TOP ACTIVATIONS
MAX = 10.145

the handle of the st roller so we won 't get
down next to the st roller . They saw Leo 's
until she got her st roller and was ready to start
happily hopped off the st roller and ran towards the playground
pushing Sammy in the st roller , and they were having
her standing near the st roller waving at him . He
fell out of her st roller . When they
She puts on her p aj amas . She hugs her
pushing Tommy in his st roller . Suddenly , Tommy saw
Ben and Lily to the hospital . Ben has
have to go to the hospital . Mom and Dad are
have to go back to heaven ," said the angel .
Sara was taken to the hospital . She had a bad
put him in his st roller and they went to the
happy . They thanked the vet and praised him for his
Ben and Lily to the hospital . Ben had to get
and took her to the hospital . They were angry and
and takes Tim to the hospital . Tim has a big
Ben and Sam to the hospital . They had to stay
Tim and Kim to the hospital . They threw away the

MIN ACTIVATIONS
MIN = -13.462

for some funny little sun fl owers . Her m ama
, â said T od . â
stretch out and enormous sun fl owers would show their bright
replied his mum . T od nodded and agreed to put
, there was a do od le who wanted to make
thing , John . N od if you 're tired ,
<|endoftext|> L ily loved sun fl owers . She liked to
a garden full of wild fl owers . He decided to
the biggest and brightest sun fl owers she could find .
The next day , T od and his mom opened the
garden and admire his sun fl owers . They also see
da is ies , sun fl owers , and tul ips
El od ie . El od ie loved to think .
think . El od ie would think about all
so excited to start do od ling with it .
back with lots of sun fl owers to fill the prince
other . Jim and J od ie learned that working together
<|endoftext|> Sam was picking wild fl owers in the me adow
a gifted girl called El od ie . El od ie
But what El od ie liked to think about

INTERVAL 4.243 - 10.145
CONTAINS 2.081%

!" And they hugged and filled each other with love .
a great job cleaning the garage . Tim my felt proud
you can ! We can start by getting a more comfortable
do you want some of my snacks ? I have a
, he saw the terrible mess . His mar bles were

INTERVAL -1.659 - 4.243
CONTAINS 82.491%

She knows she is strong and happy . She
cr ay ons she had forgotten about . She wanted to
but the small square box had a hard time rolling .
eating , Lily 's mom suggested they go for a walk
and even ignored Tim my when he said he was still

INTERVAL -7.561 - -1.659
CONTAINS 15.305%

grew bigger . Anna was amazed and full of joy .
the rabbit said to the sheep , " I think your
day at the park , playing tag and chasing each other
. She saw a big bear and thought it was cute
it anywhere . The creatures in the jar were gone ,

INTERVAL -13.462 - -7.561
CONTAINS 0.123%

excited and began to count how many patterns were on the
flew in circles , swoop ing and diving . Then ,
the hook and set Bub bles free . Bub bles was
son . This lovely muff in isn 't for us ,
like bright yellow neck l aces and flower crown s .
Token the
Token ID262
Feature activation+2.33e+00

Pos logit contributions
ling+0.019
led+0.016
ie+0.015
 shining+0.015
 around+0.015
Neg logit contributions
roller-0.012
 salon-0.010
ara-0.010
post-0.010
atters-0.009
Token handle
Token ID5412
Feature activation+1.98e+00

Pos logit contributions
roller+0.112
 salon+0.106
ara+0.099
 pitcher+0.096
 heaven+0.096
Neg logit contributions
ling-0.119
 around-0.106
 shining-0.103
led-0.102
 up-0.096
Token of
Token ID286
Feature activation+1.92e+00

Pos logit contributions
roller+0.117
 salon+0.111
ara+0.107
 heaven+0.107
 clinic+0.105
Neg logit contributions
ling-0.078
 around-0.070
 shining-0.066
led-0.062
 up-0.061
Token the
Token ID262
Feature activation+1.21e-01

Pos logit contributions
roller+0.124
 salon+0.119
ara+0.113
 heaven+0.111
 pitcher+0.111
Neg logit contributions
ling-0.067
 around-0.055
 shining-0.053
led-0.051
 up-0.047
Token st
Token ID336
Feature activation+1.17e+00

Pos logit contributions
roller+0.005
 salon+0.005
ara+0.005
 heaven+0.005
 pitcher+0.005
Neg logit contributions
ling-0.007
 around-0.006
 shining-0.006
led-0.006
 up-0.005
Tokenroller
Token ID10646
Feature activation+9.08e+00

Pos logit contributions
 heaven+0.036
 salon+0.033
ara+0.032
S+0.030
Anna+0.030
Neg logit contributions
ling-0.076
 around-0.069
 up-0.065
led-0.063
 shining-0.063
Token so
Token ID523
Feature activation-5.56e-01

Pos logit contributions
roller+0.547
 salon+0.520
ara+0.503
 heaven+0.490
atters+0.487
Neg logit contributions
ling-0.355
 around-0.309
 shining-0.295
led-0.276
 up-0.272
Token we
Token ID356
Feature activation+1.76e+00

Pos logit contributions
ling+0.021
 around+0.018
 shining+0.017
led+0.016
 up+0.015
Neg logit contributions
roller-0.034
 salon-0.033
ara-0.031
 heaven-0.030
 pitcher-0.030
Token won
Token ID1839
Feature activation+2.43e+00

Pos logit contributions
roller+0.090
 salon+0.086
ara+0.081
atters+0.079
 heaven+0.079
Neg logit contributions
ling-0.083
 shining-0.071
 around-0.070
led-0.069
 up-0.065
Token't
Token ID470
Feature activation+9.56e-01

Pos logit contributions
roller+0.128
 salon+0.121
ara+0.115
 heaven+0.111
atters+0.111
Neg logit contributions
ling-0.113
 around-0.099
 shining-0.097
led-0.093
 up-0.090
Token get
Token ID651
Feature activation+1.42e+00

Pos logit contributions
roller+0.049
 salon+0.046
ara+0.044
post+0.042
 heaven+0.042
Neg logit contributions
ling-0.046
 around-0.041
 shining-0.040
 up-0.037
 him-0.037
Token down
Token ID866
Feature activation-3.24e+00

Pos logit contributions
ling+0.087
 shining+0.066
bee+0.066
 mattered+0.062
itive+0.061
Neg logit contributions
roller-0.184
 salon-0.181
ara-0.175
ress-0.170
post-0.166
Token next
Token ID1306
Feature activation-2.45e-01

Pos logit contributions
ling+0.127
 around+0.105
 shining+0.105
led+0.100
ie+0.096
Neg logit contributions
roller-0.193
 salon-0.184
ara-0.176
 heaven-0.171
post-0.171
Token to
Token ID284
Feature activation+4.96e-01

Pos logit contributions
ling+0.009
 around+0.009
 shining+0.008
 up+0.008
 in+0.007
Neg logit contributions
roller-0.015
atters-0.015
 salon-0.014
 pitcher-0.014
ara-0.014
Token the
Token ID262
Feature activation+4.23e-03

Pos logit contributions
 heaven+0.029
ara+0.026
 salon+0.026
 clinic+0.026
 vet+0.026
Neg logit contributions
ling-0.021
led-0.019
 shining-0.016
itive-0.016
b-0.015
Token st
Token ID336
Feature activation+7.74e-01

Pos logit contributions
roller+0.000
 salon+0.000
ara+0.000
 heaven+0.000
 clinic+0.000
Neg logit contributions
ling-0.000
 around-0.000
 shining-0.000
led-0.000
 up-0.000
Tokenroller
Token ID10646
Feature activation+8.92e+00

Pos logit contributions
Ble+0.019
Y+0.018
 pitcher+0.016
Uh+0.016
 pic+0.016
Neg logit contributions
ling-0.049
 up-0.046
 around-0.045
 curiously-0.044
ing-0.043
Token.
Token ID13
Feature activation+1.39e-01

Pos logit contributions
roller+0.559
 salon+0.524
ara+0.514
atters+0.506
 pitcher+0.501
Neg logit contributions
ling-0.331
 around-0.294
 shining-0.269
 up-0.258
 in-0.245
Token They
Token ID1119
Feature activation+2.99e+00

Pos logit contributions
 salon+0.007
roller+0.006
lor+0.006
ird+0.006
ress+0.006
Neg logit contributions
 around-0.007
 up-0.006
 him-0.006
ling-0.006
 curiously-0.006
Token saw
Token ID2497
Feature activation-5.51e-01

Pos logit contributions
roller+0.166
 salon+0.161
ara+0.151
 heaven+0.149
 clinic+0.147
Neg logit contributions
ling-0.131
 shining-0.110
 around-0.110
led-0.104
 up-0.100
Token Leo
Token ID19632
Feature activation-3.21e+00

Pos logit contributions
ling+0.027
 shining+0.022
ie+0.022
itive+0.022
led+0.021
Neg logit contributions
 salon-0.025
 heaven-0.025
 vet-0.024
 clinic-0.023
ara-0.023
Token's
Token ID338
Feature activation+1.34e+00

Pos logit contributions
ling+0.132
led+0.120
 around+0.112
itive+0.110
 shining+0.108
Neg logit contributions
 salon-0.160
roller-0.158
 heaven-0.155
 clinic-0.155
ara-0.149
Token until
Token ID1566
Feature activation-1.01e-01

Pos logit contributions
roller+0.079
 salon+0.077
post+0.072
 heaven+0.071
 clinic+0.071
Neg logit contributions
ling-0.054
 shining-0.045
 around-0.044
led-0.044
ie-0.042
Token she
Token ID673
Feature activation-1.48e-01

Pos logit contributions
ling+0.003
 around+0.002
 shining+0.002
led+0.002
ie+0.002
Neg logit contributions
roller-0.007
 salon-0.007
ara-0.006
 heaven-0.006
 clinic-0.006
Token got
Token ID1392
Feature activation+5.40e-01

Pos logit contributions
ling+0.006
 around+0.005
 shining+0.004
led+0.004
 up+0.004
Neg logit contributions
roller-0.009
 salon-0.009
atters-0.008
ara-0.008
 heaven-0.008
Token her
Token ID607
Feature activation+1.99e-01

Pos logit contributions
roller+0.030
 salon+0.029
 heaven+0.028
 pitcher+0.028
 clinic+0.026
Neg logit contributions
ling-0.023
bee-0.019
itive-0.018
led-0.018
 shining-0.017
Token st
Token ID336
Feature activation+6.43e-01

Pos logit contributions
roller+0.010
 salon+0.010
 pitcher+0.009
 clinic+0.009
 heaven+0.009
Neg logit contributions
ling-0.010
led-0.009
 around-0.008
 shining-0.008
ie-0.008
Tokenroller
Token ID10646
Feature activation+8.86e+00

Pos logit contributions
 heaven+0.018
 salon+0.017
ara+0.016
 bidder+0.015
Anna+0.014
Neg logit contributions
ling-0.043
 around-0.039
 up-0.039
led-0.037
 shining-0.037
Token and
Token ID290
Feature activation-7.01e-01

Pos logit contributions
roller+0.598
 salon+0.573
ara+0.550
 heaven+0.538
atters+0.536
Neg logit contributions
ling-0.282
 around-0.233
 shining-0.225
led-0.208
 up-0.199
Token was
Token ID373
Feature activation+7.47e-02

Pos logit contributions
ling+0.028
 around+0.024
 shining+0.023
led+0.022
 up+0.021
Neg logit contributions
roller-0.042
 salon-0.040
ara-0.038
atters-0.037
 heaven-0.037
Token ready
Token ID3492
Feature activation+3.03e-01

Pos logit contributions
roller+0.005
 salon+0.004
ara+0.004
 heaven+0.004
atters+0.004
Neg logit contributions
ling-0.003
 around-0.002
 shining-0.002
led-0.002
 up-0.002
Token to
Token ID284
Feature activation+2.45e-01

Pos logit contributions
roller+0.019
 salon+0.019
ara+0.018
 heaven+0.017
atters+0.017
Neg logit contributions
ling-0.011
 around-0.009
 shining-0.009
led-0.008
 up-0.008
Token start
Token ID923
Feature activation+5.67e+00

Pos logit contributions
roller+0.014
 salon+0.013
ara+0.013
post+0.013
atters+0.012
Neg logit contributions
ling-0.010
 around-0.009
 shining-0.009
 up-0.009
 him-0.008
Token happily
Token ID18177
Feature activation+1.39e+00

Pos logit contributions
ling+0.139
 around+0.118
 shining+0.114
led+0.109
 up+0.104
Neg logit contributions
roller-0.222
 salon-0.212
ara-0.201
 heaven-0.197
atters-0.197
Token hopped
Token ID45230
Feature activation-2.13e+00

Pos logit contributions
roller+0.083
 salon+0.078
 heaven+0.075
ara+0.074
 clinic+0.073
Neg logit contributions
ling-0.057
 around-0.051
 shining-0.047
led-0.046
 up-0.044
Token off
Token ID572
Feature activation+2.84e+00

Pos logit contributions
ling+0.062
 around+0.052
 shining+0.048
led+0.044
 up+0.043
Neg logit contributions
roller-0.151
 salon-0.145
ara-0.139
 heaven-0.137
atters-0.136
Token the
Token ID262
Feature activation-5.22e-01

Pos logit contributions
roller+0.198
 salon+0.190
ara+0.182
atters+0.178
 heaven+0.178
Neg logit contributions
ling-0.085
 around-0.070
 shining-0.067
led-0.061
 up-0.059
Token st
Token ID336
Feature activation+6.68e-01

Pos logit contributions
ling+0.031
 around+0.028
 shining+0.028
 up+0.026
led+0.026
Neg logit contributions
roller-0.021
 salon-0.019
 heaven-0.018
ara-0.018
atters-0.018
Tokenroller
Token ID10646
Feature activation+8.86e+00

Pos logit contributions
 heaven+0.018
Ble+0.016
prise+0.016
Uh+0.015
Gu+0.015
Neg logit contributions
ling-0.040
 around-0.039
 up-0.038
bles-0.035
ing-0.035
Token and
Token ID290
Feature activation-3.21e-01

Pos logit contributions
roller+0.547
ara+0.520
 salon+0.519
atters+0.500
 heaven+0.499
Neg logit contributions
ling-0.329
 around-0.299
 shining-0.271
 up-0.260
 in-0.245
Token ran
Token ID4966
Feature activation+8.68e-01

Pos logit contributions
ling+0.013
 around+0.011
 shining+0.011
led+0.010
 up+0.010
Neg logit contributions
roller-0.019
 salon-0.018
ara-0.017
 heaven-0.017
atters-0.017
Token towards
Token ID3371
Feature activation-8.48e-01

Pos logit contributions
roller+0.062
 salon+0.061
ara+0.058
 clinic+0.057
 heaven+0.057
Neg logit contributions
ling-0.024
 around-0.023
 up-0.019
 shining-0.018
 in-0.017
Token the
Token ID262
Feature activation-4.07e-01

Pos logit contributions
ling+0.028
 shining+0.022
led+0.021
 around+0.021
ie+0.018
Neg logit contributions
roller-0.057
 salon-0.055
 heaven-0.053
ara-0.052
 clinic-0.051
Token playground
Token ID24817
Feature activation+4.57e+00

Pos logit contributions
ling+0.022
 around+0.021
 shining+0.020
 up+0.019
 in+0.018
Neg logit contributions
roller-0.018
 heaven-0.016
lor-0.016
atters-0.016
ara-0.016
Token pushing
Token ID7796
Feature activation-2.42e+00

Pos logit contributions
roller+0.001
 salon+0.001
ara+0.001
atters+0.001
 heaven+0.001
Neg logit contributions
ling-0.001
 around-0.001
 shining-0.001
led-0.000
 up-0.000
Token Sammy
Token ID44067
Feature activation-6.79e+00

Pos logit contributions
ling+0.096
 shining+0.084
itive+0.081
led+0.079
 atop+0.075
Neg logit contributions
roller-0.125
ara-0.122
S-0.121
 heaven-0.118
 jail-0.116
Token in
Token ID287
Feature activation-1.91e+00

Pos logit contributions
ling+0.244
led+0.230
itive+0.214
ie+0.207
 shining+0.205
Neg logit contributions
roller-0.391
 salon-0.375
stand-0.330
 clinic-0.328
 pitcher-0.328
Token the
Token ID262
Feature activation-6.04e-01

Pos logit contributions
ling+0.044
 around+0.038
 shining+0.035
 him+0.031
 up+0.031
Neg logit contributions
roller-0.145
 salon-0.139
ara-0.135
atters-0.133
 pitcher-0.130
Token st
Token ID336
Feature activation+6.79e-01

Pos logit contributions
ling+0.034
 around+0.033
 up+0.032
 shining+0.032
 in+0.029
Neg logit contributions
roller-0.025
ress-0.024
lor-0.024
post-0.023
 salon-0.023
Tokenroller
Token ID10646
Feature activation+8.76e+00

Pos logit contributions
 heaven+0.016
jar+0.014
Ble+0.014
 anniversary+0.013
Y+0.013
Neg logit contributions
ling-0.043
 around-0.042
 all-0.040
 up-0.040
uses-0.039
Token,
Token ID11
Feature activation-2.95e-01

Pos logit contributions
roller+0.561
ara+0.534
 salon+0.532
 pitcher+0.515
atters+0.513
Neg logit contributions
ling-0.308
 around-0.280
 shining-0.240
 up-0.240
ie-0.228
Token and
Token ID290
Feature activation-5.23e-01

Pos logit contributions
ling+0.012
led+0.010
ie+0.010
 around+0.009
 shining+0.009
Neg logit contributions
roller-0.016
 salon-0.016
ara-0.015
post-0.015
 pitcher-0.015
Token they
Token ID484
Feature activation+4.87e-01

Pos logit contributions
ling+0.021
led+0.017
itive+0.016
 around+0.016
 shining+0.016
Neg logit contributions
roller-0.026
ara-0.025
 pitcher-0.025
 salon-0.025
S-0.024
Token were
Token ID547
Feature activation-1.09e+00

Pos logit contributions
 salon+0.024
roller+0.023
ird+0.021
ara+0.020
post+0.020
Neg logit contributions
ling-0.023
led-0.019
 shining-0.019
 spear-0.018
itive-0.018
Token having
Token ID1719
Feature activation-4.68e+00

Pos logit contributions
ling+0.036
led+0.028
ie+0.027
 shining+0.026
 around+0.026
Neg logit contributions
roller-0.071
 salon-0.068
 heaven-0.064
 clinic-0.064
atters-0.063
Token her
Token ID607
Feature activation+3.02e-01

Pos logit contributions
ling+0.026
led+0.021
 shining+0.019
ie+0.018
bee+0.018
Neg logit contributions
 salon-0.037
roller-0.035
 vet-0.034
 heaven-0.034
 jail-0.034
Token standing
Token ID5055
Feature activation-5.14e+00

Pos logit contributions
roller+0.016
 salon+0.016
 clinic+0.014
 pitcher+0.014
ara+0.014
Neg logit contributions
ling-0.014
led-0.012
 shining-0.011
 around-0.011
ie-0.011
Token near
Token ID1474
Feature activation-1.51e+00

Pos logit contributions
ling+0.152
 shining+0.120
 around+0.118
led+0.114
ie+0.109
Neg logit contributions
roller-0.350
 salon-0.343
ara-0.327
 pitcher-0.318
post-0.313
Token the
Token ID262
Feature activation-3.63e-01

Pos logit contributions
ling+0.047
 around+0.038
 shining+0.037
led+0.035
 up+0.032
Neg logit contributions
roller-0.103
 salon-0.099
ara-0.095
 heaven-0.093
 clinic-0.092
Token st
Token ID336
Feature activation+6.65e-01

Pos logit contributions
ling+0.019
 around+0.018
 shining+0.018
 up+0.017
 in+0.017
Neg logit contributions
roller-0.016
lor-0.015
ress-0.015
 heaven-0.015
ara-0.014
Tokenroller
Token ID10646
Feature activation+8.76e+00

Pos logit contributions
 heaven+0.018
prise+0.016
 anniversary+0.016
des+0.014
Ble+0.014
Neg logit contributions
ling-0.040
 around-0.040
 up-0.039
 all-0.037
 through-0.036
Token waving
Token ID25849
Feature activation-3.11e+00

Pos logit contributions
roller+0.532
 salon+0.500
ara+0.500
 heaven+0.489
atters+0.488
Neg logit contributions
ling-0.334
 around-0.300
 shining-0.278
 up-0.264
ie-0.251
Token at
Token ID379
Feature activation-1.69e+00

Pos logit contributions
ling+0.111
 shining+0.091
 around+0.089
led+0.085
ie+0.082
Neg logit contributions
roller-0.195
 salon-0.185
ara-0.178
post-0.173
 pitcher-0.173
Token him
Token ID683
Feature activation-4.54e+00

Pos logit contributions
ling+0.073
led+0.061
 shining+0.057
 around+0.052
uses+0.052
Neg logit contributions
 heaven-0.088
roller-0.086
 salon-0.085
ara-0.083
 vet-0.083
Token.
Token ID13
Feature activation-1.39e-01

Pos logit contributions
ling+0.203
led+0.177
 shining+0.174
 around+0.166
ie+0.161
Neg logit contributions
roller-0.239
 salon-0.226
 heaven-0.209
 clinic-0.207
ara-0.207
Token He
Token ID679
Feature activation-7.90e-01

Pos logit contributions
ling+0.007
 around+0.007
 shining+0.006
led+0.006
ie+0.006
Neg logit contributions
roller-0.007
 salon-0.006
ara-0.006
 heaven-0.006
 clinic-0.005
Token fell
Token ID3214
Feature activation+2.71e+00

Pos logit contributions
ling+0.195
 around+0.166
 shining+0.162
led+0.156
 up+0.147
Neg logit contributions
roller-0.275
 salon-0.264
ara-0.250
 clinic-0.242
 heaven-0.241
Token out
Token ID503
Feature activation-4.33e-01

Pos logit contributions
roller+0.194
 salon+0.184
ara+0.175
post+0.174
atters+0.173
Neg logit contributions
ling-0.076
 around-0.055
 shining-0.055
led-0.054
ie-0.046
Token of
Token ID286
Feature activation+1.60e+00

Pos logit contributions
ling+0.018
 around+0.016
 shining+0.015
led+0.014
 up+0.014
Neg logit contributions
roller-0.026
 salon-0.025
ara-0.024
 clinic-0.023
 pitcher-0.023
Token her
Token ID607
Feature activation+2.42e-01

Pos logit contributions
roller+0.120
 salon+0.115
atters+0.112
lor+0.112
ara+0.111
Neg logit contributions
 shining-0.034
 around-0.032
 him-0.031
ling-0.031
 all-0.029
Token st
Token ID336
Feature activation+6.41e-01

Pos logit contributions
roller+0.010
ara+0.010
 salon+0.010
atters+0.010
post+0.010
Neg logit contributions
ling-0.012
 around-0.012
 shining-0.011
 up-0.011
 in-0.011
Tokenroller
Token ID10646
Feature activation+8.74e+00

Pos logit contributions
 salon+0.015
 heaven+0.014
Anna+0.013
 inconvenience+0.012
 anniversary+0.012
Neg logit contributions
ling-0.041
 around-0.040
 up-0.038
ully-0.038
ie-0.037
Token.
Token ID13
Feature activation-1.66e-01

Pos logit contributions
ara+0.610
Ble+0.604
 bidder+0.595
 clinic+0.594
 salon+0.593
Neg logit contributions
 around-0.243
ling-0.227
 up-0.203
 in-0.193
 through-0.169
Token
Token ID198
Feature activation+1.52e-01

Pos logit contributions
ling+0.007
 around+0.007
 shining+0.006
 up+0.006
 him+0.006
Neg logit contributions
roller-0.009
 salon-0.009
 heaven-0.008
 clinic-0.008
atters-0.008
Token
Token ID198
Feature activation+1.36e-01

Pos logit contributions
roller+0.007
 salon+0.006
ara+0.006
atters+0.006
 heaven+0.006
Neg logit contributions
ling-0.008
 shining-0.007
 around-0.007
led-0.007
 up-0.007
TokenWhen
Token ID2215
Feature activation+3.70e+00

Pos logit contributions
roller+0.007
 salon+0.006
 clinic+0.006
 pitcher+0.006
 heaven+0.006
Neg logit contributions
ling-0.007
 around-0.007
 up-0.006
 in-0.006
 shining-0.006
Token they
Token ID484
Feature activation+5.51e-01

Pos logit contributions
roller+0.245
lor+0.230
atters+0.226
 salon+0.226
 clinic+0.225
Neg logit contributions
ling-0.118
 around-0.118
 shining-0.115
 up-0.108
 in-0.107
Token She
Token ID1375
Feature activation+2.56e+00

Pos logit contributions
roller+0.017
 salon+0.016
atters+0.015
lor+0.014
 pitcher+0.014
Neg logit contributions
 around-0.018
ling-0.017
 him-0.017
 shining-0.017
 up-0.016
Token puts
Token ID7584
Feature activation+6.46e-01

Pos logit contributions
roller+0.125
 salon+0.115
ara+0.107
post+0.106
 pitcher+0.105
Neg logit contributions
ling-0.131
 around-0.118
 shining-0.114
led-0.111
 up-0.109
Token on
Token ID319
Feature activation-2.51e+00

Pos logit contributions
roller+0.036
 salon+0.036
 heaven+0.035
post+0.034
ara+0.034
Neg logit contributions
ling-0.026
 shining-0.020
itive-0.020
led-0.019
 around-0.019
Token her
Token ID607
Feature activation+8.78e-01

Pos logit contributions
ling+0.073
 around+0.062
 shining+0.058
 up+0.054
led+0.052
Neg logit contributions
roller-0.177
 salon-0.169
ara-0.162
atters-0.160
 clinic-0.159
Token p
Token ID279
Feature activation+4.62e+00

Pos logit contributions
roller+0.035
 salon+0.033
 heaven+0.030
 clinic+0.030
ara+0.029
Neg logit contributions
ling-0.054
led-0.046
 around-0.046
 shining-0.045
ie-0.045
Tokenaj
Token ID1228
Feature activation+8.72e+00

Pos logit contributions
 heaven+0.169
 salon+0.166
 vet+0.162
 bidder+0.162
roller+0.162
Neg logit contributions
ling-0.263
 around-0.236
ie-0.230
led-0.227
 up-0.226
Tokenamas
Token ID17485
Feature activation+2.07e+00

Pos logit contributions
 salon+0.357
roller+0.344
 heaven+0.326
atters+0.319
ara+0.318
Neg logit contributions
ling-0.494
 shining-0.448
 around-0.442
led-0.422
 up-0.420
Token.
Token ID13
Feature activation-5.58e-02

Pos logit contributions
roller+0.135
 salon+0.130
ara+0.126
 pitcher+0.125
atters+0.124
Neg logit contributions
ling-0.070
 around-0.066
 shining-0.059
 up-0.057
 in-0.056
Token She
Token ID1375
Feature activation+2.64e+00

Pos logit contributions
 around+0.003
ling+0.003
 shining+0.003
 him+0.003
 up+0.003
Neg logit contributions
roller-0.002
 salon-0.002
atters-0.002
 pitcher-0.002
ara-0.002
Token hugs
Token ID40657
Feature activation+2.79e+00

Pos logit contributions
roller+0.127
 salon+0.114
post+0.108
ress+0.106
 pitcher+0.106
Neg logit contributions
ling-0.138
 around-0.125
 shining-0.120
led-0.118
 up-0.117
Token her
Token ID607
Feature activation+7.76e-01

Pos logit contributions
roller+0.163
 salon+0.159
ara+0.153
 heaven+0.151
post+0.149
Neg logit contributions
ling-0.115
led-0.093
 shining-0.091
 around-0.090
 atop-0.083
Token pushing
Token ID7796
Feature activation-2.44e+00

Pos logit contributions
roller+0.006
 salon+0.006
ara+0.006
atters+0.006
 heaven+0.006
Neg logit contributions
ling-0.003
 around-0.003
 shining-0.002
led-0.002
 up-0.002
Token Tommy
Token ID19919
Feature activation-8.63e-01

Pos logit contributions
ling+0.083
 shining+0.072
itive+0.069
led+0.064
bee+0.061
Neg logit contributions
roller-0.137
 heaven-0.135
ara-0.134
 jail-0.130
post-0.129
Token in
Token ID287
Feature activation-1.79e+00

Pos logit contributions
ling+0.026
led+0.026
itive+0.024
ie+0.022
agles+0.021
Neg logit contributions
roller-0.052
 salon-0.051
stand-0.045
 rack-0.044
ress-0.044
Token his
Token ID465
Feature activation-1.78e+00

Pos logit contributions
ling+0.050
 around+0.043
 shining+0.040
 up+0.035
 him+0.035
Neg logit contributions
roller-0.127
 salon-0.122
ara-0.117
atters-0.116
 clinic-0.114
Token st
Token ID336
Feature activation+6.13e-01

Pos logit contributions
ling+0.103
 around+0.094
 shining+0.091
led+0.088
 up+0.087
Neg logit contributions
roller-0.074
 salon-0.069
ara-0.065
atters-0.062
 heaven-0.062
Tokenroller
Token ID10646
Feature activation+8.67e+00

Pos logit contributions
 heaven+0.012
Ble+0.012
jar+0.011
S+0.010
Anna+0.010
Neg logit contributions
ling-0.041
 around-0.040
 up-0.039
 all-0.038
 upon-0.037
Token.
Token ID13
Feature activation-2.95e-01

Pos logit contributions
roller+0.533
 salon+0.506
ara+0.499
atters+0.485
 pitcher+0.481
Neg logit contributions
ling-0.328
 around-0.291
 shining-0.266
 up-0.254
led-0.246
Token Suddenly
Token ID24975
Feature activation+1.91e+00

Pos logit contributions
ling+0.013
 around+0.013
 shining+0.012
 him+0.012
 up+0.012
Neg logit contributions
roller-0.014
 salon-0.014
atters-0.013
 heaven-0.013
 clinic-0.013
Token,
Token ID11
Feature activation-2.13e-01

Pos logit contributions
roller+0.103
 salon+0.098
ara+0.095
post+0.088
 heaven+0.088
Neg logit contributions
ling-0.088
 around-0.073
 shining-0.071
led-0.071
ie-0.066
Token Tommy
Token ID19919
Feature activation-8.54e-01

Pos logit contributions
ling+0.009
 around+0.008
led+0.007
 shining+0.007
ie+0.007
Neg logit contributions
roller-0.012
 salon-0.012
ara-0.011
 heaven-0.011
 pitcher-0.011
Token saw
Token ID2497
Feature activation-6.16e-01

Pos logit contributions
ling+0.039
 minerals+0.034
nam+0.034
led+0.034
 shining+0.034
Neg logit contributions
roller-0.042
ara-0.037
 salon-0.037
 pitcher-0.033
stand-0.032
Token Ben
Token ID3932
Feature activation+1.69e+00

Pos logit contributions
ling+0.011
 around+0.010
 up+0.009
 him+0.009
led+0.008
Neg logit contributions
roller-0.027
 salon-0.027
ara-0.025
 clinic-0.025
 pitcher-0.024
Token and
Token ID290
Feature activation-3.86e-02

Pos logit contributions
roller+0.094
 salon+0.089
ara+0.084
 heaven+0.084
atters+0.083
Neg logit contributions
ling-0.072
 shining-0.063
 around-0.062
led-0.060
ie-0.056
Token Lily
Token ID20037
Feature activation+4.35e+00

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
 up+0.002
 him+0.002
Neg logit contributions
roller-0.002
 salon-0.002
ara-0.002
post-0.002
lor-0.002
Token to
Token ID284
Feature activation+6.27e-01

Pos logit contributions
roller+0.254
 salon+0.240
 heaven+0.229
atters+0.225
ara+0.224
Neg logit contributions
ling-0.176
 shining-0.150
led-0.144
 around-0.143
ie-0.132
Token the
Token ID262
Feature activation-1.06e-01

Pos logit contributions
roller+0.046
 salon+0.044
ara+0.042
atters+0.041
 pitcher+0.041
Neg logit contributions
ling-0.017
 around-0.014
 shining-0.013
 him-0.013
 up-0.012
Token hospital
Token ID4436
Feature activation+8.65e+00

Pos logit contributions
ling+0.005
 around+0.005
 shining+0.005
 up+0.004
 in+0.004
Neg logit contributions
roller-0.005
ress-0.005
atters-0.005
 salon-0.005
lor-0.005
Token.
Token ID13
Feature activation+5.72e-02

Pos logit contributions
roller+0.551
 salon+0.523
ara+0.515
 pitcher+0.497
atters+0.488
Neg logit contributions
ling-0.312
 around-0.279
 shining-0.255
 up-0.246
led-0.238
Token
Token ID198
Feature activation+4.01e-01

Pos logit contributions
 salon+0.004
roller+0.004
lor+0.003
ress+0.003
ction+0.003
Neg logit contributions
 around-0.002
 him-0.002
ling-0.002
 his-0.002
 in-0.002
Token
Token ID198
Feature activation+4.01e-01

Pos logit contributions
roller+0.021
 salon+0.020
ara+0.018
 pitcher+0.018
 heaven+0.018
Neg logit contributions
ling-0.019
 around-0.017
 shining-0.016
 him-0.015
 up-0.015
TokenBen
Token ID11696
Feature activation+2.38e+00

Pos logit contributions
roller+0.022
 salon+0.021
ress+0.019
 bidder+0.019
lor+0.018
Neg logit contributions
ling-0.019
 around-0.017
 him-0.016
 his-0.015
 in-0.015
Token has
Token ID468
Feature activation+1.77e+00

Pos logit contributions
roller+0.165
 salon+0.159
ress+0.149
ction+0.149
post+0.149
Neg logit contributions
ling-0.078
 around-0.070
 up-0.061
led-0.058
 in-0.058
Token have
Token ID423
Feature activation+1.92e+00

Pos logit contributions
roller+0.174
 salon+0.160
ara+0.155
ress+0.154
lor+0.148
Neg logit contributions
ling-0.138
 around-0.127
 shining-0.117
 him-0.115
 up-0.112
Token to
Token ID284
Feature activation+6.00e-01

Pos logit contributions
roller+0.126
 salon+0.121
ara+0.117
atters+0.113
ress+0.112
Neg logit contributions
ling-0.061
 around-0.057
 shining-0.054
 him-0.053
 in-0.050
Token go
Token ID467
Feature activation+2.45e+00

Pos logit contributions
roller+0.038
lor+0.036
 salon+0.035
ase+0.035
ress+0.035
Neg logit contributions
 him-0.020
ling-0.019
 all-0.019
 up-0.019
 in-0.018
Token to
Token ID284
Feature activation+5.99e-01

Pos logit contributions
roller+0.173
ara+0.168
lor+0.165
Ble+0.164
 salon+0.163
Neg logit contributions
 around-0.075
ling-0.070
 in-0.065
 up-0.064
 him-0.056
Token the
Token ID262
Feature activation+2.01e-01

Pos logit contributions
roller+0.040
lor+0.038
 salon+0.037
ara+0.037
 pitcher+0.036
Neg logit contributions
 him-0.018
 up-0.016
 around-0.016
ling-0.016
 his-0.015
Token hospital
Token ID4436
Feature activation+8.63e+00

Pos logit contributions
roller+0.008
atters+0.008
lor+0.008
ress+0.007
ird+0.007
Neg logit contributions
 up-0.011
 shining-0.011
 around-0.010
 in-0.010
ling-0.010
Token.
Token ID13
Feature activation+1.00e-01

Pos logit contributions
roller+0.571
ara+0.533
 salon+0.530
 pitcher+0.522
atters+0.510
Neg logit contributions
ling-0.292
 around-0.269
 up-0.235
 shining-0.231
 in-0.230
Token Mom
Token ID11254
Feature activation+2.21e+00

Pos logit contributions
 salon+0.006
roller+0.006
lor+0.006
 pitcher+0.006
ress+0.006
Neg logit contributions
 him-0.003
 around-0.003
 his-0.003
 in-0.003
 up-0.003
Token and
Token ID290
Feature activation+1.15e-02

Pos logit contributions
roller+0.162
 pitcher+0.147
post+0.147
 salon+0.146
lor+0.144
Neg logit contributions
ling-0.066
 around-0.059
 up-0.050
 in-0.049
ie-0.044
Token Dad
Token ID17415
Feature activation-9.55e-01

Pos logit contributions
roller+0.001
 salon+0.001
atters+0.001
ara+0.001
 clinic+0.001
Neg logit contributions
ling-0.000
 around-0.000
 shining-0.000
led-0.000
 up-0.000
Token are
Token ID389
Feature activation+1.70e+00

Pos logit contributions
ling+0.047
 around+0.043
 shining+0.041
led+0.039
 up+0.039
Neg logit contributions
roller-0.048
 salon-0.045
ara-0.043
 pitcher-0.042
post-0.041
Token have
Token ID423
Feature activation+1.41e+00

Pos logit contributions
roller+0.099
 salon+0.094
ara+0.089
 pitcher+0.087
 heaven+0.087
Neg logit contributions
ling-0.073
 around-0.064
 shining-0.061
led-0.058
 up-0.057
Token to
Token ID284
Feature activation+3.33e-01

Pos logit contributions
roller+0.090
 salon+0.084
ara+0.081
atters+0.080
post+0.079
Neg logit contributions
ling-0.049
 around-0.045
 shining-0.042
led-0.039
 him-0.039
Token go
Token ID467
Feature activation+2.04e+00

Pos logit contributions
roller+0.021
lor+0.019
ara+0.019
 salon+0.019
ress+0.018
Neg logit contributions
ling-0.011
 him-0.011
 around-0.011
 all-0.011
 in-0.011
Token back
Token ID736
Feature activation+2.40e+00

Pos logit contributions
roller+0.134
 salon+0.127
ara+0.124
 pitcher+0.122
post+0.120
Neg logit contributions
ling-0.069
 around-0.065
 shining-0.057
 up-0.057
 in-0.053
Token to
Token ID284
Feature activation+3.42e-01

Pos logit contributions
roller+0.152
 salon+0.144
 pitcher+0.138
ara+0.137
atters+0.135
Neg logit contributions
ling-0.086
 around-0.079
 shining-0.071
 up-0.071
 him-0.066
Token heaven
Token ID9538
Feature activation+8.63e+00

Pos logit contributions
roller+0.022
 salon+0.020
ara+0.020
 pitcher+0.019
post+0.019
Neg logit contributions
ling-0.012
 around-0.011
 him-0.011
 shining-0.010
 up-0.010
Token,"
Token ID553
Feature activation+5.99e-01

Pos logit contributions
roller+0.496
 salon+0.458
 pitcher+0.447
atters+0.435
ara+0.432
Neg logit contributions
ling-0.371
 around-0.353
 up-0.316
 in-0.315
 shining-0.299
Token said
Token ID531
Feature activation-1.35e+00

Pos logit contributions
 salon+0.029
roller+0.028
ara+0.028
 heaven+0.027
 vet+0.026
Neg logit contributions
ling-0.030
 around-0.026
 shining-0.025
led-0.025
 up-0.023
Token the
Token ID262
Feature activation-2.22e-01

Pos logit contributions
ling+0.064
led+0.052
 shining+0.050
 around+0.049
uses+0.046
Neg logit contributions
 salon-0.069
roller-0.069
 heaven-0.066
ara-0.064
 vet-0.064
Token angel
Token ID18304
Feature activation-1.88e+00

Pos logit contributions
ling+0.010
 around+0.009
 shining+0.008
led+0.008
 him+0.008
Neg logit contributions
roller-0.012
 salon-0.012
ara-0.011
 heaven-0.011
 clinic-0.010
Token.
Token ID13
Feature activation+6.37e-02

Pos logit contributions
ling+0.086
 shining+0.077
led+0.073
 around+0.072
 him+0.069
Neg logit contributions
roller-0.092
 salon-0.091
 heaven-0.086
ress-0.085
 clinic-0.085
Token Sara
Token ID24799
Feature activation+4.25e+00

Pos logit contributions
 salon+0.011
lor+0.011
ction+0.010
roller+0.010
ress+0.010
Neg logit contributions
 around-0.007
 in-0.007
 him-0.007
 his-0.007
ling-0.006
Token was
Token ID373
Feature activation+4.69e-01

Pos logit contributions
roller+0.256
 salon+0.244
ara+0.234
 heaven+0.227
atters+0.226
Neg logit contributions
ling-0.166
 around-0.142
 shining-0.140
led-0.130
 up-0.125
Token taken
Token ID2077
Feature activation+5.95e-01

Pos logit contributions
roller+0.028
 salon+0.026
 heaven+0.026
 clinic+0.025
atters+0.025
Neg logit contributions
ling-0.018
 shining-0.015
 around-0.015
led-0.015
ie-0.014
Token to
Token ID284
Feature activation+5.06e-01

Pos logit contributions
roller+0.047
 salon+0.045
 pitcher+0.042
ara+0.042
atters+0.042
Neg logit contributions
ling-0.014
 around-0.013
 up-0.011
 him-0.011
 in-0.010
Token the
Token ID262
Feature activation-7.30e-02

Pos logit contributions
roller+0.041
lor+0.038
 salon+0.038
atters+0.038
 pitcher+0.037
Neg logit contributions
 him-0.008
 up-0.008
 around-0.007
ling-0.007
 all-0.006
Token hospital
Token ID4436
Feature activation+8.61e+00

Pos logit contributions
ling+0.004
 shining+0.004
 around+0.004
 up+0.004
 in+0.003
Neg logit contributions
roller-0.003
lor-0.003
atters-0.003
ress-0.003
post-0.003
Token.
Token ID13
Feature activation+1.48e-01

Pos logit contributions
roller+0.539
 salon+0.511
ara+0.496
 pitcher+0.482
atters+0.480
Neg logit contributions
ling-0.320
 around-0.282
 shining-0.264
 up-0.248
led-0.246
Token She
Token ID1375
Feature activation+2.52e+00

Pos logit contributions
 salon+0.008
ction+0.008
roller+0.008
lor+0.008
ress+0.008
Neg logit contributions
 him-0.005
 around-0.005
 in-0.005
 his-0.004
 up-0.004
Token had
Token ID550
Feature activation-7.11e-01

Pos logit contributions
roller+0.164
 salon+0.154
 pitcher+0.146
atters+0.145
ara+0.145
Neg logit contributions
ling-0.088
 around-0.076
 shining-0.071
led-0.071
 him-0.066
Token a
Token ID257
Feature activation+2.24e-01

Pos logit contributions
ling+0.021
 around+0.019
 shining+0.019
 him+0.017
led+0.016
Neg logit contributions
roller-0.049
 salon-0.047
ara-0.045
atters-0.044
ress-0.043
Token bad
Token ID2089
Feature activation+5.86e-01

Pos logit contributions
roller+0.013
 salon+0.013
atters+0.013
ara+0.013
ress+0.012
Neg logit contributions
ling-0.008
 shining-0.008
 around-0.007
 up-0.007
 in-0.006
Token put
Token ID1234
Feature activation+5.43e-01

Pos logit contributions
ling+0.165
 around+0.142
 shining+0.139
led+0.132
 up+0.126
Neg logit contributions
roller-0.231
 salon-0.220
ara-0.209
 heaven-0.203
 pitcher-0.203
Token him
Token ID683
Feature activation-4.58e+00

Pos logit contributions
 heaven+0.030
roller+0.029
 pitcher+0.029
post+0.029
 jail+0.028
Neg logit contributions
ling-0.021
itive-0.017
bee-0.016
ie-0.016
 curiously-0.015
Token in
Token ID287
Feature activation-1.68e+00

Pos logit contributions
ling+0.135
 Barry+0.094
itive+0.094
 curiously+0.093
uses+0.093
Neg logit contributions
roller-0.286
 heaven-0.276
 salon-0.273
 jail-0.263
 clinic-0.261
Token his
Token ID465
Feature activation-1.67e+00

Pos logit contributions
 shining+0.028
 him+0.025
 around+0.023
 in+0.023
 his+0.023
Neg logit contributions
roller-0.136
atters-0.132
 salon-0.130
lor-0.128
ara-0.128
Token st
Token ID336
Feature activation+6.41e-01

Pos logit contributions
ling+0.094
 around+0.091
 shining+0.089
 up+0.086
 in+0.083
Neg logit contributions
roller-0.066
 salon-0.062
atters-0.061
ara-0.061
post-0.059
Tokenroller
Token ID10646
Feature activation+8.60e+00

Pos logit contributions
 inconvenience+0.015
 salon+0.014
Ble+0.014
Uh+0.014
 heaven+0.013
Neg logit contributions
 around-0.038
ling-0.038
 up-0.037
 all-0.037
 upon-0.036
Token and
Token ID290
Feature activation-3.38e-01

Pos logit contributions
roller+0.552
ara+0.540
 salon+0.535
atters+0.530
 bidder+0.528
Neg logit contributions
ling-0.286
 around-0.280
 up-0.242
 in-0.228
 shining-0.222
Token they
Token ID484
Feature activation+5.14e-01

Pos logit contributions
ling+0.015
 around+0.012
 shining+0.012
led+0.012
ie+0.011
Neg logit contributions
roller-0.019
 salon-0.018
ara-0.017
 heaven-0.017
 clinic-0.017
Token went
Token ID1816
Feature activation+6.37e-01

Pos logit contributions
roller+0.030
 salon+0.029
ara+0.027
 pitcher+0.026
 heaven+0.026
Neg logit contributions
ling-0.021
 around-0.017
 shining-0.017
led-0.016
 up-0.015
Token to
Token ID284
Feature activation+5.54e-01

Pos logit contributions
roller+0.046
 salon+0.045
ara+0.045
 bidder+0.043
 pitcher+0.043
Neg logit contributions
ling-0.016
 around-0.016
 shining-0.014
 up-0.014
 in-0.013
Token the
Token ID262
Feature activation+9.77e-02

Pos logit contributions
roller+0.040
 salon+0.038
ara+0.037
post+0.036
atters+0.036
Neg logit contributions
ling-0.014
 around-0.013
 shining-0.012
 up-0.011
 him-0.011
Token happy
Token ID3772
Feature activation+4.23e-01

Pos logit contributions
roller+0.031
 salon+0.029
ara+0.028
 heaven+0.027
 clinic+0.027
Neg logit contributions
ling-0.023
 around-0.020
 shining-0.019
led-0.018
 up-0.018
Token.
Token ID13
Feature activation+1.03e+00

Pos logit contributions
roller+0.025
 salon+0.024
ara+0.023
 heaven+0.022
 clinic+0.022
Neg logit contributions
ling-0.017
 around-0.015
 shining-0.014
led-0.014
 up-0.013
Token They
Token ID1119
Feature activation+3.78e+00

Pos logit contributions
 salon+0.055
roller+0.054
 pitcher+0.049
ction+0.048
atters+0.047
Neg logit contributions
 around-0.042
 Joey-0.041
ling-0.041
 him-0.040
 Peter-0.039
Token thanked
Token ID26280
Feature activation+2.32e+00

Pos logit contributions
roller+0.219
 salon+0.204
ara+0.193
ress+0.189
post+0.189
Neg logit contributions
ling-0.161
 around-0.145
 shining-0.135
led-0.131
 him-0.128
Token the
Token ID262
Feature activation+1.58e-01

Pos logit contributions
roller+0.140
 salon+0.136
 heaven+0.133
ara+0.132
 clinic+0.127
Neg logit contributions
ling-0.089
led-0.071
 around-0.071
 shining-0.070
 atop-0.060
Token vet
Token ID20202
Feature activation+8.60e+00

Pos logit contributions
roller+0.008
 salon+0.008
ara+0.007
 clinic+0.007
 heaven+0.007
Neg logit contributions
ling-0.008
 around-0.007
led-0.006
 shining-0.006
ie-0.006
Token and
Token ID290
Feature activation-5.83e-02

Pos logit contributions
roller+0.496
 salon+0.478
 clinic+0.449
 heaven+0.449
ara+0.448
Neg logit contributions
ling-0.345
 shining-0.294
 around-0.287
led-0.281
 up-0.252
Token praised
Token ID15342
Feature activation+7.78e-01

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
led+0.002
 up+0.002
Neg logit contributions
roller-0.004
 salon-0.003
ara-0.003
atters-0.003
post-0.003
Token him
Token ID683
Feature activation-4.18e+00

Pos logit contributions
roller+0.048
 heaven+0.047
 salon+0.047
ara+0.046
post+0.044
Neg logit contributions
ling-0.027
 shining-0.022
led-0.020
 around-0.020
 atop-0.018
Token for
Token ID329
Feature activation-1.77e+00

Pos logit contributions
ling+0.151
 shining+0.132
led+0.125
 around+0.118
ie+0.109
Neg logit contributions
roller-0.245
 salon-0.236
 heaven-0.235
 clinic-0.230
ara-0.226
Token his
Token ID465
Feature activation-1.09e+00

Pos logit contributions
ling+0.066
 around+0.054
 shining+0.051
led+0.051
ie+0.047
Neg logit contributions
roller-0.110
 salon-0.106
 heaven-0.102
ara-0.100
 clinic-0.100
Token Ben
Token ID3932
Feature activation+1.71e+00

Pos logit contributions
ling+0.008
 around+0.007
 shining+0.007
led+0.006
 up+0.006
Neg logit contributions
roller-0.017
 salon-0.016
ara-0.015
 heaven-0.015
atters-0.015
Token and
Token ID290
Feature activation-2.70e-01

Pos logit contributions
roller+0.096
 salon+0.089
 heaven+0.086
 clinic+0.083
ara+0.083
Neg logit contributions
ling-0.073
 shining-0.065
led-0.062
 around-0.059
ie-0.056
Token Lily
Token ID20037
Feature activation+4.43e+00

Pos logit contributions
ling+0.011
 around+0.010
 shining+0.010
led+0.010
 up+0.009
Neg logit contributions
roller-0.016
 salon-0.015
ara-0.014
atters-0.014
post-0.013
Token to
Token ID284
Feature activation+5.61e-01

Pos logit contributions
roller+0.279
 salon+0.266
 heaven+0.251
ara+0.250
 clinic+0.248
Neg logit contributions
ling-0.161
 shining-0.133
 around-0.129
led-0.126
ie-0.116
Token the
Token ID262
Feature activation-1.35e-01

Pos logit contributions
roller+0.039
 salon+0.037
ara+0.036
atters+0.035
 pitcher+0.035
Neg logit contributions
ling-0.017
 around-0.014
 shining-0.013
 him-0.012
 up-0.012
Token hospital
Token ID4436
Feature activation+8.60e+00

Pos logit contributions
ling+0.007
 around+0.007
 shining+0.007
 up+0.006
 in+0.006
Neg logit contributions
roller-0.006
atters-0.005
 salon-0.005
ress-0.005
ara-0.005
Token.
Token ID13
Feature activation+1.14e-03

Pos logit contributions
roller+0.526
 salon+0.501
ara+0.482
 pitcher+0.467
atters+0.465
Neg logit contributions
ling-0.329
 around-0.287
 shining-0.274
led-0.258
 up-0.254
Token Ben
Token ID3932
Feature activation+1.66e+00

Pos logit contributions
 salon+0.000
roller+0.000
lor+0.000
ress+0.000
ction+0.000
Neg logit contributions
 around-0.000
 him-0.000
 his-0.000
ling-0.000
 all-0.000
Token had
Token ID550
Feature activation-8.20e-01

Pos logit contributions
roller+0.110
 salon+0.106
atters+0.099
 pitcher+0.099
ara+0.097
Neg logit contributions
ling-0.057
 around-0.051
led-0.045
 in-0.043
 shining-0.043
Token to
Token ID284
Feature activation+4.65e-01

Pos logit contributions
ling+0.025
 around+0.023
 shining+0.022
 him+0.021
led+0.020
Neg logit contributions
roller-0.055
 salon-0.053
ara-0.051
atters-0.049
 pitcher-0.049
Token get
Token ID651
Feature activation+1.64e+00

Pos logit contributions
roller+0.030
 salon+0.028
ara+0.027
ase+0.027
lor+0.027
Neg logit contributions
 him-0.016
 his-0.014
 all-0.014
ling-0.014
 up-0.014
Token and
Token ID290
Feature activation-6.46e-02

Pos logit contributions
roller+0.280
 salon+0.262
ara+0.259
 pitcher+0.258
atters+0.252
Neg logit contributions
ling-0.137
 around-0.131
 shining-0.118
 in-0.114
 up-0.111
Token took
Token ID1718
Feature activation-2.50e-01

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
led+0.002
 up+0.002
Neg logit contributions
roller-0.004
 salon-0.004
ara-0.004
atters-0.003
 pitcher-0.003
Token her
Token ID607
Feature activation+7.10e-01

Pos logit contributions
ling+0.008
 around+0.007
 him+0.007
 up+0.006
 shining+0.006
Neg logit contributions
roller-0.017
 salon-0.016
ara-0.015
 pitcher-0.015
 clinic-0.015
Token to
Token ID284
Feature activation+5.71e-01

Pos logit contributions
roller+0.041
 salon+0.038
 heaven+0.037
 clinic+0.035
 pitcher+0.035
Neg logit contributions
ling-0.031
led-0.025
 shining-0.025
ie-0.024
itive-0.023
Token the
Token ID262
Feature activation+2.09e-01

Pos logit contributions
roller+0.048
lor+0.047
 salon+0.046
ara+0.045
atters+0.045
Neg logit contributions
 him-0.008
 up-0.006
 around-0.006
 all-0.005
 his-0.005
Token hospital
Token ID4436
Feature activation+8.60e+00

Pos logit contributions
roller+0.008
lor+0.008
ara+0.007
atters+0.007
ress+0.007
Neg logit contributions
 up-0.012
ling-0.011
 around-0.011
 shining-0.011
 in-0.011
Token.
Token ID13
Feature activation+2.35e-01

Pos logit contributions
roller+0.542
 salon+0.516
ara+0.504
 pitcher+0.489
atters+0.484
Neg logit contributions
ling-0.312
 around-0.276
 shining-0.258
 up-0.244
led-0.240
Token They
Token ID1119
Feature activation+3.33e+00

Pos logit contributions
 salon+0.015
lor+0.015
ction+0.015
ress+0.014
 pitcher+0.014
Neg logit contributions
 his-0.007
 around-0.007
 him-0.007
 in-0.006
 up-0.006
Token were
Token ID547
Feature activation-6.49e-01

Pos logit contributions
roller+0.194
 salon+0.183
ara+0.175
 pitcher+0.171
atters+0.171
Neg logit contributions
ling-0.138
 around-0.121
 shining-0.116
led-0.112
 up-0.106
Token angry
Token ID7954
Feature activation-1.11e+00

Pos logit contributions
ling+0.026
 shining+0.022
 around+0.022
led+0.021
ie+0.021
Neg logit contributions
roller-0.037
 salon-0.036
 heaven-0.034
 clinic-0.034
atters-0.034
Token and
Token ID290
Feature activation-4.11e-02

Pos logit contributions
ling+0.044
 around+0.038
 shining+0.036
led+0.034
 up+0.034
Neg logit contributions
roller-0.068
 salon-0.065
ara-0.061
 pitcher-0.060
 clinic-0.060
Token and
Token ID290
Feature activation-1.20e-01

Pos logit contributions
roller+0.007
 salon+0.007
ara+0.006
 pitcher+0.006
 clinic+0.006
Neg logit contributions
ling-0.004
 around-0.004
 shining-0.004
 up-0.003
 him-0.003
Token takes
Token ID2753
Feature activation-3.28e-01

Pos logit contributions
ling+0.004
 around+0.003
 shining+0.003
led+0.003
 up+0.003
Neg logit contributions
roller-0.008
 salon-0.008
ara-0.007
post-0.007
 pitcher-0.007
Token Tim
Token ID5045
Feature activation-7.64e-01

Pos logit contributions
ling+0.008
 around+0.007
 him+0.006
 shining+0.006
 up+0.006
Neg logit contributions
roller-0.025
 salon-0.024
ara-0.023
 clinic-0.023
post-0.023
Token to
Token ID284
Feature activation+6.69e-01

Pos logit contributions
ling+0.029
 around+0.025
 shining+0.024
led+0.023
 up+0.022
Neg logit contributions
roller-0.047
 salon-0.045
ara-0.043
 heaven-0.042
atters-0.042
Token the
Token ID262
Feature activation+1.25e-01

Pos logit contributions
roller+0.057
lor+0.055
 salon+0.054
ara+0.053
atters+0.053
Neg logit contributions
 him-0.010
 his-0.007
 up-0.007
 around-0.006
ling-0.005
Token hospital
Token ID4436
Feature activation+8.59e+00

Pos logit contributions
roller+0.005
atters+0.004
lor+0.004
post+0.004
ird+0.004
Neg logit contributions
ling-0.007
 up-0.007
 around-0.007
 shining-0.007
 all-0.006
Token.
Token ID13
Feature activation-1.32e-02

Pos logit contributions
roller+0.554
 salon+0.526
ara+0.519
 pitcher+0.502
atters+0.493
Neg logit contributions
ling-0.301
 around-0.271
 shining-0.244
 up-0.239
led-0.227
Token Tim
Token ID5045
Feature activation-7.60e-01

Pos logit contributions
 around+0.001
 him+0.001
ling+0.001
 up+0.001
 in+0.001
Neg logit contributions
 salon-0.001
lor-0.001
roller-0.001
ction-0.001
ara-0.001
Token has
Token ID468
Feature activation+1.70e+00

Pos logit contributions
ling+0.039
 around+0.035
 shining+0.033
 up+0.033
led+0.033
Neg logit contributions
roller-0.038
 salon-0.035
post-0.032
ara-0.032
 pitcher-0.032
Token a
Token ID257
Feature activation+3.20e-01

Pos logit contributions
roller+0.119
 salon+0.114
ara+0.112
atters+0.108
ress+0.107
Neg logit contributions
ling-0.042
 him-0.041
 around-0.041
 shining-0.040
 in-0.037
Token big
Token ID1263
Feature activation+1.74e+00

Pos logit contributions
roller+0.018
 salon+0.017
ara+0.016
atters+0.016
ress+0.016
Neg logit contributions
ling-0.013
 shining-0.012
 around-0.012
 up-0.011
 in-0.010
Token Ben
Token ID3932
Feature activation+1.79e+00

Pos logit contributions
ling+0.009
 around+0.008
 up+0.007
 shining+0.007
 him+0.007
Neg logit contributions
roller-0.020
 salon-0.019
ara-0.018
 clinic-0.018
atters-0.018
Token and
Token ID290
Feature activation+2.33e-02

Pos logit contributions
roller+0.109
 salon+0.104
ara+0.099
 pitcher+0.097
atters+0.096
Neg logit contributions
ling-0.070
 around-0.061
 shining-0.057
 up-0.054
led-0.054
Token Sam
Token ID3409
Feature activation-1.00e+00

Pos logit contributions
roller+0.001
 salon+0.001
ara+0.001
post+0.001
lor+0.001
Neg logit contributions
ling-0.001
 around-0.001
 shining-0.001
 up-0.001
led-0.001
Token to
Token ID284
Feature activation+6.30e-01

Pos logit contributions
ling+0.038
 around+0.032
 shining+0.032
led+0.030
 up+0.028
Neg logit contributions
roller-0.062
 salon-0.059
ara-0.056
 heaven-0.055
atters-0.055
Token the
Token ID262
Feature activation+3.47e-01

Pos logit contributions
roller+0.047
 salon+0.045
lor+0.044
ara+0.044
atters+0.043
Neg logit contributions
 him-0.014
ling-0.014
 around-0.013
 up-0.012
 shining-0.012
Token hospital
Token ID4436
Feature activation+8.59e+00

Pos logit contributions
roller+0.015
ress+0.015
atters+0.015
ara+0.015
lor+0.014
Neg logit contributions
 around-0.016
 shining-0.016
 up-0.016
ling-0.016
 in-0.016
Token.
Token ID13
Feature activation+2.39e-01

Pos logit contributions
roller+0.556
 salon+0.531
ara+0.523
 pitcher+0.503
atters+0.494
Neg logit contributions
ling-0.298
 around-0.270
 shining-0.243
 up-0.239
led-0.229
Token They
Token ID1119
Feature activation+3.14e+00

Pos logit contributions
 salon+0.014
lor+0.013
roller+0.013
ress+0.013
ction+0.013
Neg logit contributions
 around-0.009
 him-0.009
 his-0.009
 all-0.008
 in-0.008
Token had
Token ID550
Feature activation-5.87e-01

Pos logit contributions
roller+0.193
 salon+0.179
ara+0.175
 pitcher+0.170
atters+0.169
Neg logit contributions
ling-0.119
 around-0.109
 shining-0.101
led-0.098
 in-0.097
Token to
Token ID284
Feature activation+6.12e-01

Pos logit contributions
ling+0.017
 shining+0.017
 around+0.017
 him+0.016
 in+0.015
Neg logit contributions
roller-0.039
 salon-0.037
ara-0.036
atters-0.035
lor-0.035
Token stay
Token ID2652
Feature activation+8.72e-01

Pos logit contributions
roller+0.040
lor+0.039
ase+0.038
 salon+0.038
ara+0.037
Neg logit contributions
 him-0.020
 up-0.018
 all-0.018
 in-0.018
 his-0.017
Token Tim
Token ID5045
Feature activation-6.04e-01

Pos logit contributions
ling+0.006
 around+0.005
 shining+0.005
 up+0.005
 him+0.004
Neg logit contributions
roller-0.015
 salon-0.014
ara-0.014
 clinic-0.013
 pitcher-0.013
Token and
Token ID290
Feature activation+2.24e-02

Pos logit contributions
ling+0.023
 around+0.020
 up+0.018
 shining+0.018
led+0.017
Neg logit contributions
roller-0.038
 salon-0.037
ara-0.035
 pitcher-0.034
 clinic-0.034
Token Kim
Token ID6502
Feature activation+2.99e-01

Pos logit contributions
roller+0.001
 salon+0.001
lor+0.001
ara+0.001
 clinic+0.001
Neg logit contributions
ling-0.001
 shining-0.001
 around-0.001
 him-0.001
 in-0.001
Token to
Token ID284
Feature activation+6.18e-01

Pos logit contributions
roller+0.018
 salon+0.018
ara+0.017
 pitcher+0.016
post+0.016
Neg logit contributions
ling-0.012
 around-0.010
 shining-0.009
led-0.009
 up-0.009
Token the
Token ID262
Feature activation+1.44e-01

Pos logit contributions
roller+0.046
 salon+0.044
ara+0.044
lor+0.043
post+0.042
Neg logit contributions
ling-0.014
 him-0.014
 around-0.013
 up-0.012
 shining-0.011
Token hospital
Token ID4436
Feature activation+8.59e+00

Pos logit contributions
roller+0.006
ress+0.006
lor+0.006
ird+0.005
ara+0.005
Neg logit contributions
 up-0.007
 around-0.007
 in-0.007
ling-0.007
 shining-0.007
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
roller+0.553
 salon+0.526
ara+0.521
 pitcher+0.501
atters+0.491
Neg logit contributions
ling-0.302
 around-0.272
 shining-0.246
 up-0.242
led-0.231
Token They
Token ID1119
Feature activation+3.23e+00

Pos logit contributions
 salon+0.018
lor+0.017
roller+0.016
ction+0.016
 pitcher+0.016
Neg logit contributions
 him-0.012
 around-0.011
 his-0.010
 Joey-0.010
 in-0.010
Token threw
Token ID9617
Feature activation+7.02e-01

Pos logit contributions
roller+0.199
 salon+0.187
ara+0.181
 pitcher+0.175
atters+0.175
Neg logit contributions
ling-0.123
 around-0.111
 shining-0.103
led-0.101
 him-0.098
Token away
Token ID1497
Feature activation+4.31e+00

Pos logit contributions
roller+0.045
 salon+0.043
ara+0.041
 heaven+0.041
atters+0.041
Neg logit contributions
ling-0.024
 shining-0.019
 around-0.019
led-0.018
ie-0.017
Token the
Token ID262
Feature activation+1.32e-01

Pos logit contributions
roller+0.321
 salon+0.310
ara+0.292
atters+0.290
 pitcher+0.289
Neg logit contributions
ling-0.112
 around-0.097
 shining-0.087
 in-0.084
 up-0.083
Token for
Token ID329
Feature activation-1.92e+00

Pos logit contributions
ling+0.102
led+0.090
 shining+0.086
 around+0.081
itive+0.079
Neg logit contributions
 salon-0.114
ara-0.108
 clinic-0.108
roller-0.108
ress-0.107
Token some
Token ID617
Feature activation+1.27e+00

Pos logit contributions
ling+0.072
 around+0.060
 shining+0.057
led+0.055
 up+0.051
Neg logit contributions
roller-0.120
 salon-0.116
ara-0.110
 heaven-0.109
 clinic-0.107
Token funny
Token ID8258
Feature activation-3.49e-01

Pos logit contributions
roller+0.060
 salon+0.059
 heaven+0.055
 clinic+0.054
ara+0.054
Neg logit contributions
ling-0.064
 around-0.055
led-0.054
 shining-0.054
 up-0.050
Token little
Token ID1310
Feature activation+1.20e+00

Pos logit contributions
ling+0.016
 shining+0.014
 around+0.013
led+0.013
 atop+0.013
Neg logit contributions
roller-0.018
 salon-0.017
 heaven-0.016
ara-0.016
 pitcher-0.016
Token sun
Token ID4252
Feature activation+1.45e+00

Pos logit contributions
roller+0.065
 salon+0.063
ara+0.059
 heaven+0.059
 pitcher+0.058
Neg logit contributions
ling-0.052
 around-0.045
 shining-0.044
led-0.043
 up-0.040
Tokenfl
Token ID2704
Feature activation-1.32e+01

Pos logit contributions
roller+0.080
 salon+0.078
ara+0.073
 pitcher+0.071
 heaven+0.071
Neg logit contributions
ling-0.061
 around-0.050
 shining-0.049
led-0.049
 him-0.045
Tokenowers
Token ID3618
Feature activation-6.79e+00

Pos logit contributions
ling+0.601
 around+0.523
 shining+0.515
led+0.484
 up+0.471
Neg logit contributions
roller-0.706
 salon-0.671
ara-0.635
 pitcher-0.620
atters-0.610
Token.
Token ID13
Feature activation-6.68e-01

Pos logit contributions
ling+0.302
led+0.251
 shining+0.247
 around+0.241
itive+0.229
Neg logit contributions
roller-0.333
 salon-0.325
 pitcher-0.305
 heaven-0.302
post-0.299
Token Her
Token ID2332
Feature activation+3.66e-01

Pos logit contributions
ling+0.026
 around+0.026
 him+0.024
 Joey+0.024
 shining+0.024
Neg logit contributions
roller-0.036
 salon-0.034
atters-0.034
post-0.031
 pitcher-0.031
Token m
Token ID285
Feature activation-3.53e+00

Pos logit contributions
roller+0.019
 salon+0.018
ara+0.018
 vet+0.017
 clinic+0.017
Neg logit contributions
ling-0.018
led-0.015
 around-0.015
ging-0.013
 shining-0.013
Tokenama
Token ID1689
Feature activation-2.45e+00

Pos logit contributions
ling+0.180
 around+0.162
 shining+0.157
led+0.151
 up+0.146
Neg logit contributions
roller-0.174
 salon-0.164
ara-0.154
ress-0.147
atters-0.147
Token,
Token ID11
Feature activation+2.92e+00

Pos logit contributions
ling+0.033
 shining+0.028
 around+0.026
led+0.026
 usual+0.026
Neg logit contributions
 heaven-0.037
ara-0.037
S-0.034
 salon-0.033
 jail-0.033
Tokenâ
Token ID22940
Feature activation+3.69e+00

Pos logit contributions
roller+0.162
 salon+0.158
ara+0.147
 pitcher+0.146
 heaven+0.146
Neg logit contributions
ling-0.126
led-0.106
 around-0.102
 shining-0.099
ie-0.094
Token
Token ID26391
Feature activation-2.32e+00

Pos logit contributions
roller+0.181
 salon+0.166
 heaven+0.155
 pitcher+0.154
ara+0.153
Neg logit contributions
ling-0.199
 around-0.173
 shining-0.165
led-0.160
ie-0.160
Token said
Token ID531
Feature activation-9.09e-01

Pos logit contributions
ling+0.100
 around+0.093
 shining+0.088
 up+0.084
 in+0.081
Neg logit contributions
roller-0.114
 salon-0.114
 heaven-0.112
 pitcher-0.108
 clinic-0.107
Token T
Token ID309
Feature activation-4.49e+00

Pos logit contributions
ling+0.039
 shining+0.033
 around+0.033
led+0.032
ie+0.030
Neg logit contributions
roller-0.051
 salon-0.048
ara-0.046
 heaven-0.045
post-0.044
Tokenod
Token ID375
Feature activation-1.32e+01

Pos logit contributions
ling+0.226
 around+0.201
 shining+0.198
led+0.189
 up+0.183
Neg logit contributions
roller-0.220
 salon-0.207
ara-0.196
 heaven-0.188
atters-0.188
Token.
Token ID13
Feature activation-2.25e-01

Pos logit contributions
ling+0.554
 shining+0.539
 around+0.501
 up+0.470
ging+0.458
Neg logit contributions
ara-0.662
roller-0.655
 salon-0.654
ress-0.596
lor-0.595
Token
Token ID198
Feature activation-1.22e-01

Pos logit contributions
 around+0.008
 him+0.008
 shining+0.008
ling+0.008
 Joey+0.008
Neg logit contributions
 salon-0.012
roller-0.012
atters-0.012
 pitcher-0.011
 heaven-0.011
Token
Token ID198
Feature activation-1.31e-01

Pos logit contributions
ling+0.005
 around+0.004
 him+0.004
 shining+0.004
 up+0.004
Neg logit contributions
 salon-0.007
roller-0.007
 pitcher-0.006
atters-0.006
 heaven-0.006
Tokenâ
Token ID22940
Feature activation+2.44e+00

Pos logit contributions
ling+0.006
 around+0.005
 him+0.005
 shining+0.005
 Barry+0.005
Neg logit contributions
roller-0.007
atters-0.006
 salon-0.006
 heaven-0.006
 pitcher-0.006
Token
Token ID26391
Feature activation-2.84e+00

Pos logit contributions
roller+0.119
 salon+0.114
 pitcher+0.105
 clinic+0.103
 heaven+0.102
Neg logit contributions
ling-0.128
 around-0.114
 shining-0.107
led-0.107
 up-0.104
Token stretch
Token ID7539
Feature activation-7.50e-02

Pos logit contributions
ling+0.153
 around+0.135
 shining+0.132
led+0.126
 up+0.122
Neg logit contributions
roller-0.165
 salon-0.156
ara-0.147
atters-0.142
 heaven-0.142
Token out
Token ID503
Feature activation-1.06e+00

Pos logit contributions
ling+0.003
 shining+0.002
led+0.002
 around+0.002
itive+0.002
Neg logit contributions
roller-0.005
 salon-0.004
post-0.004
ara-0.004
 heaven-0.004
Token and
Token ID290
Feature activation-8.02e-01

Pos logit contributions
ling+0.042
 shining+0.035
 around+0.035
led+0.034
ie+0.031
Neg logit contributions
roller-0.061
 salon-0.057
 heaven-0.055
ara-0.054
post-0.054
Token enormous
Token ID9812
Feature activation-2.42e+00

Pos logit contributions
ling+0.037
led+0.032
 shining+0.030
 around+0.030
ie+0.029
Neg logit contributions
roller-0.040
 salon-0.039
 heaven-0.037
 pitcher-0.036
 clinic-0.036
Token sun
Token ID4252
Feature activation+1.67e+00

Pos logit contributions
ling+0.101
 around+0.088
 shining+0.086
led+0.081
 up+0.078
Neg logit contributions
roller-0.139
 salon-0.132
ara-0.125
atters-0.122
 heaven-0.122
Tokenfl
Token ID2704
Feature activation-1.31e+01

Pos logit contributions
roller+0.104
 salon+0.100
ara+0.096
post+0.094
 pitcher+0.093
Neg logit contributions
ling-0.056
 around-0.044
 shining-0.043
led-0.042
 him-0.040
Tokenowers
Token ID3618
Feature activation-6.62e+00

Pos logit contributions
ling+0.507
 around+0.432
 shining+0.421
led+0.397
 up+0.381
Neg logit contributions
roller-0.791
 salon-0.753
ara-0.716
 heaven-0.698
atters-0.698
Token would
Token ID561
Feature activation-3.18e+00

Pos logit contributions
ling+0.239
 shining+0.190
 around+0.186
led+0.186
 him+0.175
Neg logit contributions
roller-0.384
 salon-0.372
post-0.352
 heaven-0.351
ara-0.347
Token show
Token ID905
Feature activation+2.31e+00

Pos logit contributions
ling+0.154
 around+0.138
 shining+0.135
led+0.127
 up+0.125
Neg logit contributions
roller-0.162
 salon-0.152
ara-0.144
atters-0.139
 pitcher-0.139
Token their
Token ID511
Feature activation+2.43e+00

Pos logit contributions
roller+0.127
 salon+0.121
 heaven+0.121
ara+0.118
post+0.117
Neg logit contributions
ling-0.089
 shining-0.074
led-0.071
 around-0.069
 atop-0.064
Token bright
Token ID6016
Feature activation-1.68e+00

Pos logit contributions
roller+0.129
 salon+0.124
ara+0.115
 heaven+0.114
 clinic+0.114
Neg logit contributions
ling-0.108
led-0.091
 around-0.090
 shining-0.087
 him-0.083
Token replied
Token ID8712
Feature activation-2.34e+00

Pos logit contributions
 around+0.129
ling+0.123
 in+0.115
 couldn+0.115
 walking+0.114
Neg logit contributions
 salon-0.130
 vet-0.123
roller-0.122
 heaven-0.119
 pitcher-0.119
Token his
Token ID465
Feature activation-1.99e+00

Pos logit contributions
ling+0.092
 around+0.081
 shining+0.078
led+0.073
 up+0.072
Neg logit contributions
roller-0.140
 salon-0.134
ara-0.127
atters-0.124
 heaven-0.124
Token mum
Token ID25682
Feature activation-1.73e+00

Pos logit contributions
ling+0.082
 around+0.072
 shining+0.070
 up+0.065
led+0.065
Neg logit contributions
roller-0.114
 salon-0.109
ara-0.103
 heaven-0.101
atters-0.101
Token.
Token ID13
Feature activation-7.35e-01

Pos logit contributions
ling+0.077
 shining+0.066
 around+0.064
led+0.063
 up+0.058
Neg logit contributions
roller-0.093
 salon-0.089
ara-0.086
 heaven-0.082
ress-0.082
Token T
Token ID309
Feature activation-4.98e+00

Pos logit contributions
 around+0.030
 Joey+0.029
 him+0.028
 Barry+0.028
 Bobby+0.027
Neg logit contributions
 salon-0.037
roller-0.036
atters-0.035
 heaven-0.034
 bidder-0.033
Tokenod
Token ID375
Feature activation-1.30e+01

Pos logit contributions
ling+0.186
 around+0.158
 shining+0.153
led+0.144
 up+0.137
Neg logit contributions
roller-0.310
 salon-0.296
ara-0.283
atters-0.275
 heaven-0.273
Token nodded
Token ID14464
Feature activation-6.90e-02

Pos logit contributions
ling+0.459
 shining+0.440
 around+0.406
 up+0.380
ging+0.352
Neg logit contributions
ara-0.750
roller-0.749
 salon-0.744
post-0.679
lor-0.678
Token and
Token ID290
Feature activation-8.29e-01

Pos logit contributions
ling+0.003
led+0.003
 around+0.002
 shining+0.002
uses+0.002
Neg logit contributions
roller-0.004
 salon-0.003
ara-0.003
 heaven-0.003
post-0.003
Token agreed
Token ID4987
Feature activation+5.28e-01

Pos logit contributions
ling+0.033
 around+0.029
 shining+0.028
led+0.027
 up+0.026
Neg logit contributions
roller-0.050
 salon-0.047
ara-0.045
atters-0.044
 clinic-0.044
Token to
Token ID284
Feature activation-1.14e-01

Pos logit contributions
roller+0.033
 salon+0.031
ara+0.030
atters+0.029
 heaven+0.029
Neg logit contributions
ling-0.020
 around-0.017
 shining-0.017
led-0.016
 up-0.015
Token put
Token ID1234
Feature activation+2.34e-01

Pos logit contributions
ling+0.005
 around+0.005
 shining+0.004
 him+0.004
 up+0.004
Neg logit contributions
roller-0.006
 salon-0.006
ara-0.006
 pitcher-0.005
post-0.005
Token,
Token ID11
Feature activation-4.98e-01

Pos logit contributions
roller+0.084
ara+0.077
 salon+0.077
atters+0.070
S+0.070
Neg logit contributions
ling-0.093
led-0.080
 around-0.076
 shining-0.075
uses-0.072
Token there
Token ID612
Feature activation-5.61e-01

Pos logit contributions
uses+0.020
ling+0.019
 him+0.017
 Barry+0.017
 questions+0.016
Neg logit contributions
ara-0.026
Anna-0.025
roller-0.024
S-0.024
 salon-0.023
Token was
Token ID373
Feature activation-6.03e-02

Pos logit contributions
ling+0.025
 shining+0.021
led+0.021
 around+0.021
ging+0.019
Neg logit contributions
roller-0.030
ara-0.029
 salon-0.029
post-0.026
 heaven-0.026
Token a
Token ID257
Feature activation-3.11e-01

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
led+0.002
ie+0.002
Neg logit contributions
roller-0.004
 salon-0.004
ara-0.004
 heaven-0.004
 pitcher-0.003
Token do
Token ID466
Feature activation-1.77e+00

Pos logit contributions
ling+0.012
 around+0.010
led+0.010
 shining+0.010
 him+0.009
Neg logit contributions
roller-0.019
 salon-0.018
ara-0.018
 pitcher-0.017
 clinic-0.017
Tokenod
Token ID375
Feature activation-1.30e+01

Pos logit contributions
ling+0.105
 around+0.095
 shining+0.094
led+0.090
 up+0.088
Neg logit contributions
roller-0.070
 salon-0.066
ara-0.060
atters-0.059
 heaven-0.058
Tokenle
Token ID293
Feature activation-5.25e+00

Pos logit contributions
 shining+0.402
 Joey+0.376
 Barry+0.368
 around+0.365
ling+0.356
Neg logit contributions
 salon-0.811
ara-0.780
ase-0.732
S-0.726
ress-0.720
Token who
Token ID508
Feature activation-6.44e+00

Pos logit contributions
ling+0.258
 shining+0.239
led+0.234
 around+0.231
 up+0.221
Neg logit contributions
 salon-0.223
roller-0.216
 heaven-0.212
 clinic-0.210
ara-0.210
Token wanted
Token ID2227
Feature activation-2.96e+00

Pos logit contributions
ling+0.214
 shining+0.170
 around+0.169
led+0.157
 up+0.151
Neg logit contributions
roller-0.422
 salon-0.403
ara-0.392
 heaven-0.389
post-0.384
Token to
Token ID284
Feature activation-5.15e-02

Pos logit contributions
ling+0.131
led+0.106
 around+0.104
 shining+0.102
ie+0.096
Neg logit contributions
roller-0.172
 salon-0.165
 heaven-0.154
ara-0.153
 clinic-0.152
Token make
Token ID787
Feature activation+3.29e-01

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
 him+0.002
 up+0.002
Neg logit contributions
roller-0.003
ara-0.003
 salon-0.003
post-0.003
atters-0.003
Token thing
Token ID1517
Feature activation-4.78e-01

Pos logit contributions
roller+0.252
 salon+0.240
ara+0.227
 heaven+0.222
 clinic+0.220
Neg logit contributions
ling-0.197
 around-0.171
 shining-0.167
led-0.160
 up-0.153
Token,
Token ID11
Feature activation-2.91e-01

Pos logit contributions
ling+0.023
 shining+0.019
 around+0.018
led+0.018
ie+0.017
Neg logit contributions
roller-0.025
 salon-0.024
ara-0.022
 heaven-0.022
atters-0.022
Token John
Token ID1757
Feature activation-2.15e+00

Pos logit contributions
ling+0.013
 around+0.011
 shining+0.011
 up+0.010
led+0.010
Neg logit contributions
roller-0.016
 salon-0.015
ara-0.014
atters-0.014
 clinic-0.014
Token.
Token ID13
Feature activation-7.15e-01

Pos logit contributions
ling+0.098
 around+0.085
 shining+0.084
led+0.080
 up+0.077
Neg logit contributions
roller-0.115
 salon-0.109
ara-0.104
 heaven-0.101
 clinic-0.100
Token N
Token ID399
Feature activation-9.40e-02

Pos logit contributions
 around+0.032
ling+0.031
 him+0.031
 Joey+0.030
 shining+0.030
Neg logit contributions
roller-0.033
 salon-0.033
post-0.030
 pitcher-0.030
ara-0.029
Tokenod
Token ID375
Feature activation-1.29e+01

Pos logit contributions
ling+0.005
 around+0.005
led+0.004
 in+0.004
ie+0.004
Neg logit contributions
roller-0.004
 salon-0.004
 pitcher-0.004
atters-0.003
post-0.003
Token if
Token ID611
Feature activation-3.63e+00

Pos logit contributions
ling+0.651
 shining+0.604
 around+0.582
 up+0.546
led+0.538
Neg logit contributions
roller-0.572
 salon-0.563
ara-0.542
atters-0.510
 heaven-0.506
Token you
Token ID345
Feature activation-4.62e-01

Pos logit contributions
ling+0.198
 around+0.174
 shining+0.170
led+0.167
 up+0.158
Neg logit contributions
roller-0.163
 salon-0.157
ara-0.146
 heaven-0.142
 clinic-0.139
Token're
Token ID821
Feature activation-2.06e+00

Pos logit contributions
ling+0.023
 shining+0.020
 around+0.019
led+0.019
 up+0.018
Neg logit contributions
roller-0.023
 salon-0.022
ara-0.020
 heaven-0.020
atters-0.020
Token tired
Token ID10032
Feature activation+3.56e-01

Pos logit contributions
ling+0.086
led+0.073
 around+0.070
 shining+0.068
ie+0.067
Neg logit contributions
roller-0.113
 salon-0.111
 heaven-0.107
 clinic-0.105
ara-0.101
Token,
Token ID11
Feature activation-5.83e-01

Pos logit contributions
roller+0.020
 salon+0.019
ara+0.018
 heaven+0.018
atters+0.018
Neg logit contributions
ling-0.015
 around-0.013
 shining-0.013
led-0.012
 up-0.012
Token<|endoftext|>
Token ID50256
Feature activation+1.13e+00

Pos logit contributions
roller+0.136
 salon+0.130
atters+0.119
ara+0.118
 pitcher+0.117
Neg logit contributions
ling-0.130
 around-0.127
 shining-0.118
 him-0.117
 up-0.113
TokenL
Token ID43
Feature activation+3.69e+00

Pos logit contributions
roller+0.052
 salon+0.048
ara+0.047
 heaven+0.045
post+0.044
Neg logit contributions
ling-0.060
 shining-0.052
 around-0.052
led-0.051
ie-0.048
Tokenily
Token ID813
Feature activation+1.58e+00

Pos logit contributions
roller+0.174
 salon+0.162
ara+0.148
 clinic+0.148
 pitcher+0.148
Neg logit contributions
ling-0.195
 around-0.173
 shining-0.168
led-0.163
ie-0.162
Token loved
Token ID6151
Feature activation-1.74e+00

Pos logit contributions
roller+0.096
 salon+0.091
ara+0.085
post+0.085
 pitcher+0.085
Neg logit contributions
ling-0.062
 around-0.054
 shining-0.050
led-0.049
 up-0.047
Token sun
Token ID4252
Feature activation+1.69e+00

Pos logit contributions
ling+0.088
led+0.074
 around+0.068
 atop+0.068
ging+0.067
Neg logit contributions
 salon-0.090
roller-0.083
 heaven-0.079
 pitcher-0.077
ara-0.076
Tokenfl
Token ID2704
Feature activation-1.28e+01

Pos logit contributions
 salon+0.088
roller+0.085
ara+0.080
 pitcher+0.077
 clinic+0.075
Neg logit contributions
ling-0.073
 around-0.064
led-0.063
 spear-0.060
 Barry-0.058
Tokenowers
Token ID3618
Feature activation-6.09e+00

Pos logit contributions
ling+0.542
 around+0.471
 shining+0.455
led+0.433
 up+0.420
Neg logit contributions
roller-0.725
 salon-0.693
ara-0.659
 pitcher-0.637
atters-0.634
Token.
Token ID13
Feature activation-7.20e-01

Pos logit contributions
ling+0.255
led+0.232
ging+0.217
ged+0.211
 around+0.208
Neg logit contributions
 salon-0.289
roller-0.284
 heaven-0.272
 pitcher-0.271
 clinic-0.269
Token She
Token ID1375
Feature activation+1.64e+00

Pos logit contributions
ling+0.036
 around+0.035
 him+0.034
 shining+0.034
 Joey+0.034
Neg logit contributions
roller-0.033
 salon-0.031
 clinic-0.028
atters-0.028
lor-0.028
Token liked
Token ID8288
Feature activation-2.47e+00

Pos logit contributions
roller+0.108
 salon+0.104
ara+0.101
 heaven+0.097
post+0.097
Neg logit contributions
ling-0.054
 around-0.043
 shining-0.042
led-0.039
 up-0.038
Token to
Token ID284
Feature activation+1.03e-01

Pos logit contributions
ling+0.130
led+0.115
ging+0.108
 around+0.101
itive+0.101
Neg logit contributions
 salon-0.109
roller-0.104
 heaven-0.098
 pitcher-0.096
ara-0.096
Token a
Token ID257
Feature activation-3.56e-01

Pos logit contributions
roller+0.065
 heaven+0.061
 salon+0.060
 jail+0.058
post+0.058
Neg logit contributions
ling-0.048
led-0.036
 around-0.036
ie-0.034
 shining-0.033
Token garden
Token ID11376
Feature activation+3.17e+00

Pos logit contributions
ling+0.015
 shining+0.013
 around+0.013
 up+0.012
led+0.011
Neg logit contributions
roller-0.020
 salon-0.019
ara-0.019
atters-0.019
 heaven-0.018
Token full
Token ID1336
Feature activation+2.09e+00

Pos logit contributions
roller+0.207
ara+0.201
 salon+0.199
atters+0.199
 heaven+0.185
Neg logit contributions
ling-0.098
 around-0.098
 shining-0.091
 in-0.089
 up-0.079
Token of
Token ID286
Feature activation+1.20e+00

Pos logit contributions
roller+0.116
 salon+0.112
ara+0.106
atters+0.105
 heaven+0.101
Neg logit contributions
ling-0.090
 around-0.083
 shining-0.081
 in-0.073
 up-0.072
Token wild
Token ID4295
Feature activation-1.21e+00

Pos logit contributions
roller+0.060
 salon+0.057
 heaven+0.057
 jail+0.055
 pitcher+0.055
Neg logit contributions
ling-0.057
led-0.047
 around-0.045
ging-0.043
ie-0.043
Tokenfl
Token ID2704
Feature activation-1.28e+01

Pos logit contributions
ling+0.047
 around+0.038
led+0.037
 shining+0.036
bled+0.036
Neg logit contributions
roller-0.068
 heaven-0.062
ara-0.062
 jail-0.061
 clinic-0.061
Tokenowers
Token ID3618
Feature activation-6.18e+00

Pos logit contributions
ling+0.528
 around+0.460
 shining+0.442
led+0.416
 up+0.411
Neg logit contributions
roller-0.731
 salon-0.688
ara-0.655
atters-0.646
 pitcher-0.641
Token.
Token ID13
Feature activation-4.17e-01

Pos logit contributions
ling+0.278
led+0.237
 around+0.226
 shining+0.226
 him+0.216
Neg logit contributions
roller-0.308
 salon-0.285
 heaven-0.283
 pitcher-0.273
 clinic-0.272
Token He
Token ID679
Feature activation-1.16e+00

Pos logit contributions
ling+0.017
 around+0.016
 shining+0.015
 him+0.015
 up+0.015
Neg logit contributions
roller-0.023
 salon-0.022
ara-0.020
atters-0.020
 pitcher-0.020
Token decided
Token ID3066
Feature activation+7.12e-01

Pos logit contributions
ling+0.046
 shining+0.039
 around+0.039
led+0.036
 up+0.035
Neg logit contributions
roller-0.068
 salon-0.066
ara-0.063
 clinic-0.061
 heaven-0.060
Token to
Token ID284
Feature activation+1.53e-02

Pos logit contributions
roller+0.036
 salon+0.034
post+0.032
 vet+0.031
 heaven+0.031
Neg logit contributions
ling-0.034
led-0.029
 shining-0.028
ie-0.027
 around-0.027
Token the
Token ID262
Feature activation-7.50e-01

Pos logit contributions
ling+0.036
 shining+0.028
led+0.028
 around+0.027
bee+0.026
Neg logit contributions
roller-0.044
 salon-0.043
 heaven-0.042
 vet-0.041
ara-0.041
Token biggest
Token ID4094
Feature activation-4.52e+00

Pos logit contributions
ling+0.034
led+0.030
 around+0.029
 him+0.026
ie+0.026
Neg logit contributions
roller-0.039
 salon-0.039
 pitcher-0.036
ara-0.036
 clinic-0.035
Token and
Token ID290
Feature activation-6.85e-01

Pos logit contributions
ling+0.200
 around+0.175
 shining+0.172
led+0.161
 up+0.157
Neg logit contributions
roller-0.252
 salon-0.239
ara-0.226
atters-0.220
 heaven-0.220
Token brightest
Token ID33871
Feature activation-6.09e+00

Pos logit contributions
ling+0.033
led+0.028
ging+0.027
ged+0.027
 around+0.027
Neg logit contributions
 salon-0.031
 pitcher-0.030
roller-0.030
 heaven-0.028
ara-0.028
Token sun
Token ID4252
Feature activation+1.88e+00

Pos logit contributions
ling+0.265
 around+0.228
led+0.220
 shining+0.218
 up+0.204
Neg logit contributions
roller-0.333
 salon-0.318
ara-0.304
 pitcher-0.295
 heaven-0.293
Tokenfl
Token ID2704
Feature activation-1.27e+01

Pos logit contributions
roller+0.124
ara+0.119
 salon+0.118
post+0.114
 pitcher+0.114
Neg logit contributions
ling-0.050
led-0.036
 around-0.036
 spear-0.036
nam-0.035
Tokenowers
Token ID3618
Feature activation-6.27e+00

Pos logit contributions
ling+0.572
 around+0.507
 shining+0.487
 up+0.465
 him+0.457
Neg logit contributions
roller-0.637
 salon-0.601
ara-0.588
 pitcher-0.572
atters-0.566
Token she
Token ID673
Feature activation-2.54e-01

Pos logit contributions
ling+0.235
led+0.187
 shining+0.184
itive+0.181
 around+0.180
Neg logit contributions
roller-0.333
 salon-0.322
 pitcher-0.312
post-0.311
 heaven-0.311
Token could
Token ID714
Feature activation-2.81e+00

Pos logit contributions
ling+0.010
 shining+0.008
 around+0.008
 up+0.008
led+0.008
Neg logit contributions
roller-0.014
 salon-0.013
ara-0.013
post-0.013
 heaven-0.013
Token find
Token ID1064
Feature activation-1.28e+00

Pos logit contributions
ling+0.129
 around+0.114
 shining+0.111
led+0.104
 up+0.102
Neg logit contributions
roller-0.152
 salon-0.143
ara-0.136
 heaven-0.131
atters-0.131
Token.
Token ID13
Feature activation-4.32e-01

Pos logit contributions
ling+0.055
 shining+0.046
 around+0.044
led+0.044
 spear+0.040
Neg logit contributions
roller-0.066
 salon-0.064
 heaven-0.063
ara-0.062
post-0.061
TokenThe
Token ID464
Feature activation+7.33e-01

Pos logit contributions
ling+0.020
 around+0.017
 shining+0.017
 him+0.016
 in+0.016
Neg logit contributions
roller-0.021
 salon-0.019
atters-0.019
 heaven-0.019
 clinic-0.018
Token next
Token ID1306
Feature activation-5.95e-01

Pos logit contributions
roller+0.042
 salon+0.041
ara+0.038
 pitcher+0.038
 heaven+0.037
Neg logit contributions
ling-0.031
 around-0.026
led-0.026
 shining-0.025
 up-0.023
Token day
Token ID1110
Feature activation-1.70e+00

Pos logit contributions
ling+0.017
 around+0.015
 shining+0.014
 in+0.012
 up+0.012
Neg logit contributions
roller-0.043
 salon-0.041
atters-0.040
 heaven-0.039
 clinic-0.039
Token,
Token ID11
Feature activation-5.59e-01

Pos logit contributions
ling+0.053
 around+0.048
 shining+0.044
 in+0.044
 him+0.043
Neg logit contributions
roller-0.114
 salon-0.110
atters-0.109
 heaven-0.105
 clinic-0.104
Token T
Token ID309
Feature activation-4.85e+00

Pos logit contributions
ling+0.023
 around+0.021
 shining+0.020
 up+0.019
 him+0.019
Neg logit contributions
roller-0.032
 salon-0.031
atters-0.030
 clinic-0.029
ara-0.029
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.167
 around+0.140
 shining+0.136
led+0.127
 up+0.120
Neg logit contributions
roller-0.316
 salon-0.302
ara-0.289
atters-0.281
 heaven-0.281
Token and
Token ID290
Feature activation-6.30e-01

Pos logit contributions
ling+0.254
 shining+0.237
 around+0.225
 up+0.187
nam+0.160
Neg logit contributions
ara-0.928
roller-0.924
 salon-0.924
ress-0.861
lor-0.857
Token his
Token ID465
Feature activation-1.81e+00

Pos logit contributions
ling+0.023
 around+0.019
 shining+0.019
led+0.018
 up+0.017
Neg logit contributions
roller-0.040
 salon-0.038
ara-0.036
atters-0.036
 clinic-0.035
Token mom
Token ID1995
Feature activation+7.33e-01

Pos logit contributions
ling+0.077
led+0.063
 around+0.063
 shining+0.058
ging+0.057
Neg logit contributions
roller-0.104
 salon-0.098
ara-0.094
 pitcher-0.094
 vet-0.091
Token opened
Token ID4721
Feature activation+4.26e+00

Pos logit contributions
roller+0.044
 salon+0.042
ara+0.040
 pitcher+0.039
atters+0.038
Neg logit contributions
ling-0.029
 around-0.024
 shining-0.024
led-0.023
 up-0.021
Token the
Token ID262
Feature activation-5.82e-01

Pos logit contributions
roller+0.279
 salon+0.264
ara+0.256
 heaven+0.252
 pitcher+0.247
Neg logit contributions
ling-0.146
led-0.110
 shining-0.109
 around-0.107
ie-0.099
Token garden
Token ID11376
Feature activation+3.47e+00

Pos logit contributions
roller+0.085
 salon+0.081
ara+0.075
 clinic+0.074
 heaven+0.074
Neg logit contributions
ling-0.078
 around-0.068
 shining-0.066
led-0.066
ie-0.062
Token and
Token ID290
Feature activation-4.09e-01

Pos logit contributions
roller+0.207
 salon+0.196
ara+0.188
atters+0.183
 heaven+0.182
Neg logit contributions
ling-0.138
 around-0.119
 shining-0.115
led-0.109
 up-0.106
Token admire
Token ID21099
Feature activation+5.21e-01

Pos logit contributions
ling+0.017
 around+0.014
 shining+0.014
led+0.014
 up+0.013
Neg logit contributions
roller-0.024
 salon-0.023
ara-0.021
 heaven-0.021
 pitcher-0.021
Token his
Token ID465
Feature activation-1.22e+00

Pos logit contributions
 heaven+0.028
 salon+0.028
roller+0.028
ara+0.028
 yourselves+0.026
Neg logit contributions
ling-0.020
led-0.016
 around-0.015
 shining-0.015
 atop-0.014
Token sun
Token ID4252
Feature activation+1.81e+00

Pos logit contributions
ling+0.057
led+0.049
 around+0.049
 him+0.045
 shining+0.045
Neg logit contributions
roller-0.061
 salon-0.058
 heaven-0.055
 pitcher-0.055
 clinic-0.054
Tokenfl
Token ID2704
Feature activation-1.27e+01

Pos logit contributions
roller+0.099
 salon+0.098
ara+0.090
 pitcher+0.090
 heaven+0.088
Neg logit contributions
ling-0.064
led-0.055
 around-0.054
 Barry-0.051
ging-0.050
Tokenowers
Token ID3618
Feature activation-6.22e+00

Pos logit contributions
ling+0.561
 around+0.501
 shining+0.477
 him+0.448
led+0.448
Neg logit contributions
roller-0.649
 salon-0.625
ara-0.597
 pitcher-0.583
atters-0.566
Token.
Token ID13
Feature activation-4.74e-02

Pos logit contributions
ling+0.263
led+0.231
 around+0.220
 shining+0.212
ged+0.208
Neg logit contributions
roller-0.293
 salon-0.292
 heaven-0.280
 pitcher-0.272
 clinic-0.270
Token They
Token ID1119
Feature activation+3.01e+00

Pos logit contributions
 around+0.002
 him+0.002
ling+0.002
 Joey+0.002
 up+0.002
Neg logit contributions
 salon-0.003
roller-0.003
post-0.002
ction-0.002
 pitcher-0.002
Token also
Token ID635
Feature activation+1.57e+00

Pos logit contributions
 salon+0.148
roller+0.145
ara+0.135
 heaven+0.134
 pitcher+0.132
Neg logit contributions
ling-0.143
 shining-0.122
led-0.120
 around-0.118
 up-0.111
Token see
Token ID766
Feature activation+6.12e-01

Pos logit contributions
roller+0.086
 salon+0.081
ara+0.077
 heaven+0.075
atters+0.075
Neg logit contributions
ling-0.070
 around-0.062
 shining-0.060
led-0.058
 up-0.055
Token da
Token ID12379
Feature activation-2.91e+00

Pos logit contributions
roller+0.011
 salon+0.011
ara+0.010
 heaven+0.010
 pitcher+0.010
Neg logit contributions
ling-0.009
 around-0.007
led-0.007
 shining-0.007
ie-0.006
Tokenis
Token ID271
Feature activation-3.83e+00

Pos logit contributions
ling+0.157
 around+0.141
 shining+0.138
led+0.133
 up+0.130
Neg logit contributions
roller-0.131
 salon-0.123
ara-0.116
 heaven-0.111
atters-0.110
Tokenies
Token ID444
Feature activation-2.89e+00

Pos logit contributions
ling+0.101
 around+0.089
 shining+0.084
 up+0.074
 spear+0.072
Neg logit contributions
roller-0.258
ara-0.252
 salon-0.251
atters-0.235
post-0.234
Token,
Token ID11
Feature activation-2.94e-01

Pos logit contributions
ling+0.120
 around+0.102
 shining+0.099
led+0.098
 up+0.091
Neg logit contributions
roller-0.163
 salon-0.157
ara-0.149
 heaven-0.144
 clinic-0.144
Token sun
Token ID4252
Feature activation+1.91e+00

Pos logit contributions
ling+0.013
 around+0.011
led+0.011
 shining+0.011
 up+0.010
Neg logit contributions
roller-0.015
 salon-0.015
ara-0.015
 heaven-0.014
 pitcher-0.014
Tokenfl
Token ID2704
Feature activation-1.27e+01

Pos logit contributions
roller+0.120
 salon+0.115
ara+0.114
 pitcher+0.108
post+0.106
Neg logit contributions
ling-0.064
 around-0.053
led-0.048
 shining-0.047
 him-0.044
Tokenowers
Token ID3618
Feature activation-6.38e+00

Pos logit contributions
ling+0.505
 around+0.432
 shining+0.418
led+0.393
 up+0.382
Neg logit contributions
roller-0.750
 salon-0.718
ara-0.688
atters-0.665
 pitcher-0.664
Token,
Token ID11
Feature activation-4.02e-01

Pos logit contributions
ling+0.269
 around+0.227
led+0.221
 shining+0.219
 him+0.206
Neg logit contributions
roller-0.343
 salon-0.329
ara-0.316
 pitcher-0.306
 heaven-0.306
Token and
Token ID290
Feature activation-4.24e-01

Pos logit contributions
ling+0.018
 around+0.016
led+0.015
 shining+0.015
 up+0.014
Neg logit contributions
roller-0.020
 salon-0.020
ara-0.019
 heaven-0.019
 pitcher-0.019
Token tul
Token ID48373
Feature activation-2.67e+00

Pos logit contributions
ling+0.018
led+0.015
 around+0.014
 shining+0.014
ie+0.013
Neg logit contributions
 salon-0.022
roller-0.022
 heaven-0.022
 pitcher-0.021
ara-0.021
Tokenips
Token ID2419
Feature activation+7.70e-01

Pos logit contributions
ling+0.129
 around+0.112
 shining+0.109
led+0.109
ie+0.105
Neg logit contributions
roller-0.142
 salon-0.128
post-0.121
 heaven-0.121
 clinic-0.119
Token El
Token ID2574
Feature activation-3.49e+00

Pos logit contributions
roller+0.066
 salon+0.062
atters+0.058
 clinic+0.058
 pitcher+0.057
Neg logit contributions
 around-0.054
ling-0.052
 in-0.050
 shining-0.050
 him-0.050
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.167
 shining+0.166
 around+0.153
 him+0.147
 spear+0.143
Neg logit contributions
ara-0.164
 salon-0.157
roller-0.150
 heaven-0.142
 clinic-0.138
Tokenie
Token ID494
Feature activation-6.27e+00

Pos logit contributions
 shining+0.379
 Barry+0.326
ging+0.323
 Joey+0.320
ling+0.304
Neg logit contributions
ara-0.804
 salon-0.772
S-0.736
roller-0.705
ress-0.704
Token.
Token ID13
Feature activation-5.25e-01

Pos logit contributions
ling+0.282
 shining+0.242
led+0.239
 around+0.238
 up+0.216
Neg logit contributions
roller-0.335
 salon-0.318
ara-0.302
 heaven-0.301
 clinic-0.296
Token El
Token ID2574
Feature activation-3.52e+00

Pos logit contributions
ling+0.034
 around+0.031
 shining+0.030
led+0.029
 up+0.029
Neg logit contributions
roller-0.019
 salon-0.017
ara-0.016
 heaven-0.015
atters-0.015
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.177
 shining+0.162
 around+0.157
 him+0.148
 up+0.145
Neg logit contributions
roller-0.160
 salon-0.159
ara-0.157
atters-0.142
 pitcher-0.142
Tokenie
Token ID494
Feature activation-6.29e+00

Pos logit contributions
 shining+0.506
ging+0.460
ling+0.437
 Barry+0.422
nam+0.421
Neg logit contributions
ara-0.695
 salon-0.614
S-0.592
ase-0.590
Anna-0.586
Token loved
Token ID6151
Feature activation-2.37e+00

Pos logit contributions
ling+0.165
 around+0.125
 shining+0.123
led+0.112
 up+0.102
Neg logit contributions
roller-0.458
 salon-0.441
ara-0.426
 heaven-0.416
atters-0.415
Token to
Token ID284
Feature activation+1.55e-01

Pos logit contributions
ling+0.117
led+0.101
 around+0.090
 atop+0.086
ging+0.086
Neg logit contributions
 salon-0.123
roller-0.119
 heaven-0.113
 clinic-0.107
 pitcher-0.107
Token think
Token ID892
Feature activation-3.64e+00

Pos logit contributions
roller+0.008
 salon+0.007
 heaven+0.007
ara+0.007
 pitcher+0.007
Neg logit contributions
ling-0.008
 around-0.007
led-0.007
 shining-0.007
 up-0.006
Token.
Token ID13
Feature activation-5.31e-01

Pos logit contributions
ling+0.185
led+0.157
 around+0.148
 shining+0.147
ging+0.142
Neg logit contributions
roller-0.172
 salon-0.172
post-0.156
ress-0.152
 heaven-0.152
Token think
Token ID892
Feature activation-3.64e+00

Pos logit contributions
ling+0.029
led+0.025
 shining+0.024
 around+0.023
ging+0.023
Neg logit contributions
 salon-0.023
roller-0.023
 heaven-0.022
 clinic-0.021
ara-0.020
Token.
Token ID13
Feature activation-5.22e-01

Pos logit contributions
ling+0.167
led+0.138
 shining+0.129
uses+0.129
 spear+0.128
Neg logit contributions
roller-0.190
 salon-0.188
post-0.179
ress-0.170
ara-0.168
Token
Token ID198
Feature activation-2.14e-01

Pos logit contributions
ling+0.030
led+0.026
 shining+0.026
 around+0.026
ie+0.024
Neg logit contributions
roller-0.023
 salon-0.021
ara-0.020
 clinic-0.019
post-0.019
Token
Token ID198
Feature activation-2.01e-01

Pos logit contributions
ling+0.014
led+0.013
 shining+0.012
itive+0.012
 around+0.012
Neg logit contributions
roller-0.007
ara-0.007
 salon-0.006
post-0.006
S-0.006
TokenEl
Token ID9527
Feature activation-5.81e-01

Pos logit contributions
ling+0.013
led+0.012
 shining+0.011
 around+0.011
 up+0.011
Neg logit contributions
roller-0.007
 salon-0.007
ara-0.006
post-0.006
 clinic-0.006
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.028
 shining+0.025
 around+0.025
 him+0.024
led+0.023
Neg logit contributions
roller-0.028
 salon-0.028
ara-0.027
ress-0.025
atters-0.025
Tokenie
Token ID494
Feature activation-6.31e+00

Pos logit contributions
 shining+0.318
nam+0.255
ging+0.252
ling+0.246
 Barry+0.244
Neg logit contributions
ara-0.884
 salon-0.830
Anna-0.782
ue-0.759
lor-0.758
Token would
Token ID561
Feature activation-2.72e+00

Pos logit contributions
ling+0.272
 shining+0.241
 around+0.225
led+0.221
ging+0.211
Neg logit contributions
 salon-0.336
roller-0.335
ara-0.320
 clinic-0.309
 heaven-0.307
Token think
Token ID892
Feature activation-3.64e+00

Pos logit contributions
ling+0.146
led+0.129
 shining+0.123
 around+0.122
ie+0.114
Neg logit contributions
 salon-0.124
roller-0.124
 heaven-0.111
 clinic-0.110
ara-0.108
Token about
Token ID546
Feature activation-2.24e+00

Pos logit contributions
ling+0.176
led+0.149
 shining+0.141
ging+0.141
 spear+0.134
Neg logit contributions
 salon-0.184
roller-0.173
 heaven-0.161
ress-0.157
post-0.157
Token all
Token ID477
Feature activation-1.20e+00

Pos logit contributions
ling+0.099
led+0.091
ging+0.072
bled+0.071
 around+0.071
Neg logit contributions
 salon-0.126
 heaven-0.124
 clinic-0.114
 jail-0.114
roller-0.113
Token so
Token ID523
Feature activation-4.41e-01

Pos logit contributions
ling+0.001
led+0.001
 around+0.001
uses+0.001
ie+0.001
Neg logit contributions
 salon-0.002
roller-0.002
 heaven-0.002
 pitcher-0.002
 clinic-0.002
Token excited
Token ID6568
Feature activation-5.90e-01

Pos logit contributions
ling+0.018
 around+0.015
led+0.014
 shining+0.014
ie+0.013
Neg logit contributions
roller-0.025
 salon-0.025
ara-0.023
 pitcher-0.023
atters-0.023
Token to
Token ID284
Feature activation+3.16e-01

Pos logit contributions
ling+0.029
led+0.023
 shining+0.023
 spear+0.023
ging+0.022
Neg logit contributions
roller-0.028
 salon-0.027
ara-0.026
post-0.026
 heaven-0.026
Token start
Token ID923
Feature activation+5.65e+00

Pos logit contributions
roller+0.014
 salon+0.013
atters+0.012
ara+0.012
 pitcher+0.012
Neg logit contributions
ling-0.018
led-0.016
 shining-0.015
 around-0.015
ging-0.014
Token do
Token ID466
Feature activation-1.41e+00

Pos logit contributions
roller+0.298
 salon+0.288
 pitcher+0.263
 heaven+0.263
stand+0.257
Neg logit contributions
ling-0.270
led-0.232
itive-0.207
 around-0.203
ie-0.195
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.052
 around+0.043
 shining+0.042
led+0.041
 up+0.038
Neg logit contributions
roller-0.088
 salon-0.085
ara-0.080
 heaven-0.079
atters-0.078
Tokenling
Token ID1359
Feature activation-9.46e+00

Pos logit contributions
 shining+0.550
ling+0.543
 Joey+0.496
 around+0.491
 Barry+0.475
Neg logit contributions
 salon-0.627
ara-0.604
roller-0.576
ress-0.574
atters-0.561
Token with
Token ID351
Feature activation+1.22e+00

Pos logit contributions
ling+0.390
led+0.320
 shining+0.320
 around+0.293
ie+0.285
Neg logit contributions
roller-0.525
 salon-0.520
 heaven-0.494
post-0.478
 clinic-0.470
Token it
Token ID340
Feature activation-3.16e-01

Pos logit contributions
 salon+0.060
roller+0.060
 heaven+0.057
ara+0.056
 pitcher+0.054
Neg logit contributions
ling-0.062
led-0.052
 shining-0.046
 around-0.045
itive-0.044
Token.
Token ID13
Feature activation-4.19e-01

Pos logit contributions
ling+0.014
 shining+0.012
led+0.012
 around+0.011
 him+0.011
Neg logit contributions
roller-0.017
 salon-0.017
ara-0.015
post-0.015
 heaven-0.015
Token
Token ID198
Feature activation-9.65e-02

Pos logit contributions
ling+0.020
 around+0.019
 shining+0.018
 up+0.017
 him+0.017
Neg logit contributions
roller-0.021
 salon-0.020
 heaven-0.018
 pitcher-0.018
ara-0.018
Token back
Token ID736
Feature activation+2.53e+00

Pos logit contributions
ling+0.017
led+0.015
 shining+0.014
 around+0.014
bee+0.013
Neg logit contributions
roller-0.027
 salon-0.024
 heaven-0.024
ara-0.024
 clinic-0.024
Token with
Token ID351
Feature activation+1.27e+00

Pos logit contributions
roller+0.156
 salon+0.149
ara+0.141
atters+0.139
post+0.138
Neg logit contributions
ling-0.096
 around-0.084
 shining-0.081
 up-0.075
led-0.074
Token lots
Token ID6041
Feature activation+1.01e-01

Pos logit contributions
roller+0.079
 salon+0.075
 heaven+0.073
ara+0.072
 pitcher+0.072
Neg logit contributions
ling-0.047
 around-0.038
 shining-0.036
led-0.036
ie-0.033
Token of
Token ID286
Feature activation+1.47e+00

Pos logit contributions
roller+0.005
 salon+0.005
ara+0.005
atters+0.005
 heaven+0.004
Neg logit contributions
ling-0.005
 around-0.004
 shining-0.004
led-0.004
 up-0.004
Token sun
Token ID4252
Feature activation+1.86e+00

Pos logit contributions
 heaven+0.065
roller+0.065
 salon+0.063
 pitcher+0.062
 vet+0.061
Neg logit contributions
ling-0.069
 around-0.059
led-0.057
 shining-0.055
ging-0.054
Tokenfl
Token ID2704
Feature activation-1.27e+01

Pos logit contributions
roller+0.112
 salon+0.105
ara+0.104
 pitcher+0.102
post+0.098
Neg logit contributions
ling-0.062
 around-0.053
 shining-0.047
 spear-0.047
led-0.047
Tokenowers
Token ID3618
Feature activation-6.04e+00

Pos logit contributions
ling+0.532
 around+0.467
 shining+0.454
 up+0.420
led+0.412
Neg logit contributions
roller-0.702
 salon-0.661
ara-0.638
 pitcher-0.623
atters-0.619
Token to
Token ID284
Feature activation+7.56e-05

Pos logit contributions
ling+0.252
 around+0.210
 shining+0.204
led+0.202
 him+0.194
Neg logit contributions
roller-0.316
 salon-0.299
 pitcher-0.285
 heaven-0.285
ara-0.282
Token fill
Token ID6070
Feature activation+2.07e+00

Pos logit contributions
roller+0.000
 salon+0.000
 heaven+0.000
ara+0.000
 clinic+0.000
Neg logit contributions
ling-0.000
 around-0.000
led-0.000
 shining-0.000
 up-0.000
Token the
Token ID262
Feature activation-1.08e+00

Pos logit contributions
roller+0.123
 salon+0.117
ara+0.116
 heaven+0.116
 pitcher+0.113
Neg logit contributions
ling-0.073
 shining-0.060
 around-0.058
led-0.057
itive-0.052
Token prince
Token ID19716
Feature activation-3.07e+00

Pos logit contributions
ling+0.056
 around+0.050
 shining+0.048
led+0.048
 him+0.046
Neg logit contributions
roller-0.050
 salon-0.047
ara-0.045
 pitcher-0.043
 heaven-0.043
Token other
Token ID584
Feature activation+1.22e+00

Pos logit contributions
ling+0.211
 shining+0.175
 around+0.173
led+0.171
 up+0.150
Neg logit contributions
roller-0.356
 salon-0.345
ara-0.333
 heaven-0.327
 clinic-0.319
Token.
Token ID13
Feature activation-3.79e-01

Pos logit contributions
roller+0.068
 salon+0.064
 heaven+0.061
ara+0.060
post+0.059
Neg logit contributions
ling-0.050
 shining-0.044
led-0.042
 around-0.041
ie-0.039
Token Jim
Token ID5395
Feature activation-1.33e+00

Pos logit contributions
 around+0.015
 Joey+0.015
 Peter+0.015
 him+0.014
 Barry+0.014
Neg logit contributions
 salon-0.021
roller-0.019
 clinic-0.018
atters-0.018
post-0.018
Token and
Token ID290
Feature activation-7.21e-01

Pos logit contributions
ling+0.052
 around+0.044
 shining+0.043
led+0.041
 up+0.039
Neg logit contributions
roller-0.080
 salon-0.077
ara-0.073
 heaven-0.071
atters-0.071
Token J
Token ID449
Feature activation-3.46e+00

Pos logit contributions
ling+0.032
 shining+0.026
led+0.026
 around+0.025
 up+0.023
Neg logit contributions
roller-0.040
 salon-0.039
ara-0.037
 heaven-0.036
ress-0.036
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.176
 around+0.157
 shining+0.157
led+0.147
 up+0.144
Neg logit contributions
roller-0.167
 salon-0.157
ara-0.150
atters-0.143
post-0.141
Tokenie
Token ID494
Feature activation-6.35e+00

Pos logit contributions
ling+0.463
 shining+0.451
 around+0.414
 up+0.402
 him+0.355
Neg logit contributions
roller-0.705
 salon-0.696
ara-0.681
atters-0.635
post-0.634
Token learned
Token ID4499
Feature activation+3.30e+00

Pos logit contributions
ling+0.233
 around+0.203
 shining+0.190
led+0.183
ie+0.174
Neg logit contributions
roller-0.403
 salon-0.380
ara-0.363
 heaven-0.357
post-0.357
Token that
Token ID326
Feature activation-1.11e+00

Pos logit contributions
roller+0.208
 salon+0.198
ara+0.190
atters+0.185
 pitcher+0.185
Neg logit contributions
ling-0.122
 around-0.104
 shining-0.101
led-0.094
 up-0.092
Token working
Token ID1762
Feature activation-5.11e+00

Pos logit contributions
ling+0.048
led+0.042
 around+0.037
ie+0.036
 shining+0.036
Neg logit contributions
 salon-0.059
roller-0.059
 heaven-0.056
 vet-0.054
ara-0.053
Token together
Token ID1978
Feature activation+2.19e+00

Pos logit contributions
ling+0.209
 shining+0.174
 around+0.167
led+0.164
ie+0.162
Neg logit contributions
roller-0.286
 salon-0.269
ara-0.259
post-0.257
atters-0.256
Token<|endoftext|>
Token ID50256
Feature activation-2.93e-01

Pos logit contributions
ling+0.021
 around+0.020
 him+0.019
 Joey+0.018
 up+0.018
Neg logit contributions
 salon-0.030
roller-0.029
 pitcher-0.026
ara-0.026
 heaven-0.026
TokenSam
Token ID16305
Feature activation-9.37e-01

Pos logit contributions
ling+0.016
 around+0.014
 shining+0.014
led+0.013
ie+0.013
Neg logit contributions
roller-0.014
 salon-0.012
ara-0.012
post-0.011
 heaven-0.011
Token was
Token ID373
Feature activation+1.92e-03

Pos logit contributions
ling+0.033
 around+0.027
 shining+0.027
led+0.025
 up+0.023
Neg logit contributions
roller-0.060
 salon-0.057
ara-0.056
 heaven-0.054
atters-0.053
Token picking
Token ID10868
Feature activation-2.89e+00

Pos logit contributions
roller+0.000
 salon+0.000
 heaven+0.000
 clinic+0.000
 pitcher+0.000
Neg logit contributions
ling-0.000
led-0.000
ie-0.000
 shining-0.000
 around-0.000
Token wild
Token ID4295
Feature activation-1.01e+00

Pos logit contributions
ling+0.104
led+0.085
itive+0.083
bee+0.080
 couldn+0.077
Neg logit contributions
roller-0.164
ara-0.158
 pitcher-0.156
 salon-0.151
S-0.150
Tokenfl
Token ID2704
Feature activation-1.27e+01

Pos logit contributions
ling+0.044
led+0.034
 shining+0.034
ging+0.033
bled+0.033
Neg logit contributions
roller-0.050
 salon-0.048
ara-0.048
 heaven-0.046
 jail-0.045
Tokenowers
Token ID3618
Feature activation-6.23e+00

Pos logit contributions
ling+0.503
 around+0.431
 shining+0.418
led+0.395
 up+0.379
Neg logit contributions
roller-0.762
 salon-0.726
ara-0.690
 pitcher-0.672
atters-0.670
Token in
Token ID287
Feature activation-2.09e+00

Pos logit contributions
ling+0.245
led+0.223
bee+0.204
itive+0.202
 shining+0.189
Neg logit contributions
roller-0.312
 pitcher-0.296
 salon-0.294
 heaven-0.279
 clinic-0.278
Token the
Token ID262
Feature activation-6.24e-01

Pos logit contributions
ling+0.059
 around+0.047
 shining+0.045
led+0.042
 up+0.038
Neg logit contributions
roller-0.150
 salon-0.144
ara-0.138
 heaven-0.136
 pitcher-0.135
Token me
Token ID502
Feature activation+9.13e-01

Pos logit contributions
ling+0.035
 shining+0.033
 around+0.033
 up+0.031
 in+0.029
Neg logit contributions
roller-0.027
 salon-0.025
atters-0.024
ara-0.023
post-0.023
Tokenadow
Token ID4584
Feature activation-2.31e+00

Pos logit contributions
roller+0.042
 salon+0.039
ara+0.037
 heaven+0.035
atters+0.035
Neg logit contributions
ling-0.049
 around-0.044
 shining-0.043
led-0.042
 up-0.040
Token a
Token ID257
Feature activation-2.75e-01

Pos logit contributions
ling+0.004
led+0.004
 around+0.004
uses+0.003
ie+0.003
Neg logit contributions
roller-0.005
 heaven-0.005
ara-0.005
 salon-0.005
 pitcher-0.005
Token gifted
Token ID22527
Feature activation-2.77e-01

Pos logit contributions
ling+0.010
 around+0.009
led+0.009
 shining+0.008
 up+0.008
Neg logit contributions
roller-0.017
 salon-0.016
ara-0.016
 pitcher-0.015
 clinic-0.015
Token girl
Token ID2576
Feature activation+2.13e+00

Pos logit contributions
ling+0.012
 around+0.010
 shining+0.010
led+0.010
 up+0.009
Neg logit contributions
roller-0.015
 salon-0.015
ara-0.014
 heaven-0.014
 pitcher-0.013
Token called
Token ID1444
Feature activation+1.27e+00

Pos logit contributions
roller+0.111
 salon+0.103
lor+0.094
atters+0.094
 pitcher+0.094
Neg logit contributions
ling-0.105
 around-0.101
 in-0.094
 shining-0.093
ie-0.091
Token El
Token ID2574
Feature activation-3.49e+00

Pos logit contributions
roller+0.066
 salon+0.062
atters+0.058
 clinic+0.058
 pitcher+0.057
Neg logit contributions
 around-0.054
ling-0.052
 in-0.050
 shining-0.050
 him-0.050
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.167
 shining+0.166
 around+0.153
 him+0.147
 spear+0.143
Neg logit contributions
ara-0.164
 salon-0.157
roller-0.150
 heaven-0.142
 clinic-0.138
Tokenie
Token ID494
Feature activation-6.27e+00

Pos logit contributions
 shining+0.379
 Barry+0.326
ging+0.323
 Joey+0.320
ling+0.304
Neg logit contributions
ara-0.804
 salon-0.772
S-0.736
roller-0.705
ress-0.704
Token.
Token ID13
Feature activation-5.25e-01

Pos logit contributions
ling+0.282
 shining+0.242
led+0.239
 around+0.238
 up+0.216
Neg logit contributions
roller-0.335
 salon-0.318
ara-0.302
 heaven-0.301
 clinic-0.296
Token El
Token ID2574
Feature activation-3.52e+00

Pos logit contributions
ling+0.034
 around+0.031
 shining+0.030
led+0.029
 up+0.029
Neg logit contributions
roller-0.019
 salon-0.017
ara-0.016
 heaven-0.015
atters-0.015
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.177
 shining+0.162
 around+0.157
 him+0.148
 up+0.145
Neg logit contributions
roller-0.160
 salon-0.159
ara-0.157
atters-0.142
 pitcher-0.142
Tokenie
Token ID494
Feature activation-6.29e+00

Pos logit contributions
 shining+0.506
ging+0.460
ling+0.437
 Barry+0.422
nam+0.421
Neg logit contributions
ara-0.695
 salon-0.614
S-0.592
ase-0.590
Anna-0.586
Token
Token ID198
Feature activation-1.57e-01

Pos logit contributions
ling+0.027
 around+0.024
 shining+0.024
led+0.023
 up+0.022
Neg logit contributions
roller-0.023
 salon-0.021
ara-0.020
atters-0.019
 clinic-0.019
Token
Token ID198
Feature activation-1.55e-01

Pos logit contributions
ling+0.010
led+0.009
 shining+0.009
 around+0.008
itive+0.008
Neg logit contributions
roller-0.006
ara-0.005
 salon-0.005
post-0.005
atters-0.005
TokenBut
Token ID1537
Feature activation+3.36e+00

Pos logit contributions
ling+0.009
led+0.008
 around+0.008
 shining+0.008
 up+0.008
Neg logit contributions
roller-0.006
 salon-0.006
ara-0.005
post-0.005
 clinic-0.005
Token what
Token ID644
Feature activation-7.35e+00

Pos logit contributions
 salon+0.186
roller+0.170
ara+0.168
 heaven+0.161
 clinic+0.160
Neg logit contributions
ling-0.152
led-0.124
 around-0.116
 spear-0.115
 shining-0.115
Token El
Token ID2574
Feature activation-3.55e+00

Pos logit contributions
ling+0.332
ging+0.283
 shining+0.272
led+0.268
uses+0.255
Neg logit contributions
 salon-0.384
ara-0.372
roller-0.368
ress-0.343
 heaven-0.343
Tokenod
Token ID375
Feature activation-1.27e+01

Pos logit contributions
ling+0.175
 shining+0.161
 around+0.157
 him+0.150
 up+0.144
Neg logit contributions
 salon-0.163
roller-0.161
ara-0.161
ress-0.147
 pitcher-0.146
Tokenie
Token ID494
Feature activation-6.46e+00

Pos logit contributions
 shining+0.368
nam+0.313
ging+0.288
ling+0.288
 Barry+0.278
Neg logit contributions
ara-0.824
 salon-0.768
ase-0.709
Anna-0.705
S-0.701
Token liked
Token ID8288
Feature activation-2.58e+00

Pos logit contributions
ling+0.249
 shining+0.213
 around+0.211
led+0.197
 up+0.187
Neg logit contributions
roller-0.382
 salon-0.372
ara-0.357
 heaven-0.346
 clinic-0.343
Token to
Token ID284
Feature activation+1.40e-01

Pos logit contributions
ling+0.112
led+0.095
 around+0.086
ging+0.083
 shining+0.083
Neg logit contributions
 salon-0.146
roller-0.140
 heaven-0.134
ara-0.129
 clinic-0.129
Token think
Token ID892
Feature activation-3.66e+00

Pos logit contributions
roller+0.007
 salon+0.007
ara+0.006
 heaven+0.006
 clinic+0.006
Neg logit contributions
ling-0.007
 around-0.006
 shining-0.006
led-0.006
 up-0.005
Token about
Token ID546
Feature activation-2.31e+00

Pos logit contributions
ling+0.180
led+0.154
 shining+0.147
ging+0.144
 around+0.141
Neg logit contributions
 salon-0.174
roller-0.170
post-0.156
 heaven-0.156
 clinic-0.153
Token!"
Token ID2474
Feature activation+3.24e+00

Pos logit contributions
ling+0.254
 shining+0.223
 around+0.223
led+0.215
 up+0.202
Neg logit contributions
roller-0.243
 salon-0.233
ara-0.220
 heaven-0.215
 clinic-0.211
Token And
Token ID843
Feature activation+1.88e+00

Pos logit contributions
roller+0.137
 salon+0.123
ara+0.116
 heaven+0.116
atters+0.115
Neg logit contributions
ling-0.186
led-0.160
 shining-0.160
 around-0.159
ie-0.149
Token they
Token ID484
Feature activation+5.70e-01

Pos logit contributions
roller+0.108
 salon+0.103
ress+0.097
 heaven+0.097
ara+0.095
Neg logit contributions
ling-0.077
 shining-0.062
led-0.061
 around-0.061
ie-0.055
Token hugged
Token ID41130
Feature activation+9.27e-01

Pos logit contributions
roller+0.038
 salon+0.037
ara+0.036
 heaven+0.035
atters+0.034
Neg logit contributions
ling-0.018
 shining-0.014
 around-0.014
led-0.013
 up-0.012
Token and
Token ID290
Feature activation-2.47e-01

Pos logit contributions
roller+0.049
 heaven+0.049
ara+0.046
post+0.046
 jail+0.046
Neg logit contributions
ling-0.035
 shining-0.030
led-0.029
 atop-0.025
uses-0.025
Token filled
Token ID5901
Feature activation+4.52e+00

Pos logit contributions
ling+0.010
 around+0.009
 shining+0.009
led+0.008
 up+0.008
Neg logit contributions
roller-0.014
 salon-0.013
ara-0.013
 heaven-0.012
atters-0.012
Token each
Token ID1123
Feature activation-5.70e+00

Pos logit contributions
 heaven+0.252
roller+0.251
 jail+0.238
ress+0.234
ara+0.234
Neg logit contributions
ling-0.156
itive-0.134
bee-0.133
 shining-0.131
led-0.120
Token other
Token ID584
Feature activation+1.26e+00

Pos logit contributions
ling+0.200
 around+0.169
 shining+0.169
led+0.157
 him+0.147
Neg logit contributions
roller-0.356
 salon-0.341
ara-0.331
 heaven-0.323
 pitcher-0.322
Token with
Token ID351
Feature activation+1.35e+00

Pos logit contributions
roller+0.073
 salon+0.068
ara+0.065
 heaven+0.064
atters+0.063
Neg logit contributions
ling-0.051
 shining-0.044
 around-0.044
led-0.042
 him-0.040
Token love
Token ID1842
Feature activation-1.12e+00

Pos logit contributions
roller+0.062
 heaven+0.059
 salon+0.059
ara+0.055
 pitcher+0.055
Neg logit contributions
ling-0.069
 around-0.058
led-0.057
 shining-0.057
ie-0.055
Token.
Token ID13
Feature activation-1.80e-01

Pos logit contributions
ling+0.046
 around+0.038
 shining+0.038
led+0.037
 up+0.035
Neg logit contributions
roller-0.063
 salon-0.061
 heaven-0.058
ara-0.058
post-0.057
Token a
Token ID257
Feature activation-3.94e-02

Pos logit contributions
roller+0.044
 salon+0.042
 heaven+0.040
post+0.039
ara+0.039
Neg logit contributions
ling-0.033
led-0.027
 around-0.027
 shining-0.026
ie-0.026
Token great
Token ID1049
Feature activation-4.12e-01

Pos logit contributions
ling+0.002
 around+0.001
 shining+0.001
led+0.001
 up+0.001
Neg logit contributions
roller-0.002
 salon-0.002
ara-0.002
 heaven-0.002
 pitcher-0.002
Token job
Token ID1693
Feature activation+2.98e+00

Pos logit contributions
ling+0.022
 around+0.020
 shining+0.020
led+0.019
 up+0.018
Neg logit contributions
roller-0.019
 salon-0.018
ara-0.017
atters-0.016
 heaven-0.016
Token cleaning
Token ID12724
Feature activation+7.01e-01

Pos logit contributions
roller+0.186
 salon+0.176
ara+0.171
atters+0.166
 pitcher+0.164
Neg logit contributions
ling-0.111
 around-0.097
 shining-0.093
led-0.085
 up-0.084
Token the
Token ID262
Feature activation-2.92e-01

Pos logit contributions
roller+0.046
 salon+0.044
ara+0.042
 heaven+0.041
post+0.041
Neg logit contributions
ling-0.023
 shining-0.018
led-0.017
 around-0.017
bee-0.016
Token garage
Token ID15591
Feature activation+4.38e+00

Pos logit contributions
ling+0.017
 around+0.016
 shining+0.015
 up+0.015
led+0.014
Neg logit contributions
roller-0.012
 salon-0.011
ara-0.011
atters-0.010
 heaven-0.010
Token.
Token ID13
Feature activation-2.55e-01

Pos logit contributions
roller+0.282
ara+0.271
 salon+0.265
atters+0.257
 heaven+0.255
Neg logit contributions
ling-0.151
 around-0.141
 shining-0.126
 up-0.123
 in-0.118
Token Tim
Token ID5045
Feature activation-9.83e-01

Pos logit contributions
 around+0.010
ling+0.009
 him+0.009
 Joey+0.009
 shining+0.009
Neg logit contributions
roller-0.014
 salon-0.014
lor-0.013
 clinic-0.013
 heaven-0.013
Tokenmy
Token ID1820
Feature activation-4.30e+00

Pos logit contributions
ling+0.041
 shining+0.038
 around+0.037
led+0.036
 atop+0.035
Neg logit contributions
roller-0.052
ara-0.051
 salon-0.050
ress-0.047
atters-0.045
Token felt
Token ID2936
Feature activation-2.45e+00

Pos logit contributions
ling+0.155
 around+0.132
 shining+0.130
led+0.120
 up+0.114
Neg logit contributions
roller-0.271
 salon-0.259
ara-0.248
 pitcher-0.240
atters-0.238
Token proud
Token ID6613
Feature activation+7.22e-02

Pos logit contributions
ling+0.099
 shining+0.082
 around+0.081
led+0.079
ie+0.077
Neg logit contributions
roller-0.141
 salon-0.134
 heaven-0.129
 pitcher-0.127
ara-0.127
Token you
Token ID345
Feature activation-2.78e-01

Pos logit contributions
ling+0.031
 around+0.027
 shining+0.026
ie+0.024
 him+0.024
Neg logit contributions
roller-0.053
 salon-0.051
 clinic-0.047
 pitcher-0.047
atters-0.047
Token can
Token ID460
Feature activation+6.27e-01

Pos logit contributions
ling+0.014
 shining+0.012
 around+0.012
led+0.011
 up+0.011
Neg logit contributions
roller-0.014
 salon-0.013
ara-0.013
atters-0.012
 pitcher-0.012
Token!
Token ID0
Feature activation-1.32e-01

Pos logit contributions
roller+0.037
 salon+0.035
ara+0.033
 heaven+0.032
 clinic+0.032
Neg logit contributions
ling-0.026
 around-0.023
 shining-0.022
led-0.021
 up-0.020
Token We
Token ID775
Feature activation+1.40e+00

Pos logit contributions
 around+0.006
 him+0.006
ling+0.006
 up+0.005
 Barry+0.005
Neg logit contributions
roller-0.006
 salon-0.006
ara-0.005
 pitcher-0.005
 heaven-0.005
Token can
Token ID460
Feature activation+6.30e-01

Pos logit contributions
roller+0.072
 salon+0.068
ara+0.064
atters+0.062
 heaven+0.062
Neg logit contributions
ling-0.067
 around-0.058
 shining-0.058
led-0.056
 up-0.053
Token start
Token ID923
Feature activation+5.84e+00

Pos logit contributions
roller+0.034
 salon+0.033
ara+0.031
 heaven+0.030
atters+0.030
Neg logit contributions
ling-0.028
 around-0.025
 shining-0.024
led-0.023
 up-0.022
Token by
Token ID416
Feature activation-7.93e-01

Pos logit contributions
roller+0.327
 salon+0.316
 heaven+0.302
ara+0.294
atters+0.289
Neg logit contributions
ling-0.251
led-0.209
 around-0.197
 shining-0.189
ie-0.187
Token getting
Token ID1972
Feature activation-2.01e+00

Pos logit contributions
ling+0.031
 around+0.026
led+0.025
 shining+0.025
ie+0.023
Neg logit contributions
roller-0.048
 salon-0.046
ara-0.043
 heaven-0.043
atters-0.042
Token a
Token ID257
Feature activation-4.93e-02

Pos logit contributions
ling+0.083
bee+0.070
led+0.067
itive+0.065
ie+0.064
Neg logit contributions
 salon-0.110
roller-0.109
 heaven-0.105
 pitcher-0.101
ara-0.099
Token more
Token ID517
Feature activation+4.71e+00

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
 up+0.002
led+0.002
Neg logit contributions
roller-0.003
 salon-0.002
ara-0.002
atters-0.002
post-0.002
Token comfortable
Token ID6792
Feature activation+1.65e+00

Pos logit contributions
roller+0.231
 salon+0.218
ara+0.204
 heaven+0.200
 clinic+0.198
Neg logit contributions
ling-0.235
 around-0.206
 shining-0.205
led-0.198
 him-0.188
Token do
Token ID466
Feature activation-1.22e+00

Pos logit contributions
roller+0.003
 salon+0.003
post+0.003
atters+0.003
ara+0.003
Neg logit contributions
ling-0.004
 around-0.003
 shining-0.003
 him-0.003
 up-0.003
Token you
Token ID345
Feature activation-2.00e-01

Pos logit contributions
ling+0.066
 around+0.060
 shining+0.058
led+0.056
 up+0.055
Neg logit contributions
roller-0.055
 salon-0.051
ara-0.048
 heaven-0.046
atters-0.046
Token want
Token ID765
Feature activation-3.03e+00

Pos logit contributions
ling+0.010
 around+0.008
 shining+0.008
led+0.008
 up+0.008
Neg logit contributions
roller-0.011
 salon-0.010
ara-0.009
 heaven-0.009
 clinic-0.009
Token some
Token ID617
Feature activation+9.77e-01

Pos logit contributions
ling+0.129
 around+0.110
 shining+0.107
led+0.103
 up+0.097
Neg logit contributions
roller-0.175
 salon-0.167
ara-0.158
 heaven-0.155
 pitcher-0.153
Token of
Token ID286
Feature activation+1.69e+00

Pos logit contributions
roller+0.045
 salon+0.043
ara+0.040
 heaven+0.039
atters+0.038
Neg logit contributions
ling-0.052
 around-0.046
 shining-0.046
led-0.044
 up-0.043
Token my
Token ID616
Feature activation+4.34e+00

Pos logit contributions
roller+0.081
 salon+0.074
ara+0.070
atters+0.068
post+0.067
Neg logit contributions
ling-0.087
 around-0.080
 shining-0.079
 him-0.076
 up-0.075
Token snacks
Token ID28487
Feature activation+3.45e-01

Pos logit contributions
roller+0.169
 salon+0.164
 pitcher+0.159
 clinic+0.158
ara+0.157
Neg logit contributions
ling-0.254
led-0.221
 around-0.217
 shining-0.210
ie-0.207
Token?
Token ID30
Feature activation+7.32e-01

Pos logit contributions
roller+0.021
 salon+0.020
atters+0.019
 clinic+0.019
ara+0.019
Neg logit contributions
ling-0.013
 around-0.012
 in-0.011
 shining-0.011
 up-0.011
Token I
Token ID314
Feature activation+1.78e+00

Pos logit contributions
roller+0.036
 salon+0.034
 pitcher+0.030
post+0.029
ara+0.029
Neg logit contributions
 around-0.034
 him-0.033
ling-0.033
 in-0.032
 shining-0.030
Token have
Token ID423
Feature activation+1.54e+00

Pos logit contributions
roller+0.104
 salon+0.096
ara+0.089
post+0.089
ress+0.088
Neg logit contributions
ling-0.076
 around-0.070
 shining-0.063
 him-0.062
 in-0.061
Token a
Token ID257
Feature activation+9.56e-02

Pos logit contributions
roller+0.103
 salon+0.096
atters+0.092
ara+0.092
ress+0.091
Neg logit contributions
ling-0.049
 around-0.044
 shining-0.042
 him-0.040
 up-0.038
Token,
Token ID11
Feature activation-1.68e-01

Pos logit contributions
roller+0.357
 salon+0.342
ara+0.320
post+0.316
 pitcher+0.312
Neg logit contributions
ling-0.253
 around-0.213
 shining-0.208
led-0.200
 up-0.191
Token he
Token ID339
Feature activation-3.43e+00

Pos logit contributions
ling+0.008
led+0.007
ie+0.007
bee+0.007
mo+0.006
Neg logit contributions
 jail-0.008
roller-0.008
 vet-0.008
 salon-0.007
 heaven-0.007
Token saw
Token ID2497
Feature activation-5.15e-01

Pos logit contributions
ling+0.145
 shining+0.128
 around+0.119
led+0.116
ging+0.114
Neg logit contributions
roller-0.185
 salon-0.183
ara-0.173
 clinic-0.163
 pitcher-0.161
Token the
Token ID262
Feature activation-1.58e-01

Pos logit contributions
ling+0.025
led+0.020
ie+0.019
 shining+0.019
ging+0.019
Neg logit contributions
roller-0.025
 salon-0.025
 vet-0.024
 jail-0.023
 heaven-0.023
Token terrible
Token ID7818
Feature activation-5.14e-01

Pos logit contributions
ling+0.007
led+0.006
 around+0.006
ie+0.006
 shining+0.006
Neg logit contributions
roller-0.008
 salon-0.008
 pitcher-0.008
ara-0.008
post-0.007
Token mess
Token ID2085
Feature activation+4.37e+00

Pos logit contributions
ling+0.024
 around+0.021
led+0.020
 shining+0.020
 atop+0.019
Neg logit contributions
roller-0.026
 salon-0.025
ara-0.024
 pitcher-0.023
post-0.023
Token.
Token ID13
Feature activation-3.21e-01

Pos logit contributions
roller+0.250
 salon+0.242
post+0.229
ara+0.229
 heaven+0.224
Neg logit contributions
ling-0.167
 shining-0.138
 around-0.137
led-0.129
ie-0.123
Token His
Token ID2399
Feature activation-1.58e+00

Pos logit contributions
ling+0.017
led+0.015
 around+0.015
 shining+0.014
ie+0.014
Neg logit contributions
roller-0.016
 salon-0.014
ara-0.013
post-0.013
 pitcher-0.013
Token mar
Token ID1667
Feature activation-3.76e+00

Pos logit contributions
ling+0.076
 around+0.065
led+0.065
 shining+0.061
ie+0.060
Neg logit contributions
roller-0.085
 salon-0.077
 pitcher-0.074
ara-0.074
 jail-0.070
Tokenbles
Token ID7689
Feature activation-8.30e+00

Pos logit contributions
ling+0.108
 shining+0.097
 around+0.097
 up+0.086
 him+0.084
Neg logit contributions
roller-0.253
 salon-0.238
 pitcher-0.231
ara-0.227
atters-0.226
Token were
Token ID547
Feature activation-9.84e-01

Pos logit contributions
ling+0.333
 shining+0.296
 around+0.292
 up+0.275
 Barry+0.269
Neg logit contributions
roller-0.427
 salon-0.419
 pitcher-0.388
ara-0.375
atters-0.371
Token She
Token ID1375
Feature activation+2.58e+00

Pos logit contributions
roller+0.004
 salon+0.004
 pitcher+0.004
ress+0.004
 clinic+0.004
Neg logit contributions
 around-0.004
ling-0.004
 him-0.004
 shining-0.004
 up-0.004
Token knows
Token ID4206
Feature activation-1.96e+00

Pos logit contributions
roller+0.126
 salon+0.117
ara+0.109
 pitcher+0.106
post+0.106
Neg logit contributions
ling-0.133
 around-0.118
 shining-0.115
led-0.112
 up-0.109
Token she
Token ID673
Feature activation+3.37e-01

Pos logit contributions
ling+0.100
led+0.083
 around+0.077
 shining+0.075
 atop+0.072
Neg logit contributions
 salon-0.099
 heaven-0.091
roller-0.090
ara-0.089
 vet-0.087
Token is
Token ID318
Feature activation+3.21e+00

Pos logit contributions
roller+0.012
 salon+0.012
ara+0.012
 heaven+0.011
atters+0.011
Neg logit contributions
ling-0.020
 shining-0.018
 around-0.018
led-0.017
ie-0.016
Token strong
Token ID1913
Feature activation-1.12e+00

Pos logit contributions
 heaven+0.158
 salon+0.152
 clinic+0.144
roller+0.143
 vet+0.142
Neg logit contributions
ling-0.148
led-0.127
ie-0.124
 shining-0.118
 around-0.117
Token and
Token ID290
Feature activation-6.64e-02

Pos logit contributions
ling+0.044
 around+0.039
 shining+0.036
led+0.034
 up+0.034
Neg logit contributions
roller-0.069
 salon-0.065
ara-0.061
 pitcher-0.060
atters-0.060
Token happy
Token ID3772
Feature activation-6.48e-01

Pos logit contributions
ling+0.003
ie+0.002
 shining+0.002
 around+0.002
led+0.002
Neg logit contributions
 salon-0.003
 heaven-0.003
roller-0.003
post-0.003
atters-0.003
Token.
Token ID13
Feature activation+9.65e-02

Pos logit contributions
ling+0.028
 shining+0.023
 around+0.023
led+0.022
 up+0.021
Neg logit contributions
roller-0.037
 salon-0.035
ara-0.033
 heaven-0.033
 clinic-0.032
Token
Token ID198
Feature activation+4.07e-01

Pos logit contributions
roller+0.005
 salon+0.005
 pitcher+0.005
ction+0.005
atters+0.004
Neg logit contributions
 around-0.004
ling-0.004
 him-0.004
 shining-0.004
 up-0.004
Token
Token ID198
Feature activation+4.22e-01

Pos logit contributions
roller+0.021
 salon+0.019
ara+0.019
 heaven+0.018
atters+0.018
Neg logit contributions
ling-0.020
 around-0.017
 shining-0.017
led-0.017
 up-0.016
TokenShe
Token ID3347
Feature activation+2.92e+00

Pos logit contributions
roller+0.023
 salon+0.022
 bidder+0.020
 clinic+0.020
 jail+0.019
Neg logit contributions
ling-0.020
 around-0.017
 in-0.015
 him-0.015
 up-0.015
Token cr
Token ID1067
Feature activation-1.42e+00

Pos logit contributions
roller+0.085
 salon+0.078
ara+0.075
atters+0.074
lor+0.072
Neg logit contributions
ling-0.079
 around-0.073
 shining-0.073
 him-0.067
 up-0.067
Tokenay
Token ID323
Feature activation-1.63e+00

Pos logit contributions
ling+0.065
 around+0.061
ie+0.059
 shining+0.057
led+0.055
Neg logit contributions
 salon-0.068
 clinic-0.068
 heaven-0.066
 bidder-0.065
 vet-0.064
Tokenons
Token ID684
Feature activation-3.97e+00

Pos logit contributions
ling+0.064
 around+0.055
 shining+0.054
led+0.051
 up+0.049
Neg logit contributions
roller-0.098
 salon-0.094
ara-0.089
 heaven-0.087
atters-0.087
Token she
Token ID673
Feature activation-2.40e-02

Pos logit contributions
ling+0.135
 around+0.124
 shining+0.111
 in+0.104
 up+0.103
Neg logit contributions
roller-0.264
 salon-0.247
ara-0.244
atters-0.239
 clinic-0.235
Token had
Token ID550
Feature activation-9.54e-01

Pos logit contributions
ling+0.001
 shining+0.001
 around+0.001
 up+0.001
led+0.001
Neg logit contributions
roller-0.001
 salon-0.001
ara-0.001
post-0.001
 heaven-0.001
Token forgotten
Token ID11564
Feature activation+3.19e+00

Pos logit contributions
ling+0.038
 around+0.035
 shining+0.033
led+0.031
 in+0.031
Neg logit contributions
roller-0.057
 salon-0.053
ara-0.051
 pitcher-0.051
 clinic-0.050
Token about
Token ID546
Feature activation-1.93e+00

Pos logit contributions
roller+0.189
 salon+0.180
ara+0.171
 heaven+0.167
atters+0.166
Neg logit contributions
ling-0.128
 around-0.109
 shining-0.107
led-0.102
 up-0.097
Token.
Token ID13
Feature activation-1.60e-01

Pos logit contributions
ling+0.086
 around+0.070
led+0.069
 shining+0.068
 up+0.062
Neg logit contributions
roller-0.107
 salon-0.104
ara-0.097
 heaven-0.097
post-0.096
Token She
Token ID1375
Feature activation+2.23e+00

Pos logit contributions
 around+0.008
 up+0.008
ling+0.008
 shining+0.007
 in+0.007
Neg logit contributions
roller-0.007
lor-0.007
 salon-0.007
 heaven-0.006
 clinic-0.006
Token wanted
Token ID2227
Feature activation-2.64e+00

Pos logit contributions
roller+0.137
 salon+0.130
ara+0.124
 heaven+0.121
atters+0.121
Neg logit contributions
ling-0.086
 around-0.073
 shining-0.071
led-0.067
 up-0.064
Token to
Token ID284
Feature activation+5.22e-01

Pos logit contributions
ling+0.118
led+0.096
 around+0.092
 shining+0.091
ie+0.084
Neg logit contributions
roller-0.148
 salon-0.146
ara-0.134
 heaven-0.134
post-0.131
Token but
Token ID475
Feature activation-2.57e+00

Pos logit contributions
ling+0.007
 around+0.006
 shining+0.006
led+0.005
ie+0.005
Neg logit contributions
roller-0.010
 salon-0.010
ara-0.009
 pitcher-0.009
 heaven-0.009
Token the
Token ID262
Feature activation-1.62e-01

Pos logit contributions
ling+0.106
led+0.083
 shining+0.081
 around+0.081
ie+0.076
Neg logit contributions
roller-0.147
 salon-0.147
ara-0.138
 heaven-0.135
 pitcher-0.135
Token small
Token ID1402
Feature activation+2.72e-01

Pos logit contributions
ling+0.007
 around+0.005
led+0.005
 shining+0.005
ie+0.005
Neg logit contributions
roller-0.010
 salon-0.009
 pitcher-0.009
ara-0.009
 heaven-0.009
Token square
Token ID6616
Feature activation+1.53e+00

Pos logit contributions
roller+0.014
 salon+0.013
ara+0.012
lor+0.012
atters+0.012
Neg logit contributions
ling-0.013
 around-0.012
 shining-0.012
 up-0.011
led-0.011
Token box
Token ID3091
Feature activation+3.47e+00

Pos logit contributions
roller+0.096
 salon+0.092
ara+0.088
 heaven+0.086
atters+0.086
Neg logit contributions
ling-0.056
 around-0.047
 shining-0.046
led-0.043
 up-0.041
Token had
Token ID550
Feature activation-8.38e-01

Pos logit contributions
roller+0.220
 salon+0.210
ara+0.200
 heaven+0.195
atters+0.195
Neg logit contributions
ling-0.126
 around-0.106
 shining-0.103
led-0.097
 up-0.092
Token a
Token ID257
Feature activation-2.87e-02

Pos logit contributions
 around+0.020
 shining+0.020
 all+0.019
ling+0.019
 in+0.017
Neg logit contributions
roller-0.059
 salon-0.059
ara-0.057
 clinic-0.056
ress-0.055
Token hard
Token ID1327
Feature activation-1.30e+00

Pos logit contributions
 shining+0.001
ling+0.001
 around+0.001
 spear+0.001
 all+0.001
Neg logit contributions
 salon-0.002
 heaven-0.002
ara-0.002
ress-0.002
lor-0.002
Token time
Token ID640
Feature activation+2.12e+00

Pos logit contributions
ling+0.055
 around+0.049
 shining+0.048
 up+0.044
led+0.043
Neg logit contributions
roller-0.073
 salon-0.072
ara-0.068
 heaven-0.067
 clinic-0.066
Token rolling
Token ID10708
Feature activation-2.52e+00

Pos logit contributions
roller+0.130
 salon+0.124
ara+0.119
 heaven+0.115
 pitcher+0.115
Neg logit contributions
ling-0.080
 around-0.071
 shining-0.067
led-0.063
 up-0.061
Token.
Token ID13
Feature activation-5.07e-01

Pos logit contributions
ling+0.071
 shining+0.054
 around+0.051
led+0.050
ie+0.045
Neg logit contributions
roller-0.180
 salon-0.170
ara-0.164
atters-0.163
post-0.163
Token eating
Token ID6600
Feature activation-3.57e+00

Pos logit contributions
roller+0.260
 salon+0.251
 heaven+0.235
 pitcher+0.234
ara+0.233
Neg logit contributions
ling-0.206
 around-0.171
led-0.169
 shining-0.159
ie-0.155
Token,
Token ID11
Feature activation-9.35e-02

Pos logit contributions
ling+0.138
 around+0.113
 shining+0.111
led+0.108
ie+0.100
Neg logit contributions
roller-0.215
 salon-0.207
ara-0.198
 heaven-0.194
 pitcher-0.194
Token Lily
Token ID20037
Feature activation+4.12e+00

Pos logit contributions
ling+0.006
 around+0.005
 shining+0.005
led+0.005
 up+0.005
Neg logit contributions
roller-0.004
 salon-0.003
ara-0.003
 heaven-0.003
 pitcher-0.003
Token's
Token ID338
Feature activation+1.25e+00

Pos logit contributions
roller+0.250
 salon+0.237
atters+0.226
post+0.224
ara+0.223
Neg logit contributions
ling-0.163
 around-0.139
 shining-0.132
led-0.128
 up-0.124
Token mom
Token ID1995
Feature activation+8.55e-01

Pos logit contributions
roller+0.066
 salon+0.063
ara+0.058
 heaven+0.058
 pitcher+0.058
Neg logit contributions
ling-0.060
 around-0.051
led-0.051
 shining-0.049
 atop-0.046
Token suggested
Token ID5220
Feature activation-9.26e-01

Pos logit contributions
roller+0.052
 salon+0.049
ara+0.046
atters+0.046
 heaven+0.045
Neg logit contributions
ling-0.034
 around-0.029
 shining-0.029
led-0.027
 up-0.026
Token they
Token ID484
Feature activation+5.08e-01

Pos logit contributions
ling+0.043
 around+0.037
 shining+0.037
led+0.036
ie+0.034
Neg logit contributions
roller-0.046
 salon-0.046
ara-0.043
 heaven-0.042
 pitcher-0.042
Token go
Token ID467
Feature activation+2.15e+00

Pos logit contributions
roller+0.023
 salon+0.022
 heaven+0.020
 pitcher+0.020
ara+0.020
Neg logit contributions
ling-0.027
led-0.023
 shining-0.023
 around-0.023
ie-0.021
Token for
Token ID329
Feature activation-1.84e+00

Pos logit contributions
roller+0.147
 salon+0.143
post+0.137
ara+0.136
lor+0.133
Neg logit contributions
ling-0.067
 around-0.065
 up-0.059
 shining-0.057
 in-0.056
Token a
Token ID257
Feature activation-4.19e-02

Pos logit contributions
ling+0.048
 shining+0.048
 him+0.048
 around+0.045
 up+0.042
Neg logit contributions
roller-0.131
 salon-0.125
ress-0.118
ara-0.118
atters-0.116
Token walk
Token ID2513
Feature activation-1.30e+00

Pos logit contributions
ling+0.002
 around+0.002
 shining+0.002
led+0.002
 up+0.001
Neg logit contributions
roller-0.002
 salon-0.002
ara-0.002
 heaven-0.002
atters-0.002
Token and
Token ID290
Feature activation-5.99e-01

Pos logit contributions
ling+0.070
 around+0.059
 shining+0.058
led+0.056
 up+0.052
Neg logit contributions
roller-0.108
 salon-0.103
ara-0.097
 pitcher-0.095
 heaven-0.095
Token even
Token ID772
Feature activation-2.37e+00

Pos logit contributions
ling+0.025
 around+0.020
 shining+0.020
led+0.020
ie+0.018
Neg logit contributions
 salon-0.033
roller-0.033
 heaven-0.030
 pitcher-0.030
ara-0.030
Token ignored
Token ID9514
Feature activation+7.67e-01

Pos logit contributions
ling+0.108
 shining+0.094
itive+0.090
 around+0.089
bee+0.087
Neg logit contributions
 salon-0.113
 pitcher-0.106
 heaven-0.105
ara-0.103
roller-0.101
Token Tim
Token ID5045
Feature activation-1.00e+00

Pos logit contributions
roller+0.041
 heaven+0.041
 jail+0.041
 salon+0.038
post+0.038
Neg logit contributions
ling-0.030
led-0.025
 shining-0.024
 atop-0.023
itive-0.023
Tokenmy
Token ID1820
Feature activation-4.32e+00

Pos logit contributions
ling+0.041
 shining+0.040
led+0.040
 atop+0.037
 around+0.036
Neg logit contributions
 salon-0.051
roller-0.050
ress-0.048
ara-0.047
 heaven-0.046
Token when
Token ID618
Feature activation-1.30e-01

Pos logit contributions
ling+0.190
 shining+0.170
led+0.163
 around+0.160
ie+0.148
Neg logit contributions
roller-0.227
 salon-0.219
 heaven-0.205
 clinic-0.200
ara-0.199
Token he
Token ID339
Feature activation-3.44e+00

Pos logit contributions
ling+0.005
led+0.004
 around+0.004
ie+0.004
bee+0.003
Neg logit contributions
 salon-0.007
roller-0.007
 jail-0.007
 heaven-0.007
ara-0.007
Token said
Token ID531
Feature activation-1.47e+00

Pos logit contributions
ling+0.144
 around+0.124
 shining+0.123
led+0.113
 up+0.112
Neg logit contributions
roller-0.194
 salon-0.187
ara-0.181
 pitcher-0.173
 heaven-0.172
Token he
Token ID339
Feature activation-3.41e+00

Pos logit contributions
ling+0.065
 shining+0.054
 around+0.054
led+0.053
ie+0.051
Neg logit contributions
roller-0.079
 salon-0.077
ara-0.071
 heaven-0.071
 pitcher-0.070
Token was
Token ID373
Feature activation+2.56e-01

Pos logit contributions
ling+0.141
 shining+0.122
 around+0.119
led+0.107
 up+0.107
Neg logit contributions
roller-0.190
 salon-0.183
ara-0.182
atters-0.170
 clinic-0.169
Token still
Token ID991
Feature activation+2.82e+00

Pos logit contributions
 salon+0.013
roller+0.013
 heaven+0.012
 pitcher+0.012
 jail+0.012
Neg logit contributions
ling-0.011
led-0.010
ie-0.009
 around-0.009
 shining-0.009
Token grew
Token ID6348
Feature activation+1.56e+00

Pos logit contributions
roller+0.078
 salon+0.074
ara+0.070
 heaven+0.068
atters+0.068
Neg logit contributions
ling-0.056
 around-0.048
 shining-0.047
led-0.045
 up-0.043
Token bigger
Token ID5749
Feature activation+1.56e+00

Pos logit contributions
roller+0.081
post+0.074
 salon+0.074
 pitcher+0.073
atters+0.072
Neg logit contributions
ling-0.065
 around-0.054
led-0.053
bee-0.053
uses-0.051
Token.
Token ID13
Feature activation-1.60e-01

Pos logit contributions
roller+0.095
 salon+0.090
ara+0.085
atters+0.083
 heaven+0.083
Neg logit contributions
ling-0.062
 around-0.054
 shining-0.051
led-0.049
 up-0.048
Token Anna
Token ID11735
Feature activation+4.80e+00

Pos logit contributions
 around+0.007
ling+0.006
 up+0.006
 shining+0.006
 him+0.006
Neg logit contributions
roller-0.009
 salon-0.009
 pitcher-0.008
 clinic-0.008
 heaven-0.008
Token was
Token ID373
Feature activation+2.70e-01

Pos logit contributions
roller+0.296
 salon+0.283
ara+0.269
 heaven+0.262
atters+0.262
Neg logit contributions
ling-0.180
 around-0.152
 shining-0.149
led-0.140
 up-0.133
Token amazed
Token ID24562
Feature activation-3.46e+00

Pos logit contributions
roller+0.017
 salon+0.016
ara+0.015
 heaven+0.015
atters+0.015
Neg logit contributions
ling-0.010
 around-0.009
 shining-0.009
led-0.008
 up-0.008
Token and
Token ID290
Feature activation-2.85e-01

Pos logit contributions
ling+0.153
 shining+0.132
led+0.131
bee+0.126
 spear+0.123
Neg logit contributions
ara-0.164
roller-0.162
 salon-0.160
post-0.154
 heaven-0.151
Token full
Token ID1336
Feature activation+2.22e+00

Pos logit contributions
ling+0.012
 shining+0.009
 around+0.009
led+0.009
ie+0.009
Neg logit contributions
roller-0.016
 salon-0.016
ara-0.015
post-0.015
 pitcher-0.015
Token of
Token ID286
Feature activation+1.53e+00

Pos logit contributions
roller+0.132
 salon+0.124
atters+0.118
ara+0.118
 clinic+0.115
Neg logit contributions
ling-0.091
 around-0.085
 shining-0.082
 in-0.074
 him-0.074
Token joy
Token ID8716
Feature activation+2.14e-01

Pos logit contributions
roller+0.068
 salon+0.066
 heaven+0.064
ara+0.062
 clinic+0.061
Neg logit contributions
ling-0.080
led-0.068
 around-0.068
 shining-0.067
ie-0.064
Token.
Token ID13
Feature activation-1.32e-01

Pos logit contributions
roller+0.010
 salon+0.010
post+0.009
ara+0.009
 heaven+0.009
Neg logit contributions
ling-0.010
 around-0.008
led-0.008
 shining-0.008
 spear-0.008
Token the
Token ID262
Feature activation-1.99e-01

Pos logit contributions
ling+0.011
led+0.008
ie+0.008
ering+0.008
 around+0.008
Neg logit contributions
roller-0.013
 pitcher-0.013
ara-0.013
 salon-0.013
 jail-0.012
Token rabbit
Token ID22746
Feature activation-5.27e+00

Pos logit contributions
ling+0.008
 around+0.007
led+0.007
 shining+0.006
 him+0.006
Neg logit contributions
roller-0.011
 vet-0.011
ara-0.011
 pitcher-0.011
 salon-0.011
Token said
Token ID531
Feature activation-1.54e+00

Pos logit contributions
ling+0.205
 around+0.174
 shining+0.172
led+0.161
 up+0.154
Neg logit contributions
roller-0.318
 salon-0.303
ara-0.290
 heaven-0.282
 pitcher-0.280
Token to
Token ID284
Feature activation+4.04e-01

Pos logit contributions
ling+0.077
 shining+0.065
led+0.064
ie+0.061
 around+0.061
Neg logit contributions
roller-0.073
 salon-0.069
ara-0.065
 pitcher-0.065
ress-0.064
Token the
Token ID262
Feature activation-1.81e-01

Pos logit contributions
roller+0.027
 salon+0.026
ara+0.025
 heaven+0.025
atters+0.025
Neg logit contributions
ling-0.013
 around-0.010
 shining-0.010
led-0.010
ie-0.008
Token sheep
Token ID15900
Feature activation-4.47e+00

Pos logit contributions
ling+0.007
led+0.007
 around+0.006
 atop+0.005
 shining+0.005
Neg logit contributions
roller-0.010
 vet-0.010
 salon-0.009
ara-0.009
 pitcher-0.009
Token,
Token ID11
Feature activation-2.30e-01

Pos logit contributions
ling+0.207
led+0.181
 shining+0.179
 around+0.170
ging+0.162
Neg logit contributions
roller-0.221
 salon-0.216
ara-0.200
 clinic-0.199
 heaven-0.198
Token "
Token ID366
Feature activation-1.98e+00

Pos logit contributions
ling+0.013
led+0.012
ie+0.011
itive+0.010
 shining+0.010
Neg logit contributions
roller-0.010
 salon-0.009
ara-0.009
 vet-0.009
post-0.008
TokenI
Token ID40
Feature activation+5.14e-01

Pos logit contributions
ling+0.102
 around+0.090
 shining+0.088
led+0.084
 up+0.081
Neg logit contributions
roller-0.101
 salon-0.093
atters-0.086
 clinic-0.085
 heaven-0.083
Token think
Token ID892
Feature activation-3.44e+00

Pos logit contributions
roller+0.030
 salon+0.028
ara+0.026
ress+0.026
post+0.026
Neg logit contributions
ling-0.022
 around-0.020
 shining-0.018
led-0.017
 up-0.017
Token your
Token ID534
Feature activation+3.70e+00

Pos logit contributions
ling+0.205
led+0.178
 shining+0.171
 around+0.168
 atop+0.163
Neg logit contributions
roller-0.132
 salon-0.128
 heaven-0.119
ara-0.117
 pitcher-0.113
Token day
Token ID1110
Feature activation-1.33e+00

Pos logit contributions
ling+0.014
 around+0.012
 shining+0.012
 up+0.010
led+0.010
Neg logit contributions
roller-0.025
 salon-0.024
 heaven-0.022
ress-0.022
ara-0.022
Token at
Token ID379
Feature activation-1.86e+00

Pos logit contributions
ling+0.046
 around+0.039
 shining+0.037
led+0.035
 up+0.033
Neg logit contributions
roller-0.087
 salon-0.083
ara-0.080
 heaven-0.078
atters-0.078
Token the
Token ID262
Feature activation-3.44e-01

Pos logit contributions
 around+0.027
ling+0.025
 up+0.024
 shining+0.022
 him+0.020
Neg logit contributions
roller-0.153
 salon-0.144
ara-0.142
 pitcher-0.140
lor-0.140
Token park
Token ID3952
Feature activation+4.92e+00

Pos logit contributions
ling+0.018
 up+0.018
 around+0.018
 shining+0.018
 all+0.017
Neg logit contributions
roller-0.015
atters-0.013
ress-0.013
ird-0.012
post-0.012
Token,
Token ID11
Feature activation-2.89e-01

Pos logit contributions
roller+0.309
 salon+0.293
ara+0.282
atters+0.276
 pitcher+0.274
Neg logit contributions
ling-0.182
 around-0.160
 shining-0.149
 up-0.139
led-0.138
Token playing
Token ID2712
Feature activation-4.01e+00

Pos logit contributions
ling+0.010
led+0.008
 around+0.008
 shining+0.008
ie+0.008
Neg logit contributions
roller-0.018
 salon-0.018
ara-0.017
 pitcher-0.016
 heaven-0.016
Token tag
Token ID7621
Feature activation+4.24e+00

Pos logit contributions
ling+0.166
 shining+0.142
 around+0.139
led+0.134
ie+0.128
Neg logit contributions
roller-0.227
 salon-0.217
ara-0.207
 heaven-0.203
 clinic-0.202
Token and
Token ID290
Feature activation-4.40e-01

Pos logit contributions
roller+0.219
 salon+0.209
ara+0.198
post+0.195
 clinic+0.192
Neg logit contributions
ling-0.189
 shining-0.165
led-0.157
 around-0.156
ie-0.150
Token chasing
Token ID20023
Feature activation-4.18e+00

Pos logit contributions
ling+0.017
 around+0.014
led+0.014
 shining+0.013
ie+0.013
Neg logit contributions
roller-0.027
 salon-0.026
ara-0.024
 pitcher-0.024
 heaven-0.024
Token each
Token ID1123
Feature activation-5.72e+00

Pos logit contributions
ling+0.158
 shining+0.152
led+0.126
ie+0.126
itive+0.124
Neg logit contributions
roller-0.231
ara-0.225
 heaven-0.220
 salon-0.213
 jail-0.213
Token other
Token ID584
Feature activation+1.23e+00

Pos logit contributions
ling+0.187
 around+0.154
 shining+0.150
led+0.141
 up+0.132
Neg logit contributions
roller-0.381
 salon-0.366
ara-0.350
 heaven-0.342
atters-0.340
Token.
Token ID13
Feature activation-2.21e-01

Pos logit contributions
ling+0.000
 around+0.000
 shining+0.000
 up+0.000
led+0.000
Neg logit contributions
roller-0.001
 salon-0.001
ara-0.001
 pitcher-0.001
atters-0.001
Token She
Token ID1375
Feature activation+2.15e+00

Pos logit contributions
 around+0.012
ling+0.012
 shining+0.011
 up+0.011
 him+0.011
Neg logit contributions
roller-0.009
 salon-0.008
ara-0.008
 pitcher-0.008
lor-0.008
Token saw
Token ID2497
Feature activation-5.08e-01

Pos logit contributions
 salon+0.113
roller+0.113
ara+0.107
 clinic+0.101
 heaven+0.099
Neg logit contributions
ling-0.098
 shining-0.082
led-0.078
 around-0.078
ging-0.076
Token a
Token ID257
Feature activation-3.38e-02

Pos logit contributions
ling+0.026
ging+0.021
led+0.021
 around+0.020
 atop+0.020
Neg logit contributions
 salon-0.024
 vet-0.024
 heaven-0.022
roller-0.022
 jail-0.022
Token big
Token ID1263
Feature activation+1.52e+00

Pos logit contributions
ling+0.001
 around+0.001
 shining+0.001
led+0.001
 up+0.001
Neg logit contributions
roller-0.002
 salon-0.002
ara-0.002
 heaven-0.002
 pitcher-0.002
Token bear
Token ID6842
Feature activation-4.80e+00

Pos logit contributions
roller+0.080
 salon+0.075
ara+0.071
atters+0.070
post+0.070
Neg logit contributions
ling-0.071
 around-0.063
 shining-0.062
 up-0.057
ie-0.057
Token and
Token ID290
Feature activation-4.40e-01

Pos logit contributions
ling+0.172
 around+0.142
 shining+0.138
led+0.137
 up+0.123
Neg logit contributions
roller-0.302
 salon-0.291
ara-0.275
 heaven-0.272
 clinic-0.272
Token thought
Token ID1807
Feature activation-4.53e+00

Pos logit contributions
ling+0.023
ging+0.018
 around+0.018
led+0.017
ering+0.017
Neg logit contributions
 vet-0.019
 salon-0.019
roller-0.019
 heaven-0.018
ara-0.018
Token it
Token ID340
Feature activation-4.30e-02

Pos logit contributions
ling+0.235
led+0.192
 shining+0.185
uses+0.181
 atop+0.178
Neg logit contributions
 salon-0.199
roller-0.199
post-0.196
 heaven-0.188
ara-0.185
Token was
Token ID373
Feature activation+8.73e-02

Pos logit contributions
ling+0.002
ging+0.002
nam+0.002
itive+0.002
led+0.002
Neg logit contributions
post-0.001
house-0.001
roller-0.001
 salon-0.001
ara-0.001
Token cute
Token ID13779
Feature activation-3.02e+00

Pos logit contributions
 heaven+0.004
 jail+0.004
roller+0.004
 vet+0.004
 salon+0.004
Neg logit contributions
ling-0.004
led-0.004
ging-0.003
ged-0.003
 around-0.003
Token it
Token ID340
Feature activation-5.91e-01

Pos logit contributions
ling+0.056
 shining+0.044
led+0.044
 around+0.042
itive+0.042
Neg logit contributions
roller-0.067
 heaven-0.065
 salon-0.063
 jail-0.062
ara-0.062
Token anywhere
Token ID6609
Feature activation-4.15e+00

Pos logit contributions
ling+0.029
led+0.026
 shining+0.025
bee+0.024
itive+0.023
Neg logit contributions
roller-0.026
 salon-0.024
 heaven-0.024
post-0.023
 jail-0.023
Token.
Token ID13
Feature activation-1.60e-01

Pos logit contributions
ling+0.191
 shining+0.159
led+0.159
 around+0.154
ie+0.143
Neg logit contributions
roller-0.216
 salon-0.206
 heaven-0.198
post-0.193
 jail-0.191
Token The
Token ID383
Feature activation+7.78e-01

Pos logit contributions
ling+0.006
 around+0.006
 him+0.006
 up+0.006
 shining+0.005
Neg logit contributions
roller-0.009
 salon-0.009
atters-0.008
ara-0.008
 pitcher-0.008
Token creatures
Token ID8109
Feature activation-5.50e+00

Pos logit contributions
roller+0.040
 salon+0.040
ara+0.037
 pitcher+0.037
 vet+0.036
Neg logit contributions
ling-0.035
 around-0.031
led-0.030
 shining-0.030
 him-0.028
Token in
Token ID287
Feature activation-2.10e+00

Pos logit contributions
ling+0.192
 shining+0.192
fl+0.181
 up+0.167
itive+0.166
Neg logit contributions
roller-0.257
 salon-0.256
 fountain-0.239
 pitcher-0.237
post-0.233
Token the
Token ID262
Feature activation-6.87e-01

Pos logit contributions
ling+0.062
 around+0.048
 shining+0.048
led+0.045
 up+0.041
Neg logit contributions
roller-0.147
 salon-0.141
ara-0.136
 heaven-0.134
 pitcher-0.132
Token jar
Token ID17379
Feature activation+1.33e+00

Pos logit contributions
ling+0.038
 around+0.034
 shining+0.033
 up+0.031
led+0.031
Neg logit contributions
roller-0.032
 salon-0.029
atters-0.028
ara-0.027
ress-0.027
Token were
Token ID547
Feature activation-9.25e-01

Pos logit contributions
roller+0.083
 salon+0.079
ara+0.076
 heaven+0.074
atters+0.074
Neg logit contributions
ling-0.049
 around-0.042
 shining-0.040
led-0.038
 up-0.036
Token gone
Token ID3750
Feature activation+2.22e+00

Pos logit contributions
ling+0.036
led+0.029
 shining+0.029
ie+0.029
 around+0.029
Neg logit contributions
roller-0.053
 salon-0.051
 heaven-0.048
 pitcher-0.048
 clinic-0.048
Token,
Token ID11
Feature activation-2.84e-01

Pos logit contributions
roller+0.125
 salon+0.118
ara+0.111
 heaven+0.110
post+0.109
Neg logit contributions
ling-0.093
 around-0.079
 shining-0.078
led-0.076
ie-0.071
Token excited
Token ID6568
Feature activation-5.95e-01

Pos logit contributions
ling+0.018
 around+0.015
 shining+0.015
led+0.014
ie+0.014
Neg logit contributions
roller-0.034
 salon-0.033
ara-0.031
 pitcher-0.031
atters-0.031
Token and
Token ID290
Feature activation-5.59e-01

Pos logit contributions
ling+0.027
 shining+0.023
led+0.022
 around+0.021
 atop+0.021
Neg logit contributions
roller-0.030
 salon-0.029
ara-0.028
post-0.027
 heaven-0.027
Token began
Token ID2540
Feature activation+1.92e+00

Pos logit contributions
ling+0.022
 shining+0.018
 around+0.018
led+0.017
ie+0.016
Neg logit contributions
roller-0.033
 salon-0.032
ara-0.030
 pitcher-0.029
 heaven-0.029
Token to
Token ID284
Feature activation+2.46e-01

Pos logit contributions
roller+0.118
 salon+0.112
ara+0.105
 heaven+0.105
 clinic+0.104
Neg logit contributions
ling-0.074
 around-0.061
 shining-0.060
led-0.059
ie-0.055
Token count
Token ID954
Feature activation-8.68e-01

Pos logit contributions
roller+0.013
 salon+0.013
ara+0.012
 heaven+0.012
 pitcher+0.012
Neg logit contributions
ling-0.011
 around-0.010
 shining-0.009
led-0.009
 up-0.009
Token how
Token ID703
Feature activation-7.65e+00

Pos logit contributions
ling+0.035
led+0.029
 shining+0.027
itive+0.026
bee+0.025
Neg logit contributions
roller-0.050
 salon-0.049
ara-0.045
post-0.044
ress-0.044
Token many
Token ID867
Feature activation+1.14e+00

Pos logit contributions
ling+0.344
 shining+0.283
bee+0.282
itive+0.274
led+0.273
Neg logit contributions
roller-0.385
 salon-0.380
ara-0.354
post-0.349
ress-0.344
Token patterns
Token ID7572
Feature activation-7.42e-01

Pos logit contributions
roller+0.058
 salon+0.057
ara+0.055
 pitcher+0.052
 heaven+0.052
Neg logit contributions
ling-0.051
 shining-0.043
 around-0.043
led-0.041
 up-0.040
Token were
Token ID547
Feature activation-1.14e+00

Pos logit contributions
ling+0.029
 shining+0.022
 around+0.021
led+0.021
itive+0.020
Neg logit contributions
roller-0.041
 salon-0.040
post-0.039
ara-0.038
 pitcher-0.037
Token on
Token ID319
Feature activation-3.40e+00

Pos logit contributions
ling+0.043
led+0.035
 around+0.034
 shining+0.033
ie+0.031
Neg logit contributions
roller-0.068
 salon-0.066
 pitcher-0.061
 heaven-0.061
post-0.061
Token the
Token ID262
Feature activation-3.74e-01

Pos logit contributions
ling+0.111
 around+0.091
 shining+0.088
led+0.083
 up+0.078
Neg logit contributions
roller-0.227
 salon-0.217
ara-0.208
 heaven-0.204
 pitcher-0.203
Token flew
Token ID13112
Feature activation+3.62e-01

Pos logit contributions
ling+0.012
 shining+0.010
 around+0.010
 him+0.009
led+0.009
Neg logit contributions
roller-0.016
 salon-0.016
ara-0.015
 clinic-0.015
 heaven-0.015
Token in
Token ID287
Feature activation-2.04e+00

Pos logit contributions
roller+0.023
post+0.022
 salon+0.020
ara+0.020
atters+0.020
Neg logit contributions
ling-0.011
itive-0.009
led-0.009
uses-0.008
bee-0.008
Token circles
Token ID13332
Feature activation+1.38e+00

Pos logit contributions
ling+0.047
 around+0.043
 shining+0.041
 up+0.036
 him+0.035
Neg logit contributions
roller-0.155
 salon-0.150
ara-0.144
atters-0.143
 clinic-0.141
Token,
Token ID11
Feature activation-4.68e-01

Pos logit contributions
roller+0.068
post+0.065
 salon+0.063
ara+0.060
 heaven+0.059
Neg logit contributions
ling-0.059
nam-0.050
 shining-0.048
led-0.048
uses-0.048
Token swoop
Token ID38527
Feature activation-2.70e+00

Pos logit contributions
ling+0.017
ie+0.015
led+0.015
itive+0.014
 him+0.013
Neg logit contributions
roller-0.025
 salon-0.025
post-0.024
 heaven-0.023
atters-0.022
Tokening
Token ID278
Feature activation-7.67e+00

Pos logit contributions
ling+0.086
nam+0.074
ie+0.071
 shining+0.070
uses+0.067
Neg logit contributions
roller-0.163
ara-0.147
post-0.146
 salon-0.143
lor-0.139
Token and
Token ID290
Feature activation-6.15e-01

Pos logit contributions
ling+0.254
ie+0.208
uses+0.200
itive+0.200
nam+0.199
Neg logit contributions
roller-0.447
post-0.420
 salon-0.398
ara-0.387
 pitcher-0.386
Token diving
Token ID23186
Feature activation-1.59e+00

Pos logit contributions
ling+0.028
ie+0.022
led+0.022
 around+0.021
itive+0.021
Neg logit contributions
roller-0.031
 salon-0.031
post-0.029
 heaven-0.028
ara-0.028
Token.
Token ID13
Feature activation-4.99e-01

Pos logit contributions
ling+0.048
 shining+0.038
ie+0.036
led+0.034
 around+0.033
Neg logit contributions
roller-0.104
post-0.097
 salon-0.094
ara-0.093
atters-0.092
Token Then
Token ID3244
Feature activation+2.43e+00

Pos logit contributions
ling+0.025
 around+0.024
 shining+0.022
 him+0.021
 up+0.021
Neg logit contributions
roller-0.024
 salon-0.023
ara-0.021
 heaven-0.020
 pitcher-0.020
Token,
Token ID11
Feature activation-3.36e-01

Pos logit contributions
roller+0.148
 salon+0.142
ara+0.136
 heaven+0.133
 clinic+0.132
Neg logit contributions
ling-0.093
 around-0.075
 shining-0.074
led-0.072
ie-0.066
Token the
Token ID262
Feature activation-2.31e-02

Pos logit contributions
ling+0.002
 shining+0.001
led+0.001
 around+0.001
itive+0.001
Neg logit contributions
roller-0.002
ara-0.002
post-0.002
S-0.002
 salon-0.002
Token hook
Token ID8011
Feature activation-2.33e-01

Pos logit contributions
ling+0.001
 around+0.001
 shining+0.001
led+0.001
ie+0.001
Neg logit contributions
roller-0.001
 salon-0.001
ara-0.001
 pitcher-0.001
 clinic-0.001
Token and
Token ID290
Feature activation-3.01e-01

Pos logit contributions
ling+0.010
 shining+0.009
 around+0.008
led+0.008
ie+0.008
Neg logit contributions
roller-0.013
 salon-0.012
post-0.011
 heaven-0.011
 pitcher-0.011
Token set
Token ID900
Feature activation+4.25e+00

Pos logit contributions
ling+0.012
 around+0.011
 shining+0.010
led+0.010
 up+0.009
Neg logit contributions
roller-0.017
 salon-0.017
ara-0.016
 heaven-0.015
 pitcher-0.015
Token Bub
Token ID19645
Feature activation-4.63e+00

Pos logit contributions
roller+0.276
 salon+0.255
atters+0.248
 heaven+0.247
post+0.247
Neg logit contributions
ling-0.142
 shining-0.112
led-0.109
 around-0.106
ie-0.098
Tokenbles
Token ID7689
Feature activation-8.48e+00

Pos logit contributions
ling+0.126
 around+0.120
 shining+0.120
led+0.098
 him+0.098
Neg logit contributions
roller-0.313
 salon-0.292
ara-0.290
atters-0.282
lor-0.280
Token free
Token ID1479
Feature activation+1.94e+00

Pos logit contributions
ling+0.344
 shining+0.304
led+0.288
 around+0.278
ie+0.271
Neg logit contributions
roller-0.469
 salon-0.441
atters-0.428
 heaven-0.425
 pitcher-0.418
Token.
Token ID13
Feature activation-1.26e-01

Pos logit contributions
roller+0.106
 salon+0.099
 heaven+0.098
post+0.097
ara+0.095
Neg logit contributions
ling-0.079
 shining-0.067
led-0.066
 around-0.064
ie-0.060
Token Bub
Token ID19645
Feature activation-4.64e+00

Pos logit contributions
ling+0.004
 around+0.004
 shining+0.003
 up+0.003
led+0.003
Neg logit contributions
roller-0.008
 salon-0.008
ara-0.008
 heaven-0.008
 pitcher-0.008
Tokenbles
Token ID7689
Feature activation-8.48e+00

Pos logit contributions
ling+0.130
 around+0.125
 shining+0.123
 him+0.107
led+0.100
Neg logit contributions
roller-0.308
ara-0.292
 salon-0.289
ress-0.279
atters-0.277
Token was
Token ID373
Feature activation+1.65e-01

Pos logit contributions
ling+0.348
 shining+0.316
 around+0.298
 up+0.279
led+0.276
Neg logit contributions
roller-0.451
 salon-0.425
 pitcher-0.404
ara-0.401
ress-0.396
Token son
Token ID3367
Feature activation-3.39e+00

Pos logit contributions
ling+0.016
 around+0.015
 shining+0.015
 him+0.014
 up+0.014
Neg logit contributions
roller-0.019
 salon-0.017
atters-0.016
 clinic-0.016
ara-0.016
Token.
Token ID13
Feature activation-1.05e-01

Pos logit contributions
ling+0.133
 around+0.115
 shining+0.109
led+0.104
 up+0.101
Neg logit contributions
roller-0.206
 salon-0.195
ara-0.185
 pitcher-0.181
atters-0.181
Token This
Token ID770
Feature activation-1.09e+00

Pos logit contributions
ling+0.005
 around+0.005
 him+0.005
 shining+0.005
 up+0.005
Neg logit contributions
roller-0.005
 salon-0.005
post-0.004
 pitcher-0.004
 bidder-0.004
Token lovely
Token ID14081
Feature activation-1.97e-01

Pos logit contributions
ling+0.060
 around+0.054
 shining+0.053
 up+0.050
led+0.049
Neg logit contributions
roller-0.049
 salon-0.046
ara-0.042
atters-0.042
post-0.042
Token muff
Token ID27563
Feature activation-3.64e+00

Pos logit contributions
ling+0.009
 around+0.008
 shining+0.008
led+0.008
 up+0.007
Neg logit contributions
roller-0.010
 salon-0.010
ara-0.009
 heaven-0.009
 clinic-0.009
Tokenin
Token ID259
Feature activation-7.93e+00

Pos logit contributions
ling+0.125
 shining+0.109
 around+0.108
 up+0.096
 him+0.095
Neg logit contributions
roller-0.222
 salon-0.221
ara-0.211
 pitcher-0.205
atters-0.204
Token isn
Token ID2125
Feature activation-6.47e-01

Pos logit contributions
ling+0.386
 around+0.345
 shining+0.327
led+0.321
 up+0.311
Neg logit contributions
roller-0.413
 salon-0.385
ara-0.361
atters-0.349
 pitcher-0.349
Token't
Token ID470
Feature activation+1.20e+00

Pos logit contributions
ling+0.037
 around+0.033
 shining+0.033
led+0.032
ie+0.030
Neg logit contributions
roller-0.027
 salon-0.025
ara-0.024
atters-0.023
 pitcher-0.023
Token for
Token ID329
Feature activation-2.02e+00

Pos logit contributions
roller+0.063
 salon+0.060
 heaven+0.059
ara+0.057
post+0.057
Neg logit contributions
ling-0.054
led-0.045
 around-0.045
 shining-0.044
 atop-0.040
Token us
Token ID514
Feature activation+4.78e-01

Pos logit contributions
ling+0.104
 around+0.089
led+0.087
 shining+0.087
 up+0.080
Neg logit contributions
roller-0.097
 salon-0.093
 heaven-0.086
ara-0.086
 pitcher-0.084
Token,
Token ID11
Feature activation-4.36e-01

Pos logit contributions
roller+0.024
 salon+0.023
ara+0.022
 heaven+0.021
atters+0.021
Neg logit contributions
ling-0.023
 around-0.020
 shining-0.020
led-0.019
 up-0.018
Token like
Token ID588
Feature activation-2.95e+00

Pos logit contributions
ling+0.051
 around+0.042
led+0.042
 shining+0.042
 him+0.039
Neg logit contributions
roller-0.070
 salon-0.066
 pitcher-0.063
ara-0.062
post-0.062
Token bright
Token ID6016
Feature activation-1.30e+00

Pos logit contributions
ling+0.119
 around+0.099
led+0.094
 shining+0.094
 up+0.087
Neg logit contributions
roller-0.174
 salon-0.168
ara-0.159
 heaven-0.157
 pitcher-0.156
Token yellow
Token ID7872
Feature activation-1.67e+00

Pos logit contributions
ling+0.053
 around+0.045
 shining+0.044
led+0.043
 up+0.041
Neg logit contributions
roller-0.071
ara-0.066
 salon-0.066
 pitcher-0.064
 heaven-0.063
Token neck
Token ID7393
Feature activation-3.47e+00

Pos logit contributions
ling+0.071
 around+0.062
 shining+0.059
led+0.059
 up+0.056
Neg logit contributions
roller-0.089
 salon-0.084
ara-0.083
 pitcher-0.080
 heaven-0.079
Tokenl
Token ID75
Feature activation-5.84e+00

Pos logit contributions
ling+0.142
 shining+0.123
 around+0.123
led+0.122
 him+0.117
Neg logit contributions
roller-0.186
 salon-0.174
ara-0.167
post-0.166
 pitcher-0.160
Tokenaces
Token ID2114
Feature activation-7.59e+00

Pos logit contributions
ling+0.253
 shining+0.227
 around+0.214
nam+0.211
 Barry+0.208
Neg logit contributions
roller-0.280
post-0.262
 heaven-0.259
ara-0.256
atters-0.255
Token and
Token ID290
Feature activation-3.67e-01

Pos logit contributions
ling+0.322
 shining+0.265
 around+0.263
led+0.260
 him+0.250
Neg logit contributions
roller-0.418
 salon-0.397
post-0.373
atters-0.369
ara-0.365
Token flower
Token ID15061
Feature activation+8.02e-01

Pos logit contributions
ling+0.016
 around+0.012
led+0.012
ie+0.012
 shining+0.011
Neg logit contributions
roller-0.020
 salon-0.019
 pitcher-0.019
post-0.018
ara-0.018
Token crown
Token ID12389
Feature activation-1.37e+00

Pos logit contributions
roller+0.044
 salon+0.041
ara+0.039
 pitcher+0.039
 heaven+0.039
Neg logit contributions
ling-0.034
 around-0.029
 shining-0.028
led-0.028
 up-0.026
Tokens
Token ID82
Feature activation-1.07e+00

Pos logit contributions
ling+0.060
 shining+0.052
 around+0.050
led+0.047
 up+0.047
Neg logit contributions
roller-0.072
 salon-0.068
ara-0.065
post-0.065
atters-0.062
Token.
Token ID13
Feature activation-2.27e-01

Pos logit contributions
ling+0.046
led+0.037
 around+0.037
ie+0.035
 shining+0.035
Neg logit contributions
roller-0.057
 salon-0.052
post-0.050
 pitcher-0.050
 heaven-0.049