TOP ACTIVATIONS
MAX = 8.257

One day , the sausage saw a bright light and
a time there was a fle a who lived near a
Once , there was a fle a who did not want
One day , Max got fle as all over his fur
time there was a light sausage . The sausage was looking
was very happy with his sausage , but he was also
park . Every day the fle a saw a child playing
he came across a white fle a . He looked at
. One day , a fle a came to the beach
a light sausage . The sausage was looking for someone to
it was hard for the sausage to find the perfect match
an adventure in a delicious sausage forest . She saw so
, forgetting all about the sausage he found . <|endoftext|> Once
One day , the fle a decided to give it
, she made a big sausage for her friend 's birthday
Max had a lot of fle as on his fur .
and Tom fighting over the sausage . " Stop
, there was a restless fle a . She wanted to
," and he grabbed the fle a 's legs . Together
time , there lived a fle a . He was very

MIN ACTIVATIONS
MIN = -8.402

big , beautiful rainbow - col oured butterfly ! The little
a tall tree . Un bek now st to him ,
grandma 's house for safe keeping . When
. It is the door bell . Mom my is home
it to me for safe keeping . He told me to
Soon , the door bell rang . Lily ran to
. Soon , the door bell rang and Anna ran to
. It is the door bell . She knows it is
got to ring the door bell and say hello to the
. It was the door bell ringing ! L
said the mirror was a portal to another world . He
her wand and a magical portal opened . The
. The light was a portal . It took her to
better and then the door bell rang . Mom opened the
oak tree , their secret portal . They never told anyone
. They rang the door bell to see what was going
The driver rang the door bell and her mom answered it
heard before was the door bell . From then on ,
ringing sound of the door bell was still ringing in Lucy
Soon , the door bell rang and Lily 's friends

INTERVAL 4.092 - 8.257
CONTAINS 0.805%

giving was just as special as receiving . <|endoftext|> Once there
forgot about her l oll ipop and had so much fun
" By e , fish !" his dad said ,
We can use our bikes as rockets and our helmets as
funny that the shadow was as tall as he was !

INTERVAL -0.072 - 4.092
CONTAINS 51.599%

. When they arrived , the store was full of different
. He did not know that Lily was in the barn
the grass . She went over to it and picked it
enjoyed the warm cookies together , and Tim my 's toy
little girl named Lily who loved to explore . One day

INTERVAL -4.237 - -0.072
CONTAINS 46.715%

She likes to sing songs with her guitar . Sometimes she
hid the coin in his pocket and didn 't say anything
" Sarah , your doll is so pretty . Can I
his bag . They held the treat in their hands and
my loves . We can get new toys and a new

INTERVAL -8.402 - -4.237
CONTAINS 0.881%

pretend to sleep . They hope Mom won 't notice the
could fly and explore the world . The moral
decided to throw a big party and sent out invitations to
. It pulled the fridge door and made a loud noise
a reward , I will show you something special ". Jack
Token
Token ID198
Feature activation+4.99e-01

Pos logit contributions
oyd+0.013
 pistols+0.010
 neither+0.009
 brothers+0.009
 sometimes+0.009
Neg logit contributions
igsaw-0.022
hy-0.021
 brand-0.020
 doorway-0.020
 click-0.020
TokenOne
Token ID3198
Feature activation+3.75e+00

Pos logit contributions
oyd+0.014
 pistols+0.011
 brothers+0.010
 neither+0.010
 sometimes+0.010
Neg logit contributions
igsaw-0.021
hy-0.021
 brand-0.020
 doorway-0.020
 click-0.020
Token day
Token ID1110
Feature activation+1.54e+00

Pos logit contributions
oyd+0.091
 pistols+0.071
 neither+0.061
 brothers+0.061
 sometimes+0.059
Neg logit contributions
igsaw-0.175
hy-0.171
 brand-0.164
 doorway-0.162
 click-0.161
Token,
Token ID11
Feature activation+6.97e-01

Pos logit contributions
oyd+0.046
 pistols+0.038
 neither+0.035
 brothers+0.034
 sometimes+0.034
Neg logit contributions
igsaw-0.063
hy-0.061
 brand-0.059
 doorway-0.057
 click-0.057
Token the
Token ID262
Feature activation+2.01e-01

Pos logit contributions
oyd+0.023
 pistols+0.019
 brothers+0.017
 neither+0.017
 sometimes+0.017
Neg logit contributions
igsaw-0.026
hy-0.026
 brand-0.024
 doorway-0.024
 click-0.024
Token sausage
Token ID31889
Feature activation+8.15e+00

Pos logit contributions
oyd+0.006
 pistols+0.005
 brothers+0.004
 neither+0.004
 sometimes+0.004
Neg logit contributions
igsaw-0.008
hy-0.008
 brand-0.008
 doorway-0.008
 click-0.008
Token saw
Token ID2497
Feature activation-2.01e+00

Pos logit contributions
oyd+0.257
 pistols+0.217
 neither+0.193
Rock+0.188
ained+0.185
Neg logit contributions
hy-0.293
igsaw-0.293
 doorway-0.286
 brand-0.278
 click-0.275
Token a
Token ID257
Feature activation+2.82e-02

Pos logit contributions
igsaw+0.091
hy+0.087
 click+0.083
 brand+0.082
 doorway+0.082
Neg logit contributions
oyd-0.053
 pistols-0.043
 brothers-0.039
 neither-0.038
 sometimes-0.037
Token bright
Token ID6016
Feature activation-2.54e+00

Pos logit contributions
oyd+0.001
 pistols+0.001
 brothers+0.001
ball+0.001
 neither+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
col-0.001
 doorway-0.001
 brand-0.001
Token light
Token ID1657
Feature activation-3.45e+00

Pos logit contributions
hy+0.096
igsaw+0.096
des+0.090
lic+0.089
col+0.088
Neg logit contributions
oyd-0.077
 pistols-0.063
 brothers-0.063
 sometimes-0.061
 on-0.060
Token and
Token ID290
Feature activation+4.82e-02

Pos logit contributions
igsaw+0.152
hy+0.146
 brand+0.139
 click+0.137
 disl+0.137
Neg logit contributions
oyd-0.091
 pistols-0.072
 sometimes-0.067
ball-0.066
 on-0.065
Token a
Token ID257
Feature activation-5.04e-02

Pos logit contributions
igsaw+0.009
hy+0.008
 brand+0.008
 doorway+0.008
 click+0.008
Neg logit contributions
oyd-0.006
 pistols-0.005
 neither-0.005
 brothers-0.005
 sometimes-0.004
Token time
Token ID640
Feature activation-1.41e+00

Pos logit contributions
igsaw+0.002
hy+0.002
 brand+0.002
 doorway+0.002
col+0.002
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 sometimes-0.001
 brothers-0.001
Token there
Token ID612
Feature activation-7.77e-02

Pos logit contributions
igsaw+0.053
hy+0.051
 brand+0.048
 click+0.048
 doorway+0.047
Neg logit contributions
oyd-0.047
 pistols-0.040
 brothers-0.036
 neither-0.035
ball-0.035
Token was
Token ID373
Feature activation+9.00e-01

Pos logit contributions
hy+0.003
igsaw+0.003
 brand+0.003
 doorway+0.003
 click+0.003
Neg logit contributions
oyd-0.002
 pistols-0.002
 horses-0.002
 brothers-0.002
 spears-0.002
Token a
Token ID257
Feature activation-9.26e-03

Pos logit contributions
oyd+0.032
 pistols+0.029
 neither+0.026
eper+0.026
 forth+0.025
Neg logit contributions
igsaw-0.030
hy-0.029
 brand-0.028
col-0.027
 doorway-0.026
Token fle
Token ID5104
Feature activation+8.14e+00

Pos logit contributions
igsaw+0.000
hy+0.000
 brand+0.000
 click+0.000
 doorway+0.000
Neg logit contributions
oyd-0.000
 pistols-0.000
ained-0.000
 neither-0.000
 sometimes-0.000
Tokena
Token ID64
Feature activation+1.27e+00

Pos logit contributions
 pistols+0.263
oyd+0.261
 forth+0.241
 horses+0.225
 neither+0.224
Neg logit contributions
hy-0.274
lic-0.266
igsaw-0.256
endant-0.255
col-0.249
Token who
Token ID508
Feature activation+1.12e+00

Pos logit contributions
oyd+0.047
 pistols+0.042
 neither+0.039
ained+0.038
 forth+0.037
Neg logit contributions
hy-0.041
igsaw-0.037
col-0.036
 brand-0.035
 doorway-0.034
Token lived
Token ID5615
Feature activation+1.24e+00

Pos logit contributions
oyd+0.027
 pistols+0.020
 sometimes+0.019
ball+0.019
 also+0.019
Neg logit contributions
igsaw-0.055
 doorway-0.050
 click-0.050
hy-0.048
 brand-0.048
Token near
Token ID1474
Feature activation+1.66e+00

Pos logit contributions
oyd+0.034
 pistols+0.030
 neither+0.025
 humans+0.022
 sometimes+0.022
Neg logit contributions
hy-0.050
 brand-0.049
 doorway-0.049
igsaw-0.048
col-0.047
Token a
Token ID257
Feature activation+3.61e-02

Pos logit contributions
oyd+0.058
 pistols+0.047
 neither+0.045
 forth+0.045
 goalie+0.042
Neg logit contributions
hy-0.057
igsaw-0.057
 brand-0.055
 click-0.052
col-0.051
TokenOnce
Token ID7454
Feature activation+9.94e-01

Pos logit contributions
oyd+0.066
 pistols+0.056
 neither+0.052
 forth+0.048
 sometimes+0.047
Neg logit contributions
igsaw-0.093
hy-0.090
 doorway-0.085
 brand-0.083
col-0.083
Token,
Token ID11
Feature activation+4.38e-01

Pos logit contributions
oyd+0.037
 pistols+0.032
Rock+0.029
ball+0.028
 horses+0.027
Neg logit contributions
igsaw-0.034
 brand-0.031
hy-0.030
 doorway-0.030
 Sus-0.029
Token there
Token ID612
Feature activation-2.89e-01

Pos logit contributions
oyd+0.016
 pistols+0.015
 mosquitoes+0.012
 humans+0.012
Rock+0.012
Neg logit contributions
hy-0.014
 brand-0.014
 Sus-0.013
 Jane-0.013
aney-0.013
Token was
Token ID373
Feature activation+1.04e+00

Pos logit contributions
hy+0.010
igsaw+0.010
 brand+0.010
 doorway+0.010
col+0.009
Neg logit contributions
oyd-0.009
 pistols-0.008
 forth-0.007
 horses-0.007
 brothers-0.007
Token a
Token ID257
Feature activation+8.33e-02

Pos logit contributions
oyd+0.042
 pistols+0.037
 forth+0.035
 neither+0.034
ey+0.034
Neg logit contributions
igsaw-0.031
hy-0.029
 brand-0.029
col-0.027
 Sus-0.026
Token fle
Token ID5104
Feature activation+8.08e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
ained+0.003
 neither+0.003
 sometimes+0.003
Neg logit contributions
igsaw-0.002
 brand-0.002
hy-0.002
 click-0.002
 doorway-0.002
Tokena
Token ID64
Feature activation+1.19e+00

Pos logit contributions
oyd+0.264
 pistols+0.252
 forth+0.233
 horses+0.216
 neither+0.213
Neg logit contributions
hy-0.286
lic-0.271
igsaw-0.269
endant-0.266
col-0.260
Token who
Token ID508
Feature activation+9.59e-01

Pos logit contributions
oyd+0.045
 pistols+0.039
 neither+0.036
 forth+0.034
ained+0.034
Neg logit contributions
hy-0.039
igsaw-0.036
col-0.035
 brand-0.034
 doorway-0.034
Token did
Token ID750
Feature activation+9.46e-01

Pos logit contributions
oyd+0.024
 pistols+0.019
 sometimes+0.018
ball+0.017
 also+0.017
Neg logit contributions
igsaw-0.045
 doorway-0.041
 click-0.041
hy-0.041
 brand-0.041
Token not
Token ID407
Feature activation+1.50e+00

Pos logit contributions
oyd+0.019
 pistols+0.013
 sometimes+0.012
 neither+0.012
 also+0.012
Neg logit contributions
igsaw-0.050
hy-0.047
 doorway-0.045
 click-0.045
col-0.045
Token want
Token ID765
Feature activation+1.80e+00

Pos logit contributions
oyd+0.038
 sometimes+0.033
 also+0.032
 pistols+0.030
 brothers+0.030
Neg logit contributions
igsaw-0.068
 brand-0.065
 doorway-0.065
icol-0.064
hy-0.060
Token One
Token ID1881
Feature activation+3.66e+00

Pos logit contributions
oyd+0.004
dd+0.004
ble+0.004
eper+0.004
w+0.003
Neg logit contributions
hy-0.002
igsaw-0.002
 brand-0.002
 doorway-0.002
 click-0.002
Token day
Token ID1110
Feature activation+9.90e-01

Pos logit contributions
oyd+0.138
 pistols+0.125
eper+0.115
 neither+0.112
 goalie+0.109
Neg logit contributions
hy-0.114
igsaw-0.112
 doorway-0.102
des-0.100
 sill-0.100
Token,
Token ID11
Feature activation+6.40e-01

Pos logit contributions
oyd+0.047
eper+0.040
opsy+0.040
 pistols+0.039
ble+0.039
Neg logit contributions
hy-0.021
 click-0.019
igsaw-0.018
 brand-0.018
lic-0.016
Token Max
Token ID5436
Feature activation+2.93e+00

Pos logit contributions
oyd+0.024
ey+0.022
opsy+0.021
eper+0.021
iles+0.020
Neg logit contributions
 brand-0.021
 click-0.020
hy-0.019
 doorway-0.016
igsaw-0.016
Token got
Token ID1392
Feature activation+3.90e-01

Pos logit contributions
oyd+0.109
 neither+0.089
 pistols+0.088
ble+0.088
ained+0.088
Neg logit contributions
hy-0.082
 doorway-0.080
 gn-0.079
 brand-0.077
col-0.077
Token fle
Token ID5104
Feature activation+8.07e+00

Pos logit contributions
oyd+0.018
elvet+0.014
ball+0.014
Le+0.014
 goalie+0.014
Neg logit contributions
 brand-0.010
hy-0.009
 aback-0.008
 doorway-0.008
 Sus-0.008
Tokenas
Token ID292
Feature activation+3.94e-01

Pos logit contributions
oyd+0.268
 pistols+0.235
 sometimes+0.208
 forth+0.206
 neither+0.206
Neg logit contributions
hy-0.283
igsaw-0.283
lic-0.276
col-0.265
 doorway-0.254
Token all
Token ID477
Feature activation+3.81e-01

Pos logit contributions
oyd+0.010
 pistols+0.007
 brothers+0.006
 neither+0.006
 sometimes+0.006
Neg logit contributions
igsaw-0.018
hy-0.018
 brand-0.017
 click-0.017
 doorway-0.017
Token over
Token ID625
Feature activation+1.44e+00

Pos logit contributions
oyd+0.009
 pistols+0.007
 sometimes+0.007
 neither+0.007
 brothers+0.007
Neg logit contributions
igsaw-0.018
hy-0.017
 click-0.017
 brand-0.017
 doorway-0.016
Token his
Token ID465
Feature activation+1.02e+00

Pos logit contributions
oyd+0.052
 pistols+0.039
 neither+0.038
ball+0.037
dd+0.035
Neg logit contributions
igsaw-0.051
 doorway-0.048
hy-0.047
 brand-0.046
col-0.044
Token fur
Token ID9230
Feature activation+1.90e+00

Pos logit contributions
oyd+0.034
 neither+0.028
dd+0.028
 sometimes+0.027
eper+0.027
Neg logit contributions
hy-0.035
igsaw-0.034
 doorway-0.034
 brand-0.034
 click-0.030
Token time
Token ID640
Feature activation-1.47e+00

Pos logit contributions
oyd+0.003
 pistols+0.003
 brothers+0.003
 neither+0.003
eper+0.003
Neg logit contributions
igsaw-0.005
hy-0.004
 brand-0.004
 doorway-0.004
col-0.004
Token there
Token ID612
Feature activation-5.33e-01

Pos logit contributions
igsaw+0.055
hy+0.053
 brand+0.050
 click+0.049
 doorway+0.048
Neg logit contributions
oyd-0.050
 pistols-0.042
 brothers-0.038
ball-0.037
 neither-0.037
Token was
Token ID373
Feature activation+1.17e+00

Pos logit contributions
hy+0.020
igsaw+0.019
 brand+0.019
 doorway+0.018
col+0.018
Neg logit contributions
oyd-0.017
 pistols-0.015
 brothers-0.013
 horses-0.013
 goalie-0.012
Token a
Token ID257
Feature activation+9.77e-02

Pos logit contributions
oyd+0.044
 pistols+0.039
 neither+0.036
 forth+0.036
eper+0.035
Neg logit contributions
igsaw-0.038
hy-0.037
 brand-0.035
col-0.033
 doorway-0.033
Token light
Token ID1657
Feature activation-3.47e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 neither+0.003
 sometimes+0.003
ained+0.003
Neg logit contributions
igsaw-0.003
hy-0.003
 brand-0.003
 click-0.003
 doorway-0.003
Token sausage
Token ID31889
Feature activation+8.06e+00

Pos logit contributions
igsaw+0.153
hy+0.145
 brand+0.140
 click+0.139
des+0.136
Neg logit contributions
oyd-0.096
 pistols-0.073
 sometimes-0.069
ball-0.069
 brothers-0.069
Token.
Token ID13
Feature activation+1.31e-01

Pos logit contributions
oyd+0.273
 pistols+0.231
 neither+0.208
Rock+0.203
 brothers+0.197
Neg logit contributions
hy-0.283
igsaw-0.273
 doorway-0.265
 brand-0.265
col-0.258
Token The
Token ID383
Feature activation-8.31e-01

Pos logit contributions
oyd+0.004
 sometimes+0.003
 brothers+0.003
 pistols+0.003
 Remy+0.003
Neg logit contributions
 brand-0.005
igsaw-0.005
col-0.005
icol-0.005
endant-0.005
Token sausage
Token ID31889
Feature activation+8.04e+00

Pos logit contributions
 brand+0.030
hy+0.029
 click+0.028
 doorway+0.028
igsaw+0.027
Neg logit contributions
oyd-0.027
 neither-0.022
 pistols-0.021
eper-0.021
Le-0.020
Token was
Token ID373
Feature activation+1.17e+00

Pos logit contributions
oyd+0.265
 pistols+0.231
 neither+0.209
 goalie+0.204
Rock+0.195
Neg logit contributions
hy-0.274
 brand-0.262
 doorway-0.261
col-0.260
igsaw-0.257
Token looking
Token ID2045
Feature activation-2.29e+00

Pos logit contributions
oyd+0.037
 pistols+0.031
 brothers+0.030
 sometimes+0.029
 also+0.029
Neg logit contributions
igsaw-0.047
hy-0.043
col-0.041
 click-0.041
 doorway-0.041
Token was
Token ID373
Feature activation+8.08e-01

Pos logit contributions
igsaw+0.011
icol+0.010
hy+0.010
 click+0.010
 doorway+0.010
Neg logit contributions
oyd-0.005
 pistols-0.004
 sometimes-0.004
 on-0.004
 neither-0.004
Token very
Token ID845
Feature activation-1.39e+00

Pos logit contributions
oyd+0.024
 pistols+0.020
 brothers+0.019
 neither+0.019
 sometimes+0.018
Neg logit contributions
igsaw-0.033
hy-0.032
 click-0.030
 brand-0.030
col-0.029
Token happy
Token ID3772
Feature activation-2.31e+00

Pos logit contributions
igsaw+0.056
hy+0.053
 doorway+0.050
icol+0.050
 click+0.050
Neg logit contributions
oyd-0.043
 also-0.035
 neither-0.035
 brothers-0.035
 friends-0.035
Token with
Token ID351
Feature activation-5.09e-01

Pos logit contributions
igsaw+0.106
hy+0.103
 click+0.097
 brand+0.097
 doorway+0.096
Neg logit contributions
oyd-0.055
 sometimes-0.046
 pistols-0.046
 friends-0.044
 brothers-0.043
Token his
Token ID465
Feature activation+8.04e-01

Pos logit contributions
hy+0.025
igsaw+0.024
lic+0.023
icol+0.023
 click+0.023
Neg logit contributions
oyd-0.009
 friends-0.009
 brothers-0.009
 sometimes-0.009
 pistols-0.009
Token sausage
Token ID31889
Feature activation+8.06e+00

Pos logit contributions
oyd+0.022
 pistols+0.018
 brothers+0.016
 neither+0.016
 sometimes+0.015
Neg logit contributions
igsaw-0.034
hy-0.034
 brand-0.032
 click-0.032
 doorway-0.032
Token,
Token ID11
Feature activation+5.09e-01

Pos logit contributions
oyd+0.253
 pistols+0.210
 brothers+0.190
 neither+0.190
 sometimes+0.187
Neg logit contributions
igsaw-0.315
hy-0.306
 brand-0.292
 doorway-0.287
 click-0.287
Token but
Token ID475
Feature activation+7.14e-01

Pos logit contributions
oyd+0.012
 pistols+0.010
 sometimes+0.009
 brothers+0.009
 also+0.009
Neg logit contributions
igsaw-0.024
hy-0.023
 doorway-0.022
lic-0.022
 click-0.021
Token he
Token ID339
Feature activation+2.94e+00

Pos logit contributions
oyd+0.022
 pistols+0.018
 brothers+0.016
ball+0.015
 neither+0.015
Neg logit contributions
igsaw-0.029
hy-0.028
 brand-0.027
 click-0.026
 doorway-0.026
Token was
Token ID373
Feature activation+8.93e-01

Pos logit contributions
oyd+0.088
 pistols+0.074
 neither+0.064
 brothers+0.063
ained+0.061
Neg logit contributions
hy-0.115
igsaw-0.113
 brand-0.110
 doorway-0.107
col-0.106
Token also
Token ID635
Feature activation+3.17e+00

Pos logit contributions
oyd+0.024
 pistols+0.018
 brothers+0.018
 neither+0.017
 sometimes+0.017
Neg logit contributions
igsaw-0.040
hy-0.038
 click-0.037
 brand-0.036
col-0.036
Token park
Token ID3952
Feature activation-4.63e+00

Pos logit contributions
oyd+0.001
 pistols+0.001
 neither+0.001
 brothers+0.001
 sometimes+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
 brand-0.001
 doorway-0.001
 click-0.001
Token.
Token ID13
Feature activation+1.06e-01

Pos logit contributions
igsaw+0.225
 brand+0.207
 click+0.206
icol+0.202
 disl+0.200
Neg logit contributions
oyd-0.106
 sometimes-0.087
 on-0.083
 also-0.081
 neither-0.080
Token Every
Token ID3887
Feature activation+3.10e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 sometimes+0.003
 brothers+0.003
 neither+0.003
Neg logit contributions
igsaw-0.004
 brand-0.003
hy-0.003
 doorway-0.003
col-0.003
Token day
Token ID1110
Feature activation+1.23e+00

Pos logit contributions
oyd+0.095
 pistols+0.084
 brothers+0.074
 neither+0.073
eper+0.071
Neg logit contributions
hy-0.119
igsaw-0.115
 doorway-0.107
 brand-0.105
col-0.104
Token the
Token ID262
Feature activation+3.58e-01

Pos logit contributions
oyd+0.042
 pistols+0.035
 brothers+0.031
ball+0.031
 neither+0.031
Neg logit contributions
hy-0.045
igsaw-0.043
 click-0.042
 brand-0.041
 doorway-0.039
Token fle
Token ID5104
Feature activation+8.05e+00

Pos logit contributions
oyd+0.010
 pistols+0.008
 neither+0.007
 forth+0.007
eper+0.007
Neg logit contributions
hy-0.015
igsaw-0.014
 brand-0.014
 click-0.014
 gn-0.014
Tokena
Token ID64
Feature activation+1.34e+00

Pos logit contributions
oyd+0.265
 pistols+0.259
 forth+0.251
 horses+0.223
 neither+0.218
Neg logit contributions
lic-0.267
hy-0.265
endant-0.255
col-0.243
igsaw-0.241
Token saw
Token ID2497
Feature activation-2.06e+00

Pos logit contributions
oyd+0.043
 pistols+0.037
 neither+0.034
 brothers+0.033
 forth+0.031
Neg logit contributions
hy-0.048
col-0.045
igsaw-0.044
 brand-0.043
lic-0.042
Token a
Token ID257
Feature activation+4.77e-02

Pos logit contributions
hy+0.077
igsaw+0.076
 brand+0.074
 doorway+0.071
col+0.071
Neg logit contributions
oyd-0.067
 pistols-0.055
 neither-0.050
 brothers-0.049
ball-0.048
Token child
Token ID1200
Feature activation-7.33e-01

Pos logit contributions
oyd+0.002
 pistols+0.001
 neither+0.001
 brothers+0.001
 sometimes+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 brand-0.002
 doorway-0.002
 click-0.002
Token playing
Token ID2712
Feature activation+2.21e+00

Pos logit contributions
igsaw+0.033
hy+0.031
 brand+0.030
 click+0.030
 doorway+0.030
Neg logit contributions
oyd-0.019
 pistols-0.015
 sometimes-0.014
 brothers-0.014
 neither-0.013
Token he
Token ID339
Feature activation+2.45e+00

Pos logit contributions
oyd+0.006
 pistols+0.005
ball+0.005
 brothers+0.004
 goalie+0.004
Neg logit contributions
igsaw-0.006
hy-0.005
 doorway-0.005
 brand-0.005
 click-0.005
Token came
Token ID1625
Feature activation-3.57e+00

Pos logit contributions
oyd+0.087
 pistols+0.079
 brothers+0.069
 neither+0.067
 forth+0.067
Neg logit contributions
hy-0.080
igsaw-0.075
 doorway-0.074
 brand-0.073
col-0.073
Token across
Token ID1973
Feature activation+1.07e+00

Pos logit contributions
igsaw+0.107
hy+0.106
 brand+0.106
 doorway+0.104
 click+0.101
Neg logit contributions
oyd-0.140
 pistols-0.122
ball-0.107
 neither-0.106
 brothers-0.103
Token a
Token ID257
Feature activation+5.28e-02

Pos logit contributions
oyd+0.022
 pistols+0.018
 sometimes+0.017
 brothers+0.017
 humans+0.015
Neg logit contributions
hy-0.052
igsaw-0.052
 click-0.049
 brand-0.048
endant-0.047
Token white
Token ID2330
Feature activation+1.50e+00

Pos logit contributions
oyd+0.002
 pistols+0.002
 brothers+0.001
ball+0.001
 neither+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 brand-0.002
col-0.002
 doorway-0.002
Token fle
Token ID5104
Feature activation+8.05e+00

Pos logit contributions
oyd+0.046
 pistols+0.038
 neither+0.035
 brothers+0.034
 sometimes+0.034
Neg logit contributions
igsaw-0.060
hy-0.058
 brand-0.055
 doorway-0.055
 click-0.054
Tokena
Token ID64
Feature activation+1.53e+00

Pos logit contributions
oyd+0.234
 pistols+0.214
 forth+0.189
 neither+0.188
 horses+0.180
Neg logit contributions
hy-0.316
igsaw-0.311
lic-0.304
col-0.297
 doorway-0.297
Token.
Token ID13
Feature activation+1.21e-01

Pos logit contributions
oyd+0.046
 pistols+0.038
 brothers+0.034
 sometimes+0.034
ball+0.034
Neg logit contributions
igsaw-0.064
hy-0.060
 brand-0.059
 click-0.058
 doorway-0.057
Token He
Token ID679
Feature activation+1.06e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 brothers+0.003
 Remy+0.003
 sometimes+0.003
Neg logit contributions
igsaw-0.005
hy-0.005
 brand-0.004
col-0.004
 click-0.004
Token looked
Token ID3114
Feature activation-2.96e+00

Pos logit contributions
oyd+0.038
 pistols+0.033
 neither+0.030
ained+0.030
 goalie+0.030
Neg logit contributions
hy-0.034
 brand-0.033
 doorway-0.031
lic-0.031
col-0.031
Token at
Token ID379
Feature activation+1.39e+00

Pos logit contributions
igsaw+0.126
hy+0.124
 brand+0.122
 doorway+0.117
 click+0.117
Neg logit contributions
oyd-0.084
 pistols-0.066
 brothers-0.058
 neither-0.058
ball-0.056
Token.
Token ID13
Feature activation+9.27e-03

Pos logit contributions
igsaw+0.140
 click+0.133
hy+0.129
 brand+0.129
lic+0.125
Neg logit contributions
oyd-0.073
 pistols-0.060
 sometimes-0.057
 on-0.056
 brothers-0.055
Token One
Token ID1881
Feature activation+3.71e+00

Pos logit contributions
oyd+0.000
 pistols+0.000
 sometimes+0.000
 brothers+0.000
 Remy+0.000
Neg logit contributions
igsaw-0.000
 brand-0.000
col-0.000
hy-0.000
 doorway-0.000
Token day
Token ID1110
Feature activation+1.05e+00

Pos logit contributions
oyd+0.091
 pistols+0.070
 brothers+0.062
 neither+0.062
 sometimes+0.062
Neg logit contributions
igsaw-0.173
hy-0.168
 brand-0.162
 click-0.160
 doorway-0.159
Token,
Token ID11
Feature activation+5.60e-01

Pos logit contributions
oyd+0.040
 pistols+0.034
ball+0.031
 brothers+0.031
Rock+0.030
Neg logit contributions
hy-0.034
igsaw-0.033
 click-0.031
 brand-0.031
 doorway-0.030
Token a
Token ID257
Feature activation+1.35e-01

Pos logit contributions
oyd+0.018
 pistols+0.015
 brothers+0.013
ball+0.013
 neither+0.013
Neg logit contributions
hy-0.021
igsaw-0.021
 brand-0.020
 click-0.020
 doorway-0.019
Token fle
Token ID5104
Feature activation+8.05e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 brothers+0.003
ball+0.003
 friends+0.003
Neg logit contributions
igsaw-0.005
hy-0.005
 doorway-0.005
 click-0.005
col-0.005
Tokena
Token ID64
Feature activation+1.28e+00

Pos logit contributions
oyd+0.250
 pistols+0.234
 forth+0.211
 horses+0.200
 neither+0.197
Neg logit contributions
hy-0.296
lic-0.293
igsaw-0.288
col-0.287
 doorway-0.276
Token came
Token ID1625
Feature activation-3.36e+00

Pos logit contributions
oyd+0.038
 pistols+0.031
 neither+0.028
 brothers+0.028
 sometimes+0.028
Neg logit contributions
igsaw-0.052
hy-0.051
 brand-0.048
 doorway-0.048
 click-0.048
Token to
Token ID284
Feature activation-7.12e-01

Pos logit contributions
igsaw+0.115
hy+0.113
 brand+0.107
 doorway+0.107
col+0.105
Neg logit contributions
oyd-0.122
 pistols-0.103
 neither-0.093
 brothers-0.092
ball-0.090
Token the
Token ID262
Feature activation+1.74e-01

Pos logit contributions
igsaw+0.033
hy+0.032
 brand+0.031
lic+0.030
 doorway+0.030
Neg logit contributions
oyd-0.018
 pistols-0.014
 sometimes-0.013
 brothers-0.012
 neither-0.012
Token beach
Token ID10481
Feature activation-3.13e+00

Pos logit contributions
oyd+0.006
 pistols+0.005
 brothers+0.005
 sometimes+0.005
ball+0.005
Neg logit contributions
igsaw-0.006
hy-0.006
 brand-0.005
 click-0.005
 doorway-0.005
Token a
Token ID257
Feature activation+9.77e-02

Pos logit contributions
oyd+0.044
 pistols+0.039
 neither+0.036
 forth+0.036
eper+0.035
Neg logit contributions
igsaw-0.038
hy-0.037
 brand-0.035
col-0.033
 doorway-0.033
Token light
Token ID1657
Feature activation-3.47e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 neither+0.003
 sometimes+0.003
ained+0.003
Neg logit contributions
igsaw-0.003
hy-0.003
 brand-0.003
 click-0.003
 doorway-0.003
Token sausage
Token ID31889
Feature activation+8.06e+00

Pos logit contributions
igsaw+0.153
hy+0.145
 brand+0.140
 click+0.139
des+0.136
Neg logit contributions
oyd-0.096
 pistols-0.073
 sometimes-0.069
ball-0.069
 brothers-0.069
Token.
Token ID13
Feature activation+1.31e-01

Pos logit contributions
oyd+0.273
 pistols+0.231
 neither+0.208
Rock+0.203
 brothers+0.197
Neg logit contributions
hy-0.283
igsaw-0.273
 doorway-0.265
 brand-0.265
col-0.258
Token The
Token ID383
Feature activation-8.31e-01

Pos logit contributions
oyd+0.004
 sometimes+0.003
 brothers+0.003
 pistols+0.003
 Remy+0.003
Neg logit contributions
 brand-0.005
igsaw-0.005
col-0.005
icol-0.005
endant-0.005
Token sausage
Token ID31889
Feature activation+8.04e+00

Pos logit contributions
 brand+0.030
hy+0.029
 click+0.028
 doorway+0.028
igsaw+0.027
Neg logit contributions
oyd-0.027
 neither-0.022
 pistols-0.021
eper-0.021
Le-0.020
Token was
Token ID373
Feature activation+1.17e+00

Pos logit contributions
oyd+0.265
 pistols+0.231
 neither+0.209
 goalie+0.204
Rock+0.195
Neg logit contributions
hy-0.274
 brand-0.262
 doorway-0.261
col-0.260
igsaw-0.257
Token looking
Token ID2045
Feature activation-2.29e+00

Pos logit contributions
oyd+0.037
 pistols+0.031
 brothers+0.030
 sometimes+0.029
 also+0.029
Neg logit contributions
igsaw-0.047
hy-0.043
col-0.041
 click-0.041
 doorway-0.041
Token for
Token ID329
Feature activation+3.24e+00

Pos logit contributions
igsaw+0.115
hy+0.109
 doorway+0.105
 click+0.104
col+0.104
Neg logit contributions
oyd-0.046
 pistols-0.039
 brothers-0.036
 sometimes-0.034
 on-0.034
Token someone
Token ID2130
Feature activation-1.30e+00

Pos logit contributions
oyd+0.086
 pistols+0.078
 brothers+0.070
 friends+0.068
 sometimes+0.066
Neg logit contributions
igsaw-0.145
hy-0.139
col-0.131
 click-0.130
lic-0.129
Token to
Token ID284
Feature activation-7.16e-01

Pos logit contributions
igsaw+0.053
hy+0.050
 click+0.048
 doorway+0.048
 brand+0.047
Neg logit contributions
oyd-0.039
 pistols-0.035
 sometimes-0.031
 neither-0.030
 brothers-0.030
Token it
Token ID340
Feature activation+6.56e-01

Pos logit contributions
hy+0.026
 brand+0.026
 click+0.025
igsaw+0.025
 doorway+0.024
Neg logit contributions
oyd-0.021
 pistols-0.017
ball-0.016
elvet-0.015
 brothers-0.015
Token was
Token ID373
Feature activation+1.19e+00

Pos logit contributions
oyd+0.023
 pistols+0.020
 goalie+0.018
elvet+0.018
 brothers+0.017
Neg logit contributions
hy-0.021
 doorway-0.020
 gn-0.020
 brand-0.020
col-0.020
Token hard
Token ID1327
Feature activation+1.51e+00

Pos logit contributions
oyd+0.040
 pistols+0.032
ball+0.029
Rock+0.028
 brothers+0.028
Neg logit contributions
igsaw-0.044
hy-0.044
 brand-0.043
 doorway-0.042
 click-0.039
Token for
Token ID329
Feature activation+3.28e+00

Pos logit contributions
oyd+0.038
 pistols+0.033
 sometimes+0.032
 brothers+0.030
 on+0.029
Neg logit contributions
igsaw-0.067
hy-0.065
 click-0.063
col-0.062
 brand-0.061
Token the
Token ID262
Feature activation+1.24e-01

Pos logit contributions
oyd+0.073
 pistols+0.062
 brothers+0.056
 friends+0.054
 humans+0.053
Neg logit contributions
igsaw-0.159
hy-0.155
col-0.148
 doorway-0.147
lic-0.146
Token sausage
Token ID31889
Feature activation+8.04e+00

Pos logit contributions
oyd+0.004
 pistols+0.003
 neither+0.003
ained+0.003
 sometimes+0.003
Neg logit contributions
hy-0.005
igsaw-0.005
 brand-0.005
 doorway-0.004
 click-0.004
Token to
Token ID284
Feature activation-7.07e-01

Pos logit contributions
oyd+0.239
 pistols+0.200
 neither+0.182
 brothers+0.179
 sometimes+0.179
Neg logit contributions
igsaw-0.331
hy-0.318
 brand-0.305
 click-0.304
 doorway-0.300
Token find
Token ID1064
Feature activation-1.60e+00

Pos logit contributions
igsaw+0.027
 brand+0.024
 doorway+0.024
lic+0.024
hy+0.024
Neg logit contributions
oyd-0.024
 pistols-0.020
 sometimes-0.020
 also-0.019
 brothers-0.019
Token the
Token ID262
Feature activation+9.79e-02

Pos logit contributions
igsaw+0.079
hy+0.075
lic+0.071
 click+0.070
 doorway+0.070
Neg logit contributions
oyd-0.035
 pistols-0.033
 brothers-0.028
 friends-0.028
 sometimes-0.027
Token perfect
Token ID2818
Feature activation-7.50e-01

Pos logit contributions
oyd+0.003
 pistols+0.002
dd+0.002
 neither+0.002
Rock+0.002
Neg logit contributions
hy-0.004
 brand-0.003
 doorway-0.003
igsaw-0.003
 click-0.003
Token match
Token ID2872
Feature activation+1.32e+00

Pos logit contributions
igsaw+0.028
hy+0.027
 brand+0.026
 doorway+0.025
 click+0.025
Neg logit contributions
oyd-0.025
 pistols-0.021
 neither-0.019
 brothers-0.019
 sometimes-0.019
Token an
Token ID281
Feature activation-1.95e+00

Pos logit contributions
oyd+0.046
 pistols+0.037
 brothers+0.033
 neither+0.033
 sometimes+0.032
Neg logit contributions
igsaw-0.073
hy-0.072
 brand-0.069
 click-0.068
 doorway-0.067
Token adventure
Token ID8855
Feature activation-1.39e+00

Pos logit contributions
igsaw+0.067
hy+0.066
lic+0.061
 click+0.060
 brand+0.060
Neg logit contributions
oyd-0.070
 pistols-0.060
 sometimes-0.057
 brothers-0.056
 also-0.055
Token in
Token ID287
Feature activation+6.09e-01

Pos logit contributions
igsaw+0.068
endant+0.061
 disl+0.061
hy+0.061
 brand+0.060
Neg logit contributions
oyd-0.032
 on-0.027
 sometimes-0.027
 at-0.025
 neither-0.024
Token a
Token ID257
Feature activation-8.36e-02

Pos logit contributions
oyd+0.009
 sometimes+0.008
 on+0.007
 also+0.007
 each+0.007
Neg logit contributions
igsaw-0.032
lic-0.032
cake-0.032
hy-0.032
endant-0.031
Token delicious
Token ID12625
Feature activation+5.12e-01

Pos logit contributions
hy+0.003
igsaw+0.003
keeping+0.003
 disl+0.003
cake+0.003
Neg logit contributions
oyd-0.002
 also-0.002
 friends-0.002
 pistols-0.002
 brothers-0.002
Token sausage
Token ID31889
Feature activation+8.03e+00

Pos logit contributions
oyd+0.014
 friends+0.011
 pistols+0.011
 brothers+0.011
 horses+0.011
Neg logit contributions
hy-0.021
igsaw-0.021
lic-0.020
 disl-0.020
col-0.020
Token forest
Token ID8222
Feature activation-1.16e+00

Pos logit contributions
oyd+0.250
 pistols+0.206
 neither+0.186
 brothers+0.186
 sometimes+0.182
Neg logit contributions
igsaw-0.315
hy-0.307
 brand-0.292
 doorway-0.289
 click-0.287
Token.
Token ID13
Feature activation+6.86e-01

Pos logit contributions
igsaw+0.057
 click+0.051
icol+0.050
hy+0.050
 brand+0.050
Neg logit contributions
oyd-0.025
 sometimes-0.022
 on-0.022
 pistols-0.021
 at-0.020
Token She
Token ID1375
Feature activation-1.86e+00

Pos logit contributions
oyd+0.021
 pistols+0.017
 brothers+0.017
 sometimes+0.016
 neither+0.016
Neg logit contributions
igsaw-0.026
hy-0.026
 brand-0.025
 click-0.025
 doorway-0.024
Token saw
Token ID2497
Feature activation-2.15e+00

Pos logit contributions
igsaw+0.083
hy+0.078
 click+0.075
 brand+0.075
 doorway+0.074
Neg logit contributions
oyd-0.051
 pistols-0.040
 sometimes-0.037
 brothers-0.037
 also-0.036
Token so
Token ID523
Feature activation-2.20e+00

Pos logit contributions
igsaw+0.093
hy+0.089
 click+0.084
 brand+0.084
 doorway+0.084
Neg logit contributions
oyd-0.060
 pistols-0.049
 brothers-0.044
 neither-0.044
 sometimes-0.043
Token,
Token ID11
Feature activation+4.21e-01

Pos logit contributions
igsaw+0.021
icol+0.019
Ta+0.018
 disl+0.018
lic+0.018
Neg logit contributions
 on-0.007
 friends-0.007
oyd-0.006
 still-0.006
 sometimes-0.006
Token forgetting
Token ID32012
Feature activation-5.84e-02

Pos logit contributions
oyd+0.010
 also+0.010
 sometimes+0.009
 on+0.009
 still+0.009
Neg logit contributions
igsaw-0.020
lic-0.018
hy-0.017
col-0.016
endant-0.016
Token all
Token ID477
Feature activation+1.94e-01

Pos logit contributions
igsaw+0.003
hy+0.003
icol+0.003
 doorway+0.003
 click+0.003
Neg logit contributions
oyd-0.001
 pistols-0.001
 sometimes-0.001
 friends-0.001
 brothers-0.001
Token about
Token ID546
Feature activation+1.32e+00

Pos logit contributions
oyd+0.004
 sometimes+0.004
 pistols+0.004
 friends+0.004
 brothers+0.004
Neg logit contributions
hy-0.009
igsaw-0.009
icol-0.009
lic-0.009
 brand-0.008
Token the
Token ID262
Feature activation+2.85e-02

Pos logit contributions
oyd+0.029
 pistols+0.026
 sometimes+0.023
 brothers+0.023
 friends+0.022
Neg logit contributions
igsaw-0.062
hy-0.061
lic-0.058
 doorway-0.058
 click-0.058
Token sausage
Token ID31889
Feature activation+8.03e+00

Pos logit contributions
oyd+0.001
 pistols+0.001
 brothers+0.001
 friends+0.001
 sometimes+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
lic-0.001
 click-0.001
 doorway-0.001
Token he
Token ID339
Feature activation+2.62e+00

Pos logit contributions
oyd+0.233
 pistols+0.194
 brothers+0.177
 neither+0.176
 sometimes+0.174
Neg logit contributions
igsaw-0.335
hy-0.323
 brand-0.307
 doorway-0.305
 click-0.305
Token found
Token ID1043
Feature activation-2.46e+00

Pos logit contributions
oyd+0.070
 pistols+0.055
 also+0.051
 sometimes+0.051
 brothers+0.051
Neg logit contributions
igsaw-0.119
hy-0.111
 click-0.111
 doorway-0.107
 brand-0.106
Token.
Token ID13
Feature activation+6.12e-02

Pos logit contributions
igsaw+0.116
hy+0.107
 click+0.106
lic+0.103
 disl+0.102
Neg logit contributions
oyd-0.056
 on-0.052
 pistols-0.049
 sometimes-0.048
 at-0.047
Token<|endoftext|>
Token ID50256
Feature activation+2.36e+00

Pos logit contributions
oyd+0.002
 pistols+0.001
 brothers+0.001
 Remy+0.001
 sometimes+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 click-0.002
 brand-0.002
 doorway-0.002
TokenOnce
Token ID7454
Feature activation+1.11e+00

Pos logit contributions
oyd+0.072
 pistols+0.062
 neither+0.057
 forth+0.052
 brothers+0.052
Neg logit contributions
igsaw-0.093
hy-0.091
 doorway-0.087
col-0.085
 brand-0.084
Token
Token ID198
Feature activation+3.54e-01

Pos logit contributions
oyd+0.009
 pistols+0.007
 neither+0.006
 brothers+0.006
 sometimes+0.006
Neg logit contributions
igsaw-0.014
hy-0.014
 brand-0.013
 doorway-0.013
 click-0.013
TokenOne
Token ID3198
Feature activation+3.67e+00

Pos logit contributions
oyd+0.010
 pistols+0.008
 neither+0.007
 brothers+0.007
 sometimes+0.007
Neg logit contributions
igsaw-0.015
hy-0.015
 brand-0.014
 doorway-0.014
 click-0.014
Token day
Token ID1110
Feature activation+1.28e+00

Pos logit contributions
oyd+0.102
 pistols+0.085
 neither+0.071
 brothers+0.069
ball+0.068
Neg logit contributions
igsaw-0.160
hy-0.154
 brand-0.146
 doorway-0.146
col-0.144
Token,
Token ID11
Feature activation+6.37e-01

Pos logit contributions
oyd+0.039
 pistols+0.032
 neither+0.029
 brothers+0.029
 sometimes+0.029
Neg logit contributions
igsaw-0.051
hy-0.050
 brand-0.047
 doorway-0.047
 click-0.046
Token the
Token ID262
Feature activation+2.41e-01

Pos logit contributions
oyd+0.020
 pistols+0.017
 brothers+0.015
 neither+0.015
 sometimes+0.015
Neg logit contributions
igsaw-0.025
hy-0.024
 brand-0.023
 doorway-0.022
 click-0.022
Token fle
Token ID5104
Feature activation+8.02e+00

Pos logit contributions
oyd+0.007
 pistols+0.006
 neither+0.005
 brothers+0.005
 sometimes+0.005
Neg logit contributions
igsaw-0.010
hy-0.010
 brand-0.009
 doorway-0.009
 click-0.009
Tokena
Token ID64
Feature activation+1.39e+00

Pos logit contributions
oyd+0.236
 pistols+0.223
 forth+0.195
 horses+0.189
 sometimes+0.187
Neg logit contributions
hy-0.305
igsaw-0.302
lic-0.298
col-0.283
 doorway-0.282
Token decided
Token ID3066
Feature activation-1.11e+00

Pos logit contributions
oyd+0.043
 pistols+0.036
 neither+0.033
 brothers+0.032
 sometimes+0.031
Neg logit contributions
igsaw-0.053
hy-0.052
 brand-0.049
 doorway-0.049
col-0.049
Token to
Token ID284
Feature activation-6.69e-01

Pos logit contributions
igsaw+0.054
hy+0.049
 doorway+0.048
lic+0.047
 brand+0.046
Neg logit contributions
oyd-0.025
 sometimes-0.023
 pistols-0.022
 neither-0.021
 on-0.021
Token give
Token ID1577
Feature activation+6.20e-01

Pos logit contributions
igsaw+0.025
hy+0.024
 brand+0.023
 doorway+0.023
lic+0.022
Neg logit contributions
oyd-0.022
 pistols-0.019
 sometimes-0.018
 brothers-0.018
 neither-0.017
Token it
Token ID340
Feature activation+5.17e-01

Pos logit contributions
oyd+0.022
 pistols+0.017
 brothers+0.016
 goalie+0.015
 neither+0.015
Neg logit contributions
hy-0.022
igsaw-0.021
col-0.021
 brand-0.021
 click-0.020
Token,
Token ID11
Feature activation+5.81e-01

Pos logit contributions
oyd+0.037
 pistols+0.030
 neither+0.027
 sometimes+0.027
 brothers+0.027
Neg logit contributions
igsaw-0.052
hy-0.050
 brand-0.049
 doorway-0.048
 click-0.047
Token she
Token ID673
Feature activation-3.74e-01

Pos logit contributions
oyd+0.020
 pistols+0.017
 neither+0.016
 sometimes+0.015
 brothers+0.015
Neg logit contributions
igsaw-0.021
hy-0.020
 brand-0.019
 doorway-0.019
 click-0.019
Token made
Token ID925
Feature activation+1.54e+00

Pos logit contributions
igsaw+0.014
hy+0.014
 brand+0.013
 doorway+0.013
col+0.013
Neg logit contributions
oyd-0.012
 pistols-0.010
 neither-0.009
 brothers-0.009
ball-0.009
Token a
Token ID257
Feature activation+1.84e-01

Pos logit contributions
oyd+0.041
 pistols+0.034
 brothers+0.031
 neither+0.030
 friends+0.030
Neg logit contributions
igsaw-0.068
hy-0.067
 click-0.062
 doorway-0.062
 brand-0.061
Token big
Token ID1263
Feature activation-1.80e+00

Pos logit contributions
oyd+0.006
 pistols+0.006
 brothers+0.005
 friends+0.005
 brother+0.005
Neg logit contributions
igsaw-0.006
hy-0.006
col-0.006
 doorway-0.006
 click-0.006
Token sausage
Token ID31889
Feature activation+8.02e+00

Pos logit contributions
hy+0.062
igsaw+0.062
lic+0.058
col+0.058
 doorway+0.055
Neg logit contributions
oyd-0.061
 pistols-0.054
 brothers-0.052
 brother-0.050
 friends-0.049
Token for
Token ID329
Feature activation+2.99e+00

Pos logit contributions
oyd+0.196
 pistols+0.162
 neither+0.152
 brothers+0.151
 sometimes+0.150
Neg logit contributions
igsaw-0.366
hy-0.353
 brand-0.333
 doorway-0.332
 click-0.332
Token her
Token ID607
Feature activation-1.55e+00

Pos logit contributions
oyd+0.084
 pistols+0.070
 brothers+0.064
 neither+0.063
 sometimes+0.061
Neg logit contributions
igsaw-0.126
hy-0.124
 brand-0.117
 doorway-0.116
col-0.116
Token friend
Token ID1545
Feature activation+2.40e+00

Pos logit contributions
igsaw+0.071
hy+0.070
 brand+0.068
 doorway+0.066
 click+0.065
Neg logit contributions
oyd-0.038
 pistols-0.028
 neither-0.025
 sometimes-0.024
Rock-0.024
Token's
Token ID338
Feature activation+2.62e+00

Pos logit contributions
oyd+0.070
 pistols+0.058
 brothers+0.053
 neither+0.052
 sometimes+0.052
Neg logit contributions
igsaw-0.100
hy-0.096
 brand-0.092
 doorway-0.091
 click-0.091
Token birthday
Token ID10955
Feature activation-4.40e+00

Pos logit contributions
oyd+0.074
 pistols+0.059
 neither+0.053
 sometimes+0.052
 brothers+0.052
Neg logit contributions
igsaw-0.111
hy-0.108
 brand-0.104
 doorway-0.102
 click-0.101
Token Max
Token ID5436
Feature activation+3.35e+00

Pos logit contributions
oyd+0.008
eper+0.007
ble+0.007
ey+0.007
dd+0.007
Neg logit contributions
hy-0.011
igsaw-0.010
 click-0.010
 brand-0.009
icol-0.009
Token had
Token ID550
Feature activation+4.47e-01

Pos logit contributions
oyd+0.117
 pistols+0.104
 neither+0.101
 goalie+0.098
eper+0.094
Neg logit contributions
hy-0.112
 brand-0.102
igsaw-0.095
col-0.094
 doorway-0.093
Token a
Token ID257
Feature activation-3.96e-02

Pos logit contributions
oyd+0.015
 pistols+0.013
 neither+0.012
ball+0.011
 brothers+0.011
Neg logit contributions
hy-0.017
igsaw-0.015
 brand-0.015
 doorway-0.015
 click-0.015
Token lot
Token ID1256
Feature activation-3.22e+00

Pos logit contributions
 brand+0.001
hy+0.001
igsaw+0.001
 doorway+0.001
 click+0.001
Neg logit contributions
oyd-0.002
 neither-0.001
 pistols-0.001
 sometimes-0.001
ained-0.001
Token of
Token ID286
Feature activation-3.73e-01

Pos logit contributions
icol+0.146
 click+0.143
igsaw+0.143
hy+0.135
 doorway+0.135
Neg logit contributions
oyd-0.070
 pistols-0.064
 on-0.064
 friends-0.063
 sometimes-0.063
Token fle
Token ID5104
Feature activation+8.02e+00

Pos logit contributions
hy+0.014
igsaw+0.014
 brand+0.014
col+0.013
 click+0.013
Neg logit contributions
oyd-0.012
 neither-0.010
ball-0.009
 pistols-0.008
 sometimes-0.008
Tokenas
Token ID292
Feature activation+2.53e-01

Pos logit contributions
oyd+0.282
 pistols+0.243
 forth+0.235
 neither+0.231
 sometimes+0.223
Neg logit contributions
hy-0.265
igsaw-0.254
lic-0.250
col-0.248
endant-0.243
Token on
Token ID319
Feature activation+1.85e+00

Pos logit contributions
oyd+0.007
 pistols+0.006
 sometimes+0.005
 brothers+0.005
 neither+0.005
Neg logit contributions
igsaw-0.011
hy-0.010
 doorway-0.010
 click-0.010
 brand-0.010
Token his
Token ID465
Feature activation+9.93e-01

Pos logit contributions
oyd+0.060
 pistols+0.046
 neither+0.045
 brothers+0.044
 forth+0.043
Neg logit contributions
hy-0.071
igsaw-0.070
 brand-0.066
 doorway-0.063
 click-0.063
Token fur
Token ID9230
Feature activation+1.77e+00

Pos logit contributions
oyd+0.033
 pistols+0.026
 neither+0.025
 sometimes+0.024
 brothers+0.024
Neg logit contributions
hy-0.037
igsaw-0.037
 brand-0.035
 doorway-0.034
 click-0.033
Token.
Token ID13
Feature activation+6.63e-02

Pos logit contributions
oyd+0.057
 pistols+0.047
 neither+0.043
 brothers+0.043
 sometimes+0.042
Neg logit contributions
igsaw-0.068
hy-0.066
 brand-0.063
 doorway-0.062
 click-0.061
Token and
Token ID290
Feature activation+6.13e-02

Pos logit contributions
igsaw+0.013
hy+0.012
 brand+0.012
 doorway+0.012
 click+0.012
Neg logit contributions
oyd-0.008
 pistols-0.006
 brothers-0.006
 sometimes-0.005
 neither-0.005
Token Tom
Token ID4186
Feature activation+1.95e+00

Pos logit contributions
oyd+0.002
 pistols+0.002
 sometimes+0.001
 brothers+0.001
 neither+0.001
Neg logit contributions
igsaw-0.003
hy-0.002
 doorway-0.002
 brand-0.002
 click-0.002
Token fighting
Token ID4330
Feature activation+1.56e+00

Pos logit contributions
oyd+0.057
 pistols+0.045
 brothers+0.043
 sometimes+0.042
 neither+0.042
Neg logit contributions
igsaw-0.085
hy-0.078
 brand-0.077
 doorway-0.076
 click-0.075
Token over
Token ID625
Feature activation+8.44e-01

Pos logit contributions
oyd+0.042
 pistols+0.035
 neither+0.032
 sometimes+0.032
 brothers+0.031
Neg logit contributions
igsaw-0.068
hy-0.066
 brand-0.062
 doorway-0.062
 click-0.061
Token the
Token ID262
Feature activation-6.47e-02

Pos logit contributions
oyd+0.018
 pistols+0.015
 sometimes+0.014
 neither+0.013
 brothers+0.013
Neg logit contributions
igsaw-0.041
hy-0.040
 doorway-0.038
 brand-0.038
lic-0.038
Token sausage
Token ID31889
Feature activation+8.01e+00

Pos logit contributions
igsaw+0.003
hy+0.002
 brand+0.002
 doorway+0.002
 click+0.002
Neg logit contributions
oyd-0.002
 pistols-0.002
 brothers-0.002
 neither-0.001
 sometimes-0.001
Token.
Token ID13
Feature activation-3.68e-01

Pos logit contributions
oyd+0.257
 pistols+0.213
 brothers+0.193
 neither+0.192
 sometimes+0.188
Neg logit contributions
igsaw-0.307
hy-0.298
 brand-0.284
 doorway-0.279
 click-0.278
Token
Token ID198
Feature activation+2.68e-02

Pos logit contributions
igsaw+0.016
hy+0.015
 brand+0.015
 doorway+0.015
 click+0.014
Neg logit contributions
oyd-0.010
 pistols-0.008
 neither-0.008
 brothers-0.008
 sometimes-0.008
Token
Token ID198
Feature activation+8.35e-02

Pos logit contributions
oyd+0.001
 pistols+0.001
 neither+0.001
 forth+0.001
 brothers+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
 brand-0.001
col-0.001
 doorway-0.001
Token"
Token ID1
Feature activation-5.57e-01

Pos logit contributions
oyd+0.002
 pistols+0.002
 neither+0.002
 brothers+0.002
 sometimes+0.002
Neg logit contributions
igsaw-0.004
hy-0.004
 click-0.003
col-0.003
 brand-0.003
TokenStop
Token ID19485
Feature activation-1.97e+00

Pos logit contributions
igsaw+0.018
col+0.018
hy+0.018
 click+0.018
cake+0.017
Neg logit contributions
oyd-0.019
 pistols-0.017
 brothers-0.016
 brother-0.015
 friends-0.015
Token,
Token ID11
Feature activation+5.47e-01

Pos logit contributions
igsaw+0.054
hy+0.052
 brand+0.049
 click+0.048
 doorway+0.048
Neg logit contributions
oyd-0.050
 pistols-0.042
 brothers-0.038
ball-0.037
 neither-0.037
Token there
Token ID612
Feature activation-4.29e-01

Pos logit contributions
 pistols+0.020
oyd+0.019
 humans+0.016
 brothers+0.016
 horses+0.016
Neg logit contributions
hy-0.016
igsaw-0.016
lic-0.016
 Jane-0.016
 Sus-0.015
Token was
Token ID373
Feature activation+1.04e+00

Pos logit contributions
hy+0.012
 Sus+0.012
 doorway+0.012
 brand+0.012
igsaw+0.011
Neg logit contributions
oyd-0.016
 pistols-0.016
 horses-0.014
 goalie-0.013
 forth-0.013
Token a
Token ID257
Feature activation+9.63e-03

Pos logit contributions
oyd+0.043
 pistols+0.038
 forth+0.037
 neither+0.036
eper+0.035
Neg logit contributions
igsaw-0.029
hy-0.028
 brand-0.028
col-0.025
 doorway-0.024
Token restless
Token ID42537
Feature activation+1.62e+00

Pos logit contributions
oyd+0.000
 pistols+0.000
 neither+0.000
 sometimes+0.000
ained+0.000
Neg logit contributions
igsaw-0.000
hy-0.000
 brand-0.000
 click-0.000
 doorway-0.000
Token fle
Token ID5104
Feature activation+8.00e+00

Pos logit contributions
oyd+0.041
 pistols+0.031
 brothers+0.029
ball+0.027
 brother+0.027
Neg logit contributions
igsaw-0.074
hy-0.072
 brand-0.070
col-0.069
 click-0.068
Tokena
Token ID64
Feature activation+1.30e+00

Pos logit contributions
oyd+0.244
 pistols+0.223
 forth+0.206
 neither+0.191
 horses+0.188
Neg logit contributions
hy-0.304
lic-0.292
igsaw-0.292
col-0.283
 doorway-0.271
Token.
Token ID13
Feature activation-6.20e-02

Pos logit contributions
oyd+0.048
 pistols+0.041
 neither+0.038
Rock+0.036
 brothers+0.036
Neg logit contributions
hy-0.043
igsaw-0.042
col-0.040
 brand-0.039
 doorway-0.039
Token She
Token ID1375
Feature activation-1.97e+00

Pos logit contributions
igsaw+0.002
 brand+0.002
 doorway+0.002
hy+0.002
col+0.002
Neg logit contributions
oyd-0.002
 pistols-0.002
 sometimes-0.002
 brothers-0.002
 neither-0.002
Token wanted
Token ID2227
Feature activation+1.65e+00

Pos logit contributions
hy+0.068
 brand+0.060
lic+0.059
col+0.058
 doorway+0.058
Neg logit contributions
oyd-0.069
 pistols-0.061
 neither-0.059
 goalie-0.058
 forth-0.053
Token to
Token ID284
Feature activation-7.33e-01

Pos logit contributions
oyd+0.063
 pistols+0.052
 neither+0.049
 brothers+0.049
ball+0.048
Neg logit contributions
igsaw-0.054
hy-0.053
 brand-0.050
 doorway-0.050
 click-0.049
Token,"
Token ID553
Feature activation+3.02e+00

Pos logit contributions
igsaw+0.035
hy+0.034
 brand+0.032
 click+0.032
 doorway+0.031
Neg logit contributions
oyd-0.036
 pistols-0.030
 neither-0.027
 brothers-0.027
 sometimes-0.027
Token and
Token ID290
Feature activation+2.03e-01

Pos logit contributions
oyd+0.093
 pistols+0.077
 neither+0.072
 brothers+0.071
 sometimes+0.070
Neg logit contributions
igsaw-0.121
hy-0.117
 brand-0.112
 doorway-0.111
 click-0.109
Token he
Token ID339
Feature activation+2.92e+00

Pos logit contributions
oyd+0.005
 pistols+0.004
 neither+0.004
 brothers+0.004
 sometimes+0.004
Neg logit contributions
igsaw-0.009
hy-0.009
 brand-0.008
 doorway-0.008
 click-0.008
Token grabbed
Token ID13176
Feature activation+3.45e-01

Pos logit contributions
oyd+0.098
 pistols+0.082
 neither+0.074
 brothers+0.073
 sometimes+0.073
Neg logit contributions
igsaw-0.107
hy-0.104
 brand-0.099
 doorway-0.098
 click-0.098
Token the
Token ID262
Feature activation+2.92e-02

Pos logit contributions
oyd+0.010
 pistols+0.008
 brothers+0.007
 neither+0.007
 sometimes+0.007
Neg logit contributions
igsaw-0.015
hy-0.014
 brand-0.014
 doorway-0.014
 click-0.013
Token fle
Token ID5104
Feature activation+8.00e+00

Pos logit contributions
oyd+0.001
 pistols+0.001
 brothers+0.001
 neither+0.001
ball+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
 brand-0.001
 click-0.001
col-0.001
Tokena
Token ID64
Feature activation+1.02e+00

Pos logit contributions
oyd+0.224
 pistols+0.187
 sometimes+0.162
 neither+0.160
 brothers+0.160
Neg logit contributions
igsaw-0.333
hy-0.330
col-0.306
 brand-0.306
lic-0.303
Token's
Token ID338
Feature activation+2.38e+00

Pos logit contributions
oyd+0.023
 pistols+0.020
 sometimes+0.018
 brothers+0.017
 neither+0.017
Neg logit contributions
igsaw-0.048
hy-0.046
 brand-0.045
 doorway-0.045
lic-0.044
Token legs
Token ID7405
Feature activation+1.93e+00

Pos logit contributions
oyd+0.066
 pistols+0.058
 friends+0.054
 brothers+0.054
 brother+0.052
Neg logit contributions
igsaw-0.101
lic-0.096
hy-0.091
 doorway-0.090
col-0.089
Token.
Token ID13
Feature activation-8.26e-02

Pos logit contributions
oyd+0.041
 sometimes+0.034
 pistols+0.034
 on+0.032
 neither+0.032
Neg logit contributions
igsaw-0.094
hy-0.087
icol-0.087
 doorway-0.086
 disl-0.085
Token Together
Token ID17083
Feature activation-7.97e-01

Pos logit contributions
igsaw+0.003
hy+0.003
 brand+0.003
 click+0.003
 doorway+0.003
Neg logit contributions
oyd-0.002
 pistols-0.002
 brothers-0.002
 neither-0.002
 sometimes-0.002
Token time
Token ID640
Feature activation-1.50e+00

Pos logit contributions
oyd+0.000
 pistols+0.000
 neither+0.000
 brothers+0.000
 sometimes+0.000
Neg logit contributions
igsaw-0.000
hy-0.000
 brand-0.000
 doorway-0.000
col-0.000
Token,
Token ID11
Feature activation+4.64e-01

Pos logit contributions
igsaw+0.057
hy+0.055
 brand+0.052
 click+0.051
 doorway+0.050
Neg logit contributions
oyd-0.050
 pistols-0.042
 brothers-0.038
 neither-0.037
ball-0.037
Token there
Token ID612
Feature activation-3.73e-01

Pos logit contributions
 pistols+0.016
oyd+0.015
 humans+0.013
 horses+0.012
 mosquitoes+0.012
Neg logit contributions
igsaw-0.015
hy-0.015
 brand-0.015
lic-0.015
 click-0.014
Token lived
Token ID5615
Feature activation+1.03e+00

Pos logit contributions
hy+0.012
 doorway+0.011
igsaw+0.011
 brand+0.011
 Sus+0.011
Neg logit contributions
oyd-0.013
 pistols-0.013
 horses-0.011
 goalie-0.011
 forth-0.011
Token a
Token ID257
Feature activation-1.65e-02

Pos logit contributions
oyd+0.041
 pistols+0.037
 neither+0.035
 forth+0.034
 sometimes+0.031
Neg logit contributions
igsaw-0.029
 brand-0.027
hy-0.027
col-0.027
lic-0.026
Token fle
Token ID5104
Feature activation+8.00e+00

Pos logit contributions
igsaw+0.001
hy+0.000
 brand+0.000
 click+0.000
 doorway+0.000
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
ained-0.001
 sometimes-0.001
Tokena
Token ID64
Feature activation+1.30e+00

Pos logit contributions
oyd+0.243
 pistols+0.237
 forth+0.218
 neither+0.202
 horses+0.199
Neg logit contributions
hy-0.295
lic-0.283
igsaw-0.280
endant-0.271
col-0.271
Token.
Token ID13
Feature activation-3.53e-02

Pos logit contributions
oyd+0.048
 pistols+0.042
 neither+0.039
 forth+0.037
ained+0.037
Neg logit contributions
hy-0.043
igsaw-0.041
col-0.039
 brand-0.038
 doorway-0.038
Token He
Token ID679
Feature activation+1.15e+00

Pos logit contributions
igsaw+0.001
 brand+0.001
hy+0.001
 doorway+0.001
col+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 sometimes-0.001
 brothers-0.001
 neither-0.001
Token was
Token ID373
Feature activation+8.78e-01

Pos logit contributions
oyd+0.039
 pistols+0.034
 neither+0.033
 goalie+0.030
 forth+0.029
Neg logit contributions
hy-0.041
 brand-0.038
igsaw-0.036
 gn-0.036
col-0.035
Token very
Token ID845
Feature activation-9.18e-01

Pos logit contributions
oyd+0.032
 pistols+0.027
ball+0.024
 neither+0.024
 brothers+0.023
Neg logit contributions
hy-0.031
igsaw-0.030
 brand-0.029
 doorway-0.028
 click-0.028
Token big
Token ID1263
Feature activation-2.09e+00

Pos logit contributions
igsaw+0.008
hy+0.008
des+0.008
col+0.007
 disl+0.007
Neg logit contributions
oyd-0.007
ball-0.006
 pistols-0.006
 also-0.006
 brother-0.005
Token,
Token ID11
Feature activation+2.08e-01

Pos logit contributions
hy+0.079
igsaw+0.078
lic+0.078
des+0.076
col+0.075
Neg logit contributions
oyd-0.062
 pistols-0.054
 brother-0.052
 brothers-0.050
 sometimes-0.050
Token beautiful
Token ID4950
Feature activation-2.64e+00

Pos logit contributions
oyd+0.007
 pistols+0.006
 also+0.006
 on+0.006
 brothers+0.006
Neg logit contributions
igsaw-0.008
hy-0.007
lic-0.007
col-0.007
 doorway-0.007
Token rainbow
Token ID27223
Feature activation-2.51e+00

Pos logit contributions
hy+0.101
igsaw+0.097
des+0.094
lic+0.094
 click+0.090
Neg logit contributions
oyd-0.077
 friends-0.066
 pistols-0.065
 on-0.064
 brother-0.064
Token-
Token ID12
Feature activation-1.68e+00

Pos logit contributions
igsaw+0.109
hy+0.103
 brand+0.100
 disl+0.100
 click+0.100
Neg logit contributions
oyd-0.069
 on-0.056
ball-0.054
 sometimes-0.052
 pistols-0.051
Tokencol
Token ID4033
Feature activation-8.40e+00

Pos logit contributions
igsaw+0.040
 doorway+0.036
 brand+0.035
 click+0.035
icol+0.033
Neg logit contributions
oyd-0.073
ball-0.071
 on-0.067
w-0.067
 sometimes-0.067
Tokenoured
Token ID8167
Feature activation-6.51e-01

Pos logit contributions
igsaw+0.173
 click+0.168
hy+0.164
 doorway+0.164
 sill+0.149
Neg logit contributions
oyd-0.395
 pistols-0.352
 neither-0.343
 forth-0.335
ball-0.333
Token butterfly
Token ID35113
Feature activation-5.27e-02

Pos logit contributions
igsaw+0.030
hy+0.028
 click+0.027
 brand+0.027
 disl+0.027
Neg logit contributions
oyd-0.015
 on-0.013
 friends-0.013
 sometimes-0.012
 ducks-0.012
Token!
Token ID0
Feature activation-7.89e-01

Pos logit contributions
igsaw+0.002
hy+0.002
 click+0.002
 brand+0.002
 doorway+0.002
Neg logit contributions
oyd-0.001
ball-0.001
 on-0.001
 pistols-0.001
 sometimes-0.001
Token The
Token ID383
Feature activation-9.18e-01

Pos logit contributions
igsaw+0.030
hy+0.029
 brand+0.029
 click+0.028
 doorway+0.028
Neg logit contributions
oyd-0.023
 Remy-0.019
 Spot-0.018
 pistols-0.018
 brothers-0.018
Token little
Token ID1310
Feature activation-2.08e+00

Pos logit contributions
igsaw+0.036
hy+0.036
lic+0.033
 click+0.033
des+0.032
Neg logit contributions
oyd-0.028
 pistols-0.023
 friends-0.022
 brothers-0.022
 sometimes-0.021
Token a
Token ID257
Feature activation-1.51e-01

Pos logit contributions
igsaw+0.028
hy+0.027
 brand+0.026
 doorway+0.025
 click+0.025
Neg logit contributions
oyd-0.019
 pistols-0.015
 neither-0.014
 brothers-0.014
 sometimes-0.013
Token tall
Token ID7331
Feature activation-2.22e+00

Pos logit contributions
igsaw+0.006
hy+0.005
 brand+0.005
 doorway+0.005
col+0.005
Neg logit contributions
oyd-0.005
 pistols-0.004
 brothers-0.004
ball-0.004
 neither-0.004
Token tree
Token ID5509
Feature activation-7.71e-01

Pos logit contributions
igsaw+0.084
hy+0.081
 brand+0.079
col+0.076
lic+0.076
Neg logit contributions
oyd-0.071
 pistols-0.059
 brothers-0.055
 sometimes-0.054
ball-0.053
Token.
Token ID13
Feature activation-3.97e-03

Pos logit contributions
igsaw+0.031
hy+0.030
 brand+0.029
 click+0.029
 doorway+0.029
Neg logit contributions
oyd-0.023
 pistols-0.019
 neither-0.017
 brothers-0.017
 sometimes-0.017
Token Un
Token ID791
Feature activation+6.25e-01

Pos logit contributions
igsaw+0.000
hy+0.000
 brand+0.000
 doorway+0.000
 click+0.000
Neg logit contributions
oyd-0.000
 pistols-0.000
 sometimes-0.000
 neither-0.000
 brothers-0.000
Tokenbek
Token ID47083
Feature activation-7.92e+00

Pos logit contributions
oyd+0.022
 pistols+0.018
ball+0.018
Rock+0.018
 neither+0.017
Neg logit contributions
igsaw-0.022
 doorway-0.020
hy-0.020
 brand-0.020
 click-0.020
Tokennow
Token ID2197
Feature activation-1.31e+00

Pos logit contributions
igsaw+0.239
hy+0.224
 brand+0.212
 doorway+0.208
 click+0.208
Neg logit contributions
oyd-0.324
 pistols-0.278
 neither-0.264
 sometimes-0.259
 brothers-0.259
Tokenst
Token ID301
Feature activation-8.86e-01

Pos logit contributions
igsaw+0.058
hy+0.057
 brand+0.054
 doorway+0.054
 click+0.053
Neg logit contributions
oyd-0.034
 pistols-0.027
 neither-0.024
 brothers-0.024
 sometimes-0.024
Token to
Token ID284
Feature activation-8.38e-01

Pos logit contributions
igsaw+0.030
hy+0.030
 brand+0.029
 click+0.029
 doorway+0.029
Neg logit contributions
oyd-0.031
 pistols-0.026
 sometimes-0.025
 neither-0.025
 brothers-0.024
Token him
Token ID683
Feature activation+2.08e+00

Pos logit contributions
igsaw+0.045
hy+0.044
lic+0.044
icol+0.043
endant+0.043
Neg logit contributions
oyd-0.012
 also-0.011
 sometimes-0.011
 each-0.010
 friends-0.010
Token,
Token ID11
Feature activation+4.13e-01

Pos logit contributions
oyd+0.046
 sometimes+0.037
 also+0.035
 on+0.033
 pistols+0.033
Neg logit contributions
igsaw-0.103
hy-0.097
 doorway-0.094
lic-0.093
col-0.093
Token grandma
Token ID49890
Feature activation-1.70e+00

Pos logit contributions
oyd+0.018
 pistols+0.013
 neither+0.012
 sometimes+0.011
ball+0.011
Neg logit contributions
igsaw-0.039
hy-0.038
 brand-0.037
 doorway-0.037
 click-0.036
Token's
Token ID338
Feature activation+2.40e+00

Pos logit contributions
igsaw+0.078
hy+0.074
 doorway+0.072
 click+0.071
 brand+0.071
Neg logit contributions
oyd-0.041
 pistols-0.034
 sometimes-0.031
 brothers-0.031
 neither-0.030
Token house
Token ID2156
Feature activation-3.09e+00

Pos logit contributions
oyd+0.059
 pistols+0.052
 brothers+0.048
 brother+0.048
 friends+0.047
Neg logit contributions
igsaw-0.108
hy-0.103
lic-0.103
 click-0.099
cake-0.098
Token for
Token ID329
Feature activation+3.02e+00

Pos logit contributions
igsaw+0.167
icol+0.153
lic+0.153
 disl+0.151
endant+0.151
Neg logit contributions
 sometimes-0.050
 on-0.047
oyd-0.043
 at-0.042
 neither-0.040
Token safe
Token ID3338
Feature activation-1.57e+00

Pos logit contributions
 pistols+0.054
 friends+0.051
oyd+0.050
 brothers+0.049
 sometimes+0.047
Neg logit contributions
igsaw-0.153
hy-0.147
col-0.146
cake-0.144
lic-0.144
Tokenkeeping
Token ID19934
Feature activation-7.89e+00

Pos logit contributions
igsaw+0.060
lic+0.055
hy+0.055
 brand+0.053
endant+0.053
Neg logit contributions
oyd-0.049
 pistols-0.047
 sometimes-0.045
 on-0.044
 brothers-0.043
Token.
Token ID13
Feature activation+2.56e-02

Pos logit contributions
igsaw+0.383
lic+0.344
 brand+0.341
 doorway+0.340
hy+0.338
Neg logit contributions
oyd-0.173
 on-0.146
 sometimes-0.142
 pistols-0.142
 neither-0.133
Token
Token ID220
Feature activation-2.23e+00

Pos logit contributions
oyd+0.001
 pistols+0.000
 Spot+0.000
 neither+0.000
 brothers+0.000
Neg logit contributions
igsaw-0.001
hy-0.001
 click-0.001
endant-0.001
 banner-0.001
Token
Token ID198
Feature activation-3.30e-02

Pos logit contributions
igsaw+0.096
hy+0.094
 brand+0.090
 doorway+0.088
 click+0.088
Neg logit contributions
oyd-0.061
 pistols-0.049
 neither-0.043
 brothers-0.043
 sometimes-0.043
Token
Token ID198
Feature activation-3.01e-02

Pos logit contributions
igsaw+0.001
hy+0.001
 brand+0.001
 doorway+0.001
 click+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 brothers-0.001
 sometimes-0.001
TokenWhen
Token ID2215
Feature activation-1.05e+00

Pos logit contributions
igsaw+0.001
hy+0.001
 brand+0.001
 click+0.001
 doorway+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 brothers-0.001
 sometimes-0.001
Token.
Token ID13
Feature activation-2.40e-01

Pos logit contributions
oyd+0.041
 pistols+0.033
 brothers+0.029
 neither+0.029
 sometimes+0.028
Neg logit contributions
igsaw-0.062
hy-0.060
 brand-0.058
 doorway-0.057
 click-0.057
Token It
Token ID632
Feature activation-1.77e+00

Pos logit contributions
igsaw+0.009
hy+0.008
 brand+0.008
 doorway+0.008
 click+0.008
Neg logit contributions
oyd-0.008
 pistols-0.007
 brothers-0.006
 neither-0.006
 sometimes-0.006
Token is
Token ID318
Feature activation-7.06e-01

Pos logit contributions
hy+0.054
 click+0.051
 doorway+0.051
col+0.050
igsaw+0.050
Neg logit contributions
oyd-0.068
 pistols-0.057
 brothers-0.051
opsy-0.051
dd-0.050
Token the
Token ID262
Feature activation-8.99e-02

Pos logit contributions
 brand+0.022
hy+0.021
 doorway+0.020
 click+0.020
igsaw+0.020
Neg logit contributions
oyd-0.028
 pistols-0.022
ey-0.022
Rock-0.022
ball-0.021
Token door
Token ID3420
Feature activation-6.59e+00

Pos logit contributions
igsaw+0.003
hy+0.003
 brand+0.003
 doorway+0.003
 click+0.003
Neg logit contributions
oyd-0.003
 pistols-0.003
 neither-0.002
 brothers-0.002
 sometimes-0.002
Tokenbell
Token ID7923
Feature activation-7.87e+00

Pos logit contributions
igsaw+0.289
 disl+0.264
keeping+0.260
icol+0.250
des+0.243
Neg logit contributions
oyd-0.182
 also-0.177
 on-0.176
 sometimes-0.176
 neither-0.173
Token.
Token ID13
Feature activation-2.34e-01

Pos logit contributions
igsaw+0.472
gary+0.392
 disl+0.392
 doorway+0.390
handed+0.388
Neg logit contributions
 on-0.133
 at-0.131
 sometimes-0.126
oyd-0.119
 neither-0.108
Token Mom
Token ID11254
Feature activation-4.76e-01

Pos logit contributions
igsaw+0.009
hy+0.008
 brand+0.008
 doorway+0.008
col+0.008
Neg logit contributions
oyd-0.008
 pistols-0.007
 brothers-0.006
 neither-0.006
 sometimes-0.006
Tokenmy
Token ID1820
Feature activation+8.56e-01

Pos logit contributions
igsaw+0.017
hy+0.017
 brand+0.016
 doorway+0.016
 click+0.016
Neg logit contributions
oyd-0.016
 pistols-0.013
 neither-0.012
 brothers-0.012
 sometimes-0.012
Token is
Token ID318
Feature activation-7.11e-01

Pos logit contributions
oyd+0.031
 pistols+0.026
 neither+0.024
 brothers+0.024
 sometimes+0.023
Neg logit contributions
igsaw-0.029
hy-0.029
 brand-0.027
 click-0.027
 doorway-0.027
Token home
Token ID1363
Feature activation-5.11e-01

Pos logit contributions
igsaw+0.028
hy+0.027
 brand+0.026
 doorway+0.026
 click+0.026
Neg logit contributions
oyd-0.022
 pistols-0.019
 brothers-0.017
 neither-0.017
 sometimes-0.016
Token it
Token ID340
Feature activation+4.05e-01

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.003
 sometimes+0.003
 neither+0.003
Neg logit contributions
igsaw-0.010
hy-0.009
 brand-0.009
 doorway-0.009
 click-0.009
Token to
Token ID284
Feature activation-1.06e+00

Pos logit contributions
oyd+0.009
 pistols+0.009
 friends+0.008
 sometimes+0.008
 brothers+0.008
Neg logit contributions
igsaw-0.018
 doorway-0.017
hy-0.017
lic-0.017
 click-0.016
Token me
Token ID502
Feature activation-2.45e+00

Pos logit contributions
igsaw+0.053
hy+0.051
 brand+0.049
 doorway+0.048
col+0.048
Neg logit contributions
oyd-0.022
 pistols-0.017
 brothers-0.017
 sometimes-0.016
 neither-0.015
Token for
Token ID329
Feature activation+3.14e+00

Pos logit contributions
igsaw+0.114
hy+0.109
 doorway+0.107
icol+0.105
 brand+0.104
Neg logit contributions
oyd-0.059
 sometimes-0.050
 pistols-0.050
 brothers-0.048
 friends-0.046
Token safe
Token ID3338
Feature activation-1.50e+00

Pos logit contributions
oyd+0.071
 pistols+0.067
 friends+0.065
 brothers+0.064
 sometimes+0.061
Neg logit contributions
igsaw-0.147
hy-0.142
col-0.136
lic-0.135
icol-0.134
Tokenkeeping
Token ID19934
Feature activation-7.85e+00

Pos logit contributions
igsaw+0.045
hy+0.042
 brand+0.041
lic+0.040
 doorway+0.038
Neg logit contributions
oyd-0.060
 pistols-0.057
 sometimes-0.054
 brothers-0.054
 friends-0.052
Token.
Token ID13
Feature activation-3.72e-01

Pos logit contributions
igsaw+0.407
 brand+0.363
 doorway+0.362
lic+0.361
icol+0.353
Neg logit contributions
oyd-0.153
 on-0.137
 sometimes-0.135
 pistols-0.125
 friends-0.124
Token He
Token ID679
Feature activation+8.93e-01

Pos logit contributions
igsaw+0.015
hy+0.013
 doorway+0.013
 brand+0.013
col+0.013
Neg logit contributions
oyd-0.011
 pistols-0.010
 brothers-0.010
 sometimes-0.010
 also-0.009
Token told
Token ID1297
Feature activation+6.00e-01

Pos logit contributions
oyd+0.028
 pistols+0.024
 brothers+0.022
 sometimes+0.022
 neither+0.021
Neg logit contributions
igsaw-0.036
hy-0.034
 brand-0.032
 doorway-0.032
 click-0.032
Token me
Token ID502
Feature activation-2.43e+00

Pos logit contributions
oyd+0.011
 sometimes+0.010
 pistols+0.010
 friends+0.009
 brothers+0.009
Neg logit contributions
igsaw-0.031
hy-0.029
lic-0.029
 brand-0.028
 doorway-0.028
Token to
Token ID284
Feature activation-1.04e+00

Pos logit contributions
igsaw+0.112
icol+0.106
 doorway+0.104
hy+0.103
col+0.101
Neg logit contributions
oyd-0.057
 sometimes-0.053
 pistols-0.052
 brothers-0.049
 friends-0.049
Token
Token ID198
Feature activation-2.04e-02

Pos logit contributions
igsaw+0.001
hy+0.001
col+0.001
 doorway+0.001
 click+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 forth-0.001
 brothers-0.001
TokenSoon
Token ID28093
Feature activation-1.83e+00

Pos logit contributions
igsaw+0.001
hy+0.001
 click+0.001
col+0.001
 doorway+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 brothers-0.000
 brother-0.000
Token,
Token ID11
Feature activation+3.18e-01

Pos logit contributions
igsaw+0.070
hy+0.068
 click+0.064
 brand+0.064
 doorway+0.064
Neg logit contributions
oyd-0.060
 pistols-0.049
 brothers-0.044
 neither-0.044
ball-0.043
Token the
Token ID262
Feature activation-2.10e-03

Pos logit contributions
oyd+0.010
 pistols+0.008
ball+0.007
ained+0.007
 brothers+0.007
Neg logit contributions
igsaw-0.013
hy-0.012
 click-0.012
 brand-0.012
 doorway-0.012
Token door
Token ID3420
Feature activation-6.57e+00

Pos logit contributions
igsaw+0.000
hy+0.000
lic+0.000
col+0.000
 brand+0.000
Neg logit contributions
oyd-0.000
 pistols-0.000
 brothers-0.000
 friends-0.000
 brother-0.000
Tokenbell
Token ID7923
Feature activation-7.84e+00

Pos logit contributions
igsaw+0.255
 disl+0.223
icol+0.206
des+0.206
 Acc+0.206
Neg logit contributions
oyd-0.227
 on-0.210
 sometimes-0.203
 also-0.200
 neither-0.198
Token rang
Token ID28077
Feature activation-6.74e-01

Pos logit contributions
igsaw+0.411
gary+0.339
 brand+0.330
 disl+0.329
des+0.319
Neg logit contributions
 on-0.192
oyd-0.188
 at-0.187
 sometimes-0.181
 neither-0.153
Token.
Token ID13
Feature activation-1.59e-01

Pos logit contributions
igsaw+0.034
 disl+0.032
des+0.030
hy+0.030
endant+0.030
Neg logit contributions
oyd-0.014
 on-0.012
 at-0.011
 sometimes-0.011
ball-0.010
Token Lily
Token ID20037
Feature activation-1.53e+00

Pos logit contributions
igsaw+0.007
hy+0.007
des+0.007
 brand+0.007
col+0.007
Neg logit contributions
oyd-0.004
 Ben-0.003
 Rex-0.003
 Remy-0.003
 brothers-0.003
Token ran
Token ID4966
Feature activation-1.83e+00

Pos logit contributions
igsaw+0.067
hy+0.061
 brand+0.060
 click+0.058
 doorway+0.057
Neg logit contributions
oyd-0.045
 pistols-0.035
 sometimes-0.034
 also-0.034
 brothers-0.033
Token to
Token ID284
Feature activation-1.05e+00

Pos logit contributions
igsaw+0.098
 disl+0.091
endant+0.090
 Acc+0.088
 brand+0.087
Neg logit contributions
 on-0.035
 at-0.033
 sometimes-0.029
 each-0.027
 in-0.026
Token.
Token ID13
Feature activation-2.85e-01

Pos logit contributions
igsaw+0.140
 brand+0.132
hy+0.131
icol+0.128
 doorway+0.128
Neg logit contributions
oyd-0.065
 on-0.060
 sometimes-0.060
 pistols-0.056
 at-0.054
Token Soon
Token ID15894
Feature activation-1.69e+00

Pos logit contributions
igsaw+0.012
hy+0.012
 brand+0.012
 doorway+0.011
 click+0.011
Neg logit contributions
oyd-0.008
 pistols-0.006
 sometimes-0.006
 neither-0.006
 brothers-0.006
Token,
Token ID11
Feature activation+1.91e-01

Pos logit contributions
igsaw+0.077
hy+0.074
 brand+0.072
 doorway+0.071
col+0.070
Neg logit contributions
oyd-0.043
 pistols-0.035
 neither-0.033
 sometimes-0.032
 brothers-0.032
Token the
Token ID262
Feature activation-7.96e-02

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.003
 neither+0.003
ball+0.003
Neg logit contributions
igsaw-0.009
hy-0.008
 brand-0.008
 click-0.008
 doorway-0.008
Token door
Token ID3420
Feature activation-6.53e+00

Pos logit contributions
igsaw+0.003
hy+0.003
 brand+0.003
des+0.003
 click+0.003
Neg logit contributions
oyd-0.003
 pistols-0.002
 brothers-0.002
 friends-0.002
 brother-0.002
Tokenbell
Token ID7923
Feature activation-7.83e+00

Pos logit contributions
igsaw+0.291
 disl+0.250
des+0.237
icol+0.235
keeping+0.230
Neg logit contributions
oyd-0.192
 on-0.178
 also-0.178
 sometimes-0.178
 still-0.168
Token rang
Token ID28077
Feature activation-6.40e-01

Pos logit contributions
igsaw+0.422
gary+0.338
 disl+0.333
 brand+0.326
des+0.324
Neg logit contributions
 on-0.190
 at-0.181
 sometimes-0.179
oyd-0.178
 each-0.147
Token and
Token ID290
Feature activation-1.10e-01

Pos logit contributions
igsaw+0.033
 disl+0.030
hy+0.029
des+0.029
endant+0.029
Neg logit contributions
oyd-0.013
 on-0.011
 sometimes-0.010
 at-0.010
 pistols-0.010
Token Anna
Token ID11735
Feature activation-6.56e-01

Pos logit contributions
igsaw+0.006
hy+0.005
des+0.005
 disl+0.005
 brand+0.005
Neg logit contributions
oyd-0.002
 sometimes-0.002
 neither-0.002
 on-0.002
 also-0.002
Token ran
Token ID4966
Feature activation-1.84e+00

Pos logit contributions
igsaw+0.031
 brand+0.028
hy+0.028
 disl+0.027
 click+0.027
Neg logit contributions
oyd-0.018
 pistols-0.013
ball-0.013
 sometimes-0.013
 also-0.013
Token to
Token ID284
Feature activation-1.01e+00

Pos logit contributions
igsaw+0.100
 disl+0.096
endant+0.094
 Acc+0.093
 brand+0.087
Neg logit contributions
 on-0.036
 at-0.032
 in-0.026
 each-0.025
ball-0.025
Token.
Token ID13
Feature activation-1.83e-01

Pos logit contributions
igsaw+0.051
hy+0.048
 brand+0.047
 doorway+0.046
endant+0.045
Neg logit contributions
oyd-0.030
 pistols-0.024
 sometimes-0.024
 neither-0.022
 brothers-0.022
Token It
Token ID632
Feature activation-1.66e+00

Pos logit contributions
igsaw+0.007
 brand+0.007
des+0.007
hy+0.007
 doorway+0.006
Neg logit contributions
oyd-0.006
 Remy-0.005
 Rex-0.005
 Spot-0.004
 pistols-0.004
Token is
Token ID318
Feature activation-8.65e-01

Pos logit contributions
hy+0.057
igsaw+0.055
 doorway+0.055
 click+0.054
 brand+0.054
Neg logit contributions
oyd-0.059
 pistols-0.050
 neither-0.045
 brothers-0.044
ball-0.043
Token the
Token ID262
Feature activation-1.09e-01

Pos logit contributions
hy+0.028
 doorway+0.027
 brand+0.027
igsaw+0.026
 click+0.026
Neg logit contributions
oyd-0.033
 pistols-0.027
Rock-0.025
ball-0.025
ey-0.025
Token door
Token ID3420
Feature activation-6.54e+00

Pos logit contributions
hy+0.004
igsaw+0.004
 doorway+0.004
 brand+0.004
 click+0.004
Neg logit contributions
oyd-0.004
 pistols-0.003
 neither-0.003
 sometimes-0.003
 brothers-0.003
Tokenbell
Token ID7923
Feature activation-7.83e+00

Pos logit contributions
igsaw+0.267
 disl+0.247
 majesty+0.236
icol+0.230
keeping+0.229
Neg logit contributions
oyd-0.197
 on-0.187
 sometimes-0.185
 also-0.184
 at-0.176
Token.
Token ID13
Feature activation-2.17e-01

Pos logit contributions
igsaw+0.477
gary+0.420
 disl+0.394
 brand+0.394
handed+0.393
Neg logit contributions
 on-0.131
 at-0.129
 sometimes-0.113
oyd-0.101
 in-0.085
Token She
Token ID1375
Feature activation-2.34e+00

Pos logit contributions
igsaw+0.009
 brand+0.008
hy+0.008
 doorway+0.008
des+0.008
Neg logit contributions
oyd-0.006
 Remy-0.005
 Rex-0.005
 Spot-0.005
 pistols-0.005
Token knows
Token ID4206
Feature activation-1.76e+00

Pos logit contributions
igsaw+0.077
hy+0.075
 brand+0.071
 doorway+0.070
 click+0.069
Neg logit contributions
oyd-0.088
 pistols-0.075
 neither-0.070
 brothers-0.069
 sometimes-0.068
Token it
Token ID340
Feature activation+3.55e-01

Pos logit contributions
igsaw+0.074
hy+0.072
 brand+0.069
 doorway+0.068
 click+0.068
Neg logit contributions
oyd-0.050
 pistols-0.041
 neither-0.037
 brothers-0.037
 sometimes-0.037
Token is
Token ID318
Feature activation-9.00e-01

Pos logit contributions
oyd+0.010
 pistols+0.008
 sometimes+0.007
 neither+0.007
 brothers+0.007
Neg logit contributions
igsaw-0.015
hy-0.015
 brand-0.014
 doorway-0.014
 click-0.014
Token got
Token ID1392
Feature activation-2.93e-01

Pos logit contributions
igsaw+0.016
hy+0.016
 brand+0.015
 doorway+0.015
 click+0.015
Neg logit contributions
oyd-0.013
 pistols-0.010
 neither-0.009
 brothers-0.009
 sometimes-0.009
Token to
Token ID284
Feature activation-8.94e-01

Pos logit contributions
igsaw+0.012
hy+0.012
 brand+0.011
 doorway+0.011
 click+0.011
Neg logit contributions
oyd-0.009
 pistols-0.007
 brothers-0.006
 neither-0.006
 sometimes-0.006
Token ring
Token ID5858
Feature activation-3.76e-01

Pos logit contributions
lic+0.045
igsaw+0.044
cake+0.043
col+0.042
hy+0.042
Neg logit contributions
 each-0.017
 also-0.016
 sometimes-0.016
oyd-0.015
 mom-0.015
Token the
Token ID262
Feature activation+5.03e-02

Pos logit contributions
igsaw+0.017
hy+0.016
 doorway+0.015
 brand+0.015
 click+0.015
Neg logit contributions
oyd-0.010
 pistols-0.008
 neither-0.007
 sometimes-0.007
 brothers-0.007
Token door
Token ID3420
Feature activation-6.57e+00

Pos logit contributions
oyd+0.002
 pistols+0.001
 brothers+0.001
 sometimes+0.001
 neither+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 brand-0.002
 click-0.002
lic-0.002
Tokenbell
Token ID7923
Feature activation-7.83e+00

Pos logit contributions
igsaw+0.373
icol+0.352
 disl+0.346
des+0.333
 doorway+0.329
Neg logit contributions
 sometimes-0.113
 on-0.107
 at-0.097
 still-0.090
 also-0.089
Token and
Token ID290
Feature activation+5.79e-02

Pos logit contributions
igsaw+0.431
des+0.370
 doorway+0.369
 disl+0.368
 brand+0.363
Neg logit contributions
 on-0.137
 at-0.134
 sometimes-0.133
oyd-0.123
 each-0.107
Token say
Token ID910
Feature activation+2.82e-01

Pos logit contributions
oyd+0.001
 sometimes+0.001
 also+0.001
 pistols+0.001
 on+0.001
Neg logit contributions
igsaw-0.003
hy-0.002
 doorway-0.002
 brand-0.002
icol-0.002
Token hello
Token ID23748
Feature activation-4.75e+00

Pos logit contributions
 on+0.006
 sometimes+0.006
 at+0.005
oyd+0.005
 friends+0.005
Neg logit contributions
igsaw-0.014
 doorway-0.013
des-0.013
keeping-0.013
handed-0.013
Token to
Token ID284
Feature activation-8.81e-01

Pos logit contributions
igsaw+0.239
 doorway+0.212
hy+0.207
 brand+0.205
col+0.201
Neg logit contributions
oyd-0.102
 sometimes-0.088
 at-0.085
 on-0.084
 brother-0.083
Token the
Token ID262
Feature activation+5.05e-02

Pos logit contributions
igsaw+0.039
hy+0.038
 brand+0.036
 doorway+0.036
lic+0.035
Neg logit contributions
oyd-0.023
 pistols-0.019
 sometimes-0.017
 neither-0.017
 brothers-0.017
Token.
Token ID13
Feature activation-1.22e-01

Pos logit contributions
igsaw+0.080
hy+0.071
endant+0.069
keeping+0.069
 majesty+0.069
Neg logit contributions
oyd-0.037
 on-0.033
 sometimes-0.033
 at-0.032
 also-0.030
Token It
Token ID632
Feature activation-1.71e+00

Pos logit contributions
igsaw+0.005
hy+0.005
des+0.004
 brand+0.004
col+0.004
Neg logit contributions
oyd-0.004
 Spot-0.003
 Remy-0.003
 pistols-0.003
 on-0.003
Token was
Token ID373
Feature activation+6.13e-01

Pos logit contributions
igsaw+0.079
hy+0.076
 brand+0.073
 doorway+0.072
 click+0.072
Neg logit contributions
oyd-0.043
 pistols-0.033
 neither-0.029
 brothers-0.029
 sometimes-0.029
Token the
Token ID262
Feature activation-1.90e-01

Pos logit contributions
oyd+0.018
 pistols+0.015
 sometimes+0.015
 brothers+0.015
 also+0.014
Neg logit contributions
igsaw-0.025
hy-0.023
des-0.022
 click-0.022
endant-0.022
Token door
Token ID3420
Feature activation-6.43e+00

Pos logit contributions
igsaw+0.007
hy+0.007
 brand+0.006
 click+0.006
 doorway+0.006
Neg logit contributions
oyd-0.007
 pistols-0.006
 brothers-0.005
 neither-0.005
 sometimes-0.005
Tokenbell
Token ID7923
Feature activation-7.82e+00

Pos logit contributions
igsaw+0.264
 majesty+0.247
 disl+0.245
keeping+0.223
icol+0.220
Neg logit contributions
 on-0.201
 also-0.199
 sometimes-0.194
 still-0.194
 at-0.191
Token ringing
Token ID32333
Feature activation+6.37e-01

Pos logit contributions
igsaw+0.430
gary+0.357
 majesty+0.343
 disl+0.340
handed+0.339
Neg logit contributions
 at-0.176
 on-0.170
 sometimes-0.157
oyd-0.144
 forth-0.136
Token!
Token ID0
Feature activation-8.02e-01

Pos logit contributions
oyd+0.014
 on+0.014
 sometimes+0.013
 at+0.013
 each+0.012
Neg logit contributions
igsaw-0.031
hy-0.027
des-0.027
 disl-0.027
endant-0.027
Token
Token ID198
Feature activation-6.37e-03

Pos logit contributions
igsaw+0.032
hy+0.031
 brand+0.030
 click+0.029
col+0.029
Neg logit contributions
oyd-0.023
 pistols-0.017
 neither-0.017
 sometimes-0.017
Mom-0.017
Token
Token ID198
Feature activation+2.44e-02

Pos logit contributions
igsaw+0.000
hy+0.000
 brand+0.000
 doorway+0.000
 click+0.000
Neg logit contributions
oyd-0.000
 pistols-0.000
 brothers-0.000
 sometimes-0.000
 neither-0.000
TokenL
Token ID43
Feature activation-3.46e+00

Pos logit contributions
oyd+0.001
 pistols+0.001
ball+0.000
 sometimes+0.000
Rock+0.000
Neg logit contributions
igsaw-0.001
hy-0.001
 brand-0.001
 doorway-0.001
 click-0.001
Token said
Token ID531
Feature activation+1.44e+00

Pos logit contributions
oyd+0.027
 pistols+0.023
 brothers+0.021
 sometimes+0.021
 neither+0.020
Neg logit contributions
igsaw-0.032
hy-0.030
 brand-0.029
 click-0.028
 doorway-0.028
Token the
Token ID262
Feature activation-2.65e-01

Pos logit contributions
oyd+0.033
 pistols+0.028
 sometimes+0.025
 neither+0.025
 brothers+0.024
Neg logit contributions
igsaw-0.068
hy-0.065
 doorway-0.064
 brand-0.063
 click-0.061
Token mirror
Token ID10162
Feature activation-3.51e+00

Pos logit contributions
igsaw+0.011
hy+0.010
des+0.010
lic+0.010
aney+0.010
Neg logit contributions
 brother-0.008
oyd-0.007
 friends-0.007
 brothers-0.007
 mom-0.007
Token was
Token ID373
Feature activation+6.67e-01

Pos logit contributions
igsaw+0.186
 doorway+0.170
icol+0.168
 click+0.167
 brand+0.166
Neg logit contributions
oyd-0.065
 pistols-0.052
 sometimes-0.052
 neither-0.046
 on-0.046
Token a
Token ID257
Feature activation-1.98e-01

Pos logit contributions
oyd+0.017
 pistols+0.014
 brothers+0.013
 also+0.013
 sometimes+0.013
Neg logit contributions
igsaw-0.031
hy-0.029
 click-0.027
col-0.027
 doorway-0.027
Token portal
Token ID17898
Feature activation-7.81e+00

Pos logit contributions
igsaw+0.008
hy+0.007
keeping+0.007
aney+0.007
lic+0.007
Neg logit contributions
oyd-0.005
 on-0.005
 brother-0.005
 also-0.005
 mom-0.005
Token to
Token ID284
Feature activation-1.10e+00

Pos logit contributions
igsaw+0.415
 Acc+0.393
 disl+0.385
icol+0.370
des+0.368
Neg logit contributions
 on-0.143
 at-0.125
 sometimes-0.124
oyd-0.107
 also-0.107
Token another
Token ID1194
Feature activation+1.46e+00

Pos logit contributions
igsaw+0.049
icol+0.044
lic+0.042
hy+0.042
 brand+0.042
Neg logit contributions
 each-0.028
oyd-0.027
 sometimes-0.027
 also-0.027
 friends-0.027
Token world
Token ID995
Feature activation-4.44e+00

Pos logit contributions
oyd+0.038
 sometimes+0.034
 brothers+0.033
 brother+0.032
 also+0.032
Neg logit contributions
igsaw-0.062
hy-0.059
endant-0.059
lic-0.058
 disl-0.057
Token.
Token ID13
Feature activation-3.06e-01

Pos logit contributions
igsaw+0.241
icol+0.225
 brand+0.222
 fru+0.220
 Acc+0.219
Neg logit contributions
 on-0.076
 sometimes-0.071
 at-0.068
oyd-0.059
 in-0.054
Token He
Token ID679
Feature activation+7.95e-01

Pos logit contributions
igsaw+0.014
col+0.014
 doorway+0.013
hy+0.013
chie+0.013
Neg logit contributions
oyd-0.006
 pistols-0.006
 brothers-0.006
 sometimes-0.006
 Spot-0.005
Token her
Token ID607
Feature activation-1.69e+00

Pos logit contributions
igsaw+0.042
lic+0.038
 doorway+0.038
endant+0.038
hy+0.037
Neg logit contributions
oyd-0.022
 at-0.021
 each-0.020
 on-0.019
 sometimes-0.019
Token wand
Token ID11569
Feature activation+5.71e-01

Pos logit contributions
igsaw+0.066
hy+0.064
 brand+0.060
col+0.060
 click+0.059
Neg logit contributions
oyd-0.055
 pistols-0.046
 brothers-0.042
 neither-0.040
 sometimes-0.040
Token and
Token ID290
Feature activation+4.51e-02

Pos logit contributions
oyd+0.013
 pistols+0.010
 sometimes+0.009
 neither+0.009
 brothers+0.008
Neg logit contributions
igsaw-0.028
hy-0.027
 brand-0.026
 click-0.025
 doorway-0.025
Token a
Token ID257
Feature activation-2.57e-01

Pos logit contributions
oyd+0.001
 sometimes+0.001
 pistols+0.001
 neither+0.001
 also+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 brand-0.002
endant-0.002
 disl-0.002
Token magical
Token ID10883
Feature activation-1.50e+00

Pos logit contributions
igsaw+0.010
keeping+0.010
 disl+0.010
des+0.009
endant+0.009
Neg logit contributions
oyd-0.007
 friends-0.007
 brother-0.006
 on-0.006
 brothers-0.006
Token portal
Token ID17898
Feature activation-7.81e+00

Pos logit contributions
igsaw+0.054
 disl+0.053
des+0.051
 Acc+0.051
col+0.050
Neg logit contributions
oyd-0.046
 friends-0.041
 on-0.040
 brother-0.040
 friend-0.039
Token opened
Token ID4721
Feature activation-3.93e+00

Pos logit contributions
igsaw+0.409
 Acc+0.370
 disl+0.361
icol+0.353
des+0.341
Neg logit contributions
 on-0.171
 at-0.155
oyd-0.151
 sometimes-0.144
 in-0.135
Token.
Token ID13
Feature activation-1.18e-02

Pos logit contributions
igsaw+0.202
hy+0.187
 disl+0.183
 Acc+0.183
lic+0.183
Neg logit contributions
 on-0.070
oyd-0.069
 at-0.068
 sometimes-0.062
 each-0.055
Token
Token ID198
Feature activation-1.88e-01

Pos logit contributions
hy+0.001
igsaw+0.001
 brand+0.000
endant+0.000
 disl+0.000
Neg logit contributions
oyd-0.000
 Remy-0.000
 brothers-0.000
Rock-0.000
 sometimes-0.000
Token
Token ID198
Feature activation-1.98e-01

Pos logit contributions
igsaw+0.009
hy+0.009
 brand+0.008
 doorway+0.008
 click+0.008
Neg logit contributions
oyd-0.004
 pistols-0.003
 brothers-0.003
 sometimes-0.003
ball-0.003
TokenThe
Token ID464
Feature activation-7.08e-01

Pos logit contributions
igsaw+0.009
hy+0.009
 brand+0.009
 doorway+0.009
icol+0.008
Neg logit contributions
oyd-0.005
 pistols-0.004
Rock-0.003
 sometimes-0.003
ball-0.003
Token.
Token ID13
Feature activation-2.42e-01

Pos logit contributions
igsaw+0.177
hy+0.164
 disl+0.162
des+0.161
 brand+0.161
Neg logit contributions
oyd-0.093
 on-0.071
 pistols-0.070
 sometimes-0.068
ball-0.066
Token The
Token ID383
Feature activation-1.16e+00

Pos logit contributions
igsaw+0.010
des+0.009
 brand+0.009
hy+0.009
chie+0.009
Neg logit contributions
oyd-0.007
 sometimes-0.006
 brothers-0.005
 brother-0.005
 mom-0.005
Token light
Token ID1657
Feature activation-3.79e+00

Pos logit contributions
igsaw+0.042
hy+0.038
col+0.038
 brand+0.036
icol+0.036
Neg logit contributions
oyd-0.041
 pistols-0.035
 brothers-0.034
 sometimes-0.033
 humans-0.033
Token was
Token ID373
Feature activation+6.29e-01

Pos logit contributions
igsaw+0.144
hy+0.137
 brand+0.131
 doorway+0.127
 click+0.127
Neg logit contributions
oyd-0.126
 pistols-0.104
 brothers-0.096
 sometimes-0.096
 neither-0.096
Token a
Token ID257
Feature activation-2.52e-01

Pos logit contributions
oyd+0.021
 pistols+0.017
 sometimes+0.017
 brothers+0.016
 neither+0.016
Neg logit contributions
igsaw-0.024
hy-0.022
col-0.021
 doorway-0.021
des-0.021
Token portal
Token ID17898
Feature activation-7.80e+00

Pos logit contributions
igsaw+0.009
hy+0.008
des+0.008
col+0.008
 disl+0.008
Neg logit contributions
oyd-0.009
 pistols-0.007
 also-0.007
 brothers-0.007
 brother-0.007
Token.
Token ID13
Feature activation-2.17e-01

Pos logit contributions
igsaw+0.422
 disl+0.398
 Acc+0.398
des+0.383
icol+0.372
Neg logit contributions
 on-0.145
 at-0.128
oyd-0.120
 sometimes-0.118
 also-0.118
Token It
Token ID632
Feature activation-1.83e+00

Pos logit contributions
igsaw+0.008
hy+0.008
 brand+0.008
col+0.008
des+0.008
Neg logit contributions
oyd-0.007
 sometimes-0.006
 brothers-0.006
 neither-0.005
 brother-0.005
Token took
Token ID1718
Feature activation-7.94e-01

Pos logit contributions
igsaw+0.083
hy+0.079
 brand+0.076
 doorway+0.074
 click+0.074
Neg logit contributions
oyd-0.048
 pistols-0.037
 sometimes-0.034
 neither-0.034
 brothers-0.034
Token her
Token ID607
Feature activation-1.71e+00

Pos logit contributions
igsaw+0.039
 disl+0.038
 brand+0.037
hy+0.037
endant+0.037
Neg logit contributions
oyd-0.015
 on-0.014
 friends-0.013
 at-0.013
 sometimes-0.013
Token to
Token ID284
Feature activation-1.12e+00

Pos logit contributions
igsaw+0.078
hy+0.075
 brand+0.072
 doorway+0.072
lic+0.072
Neg logit contributions
oyd-0.042
 pistols-0.034
 sometimes-0.030
 neither-0.030
 brothers-0.030
Token better
Token ID1365
Feature activation+2.61e+00

Pos logit contributions
igsaw+0.099
hy+0.093
 doorway+0.091
keeping+0.089
icol+0.087
Neg logit contributions
oyd-0.062
 sometimes-0.053
 neither-0.052
 also-0.050
 pistols-0.050
Token and
Token ID290
Feature activation-1.16e-01

Pos logit contributions
oyd+0.068
 pistols+0.052
 sometimes+0.050
 neither+0.049
 brothers+0.047
Neg logit contributions
igsaw-0.118
hy-0.113
 brand-0.110
 doorway-0.108
col-0.107
Token then
Token ID788
Feature activation+1.82e+00

Pos logit contributions
igsaw+0.006
hy+0.005
chie+0.005
lic+0.005
endant+0.005
Neg logit contributions
oyd-0.003
 sometimes-0.003
 also-0.003
 neither-0.002
 friends-0.002
Token the
Token ID262
Feature activation-3.49e-01

Pos logit contributions
oyd+0.054
 pistols+0.045
 sometimes+0.045
 neither+0.045
 friends+0.044
Neg logit contributions
igsaw-0.075
hy-0.071
lic-0.066
col-0.065
 doorway-0.065
Token door
Token ID3420
Feature activation-6.49e+00

Pos logit contributions
igsaw+0.015
hy+0.014
 brand+0.013
 click+0.013
col+0.013
Neg logit contributions
oyd-0.010
 pistols-0.008
 brothers-0.008
 sometimes-0.007
 neither-0.007
Tokenbell
Token ID7923
Feature activation-7.80e+00

Pos logit contributions
igsaw+0.268
 disl+0.227
keeping+0.214
icol+0.209
 majesty+0.203
Neg logit contributions
 on-0.218
 still-0.212
oyd-0.210
 at-0.208
 sometimes-0.207
Token rang
Token ID28077
Feature activation-6.51e-01

Pos logit contributions
igsaw+0.397
 disl+0.312
gary+0.303
aney+0.300
 brand+0.298
Neg logit contributions
oyd-0.211
 sometimes-0.188
 on-0.187
 at-0.178
 still-0.160
Token.
Token ID13
Feature activation-1.35e-01

Pos logit contributions
igsaw+0.032
 disl+0.030
des+0.030
endant+0.029
hy+0.029
Neg logit contributions
 on-0.012
oyd-0.012
 at-0.011
 sometimes-0.010
 each-0.010
Token Mom
Token ID11254
Feature activation-6.73e-01

Pos logit contributions
hy+0.005
des+0.005
igsaw+0.005
endant+0.005
chie+0.005
Neg logit contributions
oyd-0.004
 mom-0.004
 brother-0.003
Mom-0.003
 sometimes-0.003
Token opened
Token ID4721
Feature activation-4.16e+00

Pos logit contributions
igsaw+0.031
 disl+0.027
hy+0.027
 brand+0.026
des+0.026
Neg logit contributions
oyd-0.019
 also-0.016
 sometimes-0.015
 brothers-0.015
 brother-0.015
Token the
Token ID262
Feature activation-3.11e-01

Pos logit contributions
igsaw+0.199
hy+0.199
lic+0.187
des+0.185
 brand+0.183
Neg logit contributions
oyd-0.086
 pistols-0.067
 on-0.064
 sometimes-0.063
 neither-0.061
Token oak
Token ID26210
Feature activation-2.68e-01

Pos logit contributions
igsaw+0.003
hy+0.003
 doorway+0.003
 brand+0.003
 click+0.003
Neg logit contributions
oyd-0.003
 pistols-0.003
 neither-0.003
 brothers-0.002
 sometimes-0.002
Token tree
Token ID5509
Feature activation-7.53e-01

Pos logit contributions
igsaw+0.012
lic+0.011
hy+0.011
icol+0.011
 brand+0.011
Neg logit contributions
oyd-0.007
 sometimes-0.006
 also-0.006
 pistols-0.006
 on-0.006
Token,
Token ID11
Feature activation+2.30e-01

Pos logit contributions
igsaw+0.039
hy+0.036
lic+0.036
icol+0.036
 disl+0.036
Neg logit contributions
oyd-0.013
 sometimes-0.012
 also-0.011
 on-0.010
 pistols-0.010
Token their
Token ID511
Feature activation-6.51e-01

Pos logit contributions
oyd+0.006
 sometimes+0.006
 also+0.006
 pistols+0.006
 neither+0.006
Neg logit contributions
igsaw-0.010
lic-0.009
hy-0.009
col-0.009
 doorway-0.009
Token secret
Token ID3200
Feature activation-1.50e+00

Pos logit contributions
igsaw+0.028
hy+0.027
 brand+0.026
 doorway+0.025
 click+0.025
Neg logit contributions
oyd-0.018
 pistols-0.015
 brothers-0.013
 neither-0.013
 sometimes-0.013
Token portal
Token ID17898
Feature activation-7.80e+00

Pos logit contributions
igsaw+0.065
hy+0.061
lic+0.059
 click+0.057
icol+0.056
Neg logit contributions
oyd-0.041
 friends-0.038
 sometimes-0.035
 brothers-0.034
 on-0.034
Token.
Token ID13
Feature activation-2.40e-01

Pos logit contributions
igsaw+0.396
 click+0.350
hy+0.349
icol+0.346
 Acc+0.345
Neg logit contributions
oyd-0.159
 friends-0.145
 on-0.145
 sometimes-0.141
 each-0.135
Token They
Token ID1119
Feature activation-1.68e-01

Pos logit contributions
igsaw+0.011
hy+0.011
 brand+0.010
 doorway+0.010
 click+0.010
Neg logit contributions
oyd-0.006
 pistols-0.005
 brothers-0.004
 neither-0.004
 sometimes-0.004
Token never
Token ID1239
Feature activation-2.07e+00

Pos logit contributions
igsaw+0.008
hy+0.007
 brand+0.007
 doorway+0.007
 click+0.006
Neg logit contributions
oyd-0.005
 pistols-0.004
 sometimes-0.004
 also-0.004
 neither-0.004
Token told
Token ID1297
Feature activation+5.77e-01

Pos logit contributions
igsaw+0.103
 brand+0.094
 doorway+0.091
 click+0.088
icol+0.087
Neg logit contributions
oyd-0.051
 sometimes-0.042
 also-0.040
 neither-0.039
 pistols-0.037
Token anyone
Token ID2687
Feature activation+2.00e+00

Pos logit contributions
oyd+0.010
 pistols+0.009
 neither+0.008
 sometimes+0.008
 friends+0.008
Neg logit contributions
igsaw-0.030
lic-0.029
hy-0.028
 doorway-0.028
cake-0.028
Token.
Token ID13
Feature activation-4.39e-02

Pos logit contributions
igsaw+0.162
icol+0.147
hy+0.144
endant+0.144
lic+0.142
Neg logit contributions
 on-0.054
 sometimes-0.053
oyd-0.051
 at-0.050
 also-0.044
Token They
Token ID1119
Feature activation-5.29e-02

Pos logit contributions
igsaw+0.002
col+0.002
hy+0.002
 click+0.002
 brand+0.002
Neg logit contributions
oyd-0.001
 sometimes-0.001
 pistols-0.001
 brothers-0.001
 Spot-0.001
Token rang
Token ID28077
Feature activation-6.87e-01

Pos logit contributions
igsaw+0.002
hy+0.002
 brand+0.002
 doorway+0.002
icol+0.002
Neg logit contributions
oyd-0.001
 sometimes-0.001
 also-0.001
 pistols-0.001
 neither-0.001
Token the
Token ID262
Feature activation+5.26e-02

Pos logit contributions
igsaw+0.031
hy+0.030
 doorway+0.029
lic+0.028
col+0.028
Neg logit contributions
oyd-0.017
 pistols-0.014
 sometimes-0.013
 neither-0.012
 horses-0.012
Token door
Token ID3420
Feature activation-6.42e+00

Pos logit contributions
oyd+0.002
 pistols+0.002
 brothers+0.002
 horses+0.002
 humans+0.002
Neg logit contributions
igsaw-0.002
hy-0.002
lic-0.002
des-0.002
 brand-0.002
Tokenbell
Token ID7923
Feature activation-7.77e+00

Pos logit contributions
igsaw+0.345
 disl+0.332
 majesty+0.327
icol+0.316
des+0.316
Neg logit contributions
 on-0.116
 sometimes-0.110
 at-0.108
 each-0.089
 also-0.088
Token to
Token ID284
Feature activation-8.97e-01

Pos logit contributions
igsaw+0.427
gary+0.366
des+0.364
 disl+0.357
handed+0.351
Neg logit contributions
 on-0.145
 at-0.140
 sometimes-0.133
oyd-0.125
 each-0.116
Token see
Token ID766
Feature activation-3.80e+00

Pos logit contributions
igsaw+0.040
endant+0.039
lic+0.038
hy+0.038
col+0.037
Neg logit contributions
oyd-0.021
 each-0.020
 also-0.018
 sometimes-0.018
 pistols-0.018
Token what
Token ID644
Feature activation+1.18e+00

Pos logit contributions
igsaw+0.175
hy+0.165
 doorway+0.159
icol+0.154
lic+0.154
Neg logit contributions
oyd-0.091
 pistols-0.083
 sometimes-0.076
 brothers-0.073
 horses-0.071
Token was
Token ID373
Feature activation+6.18e-01

Pos logit contributions
oyd+0.028
 pistols+0.022
 sometimes+0.019
 brothers+0.019
 neither+0.019
Neg logit contributions
igsaw-0.055
hy-0.053
 brand-0.052
 doorway-0.051
 click-0.051
Token going
Token ID1016
Feature activation-1.43e+00

Pos logit contributions
 on+0.010
oyd+0.010
 at+0.009
 sometimes+0.009
 also+0.008
Neg logit contributions
igsaw-0.034
hy-0.031
 click-0.030
lic-0.029
icol-0.029
Token The
Token ID383
Feature activation-8.16e-01

Pos logit contributions
oyd+0.001
 Spot+0.001
 Ben+0.001
 Remy+0.001
 Max+0.001
Neg logit contributions
des-0.002
hy-0.002
igsaw-0.002
chie-0.002
col-0.002
Token driver
Token ID4639
Feature activation+6.46e-01

Pos logit contributions
igsaw+0.034
hy+0.032
col+0.030
endant+0.029
lic+0.029
Neg logit contributions
oyd-0.024
 pistols-0.021
 brothers-0.020
 friends-0.018
 horses-0.018
Token rang
Token ID28077
Feature activation-5.45e-01

Pos logit contributions
oyd+0.018
 pistols+0.014
 brothers+0.014
 neither+0.014
 also+0.014
Neg logit contributions
igsaw-0.029
hy-0.026
 brand-0.025
icol-0.025
col-0.025
Token the
Token ID262
Feature activation-2.49e-02

Pos logit contributions
igsaw+0.028
hy+0.026
 doorway+0.025
endant+0.025
des+0.025
Neg logit contributions
oyd-0.010
 pistols-0.009
 sometimes-0.008
 on-0.008
 neither-0.008
Token door
Token ID3420
Feature activation-6.41e+00

Pos logit contributions
igsaw+0.001
hy+0.001
col+0.001
 disl+0.001
endant+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 brothers-0.001
 horses-0.001
 brother-0.001
Tokenbell
Token ID7923
Feature activation-7.76e+00

Pos logit contributions
igsaw+0.373
 disl+0.367
 majesty+0.349
keeping+0.341
endant+0.340
Neg logit contributions
 on-0.089
 at-0.084
 sometimes-0.081
 also-0.065
oyd-0.062
Token and
Token ID290
Feature activation-3.30e-02

Pos logit contributions
igsaw+0.453
des+0.389
gary+0.389
 disl+0.384
 brand+0.378
Neg logit contributions
oyd-0.113
 at-0.111
 on-0.108
 sometimes-0.106
 each-0.082
Token her
Token ID607
Feature activation-1.72e+00

Pos logit contributions
igsaw+0.002
hy+0.001
 disl+0.001
endant+0.001
 brand+0.001
Neg logit contributions
oyd-0.001
 sometimes-0.001
 on-0.001
 also-0.001
 pistols-0.001
Token mom
Token ID1995
Feature activation-2.13e-01

Pos logit contributions
igsaw+0.086
hy+0.080
col+0.078
 click+0.077
 brand+0.077
Neg logit contributions
oyd-0.038
 pistols-0.029
 brothers-0.027
 friends-0.025
 brother-0.024
Token answered
Token ID9373
Feature activation+1.16e+00

Pos logit contributions
igsaw+0.010
 disl+0.009
hy+0.008
icol+0.008
 brand+0.008
Neg logit contributions
oyd-0.006
 brothers-0.005
 also-0.005
 pistols-0.005
 sometimes-0.005
Token it
Token ID340
Feature activation+2.85e-01

Pos logit contributions
oyd+0.022
 on+0.021
 sometimes+0.021
 at+0.021
 each+0.020
Neg logit contributions
igsaw-0.057
 doorway-0.053
endant-0.052
des-0.052
hy-0.051
Token heard
Token ID2982
Feature activation-2.84e+00

Pos logit contributions
 on+0.004
 friends+0.004
 also+0.004
 both+0.004
 each+0.004
Neg logit contributions
igsaw-0.008
 Acc-0.006
icol-0.006
hy-0.006
cake-0.006
Token before
Token ID878
Feature activation+2.73e+00

Pos logit contributions
igsaw+0.146
icol+0.128
endant+0.128
 disl+0.127
 Acc+0.126
Neg logit contributions
 on-0.055
 sometimes-0.052
 at-0.050
oyd-0.049
 each-0.043
Token was
Token ID373
Feature activation+6.65e-01

Pos logit contributions
oyd+0.064
 sometimes+0.054
 on+0.054
 pistols+0.053
 brothers+0.050
Neg logit contributions
igsaw-0.128
hy-0.120
lic-0.115
col-0.115
des-0.114
Token the
Token ID262
Feature activation-3.42e-01

Pos logit contributions
oyd+0.016
 on+0.015
 also+0.015
 brothers+0.014
 friends+0.014
Neg logit contributions
igsaw-0.031
hy-0.027
icol-0.027
des-0.027
col-0.026
Token door
Token ID3420
Feature activation-6.54e+00

Pos logit contributions
igsaw+0.014
lic+0.013
hy+0.013
col+0.013
endant+0.012
Neg logit contributions
oyd-0.010
 pistols-0.009
 brothers-0.009
 friends-0.008
 brother-0.008
Tokenbell
Token ID7923
Feature activation-7.76e+00

Pos logit contributions
igsaw+0.317
 disl+0.289
icol+0.285
 majesty+0.284
 Acc+0.264
Neg logit contributions
 on-0.174
 sometimes-0.160
 still-0.157
 also-0.156
 at-0.156
Token.
Token ID13
Feature activation-1.04e-01

Pos logit contributions
igsaw+0.386
gary+0.310
aney+0.309
 disl+0.303
col+0.300
Neg logit contributions
 on-0.199
 sometimes-0.194
 at-0.193
oyd-0.178
 still-0.170
Token From
Token ID3574
Feature activation-3.40e+00

Pos logit contributions
hy+0.004
igsaw+0.004
 click+0.004
 doorway+0.004
col+0.004
Neg logit contributions
oyd-0.003
 pistols-0.002
 brothers-0.002
 Remy-0.002
 sometimes-0.002
Token then
Token ID788
Feature activation+1.91e+00

Pos logit contributions
igsaw+0.121
 brand+0.121
hy+0.120
 doorway+0.114
col+0.114
Neg logit contributions
oyd-0.112
 pistols-0.097
 goalie-0.087
 brothers-0.087
 neither-0.084
Token on
Token ID319
Feature activation+1.71e+00

Pos logit contributions
oyd+0.037
 pistols+0.026
 goalie+0.019
ball+0.019
 brothers+0.018
Neg logit contributions
hy-0.098
igsaw-0.098
 click-0.094
col-0.092
 doorway-0.091
Token,
Token ID11
Feature activation+3.69e-01

Pos logit contributions
oyd+0.061
 pistols+0.051
 brothers+0.046
ball+0.045
 goalie+0.045
Neg logit contributions
igsaw-0.061
hy-0.059
 brand-0.056
 doorway-0.055
 click-0.055
Token ringing
Token ID32333
Feature activation+8.28e-01

Pos logit contributions
igsaw+0.059
hy+0.054
col+0.053
lic+0.053
 disl+0.052
Neg logit contributions
oyd-0.022
 pistols-0.016
 brothers-0.016
 sometimes-0.015
 friends-0.015
Token sound
Token ID2128
Feature activation-1.67e+00

Pos logit contributions
oyd+0.021
 sometimes+0.015
 neither+0.015
 on+0.014
 pistols+0.014
Neg logit contributions
igsaw-0.040
icol-0.036
 brand-0.035
hy-0.035
col-0.034
Token of
Token ID286
Feature activation-7.10e-01

Pos logit contributions
igsaw+0.082
hy+0.073
icol+0.072
endant+0.072
keeping+0.071
Neg logit contributions
oyd-0.039
 sometimes-0.032
 on-0.031
 neither-0.029
 at-0.029
Token the
Token ID262
Feature activation-3.08e-01

Pos logit contributions
igsaw+0.034
lic+0.032
hy+0.032
endant+0.031
des+0.031
Neg logit contributions
oyd-0.016
 pistols-0.013
 sometimes-0.013
 friends-0.013
 each-0.012
Token door
Token ID3420
Feature activation-6.45e+00

Pos logit contributions
igsaw+0.013
lic+0.012
hy+0.012
icol+0.011
endant+0.011
Neg logit contributions
oyd-0.009
 brothers-0.007
 friends-0.007
 brother-0.007
 pistols-0.007
Tokenbell
Token ID7923
Feature activation-7.75e+00

Pos logit contributions
igsaw+0.304
 disl+0.263
keeping+0.256
icol+0.245
 majesty+0.244
Neg logit contributions
 still-0.182
 on-0.182
 sometimes-0.177
 also-0.171
 at-0.171
Token was
Token ID373
Feature activation+7.05e-01

Pos logit contributions
igsaw+0.407
gary+0.320
 disl+0.318
handed+0.311
col+0.311
Neg logit contributions
oyd-0.196
 sometimes-0.184
 on-0.174
 at-0.170
 still-0.156
Token still
Token ID991
Feature activation+3.48e+00

Pos logit contributions
oyd+0.019
 also+0.017
 sometimes+0.017
 on+0.017
 still+0.016
Neg logit contributions
igsaw-0.031
hy-0.027
des-0.026
col-0.026
 disl-0.026
Token ringing
Token ID32333
Feature activation+8.08e-01

Pos logit contributions
oyd+0.083
 on+0.070
 friends+0.067
 sometimes+0.066
 brothers+0.065
Neg logit contributions
igsaw-0.162
hy-0.153
lic-0.146
col-0.145
des-0.144
Token in
Token ID287
Feature activation+2.94e-01

Pos logit contributions
 on+0.014
oyd+0.014
 sometimes+0.014
 at+0.014
 each+0.012
Neg logit contributions
igsaw-0.041
des-0.038
 disl-0.037
endant-0.037
 doorway-0.037
Token Lucy
Token ID22162
Feature activation-7.68e-01

Pos logit contributions
 sometimes+0.006
 each+0.006
 on+0.005
oyd+0.005
 friends+0.005
Neg logit contributions
lic-0.014
igsaw-0.014
endant-0.014
des-0.014
hy-0.013
Token
Token ID198
Feature activation-8.15e-01

Pos logit contributions
igsaw+0.034
hy+0.033
 brand+0.032
 click+0.031
 doorway+0.031
Neg logit contributions
oyd-0.024
 pistols-0.020
 neither-0.018
 brothers-0.018
 sometimes-0.017
TokenSoon
Token ID28093
Feature activation-1.65e+00

Pos logit contributions
igsaw+0.035
hy+0.034
 brand+0.032
 click+0.032
 doorway+0.032
Neg logit contributions
oyd-0.023
 pistols-0.019
 neither-0.017
 brothers-0.017
 sometimes-0.016
Token,
Token ID11
Feature activation+3.21e-01

Pos logit contributions
igsaw+0.053
hy+0.051
 click+0.050
 brand+0.048
 doorway+0.047
Neg logit contributions
oyd-0.065
 pistols-0.053
ball-0.050
Rock-0.048
 brothers-0.047
Token the
Token ID262
Feature activation-1.30e-01

Pos logit contributions
oyd+0.012
 pistols+0.010
ball+0.009
ey+0.009
Rock+0.009
Neg logit contributions
igsaw-0.011
 brand-0.011
 click-0.010
hy-0.010
 doorway-0.010
Token door
Token ID3420
Feature activation-6.54e+00

Pos logit contributions
igsaw+0.004
hy+0.004
 brand+0.004
col+0.004
lic+0.004
Neg logit contributions
oyd-0.005
 pistols-0.004
 brothers-0.004
 sometimes-0.004
 friends-0.004
Tokenbell
Token ID7923
Feature activation-7.74e+00

Pos logit contributions
igsaw+0.237
 disl+0.206
icol+0.183
 Acc+0.183
keeping+0.180
Neg logit contributions
oyd-0.239
 on-0.237
 sometimes-0.229
 also-0.228
 at-0.225
Token rang
Token ID28077
Feature activation-6.61e-01

Pos logit contributions
igsaw+0.389
 disl+0.311
 brand+0.305
gary+0.302
chie+0.300
Neg logit contributions
oyd-0.210
 sometimes-0.192
 on-0.188
 at-0.181
 neither-0.172
Token and
Token ID290
Feature activation-5.60e-03

Pos logit contributions
igsaw+0.034
 disl+0.032
des+0.031
hy+0.030
 doorway+0.030
Neg logit contributions
oyd-0.012
 on-0.012
 at-0.011
 sometimes-0.011
 each-0.009
Token Lily
Token ID20037
Feature activation-1.73e+00

Pos logit contributions
igsaw+0.000
hy+0.000
des+0.000
 disl+0.000
handed+0.000
Neg logit contributions
oyd-0.000
 sometimes-0.000
 on-0.000
 neither-0.000
 also-0.000
Token's
Token ID338
Feature activation+2.39e+00

Pos logit contributions
igsaw+0.078
 brand+0.067
hy+0.067
 disl+0.066
 click+0.064
Neg logit contributions
oyd-0.052
 also-0.040
ball-0.039
 pistols-0.039
 sometimes-0.039
Token friends
Token ID2460
Feature activation+1.43e+00

Pos logit contributions
oyd+0.043
 friends+0.033
 pistols+0.033
 brothers+0.032
 brother+0.031
Neg logit contributions
igsaw-0.123
hy-0.119
lic-0.118
col-0.115
 brand-0.114
Token giving
Token ID3501
Feature activation-2.50e-01

Pos logit contributions
oyd+0.043
 sometimes+0.040
 neither+0.037
 pistols+0.037
 also+0.036
Neg logit contributions
igsaw-0.074
lic-0.070
hy-0.069
 doorway-0.067
 brand-0.066
Token was
Token ID373
Feature activation+7.42e-01

Pos logit contributions
igsaw+0.010
hy+0.010
 doorway+0.010
 click+0.010
 brand+0.010
Neg logit contributions
oyd-0.007
 pistols-0.006
 friends-0.006
 sometimes-0.006
 each-0.005
Token just
Token ID655
Feature activation+1.25e+00

Pos logit contributions
oyd+0.017
 friends+0.015
 also+0.015
 sometimes+0.015
 pistols+0.015
Neg logit contributions
igsaw-0.035
hy-0.033
 doorway-0.032
 click-0.032
icol-0.031
Token as
Token ID355
Feature activation+4.43e+00

Pos logit contributions
 sometimes+0.026
oyd+0.026
 on+0.024
 neither+0.024
 also+0.024
Neg logit contributions
igsaw-0.058
lic-0.056
hy-0.055
 doorway-0.054
cake-0.053
Token special
Token ID2041
Feature activation-1.27e+00

Pos logit contributions
 friends+0.119
 brothers+0.109
 best+0.106
 on+0.104
 also+0.103
Neg logit contributions
igsaw-0.184
 doorway-0.181
hy-0.180
icol-0.179
 click-0.172
Token as
Token ID355
Feature activation+4.45e+00

Pos logit contributions
igsaw+0.061
 doorway+0.061
lic+0.060
hy+0.059
endant+0.058
Neg logit contributions
 sometimes-0.023
 friends-0.022
 on-0.022
 at-0.021
oyd-0.021
Token receiving
Token ID6464
Feature activation-8.31e-01

Pos logit contributions
 friends+0.105
 sometimes+0.098
 on+0.098
 brothers+0.096
oyd+0.093
Neg logit contributions
igsaw-0.193
hy-0.192
 doorway-0.184
 click-0.183
lic-0.180
Token.
Token ID13
Feature activation-3.23e-02

Pos logit contributions
igsaw+0.039
lic+0.038
 doorway+0.037
icol+0.037
hy+0.037
Neg logit contributions
 on-0.018
 friends-0.016
 at-0.016
 sometimes-0.015
 each-0.015
Token<|endoftext|>
Token ID50256
Feature activation+2.00e+00

Pos logit contributions
hy+0.002
igsaw+0.002
 rubble+0.002
endant+0.002
 click+0.002
Neg logit contributions
oyd-0.000
 also-0.000
 sometimes-0.000
Mom-0.000
 brothers-0.000
TokenOnce
Token ID7454
Feature activation+1.01e+00

Pos logit contributions
oyd+0.055
 pistols+0.047
 neither+0.043
 brothers+0.038
 forth+0.038
Neg logit contributions
igsaw-0.085
hy-0.082
 doorway-0.079
col-0.078
lic-0.077
Token there
Token ID612
Feature activation-6.01e-01

Pos logit contributions
oyd+0.032
 pistols+0.027
Rock+0.023
ball+0.023
 brothers+0.023
Neg logit contributions
igsaw-0.041
hy-0.038
 brand-0.037
 doorway-0.037
 click-0.035
Token forgot
Token ID16453
Feature activation-2.00e+00

Pos logit contributions
hy+0.066
igsaw+0.065
 brand+0.064
 doorway+0.063
 click+0.061
Neg logit contributions
oyd-0.052
 pistols-0.045
 neither-0.039
 brothers-0.038
 sometimes-0.037
Token about
Token ID546
Feature activation+1.07e+00

Pos logit contributions
igsaw+0.098
hy+0.096
 doorway+0.089
 click+0.089
col+0.088
Neg logit contributions
oyd-0.042
 pistols-0.035
 sometimes-0.033
 neither-0.031
 brothers-0.030
Token her
Token ID607
Feature activation-1.66e+00

Pos logit contributions
oyd+0.028
 pistols+0.023
 neither+0.021
 sometimes+0.021
 brothers+0.020
Neg logit contributions
igsaw-0.047
hy-0.046
 click-0.044
 doorway-0.044
lic-0.043
Token l
Token ID300
Feature activation+2.53e+00

Pos logit contributions
igsaw+0.053
hy+0.052
 brand+0.050
 doorway+0.047
 click+0.047
Neg logit contributions
oyd-0.065
 pistols-0.055
 neither-0.051
 brothers-0.051
ball-0.050
Tokenoll
Token ID692
Feature activation+4.64e+00

Pos logit contributions
oyd+0.079
 pistols+0.065
 neither+0.059
 brothers+0.059
 sometimes+0.058
Neg logit contributions
igsaw-0.100
hy-0.098
 brand-0.093
 doorway-0.092
 click-0.091
Tokenipop
Token ID42800
Feature activation+4.55e+00

Pos logit contributions
oyd+0.184
 pistols+0.152
 neither+0.143
 sometimes+0.141
 brothers+0.141
Neg logit contributions
igsaw-0.144
hy-0.142
 brand-0.133
endant-0.126
 doorway-0.124
Token and
Token ID290
Feature activation+9.24e-02

Pos logit contributions
oyd+0.119
 pistols+0.095
 neither+0.083
 brothers+0.083
 sometimes+0.082
Neg logit contributions
igsaw-0.203
hy-0.198
 brand-0.189
 doorway-0.187
 click-0.187
Token had
Token ID550
Feature activation+3.00e-01

Pos logit contributions
oyd+0.003
 pistols+0.002
 sometimes+0.002
 neither+0.002
 brothers+0.002
Neg logit contributions
igsaw-0.004
hy-0.004
 brand-0.004
 click-0.004
 doorway-0.004
Token so
Token ID523
Feature activation-2.18e+00

Pos logit contributions
oyd+0.007
 also+0.007
 sometimes+0.007
 friends+0.006
 pistols+0.006
Neg logit contributions
igsaw-0.014
hy-0.013
 click-0.013
col-0.013
icol-0.012
Token much
Token ID881
Feature activation-7.81e-01

Pos logit contributions
igsaw+0.055
hy+0.053
 brand+0.048
 click+0.045
col+0.045
Neg logit contributions
oyd-0.101
 pistols-0.089
 brothers-0.082
 neither-0.082
ball-0.081
Token fun
Token ID1257
Feature activation-1.18e+00

Pos logit contributions
igsaw+0.028
hy+0.027
 doorway+0.025
 click+0.025
 brand+0.025
Neg logit contributions
oyd-0.027
 pistols-0.024
 brothers-0.022
 sometimes-0.022
 friends-0.022
Token
Token ID198
Feature activation-6.11e-01

Pos logit contributions
igsaw+0.022
 brand+0.022
hy+0.022
lic+0.021
col+0.021
Neg logit contributions
oyd-0.021
 pistols-0.018
 neither-0.017
 forth-0.016
 goalie-0.015
Token"
Token ID1
Feature activation-8.95e-01

Pos logit contributions
igsaw+0.024
hy+0.023
 click+0.023
 brand+0.023
col+0.022
Neg logit contributions
oyd-0.019
 pistols-0.015
 neither-0.014
 brothers-0.013
 humans-0.013
TokenBy
Token ID3886
Feature activation+8.59e-01

Pos logit contributions
igsaw+0.034
hy+0.034
 brand+0.032
 click+0.032
 doorway+0.032
Neg logit contributions
oyd-0.028
 pistols-0.024
 brothers-0.021
 neither-0.021
 sometimes-0.020
Tokene
Token ID68
Feature activation+4.21e-01

Pos logit contributions
oyd+0.032
 pistols+0.029
 neither+0.025
ball+0.025
 horses+0.024
Neg logit contributions
igsaw-0.026
 click-0.025
hy-0.025
col-0.024
 doorway-0.024
Token,
Token ID11
Feature activation+2.60e-01

Pos logit contributions
oyd+0.013
 pistols+0.010
 neither+0.010
 brothers+0.009
 sometimes+0.009
Neg logit contributions
igsaw-0.017
hy-0.016
 brand-0.016
 doorway-0.016
 click-0.015
Token fish
Token ID5916
Feature activation+4.87e+00

Pos logit contributions
oyd+0.007
 pistols+0.006
Rock+0.005
ained+0.005
 neither+0.005
Neg logit contributions
igsaw-0.011
hy-0.011
 brand-0.010
 click-0.010
 doorway-0.010
Token!"
Token ID2474
Feature activation+4.09e-01

Pos logit contributions
oyd+0.164
 pistols+0.137
 neither+0.126
 brothers+0.126
 sometimes+0.125
Neg logit contributions
igsaw-0.182
hy-0.175
 brand-0.167
 doorway-0.165
 click-0.163
Token his
Token ID465
Feature activation+8.89e-01

Pos logit contributions
oyd+0.009
 pistols+0.008
 brothers+0.007
 neither+0.007
 sometimes+0.007
Neg logit contributions
igsaw-0.020
hy-0.019
 doorway-0.018
 brand-0.018
col-0.018
Token dad
Token ID9955
Feature activation-5.55e-01

Pos logit contributions
oyd+0.021
 pistols+0.016
 sometimes+0.015
 neither+0.014
Rock+0.014
Neg logit contributions
igsaw-0.041
hy-0.041
 brand-0.040
 doorway-0.039
 click-0.038
Token said
Token ID531
Feature activation+1.52e+00

Pos logit contributions
igsaw+0.022
hy+0.021
 brand+0.020
 doorway+0.020
 click+0.020
Neg logit contributions
oyd-0.018
 pistols-0.015
 brothers-0.013
 neither-0.013
 sometimes-0.013
Token,
Token ID11
Feature activation+3.26e-01

Pos logit contributions
oyd+0.030
 neither+0.027
 on+0.026
 friends+0.024
 at+0.024
Neg logit contributions
igsaw-0.079
 doorway-0.072
 brand-0.071
endant-0.071
 disl-0.070
TokenWe
Token ID1135
Feature activation-1.74e+00

Pos logit contributions
igsaw+0.018
hy+0.017
 brand+0.016
 doorway+0.016
 click+0.016
Neg logit contributions
oyd-0.011
 pistols-0.008
 sometimes-0.008
 neither-0.008
ball-0.007
Token can
Token ID460
Feature activation-4.48e-01

Pos logit contributions
hy+0.052
igsaw+0.049
 click+0.049
col+0.048
 brand+0.046
Neg logit contributions
oyd-0.067
 pistols-0.057
ained-0.053
 neither-0.052
 mosquitoes-0.051
Token use
Token ID779
Feature activation+1.91e+00

Pos logit contributions
 click+0.012
hy+0.012
igsaw+0.011
 brand+0.011
 throne+0.011
Neg logit contributions
oyd-0.018
 pistols-0.017
ained-0.016
 mosquitoes-0.015
dd-0.015
Token our
Token ID674
Feature activation+1.51e-01

Pos logit contributions
oyd+0.050
 pistols+0.046
 sometimes+0.041
 brothers+0.040
 neither+0.039
Neg logit contributions
igsaw-0.081
hy-0.080
 doorway-0.076
lic-0.076
 click-0.075
Token bikes
Token ID16715
Feature activation-1.02e+00

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.004
 friends+0.004
 sometimes+0.004
Neg logit contributions
igsaw-0.006
hy-0.005
lic-0.005
 doorway-0.005
 click-0.005
Token as
Token ID355
Feature activation+4.56e+00

Pos logit contributions
igsaw+0.045
hy+0.043
 doorway+0.042
 brand+0.041
 click+0.041
Neg logit contributions
oyd-0.027
 pistols-0.023
 sometimes-0.022
 neither-0.021
 brothers-0.020
Token rockets
Token ID20293
Feature activation+1.81e+00

Pos logit contributions
oyd+0.111
 pistols+0.101
 brothers+0.091
 friends+0.089
 sometimes+0.089
Neg logit contributions
igsaw-0.205
hy-0.200
 doorway-0.195
 click-0.191
 brand-0.186
Token and
Token ID290
Feature activation+3.68e-02

Pos logit contributions
oyd+0.047
 pistols+0.038
 sometimes+0.037
 neither+0.035
ball+0.034
Neg logit contributions
igsaw-0.083
hy-0.077
 brand-0.076
 doorway-0.076
icol-0.074
Token our
Token ID674
Feature activation+1.41e-01

Pos logit contributions
oyd+0.001
 pistols+0.001
 sometimes+0.001
 brothers+0.001
 horses+0.001
Neg logit contributions
igsaw-0.001
hy-0.001
lic-0.001
 doorway-0.001
 brand-0.001
Token helmets
Token ID28359
Feature activation-4.45e-01

Pos logit contributions
oyd+0.004
 pistols+0.004
 brothers+0.003
 friends+0.003
 horses+0.003
Neg logit contributions
igsaw-0.006
lic-0.005
hy-0.005
 doorway-0.005
 brand-0.005
Token as
Token ID355
Feature activation+4.55e+00

Pos logit contributions
igsaw+0.022
 doorway+0.019
 brand+0.018
lic+0.018
hy+0.018
Neg logit contributions
oyd-0.011
 sometimes-0.010
 pistols-0.009
 neither-0.009
ball-0.009
Token funny
Token ID8258
Feature activation+3.41e-01

Pos logit contributions
igsaw+0.079
hy+0.076
 doorway+0.076
keeping+0.073
cake+0.072
Neg logit contributions
oyd-0.065
 also-0.057
 sometimes-0.055
 brothers-0.053
 pistols-0.053
Token that
Token ID326
Feature activation+1.80e+00

Pos logit contributions
oyd+0.009
 pistols+0.008
 sometimes+0.007
 neither+0.007
 brothers+0.007
Neg logit contributions
igsaw-0.015
hy-0.014
 doorway-0.014
 click-0.013
 brand-0.013
Token the
Token ID262
Feature activation-1.06e-01

Pos logit contributions
oyd+0.050
 pistols+0.039
 neither+0.035
 brothers+0.034
ball+0.034
Neg logit contributions
igsaw-0.077
hy-0.076
 brand-0.073
 click-0.072
 doorway-0.071
Token shadow
Token ID9082
Feature activation+4.00e-01

Pos logit contributions
igsaw+0.004
hy+0.004
 brand+0.004
 click+0.004
 doorway+0.004
Neg logit contributions
oyd-0.004
 pistols-0.003
 brothers-0.003
 neither-0.003
 sometimes-0.003
Token was
Token ID373
Feature activation+8.87e-01

Pos logit contributions
oyd+0.011
 pistols+0.009
 sometimes+0.008
 neither+0.008
 brothers+0.008
Neg logit contributions
igsaw-0.018
hy-0.017
 brand-0.016
 doorway-0.016
 click-0.016
Token as
Token ID355
Feature activation+4.51e+00

Pos logit contributions
oyd+0.029
 pistols+0.025
 brothers+0.023
 sometimes+0.022
 neither+0.022
Neg logit contributions
igsaw-0.034
hy-0.032
 click-0.030
 brand-0.030
 doorway-0.030
Token tall
Token ID7331
Feature activation-2.16e+00

Pos logit contributions
oyd+0.156
 pistols+0.137
 brothers+0.130
 sometimes+0.127
 neither+0.126
Neg logit contributions
igsaw-0.165
hy-0.157
 doorway-0.148
 click-0.143
 brand-0.142
Token as
Token ID355
Feature activation+4.51e+00

Pos logit contributions
igsaw+0.103
 doorway+0.095
hy+0.092
 brand+0.091
aney+0.091
Neg logit contributions
oyd-0.050
 sometimes-0.045
 pistols-0.044
 on-0.041
 at-0.040
Token he
Token ID339
Feature activation+2.32e+00

Pos logit contributions
oyd+0.091
 pistols+0.081
 sometimes+0.073
 brothers+0.073
 neither+0.072
Neg logit contributions
igsaw-0.226
hy-0.216
 doorway-0.210
 click-0.204
cake-0.201
Token was
Token ID373
Feature activation+9.24e-01

Pos logit contributions
oyd+0.052
 sometimes+0.037
 pistols+0.036
 also+0.033
ball+0.033
Neg logit contributions
igsaw-0.121
hy-0.107
 doorway-0.106
 brand-0.105
 click-0.104
Token!
Token ID0
Feature activation-7.57e-01

Pos logit contributions
oyd+0.026
 pistols+0.021
 sometimes+0.020
 on+0.020
 also+0.020
Neg logit contributions
igsaw-0.041
hy-0.037
 click-0.035
 brand-0.035
icol-0.035
Token.
Token ID13
Feature activation+1.25e-01

Pos logit contributions
igsaw+0.190
hy+0.177
 click+0.172
icol+0.171
lic+0.171
Neg logit contributions
oyd-0.085
 neither-0.072
 pistols-0.071
 sometimes-0.071
 friends-0.069
Token When
Token ID1649
Feature activation+2.44e-02

Pos logit contributions
oyd+0.003
 sometimes+0.002
 neither+0.002
 each+0.002
 brothers+0.002
Neg logit contributions
hy-0.006
endant-0.005
 click-0.005
 brand-0.005
col-0.005
Token they
Token ID484
Feature activation+8.60e-01

Pos logit contributions
oyd+0.001
 pistols+0.000
 brothers+0.000
 neither+0.000
 sometimes+0.000
Neg logit contributions
igsaw-0.001
hy-0.001
 brand-0.001
 doorway-0.001
 click-0.001
Token arrived
Token ID5284
Feature activation-2.38e+00

Pos logit contributions
oyd+0.025
 pistols+0.020
 also+0.020
 neither+0.020
 sometimes+0.020
Neg logit contributions
igsaw-0.036
hy-0.034
 click-0.032
 brand-0.032
icol-0.032
Token,
Token ID11
Feature activation+4.81e-01

Pos logit contributions
igsaw+0.102
hy+0.100
 brand+0.095
 click+0.094
 doorway+0.094
Neg logit contributions
oyd-0.066
 pistols-0.053
 neither-0.048
 brothers-0.048
 sometimes-0.047
Token the
Token ID262
Feature activation-4.98e-02

Pos logit contributions
oyd+0.014
 pistols+0.011
Rock+0.010
 brothers+0.010
ball+0.010
Neg logit contributions
igsaw-0.020
 brand-0.019
hy-0.019
 doorway-0.019
 click-0.019
Token store
Token ID3650
Feature activation-3.92e+00

Pos logit contributions
igsaw+0.002
hy+0.002
 brand+0.002
 doorway+0.002
 click+0.002
Neg logit contributions
oyd-0.002
 pistols-0.001
 neither-0.001
 brothers-0.001
 sometimes-0.001
Token was
Token ID373
Feature activation+1.01e+00

Pos logit contributions
igsaw+0.168
hy+0.157
 brand+0.151
col+0.149
 click+0.148
Neg logit contributions
oyd-0.112
 pistols-0.090
 neither-0.087
 brothers-0.085
 sometimes-0.083
Token full
Token ID1336
Feature activation-2.00e+00

Pos logit contributions
oyd+0.033
 also+0.030
 on+0.029
 sometimes+0.028
 brothers+0.028
Neg logit contributions
igsaw-0.039
hy-0.035
col-0.034
 disl-0.034
icol-0.033
Token of
Token ID286
Feature activation-4.83e-01

Pos logit contributions
igsaw+0.100
 disl+0.094
icol+0.092
hy+0.090
 majesty+0.090
Neg logit contributions
 sometimes-0.038
oyd-0.038
 on-0.037
 at-0.036
 brothers-0.035
Token different
Token ID1180
Feature activation-9.82e-01

Pos logit contributions
igsaw+0.016
hy+0.016
 doorway+0.014
lic+0.014
col+0.014
Neg logit contributions
oyd-0.018
 pistols-0.016
 brothers-0.015
 sometimes-0.014
 friends-0.014
Token.
Token ID13
Feature activation-3.65e-01

Pos logit contributions
oyd+0.066
 pistols+0.055
 neither+0.049
 brothers+0.049
 sometimes+0.049
Neg logit contributions
igsaw-0.101
hy-0.097
 brand-0.093
 click-0.091
 doorway-0.090
Token He
Token ID679
Feature activation+7.61e-01

Pos logit contributions
igsaw+0.018
col+0.017
hy+0.017
quet+0.017
endant+0.017
Neg logit contributions
oyd-0.007
 sometimes-0.006
 brothers-0.006
 pistols-0.006
 Spot-0.005
Token did
Token ID750
Feature activation+5.49e-01

Pos logit contributions
oyd+0.025
 pistols+0.021
 neither+0.019
 brothers+0.019
ball+0.019
Neg logit contributions
hy-0.027
igsaw-0.027
 brand-0.026
 doorway-0.025
 click-0.025
Token not
Token ID407
Feature activation+1.38e+00

Pos logit contributions
oyd+0.010
 pistols+0.007
 neither+0.007
 sometimes+0.007
 also+0.006
Neg logit contributions
igsaw-0.029
hy-0.028
 doorway-0.028
col-0.027
 brand-0.027
Token know
Token ID760
Feature activation-1.19e+00

Pos logit contributions
oyd+0.035
 sometimes+0.034
 also+0.032
 each+0.030
 on+0.030
Neg logit contributions
igsaw-0.063
 brand-0.060
 doorway-0.060
icol-0.058
endant-0.056
Token that
Token ID326
Feature activation+1.63e+00

Pos logit contributions
igsaw+0.048
hy+0.046
 doorway+0.043
 brand+0.043
 click+0.043
Neg logit contributions
oyd-0.037
 pistols-0.032
 sometimes-0.030
 neither-0.029
 brothers-0.029
Token Lily
Token ID20037
Feature activation-1.77e+00

Pos logit contributions
oyd+0.042
 pistols+0.036
 sometimes+0.034
 neither+0.033
 brothers+0.032
Neg logit contributions
igsaw-0.073
hy-0.070
 brand-0.067
 doorway-0.066
 click-0.066
Token was
Token ID373
Feature activation+4.18e-01

Pos logit contributions
hy+0.074
igsaw+0.072
 brand+0.068
 doorway+0.068
col+0.067
Neg logit contributions
oyd-0.049
 pistols-0.040
 brothers-0.035
 neither-0.035
Rock-0.034
Token in
Token ID287
Feature activation+1.82e-01

Pos logit contributions
oyd+0.014
 pistols+0.011
 neither+0.010
ball+0.010
 sometimes+0.010
Neg logit contributions
igsaw-0.015
hy-0.015
 brand-0.015
 doorway-0.014
 click-0.014
Token the
Token ID262
Feature activation-3.31e-01

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.004
 neither+0.004
 sometimes+0.004
Neg logit contributions
igsaw-0.008
hy-0.008
 brand-0.007
 click-0.007
col-0.007
Token barn
Token ID25203
Feature activation-2.00e+00

Pos logit contributions
igsaw+0.014
hy+0.014
 brand+0.013
col+0.013
 click+0.013
Neg logit contributions
oyd-0.009
 pistols-0.008
 brothers-0.007
 sometimes-0.007
 neither-0.007
Token the
Token ID262
Feature activation-1.15e-01

Pos logit contributions
oyd+0.034
 pistols+0.027
 sometimes+0.025
 brothers+0.025
 neither+0.023
Neg logit contributions
hy-0.073
igsaw-0.073
 brand-0.070
lic-0.070
des-0.069
Token grass
Token ID8701
Feature activation-1.22e+00

Pos logit contributions
igsaw+0.004
hy+0.004
lic+0.004
 brand+0.004
 click+0.004
Neg logit contributions
oyd-0.004
 pistols-0.003
ball-0.003
 brothers-0.003
 sometimes-0.003
Token.
Token ID13
Feature activation+4.78e-02

Pos logit contributions
igsaw+0.058
hy+0.054
 brand+0.053
 click+0.053
 disl+0.052
Neg logit contributions
oyd-0.030
 sometimes-0.023
 pistols-0.022
ball-0.022
 neither-0.022
Token She
Token ID1375
Feature activation-2.03e+00

Pos logit contributions
oyd+0.002
 pistols+0.001
 brothers+0.001
 sometimes+0.001
 neither+0.001
Neg logit contributions
igsaw-0.002
hy-0.002
 brand-0.002
 click-0.002
 doorway-0.002
Token went
Token ID1816
Feature activation-1.35e+00

Pos logit contributions
igsaw+0.083
hy+0.079
 brand+0.075
 click+0.074
 doorway+0.074
Neg logit contributions
oyd-0.063
 pistols-0.051
 brothers-0.047
 sometimes-0.047
 neither-0.047
Token over
Token ID625
Feature activation+1.04e+00

Pos logit contributions
igsaw+0.060
hy+0.059
 click+0.056
 brand+0.054
 disl+0.054
Neg logit contributions
oyd-0.032
 pistols-0.028
 on-0.028
 sometimes-0.026
 also-0.026
Token to
Token ID284
Feature activation-8.57e-01

Pos logit contributions
oyd+0.035
 pistols+0.030
 neither+0.027
 brothers+0.027
 sometimes+0.026
Neg logit contributions
igsaw-0.039
hy-0.037
 brand-0.036
 doorway-0.035
 click-0.035
Token it
Token ID340
Feature activation+5.78e-01

Pos logit contributions
igsaw+0.036
hy+0.035
 brand+0.033
lic+0.033
 doorway+0.032
Neg logit contributions
oyd-0.025
 pistols-0.020
 sometimes-0.019
 also-0.019
 brothers-0.019
Token and
Token ID290
Feature activation-4.38e-02

Pos logit contributions
oyd+0.009
 sometimes+0.007
 on+0.006
 also+0.006
ball+0.006
Neg logit contributions
hy-0.031
igsaw-0.031
lic-0.029
icol-0.029
 disl-0.029
Token picked
Token ID6497
Feature activation-8.92e-01

Pos logit contributions
igsaw+0.002
hy+0.002
 brand+0.002
 click+0.002
 doorway+0.002
Neg logit contributions
oyd-0.001
 pistols-0.001
 sometimes-0.001
 neither-0.001
 also-0.001
Token it
Token ID340
Feature activation+5.93e-01

Pos logit contributions
igsaw+0.038
hy+0.037
 brand+0.036
col+0.035
 click+0.035
Neg logit contributions
oyd-0.025
 pistols-0.020
 brothers-0.017
 neither-0.017
 sometimes-0.017
Token enjoyed
Token ID8359
Feature activation+2.53e+00

Pos logit contributions
igsaw+0.014
hy+0.013
 brand+0.012
 doorway+0.012
icol+0.012
Neg logit contributions
oyd-0.008
 brothers-0.007
 sometimes-0.007
 also-0.007
 pistols-0.006
Token the
Token ID262
Feature activation-3.03e-01

Pos logit contributions
oyd+0.073
 pistols+0.059
 neither+0.052
 brothers+0.051
 sometimes+0.050
Neg logit contributions
igsaw-0.107
hy-0.104
 brand-0.101
 click-0.099
 doorway-0.098
Token warm
Token ID5814
Feature activation-1.45e+00

Pos logit contributions
igsaw+0.011
hy+0.011
 doorway+0.010
endant+0.010
col+0.010
Neg logit contributions
oyd-0.010
 pistols-0.008
 brothers-0.008
 friends-0.008
 sometimes-0.008
Token cookies
Token ID14746
Feature activation+1.51e+00

Pos logit contributions
igsaw+0.056
hy+0.054
 brand+0.052
 doorway+0.051
col+0.050
Neg logit contributions
oyd-0.046
 pistols-0.039
 sometimes-0.036
 brothers-0.036
 neither-0.035
Token together
Token ID1978
Feature activation-6.88e-01

Pos logit contributions
oyd+0.029
 on+0.023
 sometimes+0.023
 pistols+0.022
ball+0.021
Neg logit contributions
igsaw-0.076
hy-0.074
 doorway-0.071
 disl-0.070
icol-0.069
Token,
Token ID11
Feature activation+4.58e-01

Pos logit contributions
igsaw+0.032
hy+0.031
 brand+0.029
 doorway+0.029
 click+0.029
Neg logit contributions
oyd-0.017
 pistols-0.013
 sometimes-0.012
 neither-0.012
 brothers-0.012
Token and
Token ID290
Feature activation+1.50e-01

Pos logit contributions
oyd+0.011
 pistols+0.010
 sometimes+0.010
 also+0.009
 brothers+0.009
Neg logit contributions
igsaw-0.022
hy-0.020
 doorway-0.019
lic-0.019
col-0.018
Token Tim
Token ID5045
Feature activation+4.06e-01

Pos logit contributions
oyd+0.005
 pistols+0.004
ball+0.003
Rock+0.003
 brothers+0.003
Neg logit contributions
hy-0.006
igsaw-0.006
 brand-0.006
 click-0.006
col-0.006
Tokenmy
Token ID1820
Feature activation+1.08e+00

Pos logit contributions
oyd+0.011
 pistols+0.009
 brothers+0.008
 neither+0.008
 sometimes+0.008
Neg logit contributions
igsaw-0.018
hy-0.017
 brand-0.016
 doorway-0.016
 click-0.016
Token's
Token ID338
Feature activation+2.48e+00

Pos logit contributions
oyd+0.030
 pistols+0.023
 brothers+0.022
 sometimes+0.022
 neither+0.021
Neg logit contributions
igsaw-0.048
hy-0.046
 doorway-0.044
 brand-0.043
 click-0.043
Token toy
Token ID13373
Feature activation+2.36e+00

Pos logit contributions
oyd+0.050
 pistols+0.036
 neither+0.030
ball+0.029
 sometimes+0.029
Neg logit contributions
igsaw-0.125
hy-0.123
 brand-0.122
 doorway-0.118
 click-0.118
Token little
Token ID1310
Feature activation-2.01e+00

Pos logit contributions
igsaw+0.003
hy+0.003
 brand+0.003
 click+0.003
 doorway+0.002
Neg logit contributions
oyd-0.003
 pistols-0.003
 neither-0.003
 sometimes-0.003
ained-0.003
Token girl
Token ID2576
Feature activation-2.37e+00

Pos logit contributions
igsaw+0.080
hy+0.078
 brand+0.074
 click+0.073
 doorway+0.073
Neg logit contributions
oyd-0.062
 pistols-0.051
 neither-0.046
 sometimes-0.045
 brothers-0.045
Token named
Token ID3706
Feature activation-2.03e+00

Pos logit contributions
igsaw+0.093
hy+0.089
 brand+0.086
 click+0.084
 doorway+0.084
Neg logit contributions
oyd-0.075
 pistols-0.062
 brothers-0.057
 sometimes-0.057
 neither-0.057
Token Lily
Token ID20037
Feature activation-1.45e+00

Pos logit contributions
hy+0.089
igsaw+0.089
 brand+0.088
 click+0.085
icol+0.084
Neg logit contributions
oyd-0.045
 sometimes-0.041
 also-0.038
 neither-0.037
 brother-0.036
Token who
Token ID508
Feature activation+8.03e-01

Pos logit contributions
hy+0.050
igsaw+0.050
 brand+0.047
col+0.046
 doorway+0.046
Neg logit contributions
oyd-0.051
 pistols-0.043
 neither-0.040
 brothers-0.039
Rock-0.037
Token loved
Token ID6151
Feature activation+7.98e-01

Pos logit contributions
oyd+0.022
 pistols+0.018
 brothers+0.016
 sometimes+0.016
 neither+0.015
Neg logit contributions
igsaw-0.035
hy-0.034
 brand-0.033
 doorway-0.032
 click-0.032
Token to
Token ID284
Feature activation-8.96e-01

Pos logit contributions
oyd+0.028
 pistols+0.024
 neither+0.022
 brothers+0.022
 sometimes+0.021
Neg logit contributions
igsaw-0.028
hy-0.028
 brand-0.026
 doorway-0.026
 click-0.026
Token explore
Token ID7301
Feature activation-1.30e+00

Pos logit contributions
igsaw+0.039
 doorway+0.039
lic+0.038
hy+0.038
endant+0.037
Neg logit contributions
oyd-0.021
 sometimes-0.017
 pistols-0.017
 brothers-0.017
 friends-0.017
Token.
Token ID13
Feature activation-9.57e-02

Pos logit contributions
igsaw+0.049
hy+0.047
 brand+0.046
 doorway+0.045
 click+0.044
Neg logit contributions
oyd-0.044
 pistols-0.035
 neither-0.034
 brothers-0.032
ball-0.031
Token One
Token ID1881
Feature activation+3.54e+00

Pos logit contributions
igsaw+0.004
hy+0.004
 brand+0.004
col+0.004
 doorway+0.004
Neg logit contributions
oyd-0.003
 sometimes-0.002
 pistols-0.002
 also-0.002
 brother-0.002
Token day
Token ID1110
Feature activation+8.94e-01

Pos logit contributions
oyd+0.105
 pistols+0.089
 neither+0.078
 brothers+0.076
eper+0.074
Neg logit contributions
igsaw-0.147
hy-0.143
 doorway-0.135
 brand-0.134
col-0.132
Token She
Token ID1375
Feature activation-2.30e+00

Pos logit contributions
igsaw+0.008
hy+0.007
 doorway+0.007
 brand+0.007
 click+0.007
Neg logit contributions
oyd-0.014
 pistols-0.012
ball-0.011
 neither-0.011
 forth-0.011
Token likes
Token ID7832
Feature activation+1.43e+00

Pos logit contributions
hy+0.072
 brand+0.068
igsaw+0.065
col+0.065
 click+0.063
Neg logit contributions
oyd-0.086
 pistols-0.073
 neither-0.067
ball-0.067
ained-0.065
Token to
Token ID284
Feature activation-1.05e+00

Pos logit contributions
oyd+0.056
 pistols+0.044
 neither+0.043
ball+0.041
 brothers+0.041
Neg logit contributions
igsaw-0.045
hy-0.044
 brand-0.043
 click-0.041
col-0.041
Token sing
Token ID1702
Feature activation-8.95e-01

Pos logit contributions
igsaw+0.041
hy+0.039
 doorway+0.037
 brand+0.037
col+0.037
Neg logit contributions
oyd-0.034
 pistols-0.028
 sometimes-0.026
 brothers-0.025
 neither-0.025
Token songs
Token ID7259
Feature activation-1.27e+00

Pos logit contributions
igsaw+0.040
hy+0.039
 doorway+0.037
 brand+0.037
col+0.036
Neg logit contributions
oyd-0.023
 pistols-0.019
 sometimes-0.017
 brothers-0.017
 neither-0.016
Token with
Token ID351
Feature activation-4.23e-01

Pos logit contributions
igsaw+0.058
hy+0.056
 brand+0.054
 doorway+0.053
col+0.053
Neg logit contributions
oyd-0.032
 pistols-0.025
 sometimes-0.023
 neither-0.022
 brothers-0.022
Token her
Token ID607
Feature activation-1.60e+00

Pos logit contributions
igsaw+0.014
hy+0.014
 brand+0.014
col+0.013
 click+0.013
Neg logit contributions
oyd-0.015
 pistols-0.013
 neither-0.012
ained-0.011
ball-0.011
Token guitar
Token ID10047
Feature activation+1.52e+00

Pos logit contributions
 brand+0.064
hy+0.063
igsaw+0.062
 click+0.060
 doorway+0.058
Neg logit contributions
oyd-0.043
dd-0.033
ained-0.033
 pistols-0.032
eper-0.031
Token.
Token ID13
Feature activation-2.24e-01

Pos logit contributions
oyd+0.037
 pistols+0.030
 sometimes+0.030
 brothers+0.029
 friends+0.028
Neg logit contributions
igsaw-0.071
hy-0.067
 doorway-0.064
des-0.064
col-0.064
Token Sometimes
Token ID8975
Feature activation+2.84e-01

Pos logit contributions
igsaw+0.006
hy+0.006
 doorway+0.005
 Acc+0.005
 sill+0.005
Neg logit contributions
oyd-0.009
 pistols-0.009
 forth-0.008
ball-0.008
 neither-0.008
Token she
Token ID673
Feature activation-7.99e-01

Pos logit contributions
oyd+0.013
 pistols+0.010
ball+0.010
ained+0.010
 goalie+0.010
Neg logit contributions
hy-0.007
 click-0.007
igsaw-0.007
 brand-0.006
 doorway-0.006
Token hid
Token ID24519
Feature activation+1.42e+00

Pos logit contributions
oyd+0.039
 pistols+0.032
 neither+0.029
ained+0.028
 brothers+0.028
Neg logit contributions
hy-0.039
 brand-0.038
igsaw-0.036
 doorway-0.036
col-0.036
Token the
Token ID262
Feature activation-8.19e-02

Pos logit contributions
oyd+0.043
 pistols+0.033
 brothers+0.030
 neither+0.029
ball+0.029
Neg logit contributions
igsaw-0.058
hy-0.056
 brand-0.056
 doorway-0.055
 click-0.054
Token coin
Token ID10752
Feature activation+3.29e+00

Pos logit contributions
igsaw+0.003
hy+0.003
 brand+0.002
 click+0.002
col+0.002
Neg logit contributions
oyd-0.003
 pistols-0.003
 brothers-0.003
 sometimes-0.002
 neither-0.002
Token in
Token ID287
Feature activation+3.84e-01

Pos logit contributions
oyd+0.082
 pistols+0.063
 brothers+0.055
 neither+0.053
 humans+0.050
Neg logit contributions
igsaw-0.149
hy-0.147
 brand-0.142
 doorway-0.140
 click-0.139
Token his
Token ID465
Feature activation+8.49e-01

Pos logit contributions
oyd+0.006
 pistols+0.006
 sometimes+0.005
 brothers+0.005
 each+0.004
Neg logit contributions
hy-0.020
igsaw-0.020
lic-0.020
endant-0.019
cake-0.019
Token pocket
Token ID10000
Feature activation-1.23e+00

Pos logit contributions
oyd+0.028
 pistols+0.025
 friends+0.024
 brothers+0.024
 brother+0.024
Neg logit contributions
igsaw-0.032
lic-0.032
hy-0.030
col-0.030
cake-0.030
Token and
Token ID290
Feature activation+9.36e-02

Pos logit contributions
igsaw+0.067
 disl+0.063
icol+0.062
hy+0.062
 click+0.061
Neg logit contributions
oyd-0.019
 sometimes-0.017
 pistols-0.016
 on-0.016
 also-0.016
Token didn
Token ID1422
Feature activation+9.11e-01

Pos logit contributions
oyd+0.003
 pistols+0.002
 neither+0.002
 sometimes+0.002
 brothers+0.002
Neg logit contributions
igsaw-0.004
hy-0.004
 brand-0.003
 doorway-0.003
col-0.003
Token't
Token ID470
Feature activation+2.24e+00

Pos logit contributions
oyd+0.035
 pistols+0.031
 forth+0.029
 brothers+0.028
 horses+0.028
Neg logit contributions
igsaw-0.030
hy-0.029
 click-0.027
 brand-0.027
bell-0.025
Token say
Token ID910
Feature activation+3.89e-01

Pos logit contributions
oyd+0.075
 sometimes+0.066
 pistols+0.064
 also+0.063
 friends+0.062
Neg logit contributions
igsaw-0.082
 brand-0.077
 doorway-0.076
hy-0.075
icol-0.072
Token anything
Token ID1997
Feature activation+6.37e-01

Pos logit contributions
oyd+0.009
 neither+0.009
 on+0.009
 at+0.008
 sometimes+0.008
Neg logit contributions
igsaw-0.017
 doorway-0.017
col-0.016
hy-0.016
keeping-0.016
Token "
Token ID366
Feature activation-5.51e-01

Pos logit contributions
oyd+0.013
 pistols+0.011
 sometimes+0.010
 neither+0.010
 brothers+0.010
Neg logit contributions
igsaw-0.018
hy-0.017
 doorway-0.017
col-0.016
 brand-0.016
TokenSarah
Token ID29284
Feature activation-3.77e+00

Pos logit contributions
igsaw+0.024
hy+0.023
 brand+0.022
 doorway+0.022
 click+0.021
Neg logit contributions
oyd-0.016
 pistols-0.012
 sometimes-0.011
 neither-0.011
ball-0.011
Token,
Token ID11
Feature activation+4.18e-01

Pos logit contributions
igsaw+0.182
 doorway+0.172
hy+0.168
 click+0.168
des+0.167
Neg logit contributions
oyd-0.083
 brother-0.067
 sometimes-0.063
 pistols-0.063
 brothers-0.061
Token your
Token ID534
Feature activation-8.78e-01

Pos logit contributions
 sometimes+0.011
oyd+0.011
 also+0.010
 pistols+0.010
 on+0.009
Neg logit contributions
igsaw-0.018
hy-0.017
 doorway-0.017
lic-0.016
icol-0.015
Token doll
Token ID3654
Feature activation-3.27e-01

Pos logit contributions
igsaw+0.039
lic+0.036
des+0.036
hy+0.036
 click+0.035
Neg logit contributions
oyd-0.021
 friends-0.020
 brother-0.020
 mom-0.019
 best-0.018
Token is
Token ID318
Feature activation-3.92e-01

Pos logit contributions
igsaw+0.019
des+0.017
 doorway+0.017
 click+0.016
leaf+0.016
Neg logit contributions
 brother-0.005
 friends-0.005
oyd-0.004
 sometimes-0.004
 still-0.004
Token so
Token ID523
Feature activation-2.32e+00

Pos logit contributions
igsaw+0.016
hy+0.015
 click+0.014
des+0.014
 doorway+0.014
Neg logit contributions
oyd-0.011
 also-0.011
 sometimes-0.010
 on-0.010
 friends-0.010
Token pretty
Token ID2495
Feature activation-3.21e+00

Pos logit contributions
igsaw+0.101
 doorway+0.100
icol+0.097
 aback+0.096
 click+0.095
Neg logit contributions
 also-0.060
 friends-0.058
 mom-0.052
 still-0.051
 on-0.050
Token.
Token ID13
Feature activation-2.75e-01

Pos logit contributions
igsaw+0.114
hy+0.111
 doorway+0.104
 brand+0.103
 click+0.103
Neg logit contributions
oyd-0.110
 pistols-0.095
 sometimes-0.091
 brothers-0.089
 neither-0.088
Token Can
Token ID1680
Feature activation-2.78e+00

Pos logit contributions
igsaw+0.009
hy+0.009
 doorway+0.008
 brand+0.008
col+0.008
Neg logit contributions
oyd-0.010
 sometimes-0.009
 pistols-0.008
 brothers-0.008
 also-0.008
Token I
Token ID314
Feature activation-1.67e+00

Pos logit contributions
hy+0.077
igsaw+0.077
 brand+0.071
 click+0.070
col+0.067
Neg logit contributions
oyd-0.119
 pistols-0.103
 brothers-0.096
ball-0.094
 neither-0.093
Token his
Token ID465
Feature activation+7.43e-01

Pos logit contributions
oyd+0.008
 pistols+0.006
 neither+0.006
ball+0.006
 brothers+0.006
Neg logit contributions
igsaw-0.013
hy-0.013
 brand-0.012
 doorway-0.012
 click-0.012
Token bag
Token ID6131
Feature activation-1.15e+00

Pos logit contributions
oyd+0.019
 pistols+0.016
 brothers+0.014
 friends+0.014
 brother+0.013
Neg logit contributions
igsaw-0.034
hy-0.032
lic-0.031
col-0.031
 click-0.031
Token.
Token ID13
Feature activation-4.93e-01

Pos logit contributions
igsaw+0.059
icol+0.055
 disl+0.055
 brand+0.053
 click+0.053
Neg logit contributions
oyd-0.022
 sometimes-0.020
 on-0.019
 pistols-0.019
 brothers-0.019
Token They
Token ID1119
Feature activation-2.35e-01

Pos logit contributions
igsaw+0.024
hy+0.023
 brand+0.022
 doorway+0.022
des+0.022
Neg logit contributions
oyd-0.010
 pistols-0.008
 brothers-0.008
 Rex-0.008
 Remy-0.008
Token held
Token ID2714
Feature activation+2.53e+00

Pos logit contributions
igsaw+0.008
hy+0.008
 brand+0.007
 doorway+0.007
 click+0.007
Neg logit contributions
oyd-0.008
 pistols-0.007
 brothers-0.007
 neither-0.007
 sometimes-0.007
Token the
Token ID262
Feature activation-1.70e-01

Pos logit contributions
oyd+0.091
ained+0.071
 pistols+0.068
Sometimes+0.064
 sometimes+0.064
Neg logit contributions
hy-0.085
igsaw-0.083
 brand-0.081
 click-0.080
col-0.078
Token treat
Token ID2190
Feature activation+1.15e+00

Pos logit contributions
igsaw+0.006
hy+0.006
 brand+0.006
 doorway+0.006
col+0.006
Neg logit contributions
oyd-0.006
 pistols-0.005
 brothers-0.004
ball-0.004
 neither-0.004
Token in
Token ID287
Feature activation+2.24e-01

Pos logit contributions
oyd+0.032
 pistols+0.026
 brothers+0.023
 neither+0.022
 sometimes+0.022
Neg logit contributions
igsaw-0.049
hy-0.048
 brand-0.045
 click-0.045
 doorway-0.045
Token their
Token ID511
Feature activation-8.06e-01

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.004
 neither+0.004
 sometimes+0.003
Neg logit contributions
igsaw-0.011
hy-0.010
lic-0.010
 brand-0.010
 doorway-0.010
Token hands
Token ID2832
Feature activation+6.70e-01

Pos logit contributions
igsaw+0.034
hy+0.033
lic+0.032
 click+0.032
 brand+0.031
Neg logit contributions
oyd-0.022
 pistols-0.019
 brothers-0.017
 friends-0.017
 neither-0.016
Token and
Token ID290
Feature activation-2.48e-01

Pos logit contributions
oyd+0.019
 pistols+0.015
 sometimes+0.014
 neither+0.014
 brothers+0.014
Neg logit contributions
igsaw-0.029
hy-0.027
 brand-0.026
 doorway-0.026
 click-0.026
Token my
Token ID616
Feature activation-1.04e+00

Pos logit contributions
oyd+0.000
 pistols+0.000
 neither+0.000
 brothers+0.000
 sometimes+0.000
Neg logit contributions
igsaw-0.000
hy-0.000
 brand-0.000
 doorway-0.000
col-0.000
Token loves
Token ID10408
Feature activation+5.37e-01

Pos logit contributions
igsaw+0.040
hy+0.039
 doorway+0.037
 brand+0.036
col+0.036
Neg logit contributions
oyd-0.034
 pistols-0.029
 brothers-0.027
 neither-0.026
 friends-0.026
Token.
Token ID13
Feature activation-6.32e-01

Pos logit contributions
oyd+0.015
 pistols+0.013
 sometimes+0.012
 brothers+0.012
 neither+0.011
Neg logit contributions
igsaw-0.023
hy-0.022
 brand-0.021
 doorway-0.021
lic-0.021
Token We
Token ID775
Feature activation-2.42e+00

Pos logit contributions
igsaw+0.022
 brand+0.021
hy+0.021
 doorway+0.021
lic+0.020
Neg logit contributions
oyd-0.023
 pistols-0.019
 sometimes-0.018
 neither-0.018
 brothers-0.018
Token can
Token ID460
Feature activation-7.32e-01

Pos logit contributions
hy+0.085
igsaw+0.083
 click+0.080
 brand+0.079
col+0.078
Neg logit contributions
oyd-0.084
 pistols-0.071
 brothers-0.062
 neither-0.062
 horses-0.062
Token get
Token ID651
Feature activation-6.22e-01

Pos logit contributions
igsaw+0.021
hy+0.021
 click+0.021
 brand+0.020
col+0.019
Neg logit contributions
oyd-0.029
 pistols-0.026
 neither-0.023
 brothers-0.023
 humans-0.023
Token new
Token ID649
Feature activation-1.81e+00

Pos logit contributions
igsaw+0.024
hy+0.023
 brand+0.023
 click+0.022
col+0.022
Neg logit contributions
oyd-0.020
 pistols-0.016
 neither-0.014
 brothers-0.014
ball-0.014
Token toys
Token ID14958
Feature activation+6.36e-01

Pos logit contributions
igsaw+0.079
hy+0.077
 brand+0.074
 doorway+0.073
 click+0.073
Neg logit contributions
oyd-0.048
 pistols-0.039
 neither-0.034
 brothers-0.034
 sometimes-0.033
Token and
Token ID290
Feature activation-2.20e-01

Pos logit contributions
oyd+0.016
 pistols+0.014
 sometimes+0.014
 neither+0.013
 on+0.012
Neg logit contributions
igsaw-0.028
hy-0.027
 doorway-0.026
 brand-0.025
lic-0.025
Token a
Token ID257
Feature activation-3.33e-01

Pos logit contributions
igsaw+0.009
 doorway+0.008
hy+0.008
lic+0.008
col+0.008
Neg logit contributions
 pistols-0.006
oyd-0.006
 sometimes-0.006
 friends-0.005
 brothers-0.005
Token new
Token ID649
Feature activation-1.83e+00

Pos logit contributions
igsaw+0.013
hy+0.013
lic+0.012
col+0.012
 doorway+0.011
Neg logit contributions
oyd-0.009
 pistols-0.009
 friends-0.008
 brother-0.008
 brothers-0.008
Token pretend
Token ID16614
Feature activation-6.68e-01

Pos logit contributions
igsaw+0.011
hy+0.010
 brand+0.010
 doorway+0.010
 click+0.009
Neg logit contributions
oyd-0.013
 pistols-0.011
 neither-0.010
 brothers-0.010
 sometimes-0.010
Token to
Token ID284
Feature activation-1.08e+00

Pos logit contributions
igsaw+0.031
hy+0.029
 brand+0.027
 doorway+0.027
col+0.027
Neg logit contributions
oyd-0.017
 pistols-0.014
 sometimes-0.013
 neither-0.013
 brothers-0.013
Token sleep
Token ID3993
Feature activation+1.46e+00

Pos logit contributions
igsaw+0.042
hy+0.038
 brand+0.038
 doorway+0.037
col+0.037
Neg logit contributions
oyd-0.036
 pistols-0.029
 sometimes-0.029
 also-0.028
 neither-0.028
Token.
Token ID13
Feature activation-1.94e-01

Pos logit contributions
oyd+0.034
 pistols+0.026
 neither+0.022
 brothers+0.022
 sometimes+0.020
Neg logit contributions
igsaw-0.068
hy-0.068
 brand-0.065
 click-0.065
 doorway-0.064
Token They
Token ID1119
Feature activation-2.21e-01

Pos logit contributions
igsaw+0.008
hy+0.007
 brand+0.007
 click+0.007
 doorway+0.007
Neg logit contributions
oyd-0.006
 pistols-0.005
 brothers-0.005
 neither-0.005
ball-0.004
Token hope
Token ID2911
Feature activation-4.80e+00

Pos logit contributions
igsaw+0.008
hy+0.007
 brand+0.007
 doorway+0.007
col+0.007
Neg logit contributions
oyd-0.008
 pistols-0.007
 brothers-0.006
 sometimes-0.006
 neither-0.006
Token Mom
Token ID11254
Feature activation-4.73e-01

Pos logit contributions
igsaw+0.247
hy+0.237
 brand+0.228
 doorway+0.225
lic+0.223
Neg logit contributions
oyd-0.096
 pistols-0.073
 sometimes-0.066
 neither-0.066
 brothers-0.063
Token won
Token ID1839
Feature activation+1.09e+00

Pos logit contributions
igsaw+0.019
hy+0.019
 brand+0.018
 click+0.018
 doorway+0.018
Neg logit contributions
oyd-0.014
 pistols-0.011
 neither-0.010
 brothers-0.010
 sometimes-0.010
Token't
Token ID470
Feature activation+1.93e+00

Pos logit contributions
oyd+0.039
 pistols+0.034
 brothers+0.030
 neither+0.029
 goalie+0.029
Neg logit contributions
 click-0.037
igsaw-0.036
hy-0.036
col-0.033
bell-0.033
Token notice
Token ID4003
Feature activation-4.71e+00

Pos logit contributions
oyd+0.080
 pistols+0.069
 brothers+0.064
 neither+0.063
ball+0.061
Neg logit contributions
hy-0.057
igsaw-0.056
 click-0.053
 doorway-0.050
 brand-0.050
Token the
Token ID262
Feature activation-1.95e-01

Pos logit contributions
igsaw+0.229
hy+0.217
 doorway+0.208
lic+0.206
 brand+0.205
Neg logit contributions
oyd-0.108
 sometimes-0.093
 pistols-0.091
 neither-0.088
 each-0.084
Token could
Token ID714
Feature activation+7.13e-01

Pos logit contributions
oyd+0.058
 pistols+0.045
 sometimes+0.044
 also+0.043
 brothers+0.041
Neg logit contributions
igsaw-0.109
hy-0.097
 click-0.096
 brand-0.096
 doorway-0.095
Token fly
Token ID6129
Feature activation+5.09e-01

Pos logit contributions
oyd+0.027
 pistols+0.023
 brothers+0.021
 neither+0.021
ball+0.020
Neg logit contributions
hy-0.023
igsaw-0.023
 click-0.021
 brand-0.021
 doorway-0.021
Token and
Token ID290
Feature activation+1.62e-01

Pos logit contributions
oyd+0.010
 sometimes+0.010
 on+0.010
 pistols+0.009
 at+0.009
Neg logit contributions
igsaw-0.025
hy-0.023
 doorway-0.023
lic-0.022
cake-0.022
Token explore
Token ID7301
Feature activation-1.30e+00

Pos logit contributions
oyd+0.005
 sometimes+0.004
 also+0.004
 friends+0.004
 pistols+0.004
Neg logit contributions
igsaw-0.007
hy-0.006
lic-0.006
 brand-0.006
endant-0.006
Token the
Token ID262
Feature activation-1.90e-01

Pos logit contributions
igsaw+0.065
hy+0.064
lic+0.060
cake+0.058
endant+0.058
Neg logit contributions
oyd-0.025
 sometimes-0.023
 pistols-0.022
 on-0.021
 friends-0.020
Token world
Token ID995
Feature activation-4.30e+00

Pos logit contributions
igsaw+0.008
hy+0.007
 brand+0.007
 click+0.007
 doorway+0.007
Neg logit contributions
oyd-0.006
 pistols-0.005
 sometimes-0.005
 friends-0.005
 brothers-0.005
Token.
Token ID13
Feature activation+2.23e-01

Pos logit contributions
igsaw+0.223
icol+0.223
 doorway+0.211
 majesty+0.211
 Acc+0.211
Neg logit contributions
 on-0.071
 sometimes-0.070
 at-0.059
oyd-0.056
ball-0.052
Token
Token ID198
Feature activation-8.28e-03

Pos logit contributions
oyd+0.005
 pistols+0.004
 brothers+0.004
 sometimes+0.004
 Remy+0.004
Neg logit contributions
hy-0.010
igsaw-0.010
 click-0.010
 brand-0.010
col-0.010
Token
Token ID198
Feature activation-2.71e-02

Pos logit contributions
igsaw+0.000
 brand+0.000
hy+0.000
lic+0.000
 doorway+0.000
Neg logit contributions
oyd-0.000
 pistols-0.000
 neither-0.000
 forth-0.000
 goalie-0.000
TokenThe
Token ID464
Feature activation-5.15e-01

Pos logit contributions
igsaw+0.001
hy+0.001
 brand+0.001
 click+0.001
col+0.001
Neg logit contributions
oyd-0.001
 pistols-0.001
 neither-0.001
 brothers-0.001
 brother-0.001
Token moral
Token ID6573
Feature activation+2.60e+00

Pos logit contributions
igsaw+0.022
hy+0.021
 brand+0.020
 doorway+0.020
 click+0.020
Neg logit contributions
oyd-0.015
 pistols-0.012
 neither-0.011
 brothers-0.011
 sometimes-0.011
Token decided
Token ID3066
Feature activation-1.04e+00

Pos logit contributions
oyd+0.034
 pistols+0.030
 neither+0.027
 goalie+0.026
 brothers+0.026
Neg logit contributions
hy-0.034
col-0.033
 brand-0.033
lic-0.031
igsaw-0.031
Token to
Token ID284
Feature activation-8.31e-01

Pos logit contributions
igsaw+0.045
hy+0.042
 doorway+0.040
 click+0.039
 brand+0.039
Neg logit contributions
oyd-0.029
 pistols-0.026
 sometimes-0.025
 neither-0.023
 on-0.022
Token throw
Token ID3714
Feature activation+2.23e+00

Pos logit contributions
igsaw+0.031
hy+0.031
 brand+0.029
 doorway+0.029
lic+0.028
Neg logit contributions
oyd-0.027
 pistols-0.023
 sometimes-0.022
 brothers-0.022
 also-0.022
Token a
Token ID257
Feature activation-6.79e-02

Pos logit contributions
oyd+0.076
 pistols+0.057
 neither+0.056
 brothers+0.052
ained+0.051
Neg logit contributions
igsaw-0.085
hy-0.081
 brand-0.080
col-0.079
 click-0.077
Token big
Token ID1263
Feature activation-2.04e+00

Pos logit contributions
igsaw+0.003
hy+0.003
 doorway+0.003
 brand+0.003
 click+0.003
Neg logit contributions
oyd-0.002
 pistols-0.001
ball-0.001
 brothers-0.001
 neither-0.001
Token party
Token ID2151
Feature activation-5.36e+00

Pos logit contributions
hy+0.089
igsaw+0.089
lic+0.085
col+0.084
des+0.083
Neg logit contributions
oyd-0.050
 pistols-0.044
 brothers-0.037
 brother-0.037
ball-0.036
Token and
Token ID290
Feature activation+2.40e-01

Pos logit contributions
igsaw+0.268
hy+0.254
 brand+0.248
 disl+0.248
 Acc+0.247
Neg logit contributions
oyd-0.106
 on-0.086
 pistols-0.085
 sometimes-0.080
 brothers-0.080
Token sent
Token ID1908
Feature activation-3.35e+00

Pos logit contributions
oyd+0.007
 pistols+0.006
 sometimes+0.006
 friends+0.006
 also+0.006
Neg logit contributions
igsaw-0.010
hy-0.009
endant-0.009
 doorway-0.009
 brand-0.008
Token out
Token ID503
Feature activation-2.41e+00

Pos logit contributions
igsaw+0.174
hy+0.165
 doorway+0.159
endant+0.157
 brand+0.156
Neg logit contributions
oyd-0.057
 pistols-0.050
 friends-0.050
 brothers-0.048
 on-0.046
Token invitations
Token ID42851
Feature activation-4.14e+00

Pos logit contributions
igsaw+0.131
icol+0.130
 Acc+0.125
 doorway+0.125
hy+0.124
Neg logit contributions
 on-0.034
 friends-0.033
 brothers-0.032
 at-0.028
 pistols-0.027
Token to
Token ID284
Feature activation-8.54e-01

Pos logit contributions
igsaw+0.188
hy+0.171
 disl+0.167
 Acc+0.166
 brand+0.165
Neg logit contributions
oyd-0.104
 on-0.090
 pistols-0.089
 sometimes-0.086
 brothers-0.085
Token.
Token ID13
Feature activation-1.21e-01

Pos logit contributions
oyd+0.012
 pistols+0.010
 brothers+0.009
 sometimes+0.009
 neither+0.009
Neg logit contributions
igsaw-0.020
hy-0.019
 brand-0.018
 doorway-0.018
 click-0.018
Token It
Token ID632
Feature activation-1.70e+00

Pos logit contributions
igsaw+0.005
hy+0.005
 brand+0.005
des+0.005
col+0.005
Neg logit contributions
oyd-0.003
 brothers-0.003
 Remy-0.003
 pistols-0.003
 neither-0.003
Token pulled
Token ID5954
Feature activation+1.51e+00

Pos logit contributions
igsaw+0.072
hy+0.068
 brand+0.065
 click+0.064
 doorway+0.064
Neg logit contributions
oyd-0.050
 pistols-0.041
 sometimes-0.038
 neither-0.037
 brothers-0.037
Token the
Token ID262
Feature activation-2.62e-01

Pos logit contributions
oyd+0.038
 pistols+0.031
 sometimes+0.028
 neither+0.028
 brothers+0.027
Neg logit contributions
igsaw-0.069
hy-0.067
 doorway-0.064
 brand-0.063
 disl-0.062
Token fridge
Token ID25772
Feature activation-3.08e+00

Pos logit contributions
igsaw+0.010
hy+0.009
 brand+0.009
 doorway+0.009
 click+0.009
Neg logit contributions
oyd-0.009
 pistols-0.007
 brothers-0.007
 neither-0.007
 sometimes-0.007
Token door
Token ID3420
Feature activation-6.54e+00

Pos logit contributions
 disl+0.145
igsaw+0.144
 Acc+0.135
leaf+0.132
hy+0.132
Neg logit contributions
oyd-0.067
 on-0.066
 neither-0.065
 sometimes-0.062
 at-0.061
Token and
Token ID290
Feature activation-1.75e-01

Pos logit contributions
 disl+0.344
igsaw+0.340
 majesty+0.325
 Acc+0.314
icol+0.308
Neg logit contributions
 on-0.142
 sometimes-0.122
 at-0.122
 still-0.116
 also-0.108
Token made
Token ID925
Feature activation+1.09e+00

Pos logit contributions
igsaw+0.008
hy+0.008
 brand+0.007
col+0.007
des+0.007
Neg logit contributions
oyd-0.005
 pistols-0.004
 sometimes-0.004
 neither-0.004
 also-0.004
Token a
Token ID257
Feature activation-2.79e-01

Pos logit contributions
oyd+0.023
 pistols+0.018
 brothers+0.017
 friends+0.017
 neither+0.017
Neg logit contributions
igsaw-0.054
hy-0.053
 doorway-0.049
col-0.048
des-0.047
Token loud
Token ID7812
Feature activation-2.05e+00

Pos logit contributions
igsaw+0.011
hy+0.011
col+0.010
keeping+0.010
des+0.010
Neg logit contributions
oyd-0.008
 pistols-0.007
 brothers-0.007
 friends-0.007
 brother-0.006
Token noise
Token ID7838
Feature activation-1.27e+00

Pos logit contributions
igsaw+0.090
endant+0.088
des+0.083
keeping+0.083
col+0.082
Neg logit contributions
oyd-0.047
 on-0.044
 sometimes-0.040
 forth-0.039
 at-0.038
Token a
Token ID257
Feature activation-2.99e-01

Pos logit contributions
oyd+0.054
 pistols+0.045
 brothers+0.040
 sometimes+0.040
 neither+0.040
Neg logit contributions
igsaw-0.078
hy-0.076
 brand-0.072
 doorway-0.071
 click-0.071
Token reward
Token ID6721
Feature activation-2.05e-03

Pos logit contributions
igsaw+0.010
hy+0.010
col+0.010
 doorway+0.009
keeping+0.009
Neg logit contributions
oyd-0.010
 pistols-0.009
 friends-0.008
ball-0.008
 neither-0.008
Token,
Token ID11
Feature activation+3.27e-01

Pos logit contributions
igsaw+0.000
hy+0.000
 brand+0.000
 doorway+0.000
icol+0.000
Neg logit contributions
oyd-0.000
 sometimes-0.000
 pistols-0.000
 on-0.000
 neither-0.000
Token I
Token ID314
Feature activation-1.83e+00

Pos logit contributions
oyd+0.012
 pistols+0.010
 neither+0.009
 brothers+0.009
ball+0.009
Neg logit contributions
igsaw-0.011
hy-0.011
 brand-0.011
 click-0.010
 doorway-0.010
Token will
Token ID481
Feature activation-3.51e-01

Pos logit contributions
igsaw+0.081
hy+0.076
 brand+0.073
 doorway+0.073
 click+0.071
Neg logit contributions
oyd-0.050
 also-0.041
 sometimes-0.041
 still-0.038
ball-0.038
Token show
Token ID905
Feature activation-4.50e+00

Pos logit contributions
igsaw+0.016
lic+0.014
 doorway+0.014
 brand+0.014
hy+0.013
Neg logit contributions
oyd-0.009
 also-0.009
 still-0.009
 sometimes-0.009
 friends-0.009
Token you
Token ID345
Feature activation-2.26e+00

Pos logit contributions
igsaw+0.173
hy+0.166
 brand+0.159
 doorway+0.158
 click+0.155
Neg logit contributions
oyd-0.145
 pistols-0.123
 brothers-0.114
 neither-0.113
 sometimes-0.112
Token something
Token ID1223
Feature activation-2.94e-01

Pos logit contributions
hy+0.097
igsaw+0.096
 doorway+0.089
 click+0.088
icol+0.088
Neg logit contributions
oyd-0.060
 pistols-0.052
 sometimes-0.051
 friends-0.050
 brothers-0.049
Token special
Token ID2041
Feature activation-1.37e+00

Pos logit contributions
igsaw+0.013
hy+0.012
 doorway+0.011
icol+0.011
 click+0.011
Neg logit contributions
oyd-0.008
 friends-0.008
 sometimes-0.008
 pistols-0.008
 on-0.007
Token".
Token ID1911
Feature activation-1.44e+00

Pos logit contributions
igsaw+0.054
hy+0.053
 brand+0.049
 doorway+0.049
 click+0.049
Neg logit contributions
oyd-0.042
 pistols-0.036
 sometimes-0.035
 brothers-0.033
 neither-0.033
Token Jack
Token ID3619
Feature activation+1.26e+00

Pos logit contributions
hy+0.058
igsaw+0.057
col+0.053
 brand+0.053
 doorway+0.053
Neg logit contributions
oyd-0.044
 pistols-0.036
 brothers-0.034
 neither-0.032
 sometimes-0.032