ever seen ;
and Lucy had the most things , so they were the
all of Alice 's favourite things , which made Alice so
he found the most amazing things to look at . He

MIN ACTIVATIONS
MIN = -0.308

time ! â <|endoftext|> Pa olo was very excited .
he shook it . <|endoftext|> Pa ola and her mother were
, D ucky !" replied Fro ggy . F
S ally thanked Fro ggie and waved goodbye as
friends are for ," replied Fro ggy . And they lived
stories said . Mama helped Ci ara and together they read
He was happy and thanked Fro ggy for helping him .
was so happy and thanked Fro ggy . They played together
to come with me ," Cic i said to her friends
her brown fur . <|endoftext|> Ark ie wanted his own equipment
their loving friendship . <|endoftext|> Pe ggy was very excited .
dried his hands and followed Jol ene . When Jol ene
lot brighter . <|endoftext|> S oph ie and her dad were
any conflict . <|endoftext|> O liv ia was feeling very restless
bit of magic . <|endoftext|> Isa ac was preparing breakfast in
, a gentle girl named Ci ara stumbled upon a lovely
The man gave Fro ggie a soft cloth and
, Nem o !" said Oscar . " Would you like
bles . Bub bles asked P inch y , " Why
T oby smiled and thanked Ma is ie for being so

INTERVAL 0.135 - 0.283
CONTAINS 2.884%

yellow . And this is a sky . It is blue
the sunlight reflects off the ocean ." The
are so helpful , and I 'm so proud of you
scissors , but it is too hard . She pushes harder
are afraid of everything . I don 't want to play

INTERVAL -0.013 - 0.135
CONTAINS 55.159%

became even happier . The girl never stopped
happy and warm , and they had sweet dreams . <|endoftext|>
with her cr ay ons . One day , her mom
atted her on the arm and said goodbye .
mom and they both enjoyed them . Lily 's mom hugged

INTERVAL -0.160 - -0.013
CONTAINS 39.068%

. She sat down on the bench and started to play
excited to eat it . As Tim my was
ll lead him away from the lake and we can all
time , there was a happy clay . He lived in
he decided to take it home and show his owners .

INTERVAL -0.308 - -0.160
CONTAINS 2.889%

of fun !" Billy thought about this for a
to him ." L ily looked at Sam .
to the truck driver and said , " I can help
L ily felt bad and said , " I know ,
that noise ?" Her friend said , " It 's thunder
Token make
Token ID787
Feature activation+9.16e-03

Pos logit contributions
 spe+0.126
fall+0.122
 long+0.122
 view+0.116
 single+0.112
Neg logit contributions
ma-0.258
aj-0.256
ppa-0.245
annah-0.242
aret-0.234
Token even
Token ID772
Feature activation+2.47e-02

Pos logit contributions
fall+0.012
 spe+0.012
 seem+0.012
 long+0.011
 smell+0.011
Neg logit contributions
aj-0.023
ma-0.022
ppa-0.021
annah-0.021
atted-0.021
Token the
Token ID262
Feature activation+8.15e-03

Pos logit contributions
fall+0.036
 spe+0.034
storm+0.034
 seem+0.034
t+0.034
Neg logit contributions
aj-0.058
ppa-0.054
ma-0.053
ins-0.052
atted-0.052
Token most
Token ID749
Feature activation+1.66e-01

Pos logit contributions
fall+0.014
 seem+0.013
 spe+0.013
 long+0.012
 Pretty+0.012
Neg logit contributions
aj-0.018
ma-0.017
ppa-0.016
annah-0.016
atted-0.016
Token ordinary
Token ID8850
Feature activation+8.01e-02

Pos logit contributions
fall+0.233
 seem+0.224
storm+0.215
 Pretty+0.206
t+0.205
Neg logit contributions
aj-0.407
ma-0.385
ins-0.382
atted-0.382
ppa-0.357
Token things
Token ID1243
Feature activation+2.83e-01

Pos logit contributions
fall+0.125
 spe+0.123
 seem+0.123
 Pretty+0.122
 long+0.114
Neg logit contributions
ma-0.172
aj-0.172
atted-0.162
ins-0.159
ppa-0.157
Token special
Token ID2041
Feature activation+5.22e-02

Pos logit contributions
 spe+0.383
 dot+0.376
fall+0.364
 view+0.349
 long+0.349
Neg logit contributions
ma-0.666
aj-0.639
annah-0.622
atted-0.600
ppa-0.596
Token.
Token ID13
Feature activation+8.52e-02

Pos logit contributions
fall+0.070
 spe+0.070
 long+0.068
 seem+0.066
 smell+0.065
Neg logit contributions
aj-0.129
ma-0.127
annah-0.118
ppa-0.117
atted-0.117
Token<|endoftext|>
Token ID50256
Feature activation-1.16e-01

Pos logit contributions
 long+0.107
 seem+0.107
 spe+0.105
fall+0.105
 smell+0.103
Neg logit contributions
aj-0.227
ppa-0.209
ma-0.207
annah-0.206
atted-0.203
TokenOnce
Token ID7454
Feature activation-1.03e-01

Pos logit contributions
aj+0.301
ma+0.279
ppa+0.276
annah+0.274
atted+0.268
Neg logit contributions
 seem-0.150
 spe-0.149
 long-0.149
fall-0.147
 smell-0.146
Token upon
Token ID2402
Feature activation-7.29e-02

Pos logit contributions
aj+0.257
ppa+0.251
annah+0.231
ins+0.229
anut+0.227
Neg logit contributions
 long-0.163
 seem-0.150
t-0.141
 dark-0.141
 single-0.139
Token some
Token ID617
Feature activation+1.31e-01

Pos logit contributions
aj+0.026
ma+0.024
ppa+0.024
annah+0.023
atted+0.023
Neg logit contributions
 seem-0.012
fall-0.012
 long-0.012
 spe-0.012
 smell-0.011
Token of
Token ID286
Feature activation+8.69e-02

Pos logit contributions
fall+0.168
 seem+0.157
 Pretty+0.155
 dot+0.153
 spe+0.149
Neg logit contributions
aj-0.326
ma-0.324
atted-0.310
ins-0.302
ppa-0.302
Token the
Token ID262
Feature activation+2.88e-02

Pos logit contributions
 seem+0.125
fall+0.124
 long+0.123
 spe+0.122
 smell+0.120
Neg logit contributions
aj-0.213
ma-0.196
ppa-0.196
annah-0.193
atted-0.190
Token most
Token ID749
Feature activation+1.67e-01

Pos logit contributions
 seem+0.035
fall+0.034
 spe+0.031
 long+0.030
 Pretty+0.029
Neg logit contributions
aj-0.077
ma-0.075
ppa-0.071
annah-0.069
atted-0.069
Token beautiful
Token ID4950
Feature activation+8.01e-02

Pos logit contributions
fall+0.209
 seem+0.206
 Pretty+0.184
 long+0.180
 spe+0.179
Neg logit contributions
aj-0.442
ma-0.419
atted-0.413
ins-0.404
ppa-0.398
Token things
Token ID1243
Feature activation+2.79e-01

Pos logit contributions
 seem+0.092
fall+0.088
 long+0.087
 spe+0.086
 smell+0.081
Neg logit contributions
aj-0.221
ma-0.207
atted-0.202
ppa-0.201
annah-0.201
Token in
Token ID287
Feature activation+8.96e-02

Pos logit contributions
 dot+0.358
fall+0.354
 long+0.349
 spe+0.345
 view+0.339
Neg logit contributions
ma-0.670
aj-0.640
annah-0.625
atted-0.621
ppa-0.614
Token the
Token ID262
Feature activation+4.63e-02

Pos logit contributions
fall+0.137
 seem+0.135
 long+0.132
 spe+0.130
 smell+0.128
Neg logit contributions
aj-0.208
ma-0.197
ppa-0.192
annah-0.191
atted-0.188
Token world
Token ID995
Feature activation+2.02e-01

Pos logit contributions
 seem+0.057
fall+0.054
 spe+0.053
 long+0.050
 smell+0.049
Neg logit contributions
aj-0.124
ma-0.119
annah-0.113
ppa-0.112
atted-0.109
Token.
Token ID13
Feature activation+9.29e-02

Pos logit contributions
 long+0.242
 spe+0.241
fall+0.236
 seem+0.236
 smell+0.234
Neg logit contributions
aj-0.527
ma-0.507
annah-0.492
ppa-0.487
atted-0.482
Token<|endoftext|>
Token ID50256
Feature activation-1.21e-01

Pos logit contributions
 long+0.128
 seem+0.127
fall+0.126
 spe+0.126
 smell+0.126
Neg logit contributions
aj-0.233
ma-0.216
ppa-0.215
annah-0.215
atted-0.210
Token I
Token ID314
Feature activation+1.32e-01

Pos logit contributions
 spe+0.030
fall+0.030
 seem+0.029
 smell+0.029
 long+0.029
Neg logit contributions
aj-0.075
ma-0.072
annah-0.070
ppa-0.069
atted-0.068
Token can
Token ID460
Feature activation+1.18e-01

Pos logit contributions
 long+0.118
fall+0.115
 spe+0.114
 view+0.107
 sunset+0.105
Neg logit contributions
aj-0.388
ma-0.370
ppa-0.360
annah-0.357
atted-0.356
Token find
Token ID1064
Feature activation-2.44e-02

Pos logit contributions
 spe+0.159
fall+0.149
 long+0.148
 view+0.139
 dot+0.138
Neg logit contributions
ma-0.288
aj-0.287
annah-0.272
ppa-0.271
atted-0.258
Token the
Token ID262
Feature activation+1.73e-02

Pos logit contributions
aj+0.068
ma+0.064
ppa+0.063
annah+0.062
atted+0.062
Neg logit contributions
 seem-0.026
fall-0.026
 spe-0.026
 long-0.025
 smell-0.025
Token most
Token ID749
Feature activation+1.71e-01

Pos logit contributions
 spe+0.022
 seem+0.022
fall+0.022
 dot+0.020
 long+0.020
Neg logit contributions
ma-0.042
aj-0.042
annah-0.039
ppa-0.039
ins-0.037
Token things
Token ID1243
Feature activation+2.74e-01

Pos logit contributions
fall+0.220
 seem+0.216
 spe+0.206
 smell+0.198
 dot+0.198
Neg logit contributions
aj-0.433
ma-0.418
annah-0.395
ins-0.394
ppa-0.388
Token."
Token ID526
Feature activation-1.06e-02

Pos logit contributions
 spe+0.347
fall+0.339
 dot+0.339
 long+0.324
 view+0.320
Neg logit contributions
aj-0.689
ma-0.678
annah-0.654
ppa-0.636
atted-0.619
Token So
Token ID1406
Feature activation-8.51e-03

Pos logit contributions
aj+0.028
ppa+0.026
ma+0.025
annah+0.025
atted+0.025
Neg logit contributions
 seem-0.014
fall-0.013
 long-0.013
 smell-0.013
 view-0.013
Token they
Token ID484
Feature activation+2.85e-02

Pos logit contributions
aj+0.019
ma+0.018
ppa+0.018
annah+0.018
atted+0.018
Neg logit contributions
 spe-0.013
 seem-0.013
fall-0.013
 long-0.013
 smell-0.013
Token started
Token ID2067
Feature activation+1.23e-03

Pos logit contributions
 long+0.039
 seem+0.039
 spe+0.038
 smell+0.038
fall+0.038
Neg logit contributions
aj-0.074
ppa-0.067
ma-0.067
annah-0.066
atted-0.065
Token counting
Token ID14143
Feature activation-5.85e-02

Pos logit contributions
 seem+0.002
fall+0.002
 spe+0.002
 long+0.002
 smell+0.002
Neg logit contributions
aj-0.003
ma-0.003
atted-0.003
annah-0.003
ppa-0.003
Token it
Token ID340
Feature activation+9.37e-04

Pos logit contributions
aj+0.447
ma+0.411
ppa+0.409
annah+0.408
atted+0.406
Neg logit contributions
 seem-0.236
 spe-0.234
fall-0.232
 long-0.231
 smell-0.227
Token was
Token ID373
Feature activation+3.44e-02

Pos logit contributions
 spe+0.001
fall+0.001
 seem+0.001
 long+0.001
 smell+0.001
Neg logit contributions
aj-0.002
ma-0.002
ppa-0.002
annah-0.002
atted-0.002
Token the
Token ID262
Feature activation+1.62e-02

Pos logit contributions
 seem+0.037
fall+0.036
 spe+0.035
 long+0.035
 smell+0.034
Neg logit contributions
aj-0.098
ma-0.092
ppa-0.090
annah-0.089
atted-0.088
Token most
Token ID749
Feature activation+1.73e-01

Pos logit contributions
fall+0.024
 seem+0.023
 spe+0.022
 Pretty+0.021
t+0.020
Neg logit contributions
aj-0.039
ma-0.039
ppa-0.035
annah-0.034
anut-0.034
Token lovely
Token ID14081
Feature activation+6.18e-02

Pos logit contributions
 seem+0.250
fall+0.244
 spe+0.218
storm+0.217
 Pretty+0.216
Neg logit contributions
aj-0.424
ma-0.402
atted-0.382
ins-0.376
annah-0.367
Token things
Token ID1243
Feature activation+2.74e-01

Pos logit contributions
 seem+0.081
 spe+0.078
fall+0.077
 Pretty+0.073
 dot+0.072
Neg logit contributions
aj-0.154
ma-0.152
annah-0.138
atted-0.138
aret-0.134
Token they
Token ID484
Feature activation+1.54e-01

Pos logit contributions
 dot+0.327
 spe+0.323
 view+0.313
fall+0.311
 long+0.306
Neg logit contributions
ma-0.690
aj-0.682
annah-0.653
atted-0.647
ppa-0.634
Token ever
Token ID1683
Feature activation+6.15e-02

Pos logit contributions
 long+0.179
fall+0.178
 spe+0.177
 seem+0.176
 view+0.171
Neg logit contributions
aj-0.412
ppa-0.389
ma-0.389
annah-0.382
atted-0.371
Token built
Token ID3170
Feature activation-7.25e-03

Pos logit contributions
 spe+0.101
 long+0.100
fall+0.099
 seem+0.097
 view+0.097
Neg logit contributions
aj-0.134
ppa-0.125
ma-0.124
atted-0.119
annah-0.118
Token.
Token ID13
Feature activation+7.45e-02

Pos logit contributions
aj+0.019
ma+0.018
annah+0.018
ppa+0.017
atted+0.017
Neg logit contributions
 spe-0.009
 seem-0.009
fall-0.009
 smell-0.009
 long-0.008
Token They
Token ID1119
Feature activation+7.79e-02

Pos logit contributions
 long+0.094
 seem+0.093
fall+0.092
 spe+0.089
 smell+0.088
Neg logit contributions
aj-0.199
ppa-0.186
ma-0.182
annah-0.179
atted-0.178
Token creative
Token ID7325
Feature activation-6.00e-02

Pos logit contributions
 long+0.068
 spe+0.065
 smell+0.065
 seem+0.065
fall+0.062
Neg logit contributions
aj-0.116
annah-0.105
ppa-0.105
ma-0.104
atted-0.101
Token were
Token ID547
Feature activation+8.77e-02

Pos logit contributions
aj+0.158
ppa+0.145
ma+0.144
annah+0.143
atted+0.141
Neg logit contributions
 seem-0.077
 long-0.076
 spe-0.075
fall-0.075
 smell-0.074
Token the
Token ID262
Feature activation+2.74e-02

Pos logit contributions
 seem+0.109
 long+0.108
fall+0.107
 spe+0.107
 smell+0.105
Neg logit contributions
aj-0.234
ppa-0.216
ma-0.215
annah-0.213
atted-0.210
Token most
Token ID749
Feature activation+1.84e-01

Pos logit contributions
fall+0.046
 spe+0.043
 seem+0.043
 Pretty+0.040
 dot+0.040
Neg logit contributions
ma-0.060
aj-0.059
ppa-0.055
aret-0.053
annah-0.053
Token important
Token ID1593
Feature activation+6.35e-02

Pos logit contributions
fall+0.298
 seem+0.297
 spe+0.278
storm+0.268
 smell+0.267
Neg logit contributions
aj-0.416
ma-0.396
ins-0.379
ppa-0.373
atted-0.371
Token things
Token ID1243
Feature activation+2.74e-01

Pos logit contributions
 spe+0.078
 seem+0.076
fall+0.073
 Pretty+0.067
 long+0.067
Neg logit contributions
aj-0.162
ma-0.160
annah-0.152
ins-0.147
aret-0.146
Token in
Token ID287
Feature activation+8.46e-02

Pos logit contributions
 spe+0.336
 dot+0.313
fall+0.307
 view+0.304
 long+0.294
Neg logit contributions
ma-0.720
aj-0.702
annah-0.674
ppa-0.658
atted-0.653
Token life
Token ID1204
Feature activation+1.03e-01

Pos logit contributions
 long+0.115
 seem+0.114
 spe+0.112
fall+0.112
 smell+0.111
Neg logit contributions
aj-0.218
ppa-0.200
ma-0.198
annah-0.197
atted-0.195
Token.
Token ID13
Feature activation+7.49e-02

Pos logit contributions
 spe+0.124
fall+0.118
 long+0.117
 seem+0.116
 smell+0.116
Neg logit contributions
aj-0.275
ma-0.261
annah-0.253
ppa-0.253
atted-0.246
Token
Token ID220
Feature activation-9.70e-03

Pos logit contributions
 long+0.088
 seem+0.088
fall+0.085
 spe+0.083
 smell+0.082
Neg logit contributions
aj-0.206
ppa-0.193
ma-0.188
annah-0.187
atted-0.185
Token
Token ID198
Feature activation-7.96e-03

Pos logit contributions
ppa+0.030
aj+0.030
atted+0.029
ma+0.028
aret+0.028
Neg logit contributions
 seem-0.007
fall-0.007
 long-0.007
 Pretty-0.006
 A-0.005
Token always
Token ID1464
Feature activation-3.96e-02

Pos logit contributions
 long+0.054
 smell+0.050
 spe+0.050
 seem+0.049
 quite+0.048
Neg logit contributions
aj-0.134
annah-0.118
ppa-0.118
atted-0.115
aret-0.115
Token creating
Token ID4441
Feature activation+1.03e-02

Pos logit contributions
aj+0.100
ppa+0.090
annah+0.089
ma+0.089
atted+0.087
Neg logit contributions
 long-0.058
 seem-0.056
 spe-0.056
 smell-0.055
fall-0.055
Token the
Token ID262
Feature activation-1.00e-02

Pos logit contributions
fall+0.011
 seem+0.011
 spe+0.011
 long+0.011
 smell+0.011
Neg logit contributions
aj-0.029
ma-0.027
ppa-0.026
annah-0.026
atted-0.026
Token most
Token ID749
Feature activation+1.57e-01

Pos logit contributions
aj+0.026
ma+0.025
ppa+0.023
annah+0.023
aret+0.023
Neg logit contributions
fall-0.013
 seem-0.012
 spe-0.012
 long-0.011
t-0.011
Token wonderful
Token ID7932
Feature activation+4.61e-02

Pos logit contributions
fall+0.208
 seem+0.204
 spe+0.181
 smell+0.179
storm+0.177
Neg logit contributions
aj-0.406
ma-0.388
atted-0.377
ins-0.374
ppa-0.367
Token things
Token ID1243
Feature activation+2.73e-01

Pos logit contributions
 seem+0.066
fall+0.065
 spe+0.060
t+0.060
 Pretty+0.058
Neg logit contributions
ma-0.108
aj-0.107
atted-0.100
aret-0.098
annah-0.098
Token with
Token ID351
Feature activation+5.71e-02

Pos logit contributions
 dot+0.354
 spe+0.353
fall+0.343
 long+0.338
 view+0.337
Neg logit contributions
ma-0.667
aj-0.641
annah-0.628
ppa-0.614
atted-0.604
Token it
Token ID340
Feature activation+1.17e-01

Pos logit contributions
 seem+0.088
 long+0.088
 spe+0.086
fall+0.086
 smell+0.085
Neg logit contributions
aj-0.136
ppa-0.124
ma-0.123
annah-0.122
atted-0.120
Token.
Token ID13
Feature activation+8.01e-02

Pos logit contributions
 spe+0.134
fall+0.131
 seem+0.130
 long+0.130
 smell+0.127
Neg logit contributions
aj-0.320
ma-0.301
ppa-0.298
annah-0.296
atted-0.288
Token<|endoftext|>
Token ID50256
Feature activation-1.25e-01

Pos logit contributions
 long+0.106
 spe+0.105
 seem+0.105
fall+0.104
 smell+0.103
Neg logit contributions
aj-0.205
ma-0.191
ppa-0.190
annah-0.189
atted-0.184
TokenOnce
Token ID7454
Feature activation-1.11e-01

Pos logit contributions
aj+0.301
ma+0.283
annah+0.272
ppa+0.270
atted+0.266
Neg logit contributions
 spe-0.184
 seem-0.180
 long-0.178
 smell-0.177
fall-0.173
Token one
Token ID530
Feature activation+1.29e-01

Pos logit contributions
 long+0.083
 spe+0.080
 seem+0.078
 smell+0.078
fall+0.076
Neg logit contributions
aj-0.212
ppa-0.197
annah-0.196
ma-0.194
atted-0.190
Token of
Token ID286
Feature activation+7.66e-02

Pos logit contributions
 seem+0.159
fall+0.157
 smell+0.148
 long+0.148
 spe+0.146
Neg logit contributions
aj-0.342
ma-0.330
atted-0.318
ppa-0.315
annah-0.313
Token the
Token ID262
Feature activation+1.49e-02

Pos logit contributions
 long+0.110
 seem+0.108
 spe+0.108
 smell+0.105
fall+0.104
Neg logit contributions
aj-0.196
ppa-0.178
annah-0.176
ma-0.174
atted-0.171
Token most
Token ID749
Feature activation+1.65e-01

Pos logit contributions
 seem+0.019
fall+0.018
 spe+0.017
 smell+0.016
 Pretty+0.016
Neg logit contributions
aj-0.039
ma-0.037
ppa-0.035
annah-0.034
atted-0.034
Token beautiful
Token ID4950
Feature activation+7.20e-02

Pos logit contributions
fall+0.199
 seem+0.199
 spe+0.183
 smell+0.181
 long+0.180
Neg logit contributions
aj-0.445
ma-0.417
atted-0.412
ppa-0.397
annah-0.396
Token things
Token ID1243
Feature activation+2.73e-01

Pos logit contributions
 seem+0.087
fall+0.082
 spe+0.081
 long+0.078
 Pretty+0.076
Neg logit contributions
aj-0.181
ma-0.180
atted-0.173
aret-0.167
annah-0.166
Token little
Token ID1310
Feature activation+4.03e-02

Pos logit contributions
 spe+0.294
 dot+0.293
fall+0.290
 long+0.289
 view+0.288
Neg logit contributions
ma-0.707
aj-0.694
atted-0.672
annah-0.663
ppa-0.657
Token Mark
Token ID2940
Feature activation-4.23e-02

Pos logit contributions
fall+0.059
 seem+0.057
 spe+0.056
 smell+0.054
 long+0.053
Neg logit contributions
aj-0.092
ma-0.090
atted-0.086
annah-0.082
ppa-0.081
Token had
Token ID550
Feature activation+1.36e-01

Pos logit contributions
aj+0.120
ppa+0.109
annah+0.109
ma+0.108
atted+0.107
Neg logit contributions
 long-0.047
 seem-0.047
 spe-0.046
fall-0.045
 smell-0.045
Token ever
Token ID1683
Feature activation+5.35e-02

Pos logit contributions
 long+0.130
 seem+0.129
fall+0.125
 smell+0.121
 spe+0.119
Neg logit contributions
aj-0.406
annah-0.375
ppa-0.375
ma-0.371
atted-0.370
Token seen
Token ID1775
Feature activation-2.04e-03

Pos logit contributions
 spe+0.092
 long+0.091
fall+0.090
 seem+0.090
 smell+0.088
Neg logit contributions
aj-0.116
ppa-0.106
ma-0.105
atted-0.102
annah-0.102
Token loved
Token ID6151
Feature activation-3.35e-02

Pos logit contributions
 long+0.038
 seem+0.038
 smell+0.038
 spe+0.037
fall+0.036
Neg logit contributions
aj-0.076
ppa-0.070
annah-0.069
ma-0.068
atted-0.067
Token to
Token ID284
Feature activation+1.46e-02

Pos logit contributions
aj+0.084
ma+0.080
atted+0.077
ppa+0.076
annah+0.075
Neg logit contributions
 seem-0.044
fall-0.043
 dot-0.041
 spe-0.040
t-0.039
Token imagine
Token ID5967
Feature activation+2.12e-03

Pos logit contributions
fall+0.015
 spe+0.015
 long+0.014
 seem+0.014
 dot+0.014
Neg logit contributions
aj-0.041
ma-0.038
ppa-0.038
annah-0.038
atted-0.037
Token the
Token ID262
Feature activation+6.13e-03

Pos logit contributions
 spe+0.003
fall+0.003
 dot+0.003
 seem+0.003
 smell+0.003
Neg logit contributions
aj-0.005
ma-0.005
atted-0.005
ppa-0.005
ins-0.004
Token coolest
Token ID38889
Feature activation+8.15e-02

Pos logit contributions
fall+0.008
 seem+0.008
 Pretty+0.007
 spe+0.007
t+0.007
Neg logit contributions
aj-0.015
ma-0.015
ppa-0.014
aret-0.013
annah-0.013
Token things
Token ID1243
Feature activation+2.72e-01

Pos logit contributions
 seem+0.110
fall+0.106
 spe+0.103
 Pretty+0.094
 long+0.093
Neg logit contributions
ma-0.204
aj-0.198
ins-0.184
atted-0.183
annah-0.180
Token.
Token ID13
Feature activation+8.33e-02

Pos logit contributions
 spe+0.299
 dot+0.298
 long+0.288
fall+0.287
 view+0.280
Neg logit contributions
ma-0.710
aj-0.706
annah-0.669
ppa-0.665
atted-0.657
Token
Token ID198
Feature activation-4.43e-04

Pos logit contributions
 long+0.095
 seem+0.095
fall+0.092
 spe+0.092
 smell+0.090
Neg logit contributions
aj-0.231
ppa-0.215
ma-0.213
annah-0.211
atted-0.209
Token
Token ID198
Feature activation-3.82e-03

Pos logit contributions
aj+0.001
ppa+0.001
atted+0.001
annah+0.001
aret+0.001
Neg logit contributions
 long-0.000
 view-0.000
 single-0.000
 lot-0.000
 much-0.000
TokenOne
Token ID3198
Feature activation+2.88e-02

Pos logit contributions
aj+0.012
atted+0.011
ppa+0.011
annah+0.011
ma+0.010
Neg logit contributions
 long-0.004
 view-0.003
 seem-0.003
 single-0.003
fall-0.003
Token day
Token ID1110
Feature activation+5.34e-02

Pos logit contributions
fall+0.024
 seem+0.024
 spe+0.024
 smell+0.021
 dot+0.021
Neg logit contributions
aj-0.087
ma-0.085
annah-0.081
ppa-0.080
atted-0.080
Token love
Token ID1842
Feature activation+2.85e-02

Pos logit contributions
fall+0.052
 spe+0.050
 seem+0.049
 long+0.049
 view+0.048
Neg logit contributions
aj-0.087
ma-0.083
ppa-0.081
annah-0.080
atted-0.080
Token are
Token ID389
Feature activation+1.13e-01

Pos logit contributions
 seem+0.039
fall+0.039
 spe+0.039
 smell+0.038
 long+0.037
Neg logit contributions
aj-0.071
ma-0.067
ppa-0.065
annah-0.065
atted-0.063
Token the
Token ID262
Feature activation+3.30e-02

Pos logit contributions
 long+0.137
 seem+0.137
 spe+0.134
fall+0.133
 smell+0.132
Neg logit contributions
aj-0.304
ppa-0.281
ma-0.279
annah-0.278
atted-0.273
Token most
Token ID749
Feature activation+1.81e-01

Pos logit contributions
fall+0.049
 seem+0.046
 spe+0.045
 Pretty+0.044
 dot+0.043
Neg logit contributions
ma-0.076
aj-0.074
ppa-0.069
aret-0.067
annah-0.067
Token important
Token ID1593
Feature activation+6.16e-02

Pos logit contributions
fall+0.255
 seem+0.254
 smell+0.237
 spe+0.235
 long+0.235
Neg logit contributions
aj-0.449
ma-0.423
atted-0.409
ppa-0.407
annah-0.404
Token things
Token ID1243
Feature activation+2.69e-01

Pos logit contributions
 seem+0.074
fall+0.072
 spe+0.071
 Pretty+0.066
 dot+0.065
Neg logit contributions
aj-0.151
ma-0.148
annah-0.141
aret-0.138
atted-0.137
Token in
Token ID287
Feature activation+7.56e-02

Pos logit contributions
 spe+0.353
fall+0.342
 long+0.337
 dot+0.335
 view+0.331
Neg logit contributions
aj-0.676
ma-0.654
annah-0.635
ppa-0.630
atted-0.608
Token life
Token ID1204
Feature activation+9.48e-02

Pos logit contributions
fall+0.105
 seem+0.105
 spe+0.104
 long+0.103
 smell+0.101
Neg logit contributions
aj-0.187
ma-0.174
ppa-0.173
annah-0.171
atted-0.167
Token -
Token ID532
Feature activation+7.29e-02

Pos logit contributions
 seem+0.120
 long+0.120
 spe+0.119
fall+0.118
 smell+0.117
Neg logit contributions
aj-0.249
ma-0.229
ppa-0.229
annah-0.227
atted-0.222
Token always
Token ID1464
Feature activation-3.08e-02

Pos logit contributions
fall+0.091
 seem+0.088
 spe+0.087
 long+0.085
 smell+0.084
Neg logit contributions
aj-0.187
ma-0.177
ppa-0.170
atted-0.170
annah-0.170
Token remember
Token ID3505
Feature activation-2.73e-02

Pos logit contributions
aj+0.075
ma+0.070
ppa+0.069
annah+0.068
atted+0.067
Neg logit contributions
fall-0.043
 spe-0.043
 seem-0.043
 long-0.043
 smell-0.042
Token even
Token ID772
Feature activation+4.58e-02

Pos logit contributions
fall+0.059
 spe+0.059
 dot+0.056
 seem+0.056
 long+0.055
Neg logit contributions
aj-0.118
ma-0.111
ppa-0.109
annah-0.108
atted-0.106
Token find
Token ID1064
Feature activation-2.83e-02

Pos logit contributions
fall+0.060
 spe+0.059
 long+0.058
 seem+0.058
 view+0.057
Neg logit contributions
aj-0.116
ma-0.108
ppa-0.108
annah-0.105
atted-0.105
Token the
Token ID262
Feature activation+5.89e-03

Pos logit contributions
aj+0.083
ppa+0.077
ma+0.076
annah+0.076
atted+0.075
Neg logit contributions
 long-0.028
 seem-0.028
 spe-0.027
fall-0.027
 smell-0.027
Token most
Token ID749
Feature activation+1.66e-01

Pos logit contributions
fall+0.008
 seem+0.008
 spe+0.008
 Pretty+0.007
 dot+0.007
Neg logit contributions
aj-0.014
ma-0.014
ppa-0.013
annah-0.013
aret-0.012
Token wonderful
Token ID7932
Feature activation+4.44e-02

Pos logit contributions
fall+0.224
 seem+0.218
 spe+0.203
 smell+0.199
 dot+0.195
Neg logit contributions
aj-0.418
ma-0.396
ins-0.376
atted-0.373
annah-0.372
Token things
Token ID1243
Feature activation+2.68e-01

Pos logit contributions
 seem+0.054
fall+0.053
 spe+0.051
t+0.049
 long+0.048
Neg logit contributions
aj-0.116
ma-0.111
annah-0.105
atted-0.104
ins-0.103
Token along
Token ID1863
Feature activation+1.34e-03

Pos logit contributions
 spe+0.335
 long+0.328
fall+0.325
 dot+0.321
 view+0.314
Neg logit contributions
aj-0.691
ma-0.664
annah-0.647
ppa-0.640
atted-0.625
Token the
Token ID262
Feature activation+4.93e-03

Pos logit contributions
 seem+0.002
 long+0.002
 spe+0.002
fall+0.002
 smell+0.002
Neg logit contributions
aj-0.003
ppa-0.003
ma-0.003
annah-0.003
atted-0.003
Token way
Token ID835
Feature activation+6.90e-02

Pos logit contributions
fall+0.007
 seem+0.007
 spe+0.007
 smell+0.006
 long+0.006
Neg logit contributions
aj-0.012
ma-0.012
ppa-0.011
annah-0.011
aret-0.011
Token!
Token ID0
Feature activation+6.95e-02

Pos logit contributions
 seem+0.077
 long+0.076
 spe+0.076
fall+0.075
 smell+0.074
Neg logit contributions
aj-0.192
ma-0.178
ppa-0.178
annah-0.176
atted-0.173
Token<|endoftext|>
Token ID50256
Feature activation-1.14e-01

Pos logit contributions
 long+0.090
 seem+0.089
 spe+0.089
fall+0.088
 smell+0.088
Neg logit contributions
aj-0.181
ma-0.167
ppa-0.167
annah-0.166
atted-0.162
Token day
Token ID1110
Feature activation+7.96e-02

Pos logit contributions
 long+0.069
 seem+0.067
 spe+0.065
 smell+0.062
fall+0.062
Neg logit contributions
aj-0.383
ppa-0.357
annah-0.352
ma-0.351
atted-0.347
Token finding
Token ID4917
Feature activation-1.18e-02

Pos logit contributions
 seem+0.120
 spe+0.118
fall+0.118
 long+0.117
 smell+0.117
Neg logit contributions
aj-0.190
ppa-0.174
ma-0.173
annah-0.172
atted-0.168
Token the
Token ID262
Feature activation-7.22e-03

Pos logit contributions
aj+0.030
ma+0.029
atted+0.028
ppa+0.027
ins+0.027
Neg logit contributions
fall-0.016
 seem-0.016
 spe-0.015
storm-0.015
 long-0.014
Token prett
Token ID46442
Feature activation+7.31e-02

Pos logit contributions
aj+0.017
ma+0.017
ins+0.015
annah+0.014
ppa+0.014
Neg logit contributions
 seem-0.011
fall-0.011
storm-0.010
 spe-0.010
 Pretty-0.010
Tokeniest
Token ID6386
Feature activation+9.06e-02

Pos logit contributions
 long+0.081
 spe+0.081
 seem+0.081
fall+0.080
 smell+0.078
Neg logit contributions
aj-0.205
ma-0.193
ppa-0.190
annah-0.188
atted-0.187
Token things
Token ID1243
Feature activation+2.68e-01

Pos logit contributions
 seem+0.158
fall+0.152
 spe+0.143
storm+0.139
 sunset+0.135
Neg logit contributions
aj-0.206
ma-0.198
ins-0.177
atted-0.174
annah-0.173
Token to
Token ID284
Feature activation+6.95e-02

Pos logit contributions
 dot+0.352
 spe+0.347
fall+0.330
 view+0.322
 sunset+0.313
Neg logit contributions
ma-0.656
aj-0.626
atted-0.599
annah-0.594
ppa-0.591
Token put
Token ID1234
Feature activation-4.57e-02

Pos logit contributions
fall+0.086
 long+0.083
 spe+0.083
 seem+0.082
 view+0.079
Neg logit contributions
aj-0.179
ma-0.172
ppa-0.167
annah-0.167
atted-0.160
Token around
Token ID1088
Feature activation-5.94e-03

Pos logit contributions
aj+0.122
ma+0.116
ppa+0.113
annah+0.112
atted+0.110
Neg logit contributions
 seem-0.055
fall-0.054
 spe-0.053
 long-0.052
 smell-0.052
Token the
Token ID262
Feature activation-1.44e-02

Pos logit contributions
aj+0.014
ma+0.014
ppa+0.013
annah+0.013
atted+0.013
Neg logit contributions
 seem-0.008
fall-0.008
 spe-0.008
 smell-0.008
 long-0.008
Token house
Token ID2156
Feature activation+9.63e-02

Pos logit contributions
aj+0.032
ma+0.030
annah+0.028
ppa+0.027
ins+0.026
Neg logit contributions
 seem-0.026
fall-0.025
 spe-0.025
 smell-0.023
 long-0.023
Token to
Token ID284
Feature activation+1.88e-02

Pos logit contributions
aj+0.084
ma+0.081
ppa+0.078
atted+0.076
annah+0.075
Neg logit contributions
 seem-0.040
fall-0.039
 dot-0.038
 spe-0.037
 smell-0.035
Token eat
Token ID4483
Feature activation-5.99e-02

Pos logit contributions
 spe+0.020
fall+0.020
 seem+0.019
 long+0.019
 dot+0.019
Neg logit contributions
aj-0.052
ma-0.049
ppa-0.048
annah-0.048
aret-0.046
Token the
Token ID262
Feature activation-3.29e-02

Pos logit contributions
aj+0.155
ma+0.144
ppa+0.141
annah+0.140
atted+0.137
Neg logit contributions
 seem-0.078
fall-0.077
 spe-0.076
 dot-0.073
 long-0.073
Token most
Token ID749
Feature activation+1.53e-01

Pos logit contributions
aj+0.076
ma+0.073
ins+0.068
ppa+0.066
aret+0.065
Neg logit contributions
 seem-0.050
 spe-0.048
fall-0.047
 Pretty-0.045
 dot-0.045
Token disgusting
Token ID23374
Feature activation+7.26e-02

Pos logit contributions
 seem+0.235
fall+0.231
 spe+0.224
 dot+0.217
 Pretty+0.215
Neg logit contributions
aj-0.360
ins-0.342
ma-0.341
atted-0.321
ppa-0.315
Token things
Token ID1243
Feature activation+2.67e-01

Pos logit contributions
 seem+0.103
 Pretty+0.102
 spe+0.101
fall+0.098
t+0.093
Neg logit contributions
ma-0.164
aj-0.162
ins-0.156
annah-0.146
atted-0.145
Token.
Token ID13
Feature activation+8.14e-02

Pos logit contributions
 spe+0.377
 dot+0.370
fall+0.357
 view+0.348
 Pretty+0.332
Neg logit contributions
ma-0.638
aj-0.610
annah-0.587
ppa-0.583
atted-0.565
Token So
Token ID1406
Feature activation+1.39e-02

Pos logit contributions
 long+0.094
 seem+0.093
fall+0.091
 spe+0.090
 smell+0.090
Neg logit contributions
aj-0.226
ppa-0.211
ma-0.208
annah-0.207
atted-0.205
Token one
Token ID530
Feature activation+7.94e-02

Pos logit contributions
 seem+0.021
 spe+0.021
fall+0.021
 smell+0.020
 dot+0.020
Neg logit contributions
aj-0.033
ma-0.031
ppa-0.030
annah-0.030
atted-0.029
Token day
Token ID1110
Feature activation+6.26e-02

Pos logit contributions
 seem+0.074
fall+0.073
 spe+0.071
 smell+0.061
 long+0.060
Neg logit contributions
aj-0.233
ma-0.232
ppa-0.220
annah-0.211
atted-0.211
Token he
Token ID339
Feature activation+2.76e-02

Pos logit contributions
 seem+0.125
 spe+0.125
 dot+0.123
 smell+0.121
fall+0.117
Neg logit contributions
aj-0.109
ma-0.105
atted-0.103
annah-0.098
ppa-0.097
Token perfect
Token ID2818
Feature activation+3.00e-02

Pos logit contributions
 seem+0.181
 spe+0.181
fall+0.177
 dot+0.173
 smell+0.169
Neg logit contributions
aj-0.335
ma-0.313
ppa-0.311
annah-0.308
atted-0.295
Token in
Token ID287
Feature activation+7.69e-02

Pos logit contributions
 spe+0.042
 seem+0.042
fall+0.040
 long+0.040
 smell+0.039
Neg logit contributions
aj-0.075
ma-0.070
annah-0.069
ppa-0.068
atted-0.065
Token life
Token ID1204
Feature activation+9.55e-02

Pos logit contributions
 seem+0.125
fall+0.124
 long+0.124
 spe+0.123
 smell+0.122
Neg logit contributions
aj-0.175
ma-0.161
ppa-0.159
annah-0.158
atted-0.153
Token are
Token ID389
Feature activation+1.51e-01

Pos logit contributions
 spe+0.122
 seem+0.121
fall+0.119
 long+0.119
 smell+0.118
Neg logit contributions
aj-0.251
ma-0.233
ppa-0.230
annah-0.230
atted-0.221
Token the
Token ID262
Feature activation+6.05e-02

Pos logit contributions
 long+0.185
 seem+0.185
 spe+0.181
fall+0.181
 smell+0.178
Neg logit contributions
aj-0.403
ppa-0.372
ma-0.370
annah-0.368
atted-0.362
Token things
Token ID1243
Feature activation+2.67e-01

Pos logit contributions
 seem+0.081
fall+0.080
 spe+0.076
 dot+0.070
 smell+0.068
Neg logit contributions
ma-0.154
aj-0.151
annah-0.135
ppa-0.133
eline-0.131
Token you
Token ID345
Feature activation+1.39e-01

Pos logit contributions
 spe+0.361
 seem+0.339
fall+0.339
 dot+0.337
 view+0.335
Neg logit contributions
aj-0.670
ma-0.645
annah-0.632
ppa-0.623
atted-0.586
Token can
Token ID460
Feature activation+1.48e-01

Pos logit contributions
 spe+0.159
fall+0.150
 long+0.149
 dot+0.143
 view+0.143
Neg logit contributions
aj-0.381
ppa-0.355
ma-0.355
annah-0.352
atted-0.339
Token't
Token ID470
Feature activation+7.85e-02

Pos logit contributions
 spe+0.200
 long+0.188
fall+0.184
 view+0.177
 dot+0.177
Neg logit contributions
aj-0.373
ma-0.357
ppa-0.346
annah-0.345
aret-0.328
Token buy
Token ID2822
Feature activation-2.70e-02

Pos logit contributions
 spe+0.123
 dot+0.106
fall+0.104
 Pretty+0.104
 long+0.103
Neg logit contributions
ma-0.179
ppa-0.172
annah-0.170
aj-0.165
eline-0.162
Token".
Token ID1911
Feature activation+1.83e-02

Pos logit contributions
aj+0.070
ma+0.065
ppa+0.064
annah+0.064
atted+0.062
Neg logit contributions
 spe-0.036
 seem-0.035
fall-0.035
 long-0.035
 smell-0.034
Token sometimes
Token ID3360
Feature activation+5.40e-02

Pos logit contributions
fall+0.036
 spe+0.035
 long+0.034
 smell+0.033
 view+0.033
Neg logit contributions
aj-0.059
ma-0.059
atted-0.055
ppa-0.054
annah-0.054
Token even
Token ID772
Feature activation+2.55e-02

Pos logit contributions
fall+0.069
 spe+0.068
 dot+0.063
 smell+0.063
 view+0.062
Neg logit contributions
aj-0.131
ma-0.130
annah-0.124
ppa-0.122
atted-0.120
Token the
Token ID262
Feature activation+5.05e-03

Pos logit contributions
fall+0.036
 spe+0.035
 view+0.033
storm+0.033
 dot+0.033
Neg logit contributions
aj-0.059
ma-0.055
atted-0.054
ppa-0.054
annah-0.052
Token most
Token ID749
Feature activation+1.51e-01

Pos logit contributions
fall+0.008
 spe+0.008
 seem+0.007
 Pretty+0.007
br+0.007
Neg logit contributions
ma-0.011
aj-0.010
annah-0.009
ppa-0.009
aret-0.009
Token disgusting
Token ID23374
Feature activation+7.96e-02

Pos logit contributions
fall+0.236
 seem+0.217
 spe+0.213
 smell+0.210
 dot+0.208
Neg logit contributions
aj-0.333
ma-0.319
atted-0.317
ins-0.309
annah-0.298
Token things
Token ID1243
Feature activation+2.66e-01

Pos logit contributions
 spe+0.105
fall+0.101
 Pretty+0.096
 long+0.095
 seem+0.094
Neg logit contributions
ma-0.190
aj-0.188
annah-0.178
atted-0.173
ins-0.173
Token can
Token ID460
Feature activation+1.48e-01

Pos logit contributions
 spe+0.305
fall+0.294
 long+0.293
 dot+0.291
 view+0.289
Neg logit contributions
aj-0.689
ma-0.684
annah-0.658
atted-0.653
ppa-0.635
Token be
Token ID307
Feature activation+7.62e-02

Pos logit contributions
 spe+0.191
fall+0.178
 long+0.178
 view+0.170
 dot+0.167
Neg logit contributions
ma-0.364
aj-0.360
ppa-0.346
annah-0.346
aret-0.331
Token made
Token ID925
Feature activation+3.26e-02

Pos logit contributions
 seem+0.098
fall+0.097
 long+0.097
 spe+0.097
 smell+0.095
Neg logit contributions
aj-0.198
ppa-0.183
ma-0.182
annah-0.180
atted-0.177
Token better
Token ID1365
Feature activation-3.41e-02

Pos logit contributions
 spe+0.042
 long+0.041
 seem+0.041
fall+0.041
 smell+0.040
Neg logit contributions
aj-0.085
ma-0.079
ppa-0.078
annah-0.078
atted-0.076
Token.
Token ID13
Feature activation+7.92e-02

Pos logit contributions
aj+0.077
ma+0.075
ins+0.073
atted+0.071
ppa+0.069
Neg logit contributions
 spe-0.050
fall-0.048
 Pretty-0.045
 smell-0.045
 dot-0.045
Token full
Token ID1336
Feature activation+3.12e-02

Pos logit contributions
 long+0.051
 seem+0.049
 spe+0.048
 smell+0.047
fall+0.047
Neg logit contributions
aj-0.208
ppa-0.194
ma-0.193
annah-0.193
atted-0.190
Token of
Token ID286
Feature activation+5.99e-02

Pos logit contributions
 seem+0.040
fall+0.040
 spe+0.039
 view+0.037
 smell+0.037
Neg logit contributions
aj-0.081
ma-0.076
ppa-0.074
atted-0.074
annah-0.074
Token the
Token ID262
Feature activation+1.37e-03

Pos logit contributions
fall+0.096
 seem+0.089
 spe+0.085
storm+0.082
t+0.082
Neg logit contributions
aj-0.141
ma-0.132
ppa-0.129
atted-0.128
annah-0.126
Token most
Token ID749
Feature activation+1.59e-01

Pos logit contributions
fall+0.002
 seem+0.002
 spe+0.002
 Pretty+0.002
t+0.002
Neg logit contributions
aj-0.003
ma-0.003
ppa-0.003
annah-0.003
eline-0.003
Token wonderful
Token ID7932
Feature activation+3.94e-02

Pos logit contributions
fall+0.252
 seem+0.242
 spe+0.221
 Pretty+0.220
storm+0.217
Neg logit contributions
aj-0.386
ma-0.354
ins-0.344
atted-0.343
ppa-0.341
Token things
Token ID1243
Feature activation+2.66e-01

Pos logit contributions
fall+0.063
 seem+0.060
 spe+0.057
 Pretty+0.057
t+0.056
Neg logit contributions
aj-0.089
ma-0.084
annah-0.083
ppa-0.080
atted-0.080
Token.
Token ID13
Feature activation+8.60e-02

Pos logit contributions
 dot+0.369
fall+0.364
 spe+0.363
 view+0.362
 long+0.347
Neg logit contributions
ma-0.613
aj-0.586
ppa-0.567
annah-0.564
atted-0.562
Token The
Token ID383
Feature activation-4.88e-02

Pos logit contributions
 spe+0.114
fall+0.112
 smell+0.108
 seem+0.107
 long+0.105
Neg logit contributions
aj-0.213
ma-0.200
annah-0.196
ppa-0.185
atted-0.184
Token sack
Token ID23704
Feature activation+8.61e-02

Pos logit contributions
aj+0.114
ma+0.105
ppa+0.104
annah+0.102
aret+0.096
Neg logit contributions
fall-0.077
 spe-0.076
 seem-0.075
 Pretty-0.067
 single-0.066
Token was
Token ID373
Feature activation+8.07e-02

Pos logit contributions
 long+0.092
 smell+0.092
 seem+0.090
 spe+0.087
 view+0.083
Neg logit contributions
aj-0.253
ppa-0.233
annah-0.229
atted-0.223
aret-0.222
Token so
Token ID523
Feature activation+1.22e-01

Pos logit contributions
 long+0.073
 smell+0.064
 spe+0.064
 seem+0.059
 view+0.057
Neg logit contributions
aj-0.257
ppa-0.240
annah-0.239
ma-0.237
atted-0.236
Token and
Token ID290
Feature activation+1.27e-02

Pos logit contributions
 long+0.045
 seem+0.044
 spe+0.044
 smell+0.043
fall+0.043
Neg logit contributions
aj-0.088
ppa-0.081
ma-0.081
annah-0.080
atted-0.079
Token bird
Token ID6512
Feature activation+6.44e-03

Pos logit contributions
 seem+0.016
fall+0.016
 long+0.016
 spe+0.015
 smell+0.015
Neg logit contributions
aj-0.033
ma-0.031
ppa-0.031
annah-0.030
atted-0.030
Token had
Token ID550
Feature activation+8.50e-02

Pos logit contributions
 long+0.009
 seem+0.009
 smell+0.009
 spe+0.009
fall+0.008
Neg logit contributions
aj-0.016
annah-0.015
ppa-0.015
ma-0.015
aret-0.014
Token a
Token ID257
Feature activation+1.63e-01

Pos logit contributions
 seem+0.091
 long+0.090
fall+0.089
 spe+0.089
 smell+0.087
Neg logit contributions
aj-0.241
ppa-0.224
ma-0.223
annah-0.221
atted-0.218
Token special
Token ID2041
Feature activation+6.19e-02

Pos logit contributions
 seem+0.197
fall+0.190
 spe+0.160
 Pretty+0.158
 dot+0.158
Neg logit contributions
aj-0.446
ma-0.415
ppa-0.399
annah-0.398
atted-0.384
Token something
Token ID1223
Feature activation+2.64e-01

Pos logit contributions
 seem+0.086
fall+0.081
 spe+0.077
 Pretty+0.073
 dot+0.071
Neg logit contributions
aj-0.157
ma-0.148
annah-0.139
atted-0.138
ppa-0.135
Token to
Token ID284
Feature activation+5.64e-02

Pos logit contributions
fall+0.330
 dot+0.330
 seem+0.322
 view+0.313
 spe+0.310
Neg logit contributions
aj-0.657
ma-0.630
annah-0.606
atted-0.596
ppa-0.596
Token make
Token ID787
Feature activation+3.42e-02

Pos logit contributions
 seem+0.061
 long+0.061
 spe+0.059
fall+0.059
 smell+0.059
Neg logit contributions
aj-0.160
ppa-0.149
ma-0.148
annah-0.147
atted-0.145
Token the
Token ID262
Feature activation-3.34e-03

Pos logit contributions
 spe+0.039
 seem+0.039
 long+0.039
fall+0.039
 smell+0.037
Neg logit contributions
aj-0.093
ma-0.087
ppa-0.086
annah-0.086
atted-0.084
Token garden
Token ID11376
Feature activation+1.29e-01

Pos logit contributions
aj+0.008
ma+0.008
ppa+0.008
annah+0.007
anut+0.007
Neg logit contributions
 seem-0.004
fall-0.004
 spe-0.004
 Pretty-0.004
t-0.004
Token shine
Token ID18340
Feature activation+9.61e-02

Pos logit contributions
fall+0.156
 spe+0.151
 Pretty+0.148
 seem+0.147
 dot+0.144
Neg logit contributions
aj-0.334
ma-0.317
ppa-0.302
annah-0.300
atted-0.296
Token confronted
Token ID16675
Feature activation-1.35e-01

Pos logit contributions
 long+0.049
 seem+0.049
 spe+0.048
fall+0.048
 smell+0.047
Neg logit contributions
aj-0.137
ppa-0.128
ma-0.127
annah-0.126
atted-0.124
Token with
Token ID351
Feature activation-2.22e-02

Pos logit contributions
aj+0.368
ppa+0.339
annah+0.334
ma+0.329
atted+0.327
Neg logit contributions
 long-0.171
 seem-0.161
 smell-0.160
fall-0.159
 spe-0.159
Token the
Token ID262
Feature activation-1.63e-02

Pos logit contributions
aj+0.063
ma+0.058
ppa+0.058
annah+0.057
atted+0.057
Neg logit contributions
 seem-0.024
fall-0.023
 spe-0.023
 long-0.023
 smell-0.022
Token most
Token ID749
Feature activation+1.52e-01

Pos logit contributions
aj+0.041
ma+0.039
ppa+0.037
annah+0.037
atted+0.036
Neg logit contributions
 seem-0.022
 spe-0.022
fall-0.022
 long-0.021
 dot-0.020
Token amazing
Token ID4998
Feature activation+3.47e-02

Pos logit contributions
fall+0.188
 seem+0.188
 spe+0.176
 smell+0.172
 Pretty+0.168
Neg logit contributions
aj-0.403
ma-0.371
atted-0.369
ins-0.360
annah-0.357
Token things
Token ID1243
Feature activation+2.64e-01

Pos logit contributions
 seem+0.049
fall+0.046
 spe+0.046
 Pretty+0.045
 dot+0.044
Neg logit contributions
aj-0.086
ma-0.082
atted-0.078
annah-0.078
ins-0.075
Token she
Token ID673
Feature activation+1.15e-01

Pos logit contributions
 spe+0.325
 long+0.316
fall+0.313
 dot+0.313
 view+0.309
Neg logit contributions
aj-0.666
ma-0.660
annah-0.633
atted-0.626
ppa-0.620
Token'd
Token ID1549
Feature activation+1.23e-01

Pos logit contributions
 long+0.135
 seem+0.135
 spe+0.133
fall+0.133
 smell+0.130
Neg logit contributions
aj-0.312
ppa-0.289
ma-0.289
annah-0.286
atted-0.280
Token ever
Token ID1683
Feature activation+5.53e-02

Pos logit contributions
 spe+0.085
 long+0.084
fall+0.083
 seem+0.081
 smell+0.076
Neg logit contributions
aj-0.386
ma-0.367
ppa-0.364
annah-0.362
atted-0.354
Token seen
Token ID1775
Feature activation+6.38e-03

Pos logit contributions
 spe+0.095
 long+0.094
fall+0.093
 seem+0.092
 smell+0.091
Neg logit contributions
aj-0.119
ppa-0.109
ma-0.108
annah-0.105
atted-0.104
Token;
Token ID26
Feature activation+3.30e-02

Pos logit contributions
 long+0.006
 seem+0.006
fall+0.006
 spe+0.006
 smell+0.006
Neg logit contributions
aj-0.019
ppa-0.018
annah-0.017
ma-0.017
atted-0.017
Token and
Token ID290
Feature activation-7.97e-03

Pos logit contributions
aj+0.364
ppa+0.325
annah+0.322
aret+0.317
ma+0.308
Neg logit contributions
 spe-0.165
 long-0.162
 smell-0.161
 seem-0.160
 ever-0.150
Token Lucy
Token ID22162
Feature activation-1.34e-01

Pos logit contributions
aj+0.019
ppa+0.017
annah+0.016
ma+0.016
atted+0.016
Neg logit contributions
 seem-0.013
 long-0.013
 smell-0.013
 spe-0.013
fall-0.013
Token had
Token ID550
Feature activation+6.57e-02

Pos logit contributions
aj+0.344
ppa+0.317
ma+0.316
annah+0.313
atted+0.308
Neg logit contributions
 long-0.178
 seem-0.178
fall-0.175
 spe-0.175
 smell-0.172
Token the
Token ID262
Feature activation+6.18e-05

Pos logit contributions
fall+0.061
 seem+0.060
 spe+0.059
 long+0.059
 smell+0.057
Neg logit contributions
aj-0.194
ma-0.182
ppa-0.181
annah-0.179
atted-0.177
Token most
Token ID749
Feature activation+1.60e-01

Pos logit contributions
fall+0.000
 spe+0.000
 seem+0.000
 dot+0.000
 long+0.000
Neg logit contributions
aj-0.000
ma-0.000
annah-0.000
ppa-0.000
atted-0.000
Token things
Token ID1243
Feature activation+2.64e-01

Pos logit contributions
fall+0.199
 seem+0.189
 spe+0.174
 dot+0.174
storm+0.171
Neg logit contributions
aj-0.418
ma-0.405
atted-0.388
annah-0.386
ins-0.385
Token,
Token ID11
Feature activation+1.29e-02

Pos logit contributions
fall+0.289
 dot+0.288
 spe+0.285
 view+0.280
 long+0.277
Neg logit contributions
ma-0.689
aj-0.687
annah-0.661
atted-0.658
ppa-0.651
Token so
Token ID523
Feature activation+1.35e-01

Pos logit contributions
fall+0.013
 seem+0.012
 spe+0.012
 dot+0.012
 smell+0.012
Neg logit contributions
aj-0.036
ma-0.034
ppa-0.034
annah-0.033
atted-0.033
Token they
Token ID484
Feature activation+1.00e-01

Pos logit contributions
 seem+0.126
fall+0.123
 spe+0.121
 long+0.118
 smell+0.118
Neg logit contributions
aj-0.401
ma-0.376
ppa-0.373
annah-0.369
atted-0.368
Token were
Token ID547
Feature activation+9.66e-02

Pos logit contributions
 long+0.126
fall+0.125
 seem+0.124
 spe+0.123
 smell+0.119
Neg logit contributions
aj-0.261
ppa-0.246
ma-0.244
annah-0.240
atted-0.237
Token the
Token ID262
Feature activation+3.03e-02

Pos logit contributions
 long+0.125
 seem+0.123
 spe+0.121
 smell+0.120
fall+0.119
Neg logit contributions
aj-0.258
ppa-0.236
ma-0.234
annah-0.234
atted-0.229
Token all
Token ID477
Feature activation+1.21e-01

Pos logit contributions
 long+0.065
 seem+0.063
 smell+0.062
 spe+0.061
fall+0.060
Neg logit contributions
aj-0.217
ppa-0.202
annah-0.200
ma-0.199
atted-0.197
Token of
Token ID286
Feature activation+7.44e-02

Pos logit contributions
 long+0.157
 seem+0.154
 spe+0.153
 smell+0.152
fall+0.148
Neg logit contributions
aj-0.323
ppa-0.299
annah-0.291
ma-0.289
atted-0.286
Token Alice
Token ID14862
Feature activation-1.06e-01

Pos logit contributions
 seem+0.112
 long+0.111
fall+0.110
 spe+0.110
 smell+0.108
Neg logit contributions
aj-0.178
ppa-0.163
ma-0.163
annah-0.161
atted-0.158
Token's
Token ID338
Feature activation+2.21e-02

Pos logit contributions
aj+0.259
ppa+0.231
annah+0.227
atted+0.225
ma+0.225
Neg logit contributions
 long-0.167
 seem-0.163
 smell-0.158
 spe-0.157
fall-0.157
Token favourite
Token ID12507
Feature activation+6.68e-02

Pos logit contributions
 seem+0.035
 spe+0.035
fall+0.035
 long+0.034
 smell+0.033
Neg logit contributions
aj-0.052
ma-0.047
ppa-0.047
annah-0.046
atted-0.045
Token things
Token ID1243
Feature activation+2.63e-01

Pos logit contributions
 seem+0.090
 spe+0.089
fall+0.089
 long+0.085
 dot+0.084
Neg logit contributions
aj-0.171
ma-0.161
annah-0.156
ppa-0.154
atted-0.152
Token,
Token ID11
Feature activation+3.59e-03

Pos logit contributions
 spe+0.367
 dot+0.360
 view+0.354
fall+0.353
 long+0.345
Neg logit contributions
aj-0.615
ma-0.611
annah-0.581
atted-0.573
ppa-0.561
Token which
Token ID543
Feature activation+1.11e-01

Pos logit contributions
fall+0.004
 seem+0.004
 spe+0.004
 long+0.004
 view+0.004
Neg logit contributions
aj-0.009
ma-0.009
ppa-0.009
annah-0.009
atted-0.008
Token made
Token ID925
Feature activation+3.47e-02

Pos logit contributions
fall+0.137
 spe+0.136
 dot+0.134
 long+0.131
 seem+0.129
Neg logit contributions
ma-0.282
aj-0.278
annah-0.265
ppa-0.263
aret-0.253
Token Alice
Token ID14862
Feature activation-1.29e-01

Pos logit contributions
 long+0.059
 seem+0.058
 smell+0.057
fall+0.057
 spe+0.057
Neg logit contributions
aj-0.078
ppa-0.071
annah-0.069
ma-0.069
atted-0.068
Token so
Token ID523
Feature activation+1.10e-01

Pos logit contributions
aj+0.333
ma+0.310
ppa+0.306
annah+0.303
atted+0.302
Neg logit contributions
 spe-0.163
fall-0.162
 long-0.154
 view-0.152
 seem-0.151
Token he
Token ID339
Feature activation+1.14e-02

Pos logit contributions
aj+0.140
ma+0.139
ppa+0.133
atted+0.132
annah+0.130
Neg logit contributions
 seem-0.074
 spe-0.073
fall-0.072
 dot-0.072
 smell-0.069
Token found
Token ID1043
Feature activation-8.61e-02

Pos logit contributions
 seem+0.014
 spe+0.014
 long+0.014
 smell+0.014
fall+0.014
Neg logit contributions
aj-0.030
annah-0.028
ppa-0.028
ma-0.028
atted-0.027
Token the
Token ID262
Feature activation-1.81e-02

Pos logit contributions
aj+0.253
ma+0.238
ppa+0.236
annah+0.233
atted+0.233
Neg logit contributions
 seem-0.082
fall-0.081
 spe-0.079
 long-0.079
 smell-0.077
Token most
Token ID749
Feature activation+1.52e-01

Pos logit contributions
aj+0.047
ma+0.046
aret+0.041
ppa+0.041
annah+0.041
Neg logit contributions
fall-0.022
 spe-0.020
 seem-0.019
Ice-0.017
 smell-0.017
Token amazing
Token ID4998
Feature activation+3.58e-02

Pos logit contributions
fall+0.220
 seem+0.210
storm+0.187
t+0.184
 Pretty+0.183
Neg logit contributions
aj-0.375
ma-0.358
atted-0.341
ins-0.328
ppa-0.318
Token things
Token ID1243
Feature activation+2.63e-01

Pos logit contributions
fall+0.049
 seem+0.046
 spe+0.045
Ice+0.042
t+0.042
Neg logit contributions
ma-0.087
aj-0.086
atted-0.078
aret-0.076
lee-0.076
Token to
Token ID284
Feature activation+6.01e-02

Pos logit contributions
 spe+0.300
 dot+0.299
fall+0.292
 long+0.290
 view+0.283
Neg logit contributions
ma-0.673
aj-0.668
atted-0.637
annah-0.636
ppa-0.629
Token look
Token ID804
Feature activation+8.95e-02

Pos logit contributions
 long+0.057
 spe+0.056
fall+0.056
 seem+0.056
 smell+0.053
Neg logit contributions
aj-0.176
ma-0.165
ppa-0.164
annah-0.163
atted-0.160
Token at
Token ID379
Feature activation-1.57e-02

Pos logit contributions
 seem+0.119
 long+0.119
 spe+0.118
fall+0.117
 smell+0.115
Neg logit contributions
aj-0.231
ppa-0.212
ma-0.211
annah-0.209
atted-0.206
Token.
Token ID13
Feature activation+7.79e-02

Pos logit contributions
aj+0.042
ma+0.040
ppa+0.039
annah+0.038
atted+0.038
Neg logit contributions
fall-0.019
 seem-0.019
 long-0.018
 spe-0.018
 smell-0.018
Token He
Token ID679
Feature activation+2.79e-02

Pos logit contributions
 long+0.087
 seem+0.087
fall+0.086
 Pretty+0.082
 view+0.081
Neg logit contributions
aj-0.218
ppa-0.206
ma-0.201
annah-0.196
atted-0.195
Token time
Token ID640
Feature activation+5.47e-02

Pos logit contributions
aj+0.084
ma+0.080
annah+0.078
ppa+0.077
atted+0.075
Neg logit contributions
 seem-0.030
 spe-0.030
fall-0.029
 smell-0.028
 long-0.027
Token!
Token ID0
Feature activation+6.29e-02

Pos logit contributions
 long+0.061
 seem+0.059
fall+0.058
 spe+0.057
 smell+0.056
Neg logit contributions
aj-0.154
ppa-0.143
ma-0.142
atted-0.140
annah-0.140
Tokenâ
Token ID22940
Feature activation-1.36e-01

Pos logit contributions
 long+0.080
 seem+0.080
 spe+0.079
fall+0.078
 smell+0.078
Neg logit contributions
aj-0.166
ppa-0.152
ma-0.152
annah-0.151
atted-0.148
Token
Token ID26391
Feature activation-1.17e-01

Pos logit contributions
aj+0.289
ppa+0.260
ma+0.256
atted+0.251
annah+0.250
Neg logit contributions
 seem-0.247
fall-0.243
 long-0.241
 smell-0.238
 spe-0.236
Token<|endoftext|>
Token ID50256
Feature activation-1.35e-01

Pos logit contributions
aj+0.299
ppa+0.276
ma+0.274
annah+0.271
atted+0.268
Neg logit contributions
 long-0.158
 seem-0.158
fall-0.156
 spe-0.155
 smell-0.153
TokenPa
Token ID28875
Feature activation-3.08e-01

Pos logit contributions
aj+0.326
ma+0.305
annah+0.293
ppa+0.290
atted+0.286
Neg logit contributions
 spe-0.200
 seem-0.196
 smell-0.194
 long-0.194
fall-0.188
Tokenolo
Token ID14057
Feature activation-1.27e-01

Pos logit contributions
aj+0.659
ma+0.585
ins+0.585
aret+0.582
ppa+0.576
Neg logit contributions
 long-0.568
 view-0.525
 smell-0.523
 seem-0.521
fall-0.512
Token was
Token ID373
Feature activation+1.55e-02

Pos logit contributions
aj+0.337
ppa+0.316
annah+0.307
ma+0.304
aret+0.300
Neg logit contributions
 smell-0.160
 seem-0.159
 long-0.157
fall-0.153
 spe-0.153
Token very
Token ID845
Feature activation+6.69e-02

Pos logit contributions
 long+0.017
 smell+0.015
 spe+0.014
 view+0.014
fall+0.014
Neg logit contributions
aj-0.046
ppa-0.043
annah-0.042
ma-0.042
atted-0.041
Token excited
Token ID6568
Feature activation-9.15e-02

Pos logit contributions
 long+0.093
 seem+0.088
 smell+0.088
 spe+0.087
fall+0.086
Neg logit contributions
aj-0.174
ppa-0.163
ma-0.160
annah-0.158
atted-0.155
Token.
Token ID13
Feature activation+4.74e-02

Pos logit contributions
aj+0.234
ma+0.222
annah+0.216
ppa+0.211
atted+0.209
Neg logit contributions
 spe-0.129
 seem-0.118
 dot-0.112
 long-0.112
fall-0.112
Token he
Token ID339
Feature activation+7.59e-02

Pos logit contributions
fall+0.197
 spe+0.197
 dot+0.195
 Pretty+0.179
 view+0.175
Neg logit contributions
aj-0.408
ma-0.405
annah-0.384
atted-0.374
ins-0.366
Token shook
Token ID14682
Feature activation-8.18e-02

Pos logit contributions
 long+0.114
 seem+0.113
 spe+0.113
fall+0.112
 smell+0.110
Neg logit contributions
aj-0.182
ppa-0.166
ma-0.166
annah-0.164
atted-0.161
Token it
Token ID340
Feature activation+4.26e-02

Pos logit contributions
aj+0.205
ma+0.189
ppa+0.188
annah+0.186
atted+0.183
Neg logit contributions
 seem-0.113
fall-0.112
 spe-0.112
 long-0.111
 smell-0.109
Token.
Token ID13
Feature activation+6.72e-02

Pos logit contributions
 spe+0.055
fall+0.053
 seem+0.053
 smell+0.052
 long+0.052
Neg logit contributions
aj-0.111
ma-0.103
annah-0.101
ppa-0.101
atted-0.099
Token<|endoftext|>
Token ID50256
Feature activation-1.26e-01

Pos logit contributions
 seem+0.077
 long+0.077
fall+0.076
 spe+0.075
 smell+0.074
Neg logit contributions
aj-0.185
ppa-0.172
ma-0.171
annah-0.169
atted-0.167
TokenPa
Token ID28875
Feature activation-3.03e-01

Pos logit contributions
aj+0.311
ma+0.291
annah+0.281
ppa+0.279
atted+0.274
Neg logit contributions
 spe-0.177
 long-0.175
 seem-0.175
 smell-0.172
fall-0.170
Tokenola
Token ID5708
Feature activation-1.41e-01

Pos logit contributions
aj+0.552
ins+0.503
aret+0.478
rums+0.466
atted+0.462
Neg logit contributions
 long-0.610
 like-0.598
 view-0.567
 seem-0.556
 look-0.554
Token and
Token ID290
Feature activation-5.24e-02

Pos logit contributions
aj+0.396
ppa+0.374
annah+0.358
aret+0.358
ins+0.355
Neg logit contributions
 smell-0.157
 spe-0.153
 long-0.153
 seem-0.151
fall-0.145
Token her
Token ID607
Feature activation-1.48e-01

Pos logit contributions
aj+0.122
ppa+0.109
ma+0.108
annah+0.108
atted+0.106
Neg logit contributions
 seem-0.086
 long-0.086
 smell-0.084
fall-0.084
 spe-0.084
Token mother
Token ID2802
Feature activation-1.04e-01

Pos logit contributions
aj+0.244
ppa+0.213
ma+0.212
annah+0.209
atted+0.203
Neg logit contributions
 seem-0.339
 long-0.338
 spe-0.335
fall-0.335
 smell-0.332
Token were
Token ID547
Feature activation+2.70e-02

Pos logit contributions
aj+0.322
ppa+0.293
annah+0.292
aret+0.282
ins+0.277
Neg logit contributions
 smell-0.103
 spe-0.090
 seem-0.090
 sunset-0.090
 long-0.087
Token,
Token ID11
Feature activation-1.25e-02

Pos logit contributions
 long+0.060
 seem+0.058
 spe+0.057
 smell+0.056
fall+0.056
Neg logit contributions
aj-0.132
annah-0.122
ppa-0.122
atted-0.120
ma-0.119
Token D
Token ID360
Feature activation-1.58e-01

Pos logit contributions
aj+0.031
ppa+0.028
ma+0.028
annah+0.028
atted+0.027
Neg logit contributions
 long-0.018
 seem-0.018
 spe-0.018
fall-0.018
 smell-0.018
Tokenucky
Token ID5309
Feature activation-4.02e-02

Pos logit contributions
aj+0.323
ppa+0.291
ma+0.290
annah+0.285
atted+0.280
Neg logit contributions
 seem-0.296
 long-0.296
fall-0.292
 smell-0.289
 spe-0.289
Token!"
Token ID2474
Feature activation-3.67e-02

Pos logit contributions
aj+0.118
ppa+0.109
annah+0.108
ma+0.108
atted+0.107
Neg logit contributions
 long-0.041
 seem-0.039
 smell-0.038
 spe-0.038
fall-0.037
Token replied
Token ID8712
Feature activation-2.08e-01

Pos logit contributions
aj+0.079
ma+0.070
ppa+0.069
annah+0.069
aret+0.066
Neg logit contributions
 long-0.069
 smell-0.068
 seem-0.067
fall-0.067
 spe-0.066
Token Fro
Token ID9734
Feature activation-3.02e-01

Pos logit contributions
aj+0.518
ppa+0.470
ma+0.469
annah+0.466
atted+0.452
Neg logit contributions
 long-0.307
 seem-0.301
 spe-0.299
 smell-0.297
fall-0.296
Tokenggy
Token ID19970
Feature activation-1.56e-01

Pos logit contributions
aj+0.662
ma+0.602
annah+0.598
atted+0.578
aret+0.575
Neg logit contributions
 long-0.532
 seem-0.529
fall-0.527
 smell-0.509
 spe-0.495
Token.
Token ID13
Feature activation+3.33e-02

Pos logit contributions
aj+0.510
ma+0.476
annah+0.470
ppa+0.466
aret+0.460
Neg logit contributions
 long-0.118
fall-0.102
 seem-0.102
 spe-0.098
 smell-0.095
Token
Token ID198
Feature activation-4.01e-02

Pos logit contributions
 long+0.037
fall+0.035
 seem+0.035
 spe+0.034
 smell+0.034
Neg logit contributions
aj-0.095
ma-0.091
ppa-0.090
annah-0.088
aret-0.087
Token
Token ID198
Feature activation-3.48e-02

Pos logit contributions
aj+0.137
ppa+0.132
annah+0.129
ma+0.128
atted+0.127
Neg logit contributions
 long-0.026
 lot-0.021
t-0.019
fall-0.019
 seem-0.019
TokenF
Token ID37
Feature activation-1.34e-01

Pos logit contributions
aj+0.091
ma+0.085
annah+0.082
atted+0.082
ppa+0.081
Neg logit contributions
 long-0.051
 lot-0.046
 spe-0.046
fall-0.046
 seem-0.045
Token
Token ID198
Feature activation-2.21e-02

Pos logit contributions
aj+0.085
ppa+0.083
ma+0.080
aret+0.080
atted+0.079
Neg logit contributions
 seem-0.020
 long-0.020
fall-0.019
 Pretty-0.018
 much-0.016
Token
Token ID198
Feature activation-1.85e-02

Pos logit contributions
aj+0.079
atted+0.074
ppa+0.073
aret+0.072
ma+0.071
Neg logit contributions
 long-0.012
 much-0.010
fall-0.009
 seem-0.009
t-0.009
TokenS
Token ID50
Feature activation-1.70e-01

Pos logit contributions
aj+0.048
atted+0.043
ma+0.042
aret+0.042
ppa+0.042
Neg logit contributions
 long-0.027
fall-0.026
t-0.025
 seem-0.025
 view-0.025
Tokenally
Token ID453
Feature activation-1.71e-01

Pos logit contributions
aj+0.219
aret+0.187
ppa+0.182
annah+0.176
atted+0.167
Neg logit contributions
fall-0.431
 seem-0.427
 view-0.426
 sunset-0.422
 smell-0.418
Token thanked
Token ID26280
Feature activation-2.16e-01

Pos logit contributions
aj+0.500
ppa+0.463
annah+0.449
aret+0.447
ins+0.443
Neg logit contributions
 smell-0.167
 spe-0.165
 long-0.165
 ever-0.161
 fly-0.150
Token Fro
Token ID9734
Feature activation-2.89e-01

Pos logit contributions
aj+0.523
ppa+0.463
atted+0.454
ma+0.452
annah+0.451
Neg logit contributions
 long-0.346
 seem-0.344
 smell-0.341
fall-0.334
 spe-0.333
Tokenggie
Token ID23571
Feature activation-1.52e-01

Pos logit contributions
aj+0.598
ma+0.550
aret+0.543
annah+0.540
atted+0.533
Neg logit contributions
fall-0.542
 seem-0.532
 long-0.521
 smell-0.519
 taste-0.488
Token and
Token ID290
Feature activation-3.91e-02

Pos logit contributions
aj+0.431
ppa+0.397
ma+0.388
annah+0.384
ins+0.384
Neg logit contributions
 long-0.182
fall-0.164
 seem-0.163
 smell-0.162
 spe-0.154
Token waved
Token ID26834
Feature activation-1.13e-01

Pos logit contributions
aj+0.103
ppa+0.093
annah+0.093
ma+0.092
atted+0.090
Neg logit contributions
 long-0.053
 spe-0.051
 smell-0.051
fall-0.050
 seem-0.050
Token goodbye
Token ID24829
Feature activation-1.50e-01

Pos logit contributions
aj+0.271
ppa+0.248
ma+0.246
annah+0.244
atted+0.239
Neg logit contributions
 long-0.175
 seem-0.174
 spe-0.172
fall-0.172
 smell-0.170
Token as
Token ID355
Feature activation+1.20e-01

Pos logit contributions
aj+0.404
ppa+0.378
annah+0.372
ma+0.367
atted+0.365
Neg logit contributions
 seem-0.178
 spe-0.169
 smell-0.169
 view-0.161
fall-0.159
Token friends
Token ID2460
Feature activation+8.38e-02

Pos logit contributions
fall+0.177
 spe+0.171
 view+0.164
 dot+0.159
 long+0.158
Neg logit contributions
aj-0.265
ma-0.255
ins-0.243
atted-0.241
ppa-0.239
Token are
Token ID389
Feature activation+1.20e-01

Pos logit contributions
 long+0.090
 spe+0.090
fall+0.089
 view+0.082
 sunset+0.080
Neg logit contributions
aj-0.229
ma-0.222
ppa-0.218
annah-0.216
atted-0.210
Token for
Token ID329
Feature activation+2.75e-02

Pos logit contributions
 spe+0.172
fall+0.169
 seem+0.168
 dot+0.163
 smell+0.162
Neg logit contributions
aj-0.294
ma-0.275
ppa-0.275
annah-0.273
atted-0.263
Token,"
Token ID553
Feature activation-6.90e-02

Pos logit contributions
 long+0.035
 seem+0.033
 smell+0.032
fall+0.032
 spe+0.030
Neg logit contributions
aj-0.075
ppa-0.069
atted-0.068
annah-0.068
ma-0.067
Token replied
Token ID8712
Feature activation-2.00e-01

Pos logit contributions
aj+0.154
annah+0.132
ma+0.130
ppa+0.128
aret+0.127
Neg logit contributions
 long-0.127
 smell-0.122
 sunset-0.120
fall-0.118
 seem-0.117
Token Fro
Token ID9734
Feature activation-2.87e-01

Pos logit contributions
aj+0.533
ma+0.482
ppa+0.479
annah+0.476
atted+0.460
Neg logit contributions
 long-0.275
fall-0.260
 spe-0.260
 smell-0.259
 seem-0.258
Tokenggy
Token ID19970
Feature activation-1.47e-01

Pos logit contributions
aj+0.632
ma+0.587
aret+0.567
annah+0.566
atted+0.556
Neg logit contributions
 seem-0.495
fall-0.490
 long-0.488
t-0.468
 smell-0.454
Token.
Token ID13
Feature activation+3.49e-02

Pos logit contributions
aj+0.485
ma+0.454
annah+0.449
aret+0.444
ppa+0.441
Neg logit contributions
 long-0.101
fall-0.089
 seem-0.086
 smell-0.079
 spe-0.073
Token And
Token ID843
Feature activation-4.33e-02

Pos logit contributions
 long+0.035
fall+0.033
 Pretty+0.033
 seem+0.030
 lot+0.028
Neg logit contributions
aj-0.101
ma-0.101
ppa-0.099
aret-0.096
annah-0.094
Token they
Token ID484
Feature activation+5.15e-02

Pos logit contributions
aj+0.135
ma+0.123
ppa+0.123
annah+0.122
aret+0.120
Neg logit contributions
 long-0.040
 seem-0.037
 spe-0.036
fall-0.036
 smell-0.036
Token lived
Token ID5615
Feature activation-1.61e-02

Pos logit contributions
 long+0.064
 spe+0.064
 smell+0.064
 seem+0.064
 view+0.060
Neg logit contributions
aj-0.142
annah-0.128
ppa-0.125
ma-0.125
ins-0.123
Token stories
Token ID3923
Feature activation+1.15e-01

Pos logit contributions
fall+0.028
 spe+0.028
 seem+0.025
Ice+0.025
 dot+0.024
Neg logit contributions
ma-0.031
aj-0.030
ppa-0.028
aret-0.027
annah-0.027
Token said
Token ID531
Feature activation-1.55e-01

Pos logit contributions
 spe+0.152
 long+0.146
fall+0.145
 view+0.142
 dot+0.140
Neg logit contributions
ma-0.283
aj-0.276
annah-0.258
ppa-0.253
atted-0.251
Token.
Token ID13
Feature activation+4.20e-02

Pos logit contributions
aj+0.430
ma+0.403
ppa+0.399
annah+0.395
atted+0.389
Neg logit contributions
 spe-0.169
 long-0.166
fall-0.165
 seem-0.165
 smell-0.160
Token Mama
Token ID40084
Feature activation-1.36e-01

Pos logit contributions
 long+0.057
fall+0.056
 smell+0.054
 seem+0.054
 spe+0.053
Neg logit contributions
aj-0.108
ppa-0.102
ma-0.102
atted-0.098
annah-0.098
Token helped
Token ID4193
Feature activation-1.32e-01

Pos logit contributions
aj+0.344
annah+0.308
aret+0.288
ppa+0.285
atted+0.283
Neg logit contributions
 smell-0.203
 spe-0.178
 ever-0.176
 taste-0.173
 seem-0.170
Token Ci
Token ID37685
Feature activation-2.82e-01

Pos logit contributions
aj+0.256
ppa+0.227
ma+0.225
annah+0.223
atted+0.219
Neg logit contributions
 long-0.262
 seem-0.262
fall-0.259
 spe-0.258
 smell-0.257
Tokenara
Token ID3301
Feature activation-1.90e-01

Pos logit contributions
aj+0.397
ppa+0.333
aret+0.330
ma+0.326
atted+0.322
Neg logit contributions
fall-0.683
 seem-0.676
 smell-0.673
 spe-0.664
 long-0.657
Token and
Token ID290
Feature activation-3.15e-02

Pos logit contributions
aj+0.535
ppa+0.496
annah+0.489
ma+0.481
atted+0.474
Neg logit contributions
 seem-0.228
 smell-0.223
 long-0.214
fall-0.206
 taste-0.198
Token together
Token ID1978
Feature activation-6.63e-03

Pos logit contributions
aj+0.085
annah+0.075
ppa+0.073
atted+0.072
aret+0.071
Neg logit contributions
 smell-0.044
 spe-0.043
 long-0.042
fall-0.039
 seem-0.039
Token they
Token ID484
Feature activation+4.31e-02

Pos logit contributions
aj+0.020
ppa+0.019
ma+0.019
annah+0.019
atted+0.019
Neg logit contributions
 seem-0.006
 long-0.006
 spe-0.006
fall-0.005
 smell-0.005
Token read
Token ID1100
Feature activation-5.85e-02

Pos logit contributions
 long+0.059
 seem+0.059
 spe+0.058
fall+0.058
 smell+0.057
Neg logit contributions
aj-0.110
ppa-0.101
ma-0.100
annah-0.100
atted-0.098
Token He
Token ID679
Feature activation+1.00e-02

Pos logit contributions
 long+0.055
 seem+0.051
fall+0.050
 Pretty+0.049
 smell+0.048
Neg logit contributions
aj-0.169
ma-0.164
ppa-0.162
annah-0.154
aret-0.154
Token was
Token ID373
Feature activation+5.22e-02

Pos logit contributions
 smell+0.014
 spe+0.013
 long+0.013
 seem+0.013
 quite+0.012
Neg logit contributions
aj-0.027
annah-0.024
ppa-0.024
ma-0.024
aret-0.023
Token happy
Token ID3772
Feature activation-8.86e-02

Pos logit contributions
 long+0.073
 smell+0.066
 quite+0.065
 spe+0.064
 dizz+0.064
Neg logit contributions
aj-0.145
annah-0.131
ma-0.129
ppa-0.127
aret-0.126
Token and
Token ID290
Feature activation+3.67e-03

Pos logit contributions
aj+0.246
ma+0.230
ppa+0.227
atted+0.227
annah+0.223
Neg logit contributions
 seem-0.099
 spe-0.099
fall-0.096
 long-0.096
 smell-0.093
Token thanked
Token ID26280
Feature activation-2.01e-01

Pos logit contributions
 long+0.006
 seem+0.006
fall+0.006
 smell+0.006
 spe+0.006
Neg logit contributions
aj-0.008
ma-0.007
ppa-0.007
annah-0.007
atted-0.007
Token Fro
Token ID9734
Feature activation-2.81e-01

Pos logit contributions
aj+0.459
ppa+0.415
ma+0.413
annah+0.410
atted+0.401
Neg logit contributions
 long-0.335
 seem-0.334
 spe-0.328
fall-0.327
 smell-0.327
Tokenggy
Token ID19970
Feature activation-1.54e-01

Pos logit contributions
aj+0.559
ma+0.523
annah+0.503
atted+0.488
aret+0.488
Neg logit contributions
 seem-0.547
fall-0.545
 long-0.543
 smell-0.524
t-0.500
Token for
Token ID329
Feature activation-3.80e-02

Pos logit contributions
aj+0.489
ma+0.459
ppa+0.456
annah+0.449
aret+0.448
Neg logit contributions
 long-0.123
fall-0.106
 seem-0.106
 smell-0.101
 like-0.094
Token helping
Token ID5742
Feature activation-1.05e-01

Pos logit contributions
aj+0.098
annah+0.087
ppa+0.086
ma+0.084
atted+0.084
Neg logit contributions
 long-0.057
 smell-0.055
 seem-0.055
 spe-0.055
fall-0.053
Token him
Token ID683
Feature activation-1.08e-01

Pos logit contributions
aj+0.269
ma+0.238
annah+0.238
ppa+0.237
atted+0.235
Neg logit contributions
 long-0.154
 seem-0.153
 smell-0.150
fall-0.147
 spe-0.145
Token.
Token ID13
Feature activation+4.35e-02

Pos logit contributions
aj+0.286
ma+0.268
ppa+0.266
annah+0.261
atted+0.259
Neg logit contributions
 spe-0.136
fall-0.128
 long-0.127
 seem-0.123
 view-0.121
Token was
Token ID373
Feature activation+5.41e-02

Pos logit contributions
aj+0.197
annah+0.185
ma+0.179
aret+0.178
ppa+0.177
Neg logit contributions
 smell-0.080
 long-0.074
 spe-0.072
 while-0.067
 fly-0.067
Token so
Token ID523
Feature activation+1.13e-01

Pos logit contributions
 long+0.060
 like+0.052
 smell+0.052
 quite+0.051
 much+0.051
Neg logit contributions
aj-0.169
annah-0.155
ma-0.151
ppa-0.151
aret-0.149
Token happy
Token ID3772
Feature activation-7.45e-02

Pos logit contributions
 long+0.174
 smell+0.162
 seem+0.161
 spe+0.154
fall+0.150
Neg logit contributions
aj-0.292
ppa-0.268
annah-0.267
ma-0.267
aret-0.257
Token and
Token ID290
Feature activation+9.32e-03

Pos logit contributions
aj+0.186
ma+0.172
atted+0.171
ppa+0.170
annah+0.169
Neg logit contributions
 spe-0.104
 seem-0.101
fall-0.097
 long-0.096
 smell-0.095
Token thanked
Token ID26280
Feature activation-1.95e-01

Pos logit contributions
 long+0.017
 seem+0.016
 smell+0.016
fall+0.016
 spe+0.016
Neg logit contributions
aj-0.020
ma-0.019
ppa-0.018
annah-0.018
atted-0.018
Token Fro
Token ID9734
Feature activation-2.79e-01

Pos logit contributions
aj+0.480
ppa+0.431
annah+0.428
ma+0.428
atted+0.418
Neg logit contributions
 long-0.300
 seem-0.296
 smell-0.295
 spe-0.289
fall-0.287
Tokenggy
Token ID19970
Feature activation-1.43e-01

Pos logit contributions
aj+0.561
ma+0.517
annah+0.504
aret+0.494
atted+0.491
Neg logit contributions
 seem-0.539
fall-0.536
 long-0.533
 smell-0.521
 spe-0.493
Token.
Token ID13
Feature activation+4.26e-02

Pos logit contributions
aj+0.440
ma+0.414
ppa+0.407
aret+0.406
annah+0.404
Neg logit contributions
 long-0.129
fall-0.116
 seem-0.112
 smell-0.109
 like-0.100
Token They
Token ID1119
Feature activation+5.74e-02

Pos logit contributions
 long+0.048
fall+0.045
 Pretty+0.044
 smell+0.043
 seem+0.043
Neg logit contributions
ma-0.119
aj-0.119
ppa-0.115
annah-0.109
aret-0.109
Token played
Token ID2826
Feature activation-5.01e-02

Pos logit contributions
 long+0.082
 smell+0.081
 spe+0.080
 seem+0.080
fall+0.076
Neg logit contributions
aj-0.148
annah-0.133
ma-0.132
ppa-0.130
atted-0.128
Token together
Token ID1978
Feature activation+1.29e-02

Pos logit contributions
aj+0.147
ppa+0.138
annah+0.136
ma+0.133
aret+0.131
Neg logit contributions
 long-0.059
 view-0.056
 smell-0.055
 like-0.051
 seem-0.051
Token to
Token ID284
Feature activation-7.86e-03

Pos logit contributions
 spe+0.035
fall+0.032
 seem+0.032
 dot+0.032
 smell+0.032
Neg logit contributions
aj-0.101
ma-0.093
ppa-0.093
annah-0.091
atted-0.091
Token come
Token ID1282
Feature activation-6.48e-02

Pos logit contributions
aj+0.020
ma+0.019
ppa+0.019
annah+0.018
aret+0.018
Neg logit contributions
 spe-0.009
 dot-0.009
 brightly-0.009
 single-0.009
 long-0.008
Token with
Token ID351
Feature activation-3.98e-02

Pos logit contributions
aj+0.151
ma+0.139
ppa+0.137
annah+0.133
atted+0.131
Neg logit contributions
 spe-0.101
 seem-0.100
fall-0.098
 long-0.097
 smell-0.095
Token me
Token ID502
Feature activation-3.76e-02

Pos logit contributions
aj+0.073
ppa+0.064
atted+0.064
annah+0.063
ins+0.061
Neg logit contributions
 long-0.089
 view-0.084
 smell-0.084
 seem-0.083
 spe-0.083
Token,"
Token ID553
Feature activation-5.79e-02

Pos logit contributions
aj+0.098
annah+0.091
ppa+0.091
atted+0.091
ma+0.088
Neg logit contributions
 long-0.051
 seem-0.047
fall-0.047
 view-0.046
 smell-0.046
Token Cic
Token ID37212
Feature activation-2.77e-01

Pos logit contributions
aj+0.115
ma+0.107
ppa+0.106
annah+0.104
atted+0.103
Neg logit contributions
fall-0.106
 seem-0.106
 long-0.105
 spe-0.104
 view-0.103
Tokeni
Token ID72
Feature activation-1.40e-01

Pos logit contributions
aj+0.601
ma+0.487
annah+0.467
eline+0.457
atted+0.428
Neg logit contributions
 spe-0.490
 seem-0.455
 smell-0.450
 Hipp-0.441
fall-0.431
Token said
Token ID531
Feature activation-1.75e-01

Pos logit contributions
aj+0.344
annah+0.294
ppa+0.289
ma+0.287
aret+0.283
Neg logit contributions
 seem-0.226
 long-0.222
fall-0.214
 smell-0.211
 spe-0.206
Token to
Token ID284
Feature activation-5.52e-02

Pos logit contributions
aj+0.466
ma+0.439
ppa+0.438
annah+0.435
atted+0.429
Neg logit contributions
 spe-0.201
 seem-0.199
 view-0.198
fall-0.195
 smell-0.195
Token her
Token ID607
Feature activation-1.48e-01

Pos logit contributions
aj+0.128
ppa+0.116
ma+0.115
annah+0.114
atted+0.112
Neg logit contributions
 seem-0.091
 long-0.090
 smell-0.089
 spe-0.089
fall-0.089
Token friends
Token ID2460
Feature activation+9.18e-03

Pos logit contributions
aj+0.285
ppa+0.254
ma+0.251
annah+0.250
atted+0.243
Neg logit contributions
 long-0.302
 seem-0.301
 spe-0.297
fall-0.296
 smell-0.296
Token her
Token ID607
Feature activation-7.03e-02

Pos logit contributions
 long+0.013
 seem+0.013
 spe+0.013
fall+0.013
 smell+0.013
Neg logit contributions
aj-0.022
ppa-0.020
ma-0.020
annah-0.020
atted-0.020
Token brown
Token ID7586
Feature activation-5.78e-02

Pos logit contributions
aj+0.188
ma+0.173
ppa+0.172
annah+0.170
atted+0.168
Neg logit contributions
 seem-0.087
fall-0.087
 spe-0.085
 long-0.085
 smell-0.084
Token fur
Token ID9230
Feature activation+9.58e-02

Pos logit contributions
aj+0.136
ppa+0.125
annah+0.123
ma+0.123
atted+0.120
Neg logit contributions
 long-0.091
 seem-0.089
 smell-0.089
 spe-0.088
fall-0.088
Token.
Token ID13
Feature activation+7.56e-02

Pos logit contributions
 seem+0.115
 spe+0.112
 long+0.111
fall+0.110
 smell+0.107
Neg logit contributions
aj-0.258
ma-0.241
ppa-0.240
annah-0.235
atted-0.232
Token<|endoftext|>
Token ID50256
Feature activation-1.22e-01

Pos logit contributions
 long+0.072
 seem+0.071
fall+0.069
 Pretty+0.067
 view+0.063
Neg logit contributions
aj-0.222
ppa-0.214
ma-0.208
atted-0.204
annah-0.203
TokenArk
Token ID42007
Feature activation-2.77e-01

Pos logit contributions
aj+0.304
ma+0.285
annah+0.275
ppa+0.274
atted+0.270
Neg logit contributions
 spe-0.171
 long-0.169
 seem-0.169
 smell-0.166
fall-0.164
Tokenie
Token ID494
Feature activation-1.24e-01

Pos logit contributions
aj+0.651
annah+0.535
atted+0.526
irling+0.507
 wounds+0.503
Neg logit contributions
 night-0.481
 day-0.394
 look-0.371
 quite-0.351
 one-0.350
Token wanted
Token ID2227
Feature activation-5.79e-02

Pos logit contributions
aj+0.361
ppa+0.328
ins+0.318
aret+0.317
annah+0.313
Neg logit contributions
 smell-0.135
 long-0.134
 seem-0.126
fall-0.122
 spe-0.121
Token his
Token ID465
Feature activation-1.26e-01

Pos logit contributions
aj+0.165
ma+0.156
ppa+0.153
annah+0.151
atted+0.149
Neg logit contributions
 spe-0.059
 seem-0.058
fall-0.058
 dot-0.054
 long-0.053
Token own
Token ID898
Feature activation-1.03e-02

Pos logit contributions
aj+0.253
ma+0.230
ppa+0.222
annah+0.220
atted+0.216
Neg logit contributions
fall-0.240
 seem-0.239
 spe-0.237
 long-0.231
 smell-0.226
Token equipment
Token ID5112
Feature activation+8.10e-02

Pos logit contributions
aj+0.026
ma+0.024
ppa+0.024
annah+0.024
atted+0.023
Neg logit contributions
 seem-0.014
fall-0.014
 spe-0.014
 long-0.014
 smell-0.013
Token their
Token ID511
Feature activation-9.36e-02

Pos logit contributions
aj+0.141
ppa+0.128
annah+0.126
ma+0.124
atted+0.122
Neg logit contributions
 long-0.086
 seem-0.084
 spe-0.083
 smell-0.082
 view-0.081
Token loving
Token ID14442
Feature activation-7.73e-02

Pos logit contributions
aj+0.232
ma+0.213
annah+0.207
ppa+0.207
atted+0.206
Neg logit contributions
fall-0.133
 seem-0.132
 spe-0.131
 long-0.128
 smell-0.127
Token friendship
Token ID14738
Feature activation+1.42e-02

Pos logit contributions
aj+0.188
ppa+0.173
annah+0.170
ma+0.169
atted+0.165
Neg logit contributions
 long-0.117
 seem-0.115
 smell-0.114
 spe-0.114
fall-0.112
Token.
Token ID13
Feature activation+8.24e-02

Pos logit contributions
 spe+0.016
 seem+0.016
fall+0.016
 long+0.016
 smell+0.016
Neg logit contributions
aj-0.038
ma-0.036
ppa-0.035
annah-0.035
atted-0.034
Token<|endoftext|>
Token ID50256
Feature activation-1.24e-01

Pos logit contributions
 seem+0.102
 long+0.102
 spe+0.100
fall+0.100
 smell+0.099
Neg logit contributions
aj-0.220
ppa-0.203
ma-0.202
annah-0.200
atted-0.197
TokenPe
Token ID6435
Feature activation-2.74e-01

Pos logit contributions
aj+0.301
ma+0.285
annah+0.273
ppa+0.270
atted+0.266
Neg logit contributions
 spe-0.179
 seem-0.176
 long-0.176
 smell-0.174
fall-0.171
Tokenggy
Token ID19970
Feature activation-1.42e-01

Pos logit contributions
aj+0.505
atted+0.481
aret+0.446
ins+0.435
lee+0.409
Neg logit contributions
 long-0.510
 ever-0.473
 fly-0.472
 seem-0.471
fall-0.466
Token was
Token ID373
Feature activation+1.23e-02

Pos logit contributions
aj+0.407
ppa+0.384
aret+0.371
ins+0.369
atted+0.366
Neg logit contributions
 long-0.160
 smell-0.154
 seem-0.151
 spe-0.147
fall-0.145
Token very
Token ID845
Feature activation+6.79e-02

Pos logit contributions
 long+0.012
 smell+0.010
 spe+0.010
 view+0.010
 dark+0.010
Neg logit contributions
aj-0.038
ppa-0.036
annah-0.035
ma-0.035
atted-0.034
Token excited
Token ID6568
Feature activation-9.22e-02

Pos logit contributions
 long+0.098
 seem+0.093
 smell+0.093
 spe+0.092
fall+0.091
Neg logit contributions
aj-0.172
ppa-0.161
ma-0.158
annah-0.156
atted-0.152
Token.
Token ID13
Feature activation+5.30e-02

Pos logit contributions
aj+0.229
ma+0.220
annah+0.213
atted+0.207
ppa+0.207
Neg logit contributions
 spe-0.135
 seem-0.119
 dot-0.116
 long-0.113
fall-0.113
Token dried
Token ID16577
Feature activation-1.69e-01

Pos logit contributions
aj+0.455
ppa+0.428
anut+0.405
ins+0.404
ma+0.396
Neg logit contributions
 smell-0.248
 seem-0.239
 look-0.223
 quite-0.221
 really-0.221
Token his
Token ID465
Feature activation-1.40e-01

Pos logit contributions
aj+0.449
ppa+0.402
ins+0.385
ma+0.384
atted+0.374
Neg logit contributions
 long-0.244
 smell-0.232
fall-0.227
 view-0.215
 sunset-0.210
Token hands
Token ID2832
Feature activation+2.06e-02

Pos logit contributions
aj+0.303
ppa+0.298
ma+0.281
annah+0.279
eline+0.275
Neg logit contributions
 long-0.243
 smell-0.237
 view-0.224
 seem-0.221
 spe-0.221
Token and
Token ID290
Feature activation-2.73e-02

Pos logit contributions
 seem+0.031
 spe+0.031
fall+0.030
 long+0.030
 smell+0.030
Neg logit contributions
aj-0.049
ma-0.045
ppa-0.045
annah-0.045
atted-0.044
Token followed
Token ID3940
Feature activation-1.41e-01

Pos logit contributions
aj+0.072
ppa+0.067
annah+0.066
ma+0.065
atted+0.064
Neg logit contributions
 long-0.037
 smell-0.036
 seem-0.034
fall-0.034
 spe-0.033
Token Jol
Token ID39329
Feature activation-2.73e-01

Pos logit contributions
aj+0.318
ma+0.294
ppa+0.291
annah+0.289
atted+0.281
Neg logit contributions
 seem-0.231
 spe-0.231
fall-0.226
 long-0.225
 smell-0.223
Tokenene
Token ID1734
Feature activation-2.09e-01

Pos logit contributions
aj+0.418
ppa+0.336
ma+0.334
ins+0.333
annah+0.318
Neg logit contributions
t-0.543
fall-0.543
 dot-0.526
 long-0.522
 seem-0.521
Token.
Token ID13
Feature activation+2.67e-02

Pos logit contributions
aj+0.612
ppa+0.569
ma+0.566
annah+0.562
atted+0.554
Neg logit contributions
 long-0.207
 seem-0.206
fall-0.201
 spe-0.199
 smell-0.197
Token When
Token ID1649
Feature activation-2.06e-02

Pos logit contributions
 long+0.033
 seem+0.033
fall+0.032
 spe+0.032
 smell+0.032
Neg logit contributions
aj-0.071
ppa-0.066
ma-0.065
annah-0.065
atted-0.064
Token Jol
Token ID39329
Feature activation-2.61e-01

Pos logit contributions
ma+0.045
aj+0.045
annah+0.040
atted+0.040
aret+0.039
Neg logit contributions
 spe-0.034
 seem-0.032
 dot-0.032
 smell-0.030
fall-0.030
Tokenene
Token ID1734
Feature activation-1.87e-01

Pos logit contributions
aj+0.363
ppa+0.323
ins+0.311
ighed+0.308
 wounds+0.306
Neg logit contributions
t-0.509
fall-0.505
pool-0.486
 long-0.485
 seem-0.482
Token lot
Token ID1256
Feature activation+1.97e-01

Pos logit contributions
 seem+0.190
fall+0.184
 Pretty+0.175
 spe+0.174
t+0.169
Neg logit contributions
aj-0.258
ma-0.246
ppa-0.238
aret-0.224
 mistake-0.223
Token brighter
Token ID26083
Feature activation+7.20e-02

Pos logit contributions
 spe+0.243
fall+0.231
 long+0.229
 seem+0.225
 smell+0.224
Neg logit contributions
aj-0.526
ma-0.491
ppa-0.481
annah-0.479
atted-0.476
Token.
Token ID13
Feature activation+6.52e-02

Pos logit contributions
 seem+0.085
 spe+0.083
fall+0.083
 long+0.082
 smell+0.081
Neg logit contributions
aj-0.196
ppa-0.180
ma-0.180
annah-0.178
atted-0.176
Token<|endoftext|>
Token ID50256
Feature activation-1.28e-01

Pos logit contributions
 seem+0.074
 long+0.073
fall+0.071
 spe+0.070
 smell+0.068
Neg logit contributions
aj-0.182
ppa-0.170
ma-0.167
atted-0.165
annah-0.164
TokenS
Token ID50
Feature activation-2.28e-01

Pos logit contributions
aj+0.309
ma+0.292
annah+0.279
ppa+0.276
atted+0.271
Neg logit contributions
 spe-0.188
 seem-0.185
 long-0.183
 smell-0.182
fall-0.179
Tokenoph
Token ID2522
Feature activation-2.73e-01

Pos logit contributions
aj+0.353
ins+0.352
atted+0.337
aret+0.335
 wounds+0.330
Neg logit contributions
 sunset-0.492
 view-0.478
 long-0.475
 night-0.466
 seem-0.464
Tokenie
Token ID494
Feature activation-1.24e-01

Pos logit contributions
aj+0.556
 wounds+0.467
atted+0.463
annah+0.462
aned+0.444
Neg logit contributions
 night-0.498
 ever-0.466
 sunset-0.443
 much-0.429
 view-0.418
Token and
Token ID290
Feature activation-5.29e-02

Pos logit contributions
aj+0.354
ppa+0.337
ins+0.328
annah+0.321
aret+0.321
Neg logit contributions
 long-0.134
 smell-0.128
 spe-0.126
 seem-0.117
 sunset-0.116
Token her
Token ID607
Feature activation-1.39e-01

Pos logit contributions
aj+0.116
ppa+0.105
ma+0.105
annah+0.104
atted+0.102
Neg logit contributions
 seem-0.091
 long-0.091
 spe-0.090
fall-0.090
 smell-0.089
Token dad
Token ID9955
Feature activation-1.04e-01

Pos logit contributions
aj+0.221
ma+0.196
ppa+0.192
annah+0.183
atted+0.179
Neg logit contributions
 seem-0.327
fall-0.324
 spe-0.317
 smell-0.307
 view-0.306
Token were
Token ID547
Feature activation+3.22e-02

Pos logit contributions
aj+0.312
ppa+0.285
annah+0.283
atted+0.275
aret+0.274
Neg logit contributions
 smell-0.104
 seem-0.102
 long-0.100
fall-0.097
 spe-0.096
Token any
Token ID597
Feature activation+1.22e-01

Pos logit contributions
aj+0.040
ppa+0.038
ma+0.037
atted+0.037
annah+0.037
Neg logit contributions
fall-0.025
 seem-0.025
 spe-0.024
 smell-0.024
 view-0.024
Token conflict
Token ID5358
Feature activation+1.36e-02

Pos logit contributions
fall+0.190
 seem+0.178
storm+0.171
 smell+0.170
 spe+0.170
Neg logit contributions
aj-0.271
ma-0.263
atted-0.263
ppa-0.248
annah-0.243
Token.
Token ID13
Feature activation+7.28e-02

Pos logit contributions
 seem+0.015
 long+0.015
 spe+0.015
fall+0.014
 smell+0.014
Neg logit contributions
aj-0.038
ma-0.035
ppa-0.035
annah-0.035
atted-0.034
Token<|endoftext|>
Token ID50256
Feature activation-1.25e-01

Pos logit contributions
 seem+0.089
 long+0.089
fall+0.086
 smell+0.085
 spe+0.084
Neg logit contributions
aj-0.199
ppa-0.185
ma-0.181
annah-0.180
atted-0.177
TokenO
Token ID46
Feature activation-1.77e-01

Pos logit contributions
aj+0.304
ma+0.286
annah+0.276
ppa+0.273
atted+0.269
Neg logit contributions
 spe-0.180
 seem-0.177
 long-0.176
 smell-0.174
fall-0.171
Tokenliv
Token ID16017
Feature activation-2.72e-01

Pos logit contributions
aj+0.289
atted+0.260
ins+0.249
ppa+0.247
 wounds+0.243
Neg logit contributions
t-0.391
 night-0.387
 seem-0.381
 long-0.379
 ever-0.378
Tokenia
Token ID544
Feature activation-1.53e-01

Pos logit contributions
aj+0.408
ins+0.405
ppa+0.385
eline+0.326
annah+0.311
Neg logit contributions
 long-0.660
 night-0.648
 view-0.630
 ever-0.624
 lot-0.609
Token was
Token ID373
Feature activation+1.08e-02

Pos logit contributions
aj+0.433
ppa+0.403
aret+0.390
annah+0.390
atted+0.386
Neg logit contributions
 smell-0.164
 long-0.164
 spe-0.160
 seem-0.158
 like-0.155
Token feeling
Token ID4203
Feature activation+2.54e-02

Pos logit contributions
 long+0.010
 smell+0.009
 spe+0.009
 dark+0.008
 view+0.008
Neg logit contributions
aj-0.033
ppa-0.032
annah-0.031
ma-0.031
atted-0.030
Token very
Token ID845
Feature activation+7.90e-02

Pos logit contributions
fall+0.030
 spe+0.028
 seem+0.027
 dot+0.027
 view+0.027
Neg logit contributions
aj-0.068
ppa-0.064
ma-0.063
annah-0.063
atted-0.062
Token restless
Token ID42537
Feature activation-2.21e-02

Pos logit contributions
 spe+0.121
fall+0.119
 seem+0.117
 view+0.116
 dot+0.115
Neg logit contributions
aj-0.186
annah-0.171
ma-0.167
ppa-0.167
atted-0.166
Token bit
Token ID1643
Feature activation+1.52e-01

Pos logit contributions
fall+0.055
 seem+0.051
storm+0.048
br+0.048
 Pretty+0.045
Neg logit contributions
aj-0.093
ma-0.091
atted-0.087
annah-0.082
 mistake-0.080
Token of
Token ID286
Feature activation+6.21e-02

Pos logit contributions
fall+0.181
 spe+0.177
 seem+0.175
 view+0.169
 smell+0.169
Neg logit contributions
aj-0.411
ma-0.381
atted-0.379
ppa-0.376
annah-0.375
Token magic
Token ID5536
Feature activation+7.86e-02

Pos logit contributions
fall+0.088
 spe+0.084
 seem+0.084
 long+0.082
 smell+0.080
Neg logit contributions
aj-0.152
ma-0.144
annah-0.140
ppa-0.139
atted-0.138
Token.
Token ID13
Feature activation+7.20e-02

Pos logit contributions
 seem+0.106
fall+0.105
 spe+0.102
 long+0.100
 smell+0.095
Neg logit contributions
aj-0.197
ma-0.184
annah-0.177
ppa-0.173
atted-0.171
Token<|endoftext|>
Token ID50256
Feature activation-1.25e-01

Pos logit contributions
 long+0.099
 spe+0.098
 seem+0.097
fall+0.097
 smell+0.097
Neg logit contributions
aj-0.183
ma-0.169
annah-0.167
ppa-0.167
atted-0.163
TokenIsa
Token ID39443
Feature activation-2.71e-01

Pos logit contributions
aj+0.307
ma+0.289
annah+0.279
ppa+0.277
atted+0.271
Neg logit contributions
 spe-0.178
 seem-0.176
 long-0.174
 smell-0.172
fall-0.170
Tokenac
Token ID330
Feature activation-1.48e-01

Pos logit contributions
aj+0.500
annah+0.443
ppa+0.415
atted+0.408
ma+0.400
Neg logit contributions
 long-0.560
 seem-0.559
 spe-0.554
 single-0.549
 smell-0.542
Token was
Token ID373
Feature activation+8.65e-03

Pos logit contributions
aj+0.300
annah+0.253
ppa+0.229
ins+0.218
etically+0.215
Neg logit contributions
 clouds-0.287
 sunset-0.286
 view-0.277
 long-0.271
 night-0.270
Token preparing
Token ID10629
Feature activation-2.80e-02

Pos logit contributions
 long+0.009
 smell+0.008
 spe+0.007
 seem+0.007
 view+0.007
Neg logit contributions
aj-0.026
ppa-0.025
annah-0.024
ma-0.024
atted-0.024
Token breakfast
Token ID12607
Feature activation+3.42e-02

Pos logit contributions
aj+0.074
ma+0.069
annah+0.067
ppa+0.067
atted+0.065
Neg logit contributions
 seem-0.036
 spe-0.036
fall-0.035
 dot-0.034
 smell-0.033
Token in
Token ID287
Feature activation-4.77e-03

Pos logit contributions
 seem+0.040
 spe+0.040
 long+0.039
fall+0.039
 smell+0.038
Neg logit contributions
aj-0.094
ma-0.087
ppa-0.087
annah-0.086
atted-0.083
Token,
Token ID11
Feature activation-7.69e-02

Pos logit contributions
 long+0.029
 seem+0.029
fall+0.027
 spe+0.026
 smell+0.026
Neg logit contributions
aj-0.061
ppa-0.058
annah-0.056
ma-0.055
atted-0.055
Token a
Token ID257
Feature activation+6.69e-02

Pos logit contributions
aj+0.241
annah+0.226
ppa+0.225
atted+0.218
ma+0.216
Neg logit contributions
 long-0.074
 seem-0.060
 spe-0.060
 dark-0.059
 smell-0.058
Token gentle
Token ID10296
Feature activation-3.40e-02

Pos logit contributions
 seem+0.089
 long+0.087
fall+0.087
 spe+0.086
 smell+0.085
Neg logit contributions
aj-0.173
ppa-0.160
ma-0.159
annah-0.157
atted-0.154
Token girl
Token ID2576
Feature activation-1.11e-02

Pos logit contributions
aj+0.088
ma+0.082
ppa+0.082
annah+0.081
atted+0.079
Neg logit contributions
 seem-0.044
fall-0.042
 long-0.042
 spe-0.041
 smell-0.040
Token named
Token ID3706
Feature activation-9.66e-02

Pos logit contributions
aj+0.025
annah+0.022
ppa+0.022
aret+0.021
ins+0.021
Neg logit contributions
 spe-0.020
 long-0.020
 smell-0.019
 seem-0.018
 ever-0.018
Token Ci
Token ID37685
Feature activation-2.71e-01

Pos logit contributions
aj+0.244
ppa+0.223
atted+0.220
ins+0.219
annah+0.211
Neg logit contributions
 long-0.156
 smell-0.145
 seem-0.143
 dark-0.133
 view-0.133
Tokenara
Token ID3301
Feature activation-1.74e-01

Pos logit contributions
aj+0.412
atted+0.375
aret+0.351
ppa+0.341
ins+0.327
Neg logit contributions
 view-0.601
 seem-0.598
fall-0.595
 smell-0.593
 night-0.588
Token stumbled
Token ID24241
Feature activation-6.24e-02

Pos logit contributions
aj+0.521
annah+0.484
ppa+0.470
atted+0.465
aret+0.455
Neg logit contributions
 long-0.175
 smell-0.170
 seem-0.163
 like-0.163
 ever-0.163
Token upon
Token ID2402
Feature activation-4.89e-02

Pos logit contributions
aj+0.186
ppa+0.174
ma+0.173
annah+0.172
atted+0.170
Neg logit contributions
 long-0.060
fall-0.057
 seem-0.056
 smell-0.055
 view-0.054
Token a
Token ID257
Feature activation+8.39e-02

Pos logit contributions
aj+0.092
ma+0.091
ena+0.080
annah+0.080
ppa+0.080
Neg logit contributions
 spe-0.085
 dot-0.077
fall-0.076
 Hipp-0.072
storm-0.071
Token lovely
Token ID14081
Feature activation-8.70e-03

Pos logit contributions
 seem+0.127
fall+0.115
t+0.099
 spe+0.099
 ever+0.098
Neg logit contributions
aj-0.210
ma-0.197
ppa-0.194
annah-0.185
aret-0.178
Token
Token ID198
Feature activation-1.52e-02

Pos logit contributions
 long+0.064
fall+0.061
 seem+0.057
 Pretty+0.056
 spe+0.054
Neg logit contributions
aj-0.187
ma-0.179
ppa-0.178
aret-0.171
annah-0.170
Token
Token ID198
Feature activation-1.18e-02

Pos logit contributions
aj+0.050
ppa+0.049
eline+0.049
aret+0.048
atted+0.048
Neg logit contributions
 long-0.009
 much-0.007
 lot-0.007
fall-0.006
br-0.005
TokenThe
Token ID464
Feature activation-5.84e-02

Pos logit contributions
aj+0.030
aret+0.028
ma+0.028
atted+0.027
ppa+0.027
Neg logit contributions
 long-0.017
br-0.015
fall-0.015
 lot-0.015
 seem-0.015
Token man
Token ID582
Feature activation-2.43e-02

Pos logit contributions
aj+0.153
ppa+0.141
ma+0.140
annah+0.140
atted+0.137
Neg logit contributions
 long-0.078
 seem-0.077
 smell-0.076
 spe-0.076
fall-0.075
Token gave
Token ID2921
Feature activation-1.44e-01

Pos logit contributions
aj+0.054
annah+0.049
ppa+0.048
aret+0.047
ins+0.045
Neg logit contributions
 spe-0.043
 long-0.042
 smell-0.042
 seem-0.040
 ever-0.039
Token Fro
Token ID9734
Feature activation-2.70e-01

Pos logit contributions
aj+0.388
ppa+0.345
annah+0.340
ma+0.332
aret+0.328
Neg logit contributions
 long-0.207
 seem-0.191
 smell-0.190
 view-0.189
 spe-0.187
Tokenggie
Token ID23571
Feature activation-1.17e-01

Pos logit contributions
aj+0.556
ma+0.512
annah+0.510
atted+0.507
aret+0.506
Neg logit contributions
 seem-0.490
fall-0.488
 long-0.475
 smell-0.467
t-0.449
Token a
Token ID257
Feature activation+9.94e-02

Pos logit contributions
aj+0.385
ppa+0.362
ma+0.356
annah+0.355
aret+0.350
Neg logit contributions
 long-0.083
 smell-0.070
 seem-0.070
 view-0.067
fall-0.064
Token soft
Token ID2705
Feature activation-1.52e-02

Pos logit contributions
 seem+0.126
fall+0.123
 spe+0.119
 long+0.111
 smell+0.109
Neg logit contributions
aj-0.267
ma-0.243
ppa-0.238
annah-0.237
atted-0.234
Token cloth
Token ID16270
Feature activation+1.34e-01

Pos logit contributions
aj+0.037
ma+0.034
annah+0.033
atted+0.033
ppa+0.032
Neg logit contributions
 seem-0.023
fall-0.023
 spe-0.021
 long-0.021
 smell-0.020
Token and
Token ID290
Feature activation+2.78e-02

Pos logit contributions
 seem+0.206
fall+0.199
 spe+0.197
 long+0.194
 smell+0.193
Neg logit contributions
aj-0.320
ma-0.292
ppa-0.291
atted-0.288
annah-0.286
Token,
Token ID11
Feature activation-7.48e-03

Pos logit contributions
aj+0.108
ppa+0.098
ma+0.098
annah+0.097
atted+0.096
Neg logit contributions
 seem-0.064
 long-0.063
 spe-0.062
fall-0.062
 smell-0.061
Token Nem
Token ID22547
Feature activation-1.06e-01

Pos logit contributions
aj+0.018
ppa+0.016
ma+0.016
annah+0.016
atted+0.016
Neg logit contributions
 spe-0.011
 seem-0.011
 long-0.011
fall-0.011
 smell-0.010
Tokeno
Token ID78
Feature activation-7.70e-02

Pos logit contributions
aj+0.166
ppa+0.144
ma+0.143
annah+0.141
atted+0.137
Neg logit contributions
 long-0.248
 seem-0.248
 spe-0.246
fall-0.246
 smell-0.244
Token!"
Token ID2474
Feature activation-2.85e-02

Pos logit contributions
aj+0.213
annah+0.194
ppa+0.192
ma+0.191
atted+0.191
Neg logit contributions
 long-0.092
 spe-0.090
 smell-0.089
fall-0.089
 seem-0.087
Token said
Token ID531
Feature activation-1.91e-01

Pos logit contributions
aj+0.084
ma+0.078
ppa+0.078
annah+0.077
atted+0.076
Neg logit contributions
 long-0.028
 seem-0.027
fall-0.027
 spe-0.027
 smell-0.027
Token Oscar
Token ID15694
Feature activation-2.70e-01

Pos logit contributions
aj+0.437
ppa+0.402
ma+0.400
annah+0.398
atted+0.391
Neg logit contributions
 seem-0.300
 long-0.295
 spe-0.295
fall-0.292
 view-0.289
Token.
Token ID13
Feature activation+4.36e-02

Pos logit contributions
aj+0.973
annah+0.878
ma+0.874
ins+0.870
ppa+0.863
Neg logit contributions
 long-0.122
 seem-0.120
 smell-0.111
 like-0.096
fall-0.089
Token "
Token ID366
Feature activation+2.15e-02

Pos logit contributions
 long+0.062
 seem+0.061
fall+0.061
 spe+0.060
 smell+0.060
Neg logit contributions
aj-0.110
ma-0.101
ppa-0.101
annah-0.099
atted-0.098
TokenWould
Token ID17353
Feature activation-1.21e-01

Pos logit contributions
 long+0.029
 spe+0.028
 seem+0.028
fall+0.028
 view+0.027
Neg logit contributions
aj-0.052
ppa-0.051
ma-0.049
annah-0.048
atted-0.047
Token you
Token ID345
Feature activation+6.09e-03

Pos logit contributions
aj+0.277
ma+0.268
ppa+0.263
annah+0.258
atted+0.252
Neg logit contributions
 dot-0.186
 spe-0.184
 smell-0.177
fall-0.174
 seem-0.171
Token like
Token ID588
Feature activation+2.39e-02

Pos logit contributions
 spe+0.005
fall+0.005
 hurricane+0.004
 tornado+0.004
storm+0.004
Neg logit contributions
aj-0.019
ma-0.018
ppa-0.017
atted-0.017
anut-0.017
Tokenbles
Token ID7689
Feature activation+1.06e-02

Pos logit contributions
aj+0.272
ma+0.239
annah+0.237
aret+0.222
atted+0.219
Neg logit contributions
 seem-0.433
 long-0.430
fall-0.422
 dot-0.422
 smell-0.419
Token.
Token ID13
Feature activation+6.06e-02

Pos logit contributions
 spe+0.012
 long+0.012
fall+0.012
 seem+0.012
 smell+0.011
Neg logit contributions
aj-0.029
ma-0.027
ppa-0.027
atted-0.027
annah-0.027
Token Bub
Token ID19645
Feature activation-1.85e-01

Pos logit contributions
 long+0.077
 seem+0.077
fall+0.076
 spe+0.075
 smell+0.075
Neg logit contributions
aj-0.160
ppa-0.148
ma-0.147
annah-0.146
atted-0.143
Tokenbles
Token ID7689
Feature activation+4.37e-03

Pos logit contributions
aj+0.282
ma+0.251
annah+0.249
aret+0.236
ppa+0.230
Neg logit contributions
 seem-0.431
 long-0.419
 dot-0.417
 smell-0.413
 view-0.411
Token asked
Token ID1965
Feature activation-1.89e-01

Pos logit contributions
 seem+0.005
 smell+0.005
 long+0.005
 sunset+0.004
 view+0.004
Neg logit contributions
aj-0.013
annah-0.012
ppa-0.012
ma-0.012
aret-0.011
Token P
Token ID350
Feature activation-2.70e-01

Pos logit contributions
aj+0.585
annah+0.525
ppa+0.524
ma+0.506
atted+0.494
Neg logit contributions
 long-0.216
 seem-0.195
 like-0.190
 smell-0.189
 lot-0.182
Tokeninch
Token ID8589
Feature activation-8.85e-02

Pos logit contributions
aj+0.383
ppa+0.356
ma+0.352
annah+0.344
aret+0.337
Neg logit contributions
 seem-0.631
 long-0.608
 smell-0.606
fall-0.603
 dot-0.603
Tokeny
Token ID88
Feature activation-1.14e-01

Pos logit contributions
aj+0.209
annah+0.197
ppa+0.192
ma+0.184
aret+0.184
Neg logit contributions
 seem-0.133
 long-0.133
t-0.126
 single-0.125
 smell-0.125
Token,
Token ID11
Feature activation-8.00e-02

Pos logit contributions
aj+0.348
ppa+0.322
annah+0.322
ma+0.308
eline+0.301
Neg logit contributions
 long-0.116
 seem-0.111
 smell-0.104
 like-0.098
 look-0.096
Token "
Token ID366
Feature activation+2.79e-02

Pos logit contributions
aj+0.166
ppa+0.154
ma+0.152
annah+0.149
atted+0.146
Neg logit contributions
 spe-0.137
fall-0.136
 seem-0.135
 view-0.133
 smell-0.131
TokenWhy
Token ID5195
Feature activation+6.21e-02

Pos logit contributions
 long+0.035
 seem+0.035
 spe+0.034
fall+0.034
 smell+0.033
Neg logit contributions
aj-0.073
ppa-0.069
ma-0.069
annah-0.066
atted-0.066
TokenT
Token ID51
Feature activation-1.93e-01

Pos logit contributions
aj+0.066
atted+0.060
annah+0.060
ppa+0.058
ma+0.058
Neg logit contributions
 long-0.046
 seem-0.043
fall-0.043
 view-0.042
 lot-0.042
Tokenoby
Token ID26730
Feature activation-1.18e-01

Pos logit contributions
aj+0.333
ppa+0.302
annah+0.290
ma+0.284
atted+0.282
Neg logit contributions
 seem-0.414
fall-0.414
 long-0.408
 smell-0.405
 dot-0.403
Token smiled
Token ID13541
Feature activation-1.43e-01

Pos logit contributions
aj+0.293
ppa+0.267
annah+0.255
ins+0.254
ma+0.254
Neg logit contributions
 smell-0.176
 long-0.173
 seem-0.171
 spe-0.168
fall-0.162
Token and
Token ID290
Feature activation-3.70e-02

Pos logit contributions
aj+0.306
atted+0.276
ma+0.273
ppa+0.272
annah+0.265
Neg logit contributions
 seem-0.249
 spe-0.246
fall-0.242
 long-0.232
 smell-0.231
Token thanked
Token ID26280
Feature activation-2.13e-01

Pos logit contributions
aj+0.079
ma+0.071
ppa+0.071
annah+0.069
ins+0.066
Neg logit contributions
 long-0.069
 smell-0.068
fall-0.065
 seem-0.065
 spe-0.065
Token Ma
Token ID6669
Feature activation-2.69e-01

Pos logit contributions
aj+0.434
ma+0.399
ppa+0.395
annah+0.389
atted+0.379
Neg logit contributions
 seem-0.390
 long-0.386
 spe-0.386
fall-0.385
 smell-0.378
Tokenis
Token ID271
Feature activation-1.69e-01

Pos logit contributions
aj+0.419
annah+0.349
ppa+0.347
atted+0.340
aret+0.325
Neg logit contributions
 seem-0.643
fall-0.632
 long-0.628
 smell-0.618
 spe-0.617
Tokenie
Token ID494
Feature activation-1.36e-01

Pos logit contributions
aj+0.310
annah+0.243
ppa+0.240
eline+0.235
etically+0.209
Neg logit contributions
 seem-0.361
 long-0.351
 smell-0.350
 taste-0.340
 ever-0.337
Token for
Token ID329
Feature activation-4.20e-02

Pos logit contributions
aj+0.394
ppa+0.361
annah+0.353
ma+0.349
atted+0.348
Neg logit contributions
 long-0.150
 seem-0.143
 smell-0.137
fall-0.134
 view-0.130
Token being
Token ID852
Feature activation+1.92e-02

Pos logit contributions
aj+0.114
annah+0.097
ppa+0.097
atted+0.095
ma+0.093
Neg logit contributions
 long-0.061
 smell-0.059
 seem-0.058
 spe-0.057
 view-0.055
Token so
Token ID523
Feature activation+8.36e-02

Pos logit contributions
 long+0.015
 spe+0.013
 smell+0.013
 seem+0.013
 quite+0.012
Neg logit contributions
aj-0.063
annah-0.058
ppa-0.058
ma-0.057
atted-0.056
Token yellow
Token ID7872
Feature activation+2.16e-03

Pos logit contributions
 seem+0.109
fall+0.108
 spe+0.101
 smell+0.098
 long+0.098
Neg logit contributions
aj-0.289
ma-0.266
ppa-0.264
annah-0.261
atted-0.254
Token.
Token ID13
Feature activation+1.04e-01

Pos logit contributions
 seem+0.004
fall+0.004
 spe+0.004
 long+0.004
 smell+0.004
Neg logit contributions
aj-0.004
ma-0.004
ins-0.003
annah-0.003
atted-0.003
Token And
Token ID843
Feature activation-4.45e-04

Pos logit contributions
 view+0.130
fall+0.128
 seem+0.128
 long+0.126
 spe+0.125
Neg logit contributions
aj-0.275
annah-0.245
ma-0.244
ppa-0.241
atted-0.236
Token this
Token ID428
Feature activation+1.07e-02

Pos logit contributions
aj+0.001
ma+0.001
ppa+0.001
annah+0.001
atted+0.001
Neg logit contributions
fall-0.001
 seem-0.001
 spe-0.001
 view-0.001
 long-0.001
Token is
Token ID318
Feature activation+1.01e-01

Pos logit contributions
fall+0.015
 seem+0.015
 spe+0.015
 long+0.014
 smell+0.014
Neg logit contributions
aj-0.027
ma-0.025
ppa-0.024
annah-0.024
atted-0.024
Token a
Token ID257
Feature activation+1.70e-01

Pos logit contributions
 seem+0.135
fall+0.134
 spe+0.130
 long+0.130
 smell+0.128
Neg logit contributions
aj-0.259
ma-0.237
ppa-0.237
annah-0.234
atted-0.229
Token sky
Token ID6766
Feature activation+2.22e-01

Pos logit contributions
 seem+0.219
fall+0.206
 spe+0.179
t+0.175
 Pretty+0.171
Neg logit contributions
aj-0.453
ma-0.427
ppa-0.412
annah-0.401
aret-0.394
Token.
Token ID13
Feature activation+1.03e-01

Pos logit contributions
 seem+0.312
fall+0.306
 long+0.301
 spe+0.299
 view+0.295
Neg logit contributions
aj-0.556
ma-0.523
ppa-0.496
annah-0.494
atted-0.482
Token It
Token ID632
Feature activation+1.25e-01

Pos logit contributions
 view+0.139
 long+0.138
fall+0.135
 seem+0.135
 smell+0.134
Neg logit contributions
aj-0.259
ma-0.229
annah-0.227
ppa-0.224
atted-0.220
Token is
Token ID318
Feature activation+9.93e-02

Pos logit contributions
fall+0.233
 long+0.228
 view+0.219
 spe+0.214
 seem+0.211
Neg logit contributions
aj-0.242
ma-0.230
atted-0.215
ins-0.212
ppa-0.206
Token blue
Token ID4171
Feature activation-6.83e-04

Pos logit contributions
 seem+0.111
fall+0.111
 spe+0.099
 smell+0.096
t+0.095
Neg logit contributions
aj-0.278
ma-0.256
ppa-0.252
annah-0.247
aret-0.244
Token the
Token ID262
Feature activation+3.39e-02

Pos logit contributions
fall+0.261
 seem+0.234
 spe+0.227
 dot+0.225
 smell+0.217
Neg logit contributions
ma-0.409
aj-0.393
atted-0.365
annah-0.357
aret-0.356
Token sunlight
Token ID19606
Feature activation+1.65e-01

Pos logit contributions
fall+0.037
 seem+0.037
t+0.035
 spe+0.032
 Pretty+0.031
Neg logit contributions
ma-0.086
aj-0.085
aret-0.079
ppa-0.077
annah-0.075
Token reflects
Token ID12497
Feature activation+4.91e-02

Pos logit contributions
 long+0.289
fall+0.284
 spe+0.279
 view+0.272
 dot+0.267
Neg logit contributions
ma-0.334
aj-0.323
atted-0.298
annah-0.292
ins-0.291
Token off
Token ID572
Feature activation-3.18e-03

Pos logit contributions
 seem+0.072
fall+0.072
 spe+0.070
 long+0.069
 smell+0.067
Neg logit contributions
aj-0.116
ma-0.106
annah-0.104
atted-0.104
ppa-0.104
Token the
Token ID262
Feature activation+7.92e-03

Pos logit contributions
aj+0.007
ma+0.007
annah+0.006
atted+0.006
ppa+0.006
Neg logit contributions
 seem-0.005
fall-0.005
 spe-0.005
 dot-0.004
 smell-0.004
Token ocean
Token ID9151
Feature activation+1.40e-01

Pos logit contributions
 seem+0.008
fall+0.007
t+0.007
 Pretty+0.007
 spe+0.007
Neg logit contributions
aj-0.022
ma-0.021
aret-0.020
annah-0.020
ppa-0.019
Token."
Token ID526
Feature activation-1.48e-02

Pos logit contributions
 spe+0.186
 seem+0.182
fall+0.178
 long+0.178
 smell+0.172
Neg logit contributions
aj-0.352
ma-0.327
annah-0.321
ppa-0.318
atted-0.308
Token
Token ID220
Feature activation-4.40e-02

Pos logit contributions
aj+0.044
ppa+0.041
ma+0.040
atted+0.040
annah+0.039
Neg logit contributions
fall-0.014
 long-0.014
 seem-0.014
 Pretty-0.013
 sunset-0.012
Token
Token ID198
Feature activation-2.75e-02

Pos logit contributions
aj+0.139
ppa+0.137
ma+0.131
atted+0.131
annah+0.130
Neg logit contributions
 seem-0.031
 long-0.031
fall-0.028
 Pretty-0.026
 much-0.024
Token
Token ID198
Feature activation-1.97e-02

Pos logit contributions
aj+0.092
atted+0.088
ppa+0.085
annah+0.082
aret+0.082
Neg logit contributions
 long-0.019
fall-0.016
 lot-0.016
 much-0.015
 seem-0.015
TokenThe
Token ID464
Feature activation-7.63e-02

Pos logit contributions
aj+0.052
atted+0.049
ppa+0.046
annah+0.045
ma+0.045
Neg logit contributions
 long-0.027
 seem-0.025
 lot-0.024
fall-0.024
 brightly-0.024
Token are
Token ID389
Feature activation+9.44e-02

Pos logit contributions
fall+0.057
 spe+0.057
 long+0.055
 single+0.052
 view+0.051
Neg logit contributions
aj-0.125
ma-0.120
ppa-0.118
annah-0.115
atted-0.113
Token so
Token ID523
Feature activation+1.33e-01

Pos logit contributions
fall+0.093
 seem+0.091
 spe+0.088
 long+0.087
 smell+0.083
Neg logit contributions
aj-0.274
ma-0.262
ppa-0.257
annah-0.256
atted-0.249
Token helpful
Token ID7613
Feature activation-3.70e-02

Pos logit contributions
 seem+0.201
fall+0.199
 long+0.198
 spe+0.197
 smell+0.194
Neg logit contributions
aj-0.321
ppa-0.294
ma-0.294
annah-0.291
atted-0.285
Token,
Token ID11
Feature activation+4.42e-03

Pos logit contributions
aj+0.094
ma+0.087
ppa+0.086
annah+0.086
atted+0.084
Neg logit contributions
 seem-0.051
 spe-0.050
 long-0.050
fall-0.050
 smell-0.049
Token and
Token ID290
Feature activation+3.23e-02

Pos logit contributions
 long+0.008
 smell+0.007
 seem+0.007
 spe+0.007
 view+0.007
Neg logit contributions
aj-0.010
ppa-0.009
atted-0.009
annah-0.009
ma-0.009
Token I
Token ID314
Feature activation+1.43e-01

Pos logit contributions
 long+0.033
 seem+0.033
 spe+0.032
fall+0.032
 smell+0.032
Neg logit contributions
aj-0.094
ppa-0.087
ma-0.086
annah-0.086
atted-0.085
Token'm
Token ID1101
Feature activation+7.65e-02

Pos logit contributions
 long+0.128
 spe+0.125
fall+0.124
 seem+0.124
 view+0.117
Neg logit contributions
aj-0.428
ma-0.400
ppa-0.398
annah-0.394
atted-0.388
Token so
Token ID523
Feature activation+1.44e-01

Pos logit contributions
 long+0.117
 seem+0.115
 smell+0.114
 spe+0.114
fall+0.111
Neg logit contributions
aj-0.189
ppa-0.169
ma-0.167
annah-0.166
atted-0.165
Token proud
Token ID6613
Feature activation-1.66e-02

Pos logit contributions
 long+0.251
 seem+0.251
 spe+0.248
fall+0.247
 smell+0.245
Neg logit contributions
aj-0.312
ppa-0.282
ma-0.281
annah-0.278
atted-0.272
Token of
Token ID286
Feature activation+4.68e-02

Pos logit contributions
aj+0.043
ma+0.041
annah+0.039
atted+0.038
ins+0.038
Neg logit contributions
 seem-0.022
 spe-0.021
fall-0.021
 smell-0.020
 dot-0.020
Token you
Token ID345
Feature activation+7.43e-02

Pos logit contributions
 long+0.049
 seem+0.047
 smell+0.044
 spe+0.044
 view+0.043
Neg logit contributions
aj-0.140
ppa-0.128
atted-0.125
annah-0.124
ma-0.122
Token scissors
Token ID41786
Feature activation+9.89e-02

Pos logit contributions
fall+0.009
 seem+0.009
 spe+0.009
 long+0.008
 smell+0.008
Neg logit contributions
aj-0.013
ma-0.012
ppa-0.012
annah-0.012
atted-0.012
Token,
Token ID11
Feature activation-1.61e-02

Pos logit contributions
fall+0.133
 spe+0.133
 view+0.129
 smell+0.128
 long+0.127
Neg logit contributions
aj-0.245
ma-0.231
annah-0.226
ppa-0.225
atted-0.223
Token but
Token ID475
Feature activation+3.69e-02

Pos logit contributions
aj+0.045
ppa+0.042
ma+0.042
annah+0.042
atted+0.041
Neg logit contributions
 seem-0.018
 long-0.017
fall-0.017
 spe-0.017
 smell-0.017
Token it
Token ID340
Feature activation+8.38e-02

Pos logit contributions
fall+0.037
 spe+0.035
 seem+0.034
 long+0.034
 view+0.034
Neg logit contributions
aj-0.106
ma-0.101
annah-0.098
ppa-0.098
atted-0.097
Token is
Token ID318
Feature activation+1.07e-01

Pos logit contributions
fall+0.143
 long+0.135
 view+0.134
 spe+0.133
 sunset+0.132
Neg logit contributions
aj-0.186
ma-0.172
atted-0.168
annah-0.160
ins-0.158
Token too
Token ID1165
Feature activation+2.09e-01

Pos logit contributions
fall+0.130
 seem+0.118
 view+0.113
 spe+0.112
 smell+0.111
Neg logit contributions
aj-0.298
ma-0.273
ppa-0.267
atted-0.263
annah-0.261
Token hard
Token ID1327
Feature activation+3.41e-02

Pos logit contributions
fall+0.219
 seem+0.218
 spe+0.205
 smell+0.203
 dot+0.187
Neg logit contributions
aj-0.605
ma-0.553
ppa-0.544
annah-0.542
atted-0.529
Token.
Token ID13
Feature activation+6.51e-02

Pos logit contributions
 spe+0.058
 hurricane+0.053
fall+0.052
 smell+0.052
 view+0.052
Neg logit contributions
aj-0.076
ma-0.071
annah-0.068
atted-0.066
ppa-0.064
Token She
Token ID1375
Feature activation+1.38e-02

Pos logit contributions
 seem+0.080
 long+0.079
 spe+0.079
fall+0.078
 smell+0.078
Neg logit contributions
aj-0.175
ma-0.160
ppa-0.160
annah-0.160
atted-0.157
Token pushes
Token ID20070
Feature activation-1.19e-01

Pos logit contributions
 seem+0.025
 long+0.025
 spe+0.025
fall+0.025
 smell+0.025
Neg logit contributions
aj-0.029
ppa-0.026
ma-0.026
annah-0.026
atted-0.025
Token harder
Token ID7069
Feature activation-3.88e-02

Pos logit contributions
aj+0.329
ppa+0.302
annah+0.297
ma+0.296
atted+0.292
Neg logit contributions
 long-0.146
 seem-0.143
 spe-0.138
 smell-0.137
fall-0.137
Token are
Token ID389
Feature activation+9.94e-02

Pos logit contributions
fall+0.104
 sunset+0.100
 spe+0.098
 view+0.098
 brightly+0.098
Neg logit contributions
aj-0.198
ma-0.194
ppa-0.191
annah-0.182
atted-0.180
Token afraid
Token ID7787
Feature activation+1.00e-02

Pos logit contributions
fall+0.160
 spe+0.150
 view+0.149
 seem+0.148
 dot+0.148
Neg logit contributions
aj-0.226
ppa-0.217
ma-0.215
annah-0.208
anut-0.202
Token of
Token ID286
Feature activation+3.63e-02

Pos logit contributions
fall+0.013
 spe+0.013
 smell+0.012
 seem+0.012
 view+0.012
Neg logit contributions
aj-0.026
ma-0.025
atted-0.024
annah-0.023
ppa-0.023
Token everything
Token ID2279
Feature activation+1.26e-01

Pos logit contributions
fall+0.057
 seem+0.055
 spe+0.055
 smell+0.054
 long+0.053
Neg logit contributions
aj-0.085
ma-0.079
ppa-0.077
annah-0.076
atted-0.075
Token.
Token ID13
Feature activation+1.02e-01

Pos logit contributions
 spe+0.161
fall+0.159
 view+0.156
 smell+0.155
 seem+0.150
Neg logit contributions
aj-0.330
ma-0.313
annah-0.303
ppa-0.299
atted-0.291
Token I
Token ID314
Feature activation+1.36e-01

Pos logit contributions
 seem+0.126
 long+0.126
fall+0.125
 spe+0.124
 smell+0.123
Neg logit contributions
aj-0.273
ppa-0.251
ma-0.250
annah-0.249
atted-0.245
Token don
Token ID836
Feature activation+2.93e-02

Pos logit contributions
 long+0.150
fall+0.150
 spe+0.148
 sunset+0.146
 view+0.146
Neg logit contributions
aj-0.372
ma-0.347
ppa-0.340
atted-0.335
annah-0.334
Token't
Token ID470
Feature activation+5.39e-02

Pos logit contributions
 spe+0.035
 smell+0.031
 view+0.031
fall+0.031
 long+0.030
Neg logit contributions
aj-0.076
ma-0.076
annah-0.074
ppa-0.073
atted-0.071
Token want
Token ID765
Feature activation+3.38e-02

Pos logit contributions
 spe+0.078
fall+0.074
 brightly+0.072
 Pretty+0.071
 ceiling+0.070
Neg logit contributions
ppa-0.127
ma-0.125
eline-0.120
annah-0.120
anut-0.116
Token to
Token ID284
Feature activation-6.91e-03

Pos logit contributions
fall+0.046
 spe+0.045
 dot+0.043
 seem+0.042
 smell+0.042
Neg logit contributions
aj-0.080
ma-0.079
atted-0.074
ppa-0.073
annah-0.072
Token play
Token ID711
Feature activation+1.65e-02

Pos logit contributions
aj+0.018
ma+0.017
ppa+0.017
annah+0.016
atted+0.016
Neg logit contributions
 spe-0.008
 brightly-0.008
fall-0.008
 dot-0.008
 single-0.007
Token became
Token ID2627
Feature activation-9.39e-03

Pos logit contributions
 seem+0.029
 long+0.029
fall+0.029
 spe+0.029
 smell+0.028
Neg logit contributions
aj-0.055
ma-0.051
ppa-0.051
annah-0.050
atted-0.050
Token even
Token ID772
Feature activation+2.57e-02

Pos logit contributions
aj+0.024
ma+0.023
ppa+0.023
annah+0.022
atted+0.022
Neg logit contributions
fall-0.012
 seem-0.012
 spe-0.011
 long-0.011
 smell-0.011
Token happier
Token ID23030
Feature activation-2.56e-02

Pos logit contributions
 seem+0.032
 long+0.032
fall+0.031
 spe+0.031
 smell+0.031
Neg logit contributions
aj-0.069
ppa-0.063
ma-0.063
annah-0.063
atted-0.062
Token.
Token ID13
Feature activation+7.63e-02

Pos logit contributions
aj+0.069
ma+0.065
ppa+0.064
annah+0.064
atted+0.063
Neg logit contributions
 seem-0.030
 spe-0.029
fall-0.029
 long-0.029
 smell-0.028
Token
Token ID220
Feature activation-9.38e-03

Pos logit contributions
 seem+0.099
 long+0.098
fall+0.096
 spe+0.095
 smell+0.094
Neg logit contributions
aj-0.202
ppa-0.187
ma-0.184
annah-0.182
atted-0.181
Token
Token ID198
Feature activation-7.16e-03

Pos logit contributions
aj+0.028
ppa+0.027
ma+0.026
atted+0.026
aret+0.026
Neg logit contributions
 seem-0.009
fall-0.008
 long-0.008
 Pretty-0.008
 smell-0.007
Token
Token ID198
Feature activation-4.98e-03

Pos logit contributions
aj+0.025
atted+0.023
ppa+0.022
aret+0.022
annah+0.022
Neg logit contributions
 long-0.005
 seem-0.004
 much-0.004
 view-0.004
 lot-0.004
TokenThe
Token ID464
Feature activation-6.16e-02

Pos logit contributions
aj+0.013
atted+0.012
ppa+0.012
annah+0.012
aret+0.011
Neg logit contributions
 long-0.007
 seem-0.007
 view-0.007
fall-0.006
 sunset-0.006
Token girl
Token ID2576
Feature activation+1.02e-02

Pos logit contributions
aj+0.133
ma+0.129
ppa+0.121
annah+0.120
atted+0.115
Neg logit contributions
fall-0.105
 seem-0.105
 spe-0.103
 long-0.098
 smell-0.094
Token never
Token ID1239
Feature activation-4.98e-02

Pos logit contributions
 spe+0.012
 smell+0.012
 long+0.012
 seem+0.012
 ever+0.012
Neg logit contributions
aj-0.029
ppa-0.026
annah-0.025
aret-0.025
ins-0.024
Token stopped
Token ID5025
Feature activation-6.58e-02

Pos logit contributions
aj+0.098
ppa+0.089
ma+0.088
annah+0.087
atted+0.086
Neg logit contributions
 long-0.095
 spe-0.094
fall-0.094
 seem-0.093
 view-0.092
Token happy
Token ID3772
Feature activation-7.52e-02

Pos logit contributions
 long+0.078
 spe+0.074
 smell+0.073
 seem+0.072
 view+0.069
Neg logit contributions
aj-0.213
ppa-0.198
annah-0.198
ma-0.195
atted-0.192
Token and
Token ID290
Feature activation+1.27e-02

Pos logit contributions
aj+0.160
atted+0.152
ma+0.151
ins+0.141
ppa+0.137
Neg logit contributions
 seem-0.130
 spe-0.120
fall-0.117
 dot-0.116
 smell-0.116
Token warm
Token ID5814
Feature activation-4.23e-02

Pos logit contributions
 seem+0.018
fall+0.018
 spe+0.017
 view+0.017
 dot+0.017
Neg logit contributions
aj-0.031
ppa-0.029
ma-0.028
atted-0.028
annah-0.028
Token,
Token ID11
Feature activation-1.89e-02

Pos logit contributions
aj+0.096
ma+0.089
atted+0.088
ppa+0.085
annah+0.083
Neg logit contributions
 seem-0.071
 spe-0.066
 dot-0.064
 long-0.063
br-0.062
Token and
Token ID290
Feature activation+1.90e-02

Pos logit contributions
aj+0.045
ma+0.041
ppa+0.041
annah+0.040
atted+0.040
Neg logit contributions
 seem-0.029
fall-0.028
 long-0.027
 spe-0.027
 smell-0.027
Token they
Token ID484
Feature activation+6.93e-02

Pos logit contributions
 seem+0.026
fall+0.026
 long+0.025
 spe+0.024
 smell+0.024
Neg logit contributions
aj-0.047
ma-0.044
ppa-0.044
atted-0.043
annah-0.043
Token had
Token ID550
Feature activation+7.90e-02

Pos logit contributions
 long+0.089
 seem+0.088
 spe+0.087
 smell+0.086
fall+0.086
Neg logit contributions
aj-0.183
ppa-0.168
ma-0.167
annah-0.167
atted-0.164
Token sweet
Token ID6029
Feature activation-5.77e-02

Pos logit contributions
 seem+0.062
fall+0.062
 spe+0.059
 smell+0.057
 long+0.057
Neg logit contributions
aj-0.247
ma-0.230
ppa-0.230
annah-0.227
atted-0.224
Token dreams
Token ID10625
Feature activation+1.43e-01

Pos logit contributions
aj+0.135
ma+0.124
atted+0.118
ins+0.109
annah+0.108
Neg logit contributions
 seem-0.095
fall-0.094
 spe-0.092
 dot-0.087
t-0.087
Token.
Token ID13
Feature activation+7.65e-02

Pos logit contributions
 spe+0.189
 dot+0.188
 seem+0.186
 smell+0.182
 single+0.174
Neg logit contributions
aj-0.368
ma-0.339
ppa-0.324
annah-0.319
atted-0.313
Token<|endoftext|>
Token ID50256
Feature activation-1.26e-01

Pos logit contributions
 long+0.091
 seem+0.089
 spe+0.088
fall+0.088
 smell+0.085
Neg logit contributions
aj-0.211
ppa-0.196
ma-0.193
annah-0.193
atted-0.188
Token with
Token ID351
Feature activation-2.95e-03

Pos logit contributions
fall+0.127
 spe+0.125
 seem+0.124
 long+0.116
 single+0.114
Neg logit contributions
aj-0.225
ma-0.217
ppa-0.204
ins-0.203
atted-0.200
Token her
Token ID607
Feature activation-8.82e-02

Pos logit contributions
aj+0.006
ppa+0.005
ma+0.005
annah+0.005
atted+0.005
Neg logit contributions
 long-0.006
 seem-0.006
 spe-0.006
 smell-0.006
fall-0.006
Token cr
Token ID1067
Feature activation+6.32e-02

Pos logit contributions
aj+0.212
ppa+0.196
annah+0.194
ma+0.192
atted+0.189
Neg logit contributions
 long-0.138
 seem-0.133
 spe-0.133
 smell-0.132
fall-0.129
Tokenay
Token ID323
Feature activation+6.77e-03

Pos logit contributions
 long+0.069
 seem+0.067
fall+0.065
 smell+0.065
 spe+0.064
Neg logit contributions
aj-0.176
ppa-0.166
annah-0.165
ma-0.163
atted-0.161
Tokenons
Token ID684
Feature activation+6.12e-02

Pos logit contributions
 long+0.015
 seem+0.014
br+0.014
 lot+0.014
t+0.014
Neg logit contributions
aj-0.012
annah-0.011
ppa-0.011
eline-0.010
etically-0.010
Token.
Token ID13
Feature activation+6.34e-02

Pos logit contributions
 spe+0.074
fall+0.074
 seem+0.073
 smell+0.069
 view+0.069
Neg logit contributions
aj-0.165
ppa-0.155
ma-0.155
annah-0.149
atted-0.147
Token One
Token ID1881
Feature activation+2.74e-02

Pos logit contributions
 long+0.047
 seem+0.046
 smell+0.043
fall+0.042
 spe+0.041
Neg logit contributions
aj-0.204
ppa-0.194
ma-0.190
atted-0.189
annah-0.189
Token day
Token ID1110
Feature activation+5.63e-02

Pos logit contributions
fall+0.023
 spe+0.022
 seem+0.022
 long+0.019
 smell+0.018
Neg logit contributions
aj-0.084
ma-0.082
ppa-0.078
annah-0.078
atted-0.076
Token,
Token ID11
Feature activation-4.12e-02

Pos logit contributions
 spe+0.098
 seem+0.098
fall+0.095
 smell+0.094
 view+0.094
Neg logit contributions
aj-0.121
ma-0.111
annah-0.108
atted-0.108
ppa-0.106
Token her
Token ID607
Feature activation-1.03e-01

Pos logit contributions
aj+0.095
ma+0.089
ppa+0.087
annah+0.087
atted+0.083
Neg logit contributions
 spe-0.064
 seem-0.064
fall-0.064
 view-0.061
 long-0.060
Token mom
Token ID1995
Feature activation-9.14e-02

Pos logit contributions
aj+0.204
ppa+0.185
annah+0.184
ma+0.181
atted+0.179
Neg logit contributions
 long-0.202
 smell-0.197
 spe-0.195
 seem-0.194
fall-0.192
Tokenatted
Token ID16898
Feature activation-1.36e-01

Pos logit contributions
aret+0.009
ppa+0.008
annah+0.008
anut+0.007
 wounds+0.007
Neg logit contributions
t-0.013
fall-0.013
 brightly-0.012
 smell-0.011
ifying-0.011
Token her
Token ID607
Feature activation-1.46e-01

Pos logit contributions
aj+0.328
ppa+0.300
ma+0.300
annah+0.296
atted+0.291
Neg logit contributions
 seem-0.205
 long-0.205
 spe-0.202
fall-0.202
 smell-0.199
Token on
Token ID319
Feature activation-3.78e-02

Pos logit contributions
aj+0.329
ma+0.299
ppa+0.295
annah+0.293
atted+0.286
Neg logit contributions
 seem-0.244
 spe-0.243
fall-0.240
 long-0.238
 smell-0.230
Token the
Token ID262
Feature activation-5.28e-02

Pos logit contributions
aj+0.097
ppa+0.089
ma+0.088
annah+0.088
atted+0.086
Neg logit contributions
 long-0.052
 seem-0.052
 spe-0.051
fall-0.051
 smell-0.050
Token arm
Token ID3211
Feature activation-6.55e-04

Pos logit contributions
aj+0.114
ma+0.104
ppa+0.104
annah+0.103
atted+0.098
Neg logit contributions
 spe-0.092
 seem-0.092
fall-0.090
 long-0.089
 smell-0.087
Token and
Token ID290
Feature activation-9.94e-03

Pos logit contributions
aj+0.002
ma+0.002
annah+0.002
ppa+0.002
atted+0.001
Neg logit contributions
 spe-0.001
 seem-0.001
 long-0.001
fall-0.001
 smell-0.001
Token said
Token ID531
Feature activation-1.78e-01

Pos logit contributions
aj+0.023
ppa+0.021
annah+0.021
ma+0.021
atted+0.020
Neg logit contributions
 long-0.016
 smell-0.016
 seem-0.016
fall-0.016
 spe-0.016
Token goodbye
Token ID24829
Feature activation-1.35e-01

Pos logit contributions
aj+0.448
ppa+0.411
ma+0.410
annah+0.406
atted+0.400
Neg logit contributions
 seem-0.247
 long-0.246
 spe-0.244
fall-0.243
 smell-0.239
Token.
Token ID13
Feature activation+4.76e-02

Pos logit contributions
aj+0.358
ppa+0.328
ma+0.326
annah+0.323
atted+0.313
Neg logit contributions
 seem-0.169
 spe-0.168
 smell-0.157
 view-0.155
fall-0.154
Token
Token ID198
Feature activation-3.08e-02

Pos logit contributions
 long+0.061
 seem+0.061
fall+0.060
 smell+0.059
 spe+0.058
Neg logit contributions
aj-0.126
ppa-0.118
ma-0.116
annah-0.114
atted-0.113
Token
Token ID198
Feature activation-2.76e-02

Pos logit contributions
aj+0.095
ppa+0.090
atted+0.088
annah+0.087
ma+0.086
Neg logit contributions
 long-0.029
fall-0.027
 seem-0.027
 view-0.025
 smell-0.024
Token mom
Token ID1995
Feature activation-7.96e-02

Pos logit contributions
aj+0.180
ppa+0.163
annah+0.162
ma+0.160
atted+0.158
Neg logit contributions
 long-0.159
 spe-0.154
 seem-0.154
 smell-0.154
fall-0.151
Token and
Token ID290
Feature activation-1.46e-02

Pos logit contributions
aj+0.236
atted+0.214
annah+0.214
ppa+0.213
aret+0.209
Neg logit contributions
 long-0.083
 smell-0.077
 seem-0.073
 like-0.072
 spe-0.072
Token they
Token ID484
Feature activation+6.58e-02

Pos logit contributions
aj+0.043
annah+0.039
ppa+0.038
atted+0.037
aret+0.036
Neg logit contributions
 smell-0.018
 long-0.016
 spe-0.016
 taste-0.016
 sunset-0.015
Token both
Token ID1111
Feature activation+4.59e-02

Pos logit contributions
 long+0.086
 seem+0.085
 spe+0.085
 smell+0.085
fall+0.083
Neg logit contributions
aj-0.173
ppa-0.157
annah-0.157
ma-0.156
atted-0.153
Token enjoyed
Token ID8359
Feature activation+3.02e-03

Pos logit contributions
 seem+0.073
 long+0.073
fall+0.073
 spe+0.072
 view+0.070
Neg logit contributions
aj-0.105
ma-0.096
ppa-0.096
annah-0.094
atted-0.094
Token them
Token ID606
Feature activation+4.33e-02

Pos logit contributions
 long+0.004
 seem+0.004
 spe+0.004
 smell+0.004
fall+0.004
Neg logit contributions
aj-0.008
ppa-0.007
ma-0.007
annah-0.007
atted-0.007
Token.
Token ID13
Feature activation+6.72e-02

Pos logit contributions
 spe+0.066
 seem+0.066
fall+0.065
 dot+0.063
 smell+0.062
Neg logit contributions
aj-0.104
ma-0.096
ppa-0.095
annah-0.092
atted-0.089
Token Lily
Token ID20037
Feature activation-1.27e-01

Pos logit contributions
 long+0.092
 Pretty+0.088
 seem+0.088
fall+0.086
 smell+0.086
Neg logit contributions
aj-0.172
ppa-0.166
ma-0.163
annah-0.154
atted-0.154
Token's
Token ID338
Feature activation-2.44e-02

Pos logit contributions
aj+0.396
annah+0.356
aret+0.352
ppa+0.346
atted+0.339
Neg logit contributions
 smell-0.127
 taste-0.107
 really-0.106
 spe-0.105
 ever-0.104
Token mom
Token ID1995
Feature activation-7.55e-02

Pos logit contributions
aj+0.050
ppa+0.046
annah+0.045
ma+0.044
atted+0.044
Neg logit contributions
 long-0.047
 smell-0.046
 seem-0.045
 spe-0.045
 view-0.044
Token hugged
Token ID41130
Feature activation-1.58e-01

Pos logit contributions
aj+0.249
annah+0.225
aret+0.214
ags+0.208
ppa+0.207
Neg logit contributions
 smell-0.072
 ever-0.064
 really-0.064
 taste-0.061
 even-0.061
Token.
Token ID13
Feature activation+5.86e-02

Pos logit contributions
aj+0.173
ma+0.162
ppa+0.161
annah+0.160
atted+0.154
Neg logit contributions
 spe-0.075
fall-0.071
 seem-0.070
 long-0.069
 view-0.069
Token She
Token ID1375
Feature activation-5.39e-04

Pos logit contributions
 long+0.071
 seem+0.067
 smell+0.067
fall+0.067
 Pretty+0.064
Neg logit contributions
aj-0.161
ppa-0.155
ma-0.152
atted-0.146
annah-0.145
Token sat
Token ID3332
Feature activation-1.12e-01

Pos logit contributions
aj+0.001
annah+0.001
aret+0.001
ppa+0.001
ins+0.001
Neg logit contributions
 smell-0.001
 long-0.001
 spe-0.001
 quite-0.001
 taste-0.001
Token down
Token ID866
Feature activation-1.54e-02

Pos logit contributions
aj+0.371
annah+0.353
aret+0.346
ppa+0.343
atted+0.333
Neg logit contributions
 view-0.066
 long-0.064
 smell-0.063
 brightly-0.061
 dark-0.060
Token on
Token ID319
Feature activation-2.11e-02

Pos logit contributions
aj+0.041
ppa+0.038
ma+0.038
annah+0.038
atted+0.037
Neg logit contributions
 long-0.019
 seem-0.019
 spe-0.019
fall-0.019
 smell-0.018
Token the
Token ID262
Feature activation-3.24e-02

Pos logit contributions
aj+0.047
ma+0.045
annah+0.044
ppa+0.043
aret+0.041
Neg logit contributions
 seem-0.033
fall-0.033
 spe-0.032
 smell-0.031
 long-0.031
Token bench
Token ID7624
Feature activation+4.68e-02

Pos logit contributions
aj+0.079
ppa+0.072
ma+0.071
atted+0.071
annah+0.070
Neg logit contributions
 long-0.049
 seem-0.048
 smell-0.047
fall-0.047
 spe-0.047
Token and
Token ID290
Feature activation-3.33e-03

Pos logit contributions
 long+0.061
 smell+0.056
fall+0.056
 view+0.055
 seem+0.055
Neg logit contributions
aj-0.127
ppa-0.118
annah-0.115
aret-0.113
ma-0.113
Token started
Token ID2067
Feature activation+1.22e-02

Pos logit contributions
aj+0.008
ppa+0.007
ma+0.007
annah+0.007
atted+0.007
Neg logit contributions
 long-0.005
 smell-0.005
 seem-0.005
fall-0.005
 spe-0.005
Token to
Token ID284
Feature activation-1.39e-02

Pos logit contributions
 seem+0.012
 long+0.012
 spe+0.012
fall+0.012
 smell+0.011
Neg logit contributions
aj-0.036
ppa-0.033
ma-0.033
annah-0.033
atted-0.033
Token play
Token ID711
Feature activation+7.04e-04

Pos logit contributions
aj+0.035
ma+0.033
ppa+0.032
annah+0.032
atted+0.031
Neg logit contributions
fall-0.018
 spe-0.018
 long-0.018
 seem-0.018
 single-0.017
Token excited
Token ID6568
Feature activation-7.57e-02

Pos logit contributions
 long+0.010
 smell+0.010
 seem+0.009
fall+0.009
 spe+0.009
Neg logit contributions
aj-0.017
ma-0.016
ppa-0.015
annah-0.015
atted-0.015
Token to
Token ID284
Feature activation-2.44e-02

Pos logit contributions
aj+0.190
ma+0.180
annah+0.177
atted+0.175
ppa+0.173
Neg logit contributions
 spe-0.117
 dot-0.101
 seem-0.093
br-0.092
 view-0.090
Token eat
Token ID4483
Feature activation-8.09e-02

Pos logit contributions
aj+0.065
ma+0.061
ppa+0.060
annah+0.060
atted+0.059
Neg logit contributions
 spe-0.029
fall-0.029
 seem-0.029
 long-0.029
 view-0.027
Token it
Token ID340
Feature activation+3.43e-02

Pos logit contributions
aj+0.181
ma+0.172
ppa+0.168
annah+0.165
atted+0.163
Neg logit contributions
 seem-0.128
 spe-0.126
fall-0.124
 dot-0.120
 long-0.118
Token.
Token ID13
Feature activation+6.74e-02

Pos logit contributions
 spe+0.061
 dot+0.054
 seem+0.052
 hurricane+0.051
fall+0.050
Neg logit contributions
aj-0.073
ma-0.072
annah-0.069
ppa-0.068
anut-0.065
Token
Token ID198
Feature activation-1.48e-02

Pos logit contributions
 long+0.068
 seem+0.065
fall+0.064
 smell+0.063
 sunset+0.061
Neg logit contributions
aj-0.198
ppa-0.191
ma-0.186
atted-0.183
annah-0.182
Token
Token ID198
Feature activation-1.28e-02

Pos logit contributions
aj+0.048
ppa+0.045
atted+0.045
annah+0.044
aret+0.042
Neg logit contributions
 long-0.012
 seem-0.011
 view-0.010
t-0.010
fall-0.010
TokenAs
Token ID1722
Feature activation+1.00e-02

Pos logit contributions
aj+0.032
atted+0.029
ppa+0.029
ma+0.029
annah+0.029
Neg logit contributions
 long-0.020
fall-0.019
 seem-0.019
 view-0.019
 smell-0.018
Token Tim
Token ID5045
Feature activation-1.75e-01

Pos logit contributions
 seem+0.017
 spe+0.015
 dot+0.015
fall+0.015
 smell+0.015
Neg logit contributions
ma-0.022
aj-0.021
ppa-0.020
atted-0.020
annah-0.019
Tokenmy
Token ID1820
Feature activation-9.21e-02

Pos logit contributions
aj+0.297
annah+0.272
ppa+0.265
aret+0.260
atted+0.251
Neg logit contributions
 smell-0.375
 taste-0.365
 view-0.361
 sunset-0.359
 Pretty-0.357
Token was
Token ID373
Feature activation+3.72e-02

Pos logit contributions
aj+0.238
ppa+0.220
ma+0.217
annah+0.217
atted+0.212
Neg logit contributions
 long-0.122
 seem-0.121
 spe-0.120
fall-0.119
 smell-0.119
Tokenll
Token ID297
Feature activation-1.32e-01

Pos logit contributions
t+0.175
 long+0.150
 ever+0.138
 much+0.135
 seem+0.131
Neg logit contributions
aj-0.404
aret-0.385
ppa-0.379
annah-0.362
atted-0.353
Token lead
Token ID1085
Feature activation-2.54e-02

Pos logit contributions
aj+0.363
ppa+0.336
ma+0.335
annah+0.332
atted+0.327
Neg logit contributions
 long-0.153
 seem-0.153
 spe-0.150
fall-0.150
 smell-0.147
Token him
Token ID683
Feature activation-6.74e-02

Pos logit contributions
aj+0.066
ppa+0.060
ma+0.059
annah+0.058
atted+0.058
Neg logit contributions
 long-0.036
 seem-0.035
fall-0.034
 smell-0.033
 view-0.033
Token away
Token ID1497
Feature activation-4.58e-02

Pos logit contributions
aj+0.183
ma+0.167
ppa+0.167
annah+0.166
atted+0.164
Neg logit contributions
 seem-0.080
 spe-0.080
fall-0.079
 long-0.078
 smell-0.077
Token from
Token ID422
Feature activation-6.62e-02

Pos logit contributions
aj+0.115
ma+0.105
annah+0.103
ppa+0.103
atted+0.101
Neg logit contributions
 seem-0.062
 spe-0.062
fall-0.060
 smell-0.059
 long-0.058
Token the
Token ID262
Feature activation-1.90e-02

Pos logit contributions
aj+0.167
ma+0.155
ppa+0.153
annah+0.152
atted+0.149
Neg logit contributions
 seem-0.091
fall-0.089
 spe-0.088
 long-0.087
 smell-0.086
Token lake
Token ID13546
Feature activation+1.32e-01

Pos logit contributions
aj+0.043
ma+0.041
ppa+0.038
annah+0.038
anut+0.037
Neg logit contributions
 seem-0.028
 spe-0.026
fall-0.025
 Pretty-0.024
br-0.024
Token and
Token ID290
Feature activation+1.16e-02

Pos logit contributions
 long+0.154
 seem+0.153
 spe+0.150
fall+0.150
 smell+0.147
Neg logit contributions
aj-0.361
ppa-0.334
ma-0.332
annah-0.330
atted-0.325
Token we
Token ID356
Feature activation+9.57e-02

Pos logit contributions
 spe+0.012
 seem+0.012
 long+0.012
fall+0.012
 smell+0.012
Neg logit contributions
aj-0.033
ma-0.030
ppa-0.030
annah-0.030
atted-0.030
Token can
Token ID460
Feature activation+9.45e-02

Pos logit contributions
fall+0.095
 spe+0.095
 long+0.093
 view+0.086
 sunset+0.085
Neg logit contributions
aj-0.270
ma-0.253
atted-0.249
ppa-0.248
annah-0.246
Token all
Token ID477
Feature activation+9.11e-02

Pos logit contributions
 spe+0.114
 long+0.107
fall+0.103
 view+0.100
 dot+0.099
Neg logit contributions
aj-0.245
ma-0.234
annah-0.229
ppa-0.227
atted-0.222
Token time
Token ID640
Feature activation+2.48e-02

Pos logit contributions
 long+0.013
 seem+0.011
 spe+0.010
 smell+0.010
 view+0.010
Neg logit contributions
aj-0.065
ppa-0.062
atted-0.061
annah-0.060
ma-0.060
Token,
Token ID11
Feature activation-8.14e-02

Pos logit contributions
 long+0.033
 seem+0.033
fall+0.031
 spe+0.031
 smell+0.031
Neg logit contributions
aj-0.065
ppa-0.061
annah-0.059
ma-0.059
atted-0.058
Token there
Token ID612
Feature activation+3.45e-05

Pos logit contributions
aj+0.216
ma+0.200
ppa+0.199
annah+0.197
atted+0.194
Neg logit contributions
 seem-0.101
fall-0.101
 spe-0.100
 long-0.100
 smell-0.098
Token was
Token ID373
Feature activation+6.67e-03

Pos logit contributions
 seem+0.000
 long+0.000
 smell+0.000
 spe+0.000
fall+0.000
Neg logit contributions
aj-0.000
ppa-0.000
annah-0.000
ma-0.000
atted-0.000
Token a
Token ID257
Feature activation+9.88e-02

Pos logit contributions
fall+0.008
 seem+0.008
 spe+0.007
 smell+0.007
t+0.007
Neg logit contributions
aj-0.017
ma-0.017
ppa-0.016
annah-0.016
anut-0.015
Token happy
Token ID3772
Feature activation-4.28e-02

Pos logit contributions
 seem+0.153
fall+0.146
 spe+0.138
 view+0.134
 long+0.134
Neg logit contributions
aj-0.236
ppa-0.219
ma-0.218
annah-0.211
atted-0.201
Token clay
Token ID21558
Feature activation+1.15e-01

Pos logit contributions
aj+0.096
ma+0.091
ppa+0.089
annah+0.086
aret+0.084
Neg logit contributions
 seem-0.074
fall-0.066
 spe-0.063
t-0.062
 Pretty-0.060
Token.
Token ID13
Feature activation+6.04e-02

Pos logit contributions
 seem+0.202
fall+0.188
 long+0.181
 spe+0.176
 view+0.175
Neg logit contributions
aj-0.246
ma-0.244
ppa-0.235
atted-0.223
ins-0.222
Token He
Token ID679
Feature activation+5.83e-03

Pos logit contributions
fall+0.073
 spe+0.072
 seem+0.071
 long+0.067
 view+0.065
Neg logit contributions
aj-0.147
ma-0.140
annah-0.140
aret-0.131
atted-0.130
Token lived
Token ID5615
Feature activation-2.88e-03

Pos logit contributions
 long+0.009
 seem+0.009
 spe+0.009
 smell+0.009
fall+0.009
Neg logit contributions
aj-0.014
ppa-0.012
ma-0.012
annah-0.012
atted-0.012
Token in
Token ID287
Feature activation+2.95e-02

Pos logit contributions
aj+0.009
annah+0.009
ppa+0.009
ma+0.008
atted+0.008
Neg logit contributions
 long-0.003
 view-0.002
 smell-0.002
 spe-0.002
 dark-0.002
Token he
Token ID339
Feature activation-2.09e-03

Pos logit contributions
aj+0.026
ma+0.025
ppa+0.024
annah+0.024
atted+0.024
Neg logit contributions
 spe-0.011
 seem-0.011
fall-0.011
 smell-0.010
 view-0.010
Token decided
Token ID3066
Feature activation-8.19e-02

Pos logit contributions
aj+0.005
annah+0.005
ma+0.005
ppa+0.005
aret+0.005
Neg logit contributions
 long-0.003
 spe-0.003
 smell-0.003
fall-0.003
 seem-0.003
Token to
Token ID284
Feature activation-3.51e-02

Pos logit contributions
aj+0.253
ppa+0.237
ma+0.236
annah+0.234
atted+0.231
Neg logit contributions
 seem-0.067
 spe-0.067
 long-0.066
fall-0.065
 smell-0.064
Token take
Token ID1011
Feature activation-9.94e-02

Pos logit contributions
aj+0.097
ma+0.091
ppa+0.090
annah+0.090
atted+0.087
Neg logit contributions
 spe-0.038
 long-0.038
fall-0.038
 seem-0.038
 smell-0.035
Token it
Token ID340
Feature activation+1.83e-02

Pos logit contributions
aj+0.257
ma+0.248
annah+0.243
ppa+0.241
atted+0.238
Neg logit contributions
 spe-0.125
 seem-0.119
fall-0.119
 smell-0.113
 dot-0.110
Token home
Token ID1363
Feature activation-7.64e-02

Pos logit contributions
 spe+0.029
 seem+0.028
fall+0.027
 smell+0.027
 long+0.026
Neg logit contributions
aj-0.044
ma-0.041
ppa-0.040
annah-0.039
atted-0.038
Token and
Token ID290
Feature activation-1.35e-02

Pos logit contributions
aj+0.213
ppa+0.197
ma+0.195
annah+0.195
atted+0.193
Neg logit contributions
 long-0.087
 seem-0.085
fall-0.084
 spe-0.083
 smell-0.082
Token show
Token ID905
Feature activation-4.13e-02

Pos logit contributions
aj+0.036
ma+0.033
ppa+0.033
annah+0.033
atted+0.032
Neg logit contributions
 spe-0.016
 seem-0.016
fall-0.016
 long-0.016
 dot-0.015
Token his
Token ID465
Feature activation-1.09e-01

Pos logit contributions
aj+0.117
ppa+0.105
ma+0.103
annah+0.103
atted+0.103
Neg logit contributions
 long-0.051
 seem-0.048
 smell-0.047
 view-0.045
fall-0.045
Token owners
Token ID4393
Feature activation-3.07e-03

Pos logit contributions
aj+0.211
ppa+0.191
annah+0.189
ma+0.187
atted+0.183
Neg logit contributions
 long-0.219
 seem-0.212
 spe-0.212
 smell-0.211
 view-0.209
Token.
Token ID13
Feature activation+5.77e-02

Pos logit contributions
aj+0.007
ppa+0.007
annah+0.007
ma+0.007
atted+0.007
Neg logit contributions
 seem-0.005
 long-0.005
 smell-0.005
fall-0.004
 spe-0.004
Token of
Token ID286
Feature activation+1.13e-01

Pos logit contributions
fall+0.220
 spe+0.217
 seem+0.211
 smell+0.205
 Pretty+0.205
Neg logit contributions
aj-0.298
ma-0.283
ins-0.276
annah-0.271
atted-0.271
Token fun
Token ID1257
Feature activation+1.07e-01

Pos logit contributions
fall+0.156
t+0.143
 Pretty+0.140
storm+0.140
 seem+0.136
Neg logit contributions
aj-0.280
ins-0.251
ma-0.251
atted-0.249
ppa-0.235
Token!"
Token ID2474
Feature activation-2.57e-02

Pos logit contributions
 seem+0.160
 spe+0.160
fall+0.159
 smell+0.149
 long+0.148
Neg logit contributions
aj-0.259
ma-0.234
atted-0.228
ins-0.226
annah-0.225
Token
Token ID198
Feature activation-3.51e-02

Pos logit contributions
aj+0.067
ppa+0.061
ma+0.061
annah+0.061
atted+0.060
Neg logit contributions
 seem-0.034
 spe-0.034
 long-0.034
fall-0.033
 smell-0.033
Token
Token ID198
Feature activation-2.72e-02

Pos logit contributions
aj+0.107
ppa+0.105
annah+0.104
atted+0.102
ins+0.100
Neg logit contributions
 long-0.031
 sunset-0.027
 view-0.026
 much-0.026
fall-0.026
TokenBilly
Token ID49640
Feature activation-1.68e-01

Pos logit contributions
aj+0.063
atted+0.058
annah+0.057
ma+0.056
ppa+0.056
Neg logit contributions
 long-0.046
fall-0.043
 view-0.043
 sunset-0.042
 seem-0.042
Token thought
Token ID1807
Feature activation-9.76e-02

Pos logit contributions
aj+0.447
ppa+0.412
annah+0.410
ma+0.398
ins+0.395
Neg logit contributions
 smell-0.219
 seem-0.213
 long-0.211
 like-0.201
 quite-0.198
Token about
Token ID546
Feature activation-4.73e-02

Pos logit contributions
aj+0.227
ppa+0.208
ma+0.205
annah+0.203
atted+0.199
Neg logit contributions
 long-0.157
 seem-0.157
 spe-0.153
fall-0.153
 smell-0.152
Token this
Token ID428
Feature activation+5.71e-03

Pos logit contributions
aj+0.135
ma+0.127
ppa+0.126
annah+0.124
atted+0.123
Neg logit contributions
 spe-0.048
fall-0.048
 seem-0.047
 long-0.047
 smell-0.046
Token for
Token ID329
Feature activation+3.11e-03

Pos logit contributions
 spe+0.009
fall+0.009
 seem+0.009
 long+0.008
 smell+0.008
Neg logit contributions
aj-0.014
ma-0.012
annah-0.012
ppa-0.012
atted-0.012
Token a
Token ID257
Feature activation+1.39e-01

Pos logit contributions
 spe+0.003
fall+0.003
 seem+0.003
 long+0.003
 smell+0.003
Neg logit contributions
aj-0.009
ma-0.009
ppa-0.009
annah-0.008
atted-0.008
Token to
Token ID284
Feature activation-5.32e-02

Pos logit contributions
aj+0.414
ppa+0.390
annah+0.386
ma+0.384
atted+0.378
Neg logit contributions
 long-0.129
 seem-0.127
 smell-0.123
fall-0.121
 spe-0.114
Token him
Token ID683
Feature activation-1.12e-01

Pos logit contributions
aj+0.129
atted+0.110
ppa+0.109
ins+0.108
annah+0.108
Neg logit contributions
 long-0.096
 seem-0.092
 smell-0.092
 taste-0.086
 look-0.085
Token."
Token ID526
Feature activation-3.48e-02

Pos logit contributions
aj+0.352
atted+0.333
ppa+0.332
annah+0.327
eline+0.319
Neg logit contributions
 long-0.091
 seem-0.089
 like-0.084
 smell-0.081
fall-0.078
Token
Token ID198
Feature activation-5.05e-02

Pos logit contributions
aj+0.110
ppa+0.104
atted+0.100
ins+0.099
annah+0.098
Neg logit contributions
 long-0.034
 seem-0.030
fall-0.028
 smell-0.028
 A-0.025
Token
Token ID198
Feature activation-4.35e-02

Pos logit contributions
aj+0.179
ppa+0.171
atted+0.171
annah+0.167
ins+0.163
Neg logit contributions
 long-0.032
 much-0.021
 seem-0.017
 lot-0.017
 dark-0.015
TokenL
Token ID43
Feature activation-1.67e-01

Pos logit contributions
aj+0.110
atted+0.103
ppa+0.099
annah+0.097
ins+0.095
Neg logit contributions
 long-0.070
 much-0.060
 lot-0.059
 single-0.058
 seem-0.058
Tokenily
Token ID813
Feature activation-1.54e-01

Pos logit contributions
aj+0.360
atted+0.315
ppa+0.303
aret+0.303
eline+0.295
Neg logit contributions
 seem-0.289
 like-0.279
 ever-0.278
t-0.269
 smell-0.264
Token looked
Token ID3114
Feature activation-1.95e-02

Pos logit contributions
aj+0.430
ppa+0.417
annah+0.402
aret+0.395
anut+0.388
Neg logit contributions
 smell-0.194
 look-0.176
 seem-0.172
 taste-0.157
 like-0.155
Token at
Token ID379
Feature activation-8.12e-02

Pos logit contributions
aj+0.057
ppa+0.057
ma+0.055
annah+0.055
aret+0.053
Neg logit contributions
 long-0.020
 dark-0.019
 like-0.019
 smell-0.018
 dizz-0.017
Token Sam
Token ID3409
Feature activation-1.81e-01

Pos logit contributions
aj+0.187
ppa+0.168
annah+0.162
atted+0.159
ins+0.158
Neg logit contributions
 long-0.141
 seem-0.138
 smell-0.137
 view-0.135
 sunset-0.133
Token.
Token ID13
Feature activation+3.93e-02

Pos logit contributions
aj+0.548
ppa+0.503
annah+0.491
ma+0.483
atted+0.482
Neg logit contributions
 seem-0.183
 long-0.178
 smell-0.165
fall-0.152
 like-0.147
Token to
Token ID284
Feature activation-6.45e-02

Pos logit contributions
aj+0.236
ma+0.220
ppa+0.220
annah+0.218
atted+0.215
Neg logit contributions
 seem-0.068
 spe-0.067
 long-0.066
fall-0.065
 smell-0.064
Token the
Token ID262
Feature activation-7.48e-02

Pos logit contributions
aj+0.167
ppa+0.154
annah+0.150
atted+0.150
ma+0.149
Neg logit contributions
 long-0.091
 seem-0.090
 smell-0.089
fall-0.087
 view-0.086
Token truck
Token ID7779
Feature activation+2.63e-02

Pos logit contributions
aj+0.158
ma+0.149
ppa+0.143
annah+0.139
aret+0.133
Neg logit contributions
 seem-0.135
 spe-0.134
fall-0.128
 dot-0.125
 long-0.122
Token driver
Token ID4639
Feature activation-4.47e-02

Pos logit contributions
 seem+0.038
 spe+0.038
 long+0.038
fall+0.037
 smell+0.037
Neg logit contributions
aj-0.065
ma-0.060
ppa-0.059
annah-0.059
atted-0.058
Token and
Token ID290
Feature activation-2.06e-02

Pos logit contributions
aj+0.116
ppa+0.107
annah+0.105
ma+0.104
atted+0.103
Neg logit contributions
 seem-0.060
 long-0.060
fall-0.058
 smell-0.058
 spe-0.057
Token said
Token ID531
Feature activation-1.92e-01

Pos logit contributions
aj+0.049
annah+0.042
atted+0.042
ppa+0.042
ins+0.041
Neg logit contributions
 smell-0.035
 long-0.035
 seem-0.034
 spe-0.033
fall-0.033
Token,
Token ID11
Feature activation-9.09e-02

Pos logit contributions
aj+0.468
ma+0.427
ppa+0.427
annah+0.423
atted+0.418
Neg logit contributions
 seem-0.279
 long-0.275
 spe-0.275
fall-0.274
 smell-0.271
Token "
Token ID366
Feature activation+2.04e-02

Pos logit contributions
aj+0.207
ppa+0.189
ma+0.188
annah+0.186
atted+0.183
Neg logit contributions
 seem-0.145
 spe-0.144
 long-0.142
fall-0.141
 smell-0.141
TokenI
Token ID40
Feature activation+1.22e-01

Pos logit contributions
 long+0.029
 seem+0.029
 spe+0.028
fall+0.028
 smell+0.027
Neg logit contributions
aj-0.051
ma-0.048
ppa-0.048
annah-0.046
atted-0.045
Token can
Token ID460
Feature activation+1.26e-01

Pos logit contributions
fall+0.146
 sunset+0.144
 view+0.139
 brightly+0.136
 spe+0.134
Neg logit contributions
aj-0.325
ma-0.311
ppa-0.290
atted-0.288
annah-0.284
Token help
Token ID1037
Feature activation-6.70e-03

Pos logit contributions
 spe+0.152
 sunset+0.152
 view+0.147
fall+0.146
 single+0.141
Neg logit contributions
ma-0.322
aj-0.321
annah-0.308
ppa-0.302
aret-0.281
TokenL
Token ID43
Feature activation-1.61e-01

Pos logit contributions
aj+0.089
atted+0.089
annah+0.080
ppa+0.079
lee+0.079
Neg logit contributions
 long-0.046
 much-0.043
 lot-0.042
 dark-0.036
 while-0.036
Tokenily
Token ID813
Feature activation-1.47e-01

Pos logit contributions
aj+0.337
atted+0.298
ppa+0.293
aret+0.288
annah+0.282
Neg logit contributions
 seem-0.283
 like-0.271
 ever-0.269
t-0.267
 long-0.264
Token felt
Token ID2936
Feature activation-7.04e-03

Pos logit contributions
aj+0.379
aret+0.367
ins+0.356
ppa+0.353
annah+0.353
Neg logit contributions
 smell-0.204
 look-0.181
 seem-0.175
 really-0.174
 like-0.169
Token bad
Token ID2089
Feature activation-5.26e-02

Pos logit contributions
aj+0.015
ppa+0.014
ma+0.014
annah+0.014
atted+0.013
Neg logit contributions
 spe-0.012
fall-0.012
 seem-0.012
 long-0.012
 smell-0.012
Token and
Token ID290
Feature activation+6.36e-03

Pos logit contributions
aj+0.118
ma+0.111
atted+0.108
annah+0.104
ins+0.104
Neg logit contributions
 spe-0.092
 view-0.084
fall-0.082
 seem-0.081
 long-0.080
Token said
Token ID531
Feature activation-1.92e-01

Pos logit contributions
 smell+0.010
 long+0.010
 seem+0.010
 dark+0.009
 spe+0.009
Neg logit contributions
aj-0.016
annah-0.014
ma-0.014
ppa-0.014
aret-0.014
Token,
Token ID11
Feature activation-9.23e-02

Pos logit contributions
aj+0.500
ppa+0.461
ma+0.455
annah+0.453
atted+0.438
Neg logit contributions
 long-0.264
 seem-0.260
fall-0.252
 spe-0.248
 smell-0.248
Token "
Token ID366
Feature activation+1.03e-02

Pos logit contributions
aj+0.232
ppa+0.212
ma+0.212
annah+0.210
atted+0.207
Neg logit contributions
 long-0.133
 seem-0.132
fall-0.129
 spe-0.128
 smell-0.128
TokenI
Token ID40
Feature activation+1.14e-01

Pos logit contributions
 seem+0.012
fall+0.012
 long+0.012
 spe+0.012
 smell+0.011
Neg logit contributions
aj-0.028
ppa-0.027
ma-0.026
annah-0.026
atted-0.025
Token know
Token ID760
Feature activation-5.08e-02

Pos logit contributions
 spe+0.107
fall+0.103
 long+0.102
 view+0.101
 seem+0.101
Neg logit contributions
aj-0.341
ma-0.316
ppa-0.314
annah-0.308
atted-0.304
Token,
Token ID11
Feature activation-8.89e-03

Pos logit contributions
aj+0.128
ma+0.119
ppa+0.115
annah+0.115
atted+0.114
Neg logit contributions
fall-0.070
 spe-0.070
 seem-0.070
 long-0.069
 smell-0.068
Token that
Token ID326
Feature activation+7.89e-02

Pos logit contributions
 seem+0.215
fall+0.207
 spe+0.204
Ice+0.198
 view+0.197
Neg logit contributions
aj-0.259
ma-0.253
ppa-0.236
annah-0.235
aret-0.235
Token noise
Token ID7838
Feature activation+1.48e-01

Pos logit contributions
 spe+0.136
 seem+0.125
fall+0.121
 long+0.120
 dot+0.116
Neg logit contributions
aj-0.172
ma-0.159
ppa-0.149
annah-0.147
aret-0.146
Token?"
Token ID1701
Feature activation-2.94e-02

Pos logit contributions
 spe+0.321
 long+0.311
 seem+0.307
fall+0.307
 view+0.304
Neg logit contributions
aj-0.251
ma-0.231
ppa-0.215
annah-0.214
atted-0.213
Token Her
Token ID2332
Feature activation-1.38e-01

Pos logit contributions
aj+0.079
ppa+0.073
annah+0.071
ma+0.071
atted+0.070
Neg logit contributions
 long-0.039
 seem-0.038
fall-0.038
 smell-0.038
 sunset-0.035
Token friend
Token ID1545
Feature activation-1.86e-02

Pos logit contributions
aj+0.292
annah+0.284
ppa+0.275
ma+0.262
atted+0.256
Neg logit contributions
 long-0.260
 smell-0.254
 seem-0.239
 dark-0.235
 view-0.234
Token said
Token ID531
Feature activation-1.90e-01

Pos logit contributions
aj+0.043
annah+0.037
ppa+0.036
ins+0.036
anut+0.035
Neg logit contributions
 smell-0.033
 long-0.032
 seem-0.032
 look-0.030
 dark-0.030
Token,
Token ID11
Feature activation-8.95e-02

Pos logit contributions
aj+0.470
ppa+0.431
ma+0.430
annah+0.425
atted+0.419
Neg logit contributions
 seem-0.273
 long-0.272
 spe-0.269
fall-0.268
 smell-0.265
Token "
Token ID366
Feature activation+1.45e-02

Pos logit contributions
aj+0.229
ppa+0.209
ma+0.208
annah+0.207
atted+0.202
Neg logit contributions
 long-0.125
 seem-0.125
fall-0.123
 smell-0.123
 spe-0.121
TokenIt
Token ID1026
Feature activation+7.92e-02

Pos logit contributions
 long+0.020
 spe+0.020
fall+0.020
 seem+0.019
 view+0.019
Neg logit contributions
aj-0.033
ma-0.032
ppa-0.032
ins-0.030
atted-0.030
Token's
Token ID338
Feature activation+1.88e-03

Pos logit contributions
 spe+0.164
fall+0.162
 long+0.162
 view+0.159
 seem+0.157
Neg logit contributions
aj-0.141
ma-0.128
atted-0.124
ppa-0.120
aret-0.118
Token thunder
Token ID18355
Feature activation+9.08e-02

Pos logit contributions
 seem+0.002
 spe+0.002
fall+0.002
 view+0.002
 dot+0.002
Neg logit contributions
aj-0.005
ma-0.005
ppa-0.004
aret-0.004
atted-0.004