TOP ACTIVATIONS
MAX = 6.296

was so amazing ! Everywhere it sang , it made people
around the world . Everywhere it went , it made people
." Tim and Lily looked around the movie place
near the clouds . Everywhere it flew , it left a
what it was , so she asked her mom .
happy and free , and it made Lily smile .
feel better . Lily felt a little bit better but still
started to spread . Everywhere it went , something happened .
to be beautiful , so it tried to make itself very
loved to sing . Everywhere it went , it would sing
that little seed . Everywhere it went , it made flowers
!" said Lily generously . They looked everywhere but couldn 't
a gentle lung . Everywhere it went , it made its
was not hurt , but it was scared . The cat
cat named Jazz . Everywhere she went , she was always
it could not . Everywhere it looked there was too much
play with it . Everywhere she went , she felt the
The horse was happy and it bowed its head to Lily
be okay . Lily felt a little weak because she was
what it was , so she asked her mom my .

MIN ACTIVATIONS
MIN = -7.315

Everyone smiled and agreed . In the game
without her mom â s help . When she
so they asked Bob . One day ,
wouldn 't allow it . Sam my was
to beat Gray â s prices , because everyone
safe behind Jack â s shield . <|endoftext|> One
â She asked . To her surprise
wouldn 't let him . " Mom ,
keep talking and ask silly questions .
of a new idea . He decided to
and try hard â . All
important ! â . They took the
pan at John â s feet . He then
with his friends instead . So Peter and
at the farmer â s market . Whenever she
â she said . But Peter wasn
you can answer my r iddle â .
â she said . Tim grabbed the
someone without first asking their permission . We should ask the
nothing seemed quite right . Then one day

INTERVAL 2.893 - 6.296
CONTAINS 7.493%

He looked up and saw a kind and generous deer named
you were so fast !" Billy replied " I wasn 't
it , Daddy ?" Her dad replied , " I think
the plants , she saw a butterfly and started chasing it
. L ily said , " I can

INTERVAL -0.510 - 2.893
CONTAINS 51.800%

said , " This is my weapon . I can use
. Sarah was upset with Mary 's decision
house . Lily saw the dog and screamed too . She
go . Soon , the wizard returned and said , "
and dad . He always did what he wanted to do

INTERVAL -3.912 - -0.510
CONTAINS 39.046%

also some furry animals drinking from the pond . They looked
it . She liked the boat and took it home with
, I â m
them ?" Mia asked . " No , Mia
other and point at things . " Look ,

INTERVAL -7.315 - -3.912
CONTAINS 1.661%

, but still beautiful . From then on
as part of their game . <|endoftext|> Once upon a time
much as he wanted . The pig quickly
way inside his new created tent . It was the perfect
returned to her own hive . The bee had
Token was
Token ID373
Feature activation+1.15e+00

Pos logit contributions
 remembered+0.082
 forgot+0.082
 sne+0.076
 dove+0.073
 chased+0.072
Neg logit contributions
".-0.139
.-0.138
'.-0.133
 This-0.132
.'-0.130
Token so
Token ID523
Feature activation+1.89e+00

Pos logit contributions
 forgot+0.038
 remembered+0.037
 blew+0.035
 sne+0.034
 wish+0.033
Neg logit contributions
 This-0.056
 If-0.055
'.-0.055
".-0.055
.'-0.055
Token amazing
Token ID4998
Feature activation-2.67e+00

Pos logit contributions
 remembered+0.049
 forgot+0.049
 blew+0.042
 sne+0.042
 chased+0.042
Neg logit contributions
 This-0.103
".-0.102
 If-0.100
'.-0.098
.'-0.098
Token!
Token ID0
Feature activation-2.35e+00

Pos logit contributions
 If+0.112
and+0.108
 Today+0.104
umbs+0.098
 This+0.096
Neg logit contributions
 sw-0.109
 dove-0.104
 fl-0.104
 blew-0.104
 ate-0.103
Token Everywhere
Token ID48201
Feature activation+1.35e+00

Pos logit contributions
fulness+0.093
and+0.089
ings+0.089
arling+0.086
ancing+0.086
Neg logit contributions
 saw-0.092
 looked-0.088
 gazed-0.087
 dove-0.086
 observed-0.085
Token it
Token ID340
Feature activation+6.30e+00

Pos logit contributions
 forgot+0.069
 remembered+0.068
 blew+0.065
 sne+0.064
 wish+0.063
Neg logit contributions
 This-0.041
".-0.040
'.-0.039
 If-0.039
.'-0.039
Token sang
Token ID25889
Feature activation-1.09e+00

Pos logit contributions
inge+0.210
ige+0.208
Ta+0.207
 punished+0.202
 boo+0.198
Neg logit contributions
.-0.263
-0.247
."-0.213
 whenever-0.210
 She-0.209
Token,
Token ID11
Feature activation+1.34e-01

Pos logit contributions
.'+0.042
'.+0.042
 If+0.042
uts+0.041
 This+0.041
Neg logit contributions
 remembered-0.046
 saw-0.045
 sw-0.045
 buzz-0.044
 knew-0.044
Token it
Token ID340
Feature activation+3.33e+00

Pos logit contributions
 forgot+0.006
 remembered+0.006
 blew+0.006
 sne+0.006
 dove+0.006
Neg logit contributions
'.-0.005
.'-0.005
 If-0.005
 This-0.005
".-0.005
Token made
Token ID925
Feature activation+3.07e-01

Pos logit contributions
 forgot+0.065
 remembered+0.065
 wish+0.063
Ta+0.063
 boo+0.060
Neg logit contributions
.-0.184
".-0.179
-0.174
."-0.173
 This-0.171
Token people
Token ID661
Feature activation+1.59e+00

Pos logit contributions
 forgot+0.013
 sne+0.013
 remembered+0.013
 blew+0.013
 wish+0.012
Neg logit contributions
'.-0.012
.'-0.012
 This-0.012
 If-0.011
".-0.011
Token around
Token ID1088
Feature activation-1.60e+00

Pos logit contributions
and+0.027
 This+0.027
 If+0.027
'.+0.027
uts+0.027
Neg logit contributions
 buzz-0.028
 sw-0.028
 remembered-0.027
 sne-0.027
 flew-0.027
Token the
Token ID262
Feature activation+1.90e+00

Pos logit contributions
uts+0.092
and+0.082
 If+0.082
umbs+0.077
but+0.077
Neg logit contributions
 sw-0.052
 sne-0.050
 buzz-0.050
 saw-0.050
 ch-0.047
Token world
Token ID995
Feature activation-2.05e+00

Pos logit contributions
 forgot+0.072
 remembered+0.071
 blew+0.066
 chased+0.063
 ate+0.063
Neg logit contributions
".-0.084
 This-0.083
.'-0.083
'.-0.082
and-0.081
Token.
Token ID13
Feature activation-1.71e+00

Pos logit contributions
 If+0.064
 This+0.058
'.+0.057
.'+0.054
 Today+0.054
Neg logit contributions
 bit-0.103
 buzz-0.102
 saw-0.101
 forgot-0.100
 ate-0.100
Token Everywhere
Token ID48201
Feature activation+7.60e-01

Pos logit contributions
ings+0.061
 anymore+0.059
.'+0.059
fulness+0.058
rier+0.055
Neg logit contributions
 saw-0.070
 buzz-0.069
 ate-0.069
 tried-0.068
 sang-0.067
Token it
Token ID340
Feature activation+6.26e+00

Pos logit contributions
 forgot+0.038
 remembered+0.038
 blew+0.036
 sne+0.035
 dove+0.035
Neg logit contributions
and-0.025
 This-0.024
 If-0.024
'.-0.024
.'-0.024
Token went
Token ID1816
Feature activation-8.61e-01

Pos logit contributions
 boo+0.195
 charged+0.190
inge+0.185
 scrape+0.183
Ta+0.183
Neg logit contributions
.-0.275
-0.264
 She-0.242
 before-0.234
."-0.229
Token,
Token ID11
Feature activation-2.82e-01

Pos logit contributions
and+0.035
uts+0.033
.'+0.033
ula+0.033
 If+0.033
Neg logit contributions
 saw-0.038
 sw-0.036
 flew-0.035
 remembered-0.035
 dove-0.035
Token it
Token ID340
Feature activation+3.89e+00

Pos logit contributions
'.+0.012
.'+0.012
 If+0.012
 This+0.012
and+0.011
Neg logit contributions
 forgot-0.012
 sne-0.011
 remembered-0.011
 dove-0.011
 blew-0.011
Token made
Token ID925
Feature activation+2.29e-01

Pos logit contributions
 wish+0.081
 boo+0.078
Ta+0.077
 tug+0.075
 remembered+0.075
Neg logit contributions
.-0.210
".-0.203
-0.200
."-0.196
.'-0.194
Token people
Token ID661
Feature activation+1.27e+00

Pos logit contributions
 sne+0.009
 forgot+0.009
 remembered+0.009
 wish+0.009
 buzz+0.009
Neg logit contributions
'.-0.009
.'-0.009
 This-0.009
 If-0.009
and-0.009
Token."
Token ID526
Feature activation-1.34e+00

Pos logit contributions
and+0.015
 If+0.013
'.+0.013
.'+0.013
 This+0.013
Neg logit contributions
 bit-0.036
 forgot-0.035
 ate-0.034
 blew-0.034
 sne-0.034
Token
Token ID198
Feature activation-1.50e+00

Pos logit contributions
fulness+0.059
.'+0.059
rier+0.058
'.+0.056
".+0.054
Neg logit contributions
 ate-0.054
 played-0.053
 saw-0.051
 threw-0.051
 forgot-0.050
Token
Token ID198
Feature activation+1.46e+00

Pos logit contributions
.'+0.036
'.+0.036
rier+0.033
".+0.033
overs+0.032
Neg logit contributions
 saw-0.082
 forgot-0.079
 remembered-0.078
 knew-0.077
 ate-0.077
TokenTim
Token ID14967
Feature activation+3.84e+00

Pos logit contributions
 forgot+0.070
 remembered+0.069
 wish+0.066
 sne+0.065
 blew+0.064
Neg logit contributions
 This-0.051
 If-0.049
".-0.049
and-0.049
'.-0.048
Token and
Token ID290
Feature activation+1.47e+00

Pos logit contributions
 wish+0.150
Uh+0.149
 flood+0.144
 groom+0.142
inge+0.141
Neg logit contributions
.-0.155
-0.149
 She-0.139
 They-0.134
 This-0.131
Token Lily
Token ID20037
Feature activation+6.26e+00

Pos logit contributions
 forgot+0.091
 remembered+0.089
 wish+0.086
 blew+0.086
 sne+0.085
Neg logit contributions
 This-0.030
 If-0.029
".-0.026
and-0.026
'.-0.026
Token looked
Token ID3114
Feature activation+3.15e+00

Pos logit contributions
 crow+0.244
 cruel+0.244
 painful+0.242
 annoying+0.241
 eagle+0.237
Neg logit contributions
.-0.252
 They-0.241
 She-0.236
-0.220
 So-0.201
Token around
Token ID1088
Feature activation-4.14e-01

Pos logit contributions
 forgot+0.206
 blew+0.203
Ta+0.200
 groom+0.198
 rocked+0.194
Neg logit contributions
.-0.056
 anymore-0.042
 together-0.042
 about-0.040
 If-0.038
Token the
Token ID262
Feature activation+1.85e+00

Pos logit contributions
uts+0.017
".+0.016
'.+0.016
 If+0.016
and+0.016
Neg logit contributions
 saw-0.019
 remembered-0.018
 sne-0.018
 forgot-0.017
 sw-0.017
Token movie
Token ID3807
Feature activation-1.04e+00

Pos logit contributions
 forgot+0.073
 remembered+0.072
 blew+0.065
 chased+0.063
 dropped+0.062
Neg logit contributions
 This-0.085
and-0.080
 If-0.079
".-0.079
'.-0.077
Token place
Token ID1295
Feature activation-1.73e+00

Pos logit contributions
'.+0.038
.'+0.038
 If+0.037
".+0.037
 This+0.034
Neg logit contributions
 forgot-0.048
 saw-0.047
 remembered-0.046
 ate-0.046
 bit-0.045
Token near
Token ID1474
Feature activation+3.25e-01

Pos logit contributions
 forgot+0.036
 remembered+0.036
 blew+0.034
 sne+0.034
 ate+0.033
Neg logit contributions
'.-0.031
 This-0.031
 If-0.031
.'-0.030
".-0.030
Token the
Token ID262
Feature activation+1.43e+00

Pos logit contributions
 remembered+0.013
 sne+0.013
 forgot+0.013
 wish+0.013
 sw+0.013
Neg logit contributions
'.-0.013
 This-0.013
.'-0.013
".-0.012
 If-0.012
Token clouds
Token ID15114
Feature activation-7.13e-01

Pos logit contributions
 forgot+0.054
 remembered+0.052
 blew+0.048
 sne+0.047
 ate+0.046
Neg logit contributions
 This-0.064
".-0.064
'.-0.063
 If-0.063
.'-0.063
Token.
Token ID13
Feature activation-1.74e+00

Pos logit contributions
 If+0.023
 This+0.022
'.+0.021
and+0.021
.'+0.020
Neg logit contributions
 blew-0.036
 ate-0.036
 remembered-0.035
 sw-0.035
 sne-0.035
Token Everywhere
Token ID48201
Feature activation+1.39e+00

Pos logit contributions
keeping+0.073
ula+0.069
ings+0.069
uts+0.068
and+0.068
Neg logit contributions
 saw-0.067
 observed-0.064
 dove-0.063
 ate-0.062
 looked-0.062
Token it
Token ID340
Feature activation+6.24e+00

Pos logit contributions
 forgot+0.073
 remembered+0.072
 blew+0.068
 sne+0.068
 wish+0.067
Neg logit contributions
 This-0.040
".-0.039
'.-0.039
 If-0.039
.'-0.038
Token flew
Token ID13112
Feature activation+5.50e-02

Pos logit contributions
Ta+0.210
inge+0.205
 punished+0.202
 boo+0.202
ige+0.202
Neg logit contributions
.-0.262
-0.255
."-0.217
 whenever-0.212
 to-0.211
Token,
Token ID11
Feature activation+4.28e-01

Pos logit contributions
 forgot+0.003
 remembered+0.003
 saw+0.003
 blew+0.003
 sw+0.003
Neg logit contributions
and-0.002
'.-0.002
.'-0.002
 If-0.002
 This-0.002
Token it
Token ID340
Feature activation+4.29e+00

Pos logit contributions
 forgot+0.020
 remembered+0.019
 blew+0.019
 sne+0.019
 dove+0.018
Neg logit contributions
'.-0.016
.'-0.015
 This-0.015
 If-0.015
and-0.015
Token left
Token ID1364
Feature activation-1.08e+00

Pos logit contributions
Ta+0.107
 wish+0.106
 boo+0.100
 spit+0.099
Uh+0.097
Neg logit contributions
.-0.217
-0.211
."-0.197
".-0.197
and-0.189
Token a
Token ID257
Feature activation+3.11e+00

Pos logit contributions
'.+0.045
 This+0.044
uts+0.044
.'+0.044
and+0.043
Neg logit contributions
 remembered-0.045
 sw-0.045
 saw-0.045
 knew-0.042
 forgot-0.042
Token what
Token ID644
Feature activation-2.30e-02

Pos logit contributions
'.+0.050
 This+0.050
.'+0.049
and+0.047
 Today+0.047
Neg logit contributions
 remembered-0.044
 pin-0.043
 forgot-0.043
 blew-0.042
 sne-0.042
Token it
Token ID340
Feature activation+2.50e+00

Pos logit contributions
 If+0.001
.'+0.001
'.+0.001
and+0.001
".+0.001
Neg logit contributions
 remembered-0.001
 forgot-0.001
 blew-0.001
 pin-0.001
 saw-0.001
Token was
Token ID373
Feature activation-2.41e-02

Pos logit contributions
 remembered+0.091
 forgot+0.090
 wish+0.088
 groom+0.087
 dove+0.087
Neg logit contributions
 This-0.106
.-0.101
-0.100
".-0.097
 If-0.097
Token,
Token ID11
Feature activation-2.52e-01

Pos logit contributions
 If+0.001
and+0.001
 This+0.001
'.+0.001
.'+0.001
Neg logit contributions
 forgot-0.001
 remembered-0.001
 blew-0.001
 sne-0.001
 dropped-0.001
Token so
Token ID523
Feature activation+2.04e+00

Pos logit contributions
'.+0.009
.'+0.008
 This+0.008
".+0.008
and+0.008
Neg logit contributions
 forgot-0.013
 remembered-0.012
 sne-0.012
 wished-0.012
 bit-0.012
Token she
Token ID673
Feature activation+6.18e+00

Pos logit contributions
 forgot+0.126
 blew+0.123
 remembered+0.123
 sne+0.122
 buzz+0.120
Neg logit contributions
 This-0.033
.-0.030
 She-0.029
 If-0.029
-0.029
Token asked
Token ID1965
Feature activation-3.33e-02

Pos logit contributions
 thrill+0.260
 merry+0.249
 happiest+0.236
 flood+0.234
ited+0.234
Neg logit contributions
.-0.204
-0.197
 to-0.175
 They-0.174
 She-0.172
Token her
Token ID607
Feature activation+1.88e+00

Pos logit contributions
'.+0.001
.'+0.001
and+0.001
".+0.001
uts+0.001
Neg logit contributions
 remembered-0.001
 forgot-0.001
 dropped-0.001
 saw-0.001
 sne-0.001
Token mom
Token ID1995
Feature activation-1.88e+00

Pos logit contributions
 forgot+0.086
 remembered+0.081
 blew+0.081
 rocked+0.080
 chased+0.078
Neg logit contributions
 This-0.067
 If-0.065
.-0.065
".-0.065
-0.064
Token.
Token ID13
Feature activation-1.12e+00

Pos logit contributions
uts+0.109
'.+0.098
 This+0.097
keeping+0.096
.'+0.095
Neg logit contributions
 saw-0.051
 remembered-0.051
 cut-0.047
 forgot-0.043
 knew-0.042
Token
Token ID198
Feature activation-1.97e+00

Pos logit contributions
and+0.029
.'+0.027
".+0.026
fulness+0.026
ions+0.026
Neg logit contributions
 saw-0.062
 remembered-0.061
 blew-0.060
 forgot-0.060
 dropped-0.058
Token happy
Token ID3772
Feature activation-8.98e-01

Pos logit contributions
 forgot+0.069
 remembered+0.069
 sne+0.061
 bit+0.059
 regretted+0.059
Neg logit contributions
 This-0.108
-0.108
 If-0.106
".-0.105
and-0.104
Token and
Token ID290
Feature activation+3.06e+00

Pos logit contributions
 If+0.037
'.+0.034
.'+0.033
 This+0.033
and+0.031
Neg logit contributions
 saw-0.039
 forgot-0.039
 remembered-0.039
 dove-0.038
 ate-0.037
Token free
Token ID1479
Feature activation-1.84e+00

Pos logit contributions
 wish+0.074
 sne+0.070
 sadd+0.070
ited+0.070
 remembered+0.069
Neg logit contributions
".-0.156
 This-0.155
-0.155
.-0.154
 She-0.152
Token,
Token ID11
Feature activation-3.12e-01

Pos logit contributions
 If+0.063
 Today+0.060
and+0.059
'.+0.057
 This+0.057
Neg logit contributions
 sw-0.090
 flew-0.088
 buzz-0.087
 chased-0.087
 bit-0.086
Token and
Token ID290
Feature activation+2.08e+00

Pos logit contributions
'.+0.012
.'+0.011
and+0.011
 If+0.011
 This+0.011
Neg logit contributions
 forgot-0.015
 remembered-0.015
 blew-0.014
 sne-0.014
 ate-0.014
Token it
Token ID340
Feature activation+6.15e+00

Pos logit contributions
 remembered+0.120
 forgot+0.120
 wish+0.119
 sne+0.117
 blew+0.116
Neg logit contributions
 This-0.039
 If-0.038
".-0.037
and-0.033
 She-0.033
Token made
Token ID925
Feature activation+1.03e+00

Pos logit contributions
Ta+0.205
affle+0.194
wy+0.192
prise+0.186
 promise+0.183
Neg logit contributions
-0.266
.-0.266
 with-0.231
 "-0.228
."-0.224
Token Lily
Token ID20037
Feature activation+1.78e+00

Pos logit contributions
 forgot+0.055
 remembered+0.054
 blew+0.051
 sne+0.051
 wish+0.050
Neg logit contributions
 This-0.030
'.-0.029
 If-0.029
".-0.029
.'-0.029
Token smile
Token ID8212
Feature activation-6.45e-01

Pos logit contributions
 forgot+0.049
 remembered+0.048
 blew+0.044
 sne+0.040
 dove+0.040
Neg logit contributions
 This-0.095
".-0.094
and-0.094
'.-0.093
 If-0.093
Token.
Token ID13
Feature activation-1.36e+00

Pos logit contributions
 If+0.015
and+0.014
 This+0.014
ion+0.013
'.+0.013
Neg logit contributions
 remembered-0.040
 forgot-0.038
 sne-0.038
 ate-0.037
 blew-0.037
Token
Token ID198
Feature activation-1.28e+00

Pos logit contributions
ions+0.041
fulness+0.040
.'+0.039
'.+0.038
and+0.037
Neg logit contributions
 forgot-0.072
 saw-0.071
 remembered-0.067
 ate-0.067
 felt-0.066
Token feel
Token ID1254
Feature activation+2.04e+00

Pos logit contributions
 forgot+0.052
 remembered+0.051
 blew+0.047
 dove+0.045
 chased+0.045
Neg logit contributions
 This-0.079
".-0.078
'.-0.077
 If-0.077
.'-0.077
Token better
Token ID1365
Feature activation-1.52e+00

Pos logit contributions
 forgot+0.080
 remembered+0.078
 blew+0.074
 dove+0.073
 chased+0.072
Neg logit contributions
".-0.084
 If-0.083
'.-0.083
.'-0.083
 This-0.081
Token.
Token ID13
Feature activation-8.17e-01

Pos logit contributions
 This+0.050
 If+0.049
ions+0.045
 Everywhere+0.045
anes+0.045
Neg logit contributions
 forgot-0.084
 remembered-0.082
 bit-0.081
 hurt-0.078
 wished-0.076
Token Lily
Token ID20037
Feature activation+5.42e+00

Pos logit contributions
and+0.024
.'+0.024
".+0.023
'.+0.023
fulness+0.023
Neg logit contributions
 forgot-0.042
 remembered-0.041
 saw-0.040
 ate-0.040
 blew-0.039
Token felt
Token ID2936
Feature activation+3.33e+00

Pos logit contributions
 flood+0.226
Uh+0.222
Ta+0.214
 spe+0.209
 holy+0.209
Neg logit contributions
-0.227
.-0.211
 She-0.188
 and-0.174
 So-0.173
Token a
Token ID257
Feature activation+6.13e+00

Pos logit contributions
 forgot+0.142
 blew+0.139
 dove+0.135
 remembered+0.134
 buzz+0.133
Neg logit contributions
-0.127
 If-0.127
 This-0.122
.-0.122
'.-0.116
Token little
Token ID1310
Feature activation+4.81e+00

Pos logit contributions
 met+0.233
 swung+0.231
 chased+0.231
 forgot+0.227
 remembered+0.222
Neg logit contributions
-0.236
.-0.235
".-0.216
'.-0.214
.'-0.211
Token bit
Token ID1643
Feature activation+2.82e+00

Pos logit contributions
lly+0.235
 forgot+0.227
ep+0.223
 swung+0.215
 blew+0.215
Neg logit contributions
.-0.162
 If-0.150
-0.149
 anymore-0.145
 She-0.141
Token better
Token ID1365
Feature activation-1.22e-01

Pos logit contributions
 forgot+0.111
 remembered+0.108
 blew+0.105
 dove+0.102
 chased+0.101
Neg logit contributions
 If-0.120
.-0.114
.'-0.111
 This-0.110
and-0.110
Token but
Token ID475
Feature activation+3.24e+00

Pos logit contributions
 This+0.004
 If+0.004
and+0.004
'.+0.004
ions+0.003
Neg logit contributions
 remembered-0.007
 forgot-0.006
 sne-0.006
 bit-0.006
 blew-0.006
Token still
Token ID991
Feature activation+1.50e+00

Pos logit contributions
 swoop+0.128
 buzz+0.125
 sne+0.124
 forgot+0.123
 bit+0.120
Neg logit contributions
-0.120
.-0.116
 She-0.114
and-0.112
 This-0.110
Token started
Token ID2067
Feature activation+2.75e-01

Pos logit contributions
 forgot+0.055
 remembered+0.052
 wish+0.050
 buzz+0.046
 sne+0.046
Neg logit contributions
 This-0.140
".-0.137
'.-0.136
 If-0.133
.-0.133
Token to
Token ID284
Feature activation+2.41e+00

Pos logit contributions
 forgot+0.016
 remembered+0.016
 sne+0.016
 blew+0.015
 bit+0.015
Neg logit contributions
'.-0.007
 If-0.007
.'-0.007
 This-0.007
".-0.007
Token spread
Token ID4104
Feature activation+7.33e-02

Pos logit contributions
 remembered+0.065
 forgot+0.065
 regretted+0.052
 ate+0.051
 dove+0.051
Neg logit contributions
 This-0.130
".-0.127
'.-0.127
.'-0.126
 If-0.125
Token.
Token ID13
Feature activation-1.69e+00

Pos logit contributions
 remembered+0.003
 forgot+0.003
 sne+0.003
 blew+0.003
 sw+0.003
Neg logit contributions
 This-0.003
and-0.003
'.-0.003
 If-0.003
uts-0.003
Token Everywhere
Token ID48201
Feature activation+1.32e+00

Pos logit contributions
fulness+0.065
and+0.064
.'+0.062
ently+0.062
ings+0.061
Neg logit contributions
 saw-0.069
 cut-0.067
 rubbed-0.067
 looked-0.067
 ate-0.067
Token it
Token ID340
Feature activation+6.10e+00

Pos logit contributions
 forgot+0.071
 remembered+0.070
 blew+0.067
 sne+0.066
 wish+0.065
Neg logit contributions
 This-0.037
'.-0.036
".-0.036
 If-0.036
.'-0.035
Token went
Token ID1816
Feature activation-2.53e-01

Pos logit contributions
ige+0.205
Ta+0.205
 traitor+0.202
inge+0.196
ited+0.195
Neg logit contributions
.-0.234
-0.217
 to-0.190
 They-0.183
 with-0.181
Token,
Token ID11
Feature activation-2.34e-01

Pos logit contributions
and+0.009
.'+0.009
'.+0.009
 If+0.008
uts+0.008
Neg logit contributions
 remembered-0.012
 saw-0.012
 forgot-0.012
 blew-0.012
 sw-0.012
Token something
Token ID1223
Feature activation+3.71e+00

Pos logit contributions
'.+0.009
and+0.009
.'+0.009
 If+0.009
 This+0.009
Neg logit contributions
 forgot-0.010
 blew-0.010
 sne-0.010
 remembered-0.010
 dove-0.010
Token happened
Token ID3022
Feature activation-2.60e+00

Pos logit contributions
 remembered+0.118
ited+0.114
 forgot+0.111
Ta+0.108
 wish+0.105
Neg logit contributions
.-0.173
and-0.158
-0.156
 This-0.153
".-0.150
Token.
Token ID13
Feature activation-2.26e+00

Pos logit contributions
 If+0.078
 This+0.077
umbs+0.073
ugs+0.073
 These+0.073
Neg logit contributions
 remembered-0.141
 saw-0.136
 lost-0.134
 fl-0.132
 sw-0.130
Token to
Token ID284
Feature activation+2.83e+00

Pos logit contributions
 forgot+0.053
 remembered+0.053
 blew+0.050
 sne+0.050
 ate+0.050
Neg logit contributions
'.-0.017
 This-0.017
.'-0.016
 If-0.016
".-0.016
Token be
Token ID307
Feature activation+2.40e+00

Pos logit contributions
 remembered+0.102
 forgot+0.102
 regretted+0.094
 blew+0.091
 wished+0.088
Neg logit contributions
".-0.134
 This-0.130
'.-0.128
.'-0.126
.-0.124
Token beautiful
Token ID4950
Feature activation-8.33e-03

Pos logit contributions
 forgot+0.091
 blew+0.086
 remembered+0.084
 regretted+0.080
 rocked+0.079
Neg logit contributions
".-0.103
.-0.099
.'-0.097
 If-0.096
."-0.096
Token,
Token ID11
Feature activation+1.99e-01

Pos logit contributions
 This+0.000
 If+0.000
'.+0.000
and+0.000
.'+0.000
Neg logit contributions
 remembered-0.000
 forgot-0.000
 sne-0.000
 wish-0.000
 bit-0.000
Token so
Token ID523
Feature activation+5.91e-01

Pos logit contributions
 forgot+0.010
 sne+0.009
 remembered+0.009
 bit+0.009
 dove+0.009
Neg logit contributions
'.-0.007
.'-0.007
 This-0.007
and-0.007
".-0.007
Token it
Token ID340
Feature activation+6.06e+00

Pos logit contributions
 forgot+0.031
 remembered+0.030
 blew+0.029
 sne+0.029
 dove+0.028
Neg logit contributions
'.-0.019
 This-0.019
.'-0.018
and-0.018
 If-0.018
Token tried
Token ID3088
Feature activation-1.26e-01

Pos logit contributions
Ta+0.170
 UFO+0.160
inge+0.159
 quarrel+0.155
 argument+0.152
Neg logit contributions
.-0.267
-0.248
."-0.233
 to-0.225
 She-0.218
Token to
Token ID284
Feature activation+2.67e+00

Pos logit contributions
'.+0.004
 This+0.004
 If+0.004
.'+0.004
".+0.003
Neg logit contributions
 remembered-0.007
 forgot-0.007
 sne-0.006
 ate-0.006
 bit-0.006
Token make
Token ID787
Feature activation+5.71e-01

Pos logit contributions
 forgot+0.084
 remembered+0.082
 regretted+0.077
 wished+0.069
 blew+0.069
Neg logit contributions
".-0.133
.'-0.129
'.-0.128
.-0.127
 If-0.125
Token itself
Token ID2346
Feature activation+1.48e+00

Pos logit contributions
 forgot+0.026
 remembered+0.026
 sne+0.025
 wish+0.025
 blew+0.025
Neg logit contributions
 This-0.020
'.-0.020
and-0.020
 If-0.020
.'-0.019
Token very
Token ID845
Feature activation+2.69e+00

Pos logit contributions
 forgot+0.044
 remembered+0.042
 blew+0.039
 sne+0.036
 chased+0.036
Neg logit contributions
 This-0.078
".-0.077
'.-0.076
.'-0.076
 If-0.076
Token loved
Token ID6151
Feature activation+1.35e+00

Pos logit contributions
 forgot+0.154
 groom+0.150
 remembered+0.148
ited+0.148
 punished+0.147
Neg logit contributions
-0.145
.-0.143
 This-0.130
".-0.126
."-0.124
Token to
Token ID284
Feature activation+2.49e+00

Pos logit contributions
 forgot+0.091
 remembered+0.090
 blew+0.086
 wish+0.085
 sne+0.084
Neg logit contributions
 This-0.020
 If-0.019
".-0.019
and-0.019
'.-0.018
Token sing
Token ID1702
Feature activation-6.34e-01

Pos logit contributions
 forgot+0.075
 remembered+0.071
 regretted+0.064
 wished+0.058
 blew+0.057
Neg logit contributions
 This-0.136
".-0.133
.-0.128
'.-0.128
.'-0.127
Token.
Token ID13
Feature activation-1.54e+00

Pos logit contributions
 If+0.021
 This+0.021
and+0.020
'.+0.019
.'+0.019
Neg logit contributions
 remembered-0.029
 buzz-0.029
 sne-0.029
 fl-0.029
 ch-0.028
Token Everywhere
Token ID48201
Feature activation+1.26e+00

Pos logit contributions
ula+0.058
arling+0.058
and+0.057
keeping+0.056
uts+0.054
Neg logit contributions
 cut-0.063
 sne-0.061
 saw-0.061
 wish-0.060
 rubbed-0.059
Token it
Token ID340
Feature activation+6.06e+00

Pos logit contributions
 forgot+0.066
 remembered+0.065
 blew+0.062
 sne+0.061
 wish+0.061
Neg logit contributions
 This-0.037
'.-0.036
".-0.035
 If-0.035
.'-0.035
Token went
Token ID1816
Feature activation+2.36e-02

Pos logit contributions
Ta+0.197
inge+0.191
ige+0.187
 stolen+0.185
 punished+0.182
Neg logit contributions
.-0.264
-0.251
."-0.226
 whenever-0.211
 She-0.211
Token,
Token ID11
Feature activation+1.57e-01

Pos logit contributions
 forgot+0.001
 remembered+0.001
 saw+0.001
 blew+0.001
 sw+0.001
Neg logit contributions
and-0.001
'.-0.001
.'-0.001
 If-0.001
".-0.001
Token it
Token ID340
Feature activation+3.89e+00

Pos logit contributions
 forgot+0.007
 remembered+0.007
 sne+0.007
 blew+0.007
 dove+0.007
Neg logit contributions
'.-0.006
.'-0.006
 If-0.006
 This-0.006
and-0.006
Token would
Token ID561
Feature activation+1.83e+00

Pos logit contributions
Ta+0.091
 wish+0.090
Uh+0.084
 tug+0.083
 spit+0.081
Neg logit contributions
.-0.203
-0.194
".-0.192
."-0.190
and-0.179
Token sing
Token ID1702
Feature activation-1.25e+00

Pos logit contributions
 forgot+0.043
 remembered+0.042
 blew+0.036
 ate+0.033
 dropped+0.033
Neg logit contributions
".-0.105
 This-0.105
 If-0.103
'.-0.103
.'-0.103
Token that
Token ID326
Feature activation+1.07e+00

Pos logit contributions
 This+0.007
'.+0.007
 If+0.007
and+0.007
.'+0.007
Neg logit contributions
 remembered-0.008
 forgot-0.008
 sne-0.007
 wish-0.007
 buzz-0.007
Token little
Token ID1310
Feature activation+1.15e+00

Pos logit contributions
 forgot+0.037
 remembered+0.036
 blew+0.033
 sne+0.032
 wish+0.032
Neg logit contributions
 This-0.051
 If-0.050
'.-0.050
".-0.050
and-0.049
Token seed
Token ID9403
Feature activation-5.18e-01

Pos logit contributions
 forgot+0.038
 remembered+0.037
 blew+0.035
 sne+0.034
 wish+0.033
Neg logit contributions
 This-0.056
 If-0.055
'.-0.055
".-0.055
.'-0.054
Token.
Token ID13
Feature activation-2.63e+00

Pos logit contributions
 If+0.023
 This+0.023
and+0.023
'.+0.023
.'+0.022
Neg logit contributions
 forgot-0.020
 remembered-0.020
 blew-0.019
 ate-0.018
 bit-0.018
Token Everywhere
Token ID48201
Feature activation+7.05e-01

Pos logit contributions
uts+0.080
 anymore+0.077
ings+0.076
.'+0.075
fulness+0.073
Neg logit contributions
 planted-0.102
 wish-0.099
 believed-0.099
 observed-0.097
 cut-0.097
Token it
Token ID340
Feature activation+6.04e+00

Pos logit contributions
 forgot+0.038
 remembered+0.038
 blew+0.036
 sne+0.035
 ate+0.035
Neg logit contributions
and-0.021
 If-0.020
 This-0.020
'.-0.020
.'-0.020
Token went
Token ID1816
Feature activation-2.24e-01

Pos logit contributions
Ta+0.187
 boo+0.185
Trump+0.167
 traitor+0.167
 wish+0.165
Neg logit contributions
.-0.256
-0.242
 whenever-0.222
 She-0.220
 together-0.216
Token,
Token ID11
Feature activation+1.11e-01

Pos logit contributions
and+0.008
 If+0.007
'.+0.007
.'+0.007
".+0.007
Neg logit contributions
 forgot-0.011
 remembered-0.011
 saw-0.011
 blew-0.010
 dove-0.010
Token it
Token ID340
Feature activation+3.84e+00

Pos logit contributions
 forgot+0.005
 remembered+0.005
 sne+0.004
 blew+0.004
 dove+0.004
Neg logit contributions
'.-0.005
 This-0.005
 If-0.004
.'-0.004
".-0.004
Token made
Token ID925
Feature activation+2.81e-01

Pos logit contributions
 wish+0.089
Ta+0.086
 boo+0.082
 remembered+0.081
 forgot+0.078
Neg logit contributions
.-0.197
".-0.189
-0.185
."-0.181
.'-0.180
Token flowers
Token ID12734
Feature activation+6.63e-01

Pos logit contributions
 forgot+0.012
 remembered+0.012
 sne+0.012
 wish+0.011
 buzz+0.011
Neg logit contributions
'.-0.011
 This-0.011
 If-0.011
.'-0.011
".-0.010
Token!"
Token ID2474
Feature activation-9.86e-01

Pos logit contributions
 This+0.028
 If+0.027
'.+0.027
ula+0.026
and+0.026
Neg logit contributions
 forgot-0.054
 remembered-0.052
 saw-0.051
 hurt-0.050
 ate-0.050
Token said
Token ID531
Feature activation+2.77e-01

Pos logit contributions
.'+0.027
'.+0.027
 anymore+0.025
fulness+0.024
".+0.024
Neg logit contributions
 forgot-0.054
 saw-0.053
 ate-0.053
 remembered-0.052
 felt-0.051
Token Lily
Token ID20037
Feature activation+1.07e+00

Pos logit contributions
 remembered+0.014
 forgot+0.014
 sne+0.014
 blew+0.013
 ate+0.013
Neg logit contributions
and-0.009
.'-0.008
 If-0.008
'.-0.008
".-0.008
Token generously
Token ID42088
Feature activation+3.64e-01

Pos logit contributions
 forgot+0.060
 remembered+0.059
 blew+0.056
 sne+0.055
 wish+0.054
Neg logit contributions
 This-0.028
 If-0.027
'.-0.027
".-0.027
.'-0.027
Token.
Token ID13
Feature activation-2.24e+00

Pos logit contributions
 remembered+0.023
 forgot+0.023
 blew+0.021
 sne+0.021
 ate+0.021
Neg logit contributions
 This-0.007
 If-0.007
and-0.007
'.-0.007
.'-0.007
Token They
Token ID1119
Feature activation+6.02e+00

Pos logit contributions
fulness+0.080
.'+0.077
'.+0.072
ion+0.072
and+0.069
Neg logit contributions
 saw-0.107
 forgot-0.105
 ate-0.105
 played-0.103
 threw-0.102
Token looked
Token ID3114
Feature activation+2.79e+00

Pos logit contributions
 crow+0.241
 eagle+0.237
 bully+0.237
 raven+0.228
 joke+0.226
Neg logit contributions
.-0.239
 They-0.230
-0.224
 together-0.221
 She-0.217
Token everywhere
Token ID8347
Feature activation-6.56e-01

Pos logit contributions
 forgot+0.167
 blew+0.166
 remembered+0.161
 sne+0.161
 spit+0.159
Neg logit contributions
.-0.056
 This-0.055
 If-0.054
 anymore-0.050
".-0.049
Token but
Token ID475
Feature activation+1.22e+00

Pos logit contributions
'.+0.022
 This+0.021
 If+0.021
.'+0.021
".+0.020
Neg logit contributions
 forgot-0.034
 remembered-0.033
 saw-0.032
 knew-0.031
 dropped-0.031
Token couldn
Token ID3521
Feature activation-4.76e-01

Pos logit contributions
 forgot+0.050
 remembered+0.049
 blew+0.046
 sne+0.046
 wish+0.045
Neg logit contributions
 This-0.050
 If-0.049
'.-0.049
".-0.048
and-0.048
Token't
Token ID470
Feature activation+1.02e+00

Pos logit contributions
and+0.017
.'+0.016
 If+0.015
'.+0.015
".+0.014
Neg logit contributions
 forgot-0.022
 blew-0.022
 remembered-0.022
 sne-0.021
 dropped-0.021
Token a
Token ID257
Feature activation+4.94e+00

Pos logit contributions
 forgot+0.069
 remembered+0.067
 blew+0.063
 sne+0.062
 wish+0.062
Neg logit contributions
 This-0.045
 If-0.044
'.-0.043
".-0.043
and-0.042
Token gentle
Token ID10296
Feature activation+1.20e+00

Pos logit contributions
 forgot+0.215
 struggled+0.213
ited+0.208
 soared+0.207
 rocked+0.207
Neg logit contributions
-0.175
.-0.173
 "-0.164
".-0.150
."-0.147
Token lung
Token ID12317
Feature activation-2.07e-01

Pos logit contributions
 forgot+0.031
 remembered+0.030
 blew+0.027
 sne+0.026
 wish+0.026
Neg logit contributions
 This-0.067
'.-0.066
 If-0.066
".-0.066
.'-0.065
Token.
Token ID13
Feature activation-1.24e+00

Pos logit contributions
 If+0.005
and+0.005
 This+0.005
'.+0.005
.'+0.004
Neg logit contributions
 forgot-0.012
 blew-0.012
 remembered-0.012
 bit-0.012
 sne-0.012
Token Everywhere
Token ID48201
Feature activation+1.18e+00

Pos logit contributions
and+0.042
but+0.036
.'+0.036
arling+0.034
ula+0.034
Neg logit contributions
 ate-0.059
 sne-0.059
 cut-0.058
 blew-0.057
 wish-0.057
Token it
Token ID340
Feature activation+6.01e+00

Pos logit contributions
 forgot+0.062
 remembered+0.061
 blew+0.058
 sne+0.057
 wish+0.057
Neg logit contributions
 This-0.034
'.-0.033
 If-0.033
".-0.033
.'-0.033
Token went
Token ID1816
Feature activation-3.03e-01

Pos logit contributions
Ta+0.214
ige+0.203
 argument+0.193
Trump+0.186
ited+0.184
Neg logit contributions
.-0.250
-0.238
 He-0.214
."-0.214
".-0.195
Token,
Token ID11
Feature activation-2.47e-01

Pos logit contributions
and+0.011
.'+0.010
'.+0.010
 If+0.009
ula+0.009
Neg logit contributions
 remembered-0.014
 blew-0.014
 sw-0.014
 forgot-0.014
 saw-0.014
Token it
Token ID340
Feature activation+4.22e+00

Pos logit contributions
and+0.009
'.+0.009
.'+0.009
 If+0.009
 This+0.009
Neg logit contributions
 forgot-0.011
 remembered-0.011
 sne-0.011
 blew-0.011
 dove-0.010
Token made
Token ID925
Feature activation+6.52e-01

Pos logit contributions
Ta+0.116
Uh+0.107
 wish+0.104
 forgot+0.096
 spit+0.095
Neg logit contributions
.-0.209
-0.200
."-0.191
".-0.191
 This-0.180
Token its
Token ID663
Feature activation+2.40e+00

Pos logit contributions
 forgot+0.028
 remembered+0.028
 sne+0.027
 blew+0.027
 wish+0.027
Neg logit contributions
'.-0.025
 This-0.025
.'-0.024
and-0.024
 If-0.024
Token was
Token ID373
Feature activation+3.61e+00

Pos logit contributions
 groom+0.154
Ta+0.142
 wish+0.136
 wave+0.127
ited+0.127
Neg logit contributions
.-0.173
-0.160
".-0.149
 She-0.148
."-0.147
Token not
Token ID407
Feature activation+1.44e+00

Pos logit contributions
 forgot+0.156
ited+0.153
Ta+0.142
 remembered+0.142
 rocked+0.140
Neg logit contributions
 If-0.135
.-0.133
 This-0.131
-0.127
 about-0.123
Token hurt
Token ID5938
Feature activation-1.29e+00

Pos logit contributions
 forgot+0.042
 remembered+0.040
 blew+0.037
 wish+0.036
 sne+0.036
Neg logit contributions
".-0.075
 This-0.075
 If-0.074
and-0.074
'.-0.073
Token,
Token ID11
Feature activation-5.06e-01

Pos logit contributions
'.+0.048
 This+0.048
.'+0.046
'."+0.045
.'"+0.044
Neg logit contributions
 hurt-0.060
 lost-0.059
 bit-0.056
 spoiled-0.056
 sne-0.055
Token but
Token ID475
Feature activation+1.60e+00

Pos logit contributions
'.+0.017
.'+0.016
and+0.016
 This+0.016
".+0.016
Neg logit contributions
 forgot-0.027
 remembered-0.026
 wished-0.025
 sne-0.025
 bit-0.025
Token it
Token ID340
Feature activation+5.99e+00

Pos logit contributions
 forgot+0.090
 remembered+0.087
 blew+0.085
 sne+0.084
 wish+0.084
Neg logit contributions
 This-0.039
 If-0.038
".-0.037
'.-0.035
and-0.035
Token was
Token ID373
Feature activation+2.35e+00

Pos logit contributions
 groom+0.311
Ta+0.300
 ferry+0.292
 bride+0.283
 Cubs+0.279
Neg logit contributions
.-0.187
-0.166
 with-0.132
 She-0.126
 "-0.122
Token scared
Token ID12008
Feature activation-1.64e+00

Pos logit contributions
 forgot+0.102
 wish+0.095
 remembered+0.095
 dove+0.094
 blew+0.093
Neg logit contributions
 If-0.085
.-0.083
".-0.080
 This-0.077
and-0.076
Token.
Token ID13
Feature activation-1.25e+00

Pos logit contributions
keeping+0.063
lor+0.063
angers+0.063
ings+0.062
'.+0.061
Neg logit contributions
 hurt-0.075
 mean-0.073
 sne-0.073
 bit-0.070
 lost-0.070
Token The
Token ID383
Feature activation+3.29e+00

Pos logit contributions
fulness+0.041
.'+0.040
reating+0.039
ions+0.038
and+0.038
Neg logit contributions
 saw-0.063
 wished-0.060
 remembered-0.059
 knew-0.059
 forgot-0.059
Token cat
Token ID3797
Feature activation+4.06e+00

Pos logit contributions
 forgot+0.155
ited+0.150
 remembered+0.149
 regretted+0.138
 rocked+0.137
Neg logit contributions
-0.113
.-0.112
 This-0.107
 "-0.103
 If-0.101
Token cat
Token ID3797
Feature activation-7.55e-01

Pos logit contributions
 forgot+0.045
 remembered+0.043
 blew+0.039
 sne+0.036
 ate+0.036
Neg logit contributions
 This-0.090
".-0.088
'.-0.088
 If-0.087
.'-0.087
Token named
Token ID3706
Feature activation-7.50e-01

Pos logit contributions
 If+0.049
and+0.048
'.+0.047
 This+0.047
.'+0.047
Neg logit contributions
 remembered-0.012
 ate-0.012
 saw-0.011
 forgot-0.011
 sw-0.011
Token Jazz
Token ID21406
Feature activation-1.72e+00

Pos logit contributions
and+0.029
.'+0.028
fulness+0.027
 If+0.027
but+0.027
Neg logit contributions
 sne-0.032
 forgot-0.031
 dove-0.031
 buzz-0.031
 ate-0.030
Token.
Token ID13
Feature activation-1.31e-01

Pos logit contributions
 If+0.053
 This+0.044
 Today+0.043
 onwards+0.043
ings+0.042
Neg logit contributions
 saw-0.092
 forgot-0.091
 remembered-0.089
 knew-0.088
 sne-0.088
Token Everywhere
Token ID48201
Feature activation+8.77e-01

Pos logit contributions
.'+0.004
'.+0.004
and+0.004
".+0.004
 If+0.004
Neg logit contributions
 remembered-0.007
 forgot-0.007
 blew-0.006
 saw-0.006
 sne-0.006
Token she
Token ID673
Feature activation+5.98e+00

Pos logit contributions
 forgot+0.045
 remembered+0.045
 blew+0.042
 sne+0.042
 wish+0.041
Neg logit contributions
 This-0.027
'.-0.026
 If-0.026
.'-0.026
".-0.026
Token went
Token ID1816
Feature activation-4.34e-01

Pos logit contributions
inge+0.244
 promise+0.228
 disgusted+0.209
 forget+0.209
 quarrel+0.204
Neg logit contributions
.-0.241
-0.208
 He-0.204
 She-0.203
 They-0.200
Token,
Token ID11
Feature activation-3.19e-01

Pos logit contributions
.'+0.015
'.+0.015
".+0.014
and+0.014
 If+0.014
Neg logit contributions
 remembered-0.020
 forgot-0.020
 saw-0.020
 chased-0.019
 dove-0.019
Token she
Token ID673
Feature activation+4.21e+00

Pos logit contributions
'.+0.012
.'+0.012
".+0.011
 If+0.011
.'"+0.011
Neg logit contributions
 forgot-0.015
 remembered-0.015
 sne-0.014
 saw-0.014
 dove-0.014
Token was
Token ID373
Feature activation+1.92e+00

Pos logit contributions
Uh+0.123
Ta+0.119
 spit+0.112
 wish+0.112
inge+0.111
Neg logit contributions
.-0.202
-0.195
 She-0.180
 This-0.179
."-0.176
Token always
Token ID1464
Feature activation+7.10e-01

Pos logit contributions
 forgot+0.070
 remembered+0.066
 wish+0.064
 blew+0.063
 bit+0.059
Neg logit contributions
 This-0.086
 If-0.084
".-0.082
and-0.081
'.-0.078
Token it
Token ID340
Feature activation+3.80e+00

Pos logit contributions
 forgot+0.066
 remembered+0.065
 blew+0.062
 sne+0.062
 wish+0.061
Neg logit contributions
 This-0.032
 If-0.031
'.-0.031
".-0.031
.'-0.031
Token could
Token ID714
Feature activation+1.88e+00

Pos logit contributions
Ta+0.162
 groom+0.157
 wish+0.138
 wave+0.136
ited+0.135
Neg logit contributions
.-0.140
-0.128
 anymore-0.123
 before-0.119
 If-0.117
Token not
Token ID407
Feature activation+1.19e+00

Pos logit contributions
 remembered+0.075
 forgot+0.074
 blew+0.069
 ate+0.067
 regretted+0.067
Neg logit contributions
 This-0.075
 If-0.075
".-0.075
'.-0.073
.'-0.072
Token.
Token ID13
Feature activation-1.32e+00

Pos logit contributions
 forgot+0.041
 remembered+0.040
 blew+0.037
 sne+0.035
 ate+0.035
Neg logit contributions
 This-0.057
'.-0.057
".-0.057
 If-0.056
.'-0.056
Token Everywhere
Token ID48201
Feature activation+1.49e+00

Pos logit contributions
and+0.040
fulness+0.040
.'+0.038
".+0.037
uts+0.036
Neg logit contributions
 sne-0.061
 remembered-0.060
 bit-0.059
 wished-0.059
 saw-0.059
Token it
Token ID340
Feature activation+5.96e+00

Pos logit contributions
 forgot+0.083
 remembered+0.081
 sne+0.078
 blew+0.077
 wish+0.077
Neg logit contributions
 This-0.038
".-0.037
'.-0.037
 If-0.036
.'-0.036
Token looked
Token ID3114
Feature activation-7.84e-02

Pos logit contributions
Ta+0.201
ige+0.194
 groom+0.187
 traitor+0.183
inge+0.182
Neg logit contributions
.-0.230
-0.219
 before-0.199
 to-0.199
 He-0.189
Token there
Token ID612
Feature activation+5.20e-01

Pos logit contributions
and+0.002
 This+0.002
 If+0.002
'.+0.002
.'+0.002
Neg logit contributions
 forgot-0.004
 remembered-0.004
 blew-0.004
 saw-0.004
 dropped-0.004
Token was
Token ID373
Feature activation+8.23e-01

Pos logit contributions
 forgot+0.015
 remembered+0.015
 blew+0.014
 sne+0.013
 dropped+0.013
Neg logit contributions
 This-0.028
 If-0.028
and-0.028
'.-0.028
.'-0.027
Token too
Token ID1165
Feature activation+8.96e-01

Pos logit contributions
 forgot+0.034
 remembered+0.034
 blew+0.032
 sne+0.032
 dropped+0.031
Neg logit contributions
 This-0.034
'.-0.033
.'-0.033
 If-0.033
".-0.033
Token much
Token ID881
Feature activation+5.82e-01

Pos logit contributions
 forgot+0.027
 remembered+0.026
 blew+0.024
 sne+0.023
 dropped+0.023
Neg logit contributions
 This-0.047
'.-0.047
 If-0.046
and-0.046
.'-0.046
Token play
Token ID711
Feature activation+2.50e-01

Pos logit contributions
 forgot+0.085
 remembered+0.084
 blew+0.081
ited+0.080
 regretted+0.079
Neg logit contributions
 This-0.126
-0.117
'.-0.116
 If-0.116
".-0.115
Token with
Token ID351
Feature activation+4.05e-01

Pos logit contributions
 forgot+0.013
 remembered+0.013
 sne+0.013
 blew+0.013
 buzz+0.013
Neg logit contributions
 If-0.007
'.-0.007
 This-0.007
.'-0.006
and-0.006
Token it
Token ID340
Feature activation-3.47e-01

Pos logit contributions
 forgot+0.019
 remembered+0.018
 dropped+0.018
 wish+0.018
 dove+0.018
Neg logit contributions
 This-0.014
'.-0.014
.'-0.014
 If-0.014
and-0.014
Token.
Token ID13
Feature activation-1.04e+00

Pos logit contributions
 This+0.009
'.+0.009
 If+0.008
.'+0.008
and+0.008
Neg logit contributions
 forgot-0.020
 remembered-0.020
 bit-0.019
 saw-0.019
 ate-0.019
Token Everywhere
Token ID48201
Feature activation+1.37e+00

Pos logit contributions
.'+0.027
ions+0.026
'.+0.026
ion+0.026
".+0.026
Neg logit contributions
 saw-0.058
 forgot-0.057
 remembered-0.056
 observed-0.055
 knew-0.053
Token she
Token ID673
Feature activation+5.96e+00

Pos logit contributions
 forgot+0.079
 remembered+0.078
 blew+0.075
 sne+0.075
 wish+0.074
Neg logit contributions
 This-0.033
 If-0.031
'.-0.031
".-0.031
.'-0.030
Token went
Token ID1816
Feature activation+8.07e-02

Pos logit contributions
 wish+0.218
 promise+0.215
inge+0.213
cod+0.212
 cruel+0.206
Neg logit contributions
.-0.219
 She-0.217
 He-0.200
-0.198
 They-0.193
Token,
Token ID11
Feature activation+2.51e-01

Pos logit contributions
 forgot+0.004
 remembered+0.004
 saw+0.004
 blew+0.004
 dove+0.004
Neg logit contributions
.'-0.002
'.-0.002
and-0.002
 If-0.002
".-0.002
Token she
Token ID673
Feature activation+4.50e+00

Pos logit contributions
 forgot+0.013
 remembered+0.013
 sne+0.012
 blew+0.012
 saw+0.012
Neg logit contributions
'.-0.008
.'-0.008
".-0.008
 If-0.008
 This-0.008
Token felt
Token ID2936
Feature activation+1.35e+00

Pos logit contributions
 wish+0.130
Ta+0.121
Uh+0.117
 spit+0.107
 promise+0.106
Neg logit contributions
.-0.212
-0.206
and-0.189
 She-0.189
 to-0.183
Token the
Token ID262
Feature activation+2.32e+00

Pos logit contributions
 forgot+0.046
 remembered+0.045
 blew+0.042
 sne+0.041
 ate+0.039
Neg logit contributions
 This-0.065
 If-0.064
".-0.063
'.-0.063
and-0.063
Token The
Token ID383
Feature activation+2.02e+00

Pos logit contributions
.'+0.005
'.+0.005
and+0.005
".+0.005
fulness+0.004
Neg logit contributions
 forgot-0.008
 ate-0.008
 remembered-0.008
 sne-0.008
 blew-0.007
Token horse
Token ID8223
Feature activation+4.32e+00

Pos logit contributions
 forgot+0.080
 remembered+0.074
 regretted+0.066
 blew+0.065
 dropped+0.064
Neg logit contributions
".-0.090
 If-0.086
 This-0.086
-0.083
'.-0.083
Token was
Token ID373
Feature activation+1.76e+00

Pos logit contributions
Ta+0.138
Uh+0.117
ited+0.115
 wish+0.113
 groom+0.109
Neg logit contributions
.-0.216
-0.199
 "-0.186
."-0.181
 This-0.179
Token happy
Token ID3772
Feature activation-6.69e-01

Pos logit contributions
 forgot+0.056
 remembered+0.053
 blew+0.049
 wish+0.049
 dove+0.047
Neg logit contributions
 If-0.087
 This-0.087
".-0.085
and-0.082
'.-0.082
Token and
Token ID290
Feature activation+3.02e+00

Pos logit contributions
 If+0.023
'.+0.022
 This+0.022
.'+0.021
 These+0.020
Neg logit contributions
 remembered-0.036
 forgot-0.035
 saw-0.034
 ate-0.033
 knew-0.032
Token it
Token ID340
Feature activation+5.95e+00

Pos logit contributions
ited+0.077
 wish+0.075
Uh+0.073
 remembered+0.070
Ta+0.069
Neg logit contributions
.-0.152
-0.151
 This-0.151
".-0.147
."-0.146
Token bowed
Token ID35737
Feature activation-5.93e-01

Pos logit contributions
Ta+0.181
Uh+0.173
 wish+0.168
ited+0.155
 sudden+0.151
Neg logit contributions
.-0.272
-0.247
 He-0.228
 She-0.225
."-0.225
Token its
Token ID663
Feature activation+2.21e+00

Pos logit contributions
 If+0.016
 This+0.016
'.+0.016
and+0.016
.'+0.016
Neg logit contributions
 remembered-0.035
 forgot-0.033
 saw-0.032
 ate-0.031
 sw-0.031
Token head
Token ID1182
Feature activation-2.45e+00

Pos logit contributions
 forgot+0.066
 remembered+0.059
 blew+0.059
 ate+0.053
 chased+0.053
Neg logit contributions
".-0.120
 This-0.115
.-0.114
 If-0.114
.'-0.113
Token to
Token ID284
Feature activation+6.25e-01

Pos logit contributions
 If+0.067
 This+0.064
 By+0.062
and+0.060
'.+0.058
Neg logit contributions
 remembered-0.134
 ate-0.131
 sw-0.130
 sne-0.130
 bit-0.129
Token Lily
Token ID20037
Feature activation-2.46e+00

Pos logit contributions
 forgot+0.028
 remembered+0.027
 sne+0.026
 blew+0.026
 wish+0.025
Neg logit contributions
 This-0.024
and-0.024
 If-0.023
'.-0.023
.'-0.023
Token be
Token ID307
Feature activation+7.14e-01

Pos logit contributions
 forgot+0.048
 remembered+0.048
 blew+0.044
 dove+0.043
 ate+0.042
Neg logit contributions
 This-0.085
".-0.085
'.-0.084
 If-0.084
.'-0.083
Token okay
Token ID8788
Feature activation-2.07e+00

Pos logit contributions
 forgot+0.033
 remembered+0.033
 blew+0.031
 sne+0.030
 dropped+0.030
Neg logit contributions
 This-0.027
'.-0.026
 If-0.026
and-0.026
".-0.025
Token.
Token ID13
Feature activation-5.92e-01

Pos logit contributions
 This+0.075
anes+0.074
ays+0.071
 Everyone+0.067
 If+0.067
Neg logit contributions
 remembered-0.104
 forgot-0.101
 knew-0.095
 sne-0.094
 did-0.093
Token Lily
Token ID20037
Feature activation+5.29e+00

Pos logit contributions
.'+0.021
'.+0.020
and+0.020
".+0.020
ings+0.019
Neg logit contributions
 forgot-0.029
 remembered-0.027
 ate-0.027
 saw-0.026
 dropped-0.026
Token felt
Token ID2936
Feature activation+3.09e+00

Pos logit contributions
 flood+0.192
Uh+0.189
Ta+0.177
 ferry+0.173
 holy+0.169
Neg logit contributions
-0.235
.-0.229
 She-0.201
 and-0.200
 "-0.188
Token a
Token ID257
Feature activation+5.94e+00

Pos logit contributions
 blew+0.133
 forgot+0.132
 remembered+0.127
 dove+0.124
 buzz+0.124
Neg logit contributions
 If-0.118
-0.116
 This-0.112
.-0.110
 She-0.106
Token little
Token ID1310
Feature activation+4.80e+00

Pos logit contributions
 met+0.238
 swung+0.235
 chased+0.231
 forgot+0.231
 remembered+0.229
Neg logit contributions
-0.215
.-0.214
".-0.191
'.-0.189
and-0.186
Token weak
Token ID4939
Feature activation+1.50e+00

Pos logit contributions
lly+0.218
 forgot+0.211
ep+0.210
 blew+0.200
 remembered+0.200
Neg logit contributions
.-0.176
 If-0.165
-0.162
 anymore-0.160
 She-0.155
Token because
Token ID780
Feature activation+6.42e-01

Pos logit contributions
 forgot+0.065
 remembered+0.064
 blew+0.060
 ate+0.058
 chased+0.058
Neg logit contributions
 This-0.057
 If-0.056
".-0.056
'.-0.056
.'-0.055
Token she
Token ID673
Feature activation+5.00e+00

Pos logit contributions
 forgot+0.034
 remembered+0.034
 blew+0.032
 sne+0.031
 wish+0.031
Neg logit contributions
'.-0.019
 This-0.019
.'-0.019
".-0.019
 If-0.018
Token was
Token ID373
Feature activation+1.99e+00

Pos logit contributions
 ferry+0.195
 groom+0.195
 swoop+0.179
 flood+0.177
 island+0.176
Neg logit contributions
-0.204
.-0.197
 She-0.187
 That-0.185
 This-0.175
Token what
Token ID644
Feature activation+3.01e-01

Pos logit contributions
'.+0.021
.'+0.021
 This+0.021
 If+0.020
".+0.019
Neg logit contributions
 forgot-0.023
 remembered-0.022
 sne-0.022
 dove-0.022
 dropped-0.021
Token it
Token ID340
Feature activation+2.54e+00

Pos logit contributions
 remembered+0.014
 forgot+0.014
 sne+0.013
 blew+0.013
 dove+0.013
Neg logit contributions
 If-0.011
.'-0.011
'.-0.011
and-0.010
".-0.010
Token was
Token ID373
Feature activation-2.56e-01

Pos logit contributions
 forgot+0.095
 remembered+0.095
 groom+0.093
 wish+0.090
 dove+0.088
Neg logit contributions
 This-0.104
-0.100
.-0.099
".-0.096
 She-0.096
Token,
Token ID11
Feature activation-4.28e-01

Pos logit contributions
 If+0.008
and+0.007
'.+0.007
 This+0.007
.'+0.007
Neg logit contributions
 forgot-0.014
 remembered-0.014
 sne-0.013
 blew-0.013
 saw-0.013
Token so
Token ID523
Feature activation+1.92e+00

Pos logit contributions
'.+0.016
.'+0.015
".+0.015
 This+0.015
ings+0.014
Neg logit contributions
 forgot-0.021
 remembered-0.020
 sne-0.020
 wished-0.020
 saw-0.019
Token she
Token ID673
Feature activation+5.94e+00

Pos logit contributions
 forgot+0.118
 remembered+0.116
 blew+0.116
 sne+0.114
 buzz+0.113
Neg logit contributions
 This-0.033
 If-0.029
.-0.028
".-0.028
 She-0.028
Token asked
Token ID1965
Feature activation-1.69e-01

Pos logit contributions
inge+0.224
 wish+0.222
 connection+0.221
 thrill+0.218
 bond+0.215
Neg logit contributions
-0.195
.-0.195
 to-0.175
 She-0.173
 He-0.169
Token her
Token ID607
Feature activation+1.55e+00

Pos logit contributions
'.+0.008
.'+0.007
uts+0.007
".+0.007
and+0.007
Neg logit contributions
 remembered-0.007
 saw-0.006
 dropped-0.006
 knew-0.006
 forgot-0.006
Token mom
Token ID1995
Feature activation-2.36e+00

Pos logit contributions
 forgot+0.067
 remembered+0.064
 blew+0.062
 ate+0.060
 chased+0.059
Neg logit contributions
 This-0.060
 If-0.059
".-0.059
and-0.058
'.-0.058
Tokenmy
Token ID1820
Feature activation-3.42e+00

Pos logit contributions
uts+0.125
keeping+0.111
'.+0.109
 This+0.108
over+0.108
Neg logit contributions
 saw-0.078
 remembered-0.075
 cut-0.068
 knew-0.066
 put-0.066
Token.
Token ID13
Feature activation-1.36e+00

Pos logit contributions
keeping+0.149
uts+0.148
umbs+0.142
over+0.141
'.+0.136
Neg logit contributions
 saw-0.124
 remembered-0.117
 put-0.112
 knew-0.112
 c-0.111
Token Everyone
Token ID11075
Feature activation+1.61e+00

Pos logit contributions
fulness+0.112
and+0.107
lights+0.099
ions+0.098
reating+0.096
Neg logit contributions
 observed-0.118
 saw-0.118
 spit-0.117
 picked-0.117
 tried-0.115
Token smiled
Token ID13541
Feature activation-1.51e+00

Pos logit contributions
 forgot+0.043
 remembered+0.041
 blew+0.037
 sne+0.037
 wish+0.036
Neg logit contributions
 This-0.088
'.-0.087
 If-0.087
".-0.086
and-0.086
Token and
Token ID290
Feature activation+1.11e+00

Pos logit contributions
and+0.070
ion+0.068
 This+0.067
 If+0.065
ions+0.065
Neg logit contributions
 remembered-0.067
 forgot-0.058
 saw-0.058
 felt-0.056
 sne-0.054
Token agreed
Token ID4987
Feature activation-2.77e+00

Pos logit contributions
 forgot+0.028
 remembered+0.027
 blew+0.025
 sne+0.024
 ate+0.023
Neg logit contributions
 This-0.063
'.-0.063
 If-0.063
".-0.062
.'-0.062
Token.
Token ID13
Feature activation-3.25e+00

Pos logit contributions
ays+0.103
 If+0.093
anes+0.092
ugs+0.089
angers+0.088
Neg logit contributions
 forgot-0.131
 bit-0.128
 knew-0.128
 sne-0.127
 remembered-0.126
Token
Token ID220
Feature activation-7.32e+00

Pos logit contributions
fulness+0.120
and+0.108
rum+0.107
reating+0.107
ions+0.100
Neg logit contributions
 saw-0.133
 sw-0.131
 threw-0.130
 observed-0.129
 picked-0.127
Token
Token ID198
Feature activation-3.09e+00

Pos logit contributions
keeping+0.332
rier+0.329
overs+0.328
 lately+0.325
fulness+0.318
Neg logit contributions
 saw-0.179
 also-0.158
 met-0.149
 lost-0.149
 felt-0.149
Token
Token ID198
Feature activation-2.85e-02

Pos logit contributions
 anymore+0.102
ment+0.096
ados+0.096
overs+0.095
anks+0.094
Neg logit contributions
 saw-0.132
 forgot-0.124
 met-0.123
 se-0.123
 knew-0.121
TokenIn
Token ID818
Feature activation+3.56e-02

Pos logit contributions
 This+0.001
".+0.001
 If+0.001
'.+0.001
.'+0.001
Neg logit contributions
 forgot-0.001
 remembered-0.001
 blew-0.001
 ate-0.001
 dove-0.001
Token the
Token ID262
Feature activation+1.28e+00

Pos logit contributions
 remembered+0.002
 blew+0.001
 forgot+0.001
 sne+0.001
 saw+0.001
Neg logit contributions
and-0.002
".-0.001
 If-0.001
uts-0.001
'.-0.001
Token game
Token ID983
Feature activation-1.65e+00

Pos logit contributions
 forgot+0.039
 remembered+0.038
 blew+0.035
 sne+0.034
 ate+0.033
Neg logit contributions
 This-0.067
'.-0.065
 If-0.065
".-0.065
.'-0.065
Token without
Token ID1231
Feature activation-9.09e-01

Pos logit contributions
 This+0.158
 Today+0.157
 If+0.156
 Those+0.155
 Do+0.152
Neg logit contributions
 bit-0.132
 packed-0.131
 saw-0.130
 ate-0.128
 cut-0.127
Token her
Token ID607
Feature activation-6.91e-01

Pos logit contributions
 This+0.041
and+0.041
 If+0.040
'.+0.040
 Today+0.039
Neg logit contributions
 sne-0.035
 remembered-0.035
 forgot-0.034
 dropped-0.033
 blew-0.033
Token mom
Token ID1995
Feature activation-2.70e+00

Pos logit contributions
 If+0.035
and+0.034
 This+0.033
'.+0.032
but+0.031
Neg logit contributions
 wish-0.024
 sne-0.023
 forgot-0.023
 bit-0.022
 remembered-0.022
Tokenâ
Token ID22940
Feature activation-4.72e-01

Pos logit contributions
eway+0.129
eways+0.129
ions+0.123
 If+0.122
and+0.122
Neg logit contributions
 forgot-0.097
 bit-0.092
 cut-0.092
 remembered-0.091
 ate-0.088
Token
Token ID26391
Feature activation-2.48e+00

Pos logit contributions
and+0.021
uts+0.021
 If+0.020
 This+0.020
".+0.019
Neg logit contributions
 forgot-0.016
 remembered-0.016
 blew-0.016
 sne-0.015
 ate-0.015
Token
Token ID8151
Feature activation-7.22e+00

Pos logit contributions
 If+0.100
ion+0.089
ions+0.086
fulness+0.086
and+0.085
Neg logit contributions
 forgot-0.120
 remembered-0.105
 sne-0.102
 saw-0.101
 dove-0.100
Tokens
Token ID82
Feature activation-2.07e+00

Pos logit contributions
ment+0.275
 itself+0.272
ures+0.257
eway+0.248
 Him+0.241
Neg logit contributions
 forgot-0.249
 knew-0.226
 hurt-0.224
 got-0.224
 spoiled-0.217
Token help
Token ID1037
Feature activation-3.51e+00

Pos logit contributions
and+0.070
 This+0.069
 If+0.068
but+0.067
ays+0.065
Neg logit contributions
 wish-0.095
 sne-0.092
 remembered-0.090
 bit-0.090
 forgot-0.090
Token.
Token ID13
Feature activation-4.59e+00

Pos logit contributions
ays+0.124
anes+0.118
aris+0.116
eways+0.116
 This+0.115
Neg logit contributions
 sne-0.165
 put-0.164
 hurt-0.162
 cut-0.162
 comb-0.157
Token When
Token ID1649
Feature activation-8.63e-01

Pos logit contributions
 anymore+0.173
ings+0.161
ently+0.160
fulness+0.158
eways+0.157
Neg logit contributions
 forgot-0.155
 remembered-0.153
 rubbed-0.152
 wished-0.151
 knew-0.149
Token she
Token ID673
Feature activation+2.13e+00

Pos logit contributions
and+0.036
 If+0.033
.'+0.032
'.+0.031
.'"+0.030
Neg logit contributions
 forgot-0.040
 sne-0.038
 remembered-0.037
 blew-0.037
 wish-0.036
Token so
Token ID523
Feature activation-1.01e+00

Pos logit contributions
.'+0.071
'.+0.071
and+0.067
.'"+0.066
".+0.066
Neg logit contributions
 forgot-0.080
 remembered-0.079
 sne-0.078
 wished-0.076
 sw-0.075
Token they
Token ID484
Feature activation+2.50e+00

Pos logit contributions
.'+0.042
'.+0.042
and+0.040
.'"+0.039
 If+0.037
Neg logit contributions
 forgot-0.044
 sw-0.044
 remembered-0.043
 dove-0.042
 ate-0.042
Token asked
Token ID1965
Feature activation-2.43e+00

Pos logit contributions
 forgot+0.061
 remembered+0.058
 sne+0.054
 blew+0.053
 wish+0.052
Neg logit contributions
'.-0.133
 This-0.132
".-0.132
 If-0.131
.'-0.131
Token Bob
Token ID5811
Feature activation-3.46e+00

Pos logit contributions
uts+0.100
ula+0.096
and+0.095
ays+0.094
 This+0.091
Neg logit contributions
 sw-0.103
 remembered-0.101
 saw-0.101
 knew-0.099
 ate-0.098
Token.
Token ID13
Feature activation-3.22e+00

Pos logit contributions
fulness+0.115
ula+0.114
and+0.114
uts+0.114
 If+0.113
Neg logit contributions
 remembered-0.161
 saw-0.154
 forgot-0.149
 knew-0.145
 sne-0.144
Token
Token ID220
Feature activation-7.17e+00

Pos logit contributions
ula+0.138
fulness+0.136
and+0.135
overs+0.122
uts+0.121
Neg logit contributions
 sw-0.121
 saw-0.117
 ate-0.115
 bark-0.114
 dove-0.111
Token
Token ID198
Feature activation-2.70e+00

Pos logit contributions
keeping+0.312
ados+0.304
overs+0.290
fulness+0.285
rs+0.276
Neg logit contributions
 saw-0.200
 also-0.187
 knew-0.169
 met-0.165
 lost-0.164
Token
Token ID198
Feature activation-6.97e-01

Pos logit contributions
anks+0.096
overs+0.090
fulness+0.088
ados+0.086
 anymore+0.086
Neg logit contributions
 saw-0.116
 dove-0.113
 sw-0.110
 observed-0.107
lly-0.105
TokenOne
Token ID3198
Feature activation+6.05e-01

Pos logit contributions
and+0.032
fulness+0.032
 This+0.032
.'+0.032
overs+0.032
Neg logit contributions
 remembered-0.021
 dove-0.020
 sw-0.020
 blew-0.020
 chased-0.020
Token day
Token ID1110
Feature activation-5.01e-01

Pos logit contributions
 forgot+0.021
 remembered+0.021
 blew+0.019
 wish+0.019
 sne+0.019
Neg logit contributions
 If-0.028
 This-0.028
'.-0.028
and-0.028
".-0.028
Token,
Token ID11
Feature activation-4.22e-01

Pos logit contributions
and+0.017
.'+0.017
'.+0.016
 If+0.016
 This+0.016
Neg logit contributions
 remembered-0.025
 forgot-0.024
 blew-0.024
 ate-0.024
 sw-0.024
Token wouldn
Token ID3636
Feature activation-2.59e+00

Pos logit contributions
 wish+0.064
 blew+0.063
 forgot+0.063
 buzz+0.062
 remembered+0.062
Neg logit contributions
 If-0.122
".-0.121
.-0.121
'.-0.121
 This-0.119
Token't
Token ID470
Feature activation+2.96e-01

Pos logit contributions
and+0.127
ment+0.112
boxes+0.109
 This+0.108
 If+0.107
Neg logit contributions
 sne-0.093
 hurt-0.093
 blew-0.086
 forgot-0.086
 dropped-0.085
Token allow
Token ID1249
Feature activation-2.29e+00

Pos logit contributions
 forgot+0.010
 remembered+0.009
 sne+0.008
 wish+0.008
 blew+0.008
Neg logit contributions
 This-0.015
and-0.015
 If-0.015
'.-0.015
.'-0.015
Token it
Token ID340
Feature activation-3.67e+00

Pos logit contributions
and+0.095
'.+0.094
.'"+0.093
.'+0.092
 If+0.091
Neg logit contributions
 sne-0.099
 pin-0.091
 hurt-0.089
 cut-0.088
 sc-0.087
Token.
Token ID13
Feature activation-2.76e+00

Pos logit contributions
aps+0.135
 If+0.132
overs+0.131
 Playing+0.130
ions+0.129
Neg logit contributions
 hurt-0.162
 dropped-0.156
 got-0.154
 forgot-0.152
 lost-0.147
Token
Token ID220
Feature activation-7.16e+00

Pos logit contributions
fulness+0.112
".+0.106
ions+0.099
.'"+0.099
.'+0.097
Neg logit contributions
 saw-0.118
 remembered-0.115
 observed-0.115
 forgot-0.111
 felt-0.110
Token
Token ID198
Feature activation-3.50e+00

Pos logit contributions
rs+0.294
keeping+0.293
overs+0.291
ados+0.281
.'"+0.280
Neg logit contributions
 saw-0.202
 knew-0.184
 felt-0.184
 forgot-0.181
 also-0.178
Token
Token ID198
Feature activation+1.43e-01

Pos logit contributions
 anymore+0.118
fulness+0.118
rier+0.115
ions+0.113
.'+0.112
Neg logit contributions
 saw-0.146
 forgot-0.138
 met-0.134
 remembered-0.133
 knew-0.132
TokenSam
Token ID16305
Feature activation+1.40e+00

Pos logit contributions
 forgot+0.006
 remembered+0.006
 blew+0.005
 sne+0.005
 dropped+0.005
Neg logit contributions
".-0.006
 This-0.006
'.-0.006
.'-0.006
 If-0.006
Tokenmy
Token ID1820
Feature activation+2.87e+00

Pos logit contributions
 forgot+0.056
 remembered+0.055
 blew+0.052
 wish+0.051
 sne+0.051
Neg logit contributions
 This-0.057
 If-0.056
'.-0.056
".-0.055
.'-0.055
Token was
Token ID373
Feature activation+1.66e+00

Pos logit contributions
 wish+0.079
 forgot+0.075
 buzz+0.072
 blew+0.072
Ta+0.070
Neg logit contributions
 This-0.138
.-0.137
 If-0.137
'.-0.137
-0.134
Token to
Token ID284
Feature activation+2.85e-01

Pos logit contributions
 If+0.118
and+0.115
 You+0.114
 This+0.113
'.+0.113
Neg logit contributions
 remembered-0.119
 forgot-0.118
 pin-0.116
 dropped-0.116
 hurt-0.115
Token beat
Token ID4405
Feature activation-2.49e+00

Pos logit contributions
 forgot+0.007
 remembered+0.006
 sne+0.006
 blew+0.006
 wish+0.006
Neg logit contributions
and-0.017
 This-0.017
 If-0.017
'.-0.016
".-0.016
Token Gray
Token ID12723
Feature activation-3.50e+00

Pos logit contributions
and+0.091
 Today+0.090
 This+0.088
but+0.088
.'+0.087
Neg logit contributions
 sne-0.107
 hurt-0.105
 pin-0.103
 bit-0.103
 lost-0.100
Tokenâ
Token ID22940
Feature activation-2.32e-01

Pos logit contributions
 If+0.126
and+0.114
ions+0.113
 This+0.112
arl+0.112
Neg logit contributions
 bit-0.157
 forgot-0.150
 sne-0.149
 remembered-0.149
 ate-0.145
Token
Token ID26391
Feature activation-2.46e+00

Pos logit contributions
and+0.010
 If+0.010
uts+0.010
".+0.010
 This+0.010
Neg logit contributions
 forgot-0.008
 remembered-0.008
 sne-0.008
 blew-0.008
 wish-0.008
Token
Token ID8151
Feature activation-7.15e+00

Pos logit contributions
ion+0.102
ions+0.101
 If+0.100
fulness+0.095
and+0.092
Neg logit contributions
 forgot-0.114
 remembered-0.102
 chased-0.096
 sne-0.095
 saw-0.095
Tokens
Token ID82
Feature activation-2.41e+00

Pos logit contributions
ment+0.249
ures+0.232
 itself+0.222
 This+0.218
 If+0.216
Neg logit contributions
 forgot-0.260
 hurt-0.245
 sc-0.241
 bit-0.238
 lost-0.232
Token prices
Token ID4536
Feature activation-4.02e+00

Pos logit contributions
 This+0.084
ays+0.082
 If+0.080
and+0.079
'.+0.078
Neg logit contributions
 wish-0.103
 bit-0.103
 plan-0.102
 forgot-0.101
 remembered-0.101
Token,
Token ID11
Feature activation-2.27e+00

Pos logit contributions
 This+0.127
 Today+0.122
 If+0.119
and+0.111
.'"+0.103
Neg logit contributions
 bit-0.200
 forgot-0.198
 remembered-0.194
 saw-0.192
 kept-0.191
Token because
Token ID780
Feature activation-1.67e+00

Pos logit contributions
.'+0.087
and+0.087
'.+0.086
 This+0.085
 If+0.082
Neg logit contributions
 forgot-0.101
 remembered-0.097
 sne-0.096
 bit-0.095
 chased-0.092
Token everyone
Token ID2506
Feature activation+1.38e+00

Pos logit contributions
and+0.067
.'+0.066
'.+0.064
 This+0.062
 If+0.060
Neg logit contributions
 forgot-0.071
 remembered-0.070
 blew-0.068
 sne-0.067
 dove-0.065
Token safe
Token ID3338
Feature activation-3.01e+00

Pos logit contributions
 forgot+0.025
 remembered+0.024
 blew+0.022
 sne+0.022
 ate+0.021
Neg logit contributions
 This-0.036
'.-0.036
 If-0.036
".-0.036
.'-0.036
Token behind
Token ID2157
Feature activation-3.33e+00

Pos logit contributions
 If+0.107
 This+0.105
ula+0.092
and+0.090
 These+0.085
Neg logit contributions
 sw-0.147
 forgot-0.145
 remembered-0.143
 saw-0.138
 bit-0.135
Token Jack
Token ID3619
Feature activation-2.83e+00

Pos logit contributions
 If+0.152
but+0.140
and+0.135
uts+0.133
 This+0.133
Neg logit contributions
 sne-0.130
 sw-0.126
 lost-0.123
 c-0.122
 spark-0.122
Tokenâ
Token ID22940
Feature activation-4.71e-01

Pos logit contributions
 If+0.117
and+0.103
 This+0.101
but+0.100
ings+0.098
Neg logit contributions
 forgot-0.131
 sne-0.130
 remembered-0.127
 ate-0.123
 saw-0.122
Token
Token ID26391
Feature activation-2.35e+00

Pos logit contributions
and+0.022
 If+0.021
uts+0.021
".+0.020
 This+0.019
Neg logit contributions
 forgot-0.017
 remembered-0.016
 blew-0.015
 ate-0.015
 sne-0.015
Token
Token ID8151
Feature activation-7.13e+00

Pos logit contributions
 If+0.103
fulness+0.087
ion+0.086
and+0.085
rier+0.084
Neg logit contributions
 forgot-0.110
 remembered-0.096
 saw-0.096
 sne-0.094
 ate-0.092
Tokens
Token ID82
Feature activation-1.54e+00

Pos logit contributions
 If+0.230
 This+0.223
 itself+0.215
 Do+0.209
ment+0.205
Neg logit contributions
 forgot-0.283
 knew-0.262
 remembered-0.253
 saw-0.249
 laughed-0.248
Token shield
Token ID7614
Feature activation-4.67e+00

Pos logit contributions
 If+0.067
 This+0.065
and+0.062
'.+0.061
but+0.059
Neg logit contributions
 wish-0.064
 sne-0.061
 remembered-0.059
 forgot-0.059
 dream-0.059
Token.
Token ID13
Feature activation-5.14e+00

Pos logit contributions
 If+0.136
 This+0.134
and+0.119
ula+0.115
 That+0.111
Neg logit contributions
 forgot-0.248
 remembered-0.247
 bit-0.243
 ate-0.236
 sne-0.232
Token<|endoftext|>
Token ID50256
Feature activation-4.07e+00

Pos logit contributions
ings+0.215
rier+0.213
ently+0.204
arling+0.190
fully+0.182
Neg logit contributions
 forgot-0.177
 knew-0.174
 remembered-0.166
 saw-0.162
 learned-0.160
TokenOne
Token ID3198
Feature activation+5.41e-01

Pos logit contributions
rier+0.212
fulness+0.210
uts+0.206
and+0.198
 If+0.197
Neg logit contributions
 saw-0.136
 met-0.120
 did-0.111
 looked-0.109
 had-0.108
Tokenâ
Token ID22940
Feature activation-3.52e-01

Pos logit contributions
ently+0.224
 anymore+0.221
ings+0.216
fulness+0.212
eways+0.206
Neg logit contributions
 dove-0.168
 zoom-0.162
 put-0.161
 wished-0.160
 tried-0.159
Token
Token ID26391
Feature activation-2.42e+00

Pos logit contributions
and+0.016
 If+0.015
uts+0.015
".+0.014
 This+0.014
Neg logit contributions
 forgot-0.013
 remembered-0.012
 blew-0.012
 ate-0.012
 rocked-0.011
Token She
Token ID1375
Feature activation+5.71e-01

Pos logit contributions
fulness+0.084
and+0.080
 If+0.077
fully+0.073
ion+0.073
Neg logit contributions
 forgot-0.125
 saw-0.116
 remembered-0.114
 dove-0.109
 ate-0.109
Token asked
Token ID1965
Feature activation-3.05e+00

Pos logit contributions
 forgot+0.017
 remembered+0.016
 blew+0.015
 sne+0.015
 wish+0.014
Neg logit contributions
 This-0.030
and-0.030
 If-0.030
'.-0.030
.'-0.029
Token.
Token ID13
Feature activation-5.94e+00

Pos logit contributions
uts+0.124
 This+0.114
and+0.114
 Today+0.105
 If+0.104
Neg logit contributions
 sne-0.126
 remembered-0.123
 knew-0.122
 saw-0.121
 forgot-0.116
Token
Token ID220
Feature activation-7.10e+00

Pos logit contributions
fulness+0.234
 anymore+0.234
ently+0.230
ings+0.229
anks+0.215
Neg logit contributions
 saw-0.222
 climbed-0.210
 put-0.205
 rubbed-0.205
 pet-0.204
Token
Token ID198
Feature activation-3.66e+00

Pos logit contributions
keeping+0.298
overs+0.286
 anymore+0.286
fulness+0.285
ados+0.279
Neg logit contributions
 saw-0.183
 climbed-0.168
 also-0.165
 knew-0.165
 felt-0.161
Token
Token ID198
Feature activation-1.45e+00

Pos logit contributions
 anymore+0.154
anks+0.140
ays+0.132
unes+0.127
rier+0.126
Neg logit contributions
 saw-0.145
 dove-0.141
 blew-0.138
lly-0.138
 knew-0.134
TokenTo
Token ID2514
Feature activation-2.43e+00

Pos logit contributions
 anymore+0.074
anks+0.067
fulness+0.066
 This+0.066
ays+0.065
Neg logit contributions
 blew-0.044
 remembered-0.044
 forgot-0.044
 dove-0.044
 chased-0.043
Token her
Token ID607
Feature activation+1.96e-01

Pos logit contributions
and+0.121
uts+0.111
 If+0.110
but+0.104
 This+0.103
Neg logit contributions
 sne-0.086
 forgot-0.082
 wish-0.079
 put-0.079
 remembered-0.078
Token surprise
Token ID5975
Feature activation-2.75e+00

Pos logit contributions
 remembered+0.005
 forgot+0.005
 wish+0.005
 sne+0.005
 bit+0.005
Neg logit contributions
and-0.011
 This-0.011
 If-0.011
".-0.010
'.-0.010
Token wouldn
Token ID3636
Feature activation-2.09e+00

Pos logit contributions
 forgot+0.033
 remembered+0.031
 blew+0.028
 sne+0.027
 wish+0.027
Neg logit contributions
 This-0.078
'.-0.078
 If-0.077
".-0.077
.'-0.077
Token't
Token ID470
Feature activation+4.19e-01

Pos logit contributions
and+0.095
 If+0.085
boxes+0.084
 This+0.083
ment+0.082
Neg logit contributions
 sne-0.080
 hurt-0.078
 blew-0.074
 dropped-0.073
 forgot-0.073
Token let
Token ID1309
Feature activation-1.50e+00

Pos logit contributions
 forgot+0.012
 remembered+0.011
 sne+0.010
 wish+0.010
 blew+0.009
Neg logit contributions
 This-0.024
 If-0.023
'.-0.023
and-0.023
.'-0.023
Token him
Token ID683
Feature activation-1.09e+00

Pos logit contributions
and+0.062
aps+0.054
uts+0.054
ions+0.052
but+0.052
Neg logit contributions
 sne-0.067
 hurt-0.063
 dropped-0.063
 remembered-0.062
 wish-0.062
Token.
Token ID13
Feature activation-1.60e+00

Pos logit contributions
 If+0.062
 This+0.058
ions+0.056
'.+0.056
and+0.056
Neg logit contributions
 sne-0.032
 forgot-0.031
 wish-0.029
 remembered-0.028
 hurt-0.028
Token
Token ID220
Feature activation-7.09e+00

Pos logit contributions
and+0.052
uts+0.046
ions+0.046
ion+0.045
fulness+0.045
Neg logit contributions
 remembered-0.084
 saw-0.081
 sw-0.079
 forgot-0.079
 wished-0.078
Token
Token ID198
Feature activation-2.86e+00

Pos logit contributions
rs+0.300
overs+0.292
ados+0.285
keeping+0.282
ting+0.276
Neg logit contributions
 saw-0.178
 also-0.169
 lost-0.167
 felt-0.167
 forgot-0.157
Token
Token ID198
Feature activation+3.71e-01

Pos logit contributions
 anymore+0.086
ion+0.083
ions+0.083
ays+0.081
unes+0.081
Neg logit contributions
 met-0.139
 saw-0.139
 remembered-0.137
 forgot-0.130
 observed-0.128
Token"
Token ID1
Feature activation+8.93e-01

Pos logit contributions
 remembered+0.016
 forgot+0.016
 blew+0.015
 dropped+0.014
 sne+0.014
Neg logit contributions
 This-0.014
 If-0.014
and-0.014
".-0.014
'.-0.014
TokenMom
Token ID29252
Feature activation-1.32e+00

Pos logit contributions
 forgot+0.034
 remembered+0.033
 blew+0.031
 sne+0.030
 ate+0.030
Neg logit contributions
 This-0.040
 If-0.039
".-0.039
'.-0.039
.'-0.038
Token,
Token ID11
Feature activation+5.86e-01

Pos logit contributions
and+0.059
uts+0.054
 This+0.053
 If+0.053
fulness+0.051
Neg logit contributions
 remembered-0.059
 forgot-0.058
 saw-0.055
 sne-0.053
 blew-0.051
Token keep
Token ID1394
Feature activation-3.90e-01

Pos logit contributions
 forgot+0.024
 remembered+0.024
 blew+0.021
 sne+0.021
 wish+0.021
Neg logit contributions
 This-0.051
and-0.050
 If-0.050
'.-0.050
".-0.049
Token talking
Token ID3375
Feature activation-4.00e+00

Pos logit contributions
 If+0.019
and+0.019
'.+0.018
.'+0.018
 This+0.018
Neg logit contributions
 remembered-0.014
 sne-0.014
 pin-0.013
 sw-0.013
 forgot-0.013
Token and
Token ID290
Feature activation+5.46e-01

Pos logit contributions
 This+0.137
 If+0.131
 These+0.131
 Those+0.129
ays+0.128
Neg logit contributions
 spark-0.161
 laughed-0.161
 zoom-0.160
 saw-0.160
 sne-0.159
Token ask
Token ID1265
Feature activation-3.17e+00

Pos logit contributions
 forgot+0.017
 remembered+0.017
 blew+0.015
 sne+0.015
 wish+0.015
Neg logit contributions
 This-0.028
'.-0.028
 If-0.028
.'-0.028
".-0.028
Token silly
Token ID14397
Feature activation-1.10e+00

Pos logit contributions
uts+0.149
umbs+0.142
ays+0.139
 This+0.123
ugs+0.122
Neg logit contributions
 knew-0.115
 remembered-0.115
 sw-0.112
 sne-0.112
 c-0.109
Token questions
Token ID2683
Feature activation-7.08e+00

Pos logit contributions
 This+0.055
 If+0.055
'.+0.053
.'+0.053
 By+0.051
Neg logit contributions
 sne-0.037
 sw-0.034
 dove-0.033
 wish-0.033
 se-0.033
Token.
Token ID13
Feature activation-3.03e+00

Pos logit contributions
keeping+0.261
arling+0.242
 This+0.238
umbs+0.236
ados+0.234
Neg logit contributions
 did-0.259
 also-0.247
 put-0.239
 fl-0.238
 saw-0.238
Token
Token ID6184
Feature activation+5.02e-01

Pos logit contributions
fulness+0.109
ions+0.108
.'+0.100
ings+0.099
ays+0.098
Neg logit contributions
 bark-0.117
 che-0.117
 l-0.114
 bit-0.113
 ate-0.111
Token
Token ID95
Feature activation+3.44e-01

Pos logit contributions
 forgot+0.022
 remembered+0.021
 blew+0.020
 sne+0.020
 wish+0.020
Neg logit contributions
 If-0.019
and-0.019
.'-0.018
 This-0.018
'.-0.018
Token
Token ID26391
Feature activation-1.18e+00

Pos logit contributions
 forgot+0.015
 remembered+0.015
 blew+0.014
 ate+0.014
 sne+0.014
Neg logit contributions
and-0.012
 If-0.012
.'-0.012
".-0.012
 This-0.012
Token
Token ID129
Feature activation-9.42e-02

Pos logit contributions
 If+0.041
and+0.039
ion+0.039
fulness+0.039
.'+0.037
Neg logit contributions
 forgot-0.058
 remembered-0.056
 saw-0.053
 ate-0.052
 sne-0.050
Token of
Token ID286
Feature activation-6.94e-02

Pos logit contributions
'.+0.061
.'+0.060
.'"+0.058
".+0.056
 This+0.056
Neg logit contributions
 sne-0.073
 forgot-0.070
 remembered-0.070
 dropped-0.070
 blew-0.069
Token a
Token ID257
Feature activation+1.38e+00

Pos logit contributions
'.+0.003
.'+0.003
 This+0.003
and+0.003
".+0.003
Neg logit contributions
 remembered-0.003
 forgot-0.003
 blew-0.003
 sne-0.003
 wish-0.003
Token new
Token ID649
Feature activation+8.12e-01

Pos logit contributions
 forgot+0.029
 remembered+0.028
 blew+0.025
 sne+0.024
 ate+0.024
Neg logit contributions
 This-0.083
'.-0.083
 If-0.083
".-0.083
.'-0.082
Token idea
Token ID2126
Feature activation-2.17e+00

Pos logit contributions
 forgot+0.021
 remembered+0.021
 sne+0.019
 blew+0.019
 wish+0.018
Neg logit contributions
 This-0.045
 If-0.045
and-0.045
'.-0.045
".-0.044
Token.
Token ID13
Feature activation-2.87e+00

Pos logit contributions
 If+0.066
 This+0.064
and+0.063
ays+0.063
angers+0.062
Neg logit contributions
 remembered-0.109
 knew-0.108
 did-0.107
 tried-0.106
 saw-0.105
Token
Token ID220
Feature activation-7.07e+00

Pos logit contributions
fulness+0.137
 anymore+0.134
ings+0.126
.'+0.124
 ourselves+0.124
Neg logit contributions
 tried-0.086
 saw-0.083
 looked-0.082
 felt-0.079
 took-0.078
Token
Token ID198
Feature activation-3.79e+00

Pos logit contributions
overs+0.326
keeping+0.321
ados+0.318
ment+0.308
rier+0.308
Neg logit contributions
 saw-0.179
 also-0.155
 lost-0.153
 felt-0.152
 knew-0.151
Token
Token ID198
Feature activation-3.68e-01

Pos logit contributions
unes+0.147
ados+0.145
 anymore+0.145
overs+0.144
rier+0.142
Neg logit contributions
 saw-0.145
 met-0.129
 knew-0.127
 tried-0.124
 forgot-0.121
TokenHe
Token ID1544
Feature activation+3.77e+00

Pos logit contributions
 anymore+0.016
 This+0.016
.'+0.016
".+0.016
'.+0.016
Neg logit contributions
 forgot-0.013
 remembered-0.013
 blew-0.012
 chased-0.012
 cut-0.011
Token decided
Token ID3066
Feature activation-6.67e-01

Pos logit contributions
 wish+0.095
 forgot+0.086
Ta+0.085
 boo+0.080
 forget+0.079
Neg logit contributions
.-0.188
-0.178
'.-0.175
".-0.173
.'-0.169
Token to
Token ID284
Feature activation+1.38e+00

Pos logit contributions
'.+0.023
.'+0.023
and+0.022
 If+0.022
 Today+0.021
Neg logit contributions
 remembered-0.034
 forgot-0.033
 saw-0.032
 dropped-0.031
 wished-0.031
Token and
Token ID290
Feature activation+8.58e-01

Pos logit contributions
and+0.178
 This+0.173
 If+0.167
 These+0.156
 As+0.156
Neg logit contributions
 remembered-0.185
 blew-0.171
 hurt-0.167
 pin-0.165
 mean-0.162
Token try
Token ID1949
Feature activation-3.70e+00

Pos logit contributions
 forgot+0.028
 remembered+0.027
 blew+0.025
 sne+0.024
 wish+0.024
Neg logit contributions
 This-0.043
'.-0.043
 If-0.043
".-0.042
and-0.042
Token hard
Token ID1327
Feature activation-4.32e+00

Pos logit contributions
ugs+0.146
umbs+0.144
overs+0.139
 This+0.135
 If+0.135
Neg logit contributions
 remembered-0.149
 pin-0.141
 broken-0.141
 forgot-0.140
 fl-0.139
Tokenâ
Token ID22940
Feature activation-7.36e-03

Pos logit contributions
 This+0.142
overs+0.136
 If+0.133
ion+0.129
and+0.127
Neg logit contributions
 sne-0.187
 remembered-0.185
 forgot-0.177
 hurt-0.175
 blew-0.173
Token
Token ID26391
Feature activation-3.25e+00

Pos logit contributions
and+0.000
 If+0.000
 This+0.000
uts+0.000
".+0.000
Neg logit contributions
 forgot-0.000
 remembered-0.000
 blew-0.000
 sne-0.000
 ate-0.000
Token
Token ID8151
Feature activation-7.04e+00

Pos logit contributions
ion+0.121
 If+0.118
fulness+0.115
ula+0.110
reating+0.107
Neg logit contributions
 forgot-0.149
 remembered-0.134
 chased-0.128
 ate-0.126
 met-0.126
Token.
Token ID13
Feature activation-3.95e+00

Pos logit contributions
ment+0.256
 itself+0.246
 If+0.242
 Him+0.241
 For+0.227
Neg logit contributions
 forgot-0.220
 hurt-0.218
 got-0.216
 lost-0.204
 spoiled-0.203
Token
Token ID220
Feature activation-5.78e+00

Pos logit contributions
 anymore+0.141
ings+0.139
uts+0.128
fulness+0.127
eways+0.126
Neg logit contributions
 remembered-0.161
 wished-0.148
 forgot-0.147
 knew-0.145
 wish-0.145
Token
Token ID198
Feature activation-2.25e+00

Pos logit contributions
fulness+0.202
 anymore+0.194
overs+0.188
fully+0.181
lights+0.179
Neg logit contributions
 forgot-0.202
 succeeded-0.192
 climbed-0.190
 met-0.188
 dropped-0.187
Token
Token ID198
Feature activation-3.18e-01

Pos logit contributions
 anymore+0.074
ays+0.064
unes+0.063
overs+0.063
.'+0.062
Neg logit contributions
 knew-0.102
 saw-0.101
 dove-0.098
 remembered-0.098
 forgot-0.097
TokenAll
Token ID3237
Feature activation+1.95e-01

Pos logit contributions
 This+0.015
".+0.014
 anymore+0.014
.'+0.014
 If+0.014
Neg logit contributions
 remembered-0.011
 forgot-0.011
 blew-0.010
 dove-0.010
 chased-0.010
Token important
Token ID1593
Feature activation-3.84e+00

Pos logit contributions
and+0.024
 If+0.023
 This+0.022
'.+0.022
.'+0.022
Neg logit contributions
 forgot-0.019
 remembered-0.018
 bit-0.018
 blew-0.018
 sne-0.018
Token!
Token ID0
Feature activation-4.83e+00

Pos logit contributions
and+0.149
 If+0.134
 This+0.130
umbs+0.120
overs+0.119
Neg logit contributions
 forgot-0.167
 remembered-0.166
 blew-0.162
 did-0.160
 knew-0.160
Tokenâ
Token ID22940
Feature activation+1.03e-01

Pos logit contributions
ings+0.202
ently+0.190
 anymore+0.185
fulness+0.183
uts+0.172
Neg logit contributions
 remembered-0.167
Uh-0.160
 threw-0.158
 saw-0.158
 sne-0.155
Token
Token ID26391
Feature activation-2.12e+00

Pos logit contributions
 forgot+0.004
 remembered+0.004
 blew+0.004
 ate+0.004
 bit+0.004
Neg logit contributions
and-0.004
 If-0.004
".-0.004
 This-0.004
uts-0.004
Token.
Token ID13
Feature activation-4.92e+00

Pos logit contributions
fulness+0.073
 If+0.070
and+0.070
ion+0.066
fully+0.065
Neg logit contributions
 forgot-0.107
 remembered-0.098
 saw-0.097
 ate-0.093
 chased-0.089
Token
Token ID220
Feature activation-7.01e+00

Pos logit contributions
ings+0.190
ently+0.177
 anymore+0.174
fulness+0.167
uts+0.164
Neg logit contributions
 remembered-0.184
 sne-0.169
 threw-0.167
 forgot-0.167
 switched-0.166
Token
Token ID198
Feature activation-3.46e+00

Pos logit contributions
 anymore+0.265
keeping+0.263
fully+0.259
fulness+0.258
overs+0.255
Neg logit contributions
 also-0.174
 forgot-0.168
 saw-0.166
 felt-0.155
 took-0.153
Token
Token ID198
Feature activation-4.94e-01

Pos logit contributions
 anymore+0.134
rier+0.126
anks+0.126
ados+0.122
unes+0.120
Neg logit contributions
 saw-0.134
lly-0.126
 met-0.123
 blew-0.120
 observed-0.120
TokenThey
Token ID2990
Feature activation+2.41e+00

Pos logit contributions
 This+0.023
 anymore+0.022
 If+0.022
".+0.022
.'+0.021
Neg logit contributions
 remembered-0.017
 forgot-0.017
 blew-0.016
 chased-0.015
 saw-0.015
Token took
Token ID1718
Feature activation+4.01e-02

Pos logit contributions
 forgot+0.050
 remembered+0.046
 blew+0.043
 wish+0.042
 sne+0.041
Neg logit contributions
'.-0.138
 This-0.136
.'-0.135
".-0.134
 If-0.134
Token the
Token ID262
Feature activation+1.66e+00

Pos logit contributions
 forgot+0.002
 remembered+0.002
 sne+0.002
 bit+0.002
 dropped+0.002
Neg logit contributions
and-0.002
 If-0.002
".-0.002
.'-0.002
'.-0.002
Token pan
Token ID3425
Feature activation-2.69e+00

Pos logit contributions
 forgot+0.041
 remembered+0.040
 blew+0.036
 sne+0.035
 ate+0.035
Neg logit contributions
 This-0.062
'.-0.062
".-0.062
 If-0.061
.'-0.061
Token at
Token ID379
Feature activation-2.48e+00

Pos logit contributions
 If+0.096
and+0.092
 This+0.088
 By+0.081
 Today+0.079
Neg logit contributions
 cut-0.132
 bit-0.131
 blew-0.131
 dropped-0.131
 sne-0.130
Token John
Token ID1757
Feature activation-2.06e+00

Pos logit contributions
and+0.115
uts+0.114
 If+0.102
aring+0.099
but+0.097
Neg logit contributions
 lost-0.092
 saw-0.090
 sne-0.088
 remembered-0.086
 blew-0.085
Tokenâ
Token ID22940
Feature activation-2.76e-01

Pos logit contributions
 If+0.084
fulness+0.081
ions+0.077
 This+0.077
and+0.076
Neg logit contributions
 sne-0.087
 dropped-0.085
 ate-0.083
 bit-0.082
 forgot-0.082
Token
Token ID26391
Feature activation-1.61e+00

Pos logit contributions
and+0.013
uts+0.012
 If+0.012
".+0.012
 This+0.011
Neg logit contributions
 remembered-0.010
 forgot-0.010
 blew-0.009
 sne-0.009
 ate-0.009
Token
Token ID8151
Feature activation-7.00e+00

Pos logit contributions
 If+0.066
and+0.060
fulness+0.059
ion+0.057
ions+0.056
Neg logit contributions
 forgot-0.077
 remembered-0.069
 saw-0.068
 dropped-0.067
 sne-0.066
Tokens
Token ID82
Feature activation-1.43e+00

Pos logit contributions
ment+0.235
 situations+0.223
 This+0.221
ures+0.221
 If+0.213
Neg logit contributions
 forgot-0.280
 knew-0.265
 remembered-0.247
 laughed-0.246
 saw-0.245
Token feet
Token ID3625
Feature activation-4.03e+00

Pos logit contributions
 If+0.060
and+0.060
 This+0.058
but+0.055
'.+0.055
Neg logit contributions
 wish-0.059
 sne-0.058
 remembered-0.053
 forgot-0.053
 dropped-0.052
Token.
Token ID13
Feature activation-4.16e+00

Pos logit contributions
 This+0.125
 If+0.123
 Today+0.110
 For+0.108
 Sometimes+0.107
Neg logit contributions
 dropped-0.213
 hurt-0.209
 forgot-0.208
 blew-0.208
 cut-0.207
Token He
Token ID679
Feature activation+3.02e+00

Pos logit contributions
ently+0.159
 anymore+0.158
ings+0.156
fulness+0.154
and+0.148
Neg logit contributions
 saw-0.141
 dropped-0.138
 sne-0.138
 forgot-0.138
 wished-0.137
Token then
Token ID788
Feature activation+1.59e+00

Pos logit contributions
 forgot+0.067
 remembered+0.063
 wish+0.059
 dove+0.055
 blew+0.055
Neg logit contributions
".-0.158
'.-0.157
 This-0.156
.-0.153
 If-0.152
Token with
Token ID351
Feature activation-2.67e+00

Pos logit contributions
uts+0.105
and+0.100
ula+0.100
 If+0.098
 This+0.095
Neg logit contributions
 bit-0.161
 sne-0.155
 wish-0.155
 ate-0.154
 remembered-0.154
Token his
Token ID465
Feature activation-7.58e-01

Pos logit contributions
and+0.118
but+0.115
uts+0.113
 If+0.109
 This+0.109
Neg logit contributions
 sne-0.097
 wish-0.096
 remembered-0.092
 bit-0.091
 cut-0.091
Token friends
Token ID2460
Feature activation-4.26e+00

Pos logit contributions
 If+0.041
and+0.041
 This+0.040
'.+0.038
.'+0.038
Neg logit contributions
 remembered-0.021
 wish-0.021
 forgot-0.021
 sne-0.021
 bit-0.020
Token instead
Token ID2427
Feature activation-4.29e+00

Pos logit contributions
and+0.114
uts+0.105
 If+0.098
angers+0.095
ula+0.093
Neg logit contributions
 remembered-0.238
 bit-0.229
 forgot-0.227
 ate-0.222
 blew-0.221
Token.
Token ID13
Feature activation-5.75e+00

Pos logit contributions
and+0.113
 If+0.096
but+0.091
angers+0.091
 This+0.088
Neg logit contributions
 forgot-0.228
 remembered-0.227
 blew-0.226
 bit-0.226
 sne-0.225
Token
Token ID220
Feature activation-6.98e+00

Pos logit contributions
ings+0.207
fulness+0.190
arling+0.189
 anymore+0.189
ently+0.188
Neg logit contributions
 remembered-0.198
 ate-0.198
 wished-0.196
 sw-0.194
 threw-0.193
Token
Token ID198
Feature activation-2.71e+00

Pos logit contributions
 anymore+0.261
fulness+0.240
 lately+0.240
keeping+0.226
rier+0.223
Neg logit contributions
 met-0.215
 saw-0.213
 dropped-0.193
 loved-0.193
 took-0.189
Token
Token ID198
Feature activation-7.71e-01

Pos logit contributions
 anymore+0.089
unes+0.082
anks+0.078
uts+0.078
rier+0.077
Neg logit contributions
 saw-0.123
 remembered-0.119
 met-0.118
 knew-0.118
 sw-0.115
TokenSo
Token ID2396
Feature activation-1.63e+00

Pos logit contributions
 anymore+0.034
 This+0.034
".+0.033
.'+0.032
 If+0.032
Neg logit contributions
 remembered-0.027
 forgot-0.026
 blew-0.025
 chased-0.025
 dove-0.024
Token Peter
Token ID5613
Feature activation+9.82e-01

Pos logit contributions
and+0.066
uts+0.060
 If+0.058
.'+0.057
'.+0.055
Neg logit contributions
 remembered-0.076
 forgot-0.076
 sw-0.071
 bit-0.070
 ate-0.069
Token and
Token ID290
Feature activation-1.96e-01

Pos logit contributions
 forgot+0.034
 remembered+0.034
 blew+0.031
 sne+0.030
 ate+0.030
Neg logit contributions
 This-0.047
 If-0.046
'.-0.046
".-0.045
and-0.045
Token at
Token ID379
Feature activation-1.22e+00

Pos logit contributions
 If+0.156
 Today+0.153
uts+0.151
ays+0.148
ula+0.143
Neg logit contributions
 saw-0.156
 knew-0.142
 lost-0.140
 remembered-0.140
 ate-0.139
Token the
Token ID262
Feature activation+6.81e-01

Pos logit contributions
 If+0.056
uts+0.056
and+0.054
'.+0.054
 This+0.054
Neg logit contributions
 wish-0.042
 remembered-0.041
 lost-0.041
 saw-0.041
 sne-0.041
Token farmer
Token ID18739
Feature activation-3.81e+00

Pos logit contributions
 forgot+0.021
 remembered+0.020
 wish+0.019
 sne+0.018
 blew+0.018
Neg logit contributions
 If-0.035
and-0.034
 This-0.034
'.-0.034
".-0.034
Tokenâ
Token ID22940
Feature activation-3.10e-01

Pos logit contributions
ula+0.129
 If+0.128
fulness+0.126
uts+0.119
arl+0.118
Neg logit contributions
 sne-0.181
 ate-0.181
 saw-0.180
 forgot-0.176
 bit-0.175
Token
Token ID26391
Feature activation-2.56e+00

Pos logit contributions
and+0.013
uts+0.013
 If+0.013
".+0.013
 This+0.012
Neg logit contributions
 forgot-0.011
 remembered-0.011
 blew-0.011
 ate-0.010
 sne-0.010
Token
Token ID8151
Feature activation-6.97e+00

Pos logit contributions
 If+0.079
fulness+0.073
ion+0.072
ions+0.071
and+0.067
Neg logit contributions
 forgot-0.148
 saw-0.139
 remembered-0.134
 ate-0.132
 met-0.131
Tokens
Token ID82
Feature activation-1.29e+00

Pos logit contributions
 If+0.242
 itself+0.239
ures+0.234
 This+0.232
ment+0.230
Neg logit contributions
 forgot-0.246
 knew-0.227
 laughed-0.224
 ate-0.222
 had-0.218
Token market
Token ID1910
Feature activation-3.86e+00

Pos logit contributions
 This+0.063
uts+0.061
 If+0.061
'.+0.061
and+0.059
Neg logit contributions
 wish-0.043
 remembered-0.041
 sne-0.040
 forgot-0.038
 saw-0.038
Token.
Token ID13
Feature activation-4.58e+00

Pos logit contributions
 This+0.118
 If+0.115
 Today+0.111
ays+0.106
uts+0.100
Neg logit contributions
 bit-0.196
 remembered-0.190
 l-0.190
 forgot-0.188
 saw-0.186
Token Whenever
Token ID21326
Feature activation-1.17e+00

Pos logit contributions
ings+0.179
ently+0.167
 anymore+0.163
rier+0.157
ays+0.157
Neg logit contributions
 knew-0.172
 forgot-0.160
 ate-0.159
 remembered-0.155
 wished-0.154
Token she
Token ID673
Feature activation+2.52e+00

Pos logit contributions
and+0.047
 If+0.045
.'+0.042
uts+0.042
'.+0.041
Neg logit contributions
 forgot-0.056
 remembered-0.054
 sne-0.050
 dove-0.049
 saw-0.049
Tokenâ
Token ID22940
Feature activation+5.46e-01

Pos logit contributions
ings+0.137
 anymore+0.131
ently+0.129
fulness+0.126
and+0.113
Neg logit contributions
 remembered-0.162
 sne-0.159
 threw-0.152
Uh-0.151
 forgot-0.149
Token
Token ID26391
Feature activation-1.76e+00

Pos logit contributions
 forgot+0.023
 remembered+0.023
 blew+0.022
 ate+0.021
 sne+0.021
Neg logit contributions
and-0.021
 If-0.021
 This-0.021
".-0.021
'.-0.020
Token she
Token ID673
Feature activation+2.82e+00

Pos logit contributions
and+0.060
fulness+0.059
 If+0.058
ion+0.057
ula+0.053
Neg logit contributions
 forgot-0.090
 remembered-0.084
 saw-0.082
 ate-0.077
 chased-0.076
Token said
Token ID531
Feature activation-1.52e+00

Pos logit contributions
 remembered+0.066
 forgot+0.065
 blew+0.059
 dove+0.058
 wish+0.056
Neg logit contributions
".-0.153
'.-0.151
 This-0.149
.'-0.147
 If-0.146
Token.
Token ID13
Feature activation-4.77e+00

Pos logit contributions
and+0.060
.'+0.056
 If+0.055
'.+0.053
.'"+0.052
Neg logit contributions
 sne-0.070
 forgot-0.068
 remembered-0.068
 dropped-0.067
 ate-0.064
Token
Token ID220
Feature activation-6.96e+00

Pos logit contributions
 anymore+0.216
ings+0.195
ently+0.190
fulness+0.188
uts+0.174
Neg logit contributions
 remembered-0.157
 saw-0.156
 switched-0.149
 tried-0.149
 rubbed-0.148
Token
Token ID198
Feature activation-3.32e+00

Pos logit contributions
overs+0.298
keeping+0.284
 anymore+0.281
rier+0.280
rs+0.277
Neg logit contributions
 saw-0.155
 also-0.149
 climbed-0.148
 felt-0.143
 forgot-0.142
Token
Token ID198
Feature activation-4.48e-01

Pos logit contributions
 anymore+0.138
unes+0.123
rier+0.123
anks+0.120
overs+0.119
Neg logit contributions
 saw-0.130
 blew-0.125
lly-0.120
 met-0.120
Quick-0.120
TokenBut
Token ID1537
Feature activation-1.73e-01

Pos logit contributions
 anymore+0.020
 This+0.020
".+0.019
.'+0.019
 If+0.019
Neg logit contributions
 remembered-0.016
 blew-0.016
 forgot-0.015
 chased-0.015
 saw-0.014
Token Peter
Token ID5613
Feature activation+9.25e-01

Pos logit contributions
and+0.007
uts+0.006
 This+0.006
.'+0.006
 If+0.006
Neg logit contributions
 remembered-0.008
 forgot-0.008
 blew-0.008
 sne-0.007
 bit-0.007
Token wasn
Token ID2492
Feature activation+1.58e-01

Pos logit contributions
 forgot+0.023
 remembered+0.022
 blew+0.020
 sne+0.019
 ate+0.018
Neg logit contributions
 This-0.054
 If-0.053
and-0.053
'.-0.053
".-0.053
Token you
Token ID345
Feature activation+1.07e+00

Pos logit contributions
and+0.058
 If+0.055
 This+0.055
.'+0.054
'.+0.054
Neg logit contributions
 forgot-0.068
 remembered-0.068
 dove-0.065
 sne-0.064
 wish-0.063
Token can
Token ID460
Feature activation-6.51e-01

Pos logit contributions
 forgot+0.032
 remembered+0.031
 blew+0.029
 sne+0.028
 wish+0.028
Neg logit contributions
 This-0.055
 If-0.055
'.-0.054
".-0.054
and-0.054
Token answer
Token ID3280
Feature activation-4.83e+00

Pos logit contributions
and+0.038
 If+0.037
 This+0.036
overs+0.034
fulness+0.034
Neg logit contributions
 forgot-0.017
 remembered-0.016
 wish-0.016
 sne-0.016
 dove-0.015
Token my
Token ID616
Feature activation-6.57e-01

Pos logit contributions
uts+0.195
and+0.192
umbs+0.182
ugs+0.179
 This+0.178
Neg logit contributions
 remembered-0.187
 bit-0.176
 wish-0.171
 saw-0.169
 pin-0.169
Token r
Token ID374
Feature activation-1.12e+00

Pos logit contributions
and+0.032
 This+0.031
 If+0.031
uts+0.030
but+0.029
Neg logit contributions
 wish-0.023
 remembered-0.022
 forgot-0.020
 sne-0.020
 bit-0.020
Tokeniddle
Token ID2509
Feature activation-6.96e+00

Pos logit contributions
 This+0.054
'.+0.050
 If+0.049
".+0.048
 anymore+0.048
Neg logit contributions
 remembered-0.033
 believed-0.031
 forgot-0.029
 sne-0.029
 planted-0.029
Tokenâ
Token ID22940
Feature activation-7.46e-01

Pos logit contributions
and+0.240
 This+0.221
 If+0.219
ula+0.200
keeping+0.199
Neg logit contributions
 remembered-0.297
 forgot-0.283
 blew-0.282
 did-0.278
 bit-0.277
Token
Token ID26391
Feature activation-3.46e+00

Pos logit contributions
and+0.033
uts+0.032
 If+0.031
".+0.029
but+0.029
Neg logit contributions
 remembered-0.026
 forgot-0.026
 blew-0.025
 sne-0.025
 ate-0.024
Token.
Token ID13
Feature activation-4.24e+00

Pos logit contributions
 If+0.141
fulness+0.139
and+0.130
ion+0.129
ions+0.126
Neg logit contributions
 forgot-0.138
 remembered-0.128
 dove-0.125
 sne-0.123
 chased-0.122
Token
Token ID220
Feature activation-5.56e+00

Pos logit contributions
 anymore+0.156
ings+0.153
lights+0.138
ently+0.138
eways+0.137
Neg logit contributions
 remembered-0.169
 forgot-0.156
 dove-0.154
 bit-0.153
 wished-0.152
Token
Token ID198
Feature activation-1.71e+00

Pos logit contributions
fulness+0.188
fully+0.178
 anymore+0.177
overs+0.175
lights+0.160
Neg logit contributions
 forgot-0.206
 climbed-0.199
 sne-0.191
 remembered-0.187
 dropped-0.184
Tokenâ
Token ID22940
Feature activation+4.48e-01

Pos logit contributions
ently+0.128
ings+0.127
 anymore+0.123
fulness+0.116
and+0.110
Neg logit contributions
 bit-0.138
 threw-0.133
 ate-0.133
 dove-0.130
 remembered-0.129
Token
Token ID26391
Feature activation-1.09e+00

Pos logit contributions
 forgot+0.019
 remembered+0.018
 blew+0.018
 ate+0.017
 sne+0.017
Neg logit contributions
and-0.017
 If-0.017
 This-0.017
".-0.017
'.-0.017
Token she
Token ID673
Feature activation+2.28e+00

Pos logit contributions
and+0.039
 If+0.038
fulness+0.037
.'+0.035
ion+0.034
Neg logit contributions
 forgot-0.051
 remembered-0.048
 ate-0.047
 saw-0.046
 chased-0.045
Token said
Token ID531
Feature activation-2.11e+00

Pos logit contributions
 remembered+0.062
 forgot+0.062
 blew+0.058
 sne+0.055
 dove+0.053
Neg logit contributions
".-0.119
 This-0.118
'.-0.118
 If-0.116
.'-0.116
Token.
Token ID13
Feature activation-5.16e+00

Pos logit contributions
and+0.092
keeping+0.080
 If+0.079
.'+0.077
ays+0.076
Neg logit contributions
 remembered-0.087
 forgot-0.087
 sne-0.086
 ate-0.086
 dropped-0.085
Token
Token ID220
Feature activation-6.94e+00

Pos logit contributions
ently+0.208
fulness+0.207
 anymore+0.203
ings+0.198
ment+0.182
Neg logit contributions
 threw-0.173
 saw-0.165
 ate-0.164
 sw-0.164
 blew-0.163
Token
Token ID198
Feature activation-3.00e+00

Pos logit contributions
overs+0.302
keeping+0.301
ment+0.290
ingly+0.289
rier+0.286
Neg logit contributions
 saw-0.158
 also-0.154
 climbed-0.143
 did-0.139
 forgot-0.138
Token
Token ID198
Feature activation-8.43e-02

Pos logit contributions
rier+0.118
 anymore+0.114
overs+0.108
ados+0.107
ment+0.103
Neg logit contributions
 saw-0.123
 sw-0.114
 blew-0.114
lly-0.112
 met-0.111
TokenTim
Token ID14967
Feature activation+4.39e-01

Pos logit contributions
 This+0.004
.'+0.004
'.+0.004
 If+0.004
".+0.004
Neg logit contributions
 forgot-0.003
 blew-0.003
 remembered-0.003
 chased-0.003
 threw-0.002
Token grabbed
Token ID13176
Feature activation+6.76e-01

Pos logit contributions
 forgot+0.018
 remembered+0.018
 blew+0.017
 ate+0.016
 sne+0.016
Neg logit contributions
 This-0.019
 If-0.018
'.-0.018
and-0.018
.'-0.018
Token the
Token ID262
Feature activation+8.25e-01

Pos logit contributions
 forgot+0.031
 remembered+0.031
 blew+0.030
 sne+0.030
 dropped+0.029
Neg logit contributions
and-0.023
'.-0.023
.'-0.023
".-0.023
 This-0.023
Token someone
Token ID2130
Feature activation-1.97e+00

Pos logit contributions
and+0.079
uts+0.076
 This+0.070
 If+0.067
'.+0.066
Neg logit contributions
 remembered-0.064
 lost-0.060
 sne-0.060
 blew-0.060
 forgot-0.060
Token without
Token ID1231
Feature activation-2.78e-01

Pos logit contributions
 This+0.075
 Today+0.074
 If+0.073
fulness+0.072
ula+0.070
Neg logit contributions
 forgot-0.095
 bit-0.089
 remembered-0.087
 saw-0.084
 dropped-0.083
Token first
Token ID717
Feature activation-2.96e+00

Pos logit contributions
 This+0.012
and+0.012
'.+0.012
 If+0.012
.'+0.011
Neg logit contributions
 remembered-0.011
 forgot-0.011
 sne-0.011
 blew-0.010
 dropped-0.010
Token asking
Token ID4737
Feature activation-4.96e+00

Pos logit contributions
and+0.088
but+0.072
ions+0.071
 This+0.070
ion+0.067
Neg logit contributions
 forgot-0.175
 remembered-0.171
 dropped-0.161
 blew-0.155
 ate-0.153
Token their
Token ID511
Feature activation-1.25e-01

Pos logit contributions
and+0.152
uts+0.151
ays+0.149
anes+0.134
igsaw+0.132
Neg logit contributions
 remembered-0.226
 knew-0.215
 naughty-0.213
 forgot-0.209
 saw-0.205
Token permission
Token ID7170
Feature activation-6.92e+00

Pos logit contributions
 This+0.007
and+0.007
 If+0.006
'.+0.006
.'+0.006
Neg logit contributions
 forgot-0.004
 remembered-0.004
 wish-0.003
 sne-0.003
 bit-0.003
Token.
Token ID13
Feature activation-3.94e+00

Pos logit contributions
 Grad+0.215
lights+0.192
igsaw+0.192
irs+0.191
irens+0.191
Neg logit contributions
 cut-0.268
 remembered-0.268
 head-0.264
 forgot-0.263
 knew-0.262
Token We
Token ID775
Feature activation+2.23e+00

Pos logit contributions
 anymore+0.143
fulness+0.129
ings+0.128
and+0.123
ently+0.119
Neg logit contributions
 remembered-0.160
 forgot-0.152
 ate-0.151
 sw-0.149
 wished-0.147
Token should
Token ID815
Feature activation+4.81e-01

Pos logit contributions
 forgot+0.073
 remembered+0.071
 blew+0.068
 sne+0.066
 dove+0.064
Neg logit contributions
'.-0.106
".-0.105
.'-0.104
 This-0.104
 If-0.104
Token ask
Token ID1265
Feature activation-3.47e+00

Pos logit contributions
 forgot+0.013
 remembered+0.012
 wish+0.011
 blew+0.011
 sne+0.011
Neg logit contributions
and-0.027
 This-0.027
 If-0.026
'.-0.026
".-0.026
Token the
Token ID262
Feature activation+1.01e+00

Pos logit contributions
uts+0.158
ays+0.147
and+0.144
umbs+0.137
 Today+0.135
Neg logit contributions
 sw-0.123
 remembered-0.121
 sne-0.121
 knew-0.117
 bit-0.115
Token nothing
Token ID2147
Feature activation+4.79e-01

Pos logit contributions
'.+0.024
.'+0.024
".+0.023
 If+0.023
.'"+0.023
Neg logit contributions
 forgot-0.029
 remembered-0.028
 sne-0.027
 saw-0.027
 ate-0.026
Token seemed
Token ID3947
Feature activation-1.19e+00

Pos logit contributions
 forgot+0.013
 remembered+0.012
 blew+0.011
 ate+0.011
 dropped+0.011
Neg logit contributions
 This-0.027
 If-0.027
'.-0.027
.'-0.027
".-0.026
Token quite
Token ID2407
Feature activation-6.84e-01

Pos logit contributions
and+0.042
 If+0.042
 This+0.042
'.+0.040
uts+0.039
Neg logit contributions
 forgot-0.059
 remembered-0.058
 bit-0.055
 lost-0.055
 saw-0.054
Token right
Token ID826
Feature activation-2.94e+00

Pos logit contributions
and+0.036
 This+0.035
 If+0.034
'.+0.034
uts+0.034
Neg logit contributions
 forgot-0.023
 remembered-0.023
 dropped-0.020
 bit-0.020
 blew-0.020
Token.
Token ID13
Feature activation-3.67e+00

Pos logit contributions
 This+0.106
and+0.099
 Everyone+0.094
 If+0.093
uts+0.093
Neg logit contributions
 remembered-0.166
 saw-0.150
 forgot-0.149
 did-0.146
 knew-0.142
Token
Token ID220
Feature activation-6.92e+00

Pos logit contributions
uts+0.154
ment+0.142
fulness+0.140
ings+0.140
rier+0.139
Neg logit contributions
 saw-0.132
 looked-0.118
 observed-0.114
 tried-0.114
 felt-0.112
Token
Token ID198
Feature activation-3.67e+00

Pos logit contributions
keeping+0.304
ados+0.297
ting+0.287
overs+0.284
rs+0.284
Neg logit contributions
 saw-0.188
 felt-0.165
 also-0.160
 knew-0.153
 did-0.152
Token
Token ID198
Feature activation-5.99e-01

Pos logit contributions
unes+0.137
ment+0.136
anks+0.136
ados+0.133
ancing+0.129
Neg logit contributions
 saw-0.154
 met-0.137
 observed-0.130
 also-0.129
 remembered-0.127
TokenThen
Token ID6423
Feature activation-1.50e-01

Pos logit contributions
".+0.027
 This+0.027
uts+0.026
 If+0.026
'.+0.026
Neg logit contributions
 remembered-0.020
 forgot-0.020
 saw-0.019
 observed-0.018
 blew-0.017
Token one
Token ID530
Feature activation+1.78e+00

Pos logit contributions
and+0.006
 If+0.005
uts+0.005
.'+0.005
".+0.005
Neg logit contributions
 remembered-0.007
 forgot-0.007
 saw-0.007
 sne-0.007
 blew-0.007
Token day
Token ID1110
Feature activation+6.00e-01

Pos logit contributions
 forgot+0.078
 remembered+0.075
 blew+0.072
 dropped+0.069
 chased+0.068
Neg logit contributions
 This-0.068
'.-0.065
and-0.065
 If-0.065
.'-0.064
Token He
Token ID679
Feature activation+4.77e+00

Pos logit contributions
and+0.005
.'+0.005
".+0.005
'.+0.005
 If+0.005
Neg logit contributions
 remembered-0.015
 forgot-0.014
 sne-0.014
 dropped-0.014
 ate-0.014
Token looked
Token ID3114
Feature activation+1.90e+00

Pos logit contributions
 groom+0.171
inge+0.157
 bride+0.152
 wedding+0.147
ited+0.146
Neg logit contributions
-0.233
.-0.218
 He-0.213
 She-0.206
 "-0.192
Token up
Token ID510
Feature activation+1.86e-01

Pos logit contributions
 forgot+0.111
 remembered+0.108
 blew+0.106
 sne+0.103
 wish+0.102
Neg logit contributions
 If-0.042
 This-0.042
".-0.041
.'-0.041
'.-0.041
Token and
Token ID290
Feature activation+1.75e+00

Pos logit contributions
 remembered+0.009
 forgot+0.009
 saw+0.009
 sne+0.009
 blew+0.009
Neg logit contributions
 If-0.006
 This-0.006
'.-0.006
".-0.006
and-0.006
Token saw
Token ID2497
Feature activation+7.43e-01

Pos logit contributions
 forgot+0.019
 remembered+0.016
 blew+0.015
 wish+0.014
 sne+0.013
Neg logit contributions
 This-0.121
'.-0.118
 If-0.118
".-0.118
.'-0.117
Token a
Token ID257
Feature activation+3.46e+00

Pos logit contributions
 remembered+0.033
 forgot+0.033
 blew+0.031
 sne+0.031
 dropped+0.030
Neg logit contributions
'.-0.028
.'-0.027
 This-0.027
and-0.027
".-0.027
Token kind
Token ID1611
Feature activation+1.41e+00

Pos logit contributions
 forgot+0.118
 remembered+0.108
 ate+0.104
 swung+0.102
ited+0.102
Neg logit contributions
".-0.159
-0.157
.-0.153
'.-0.150
 This-0.149
Token and
Token ID290
Feature activation+3.20e+00

Pos logit contributions
 forgot+0.045
 remembered+0.043
 blew+0.040
 ate+0.038
 sne+0.038
Neg logit contributions
 This-0.071
".-0.070
'.-0.070
 If-0.070
.'-0.069
Token generous
Token ID14431
Feature activation+2.03e+00

Pos logit contributions
 forgot+0.097
 blew+0.087
 remembered+0.086
ited+0.084
 ate+0.084
Neg logit contributions
-0.147
 If-0.143
 This-0.141
 "-0.139
.-0.139
Token deer
Token ID20096
Feature activation+3.61e-01

Pos logit contributions
 forgot+0.061
 remembered+0.057
 blew+0.053
 ate+0.050
 chased+0.049
Neg logit contributions
 This-0.106
".-0.106
'.-0.105
.'-0.103
 If-0.103
Token named
Token ID3706
Feature activation-1.59e+00

Pos logit contributions
 remembered+0.021
 forgot+0.021
 sne+0.020
 blew+0.020
 sw+0.020
Neg logit contributions
 If-0.009
 This-0.009
and-0.009
'.-0.008
.'-0.008
Token you
Token ID345
Feature activation+3.59e+00

Pos logit contributions
 forgot+0.087
 remembered+0.086
 blew+0.082
 sne+0.081
 ate+0.080
Neg logit contributions
 This-0.038
 If-0.036
".-0.036
'.-0.036
.'-0.034
Token were
Token ID547
Feature activation+2.21e+00

Pos logit contributions
 swoop+0.122
 dove+0.119
 buzz+0.117
 chased+0.114
 ostr+0.113
Neg logit contributions
".-0.164
 This-0.157
'.-0.152
 That-0.146
."-0.145
Token so
Token ID523
Feature activation+3.06e+00

Pos logit contributions
 remembered+0.119
 forgot+0.119
 blew+0.117
 dove+0.115
 wish+0.113
Neg logit contributions
".-0.061
 This-0.060
 If-0.054
'.-0.054
.'-0.053
Token fast
Token ID3049
Feature activation-1.32e+00

Pos logit contributions
 remembered+0.107
 feather+0.096
 dove+0.095
 forgot+0.094
 regretted+0.094
Neg logit contributions
 This-0.150
".-0.146
 If-0.135
and-0.134
."-0.134
Token!"
Token ID2474
Feature activation-4.92e-01

Pos logit contributions
and+0.044
 If+0.041
 This+0.039
fulness+0.035
 Today+0.034
Neg logit contributions
 forgot-0.069
 remembered-0.068
 blew-0.067
 sne-0.066
 bit-0.066
Token Billy
Token ID15890
Feature activation+3.63e+00

Pos logit contributions
and+0.017
.'+0.016
ions+0.016
 If+0.015
".+0.015
Neg logit contributions
 forgot-0.025
 remembered-0.024
 ate-0.023
 blew-0.023
 threw-0.023
Token replied
Token ID8712
Feature activation-5.49e-01

Pos logit contributions
 wish+0.101
Uh+0.085
 dove+0.085
 remembered+0.085
Ta+0.085
Neg logit contributions
.-0.177
-0.176
".-0.169
 This-0.168
'.-0.164
Token "
Token ID366
Feature activation+1.05e+00

Pos logit contributions
.'+0.013
 If+0.013
'.+0.013
and+0.013
".+0.012
Neg logit contributions
 remembered-0.030
 forgot-0.030
 dropped-0.030
 ate-0.030
 sne-0.030
TokenI
Token ID40
Feature activation+3.09e+00

Pos logit contributions
 forgot+0.039
 remembered+0.038
 blew+0.035
 sne+0.034
 wish+0.034
Neg logit contributions
 This-0.048
 If-0.047
'.-0.047
".-0.047
.'-0.046
Token wasn
Token ID2492
Feature activation+5.09e-01

Pos logit contributions
 remembered+0.101
 dove+0.098
 feather+0.097
 chased+0.095
 buzz+0.095
Neg logit contributions
".-0.141
'.-0.136
 This-0.135
.'-0.130
.-0.128
Token't
Token ID470
Feature activation+1.13e+00

Pos logit contributions
 forgot+0.027
 remembered+0.027
 blew+0.026
 sne+0.026
 ate+0.025
Neg logit contributions
and-0.014
 This-0.013
 If-0.013
.'-0.013
'.-0.013
Token it
Token ID340
Feature activation-8.29e-01

Pos logit contributions
'.+0.012
.'+0.012
and+0.012
 If+0.012
 This+0.011
Neg logit contributions
 pin-0.014
 remembered-0.014
 blew-0.014
 sne-0.014
 forgot-0.014
Token,
Token ID11
Feature activation-1.22e-01

Pos logit contributions
and+0.034
 If+0.033
'.+0.031
 This+0.031
.'+0.031
Neg logit contributions
 remembered-0.036
 forgot-0.035
 sne-0.035
 blew-0.034
 ate-0.033
Token Daddy
Token ID26926
Feature activation-1.23e+00

Pos logit contributions
'.+0.005
.'+0.005
and+0.004
 This+0.004
".+0.004
Neg logit contributions
 forgot-0.006
 remembered-0.005
 sne-0.005
 wish-0.005
 blew-0.005
Token?"
Token ID1701
Feature activation-1.08e+00

Pos logit contributions
and+0.061
 This+0.058
 If+0.058
uts+0.055
ula+0.054
Neg logit contributions
 remembered-0.046
 forgot-0.046
 blew-0.042
 saw-0.041
 sne-0.040
Token Her
Token ID2332
Feature activation+1.77e+00

Pos logit contributions
.'+0.045
".+0.044
'.+0.044
ings+0.043
.'"+0.042
Neg logit contributions
 forgot-0.047
 ate-0.046
 threw-0.044
 saw-0.043
 sne-0.043
Token dad
Token ID9955
Feature activation+3.11e+00

Pos logit contributions
 forgot+0.078
 remembered+0.076
 blew+0.075
 buzz+0.075
 chased+0.073
Neg logit contributions
 This-0.062
".-0.060
'.-0.059
.-0.059
.'-0.059
Token replied
Token ID8712
Feature activation-6.07e-01

Pos logit contributions
 forgot+0.090
 wish+0.090
 dove+0.087
 buzz+0.081
 remembered+0.081
Neg logit contributions
.-0.146
 This-0.143
-0.143
'.-0.139
 If-0.136
Token,
Token ID11
Feature activation+1.34e-01

Pos logit contributions
'.+0.020
.'+0.020
 If+0.020
".+0.020
and+0.019
Neg logit contributions
 remembered-0.029
 forgot-0.028
 sne-0.028
 ate-0.028
 dropped-0.027
Token "
Token ID366
Feature activation+2.54e-01

Pos logit contributions
 remembered+0.009
 forgot+0.008
 blew+0.008
 ate+0.008
 sne+0.008
Neg logit contributions
.'-0.003
'.-0.003
and-0.003
 If-0.003
".-0.002
TokenI
Token ID40
Feature activation+2.62e+00

Pos logit contributions
 forgot+0.009
 remembered+0.008
 blew+0.008
 wish+0.008
 bit+0.008
Neg logit contributions
'.-0.011
".-0.011
.'-0.011
 This-0.011
 If-0.011
Token think
Token ID892
Feature activation-2.47e+00

Pos logit contributions
 chased+0.090
 dove+0.089
 forgot+0.088
 remembered+0.088
 buzz+0.086
Neg logit contributions
 This-0.113
'.-0.112
".-0.112
-0.109
.'-0.109
Token the
Token ID262
Feature activation+1.69e+00

Pos logit contributions
and+0.033
uts+0.029
 This+0.029
 If+0.029
'.+0.029
Neg logit contributions
 forgot-0.030
 blew-0.030
 remembered-0.030
 saw-0.028
 dropped-0.028
Token plants
Token ID6134
Feature activation-3.60e-01

Pos logit contributions
 forgot+0.062
 remembered+0.061
 blew+0.056
 chased+0.055
 rocked+0.054
Neg logit contributions
".-0.076
 This-0.076
.'-0.075
'.-0.075
 If-0.073
Token,
Token ID11
Feature activation+5.37e-01

Pos logit contributions
 If+0.012
'.+0.012
and+0.012
 This+0.012
.'+0.012
Neg logit contributions
 forgot-0.018
 remembered-0.018
 blew-0.017
 saw-0.017
 ate-0.017
Token she
Token ID673
Feature activation+3.99e+00

Pos logit contributions
 forgot+0.028
 remembered+0.028
 blew+0.027
 sne+0.026
 dropped+0.026
Neg logit contributions
and-0.017
'.-0.016
.'-0.016
".-0.016
 If-0.016
Token saw
Token ID2497
Feature activation+5.19e-01

Pos logit contributions
 wish+0.099
Ta+0.091
 matter+0.087
 sadd+0.084
Trump+0.082
Neg logit contributions
.-0.206
-0.205
."-0.181
".-0.180
 "-0.177
Token a
Token ID257
Feature activation+3.42e+00

Pos logit contributions
 remembered+0.023
 forgot+0.022
 blew+0.021
 sne+0.021
 dropped+0.021
Neg logit contributions
'.-0.020
.'-0.020
 This-0.020
".-0.020
and-0.020
Token butterfly
Token ID35113
Feature activation+6.45e-01

Pos logit contributions
 forgot+0.113
 remembered+0.108
 rocked+0.102
 chased+0.100
ited+0.100
Neg logit contributions
".-0.155
.-0.149
-0.146
'.-0.146
 This-0.145
Token and
Token ID290
Feature activation+2.77e+00

Pos logit contributions
 forgot+0.031
 remembered+0.031
 blew+0.030
 sne+0.030
 dove+0.029
Neg logit contributions
 If-0.021
 This-0.021
and-0.021
'.-0.021
".-0.020
Token started
Token ID2067
Feature activation+7.68e-01

Pos logit contributions
 forgot+0.061
 wish+0.059
ited+0.059
 remembered+0.059
 rocked+0.054
Neg logit contributions
 This-0.147
".-0.146
 If-0.144
-0.142
 She-0.141
Token chasing
Token ID20023
Feature activation-8.90e-02

Pos logit contributions
 forgot+0.045
 remembered+0.045
 blew+0.043
 sne+0.043
 ate+0.042
Neg logit contributions
 This-0.017
'.-0.017
 If-0.017
.'-0.017
".-0.017
Token it
Token ID340
Feature activation-1.44e+00

Pos logit contributions
 This+0.003
 If+0.003
'.+0.003
and+0.003
".+0.003
Neg logit contributions
 sne-0.004
 buzz-0.004
 remembered-0.004
 forgot-0.004
 bit-0.004
Token.
Token ID13
Feature activation-2.30e+00

Pos logit contributions
uts+0.149
and+0.142
angers+0.131
anks+0.130
ancing+0.130
Neg logit contributions
 spoiled-0.142
 saw-0.140
 bit-0.138
 left-0.137
 ruined-0.136
Token
Token ID220
Feature activation-5.55e+00

Pos logit contributions
ions+0.095
and+0.090
ment+0.090
eway+0.089
fulness+0.088
Neg logit contributions
 saw-0.097
 forgot-0.095
 remembered-0.090
 tried-0.087
 looked-0.086
Token
Token ID198
Feature activation-2.96e+00

Pos logit contributions
keeping+0.218
ment+0.203
overs+0.200
ados+0.199
ions+0.195
Neg logit contributions
 saw-0.203
 forgot-0.186
 knew-0.178
 remembered-0.174
 dropped-0.174
Token
Token ID198
Feature activation+3.24e-01

Pos logit contributions
unes+0.102
ment+0.101
ancing+0.100
ions+0.095
uts+0.095
Neg logit contributions
 saw-0.147
 forgot-0.127
 met-0.124
 remembered-0.123
 knew-0.118
TokenL
Token ID43
Feature activation+1.87e-01

Pos logit contributions
 forgot+0.013
 remembered+0.012
 blew+0.011
 bit+0.011
 saw+0.011
Neg logit contributions
 This-0.014
 If-0.014
".-0.013
'.-0.013
.'-0.013
Tokenily
Token ID813
Feature activation+4.10e+00

Pos logit contributions
 forgot+0.009
 remembered+0.008
 blew+0.008
 dropped+0.008
 ate+0.008
Neg logit contributions
 This-0.007
'.-0.006
".-0.006
.'-0.006
 If-0.006
Token said
Token ID531
Feature activation+4.30e-01

Pos logit contributions
 wish+0.162
 rocked+0.154
Ta+0.151
 groom+0.149
 feather+0.146
Neg logit contributions
-0.165
 This-0.154
 She-0.151
.-0.150
 "-0.144
Token,
Token ID11
Feature activation-2.21e+00

Pos logit contributions
 remembered+0.023
 forgot+0.023
 ate+0.022
 blew+0.022
 dropped+0.022
Neg logit contributions
and-0.012
'.-0.012
 If-0.012
.'-0.012
".-0.012
Token "
Token ID366
Feature activation+1.25e+00

Pos logit contributions
and+0.092
ays+0.092
ings+0.091
keeping+0.089
.'+0.088
Neg logit contributions
 remembered-0.094
 saw-0.091
 knew-0.090
 blew-0.090
 sne-0.089
TokenI
Token ID40
Feature activation+2.84e+00

Pos logit contributions
 forgot+0.049
 remembered+0.047
 blew+0.044
 sne+0.044
 wish+0.043
Neg logit contributions
 This-0.055
'.-0.054
 If-0.054
and-0.053
.'-0.053
Token can
Token ID460
Feature activation+1.12e+00

Pos logit contributions
 blew+0.106
 buzz+0.101
 chased+0.101
 forgot+0.100
 rocked+0.099
Neg logit contributions
 This-0.114
'.-0.114
".-0.112
 If-0.111
.'-0.110
Token said
Token ID531
Feature activation-6.81e-01

Pos logit contributions
 wish+0.137
 flood+0.133
ited+0.128
 delay+0.122
 nightmare+0.122
Neg logit contributions
-0.203
.-0.191
 "-0.181
 He-0.171
 They-0.169
Token,
Token ID11
Feature activation-2.19e+00

Pos logit contributions
.'+0.026
'.+0.025
and+0.024
".+0.023
 If+0.023
Neg logit contributions
 remembered-0.032
 forgot-0.030
 sne-0.030
 sw-0.030
 dropped-0.029
Token "
Token ID366
Feature activation+1.04e+00

Pos logit contributions
.'+0.094
and+0.090
ings+0.087
'.+0.086
keeping+0.085
Neg logit contributions
 remembered-0.096
 saw-0.093
 blew-0.092
 sw-0.092
 knew-0.091
TokenThis
Token ID1212
Feature activation+5.07e-01

Pos logit contributions
 forgot+0.040
 remembered+0.039
 blew+0.037
 sne+0.036
 wish+0.035
Neg logit contributions
 This-0.045
 If-0.044
'.-0.044
".-0.044
.'-0.044
Token is
Token ID318
Feature activation+1.16e-01

Pos logit contributions
 forgot+0.017
 remembered+0.017
 blew+0.016
 sne+0.016
 dove+0.016
Neg logit contributions
and-0.025
 If-0.024
 This-0.024
.'-0.024
'.-0.024
Token my
Token ID616
Feature activation+7.25e-01

Pos logit contributions
 remembered+0.004
 forgot+0.004
 blew+0.004
 sne+0.004
 bit+0.004
Neg logit contributions
and-0.005
'.-0.005
.'-0.005
 This-0.005
 If-0.005
Token weapon
Token ID4282
Feature activation-2.29e+00

Pos logit contributions
 forgot+0.019
 remembered+0.018
 sne+0.017
 blew+0.017
 wish+0.017
Neg logit contributions
 This-0.041
 If-0.041
'.-0.040
and-0.040
.'-0.040
Token.
Token ID13
Feature activation-1.74e+00

Pos logit contributions
 If+0.070
and+0.067
 This+0.062
 Today+0.061
.'+0.060
Neg logit contributions
 forgot-0.123
 bit-0.121
 remembered-0.118
 ate-0.117
 cut-0.115
Token I
Token ID314
Feature activation+2.55e+00

Pos logit contributions
and+0.054
 anymore+0.048
.'+0.048
ings+0.046
ently+0.046
Neg logit contributions
 wish-0.084
 sne-0.083
 remembered-0.081
 threw-0.081
 cut-0.081
Token can
Token ID460
Feature activation+1.78e+00

Pos logit contributions
 forgot+0.092
 remembered+0.090
 chased+0.085
 blew+0.084
 dove+0.084
Neg logit contributions
".-0.109
'.-0.109
 This-0.109
.'-0.106
 If-0.106
Token use
Token ID779
Feature activation-1.94e+00

Pos logit contributions
 forgot+0.061
 remembered+0.060
 blew+0.054
 regretted+0.052
 ate+0.051
Neg logit contributions
 This-0.086
".-0.085
'.-0.084
 If-0.083
.'-0.083
Token.
Token ID13
Feature activation-2.75e+00

Pos logit contributions
 If+0.117
and+0.102
 This+0.101
fulness+0.099
ion+0.098
Neg logit contributions
 sne-0.144
 forgot-0.140
 remembered-0.139
 hurt-0.136
 saw-0.135
Token
Token ID220
Feature activation-4.88e+00

Pos logit contributions
ions+0.101
rier+0.095
fulness+0.094
rily+0.090
ays+0.089
Neg logit contributions
 knew-0.121
 forgot-0.121
 remembered-0.120
 saw-0.119
 observed-0.114
Token
Token ID198
Feature activation-2.76e+00

Pos logit contributions
fulness+0.197
overs+0.187
reating+0.182
rier+0.177
ions+0.175
Neg logit contributions
 forgot-0.169
 saw-0.162
 climbed-0.156
 dropped-0.155
 felt-0.150
Token
Token ID198
Feature activation+5.91e-01

Pos logit contributions
rier+0.084
overs+0.082
ays+0.082
 anymore+0.081
".+0.079
Neg logit contributions
 knew-0.131
 saw-0.129
 forgot-0.126
 remembered-0.125
 dropped-0.120
TokenSarah
Token ID29284
Feature activation+2.76e+00

Pos logit contributions
 forgot+0.027
 remembered+0.026
 blew+0.024
 wish+0.024
 chased+0.024
Neg logit contributions
 This-0.022
".-0.022
 If-0.022
'.-0.021
.'-0.021
Token was
Token ID373
Feature activation+1.48e+00

Pos logit contributions
 forgot+0.070
 wish+0.069
 remembered+0.066
 blew+0.065
 buzz+0.063
Neg logit contributions
'.-0.139
".-0.137
 This-0.135
.'-0.134
 If-0.134
Token upset
Token ID9247
Feature activation-2.42e+00

Pos logit contributions
 forgot+0.051
 remembered+0.049
 blew+0.047
 sne+0.044
 wish+0.044
Neg logit contributions
 This-0.069
 If-0.069
'.-0.068
".-0.068
.'-0.067
Token with
Token ID351
Feature activation-7.09e-01

Pos logit contributions
keeping+0.110
and+0.108
 This+0.108
 If+0.107
.'"+0.105
Neg logit contributions
 forgot-0.087
 remembered-0.086
 knew-0.082
 bit-0.081
 felt-0.080
Token Mary
Token ID5335
Feature activation-1.13e+00

Pos logit contributions
and+0.030
".+0.029
uts+0.028
'.+0.028
.'+0.028
Neg logit contributions
 remembered-0.029
 forgot-0.029
 sne-0.028
 dropped-0.027
 wish-0.026
Token's
Token ID338
Feature activation+8.14e-01

Pos logit contributions
 This+0.042
 If+0.040
uts+0.039
and+0.039
".+0.039
Neg logit contributions
 forgot-0.054
 remembered-0.054
 bit-0.051
 sne-0.049
 ate-0.048
Token decision
Token ID2551
Feature activation-3.97e+00

Pos logit contributions
 forgot+0.023
 remembered+0.023
 sne+0.021
 wish+0.021
 blew+0.020
Neg logit contributions
 This-0.044
 If-0.043
'.-0.043
".-0.043
.'-0.043
Token house
Token ID2156
Feature activation-1.63e+00

Pos logit contributions
 forgot+0.066
 remembered+0.064
 blew+0.059
 wish+0.056
 ate+0.056
Neg logit contributions
 This-0.080
".-0.080
 If-0.079
'.-0.079
.'-0.078
Token.
Token ID13
Feature activation-1.07e+00

Pos logit contributions
 This+0.043
'.+0.042
.'+0.039
fulness+0.039
 If+0.039
Neg logit contributions
 bit-0.094
 forgot-0.089
 saw-0.089
 sne-0.089
 cut-0.087
Token Lily
Token ID20037
Feature activation+3.58e+00

Pos logit contributions
.'+0.039
'.+0.036
fulness+0.035
and+0.035
ions+0.034
Neg logit contributions
 forgot-0.053
 saw-0.052
 dropped-0.052
 threw-0.051
 ate-0.050
Token saw
Token ID2497
Feature activation+5.54e-01

Pos logit contributions
 wish+0.152
 matter+0.141
Ta+0.140
ited+0.137
 dream+0.129
Neg logit contributions
-0.134
.-0.134
and-0.121
".-0.119
."-0.115
Token the
Token ID262
Feature activation+2.41e+00

Pos logit contributions
 forgot+0.028
 remembered+0.028
 blew+0.027
 dropped+0.027
 sne+0.027
Neg logit contributions
'.-0.018
.'-0.018
".-0.017
 This-0.017
and-0.017
Token dog
Token ID3290
Feature activation+1.04e+00

Pos logit contributions
 forgot+0.090
 remembered+0.088
 believed+0.077
 swung+0.076
 rocked+0.076
Neg logit contributions
".-0.104
 This-0.101
.-0.101
 If-0.100
.'-0.097
Token and
Token ID290
Feature activation+2.36e+00

Pos logit contributions
 forgot+0.054
 remembered+0.053
 blew+0.051
 sne+0.051
 ate+0.050
Neg logit contributions
 This-0.031
'.-0.030
 If-0.030
".-0.030
.'-0.030
Token screamed
Token ID25421
Feature activation-1.21e+00

Pos logit contributions
 wish+0.080
 forgot+0.073
ited+0.071
 remembered+0.070
Ta+0.069
Neg logit contributions
 This-0.102
-0.102
 If-0.102
.-0.101
 She-0.097
Token too
Token ID1165
Feature activation-4.84e-01

Pos logit contributions
and+0.036
keeping+0.036
 If+0.036
'.+0.035
 This+0.035
Neg logit contributions
 sne-0.065
 sw-0.060
 dropped-0.059
 remembered-0.059
 bit-0.058
Token.
Token ID13
Feature activation-1.97e+00

Pos logit contributions
 This+0.011
'.+0.010
 If+0.010
.'+0.010
and+0.009
Neg logit contributions
 forgot-0.029
 sne-0.028
 hurt-0.028
 remembered-0.028
 wish-0.028
Token She
Token ID1375
Feature activation+4.58e+00

Pos logit contributions
fulness+0.076
.'+0.072
and+0.071
ions+0.070
'.+0.067
Neg logit contributions
 saw-0.095
 dropped-0.092
 forgot-0.089
 threw-0.088
 chased-0.087
Token go
Token ID467
Feature activation-1.95e+00

Pos logit contributions
 If+0.062
 This+0.060
".+0.059
and+0.059
'.+0.058
Neg logit contributions
 sne-0.043
 wish-0.041
 forgot-0.041
 remembered-0.041
 buzz-0.039
Token.
Token ID13
Feature activation-3.61e+00

Pos logit contributions
 This+0.104
 If+0.101
and+0.099
uts+0.099
'.+0.098
Neg logit contributions
 zoom-0.058
 hurt-0.057
 hit-0.057
 saw-0.057
 swoop-0.057
Token Soon
Token ID15894
Feature activation+3.43e-01

Pos logit contributions
.'+0.134
".+0.122
 anymore+0.121
fulness+0.120
ings+0.118
Neg logit contributions
 zoom-0.141
 remembered-0.139
 knew-0.138
 climbed-0.137
 rubbed-0.137
Token,
Token ID11
Feature activation+1.93e-02

Pos logit contributions
 forgot+0.019
 remembered+0.019
 sne+0.017
 blew+0.017
 dropped+0.017
Neg logit contributions
.'-0.011
'.-0.010
 If-0.010
and-0.010
".-0.010
Token the
Token ID262
Feature activation+2.55e+00

Pos logit contributions
 remembered+0.001
 forgot+0.001
 blew+0.001
 sne+0.001
 dove+0.001
Neg logit contributions
.'-0.001
'.-0.001
".-0.001
and-0.001
 If-0.001
Token wizard
Token ID18731
Feature activation+1.99e+00

Pos logit contributions
 forgot+0.082
 remembered+0.076
 ate+0.071
 blew+0.071
 chased+0.069
Neg logit contributions
".-0.120
'.-0.119
 This-0.118
.'-0.117
 If-0.116
Token returned
Token ID4504
Feature activation-2.25e+00

Pos logit contributions
 forgot+0.045
 remembered+0.044
 blew+0.038
 wish+0.038
 sne+0.038
Neg logit contributions
 This-0.112
".-0.112
'.-0.112
.'-0.110
 If-0.110
Token and
Token ID290
Feature activation+1.66e+00

Pos logit contributions
 Today+0.095
 This+0.093
 These+0.093
'.+0.093
 If+0.088
Neg logit contributions
 remembered-0.096
 saw-0.096
 lost-0.092
 dropped-0.088
 sc-0.087
Token said
Token ID531
Feature activation-1.53e+00

Pos logit contributions
 forgot+0.069
 remembered+0.068
 blew+0.063
 wish+0.063
 sne+0.063
Neg logit contributions
 This-0.064
 If-0.063
".-0.062
and-0.062
'.-0.061
Token,
Token ID11
Feature activation-2.93e+00

Pos logit contributions
.'+0.059
'.+0.058
and+0.057
".+0.053
 Today+0.053
Neg logit contributions
 remembered-0.068
 sw-0.065
 sne-0.064
 dropped-0.064
 ate-0.063
Token "
Token ID366
Feature activation+2.86e-01

Pos logit contributions
and+0.113
ions+0.105
ays+0.105
.'+0.105
 anymore+0.104
Neg logit contributions
 remembered-0.135
 blew-0.131
 knew-0.129
 sw-0.125
 sne-0.125
Token and
Token ID290
Feature activation+2.21e+00

Pos logit contributions
 If+0.052
".+0.052
'.+0.052
uts+0.052
and+0.051
Neg logit contributions
 sne-0.054
 cut-0.053
 ate-0.052
 forgot-0.051
 saw-0.050
Token dad
Token ID9955
Feature activation-1.12e+00

Pos logit contributions
 forgot+0.080
 remembered+0.078
 buzz+0.075
 blew+0.075
 bit+0.075
Neg logit contributions
 This-0.089
".-0.087
 If-0.086
'.-0.084
.'-0.084
Token.
Token ID13
Feature activation-2.80e+00

Pos logit contributions
and+0.032
".+0.031
 If+0.031
uts+0.030
'.+0.030
Neg logit contributions
 forgot-0.058
 sne-0.057
 remembered-0.057
 ate-0.057
 cut-0.055
Token He
Token ID679
Feature activation+3.94e+00

Pos logit contributions
.'+0.117
ently+0.114
fulness+0.113
ing+0.111
'.+0.110
Neg logit contributions
 cut-0.089
 joke-0.084
 spit-0.084
 bark-0.084
 played-0.083
Token always
Token ID1464
Feature activation+3.34e+00

Pos logit contributions
Ta+0.131
Uh+0.122
ited+0.119
 wave+0.113
 se+0.112
Neg logit contributions
.-0.176
-0.171
 They-0.159
 This-0.158
".-0.156
Token did
Token ID750
Feature activation+1.08e+00

Pos logit contributions
 dove+0.060
 wish+0.059
Ta+0.057
 forgot+0.053
Uh+0.052
Neg logit contributions
 This-0.184
".-0.182
.-0.180
 If-0.178
 They-0.176
Token what
Token ID644
Feature activation-2.53e-02

Pos logit contributions
 forgot+0.049
 remembered+0.048
 blew+0.045
 sne+0.045
 wish+0.044
Neg logit contributions
 This-0.040
'.-0.039
 If-0.039
".-0.039
.'-0.039
Token he
Token ID339
Feature activation+3.67e+00

Pos logit contributions
.'+0.001
'.+0.001
 If+0.001
and+0.001
 This+0.001
Neg logit contributions
 forgot-0.001
 remembered-0.001
 sne-0.001
 blew-0.001
 ate-0.001
Token wanted
Token ID2227
Feature activation-2.21e+00

Pos logit contributions
 groom+0.083
Uh+0.082
 forgot+0.077
 se+0.075
 boo+0.075
Neg logit contributions
.-0.191
-0.185
".-0.184
 This-0.181
'.-0.176
Token to
Token ID284
Feature activation+3.85e-01

Pos logit contributions
'.+0.089
.'+0.087
.'"+0.086
 If+0.086
ays+0.084
Neg logit contributions
 saw-0.088
 forgot-0.088
 sne-0.087
 knew-0.087
 remembered-0.086
Token do
Token ID466
Feature activation-2.94e+00

Pos logit contributions
 forgot+0.007
 wish+0.006
 remembered+0.006
 sne+0.006
 blew+0.006
Neg logit contributions
 If-0.025
and-0.025
 This-0.024
'.-0.024
.'-0.024
Token also
Token ID635
Feature activation+5.84e-01

Pos logit contributions
 forgot+0.048
 remembered+0.047
 blew+0.044
 sne+0.043
 ate+0.043
Neg logit contributions
 This-0.042
'.-0.042
 If-0.042
".-0.041
.'-0.041
Token some
Token ID617
Feature activation+1.61e+00

Pos logit contributions
 forgot+0.022
 remembered+0.022
 blew+0.021
 sne+0.020
 ate+0.020
Neg logit contributions
'.-0.026
 This-0.026
.'-0.026
 If-0.026
and-0.025
Token furry
Token ID40857
Feature activation+6.25e-01

Pos logit contributions
 forgot+0.054
 remembered+0.053
 blew+0.049
 sne+0.046
 wish+0.046
Neg logit contributions
 This-0.076
".-0.076
 If-0.075
'.-0.075
and-0.074
Token animals
Token ID4695
Feature activation-1.18e+00

Pos logit contributions
 forgot+0.019
 remembered+0.019
 sne+0.018
 blew+0.018
 dove+0.017
Neg logit contributions
 This-0.032
 If-0.032
'.-0.032
.'-0.031
and-0.031
Token drinking
Token ID7722
Feature activation-1.97e+00

Pos logit contributions
 If+0.041
 This+0.038
'.+0.037
and+0.036
.'+0.035
Neg logit contributions
 forgot-0.060
 ate-0.060
 saw-0.059
 remembered-0.059
 dove-0.059
Token from
Token ID422
Feature activation-1.14e+00

Pos logit contributions
 If+0.082
 This+0.080
and+0.078
'.+0.076
 By+0.076
Neg logit contributions
 sw-0.086
 c-0.079
 ate-0.078
 dove-0.077
 bit-0.077
Token the
Token ID262
Feature activation+1.02e+00

Pos logit contributions
and+0.056
uts+0.053
 If+0.052
 This+0.052
'.+0.049
Neg logit contributions
 sw-0.045
 dove-0.044
 sne-0.043
 saw-0.043
 remembered-0.042
Token pond
Token ID16723
Feature activation-3.50e+00

Pos logit contributions
 forgot+0.029
 remembered+0.028
 blew+0.025
 sne+0.025
 wish+0.024
Neg logit contributions
 This-0.055
 If-0.054
'.-0.054
and-0.054
".-0.054
Token.
Token ID13
Feature activation-3.66e+00

Pos logit contributions
 If+0.120
 This+0.115
ula+0.109
fulness+0.107
and+0.105
Neg logit contributions
 ate-0.174
 sw-0.172
 bit-0.171
 saw-0.169
 smelled-0.164
Token They
Token ID1119
Feature activation+3.25e+00

Pos logit contributions
ently+0.164
fulness+0.159
ancing+0.155
fully+0.152
.'+0.149
Neg logit contributions
 saw-0.125
 sne-0.115
 dove-0.112
 sw-0.111
 threw-0.111
Token looked
Token ID3114
Feature activation+1.32e+00

Pos logit contributions
 wish+0.073
 boo+0.073
Ta+0.071
 forgot+0.069
 scrape+0.066
Neg logit contributions
.-0.163
-0.159
 They-0.152
".-0.150
 She-0.150
Token it
Token ID340
Feature activation-2.27e+00

Pos logit contributions
and+0.096
uts+0.095
 Today+0.095
but+0.090
 This+0.089
Neg logit contributions
 sw-0.094
 dove-0.091
 lost-0.086
 dropped-0.086
 buried-0.086
Token.
Token ID13
Feature activation-2.89e+00

Pos logit contributions
 If+0.071
 This+0.071
and+0.060
'.+0.059
 Everyone+0.059
Neg logit contributions
 remembered-0.122
 forgot-0.120
 buried-0.120
 lost-0.118
 saw-0.117
Token She
Token ID1375
Feature activation+3.48e+00

Pos logit contributions
fulness+0.128
ently+0.121
ancing+0.117
ings+0.113
ors+0.111
Neg logit contributions
 saw-0.110
 wished-0.099
 forgot-0.096
 wondered-0.094
 missed-0.094
Token liked
Token ID8288
Feature activation-5.59e-01

Pos logit contributions
 wish+0.079
Ta+0.077
 forgot+0.076
ited+0.075
 itch+0.073
Neg logit contributions
-0.170
.-0.170
".-0.164
'.-0.160
 If-0.159
Token the
Token ID262
Feature activation+1.81e+00

Pos logit contributions
'.+0.023
.'+0.022
".+0.021
 If+0.020
and+0.020
Neg logit contributions
 sne-0.025
 sw-0.024
 blew-0.024
 remembered-0.024
 forgot-0.024
Token boat
Token ID8848
Feature activation-9.82e-01

Pos logit contributions
 forgot+0.055
 remembered+0.053
 blew+0.047
 ate+0.046
 chased+0.045
Neg logit contributions
".-0.093
 This-0.092
'.-0.091
.'-0.091
 If-0.090
Token and
Token ID290
Feature activation+5.00e-01

Pos logit contributions
 If+0.036
 This+0.034
'.+0.034
and+0.033
.'+0.033
Neg logit contributions
 forgot-0.047
 remembered-0.046
 bit-0.046
 ate-0.045
 sne-0.045
Token took
Token ID1718
Feature activation-1.05e-01

Pos logit contributions
 forgot+0.017
 remembered+0.017
 blew+0.015
 sne+0.015
 ate+0.015
Neg logit contributions
'.-0.025
 This-0.025
.'-0.025
 If-0.025
and-0.025
Token it
Token ID340
Feature activation-1.35e+00

Pos logit contributions
and+0.004
'.+0.004
.'+0.004
".+0.004
 This+0.004
Neg logit contributions
 remembered-0.005
 forgot-0.004
 dropped-0.004
 sne-0.004
 wish-0.004
Token home
Token ID1363
Feature activation-1.78e+00

Pos logit contributions
'.+0.055
 This+0.055
 If+0.053
.'+0.052
 You+0.052
Neg logit contributions
 bit-0.053
 sw-0.052
 lost-0.052
 buried-0.052
 dove-0.052
Token with
Token ID351
Feature activation-2.66e-01

Pos logit contributions
 If+0.069
 This+0.065
 You+0.063
 Today+0.062
 That+0.061
Neg logit contributions
 bit-0.086
 forgot-0.080
 c-0.080
 decorated-0.078
 ate-0.078
Token,
Token ID11
Feature activation-3.27e+00

Pos logit contributions
ions+0.113
and+0.111
keeping+0.109
 If+0.105
umbs+0.104
Neg logit contributions
 sne-0.143
 dropped-0.141
 sw-0.139
 remembered-0.136
 saw-0.130
Token
Token ID6184
Feature activation-3.91e-01

Pos logit contributions
and+0.126
ions+0.118
 anymore+0.117
ays+0.111
.'+0.104
Neg logit contributions
 remembered-0.155
 blew-0.154
 sw-0.154
 sne-0.151
 forgot-0.150
Token
Token ID95
Feature activation-7.31e-01

Pos logit contributions
and+0.017
 If+0.016
.'+0.015
".+0.014
 This+0.014
Neg logit contributions
 forgot-0.015
 sne-0.015
 blew-0.014
 remembered-0.014
 dove-0.014
Token
Token ID26391
Feature activation-1.78e+00

Pos logit contributions
and+0.028
 If+0.026
uts+0.025
.'+0.025
 anymore+0.024
Neg logit contributions
 forgot-0.030
 remembered-0.030
 sne-0.028
 blew-0.028
 ate-0.028
Token
Token ID129
Feature activation-4.81e-01

Pos logit contributions
and+0.073
 If+0.072
ion+0.070
fulness+0.068
ions+0.065
Neg logit contributions
 forgot-0.080
 remembered-0.076
 saw-0.071
 sne-0.070
 blew-0.067
Token
Token ID241
Feature activation-2.28e+00

Pos logit contributions
and+0.021
 If+0.020
.'+0.020
'.+0.020
ions+0.019
Neg logit contributions
 forgot-0.018
 remembered-0.017
 sne-0.017
 dropped-0.016
 blew-0.016
TokenI
Token ID40
Feature activation+1.06e+00

Pos logit contributions
 anymore+0.102
 If+0.100
ings+0.100
 This+0.100
ummy+0.099
Neg logit contributions
 remembered-0.076
 forgot-0.075
 blew-0.073
 wished-0.072
 knew-0.072
Tokenâ
Token ID22940
Feature activation-7.41e-01

Pos logit contributions
 forgot+0.034
 remembered+0.033
 blew+0.030
 sne+0.030
 wish+0.029
Neg logit contributions
 This-0.053
 If-0.052
'.-0.052
".-0.052
.'-0.051
Token
Token ID26391
Feature activation-2.35e+00

Pos logit contributions
and+0.031
 If+0.030
uts+0.030
 anymore+0.029
.'+0.027
Neg logit contributions
 forgot-0.028
 blew-0.027
 remembered-0.026
 ate-0.026
 rocked-0.026
Token
Token ID8151
Feature activation-3.83e+00

Pos logit contributions
and+0.086
 If+0.086
fulness+0.086
ion+0.085
 itself+0.081
Neg logit contributions
 forgot-0.113
 remembered-0.108
 saw-0.105
 sne-0.103
 ate-0.098
Tokenm
Token ID76
Feature activation-1.63e+00

Pos logit contributions
 itself+0.154
ures+0.145
 If+0.145
 Him+0.143
 This+0.141
Neg logit contributions
 forgot-0.152
 remembered-0.135
 saw-0.131
 laughed-0.131
 bit-0.131
Token them
Token ID606
Feature activation-1.08e+00

Pos logit contributions
and+0.085
uts+0.081
 If+0.080
 By+0.079
 This+0.078
Neg logit contributions
 sne-0.076
 sw-0.076
 saw-0.076
 dove-0.075
 se-0.074
Token?"
Token ID1701
Feature activation-1.47e+00

Pos logit contributions
and+0.051
 If+0.050
 This+0.048
'.+0.046
.'+0.045
Neg logit contributions
 sne-0.040
 remembered-0.038
 blew-0.038
 cut-0.037
 saw-0.036
Token Mia
Token ID32189
Feature activation+7.78e-01

Pos logit contributions
and+0.065
.'+0.060
fulness+0.060
'.+0.059
".+0.059
Neg logit contributions
 sne-0.067
 saw-0.065
 remembered-0.062
 buzz-0.061
 dove-0.060
Token asked
Token ID1965
Feature activation-2.69e+00

Pos logit contributions
 forgot+0.018
 remembered+0.017
 sne+0.016
 blew+0.015
 ate+0.014
Neg logit contributions
 This-0.047
'.-0.046
 If-0.046
".-0.046
and-0.046
Token.
Token ID13
Feature activation-2.73e+00

Pos logit contributions
 This+0.093
and+0.092
uts+0.089
 By+0.088
 If+0.087
Neg logit contributions
 sne-0.138
 remembered-0.128
 saw-0.124
 sh-0.123
 ch-0.121
Token
Token ID198
Feature activation-2.76e+00

Pos logit contributions
keeping+0.118
fulness+0.114
and+0.109
.'+0.104
ently+0.102
Neg logit contributions
 saw-0.128
 sne-0.119
 dove-0.119
 buzz-0.116
 pet-0.114
Token
Token ID198
Feature activation+2.24e-01

Pos logit contributions
rier+0.095
fulness+0.095
keeping+0.095
overs+0.092
anks+0.092
Neg logit contributions
 saw-0.125
 dove-0.112
 knew-0.110
 sne-0.109
 hurt-0.109
Token"
Token ID1
Feature activation+1.39e+00

Pos logit contributions
 remembered+0.011
 forgot+0.011
 blew+0.011
 dove+0.011
 sne+0.010
Neg logit contributions
 This-0.007
and-0.007
 If-0.007
.'-0.007
'.-0.007
TokenNo
Token ID2949
Feature activation-5.87e-01

Pos logit contributions
 forgot+0.055
 remembered+0.054
 sne+0.050
 blew+0.049
 wish+0.049
Neg logit contributions
 This-0.060
 If-0.059
and-0.058
'.-0.058
".-0.057
Token,
Token ID11
Feature activation+9.94e-01

Pos logit contributions
and+0.025
but+0.022
'.+0.022
.'+0.022
 If+0.021
Neg logit contributions
 sne-0.027
 forgot-0.025
 remembered-0.025
 wish-0.025
 sw-0.025
Token Mia
Token ID32189
Feature activation-8.18e-01

Pos logit contributions
 forgot+0.052
 remembered+0.052
 blew+0.049
 sne+0.049
 wish+0.048
Neg logit contributions
 This-0.029
'.-0.029
 If-0.029
.'-0.028
".-0.028
Token other
Token ID584
Feature activation-1.75e+00

Pos logit contributions
 forgot+0.042
 remembered+0.041
 blew+0.038
 sne+0.037
 ate+0.036
Neg logit contributions
 This-0.060
 If-0.060
".-0.059
'.-0.059
and-0.059
Token and
Token ID290
Feature activation+1.49e+00

Pos logit contributions
 If+0.063
 This+0.060
and+0.055
'.+0.055
ions+0.054
Neg logit contributions
 sne-0.084
 wish-0.082
 c-0.080
 ch-0.079
 forgot-0.079
Token point
Token ID966
Feature activation-1.95e+00

Pos logit contributions
 forgot+0.044
 remembered+0.043
 blew+0.040
 sne+0.038
 dropped+0.037
Neg logit contributions
 This-0.076
 If-0.075
'.-0.075
".-0.075
.'-0.074
Token at
Token ID379
Feature activation-1.58e-02

Pos logit contributions
 This+0.080
and+0.078
 By+0.077
'.+0.077
 If+0.076
Neg logit contributions
 remembered-0.083
 sne-0.082
 saw-0.081
 sw-0.080
 dropped-0.079
Token things
Token ID1243
Feature activation-2.49e+00

Pos logit contributions
 This+0.001
'.+0.001
and+0.001
.'+0.001
".+0.001
Neg logit contributions
 remembered-0.001
 forgot-0.001
 sne-0.001
 dove-0.001
 blew-0.001
Token.
Token ID13
Feature activation-2.16e+00

Pos logit contributions
 This+0.079
 If+0.075
and+0.075
'.+0.072
 Today+0.069
Neg logit contributions
 blew-0.125
 fl-0.125
 hurt-0.124
 spark-0.122
 dove-0.122
Token
Token ID198
Feature activation-1.32e+00

Pos logit contributions
.'+0.090
ently+0.090
and+0.087
fulness+0.087
ying+0.083
Neg logit contributions
 sw-0.085
 wished-0.081
 saw-0.080
 missed-0.080
 sprayed-0.078
Token
Token ID198
Feature activation+7.26e-01

Pos logit contributions
 anymore+0.035
 This+0.034
unes+0.033
'.+0.033
.'+0.032
Neg logit contributions
 saw-0.071
 remembered-0.068
 blew-0.067
 forgot-0.067
 wish-0.067
Token"
Token ID1
Feature activation+1.63e+00

Pos logit contributions
 forgot+0.038
 remembered+0.037
 blew+0.036
 dove+0.034
 sne+0.034
Neg logit contributions
 This-0.022
'.-0.022
".-0.021
 If-0.021
.'-0.021
TokenLook
Token ID8567
Feature activation+2.40e-03

Pos logit contributions
 forgot+0.067
 remembered+0.063
 wish+0.060
 sne+0.060
 dropped+0.059
Neg logit contributions
 This-0.067
 If-0.067
'.-0.066
".-0.066
and-0.065
Token,
Token ID11
Feature activation+1.31e+00

Pos logit contributions
 forgot+0.000
 remembered+0.000
 sne+0.000
 wish+0.000
 dove+0.000
Neg logit contributions
and-0.000
 If-0.000
 This-0.000
'.-0.000
.'-0.000
Token,
Token ID11
Feature activation+5.81e-01

Pos logit contributions
 This+0.045
 If+0.043
'.+0.039
 You+0.037
.'+0.036
Neg logit contributions
 forgot-0.053
 bit-0.052
 remembered-0.052
 spark-0.048
 smelled-0.048
Token but
Token ID475
Feature activation+1.19e+00

Pos logit contributions
 forgot+0.029
 remembered+0.029
 blew+0.028
 sne+0.027
 ate+0.027
Neg logit contributions
'.-0.019
 This-0.019
 If-0.019
.'-0.019
".-0.018
Token still
Token ID991
Feature activation-1.82e-01

Pos logit contributions
 forgot+0.062
 remembered+0.061
 blew+0.059
 sne+0.058
 wish+0.057
Neg logit contributions
 This-0.035
 If-0.034
'.-0.034
".-0.034
.'-0.034
Token beautiful
Token ID4950
Feature activation-2.09e-02

Pos logit contributions
 If+0.009
and+0.009
 This+0.009
'.+0.008
.'+0.008
Neg logit contributions
 forgot-0.007
 remembered-0.007
 sne-0.006
 blew-0.006
 ate-0.006
Token.
Token ID13
Feature activation-2.62e+00

Pos logit contributions
 If+0.001
 This+0.001
and+0.001
'.+0.001
.'+0.001
Neg logit contributions
 forgot-0.001
 remembered-0.001
 sne-0.001
 blew-0.001
 bit-0.001
Token
Token ID220
Feature activation-4.87e+00

Pos logit contributions
.'+0.103
rier+0.101
 anymore+0.101
".+0.098
fulness+0.098
Neg logit contributions
 knew-0.092
 felt-0.091
 forgot-0.090
 saw-0.088
 remembered-0.087
Token
Token ID198
Feature activation-1.26e+00

Pos logit contributions
fulness+0.196
 lately+0.180
keeping+0.180
overs+0.177
reating+0.176
Neg logit contributions
 forgot-0.155
 saw-0.153
 felt-0.152
 climbed-0.148
 dropped-0.148
Token
Token ID198
Feature activation+1.00e+00

Pos logit contributions
 anymore+0.032
.'+0.032
'.+0.031
".+0.030
overs+0.028
Neg logit contributions
 forgot-0.067
 saw-0.066
 knew-0.065
 remembered-0.065
 blew-0.064
TokenFrom
Token ID4863
Feature activation+6.67e-01

Pos logit contributions
 forgot+0.045
 remembered+0.044
 blew+0.041
 sne+0.041
 wish+0.040
Neg logit contributions
 This-0.038
'.-0.037
 If-0.037
".-0.037
.'-0.037
Token then
Token ID788
Feature activation-4.55e-01

Pos logit contributions
 forgot+0.029
 remembered+0.028
 blew+0.027
 dove+0.026
 sne+0.026
Neg logit contributions
 This-0.026
and-0.026
 If-0.026
'.-0.026
".-0.025
Token on
Token ID319
Feature activation-9.64e-01

Pos logit contributions
and+0.019
.'+0.018
 If+0.018
".+0.018
'.+0.018
Neg logit contributions
 remembered-0.019
 blew-0.019
 forgot-0.018
 sne-0.018
 ate-0.018
Token as
Token ID355
Feature activation-2.29e+00

Pos logit contributions
 This+0.160
 Today+0.150
 If+0.147
 That+0.135
 Those+0.127
Neg logit contributions
 forgot-0.190
 remembered-0.189
 bit-0.189
 blew-0.183
 fl-0.181
Token part
Token ID636
Feature activation-4.52e+00

Pos logit contributions
 This+0.097
 If+0.094
 Today+0.093
'.+0.092
and+0.091
Neg logit contributions
 forgot-0.093
 remembered-0.090
 blew-0.089
 sw-0.086
 dropped-0.086
Token of
Token ID286
Feature activation-2.21e+00

Pos logit contributions
 This+0.149
 Today+0.146
 If+0.136
'.+0.134
 That+0.131
Neg logit contributions
 forgot-0.201
 blew-0.200
 bit-0.197
 pin-0.192
 sw-0.191
Token their
Token ID511
Feature activation+6.88e-01

Pos logit contributions
and+0.093
 This+0.091
 If+0.089
 Today+0.088
'.+0.087
Neg logit contributions
 sw-0.088
 sne-0.087
 remembered-0.085
 pin-0.085
 forgot-0.084
Token game
Token ID983
Feature activation-4.89e+00

Pos logit contributions
 forgot+0.022
 remembered+0.022
 sne+0.020
 blew+0.020
 wish+0.020
Neg logit contributions
 This-0.034
 If-0.034
'.-0.034
".-0.033
.'-0.033
Token.
Token ID13
Feature activation-4.27e+00

Pos logit contributions
 This+0.144
 If+0.144
 Today+0.127
 That+0.124
'.+0.117
Neg logit contributions
 forgot-0.256
 bit-0.246
 remembered-0.240
 saw-0.236
 cut-0.235
Token<|endoftext|>
Token ID50256
Feature activation-4.52e+00

Pos logit contributions
rier+0.170
ings+0.151
ently+0.151
fulness+0.150
fully+0.141
Neg logit contributions
 forgot-0.161
 knew-0.160
 played-0.156
 ate-0.155
 remembered-0.155
TokenOnce
Token ID7454
Feature activation-1.85e+00

Pos logit contributions
rier+0.212
fulness+0.199
but+0.191
uts+0.191
ancing+0.187
Neg logit contributions
 saw-0.178
 met-0.167
 had-0.154
 did-0.152
 looked-0.148
Token upon
Token ID2402
Feature activation+2.06e-01

Pos logit contributions
'.+0.089
.'+0.085
".+0.085
and+0.084
 This+0.084
Neg logit contributions
 forgot-0.068
 saw-0.066
 remembered-0.065
 knew-0.058
 lost-0.058
Token a
Token ID257
Feature activation+1.86e+00

Pos logit contributions
 remembered+0.008
 forgot+0.008
 sne+0.008
 wish+0.007
 blew+0.007
Neg logit contributions
 This-0.009
'.-0.009
 If-0.008
.'-0.008
".-0.008
Token time
Token ID640
Feature activation-1.82e+00

Pos logit contributions
 forgot+0.085
 remembered+0.084
 blew+0.082
 wished+0.077
 wish+0.077
Neg logit contributions
 This-0.065
.'-0.064
'.-0.064
and-0.064
".-0.064
Token much
Token ID881
Feature activation-3.51e-02

Pos logit contributions
 forgot+0.000
 remembered+0.000
 blew+0.000
 dove+0.000
 sne+0.000
Neg logit contributions
 This-0.000
'.-0.000
.'-0.000
 If-0.000
and-0.000
Token as
Token ID355
Feature activation-7.80e-01

Pos logit contributions
and+0.001
'.+0.001
.'+0.001
 If+0.001
 This+0.001
Neg logit contributions
 forgot-0.002
 blew-0.002
 ate-0.002
 sne-0.002
 remembered-0.001
Token he
Token ID339
Feature activation+2.30e+00

Pos logit contributions
 This+0.034
'.+0.033
.'+0.033
 If+0.033
.'"+0.032
Neg logit contributions
 forgot-0.032
 remembered-0.030
 dove-0.030
 blew-0.029
 sne-0.029
Token wanted
Token ID2227
Feature activation-2.22e+00

Pos logit contributions
 forgot+0.069
 remembered+0.066
 wish+0.065
 blew+0.062
 sne+0.059
Neg logit contributions
".-0.106
'.-0.106
and-0.105
 This-0.105
.'-0.103
Token.
Token ID13
Feature activation-2.30e+00

Pos logit contributions
 If+0.086
arl+0.084
 Today+0.084
.'"+0.083
 This+0.081
Neg logit contributions
 sne-0.095
 cut-0.094
 saw-0.088
 dropped-0.088
 sw-0.088
Token
Token ID220
Feature activation-5.03e+00

Pos logit contributions
fulness+0.093
reating+0.089
 anymore+0.088
.'"+0.087
.'+0.086
Neg logit contributions
 saw-0.084
 tried-0.083
 felt-0.082
 ate-0.081
 che-0.081
Token
Token ID198
Feature activation-2.28e+00

Pos logit contributions
fulness+0.195
overs+0.191
keeping+0.191
rs+0.191
reating+0.189
Neg logit contributions
 climbed-0.162
 saw-0.161
 dropped-0.158
 forgot-0.154
 felt-0.150
Token
Token ID198
Feature activation+5.57e-01

Pos logit contributions
 anymore+0.079
ays+0.070
overs+0.070
unes+0.069
.'+0.068
Neg logit contributions
 saw-0.105
 forgot-0.103
 knew-0.101
 remembered-0.100
 climbed-0.098
TokenThe
Token ID464
Feature activation+2.36e+00

Pos logit contributions
 forgot+0.024
 remembered+0.023
 blew+0.022
 chased+0.021
 sne+0.021
Neg logit contributions
 This-0.022
.'-0.021
'.-0.021
".-0.021
 If-0.021
Token pig
Token ID12967
Feature activation+2.56e+00

Pos logit contributions
 forgot+0.088
 remembered+0.079
 swung+0.071
 dropped+0.071
 believed+0.070
Neg logit contributions
".-0.106
.-0.102
 If-0.101
 This-0.100
-0.100
Token quickly
Token ID2952
Feature activation+2.77e+00

Pos logit contributions
 forgot+0.061
 wish+0.059
 remembered+0.058
 blew+0.054
 sne+0.053
Neg logit contributions
".-0.135
 This-0.134
'.-0.133
 If-0.131
and-0.131
Token way
Token ID835
Feature activation-2.86e+00

Pos logit contributions
 forgot+0.007
 remembered+0.007
 wish+0.007
 sne+0.006
 dove+0.006
Neg logit contributions
 This-0.013
 If-0.013
'.-0.013
and-0.013
.'-0.013
Token inside
Token ID2641
Feature activation-3.97e+00

Pos logit contributions
 If+0.119
uts+0.110
and+0.109
 This+0.107
 Those+0.107
Neg logit contributions
 lost-0.118
 saw-0.111
 wave-0.110
 hurt-0.109
 zoom-0.109
Token his
Token ID465
Feature activation+2.35e-01

Pos logit contributions
uts+0.150
 If+0.144
and+0.143
arl+0.139
arling+0.138
Neg logit contributions
 saw-0.187
 sne-0.172
 remembered-0.171
 cut-0.166
 made-0.166
Token new
Token ID649
Feature activation-2.91e-01

Pos logit contributions
 remembered+0.007
 forgot+0.007
 wish+0.007
 sne+0.006
 dove+0.006
Neg logit contributions
 If-0.013
 This-0.012
and-0.012
'.-0.012
".-0.012
Token created
Token ID2727
Feature activation-2.88e+00

Pos logit contributions
 If+0.016
 This+0.016
and+0.016
uts+0.015
'.+0.015
Neg logit contributions
 remembered-0.008
 wish-0.008
 sne-0.008
 dove-0.008
 forgot-0.008
Token tent
Token ID11105
Feature activation-4.05e+00

Pos logit contributions
 If+0.096
 This+0.086
uts+0.084
 Today+0.082
arl+0.079
Neg logit contributions
 sne-0.145
 buried-0.142
 spark-0.141
 saw-0.140
 sw-0.138
Token.
Token ID13
Feature activation-3.08e+00

Pos logit contributions
 If+0.132
 This+0.114
and+0.111
ula+0.111
 Those+0.104
Neg logit contributions
 forgot-0.205
 saw-0.202
 buried-0.202
 spark-0.201
 bit-0.200
Token It
Token ID632
Feature activation+2.43e+00

Pos logit contributions
rier+0.124
 anymore+0.117
ula+0.108
fulness+0.107
uts+0.105
Neg logit contributions
 knew-0.123
 saw-0.117
 observed-0.116
 wish-0.111
 wonder-0.109
Token was
Token ID373
Feature activation+2.25e-01

Pos logit contributions
 forgot+0.076
 remembered+0.075
 wish+0.070
 dove+0.069
 blew+0.069
Neg logit contributions
".-0.111
 This-0.110
'.-0.108
 If-0.108
.'-0.105
Token the
Token ID262
Feature activation+1.20e+00

Pos logit contributions
 remembered+0.009
 forgot+0.009
 blew+0.008
 sne+0.008
 ate+0.008
Neg logit contributions
and-0.010
'.-0.010
 This-0.010
.'-0.010
 If-0.010
Token perfect
Token ID2818
Feature activation+3.08e-01

Pos logit contributions
 forgot+0.040
 remembered+0.039
 blew+0.036
 sne+0.035
 ate+0.035
Neg logit contributions
 This-0.058
'.-0.057
 If-0.057
".-0.057
.'-0.057
Token returned
Token ID4504
Feature activation-2.56e+00

Pos logit contributions
 forgot+0.039
 remembered+0.038
 blew+0.034
 sne+0.033
 wish+0.033
Neg logit contributions
 This-0.085
 If-0.083
'.-0.083
".-0.083
.'-0.082
Token to
Token ID284
Feature activation-4.27e-01

Pos logit contributions
 Today+0.126
 If+0.121
 This+0.120
 These+0.119
 Those+0.118
Neg logit contributions
 remembered-0.083
 lost-0.083
 saw-0.082
 hurt-0.081
 pet-0.080
Token her
Token ID607
Feature activation-4.55e-02

Pos logit contributions
and+0.018
 If+0.018
uts+0.018
'.+0.017
.'+0.017
Neg logit contributions
 remembered-0.018
 sne-0.017
 forgot-0.017
 wish-0.017
 dropped-0.016
Token own
Token ID898
Feature activation-1.07e+00

Pos logit contributions
 If+0.002
 This+0.002
and+0.002
'.+0.002
".+0.002
Neg logit contributions
 remembered-0.001
 forgot-0.001
 wish-0.001
 sne-0.001
 bit-0.001
Token hive
Token ID35881
Feature activation-3.67e+00

Pos logit contributions
 If+0.055
 This+0.054
and+0.053
uts+0.052
fulness+0.051
Neg logit contributions
 wish-0.034
 bit-0.031
 remembered-0.030
 sne-0.030
 forgot-0.030
Token.
Token ID13
Feature activation-3.99e+00

Pos logit contributions
 If+0.120
 This+0.115
 That+0.102
 You+0.102
 These+0.101
Neg logit contributions
 forgot-0.186
 bit-0.186
 remembered-0.184
 ate-0.181
 saw-0.179
Token
Token ID198
Feature activation-2.05e+00

Pos logit contributions
ings+0.153
 anymore+0.148
fulness+0.147
rier+0.141
ently+0.136
Neg logit contributions
 forgot-0.152
 remembered-0.150
 ate-0.147
 wished-0.147
 saw-0.146
Token
Token ID198
Feature activation-3.79e-01

Pos logit contributions
 anymore+0.079
unes+0.074
rier+0.072
anks+0.071
".+0.069
Neg logit contributions
 saw-0.083
 remembered-0.081
 knew-0.080
 wished-0.077
 forgot-0.077
TokenThe
Token ID464
Feature activation+2.02e+00

Pos logit contributions
 anymore+0.016
 This+0.016
".+0.016
.'+0.016
 If+0.015
Neg logit contributions
 remembered-0.014
 forgot-0.014
 blew-0.013
 chased-0.013
 dove-0.012
Token bee
Token ID20697
Feature activation+2.20e+00

Pos logit contributions
 forgot+0.065
 remembered+0.061
 blew+0.058
 dropped+0.055
 sne+0.054
Neg logit contributions
".-0.098
'.-0.098
.'-0.098
 If-0.097
 This-0.097
Token had
Token ID550
Feature activation+1.69e+00

Pos logit contributions
 forgot+0.057
 remembered+0.054
 blew+0.050
 wish+0.049
 sne+0.049
Neg logit contributions
'.-0.116
 This-0.116
".-0.116
.'-0.115
and-0.113