TOP ACTIVATIONS
MAX = 0.288

ostr ich pen . Ost ric hes are big birds with
ostr ich pen . Ost ric hes are big birds with
mom and to the ost ric hes . They learned their
should not tease the ost ric hes . They are not
They do not pack ost ric hes anymore . They pack
" O st ric hes do not like people
But the ost ric hes did not like that
wanted to touch the ost ric hes . They reached out
cried . " The ost ric hes are mean ! They
shouted . " The ost ric hes are so funny !"
but be careful . Ost ric hes can be gr umpy
. It sees other ost ric hes . It makes new
. One of the ost ric hes looked different from the
and run to the ost ric hes . "
. They were called ost ric hes . Lily and Ben
that said " O st ric hes ". Sara and Tom
you count how many ost ric hes there are ?"
Yes , they are ost ric hes ," mom said .
Look , mom , ost ric hes !" "
did not know what ost ric hes were . They looked

MIN ACTIVATIONS
MIN = -0.211

some snow for a play date with his friends . He
! She finished her play date and went home , with
own toys to the play date . The first one brought
could have her play - date after all . <|endoftext|> One
and had a fun play date . <|endoftext|> Once upon a
my were on a play date . Alice was playing with
can have your play - date with R uf us now
him . But to no avail , they were too scared
for using the past - date cheese . Sam learned an
times , but to no avail . Jane 's
almost ready for the play date ! Billy opened
friends over for a play date . She showed them her
deep paws , to no avail . He finally thought of
came over for a play date . Her friend asked for
my house for a play date . I invite you ."
out , but to no avail . He felt sad and
would have a play - date and wear makeup .
let 's have a play date !" The steak was
. One day , Jill suggested that they go and pick
The fish listened carefully and suggested that the crab should explore

INTERVAL 0.163 - 0.288
CONTAINS 3.736%

is not big and yellow and happy . It is brown
laugh . <|endoftext|> L ila and True were best friends .
liked to feel their wool and to jump and run on
We have to be patient and wait for the right time
put them in a bag and pretended it was their loot

INTERVAL 0.038 - 0.163
CONTAINS 19.660%

the squirrel . The squirrel didn 't run away and instead
Grand ma . Grand ma smiled and kissed them .
looked at the car . It was ruined . They felt
with lots of leaves to keep them warm and a big
day , Charlie was in the kitchen with his mom .

INTERVAL -0.086 - 0.038
CONTAINS 67.923%

park . She did not want to make Tom unhappy .
grass in the lawn and fell down . She started to
Billy . " Let 's play again soon ." And with
it with the big B !" He picks up the card
. She loved to play with her toys and go on

INTERVAL -0.211 - -0.086
CONTAINS 8.681%

. Something even more unusual happened : a big flying elephant
, but it was too late . The bridge broke ,
, but he didn 't know how . He tried to
. Now , I want to repay her kindness and show
shared this moment . He said they had to go back
Token ostr
Token ID42387
Feature activation+8.46e-02

Pos logit contributions
 com+0.065
 icy+0.062
 warm+0.061
 grass+0.057
 soft+0.057
Neg logit contributions
mate-0.076
orman-0.072
anish-0.070
 Troll-0.069
 policy-0.069
Tokenich
Token ID488
Feature activation+4.27e-02

Pos logit contributions
 icy+0.100
 com+0.100
 clouds+0.095
 warm+0.094
 very+0.093
Neg logit contributions
mate-0.160
oker-0.152
 policy-0.152
orman-0.151
 occasions-0.150
Token pen
Token ID3112
Feature activation+7.78e-02

Pos logit contributions
 com+0.052
 icy+0.050
 warm+0.050
 grass+0.046
 soft+0.046
Neg logit contributions
mate-0.082
orman-0.079
anish-0.077
 policy-0.076
oker-0.075
Token.
Token ID13
Feature activation+7.83e-03

Pos logit contributions
 com+0.080
 icy+0.075
 warm+0.075
 grass+0.069
 soft+0.068
Neg logit contributions
mate-0.164
orman-0.159
anish-0.155
 Troll-0.153
 policy-0.153
Token Ost
Token ID38919
Feature activation+4.62e-02

Pos logit contributions
 com+0.010
 warm+0.009
 icy+0.009
 grass+0.009
 soft+0.009
Neg logit contributions
mate-0.015
orman-0.014
anish-0.014
uing-0.014
uer-0.014
Tokenric
Token ID1173
Feature activation+2.88e-01

Pos logit contributions
 com+0.064
 warm+0.057
 icy+0.055
 soft+0.054
 boun+0.052
Neg logit contributions
mate-0.084
anish-0.082
orman-0.081
 Troll-0.080
ings-0.078
Tokenhes
Token ID956
Feature activation+1.01e-01

Pos logit contributions
 com+0.216
 warm+0.188
 icy+0.188
 grass+0.162
 soft+0.159
Neg logit contributions
mate-0.680
orman-0.677
anish-0.654
 Troll-0.638
oker-0.635
Token are
Token ID389
Feature activation+5.94e-02

Pos logit contributions
 com+0.127
 warm+0.126
 icy+0.126
 very+0.120
 grass+0.119
Neg logit contributions
mate-0.188
orman-0.184
uer-0.181
 policy-0.179
date-0.177
Token big
Token ID1263
Feature activation-6.07e-03

Pos logit contributions
 com+0.046
 icy+0.041
 warm+0.040
 grass+0.037
 clouds+0.036
Neg logit contributions
mate-0.134
orman-0.130
 Troll-0.126
 password-0.126
anish-0.125
Token birds
Token ID10087
Feature activation+9.89e-02

Pos logit contributions
mate+0.015
orman+0.015
date+0.014
uing+0.014
oker+0.014
Neg logit contributions
 com-0.005
 icy-0.005
 warm-0.004
 clouds-0.004
 grass-0.004
Token with
Token ID351
Feature activation-6.54e-03

Pos logit contributions
 com+0.148
 icy+0.140
 warm+0.136
 grass+0.131
 soft+0.129
Neg logit contributions
mate-0.157
orman-0.148
anish-0.145
uing-0.142
 Troll-0.142
Token ostr
Token ID42387
Feature activation+8.51e-02

Pos logit contributions
 com+0.058
 icy+0.055
 warm+0.055
 grass+0.051
 soft+0.051
Neg logit contributions
mate-0.077
orman-0.074
anish-0.072
uing-0.071
 policy-0.071
Tokenich
Token ID488
Feature activation+4.08e-02

Pos logit contributions
 icy+0.102
 com+0.102
 clouds+0.097
 warm+0.096
 very+0.094
Neg logit contributions
mate-0.162
oker-0.153
 policy-0.152
orman-0.151
 occasions-0.150
Token pen
Token ID3112
Feature activation+7.77e-02

Pos logit contributions
 com+0.053
 icy+0.050
 warm+0.050
 grass+0.048
 very+0.047
Neg logit contributions
mate-0.077
orman-0.074
anish-0.071
 policy-0.071
oker-0.070
Token.
Token ID13
Feature activation+4.20e-03

Pos logit contributions
 com+0.083
 icy+0.078
 warm+0.078
 grass+0.072
 soft+0.071
Neg logit contributions
mate-0.160
orman-0.155
anish-0.150
 policy-0.148
 Troll-0.148
Token Ost
Token ID38919
Feature activation+4.72e-02

Pos logit contributions
 com+0.005
 warm+0.004
 icy+0.004
 grass+0.004
 soft+0.004
Neg logit contributions
mate-0.009
orman-0.008
uing-0.008
uer-0.008
anish-0.008
Tokenric
Token ID1173
Feature activation+2.87e-01

Pos logit contributions
 com+0.066
 warm+0.059
 icy+0.058
 soft+0.056
 boun+0.054
Neg logit contributions
mate-0.085
anish-0.082
orman-0.082
 Troll-0.081
ings-0.078
Tokenhes
Token ID956
Feature activation+1.00e-01

Pos logit contributions
 com+0.211
 warm+0.182
 icy+0.182
 grass+0.156
 soft+0.153
Neg logit contributions
mate-0.685
orman-0.683
anish-0.659
 Troll-0.642
oker-0.639
Token are
Token ID389
Feature activation+5.97e-02

Pos logit contributions
 icy+0.115
 warm+0.114
 com+0.112
 very+0.111
 grass+0.109
Neg logit contributions
mate-0.195
uer-0.193
orman-0.190
 avail-0.189
uers-0.187
Token big
Token ID1263
Feature activation-7.60e-03

Pos logit contributions
 com+0.046
 icy+0.040
 warm+0.039
 grass+0.037
 clouds+0.035
Neg logit contributions
mate-0.135
orman-0.132
 Troll-0.128
 password-0.127
anish-0.126
Token birds
Token ID10087
Feature activation+9.76e-02

Pos logit contributions
mate+0.019
orman+0.019
date+0.018
uing+0.018
oker+0.018
Neg logit contributions
 com-0.006
 icy-0.006
 clouds-0.005
 warm-0.005
 grass-0.005
Token with
Token ID351
Feature activation-6.95e-03

Pos logit contributions
 com+0.139
 icy+0.132
 warm+0.127
 grass+0.123
 soft+0.120
Neg logit contributions
mate-0.160
orman-0.151
anish-0.148
uing-0.146
 Troll-0.146
Token mom
Token ID1995
Feature activation-3.99e-02

Pos logit contributions
mate+0.146
 client+0.137
orman+0.132
 Troll+0.132
 begs+0.130
Neg logit contributions
 com-0.125
 refreshing-0.108
 boun-0.107
 vel-0.107
 icy-0.106
Token and
Token ID290
Feature activation+1.76e-01

Pos logit contributions
mate+0.079
orman+0.076
anish+0.074
uing+0.074
 policy+0.073
Neg logit contributions
 com-0.046
 icy-0.042
 warm-0.041
 grass-0.038
 soft-0.037
Token to
Token ID284
Feature activation-8.85e-02

Pos logit contributions
 com+0.263
 icy+0.251
 warm+0.236
 clouds+0.234
 grass+0.231
Neg logit contributions
mate-0.280
 policy-0.262
orman-0.259
anish-0.256
 Troll-0.255
Token the
Token ID262
Feature activation+4.30e-02

Pos logit contributions
mate+0.151
 client+0.133
 begs+0.133
 Troll+0.130
 jury+0.130
Neg logit contributions
 com-0.126
 icy-0.111
 vel-0.107
 refreshing-0.105
 clouds-0.102
Token ost
Token ID23619
Feature activation+8.09e-02

Pos logit contributions
 com+0.048
 icy+0.047
 warm+0.047
 grass+0.043
 very+0.043
Neg logit contributions
mate-0.090
anish-0.086
oker-0.085
orman-0.085
uing-0.084
Tokenric
Token ID1173
Feature activation+2.87e-01

Pos logit contributions
 com+0.079
 icy+0.075
 warm+0.073
 grass+0.067
 soft+0.066
Neg logit contributions
mate-0.174
orman-0.168
anish-0.164
 policy-0.161
oker-0.161
Tokenhes
Token ID956
Feature activation+9.83e-02

Pos logit contributions
 com+0.176
 icy+0.152
 warm+0.149
 grass+0.125
 soft+0.123
Neg logit contributions
mate-0.725
orman-0.709
anish-0.691
uing-0.677
 Troll-0.677
Token.
Token ID13
Feature activation+2.69e-03

Pos logit contributions
 com+0.120
 warm+0.118
 icy+0.117
 very+0.109
 grass+0.109
Neg logit contributions
mate-0.186
orman-0.180
uer-0.177
anish-0.176
 Troll-0.174
Token They
Token ID1119
Feature activation+1.30e-02

Pos logit contributions
 com+0.004
 icy+0.004
 warm+0.004
 grass+0.003
 soft+0.003
Neg logit contributions
mate-0.005
orman-0.004
anish-0.004
 policy-0.004
 Troll-0.004
Token learned
Token ID4499
Feature activation+1.56e-02

Pos logit contributions
 com+0.021
 icy+0.020
 clouds+0.020
 sun+0.019
 sunshine+0.019
Neg logit contributions
 begs-0.017
mate-0.016
orman-0.015
 password-0.015
 jury-0.015
Token their
Token ID511
Feature activation-6.17e-02

Pos logit contributions
 com+0.023
 refreshing+0.021
 icy+0.021
 clouds+0.021
 vel+0.021
Neg logit contributions
orman-0.023
 policy-0.022
mate-0.022
 client-0.022
 begs-0.022
Token should
Token ID815
Feature activation-3.28e-02

Pos logit contributions
mate+0.040
anish+0.039
orman+0.038
 purposes+0.038
 Troll+0.037
Neg logit contributions
 com-0.045
 icy-0.042
 grass-0.041
 warm-0.040
 soft-0.039
Token not
Token ID407
Feature activation+7.81e-03

Pos logit contributions
mate+0.058
orman+0.056
anish+0.054
 policy+0.053
 Troll+0.053
Neg logit contributions
 com-0.045
 icy-0.042
 warm-0.042
 grass-0.039
 soft-0.039
Token tease
Token ID34691
Feature activation+6.32e-05

Pos logit contributions
 com+0.012
 icy+0.011
 warm+0.011
 grass+0.011
 soft+0.011
Neg logit contributions
mate-0.013
orman-0.012
anish-0.012
 Troll-0.011
uing-0.011
Token the
Token ID262
Feature activation+4.28e-02

Pos logit contributions
 com+0.000
 icy+0.000
 warm+0.000
 soft+0.000
 grass+0.000
Neg logit contributions
mate-0.000
orman-0.000
anish-0.000
 Troll-0.000
 policy-0.000
Token ost
Token ID23619
Feature activation+8.11e-02

Pos logit contributions
 com+0.049
 icy+0.047
 warm+0.047
 grass+0.043
 soft+0.043
Neg logit contributions
mate-0.086
orman-0.082
anish-0.081
oker-0.079
date-0.079
Tokenric
Token ID1173
Feature activation+2.87e-01

Pos logit contributions
 com+0.090
 icy+0.082
 warm+0.082
 grass+0.075
 soft+0.075
Neg logit contributions
mate-0.166
orman-0.160
anish-0.157
 Troll-0.154
uing-0.154
Tokenhes
Token ID956
Feature activation+9.91e-02

Pos logit contributions
 com+0.198
 icy+0.175
 warm+0.172
 grass+0.149
 soft+0.146
Neg logit contributions
mate-0.703
orman-0.684
anish-0.668
oker-0.655
uing-0.655
Token.
Token ID13
Feature activation+5.40e-03

Pos logit contributions
 warm+0.116
 icy+0.116
 com+0.114
 very+0.112
 grass+0.107
Neg logit contributions
orman-0.188
uer-0.186
mate-0.185
 Majesty-0.183
uers-0.182
Token They
Token ID1119
Feature activation+1.64e-02

Pos logit contributions
 com+0.008
 icy+0.008
 warm+0.008
 grass+0.007
 soft+0.007
Neg logit contributions
mate-0.009
orman-0.008
anish-0.008
 Troll-0.008
 policy-0.008
Token are
Token ID389
Feature activation+5.45e-02

Pos logit contributions
 com+0.029
 icy+0.027
 warm+0.026
 grass+0.026
 soft+0.025
Neg logit contributions
mate-0.021
anish-0.020
 purposes-0.020
orman-0.020
 begs-0.020
Token not
Token ID407
Feature activation+6.93e-03

Pos logit contributions
 com+0.064
 clouds+0.058
 icy+0.058
 grass+0.057
 warm+0.055
Neg logit contributions
mate-0.094
 policy-0.090
orman-0.088
 password-0.088
 client-0.088
Token They
Token ID1119
Feature activation+1.61e-02

Pos logit contributions
 com+0.009
 icy+0.009
 warm+0.009
 grass+0.008
 soft+0.008
Neg logit contributions
mate-0.015
orman-0.014
uing-0.014
date-0.014
anish-0.014
Token do
Token ID466
Feature activation-3.36e-02

Pos logit contributions
 com+0.029
 icy+0.027
 warm+0.026
 grass+0.026
 orange+0.026
Neg logit contributions
mate-0.020
orman-0.019
 password-0.019
 begs-0.019
 client-0.018
Token not
Token ID407
Feature activation+7.46e-03

Pos logit contributions
orman+0.061
mate+0.059
uing+0.057
anish+0.056
 policy+0.056
Neg logit contributions
 com-0.046
 icy-0.042
 vel-0.041
 warm-0.041
 grass-0.040
Token pack
Token ID2353
Feature activation+7.59e-04

Pos logit contributions
 com+0.012
 icy+0.011
 warm+0.011
 grass+0.011
 soft+0.011
Neg logit contributions
mate-0.011
orman-0.011
anish-0.010
 policy-0.010
date-0.010
Token ost
Token ID23619
Feature activation+8.26e-02

Pos logit contributions
 com+0.001
 icy+0.001
 warm+0.001
 grass+0.001
 clouds+0.001
Neg logit contributions
mate-0.001
orman-0.001
anish-0.001
 policy-0.001
uing-0.001
Tokenric
Token ID1173
Feature activation+2.86e-01

Pos logit contributions
 com+0.115
 icy+0.107
 warm+0.107
 soft+0.100
 grass+0.100
Neg logit contributions
mate-0.145
orman-0.139
anish-0.138
uing-0.137
 Troll-0.135
Tokenhes
Token ID956
Feature activation+9.92e-02

Pos logit contributions
 com+0.232
 icy+0.199
 warm+0.197
 grass+0.177
 vel+0.177
Neg logit contributions
mate-0.664
orman-0.655
anish-0.632
uing-0.620
oker-0.618
Token anymore
Token ID7471
Feature activation-8.54e-02

Pos logit contributions
 com+0.122
 icy+0.111
 warm+0.109
 soft+0.102
 grass+0.102
Neg logit contributions
mate-0.189
orman-0.177
anish-0.174
ings-0.171
uing-0.170
Token.
Token ID13
Feature activation+7.53e-03

Pos logit contributions
mate+0.165
orman+0.160
date+0.157
oker+0.155
anish+0.155
Neg logit contributions
 com-0.105
 warm-0.100
 icy-0.099
 very-0.096
 grass-0.095
Token They
Token ID1119
Feature activation+1.61e-02

Pos logit contributions
 com+0.010
 icy+0.010
 warm+0.010
 grass+0.009
 soft+0.009
Neg logit contributions
mate-0.014
orman-0.013
anish-0.013
uing-0.012
date-0.012
Token pack
Token ID2353
Feature activation+1.10e-03

Pos logit contributions
 com+0.028
 icy+0.027
 warm+0.026
 grass+0.025
 soft+0.025
Neg logit contributions
mate-0.021
orman-0.020
 begs-0.019
 password-0.019
anish-0.019
Token
Token ID198
Feature activation+2.59e-02

Pos logit contributions
 com+0.004
 warm+0.004
 very+0.004
 soft+0.004
 grass+0.004
Neg logit contributions
mate-0.011
fer-0.011
uing-0.010
 occasions-0.010
eman-0.010
Token
Token ID198
Feature activation+2.63e-02

Pos logit contributions
 com+0.020
 warm+0.019
 icy+0.018
 very+0.017
 soft+0.017
Neg logit contributions
mate-0.061
 Majesty-0.059
anish-0.059
date-0.059
uer-0.059
Token"
Token ID1
Feature activation+4.16e-03

Pos logit contributions
 icy+0.033
 warm+0.033
 com+0.033
 grass+0.032
 very+0.032
Neg logit contributions
mate-0.049
 Majesty-0.049
anish-0.048
date-0.047
 avail-0.046
TokenO
Token ID46
Feature activation+7.47e-02

Pos logit contributions
 com+0.006
 icy+0.006
 warm+0.006
 soft+0.005
 grass+0.005
Neg logit contributions
mate-0.007
orman-0.007
anish-0.007
 jury-0.006
oker-0.006
Tokenst
Token ID301
Feature activation+4.81e-02

Pos logit contributions
 icy+0.124
 com+0.122
 warm+0.119
 grass+0.112
 very+0.112
Neg logit contributions
mate-0.109
orman-0.100
date-0.099
anish-0.097
 Majesty-0.096
Tokenric
Token ID1173
Feature activation+2.86e-01

Pos logit contributions
 icy+0.080
 com+0.079
 warm+0.078
 soft+0.075
 very+0.075
Neg logit contributions
mate-0.068
 avail-0.066
date-0.064
 policy-0.064
 jury-0.064
Tokenhes
Token ID956
Feature activation+9.76e-02

Pos logit contributions
 com+0.181
 icy+0.159
 warm+0.156
 grass+0.133
 soft+0.131
Neg logit contributions
mate-0.718
orman-0.700
anish-0.685
 Troll-0.673
uing-0.671
Token do
Token ID466
Feature activation-3.17e-02

Pos logit contributions
 icy+0.114
 warm+0.114
 very+0.112
 com+0.110
 soft+0.104
Neg logit contributions
uer-0.185
orman-0.184
 avail-0.183
mate-0.182
 Majesty-0.180
Token not
Token ID407
Feature activation+6.33e-03

Pos logit contributions
mate+0.063
orman+0.061
anish+0.059
uing+0.058
uers+0.058
Neg logit contributions
 com-0.037
 icy-0.034
 warm-0.034
 grass-0.032
 soft-0.031
Token like
Token ID588
Feature activation+2.03e-02

Pos logit contributions
 com+0.010
 icy+0.009
 warm+0.009
 grass+0.009
 soft+0.009
Neg logit contributions
mate-0.010
orman-0.010
anish-0.009
 policy-0.009
date-0.009
Token people
Token ID661
Feature activation+2.87e-02

Pos logit contributions
 com+0.025
 vel+0.022
 warm+0.022
 icy+0.021
 refreshing+0.020
Neg logit contributions
mate-0.036
orman-0.035
 policy-0.033
 purposes-0.033
 client-0.033
Token
Token ID198
Feature activation+2.44e-02

Pos logit contributions
 com+0.003
 warm+0.003
 icy+0.003
 grass+0.003
 soft+0.003
Neg logit contributions
mate-0.005
orman-0.005
uing-0.005
oker-0.005
uer-0.005
Token
Token ID198
Feature activation+2.49e-02

Pos logit contributions
 com+0.025
 warm+0.023
 icy+0.023
 soft+0.021
 grass+0.021
Neg logit contributions
mate-0.052
anish-0.049
orman-0.049
date-0.049
oker-0.049
TokenBut
Token ID1537
Feature activation-3.97e-02

Pos logit contributions
 warm+0.036
 com+0.035
 icy+0.035
 very+0.034
 soft+0.033
Neg logit contributions
mate-0.042
 Majesty-0.040
anish-0.040
 avail-0.040
 Troll-0.039
Token the
Token ID262
Feature activation+4.19e-02

Pos logit contributions
mate+0.083
orman+0.080
anish+0.077
 Troll+0.076
date+0.076
Neg logit contributions
 com-0.043
 icy-0.039
 warm-0.038
 soft-0.036
 grass-0.036
Token ost
Token ID23619
Feature activation+8.03e-02

Pos logit contributions
 com+0.047
 icy+0.046
 warm+0.045
 grass+0.042
 soft+0.042
Neg logit contributions
mate-0.086
orman-0.082
anish-0.081
uing-0.079
oker-0.079
Tokenric
Token ID1173
Feature activation+2.86e-01

Pos logit contributions
 com+0.089
 icy+0.082
 warm+0.082
 grass+0.075
 soft+0.075
Neg logit contributions
mate-0.165
orman-0.158
anish-0.156
 Troll-0.153
uing-0.153
Tokenhes
Token ID956
Feature activation+9.61e-02

Pos logit contributions
 com+0.168
 icy+0.146
 warm+0.144
 grass+0.120
 soft+0.118
Neg logit contributions
mate-0.730
orman-0.713
anish-0.697
 Troll-0.683
uing-0.682
Token did
Token ID750
Feature activation-3.78e-02

Pos logit contributions
 warm+0.090
 icy+0.089
 very+0.086
 com+0.085
 grass+0.083
Neg logit contributions
 Majesty-0.207
uer-0.202
mate-0.202
 policy-0.200
uers-0.199
Token not
Token ID407
Feature activation+3.05e-03

Pos logit contributions
mate+0.072
orman+0.070
uing+0.067
 begs+0.066
anish+0.066
Neg logit contributions
 com-0.046
 icy-0.043
 warm-0.041
 grass-0.041
 clouds-0.040
Token like
Token ID588
Feature activation+1.96e-02

Pos logit contributions
 com+0.004
 warm+0.004
 icy+0.004
 soft+0.004
 grass+0.004
Neg logit contributions
mate-0.005
orman-0.005
anish-0.005
 policy-0.005
date-0.005
Token that
Token ID326
Feature activation-1.02e-01

Pos logit contributions
 com+0.022
 warm+0.020
 icy+0.019
 vel+0.018
 grass+0.018
Neg logit contributions
mate-0.038
orman-0.037
 Troll-0.035
 purposes-0.035
 client-0.034
Token wanted
Token ID2227
Feature activation-3.02e-02

Pos logit contributions
 com+0.018
 icy+0.017
 warm+0.016
 clouds+0.016
 fresh+0.015
Neg logit contributions
mate-0.015
orman-0.015
 begs-0.015
 purposes-0.014
 password-0.014
Token to
Token ID284
Feature activation-8.91e-02

Pos logit contributions
mate+0.048
orman+0.047
 Troll+0.044
 policy+0.044
anish+0.043
Neg logit contributions
 com-0.045
 icy-0.042
 warm-0.041
 grass-0.039
 soft-0.038
Token touch
Token ID3638
Feature activation+4.54e-02

Pos logit contributions
mate+0.162
orman+0.158
anish+0.156
oker+0.154
uing+0.154
Neg logit contributions
 warm-0.121
 com-0.117
 icy-0.114
 soft-0.109
 grass-0.109
Token the
Token ID262
Feature activation+4.20e-02

Pos logit contributions
 com+0.055
 icy+0.048
 warm+0.046
 refreshing+0.044
 boun+0.044
Neg logit contributions
mate-0.086
ings-0.079
 Troll-0.077
orman-0.077
anish-0.077
Token ost
Token ID23619
Feature activation+8.03e-02

Pos logit contributions
 com+0.045
 warm+0.039
 icy+0.039
 soft+0.035
 vel+0.034
Neg logit contributions
mate-0.085
orman-0.082
 Troll-0.081
anish-0.080
 purposes-0.079
Tokenric
Token ID1173
Feature activation+2.86e-01

Pos logit contributions
 com+0.099
 warm+0.090
 icy+0.089
 soft+0.083
 grass+0.082
Neg logit contributions
mate-0.154
orman-0.148
anish-0.147
 Troll-0.146
uing-0.145
Tokenhes
Token ID956
Feature activation+9.64e-02

Pos logit contributions
 com+0.189
 icy+0.163
 warm+0.161
 grass+0.137
 soft+0.135
Neg logit contributions
mate-0.709
orman-0.696
anish-0.676
oker-0.663
uing-0.661
Token.
Token ID13
Feature activation+2.30e-03

Pos logit contributions
 warm+0.106
 icy+0.103
 com+0.102
 very+0.102
 grass+0.097
Neg logit contributions
 Majesty-0.192
orman-0.191
uer-0.190
mate-0.188
uers-0.187
Token They
Token ID1119
Feature activation+1.24e-02

Pos logit contributions
 warm+0.002
 com+0.002
 very+0.002
 soft+0.002
 grass+0.002
Neg logit contributions
mate-0.005
fer-0.005
uing-0.005
rum-0.005
oker-0.005
Token reached
Token ID4251
Feature activation-5.42e-04

Pos logit contributions
 com+0.019
 icy+0.017
 warm+0.017
 clouds+0.016
 fresh+0.016
Neg logit contributions
mate-0.019
orman-0.019
 begs-0.018
anish-0.018
 Troll-0.018
Token out
Token ID503
Feature activation-7.66e-02

Pos logit contributions
mate+0.001
orman+0.001
 Troll+0.001
anish+0.001
uing+0.001
Neg logit contributions
 com-0.001
 icy-0.001
 warm-0.001
 soft-0.001
 grass-0.001
Token cried
Token ID16896
Feature activation-1.08e-02

Pos logit contributions
mate+0.101
 policy+0.096
orman+0.095
 Majesty+0.094
 occasions+0.092
Neg logit contributions
 warm-0.068
 com-0.067
 icy-0.064
 very-0.064
 soft-0.062
Token.
Token ID13
Feature activation+7.21e-03

Pos logit contributions
mate+0.017
orman+0.015
anish+0.015
uing+0.015
 begs+0.015
Neg logit contributions
 com-0.017
 icy-0.016
 warm-0.016
 grass-0.015
 refreshing-0.015
Token "
Token ID366
Feature activation+3.81e-02

Pos logit contributions
 com+0.009
 warm+0.008
 icy+0.008
 grass+0.008
 soft+0.008
Neg logit contributions
mate-0.014
orman-0.013
uing-0.013
oker-0.013
anish-0.013
TokenThe
Token ID464
Feature activation+7.18e-02

Pos logit contributions
 com+0.048
 icy+0.043
 warm+0.042
 grass+0.041
 orange+0.040
Neg logit contributions
mate-0.070
orman-0.068
anish-0.068
oker-0.065
date-0.064
Token ost
Token ID23619
Feature activation+8.24e-02

Pos logit contributions
 com+0.105
 warm+0.096
 icy+0.095
 soft+0.090
 grass+0.088
Neg logit contributions
mate-0.121
orman-0.115
anish-0.111
 Troll-0.110
 policy-0.110
Tokenric
Token ID1173
Feature activation+2.86e-01

Pos logit contributions
 com+0.066
 icy+0.066
 warm+0.063
 grass+0.057
 clouds+0.057
Neg logit contributions
mate-0.188
orman-0.184
 policy-0.178
oker-0.176
date-0.176
Tokenhes
Token ID956
Feature activation+1.02e-01

Pos logit contributions
 com+0.196
 warm+0.168
 icy+0.167
 grass+0.144
 soft+0.139
Neg logit contributions
mate-0.702
orman-0.691
anish-0.674
oker-0.658
uing-0.653
Token are
Token ID389
Feature activation+5.58e-02

Pos logit contributions
 icy+0.127
 warm+0.125
 com+0.123
 very+0.122
 grass+0.115
Neg logit contributions
orman-0.188
uer-0.188
 Majesty-0.186
 avail-0.186
uers-0.185
Token mean
Token ID1612
Feature activation-7.24e-02

Pos logit contributions
 com+0.061
 icy+0.054
 grass+0.054
 clouds+0.053
 warm+0.051
Neg logit contributions
mate-0.105
 begs-0.097
orman-0.097
rum-0.097
anish-0.096
Token!
Token ID0
Feature activation+5.35e-03

Pos logit contributions
mate+0.136
orman+0.135
date+0.130
uer+0.130
 avail+0.129
Neg logit contributions
 com-0.091
 icy-0.090
 warm-0.090
 very-0.087
 grass-0.086
Token They
Token ID1119
Feature activation+2.06e-02

Pos logit contributions
 com+0.007
 icy+0.006
 warm+0.006
 soft+0.006
 clouds+0.006
Neg logit contributions
mate-0.010
orman-0.010
date-0.010
uing-0.010
uer-0.010
Token shouted
Token ID17293
Feature activation-6.00e-02

Pos logit contributions
mate+0.162
orman+0.152
anish+0.150
date+0.150
 policy+0.149
Neg logit contributions
 com-0.065
 icy-0.060
 warm-0.059
 grass-0.056
 very-0.054
Token.
Token ID13
Feature activation+8.39e-03

Pos logit contributions
mate+0.128
 occasions+0.127
anish+0.127
 Majesty+0.124
oker+0.123
Neg logit contributions
 warm-0.066
 very-0.066
 com-0.063
 even-0.061
 soft-0.059
Token "
Token ID366
Feature activation+3.80e-02

Pos logit contributions
 very+0.007
 com+0.007
 warm+0.006
 soft+0.006
 a+0.006
Neg logit contributions
 occasions-0.020
uing-0.020
 Majesty-0.020
oker-0.019
 avail-0.019
TokenThe
Token ID464
Feature activation+7.27e-02

Pos logit contributions
 com+0.051
 icy+0.048
 warm+0.046
 refreshing+0.045
 soft+0.043
Neg logit contributions
mate-0.069
orman-0.066
anish-0.066
fer-0.063
date-0.062
Token ost
Token ID23619
Feature activation+8.12e-02

Pos logit contributions
 com+0.093
 icy+0.084
 warm+0.084
 soft+0.078
 very+0.077
Neg logit contributions
mate-0.133
orman-0.128
 Troll-0.127
anish-0.125
 policy-0.124
Tokenric
Token ID1173
Feature activation+2.85e-01

Pos logit contributions
 com+0.089
 icy+0.082
 warm+0.081
 soft+0.074
 grass+0.074
Neg logit contributions
mate-0.168
orman-0.161
anish-0.159
 Troll-0.156
uing-0.155
Tokenhes
Token ID956
Feature activation+9.86e-02

Pos logit contributions
 com+0.182
 icy+0.158
 warm+0.157
 grass+0.132
 soft+0.130
Neg logit contributions
mate-0.713
orman-0.700
anish-0.682
 Troll-0.668
oker-0.667
Token are
Token ID389
Feature activation+5.72e-02

Pos logit contributions
 icy+0.117
 warm+0.115
 very+0.113
 com+0.112
 grass+0.107
Neg logit contributions
orman-0.186
uer-0.186
 Majesty-0.184
 avail-0.184
mate-0.183
Token so
Token ID523
Feature activation-1.82e-03

Pos logit contributions
 com+0.052
 icy+0.047
 warm+0.045
 grass+0.043
 clouds+0.042
Neg logit contributions
mate-0.121
orman-0.117
 Troll-0.113
 password-0.113
anish-0.112
Token funny
Token ID8258
Feature activation-4.71e-02

Pos logit contributions
orman+0.004
oker+0.004
uers+0.004
mate+0.004
ings+0.004
Neg logit contributions
 com-0.001
 vel-0.001
 icy-0.001
 grass-0.001
 warm-0.001
Token!"
Token ID2474
Feature activation-2.04e-02

Pos logit contributions
mate+0.071
orman+0.067
anish+0.064
date+0.064
 policy+0.064
Neg logit contributions
 com-0.078
 icy-0.075
 warm-0.075
 grass-0.072
 soft-0.071
Token but
Token ID475
Feature activation+1.23e-02

Pos logit contributions
mate+0.064
orman+0.060
anish+0.059
 policy+0.057
uing+0.057
Neg logit contributions
 com-0.058
 icy-0.056
 warm-0.055
 grass-0.052
 soft-0.052
Token be
Token ID307
Feature activation+8.25e-02

Pos logit contributions
 com+0.018
 icy+0.017
 warm+0.017
 grass+0.016
 soft+0.016
Neg logit contributions
mate-0.021
orman-0.020
anish-0.019
date-0.019
oker-0.019
Token careful
Token ID8161
Feature activation-7.09e-02

Pos logit contributions
 com+0.118
 icy+0.110
 clouds+0.107
 grass+0.106
 warm+0.104
Neg logit contributions
mate-0.130
orman-0.129
uing-0.121
 Troll-0.121
rum-0.120
Token.
Token ID13
Feature activation+5.37e-03

Pos logit contributions
mate+0.121
orman+0.115
anish+0.112
 policy+0.110
uing+0.110
Neg logit contributions
 com-0.102
 icy-0.096
 warm-0.095
 grass-0.090
 soft-0.089
Token Ost
Token ID38919
Feature activation+4.73e-02

Pos logit contributions
 com+0.009
 icy+0.008
 warm+0.008
 grass+0.008
 soft+0.008
Neg logit contributions
mate-0.008
orman-0.008
anish-0.008
 Troll-0.007
 policy-0.007
Tokenric
Token ID1173
Feature activation+2.85e-01

Pos logit contributions
 com+0.072
 warm+0.061
 icy+0.061
 boun+0.060
 soft+0.060
Neg logit contributions
mate-0.080
anish-0.079
 Troll-0.077
orman-0.077
ings-0.076
Tokenhes
Token ID956
Feature activation+9.67e-02

Pos logit contributions
 com+0.216
 icy+0.186
 warm+0.185
 grass+0.159
 vel+0.158
Neg logit contributions
mate-0.674
orman-0.670
anish-0.649
oker-0.629
 Troll-0.629
Token can
Token ID460
Feature activation+4.74e-03

Pos logit contributions
 com+0.130
 icy+0.128
 warm+0.127
 very+0.120
 grass+0.120
Neg logit contributions
mate-0.170
orman-0.166
uer-0.160
 policy-0.159
 Troll-0.158
Token be
Token ID307
Feature activation+8.27e-02

Pos logit contributions
 com+0.008
 icy+0.007
 warm+0.007
 grass+0.007
 soft+0.007
Neg logit contributions
mate-0.007
orman-0.007
anish-0.006
uing-0.006
uers-0.006
Token gr
Token ID1036
Feature activation+7.65e-02

Pos logit contributions
 com+0.086
 icy+0.077
 warm+0.074
 grass+0.072
 clouds+0.071
Neg logit contributions
mate-0.166
orman-0.161
uing-0.155
anish-0.155
uer-0.155
Tokenumpy
Token ID32152
Feature activation-5.23e-03

Pos logit contributions
 com+0.163
 warm+0.156
 icy+0.155
 grass+0.150
 soft+0.149
Neg logit contributions
mate-0.076
orman-0.073
anish-0.071
oker-0.069
 purposes-0.068
Token.
Token ID13
Feature activation+5.36e-03

Pos logit contributions
mate+0.058
orman+0.055
anish+0.054
date+0.053
 Troll+0.053
Neg logit contributions
 com-0.043
 warm-0.041
 icy-0.041
 very-0.038
 soft-0.038
Token It
Token ID632
Feature activation+6.58e-02

Pos logit contributions
 com+0.007
 icy+0.007
 warm+0.007
 grass+0.006
 soft+0.006
Neg logit contributions
mate-0.010
orman-0.009
anish-0.009
uing-0.009
 policy-0.009
Token sees
Token ID7224
Feature activation-5.41e-02

Pos logit contributions
 com+0.102
 icy+0.100
 warm+0.099
 soft+0.093
 grass+0.092
Neg logit contributions
mate-0.099
orman-0.096
 Troll-0.091
anish-0.091
 begs-0.091
Token other
Token ID584
Feature activation-3.08e-02

Pos logit contributions
mate+0.120
orman+0.116
anish+0.113
 Troll+0.112
 jury+0.109
Neg logit contributions
 com-0.048
 icy-0.044
 warm-0.042
 soft-0.036
 vel-0.036
Token ost
Token ID23619
Feature activation+7.98e-02

Pos logit contributions
mate+0.067
orman+0.064
anish+0.063
date+0.062
uing+0.062
Neg logit contributions
 com-0.031
 icy-0.029
 warm-0.029
 grass-0.026
 soft-0.026
Tokenric
Token ID1173
Feature activation+2.85e-01

Pos logit contributions
 com+0.098
 icy+0.090
 warm+0.090
 soft+0.083
 grass+0.083
Neg logit contributions
mate-0.155
orman-0.149
anish-0.148
uing-0.145
 Troll-0.145
Tokenhes
Token ID956
Feature activation+9.60e-02

Pos logit contributions
 com+0.189
 icy+0.167
 warm+0.164
 grass+0.141
 soft+0.139
Neg logit contributions
mate-0.702
orman-0.688
anish-0.670
 Troll-0.658
uing-0.655
Token.
Token ID13
Feature activation+5.74e-03

Pos logit contributions
 com+0.093
 icy+0.087
 warm+0.087
 grass+0.079
 soft+0.078
Neg logit contributions
mate-0.208
orman-0.202
anish-0.197
 Troll-0.194
 policy-0.194
Token It
Token ID632
Feature activation+6.63e-02

Pos logit contributions
 com+0.007
 icy+0.007
 warm+0.007
 grass+0.006
 soft+0.006
Neg logit contributions
mate-0.011
orman-0.010
anish-0.010
 Troll-0.010
oker-0.010
Token makes
Token ID1838
Feature activation-6.27e-03

Pos logit contributions
 com+0.104
 icy+0.101
 warm+0.101
 soft+0.094
 grass+0.094
Neg logit contributions
mate-0.098
orman-0.095
 Troll-0.091
anish-0.090
 begs-0.090
Token new
Token ID649
Feature activation+1.92e-02

Pos logit contributions
mate+0.014
orman+0.014
anish+0.014
 Troll+0.013
uing+0.013
Neg logit contributions
 com-0.005
 icy-0.005
 warm-0.005
 grass-0.004
 soft-0.004
Token.
Token ID13
Feature activation+1.61e-03

Pos logit contributions
 com+0.030
 icy+0.028
 warm+0.027
 grass+0.025
 clouds+0.025
Neg logit contributions
mate-0.057
orman-0.055
anish-0.052
 Troll-0.052
 policy-0.052
Token One
Token ID1881
Feature activation-1.56e-02

Pos logit contributions
 com+0.002
 icy+0.002
 warm+0.002
 grass+0.002
 soft+0.002
Neg logit contributions
mate-0.003
orman-0.003
uing-0.003
anish-0.003
uer-0.003
Token of
Token ID286
Feature activation-4.29e-02

Pos logit contributions
mate+0.029
orman+0.028
anish+0.027
 Troll+0.027
 policy+0.027
Neg logit contributions
 com-0.020
 icy-0.019
 warm-0.018
 grass-0.017
 soft-0.017
Token the
Token ID262
Feature activation+4.42e-02

Pos logit contributions
mate+0.088
orman+0.080
 Troll+0.080
anish+0.079
 avail+0.076
Neg logit contributions
 com-0.047
 icy-0.042
 warm-0.042
 refreshing-0.038
 vel-0.038
Token ost
Token ID23619
Feature activation+8.12e-02

Pos logit contributions
 com+0.046
 icy+0.045
 warm+0.044
 very+0.041
 grass+0.041
Neg logit contributions
mate-0.098
orman-0.093
anish-0.093
oker-0.092
uing-0.091
Tokenric
Token ID1173
Feature activation+2.85e-01

Pos logit contributions
 com+0.091
 icy+0.084
 warm+0.084
 soft+0.077
 grass+0.076
Neg logit contributions
mate-0.165
orman-0.159
anish-0.156
 Troll-0.154
uing-0.153
Tokenhes
Token ID956
Feature activation+9.63e-02

Pos logit contributions
 com+0.175
 icy+0.154
 warm+0.151
 grass+0.127
 soft+0.126
Neg logit contributions
mate-0.718
orman-0.701
anish-0.685
 Troll-0.674
uing-0.671
Token looked
Token ID3114
Feature activation+6.02e-02

Pos logit contributions
 icy+0.091
 warm+0.091
 very+0.087
 com+0.087
 grass+0.085
Neg logit contributions
 Majesty-0.207
mate-0.203
uer-0.201
 policy-0.200
orman-0.200
Token different
Token ID1180
Feature activation-1.28e-02

Pos logit contributions
 com+0.083
 icy+0.076
 warm+0.075
 clouds+0.072
 vel+0.072
Neg logit contributions
mate-0.095
orman-0.092
uing-0.089
 begs-0.089
 Troll-0.088
Token from
Token ID422
Feature activation-2.26e-02

Pos logit contributions
mate+0.029
orman+0.028
oker+0.027
date+0.027
anish+0.027
Neg logit contributions
 com-0.012
 icy-0.011
 warm-0.011
 grass-0.010
 soft-0.010
Token the
Token ID262
Feature activation+4.49e-02

Pos logit contributions
mate+0.048
orman+0.046
 Troll+0.045
anish+0.044
 Majesty+0.044
Neg logit contributions
 com-0.022
 icy-0.019
 warm-0.019
 soft-0.017
 grass-0.017
Token and
Token ID290
Feature activation+1.78e-01

Pos logit contributions
 com+0.056
 icy+0.052
 warm+0.051
 grass+0.046
 soft+0.046
Neg logit contributions
mate-0.142
orman-0.137
anish-0.134
 Troll-0.132
 policy-0.132
Token run
Token ID1057
Feature activation+7.17e-02

Pos logit contributions
 com+0.288
 icy+0.277
 warm+0.272
 soft+0.260
 grass+0.255
Neg logit contributions
mate-0.261
orman-0.248
 Troll-0.239
anish-0.238
 begs-0.237
Token to
Token ID284
Feature activation-8.71e-02

Pos logit contributions
 com+0.089
 warm+0.087
 icy+0.086
 very+0.083
 grass+0.082
Neg logit contributions
mate-0.135
orman-0.130
anish-0.130
oker-0.127
 policy-0.127
Token the
Token ID262
Feature activation+4.34e-02

Pos logit contributions
mate+0.224
anish+0.220
oker+0.218
uer+0.218
orman+0.218
Neg logit contributions
 warm-0.054
 com-0.054
 icy-0.050
 grass-0.046
 soft-0.045
Token ost
Token ID23619
Feature activation+8.01e-02

Pos logit contributions
 com+0.054
 icy+0.054
 warm+0.053
 grass+0.052
 soft+0.050
Neg logit contributions
mate-0.084
orman-0.080
uing-0.079
anish-0.079
oker-0.078
Tokenric
Token ID1173
Feature activation+2.84e-01

Pos logit contributions
 com+0.077
 icy+0.073
 warm+0.071
 grass+0.066
 soft+0.065
Neg logit contributions
mate-0.174
orman-0.167
anish-0.163
 policy-0.161
oker-0.161
Tokenhes
Token ID956
Feature activation+9.91e-02

Pos logit contributions
 com+0.192
 icy+0.166
 warm+0.165
 grass+0.140
 soft+0.139
Neg logit contributions
mate-0.700
orman-0.690
anish-0.667
oker-0.654
 Troll-0.653
Token.
Token ID13
Feature activation+3.02e-03

Pos logit contributions
 com+0.121
 warm+0.120
 icy+0.118
 very+0.113
 grass+0.111
Neg logit contributions
mate-0.185
orman-0.181
uer-0.180
 Majesty-0.178
 Troll-0.178
Token
Token ID198
Feature activation+2.22e-02

Pos logit contributions
 com+0.003
 warm+0.003
 icy+0.003
 grass+0.003
 soft+0.003
Neg logit contributions
mate-0.006
uing-0.006
anish-0.006
date-0.006
uer-0.006
Token
Token ID198
Feature activation+2.33e-02

Pos logit contributions
 com+0.025
 icy+0.024
 warm+0.024
 grass+0.022
 soft+0.022
Neg logit contributions
mate-0.044
orman-0.043
anish-0.042
 Troll-0.041
 policy-0.041
Token"
Token ID1
Feature activation+3.39e-03

Pos logit contributions
 com+0.040
 icy+0.038
 warm+0.038
 soft+0.036
 grass+0.036
Neg logit contributions
mate-0.033
orman-0.032
anish-0.030
 Troll-0.029
 policy-0.029
Token.
Token ID13
Feature activation+1.49e-03

Pos logit contributions
 com+0.119
 icy+0.112
 warm+0.110
 grass+0.104
 soft+0.104
Neg logit contributions
mate-0.116
orman-0.110
anish-0.106
uing-0.105
date-0.104
Token They
Token ID1119
Feature activation+1.17e-02

Pos logit contributions
 com+0.002
 warm+0.002
 icy+0.002
 grass+0.002
 soft+0.002
Neg logit contributions
mate-0.003
orman-0.003
uing-0.003
 policy-0.003
uer-0.003
Token were
Token ID547
Feature activation+6.07e-02

Pos logit contributions
 com+0.016
 icy+0.015
 warm+0.015
 fresh+0.014
 soft+0.014
Neg logit contributions
mate-0.019
orman-0.018
 Troll-0.018
anish-0.018
 jury-0.017
Token called
Token ID1444
Feature activation+1.03e-02

Pos logit contributions
 com+0.051
 icy+0.046
 warm+0.043
 grass+0.042
 vel+0.040
Neg logit contributions
mate-0.127
orman-0.127
 Troll-0.123
 Majesty-0.123
 password-0.120
Token ost
Token ID23619
Feature activation+7.88e-02

Pos logit contributions
 com+0.014
 icy+0.013
 warm+0.013
 grass+0.012
 soft+0.012
Neg logit contributions
mate-0.018
orman-0.018
anish-0.017
 Troll-0.017
 policy-0.017
Tokenric
Token ID1173
Feature activation+2.84e-01

Pos logit contributions
 com+0.094
 icy+0.088
 warm+0.087
 grass+0.080
 soft+0.080
Neg logit contributions
mate-0.154
orman-0.148
anish-0.144
 Troll-0.142
uing-0.142
Tokenhes
Token ID956
Feature activation+9.65e-02

Pos logit contributions
 com+0.225
 icy+0.198
 warm+0.197
 grass+0.173
 soft+0.169
Neg logit contributions
mate-0.664
orman-0.656
anish-0.632
 Troll-0.621
uing-0.616
Token.
Token ID13
Feature activation+1.40e-03

Pos logit contributions
 com+0.098
 warm+0.096
 icy+0.096
 very+0.089
 grass+0.088
Neg logit contributions
mate-0.201
orman-0.195
 policy-0.191
anish-0.191
uer-0.190
Token Lily
Token ID20037
Feature activation-5.44e-02

Pos logit contributions
 com+0.002
 warm+0.002
 icy+0.002
 grass+0.002
 soft+0.001
Neg logit contributions
mate-0.003
orman-0.003
uing-0.003
anish-0.003
uer-0.003
Token and
Token ID290
Feature activation+1.74e-01

Pos logit contributions
orman+0.104
mate+0.103
 Troll+0.099
uer+0.098
anish+0.098
Neg logit contributions
 com-0.061
 icy-0.058
 warm-0.057
 fresh-0.053
 soft-0.053
Token Ben
Token ID3932
Feature activation-7.74e-02

Pos logit contributions
 com+0.234
 icy+0.225
 warm+0.224
 grass+0.215
 soft+0.210
Neg logit contributions
mate-0.318
uing-0.302
anish-0.301
uer-0.300
 policy-0.299
Token that
Token ID326
Feature activation-1.01e-01

Pos logit contributions
 com+0.087
 icy+0.083
 warm+0.082
 grass+0.077
 soft+0.076
Neg logit contributions
mate-0.143
orman-0.137
anish-0.134
uing-0.131
date-0.131
Token said
Token ID531
Feature activation-1.31e-01

Pos logit contributions
orman+0.145
mate+0.144
 Troll+0.138
 password+0.135
anish+0.133
Neg logit contributions
 com-0.161
 icy-0.154
 warm-0.150
 soft-0.142
 fresh-0.142
Token "
Token ID366
Feature activation+3.64e-02

Pos logit contributions
mate+0.258
 occasions+0.252
uer+0.251
oker+0.250
date+0.249
Neg logit contributions
 com-0.145
 warm-0.143
 very-0.139
 icy-0.139
 soft-0.136
TokenO
Token ID46
Feature activation+7.26e-02

Pos logit contributions
 com+0.048
 warm+0.046
 icy+0.046
 clouds+0.043
 grass+0.043
Neg logit contributions
mate-0.064
orman-0.062
 Majesty-0.062
uing-0.061
 policy-0.061
Tokenst
Token ID301
Feature activation+4.90e-02

Pos logit contributions
 com+0.128
 icy+0.127
 warm+0.123
 grass+0.117
 very+0.117
Neg logit contributions
mate-0.099
orman-0.091
anish-0.090
date-0.088
 policy-0.086
Tokenric
Token ID1173
Feature activation+2.84e-01

Pos logit contributions
 com+0.082
 icy+0.081
 warm+0.080
 very+0.078
 soft+0.078
Neg logit contributions
 avail-0.066
mate-0.066
date-0.064
 policy-0.063
 occasions-0.062
Tokenhes
Token ID956
Feature activation+9.84e-02

Pos logit contributions
 com+0.233
 icy+0.213
 warm+0.210
 grass+0.187
 soft+0.185
Neg logit contributions
mate-0.657
orman-0.636
anish-0.622
 Troll-0.611
uing-0.610
Token".
Token ID1911
Feature activation-1.37e-02

Pos logit contributions
 warm+0.139
 icy+0.138
 com+0.136
 very+0.134
 soft+0.129
Neg logit contributions
mate-0.158
orman-0.158
uer-0.157
 avail-0.154
 Majesty-0.152
Token Sara
Token ID24799
Feature activation-1.39e-02

Pos logit contributions
mate+0.025
anish+0.023
uing+0.023
uer+0.023
oker+0.023
Neg logit contributions
 com-0.019
 icy-0.018
 warm-0.018
 grass-0.018
 soft-0.017
Token and
Token ID290
Feature activation+1.70e-01

Pos logit contributions
mate+0.028
orman+0.028
anish+0.027
 Troll+0.027
uer+0.026
Neg logit contributions
 com-0.015
 icy-0.014
 warm-0.014
 soft-0.013
 grass-0.013
Token Tom
Token ID4186
Feature activation-7.46e-02

Pos logit contributions
 com+0.280
 icy+0.267
 warm+0.264
 soft+0.251
 grass+0.250
Neg logit contributions
mate-0.256
orman-0.244
anish-0.236
 Troll-0.230
 policy-0.229
Token you
Token ID345
Feature activation-5.83e-02

Pos logit contributions
mate+0.081
orman+0.075
anish+0.074
 Troll+0.072
uers+0.071
Neg logit contributions
 com-0.125
 icy-0.121
 warm-0.118
 grass-0.113
 vel-0.112
Token count
Token ID954
Feature activation+7.85e-03

Pos logit contributions
mate+0.090
orman+0.086
anish+0.083
 policy+0.082
 Troll+0.081
Neg logit contributions
 com-0.093
 icy-0.089
 warm-0.089
 soft-0.084
 grass-0.084
Token how
Token ID703
Feature activation+5.17e-02

Pos logit contributions
 com+0.009
 icy+0.009
 warm+0.009
 refreshing+0.008
 grass+0.008
Neg logit contributions
mate-0.015
uing-0.014
anish-0.013
orman-0.013
ings-0.013
Token many
Token ID867
Feature activation-8.61e-03

Pos logit contributions
 com+0.066
 icy+0.061
 warm+0.060
 vel+0.059
 refreshing+0.058
Neg logit contributions
mate-0.088
orman-0.086
 purposes-0.081
anish-0.080
 Troll-0.080
Token ost
Token ID23619
Feature activation+7.98e-02

Pos logit contributions
mate+0.015
 purposes+0.014
orman+0.014
 occasions+0.014
 Troll+0.014
Neg logit contributions
 com-0.011
 icy-0.010
 warm-0.010
 vel-0.009
 grass-0.009
Tokenric
Token ID1173
Feature activation+2.84e-01

Pos logit contributions
 com+0.101
 warm+0.093
 icy+0.093
 soft+0.086
 grass+0.085
Neg logit contributions
mate-0.150
orman-0.144
anish-0.144
 Troll-0.142
uing-0.141
Tokenhes
Token ID956
Feature activation+9.75e-02

Pos logit contributions
 com+0.275
 icy+0.254
 warm+0.252
 grass+0.229
 soft+0.227
Neg logit contributions
mate-0.614
orman-0.594
anish-0.580
 Troll-0.569
oker-0.567
Token there
Token ID612
Feature activation-1.36e-01

Pos logit contributions
 com+0.119
 warm+0.116
 icy+0.116
 very+0.108
 soft+0.106
Neg logit contributions
mate-0.184
orman-0.182
uer-0.175
 policy-0.174
anish-0.173
Token are
Token ID389
Feature activation+5.63e-02

Pos logit contributions
mate+0.209
orman+0.195
anish+0.192
date+0.187
uing+0.187
Neg logit contributions
 com-0.214
 icy-0.204
 warm-0.200
 grass-0.192
 soft-0.190
Token?"
Token ID1701
Feature activation+1.14e-02

Pos logit contributions
 com+0.068
 icy+0.064
 warm+0.063
 grass+0.059
 soft+0.059
Neg logit contributions
mate-0.108
orman-0.104
anish-0.101
 Troll-0.099
 policy-0.099
Token
Token ID198
Feature activation+2.31e-02

Pos logit contributions
 com+0.014
 grass+0.013
 warm+0.013
 very+0.013
 sun+0.013
Neg logit contributions
mate-0.024
uer-0.023
fer-0.023
 policy-0.022
uing-0.022
TokenYes
Token ID5297
Feature activation-2.01e-02

Pos logit contributions
 com+0.007
 icy+0.006
 warm+0.006
 cold+0.006
 refreshing+0.006
Neg logit contributions
mate-0.006
anish-0.006
 Troll-0.006
 jury-0.006
date-0.006
Token,
Token ID11
Feature activation-4.00e-02

Pos logit contributions
orman+0.039
mate+0.039
 occasions+0.038
date+0.038
 jury+0.038
Neg logit contributions
 warm-0.026
 com-0.025
 icy-0.025
 very-0.024
 soft-0.024
Token they
Token ID484
Feature activation-6.44e-02

Pos logit contributions
mate+0.070
anish+0.066
orman+0.065
 Troll+0.064
 password+0.063
Neg logit contributions
 com-0.053
 icy-0.050
 warm-0.049
 grass-0.046
 vel-0.046
Token are
Token ID389
Feature activation+5.75e-02

Pos logit contributions
mate+0.099
orman+0.095
anish+0.094
 Troll+0.091
date+0.090
Neg logit contributions
 com-0.101
 icy-0.096
 warm-0.094
 grass-0.090
 soft-0.089
Token ost
Token ID23619
Feature activation+7.99e-02

Pos logit contributions
 com+0.089
 warm+0.087
 icy+0.085
 very+0.083
 soft+0.082
Neg logit contributions
mate-0.099
orman-0.092
anish-0.091
date-0.090
oker-0.089
Tokenric
Token ID1173
Feature activation+2.84e-01

Pos logit contributions
 com+0.092
 icy+0.086
 warm+0.085
 grass+0.079
 soft+0.079
Neg logit contributions
mate-0.160
orman-0.153
anish-0.150
 Troll-0.147
uing-0.147
Tokenhes
Token ID956
Feature activation+9.84e-02

Pos logit contributions
 com+0.207
 icy+0.187
 warm+0.184
 grass+0.161
 soft+0.159
Neg logit contributions
mate-0.682
orman-0.662
anish-0.647
 Troll-0.637
uing-0.635
Token,"
Token ID553
Feature activation-2.09e-02

Pos logit contributions
 com+0.129
 warm+0.129
 icy+0.127
 very+0.122
 soft+0.120
Neg logit contributions
mate-0.179
orman-0.177
uer-0.171
 policy-0.169
 Troll-0.168
Token mom
Token ID1995
Feature activation-4.07e-02

Pos logit contributions
mate+0.040
uer+0.038
orman+0.038
uing+0.038
 policy+0.037
Neg logit contributions
 warm-0.029
 com-0.029
 icy-0.027
 soft-0.027
 very-0.027
Token said
Token ID531
Feature activation-1.25e-01

Pos logit contributions
mate+0.076
orman+0.071
anish+0.070
 policy+0.070
uer+0.069
Neg logit contributions
 com-0.054
 warm-0.051
 icy-0.051
 grass-0.048
 soft-0.047
Token.
Token ID13
Feature activation+4.45e-03

Pos logit contributions
mate+0.272
orman+0.262
reetings+0.260
 Majesty+0.258
anish+0.257
Neg logit contributions
 warm-0.140
 very-0.134
 com-0.131
 soft-0.126
 so-0.121
TokenLook
Token ID8567
Feature activation-1.00e-01

Pos logit contributions
 com+0.049
 icy+0.047
 warm+0.046
 soft+0.043
 grass+0.042
Neg logit contributions
mate-0.067
orman-0.064
anish-0.063
 Troll-0.061
date-0.061
Token,
Token ID11
Feature activation-4.06e-02

Pos logit contributions
mate+0.184
 policy+0.182
 occasions+0.182
orman+0.179
oker+0.177
Neg logit contributions
 com-0.137
 warm-0.137
 very-0.132
 icy-0.128
 all-0.128
Token mom
Token ID1995
Feature activation-4.10e-02

Pos logit contributions
mate+0.077
orman+0.074
anish+0.072
 Troll+0.071
 policy+0.070
Neg logit contributions
 com-0.050
 icy-0.048
 warm-0.047
 grass-0.044
 soft-0.044
Token,
Token ID11
Feature activation-4.07e-02

Pos logit contributions
mate+0.069
orman+0.067
anish+0.064
uer+0.063
oker+0.063
Neg logit contributions
 com-0.061
 icy-0.059
 warm-0.058
 grass-0.055
 soft-0.055
Token ost
Token ID23619
Feature activation+8.01e-02

Pos logit contributions
mate+0.074
anish+0.071
 Troll+0.070
orman+0.070
 Majesty+0.069
Neg logit contributions
 com-0.050
 icy-0.048
 warm-0.046
 grass-0.044
 soft-0.042
Tokenric
Token ID1173
Feature activation+2.83e-01

Pos logit contributions
 com+0.082
 icy+0.077
 warm+0.076
 grass+0.069
 soft+0.069
Neg logit contributions
mate-0.170
orman-0.163
anish-0.160
 Troll-0.157
 policy-0.157
Tokenhes
Token ID956
Feature activation+9.93e-02

Pos logit contributions
 com+0.196
 icy+0.175
 warm+0.172
 grass+0.148
 soft+0.146
Neg logit contributions
mate-0.691
orman-0.675
anish-0.659
 Troll-0.648
uing-0.644
Token!"
Token ID2474
Feature activation-2.01e-02

Pos logit contributions
 com+0.127
 warm+0.125
 icy+0.124
 very+0.117
 soft+0.116
Neg logit contributions
mate-0.182
orman-0.179
 policy-0.173
uer-0.173
 Troll-0.171
Token
Token ID198
Feature activation+2.41e-02

Pos logit contributions
mate+0.033
orman+0.031
 policy+0.030
oker+0.030
uer+0.030
Neg logit contributions
 com-0.031
 warm-0.030
 icy-0.030
 grass-0.029
 soft-0.028
Token
Token ID198
Feature activation+2.43e-02

Pos logit contributions
 com+0.022
 warm+0.021
 icy+0.020
 soft+0.019
 very+0.019
Neg logit contributions
mate-0.054
anish-0.051
oker-0.051
 Majesty-0.051
orman-0.051
Token"
Token ID1
Feature activation+4.18e-03

Pos logit contributions
 com+0.029
 warm+0.029
 icy+0.029
 grass+0.027
 soft+0.027
Neg logit contributions
mate-0.046
 policy-0.044
orman-0.044
oker-0.044
uer-0.043
Token did
Token ID750
Feature activation-3.66e-02

Pos logit contributions
mate+0.132
orman+0.129
anish+0.125
 Troll+0.124
uing+0.123
Neg logit contributions
 com-0.098
 icy-0.092
 warm-0.090
 grass-0.085
 soft-0.084
Token not
Token ID407
Feature activation+6.73e-03

Pos logit contributions
mate+0.076
orman+0.073
anish+0.071
uing+0.070
 Troll+0.070
Neg logit contributions
 com-0.039
 icy-0.036
 warm-0.036
 grass-0.033
 soft-0.033
Token know
Token ID760
Feature activation-8.60e-02

Pos logit contributions
 com+0.010
 warm+0.010
 icy+0.010
 very+0.010
 soft+0.009
Neg logit contributions
mate-0.011
orman-0.010
anish-0.010
 Majesty-0.010
uer-0.010
Token what
Token ID644
Feature activation-4.06e-02

Pos logit contributions
mate+0.159
orman+0.152
anish+0.147
 Troll+0.147
uing+0.146
Neg logit contributions
 com-0.108
 icy-0.100
 warm-0.100
 soft-0.092
 grass-0.092
Token ost
Token ID23619
Feature activation+7.89e-02

Pos logit contributions
mate+0.087
orman+0.083
anish+0.083
uer+0.082
uing+0.082
Neg logit contributions
 com-0.041
 icy-0.038
 warm-0.038
 grass-0.035
 very-0.035
Tokenric
Token ID1173
Feature activation+2.83e-01

Pos logit contributions
 com+0.079
 icy+0.075
 warm+0.074
 grass+0.068
 soft+0.067
Neg logit contributions
mate-0.167
orman-0.161
anish-0.157
 policy-0.155
oker-0.154
Tokenhes
Token ID956
Feature activation+9.76e-02

Pos logit contributions
 com+0.228
 icy+0.212
 warm+0.208
 grass+0.186
 soft+0.185
Neg logit contributions
mate-0.656
orman-0.632
anish-0.620
 policy-0.615
date-0.612
Token were
Token ID547
Feature activation+5.70e-02

Pos logit contributions
 warm+0.098
 icy+0.097
 com+0.096
 very+0.091
 grass+0.089
Neg logit contributions
mate-0.202
uer-0.201
orman-0.199
 Majesty-0.196
anish-0.193
Token.
Token ID13
Feature activation-6.65e-04

Pos logit contributions
 icy+0.064
 warm+0.064
 very+0.063
 com+0.061
 grass+0.060
Neg logit contributions
mate-0.121
orman-0.114
anish-0.112
date-0.111
oker-0.111
Token They
Token ID1119
Feature activation+1.31e-02

Pos logit contributions
mate+0.001
orman+0.001
anish+0.001
 policy+0.001
uing+0.001
Neg logit contributions
 com-0.001
 icy-0.001
 warm-0.001
 grass-0.001
 soft-0.001
Token looked
Token ID3114
Feature activation+5.87e-02

Pos logit contributions
 com+0.019
 icy+0.018
 warm+0.018
 soft+0.017
 fresh+0.017
Neg logit contributions
mate-0.021
orman-0.021
 Troll-0.020
anish-0.020
uers-0.020
Token some
Token ID617
Feature activation+1.51e-02

Pos logit contributions
mate+0.012
orman+0.012
 avail+0.011
 begs+0.011
ings+0.011
Neg logit contributions
 com-0.006
 icy-0.006
 refreshing-0.006
 warm-0.005
 clouds-0.005
Token snow
Token ID6729
Feature activation+9.14e-02

Pos logit contributions
 com+0.017
 icy+0.016
 warm+0.016
 clouds+0.015
 refreshing+0.015
Neg logit contributions
mate-0.028
 occasions-0.027
 policy-0.026
date-0.026
 purposes-0.025
Token for
Token ID329
Feature activation-7.65e-02

Pos logit contributions
 com+0.124
 warm+0.121
 icy+0.121
 very+0.114
 grass+0.112
Neg logit contributions
mate-0.158
orman-0.156
anish-0.148
oker-0.148
 policy-0.148
Token a
Token ID257
Feature activation+1.89e-02

Pos logit contributions
mate+0.164
orman+0.157
anish+0.152
date+0.150
 Troll+0.150
Neg logit contributions
 com-0.077
 icy-0.072
 warm-0.071
 grass-0.065
 soft-0.064
Token play
Token ID711
Feature activation+2.23e-02

Pos logit contributions
 com+0.024
 warm+0.023
 icy+0.023
 soft+0.022
 very+0.021
Neg logit contributions
mate-0.036
orman-0.035
anish-0.035
oker-0.034
 Troll-0.034
Tokendate
Token ID4475
Feature activation-2.11e-01

Pos logit contributions
 com+0.027
 warm+0.027
 icy+0.026
 very+0.025
 grass+0.024
Neg logit contributions
orman-0.041
 Majesty-0.041
 purposes-0.041
uer-0.040
 Troll-0.040
Token with
Token ID351
Feature activation-1.33e-02

Pos logit contributions
 Majesty+0.476
anish+0.462
orman+0.461
oker+0.460
 Troll+0.459
Neg logit contributions
 very-0.204
 so-0.201
 warm-0.193
 all-0.188
 even-0.187
Token his
Token ID465
Feature activation-4.51e-02

Pos logit contributions
mate+0.025
orman+0.023
date+0.022
anish+0.022
 occasions+0.022
Neg logit contributions
 com-0.017
 icy-0.016
 warm-0.016
 soft-0.015
 grass-0.015
Token friends
Token ID2460
Feature activation-1.54e-02

Pos logit contributions
mate+0.079
orman+0.077
anish+0.075
oker+0.074
uing+0.073
Neg logit contributions
 com-0.065
 warm-0.063
 icy-0.062
 soft-0.059
 grass-0.058
Token.
Token ID13
Feature activation-5.82e-03

Pos logit contributions
orman+0.031
mate+0.031
anish+0.031
oker+0.030
 Majesty+0.030
Neg logit contributions
 very-0.016
 com-0.016
 warm-0.016
 icy-0.016
 so-0.016
Token He
Token ID679
Feature activation-6.74e-03

Pos logit contributions
mate+0.012
orman+0.011
uing+0.011
anish+0.011
 policy+0.011
Neg logit contributions
 com-0.007
 warm-0.006
 grass-0.006
 icy-0.006
 very-0.006
Token!
Token ID0
Feature activation-2.99e-03

Pos logit contributions
 com+0.077
 icy+0.073
 warm+0.073
 grass+0.069
 very+0.069
Neg logit contributions
mate-0.116
orman-0.115
anish-0.110
 Troll-0.109
oker-0.109
Token She
Token ID1375
Feature activation+3.09e-03

Pos logit contributions
mate+0.005
orman+0.005
anish+0.004
 Troll+0.004
oker+0.004
Neg logit contributions
 com-0.005
 icy-0.005
 warm-0.004
 soft-0.004
 grass-0.004
Token finished
Token ID5201
Feature activation-2.94e-02

Pos logit contributions
 icy+0.004
 clouds+0.004
 warm+0.004
 com+0.004
 fresh+0.004
Neg logit contributions
mate-0.004
orman-0.004
ra-0.004
uers-0.004
uer-0.004
Token her
Token ID607
Feature activation-4.37e-02

Pos logit contributions
mate+0.052
uing+0.049
orman+0.049
ings+0.047
anish+0.047
Neg logit contributions
 com-0.037
 icy-0.036
 warm-0.035
 clouds-0.034
 grass-0.033
Token play
Token ID711
Feature activation+2.91e-02

Pos logit contributions
mate+0.076
orman+0.073
 policy+0.070
uer+0.069
anish+0.069
Neg logit contributions
 com-0.058
 icy-0.055
 warm-0.054
 clouds-0.052
 grass-0.051
Tokendate
Token ID4475
Feature activation-2.11e-01

Pos logit contributions
 com+0.037
 warm+0.035
 icy+0.035
 grass+0.033
 soft+0.032
Neg logit contributions
orman-0.053
mate-0.053
 Troll-0.051
uer-0.051
anish-0.051
Token and
Token ID290
Feature activation+1.66e-01

Pos logit contributions
mate+0.493
orman+0.480
anish+0.474
 Troll+0.470
oker+0.465
Neg logit contributions
 com-0.174
 warm-0.163
 icy-0.157
 very-0.148
 grass-0.143
Token went
Token ID1816
Feature activation+2.65e-02

Pos logit contributions
 com+0.196
 icy+0.193
 warm+0.185
 clouds+0.175
 soft+0.174
Neg logit contributions
mate-0.307
orman-0.295
anish-0.280
oker-0.280
 policy-0.279
Token home
Token ID1363
Feature activation-8.75e-02

Pos logit contributions
 com+0.030
 warm+0.029
 icy+0.028
 very+0.027
 soft+0.026
Neg logit contributions
mate-0.054
anish-0.054
orman-0.052
 purposes-0.052
uer-0.051
Token,
Token ID11
Feature activation-4.49e-02

Pos logit contributions
anish+0.194
mate+0.193
 Troll+0.183
 avail+0.182
oker+0.181
Neg logit contributions
 com-0.100
 warm-0.097
 very-0.095
 soft-0.086
 icy-0.085
Token with
Token ID351
Feature activation-1.37e-02

Pos logit contributions
mate+0.075
orman+0.072
uer+0.071
 Majesty+0.070
 begs+0.069
Neg logit contributions
 clouds-0.053
 icy-0.051
 stars-0.047
 com-0.047
 grass-0.046
Token own
Token ID898
Feature activation-7.07e-03

Pos logit contributions
mate+0.122
orman+0.117
anish+0.113
uing+0.111
oker+0.111
Neg logit contributions
 com-0.081
 icy-0.077
 warm-0.077
 grass-0.072
 soft-0.071
Token toys
Token ID14958
Feature activation+3.54e-02

Pos logit contributions
mate+0.013
orman+0.013
anish+0.012
 Troll+0.012
uing+0.012
Neg logit contributions
 com-0.009
 icy-0.009
 warm-0.009
 grass-0.008
 soft-0.008
Token to
Token ID284
Feature activation-9.09e-02

Pos logit contributions
 com+0.046
 icy+0.043
 warm+0.042
 grass+0.040
 soft+0.040
Neg logit contributions
mate-0.064
orman-0.060
anish-0.059
uing-0.059
 policy-0.059
Token the
Token ID262
Feature activation+3.95e-02

Pos logit contributions
orman+0.175
mate+0.174
uer+0.172
anish+0.171
 policy+0.169
Neg logit contributions
 warm-0.115
 com-0.111
 icy-0.107
 soft-0.104
 very-0.101
Token play
Token ID711
Feature activation+2.59e-02

Pos logit contributions
 com+0.049
 icy+0.048
 warm+0.048
 grass+0.046
 soft+0.045
Neg logit contributions
mate-0.075
orman-0.073
rum-0.072
uing-0.071
anish-0.071
Tokendate
Token ID4475
Feature activation-2.11e-01

Pos logit contributions
 very+0.051
 warm+0.051
 icy+0.050
 grass+0.050
 com+0.049
Neg logit contributions
 Majesty-0.030
uer-0.028
 avail-0.027
 purposes-0.027
rum-0.027
Token.
Token ID13
Feature activation-2.23e-03

Pos logit contributions
 Majesty+0.447
mate+0.444
orman+0.441
 avail+0.440
reetings+0.437
Neg logit contributions
 very-0.227
 warm-0.216
 so-0.209
 all-0.204
 com-0.203
Token The
Token ID383
Feature activation+6.42e-02

Pos logit contributions
mate+0.004
orman+0.004
anish+0.004
uer+0.004
uing+0.004
Neg logit contributions
 com-0.003
 icy-0.003
 warm-0.003
 grass-0.003
 soft-0.003
Token first
Token ID717
Feature activation-8.13e-02

Pos logit contributions
 com+0.091
 warm+0.090
 icy+0.089
 soft+0.085
 grass+0.084
Neg logit contributions
mate-0.112
orman-0.109
rum-0.109
uing-0.106
anish-0.106
Token one
Token ID530
Feature activation-2.49e-02

Pos logit contributions
rum+0.132
rums+0.132
uing+0.131
mate+0.127
ides+0.126
Neg logit contributions
 com-0.119
 warm-0.117
 icy-0.116
 sun-0.114
 very-0.113
Token brought
Token ID3181
Feature activation-4.32e-02

Pos logit contributions
mate+0.053
 avail+0.051
orman+0.051
anish+0.051
 Majesty+0.051
Neg logit contributions
 com-0.026
 warm-0.025
 icy-0.024
 very-0.024
 soft-0.023
Token could
Token ID714
Feature activation-1.59e-02

Pos logit contributions
 com+0.193
 warm+0.176
 icy+0.173
 very+0.165
 grass+0.160
Neg logit contributions
mate-0.337
orman-0.322
anish-0.321
date-0.313
uer-0.313
Token have
Token ID423
Feature activation+2.60e-02

Pos logit contributions
mate+0.029
orman+0.028
anish+0.027
 policy+0.026
 Troll+0.026
Neg logit contributions
 com-0.021
 warm-0.020
 icy-0.020
 soft-0.019
 grass-0.019
Token her
Token ID607
Feature activation-4.59e-02

Pos logit contributions
 com+0.024
 icy+0.023
 warm+0.022
 grass+0.021
 soft+0.020
Neg logit contributions
mate-0.056
orman-0.054
anish-0.052
date-0.052
 occasions-0.052
Token play
Token ID711
Feature activation+2.69e-02

Pos logit contributions
mate+0.087
orman+0.082
anish+0.080
date+0.080
 policy+0.080
Neg logit contributions
 com-0.055
 icy-0.052
 warm-0.051
 grass-0.049
 soft-0.048
Token-
Token ID12
Feature activation+1.38e-01

Pos logit contributions
 com+0.035
 warm+0.034
 very+0.033
 icy+0.032
 all+0.031
Neg logit contributions
 purposes-0.049
 policy-0.049
 Troll-0.049
oke-0.048
orman-0.048
Tokendate
Token ID4475
Feature activation-2.10e-01

Pos logit contributions
 com+0.262
 icy+0.252
 warm+0.247
 grass+0.246
 soft+0.238
Neg logit contributions
mate-0.173
orman-0.165
date-0.159
oker-0.151
anish-0.151
Token after
Token ID706
Feature activation-1.16e-01

Pos logit contributions
mate+0.423
 Troll+0.422
orman+0.418
anish+0.417
 policy+0.411
Neg logit contributions
 com-0.234
 warm-0.224
 very-0.220
 icy-0.211
 even-0.208
Token all
Token ID477
Feature activation+6.25e-02

Pos logit contributions
mate+0.265
anish+0.263
uer+0.263
orman+0.258
uing+0.257
Neg logit contributions
 com-0.102
 icy-0.092
 warm-0.090
 very-0.088
 all-0.085
Token.
Token ID13
Feature activation-5.55e-03

Pos logit contributions
 com+0.060
 icy+0.057
 warm+0.055
 very+0.054
 grass+0.050
Neg logit contributions
uer-0.147
mate-0.143
anish-0.142
uers-0.142
orman-0.141
Token<|endoftext|>
Token ID50256
Feature activation-3.59e-03

Pos logit contributions
mate+0.011
orman+0.010
anish+0.010
 Troll+0.010
 policy+0.010
Neg logit contributions
 com-0.007
 icy-0.006
 warm-0.006
 grass-0.006
 soft-0.006
TokenOne
Token ID3198
Feature activation-2.71e-02

Pos logit contributions
mate+0.006
orman+0.006
uing+0.006
uer+0.006
anish+0.006
Neg logit contributions
 com-0.005
 warm-0.005
 icy-0.005
 grass-0.005
 clouds-0.004
Token and
Token ID290
Feature activation+1.65e-01

Pos logit contributions
mate+0.279
orman+0.271
anish+0.265
 Troll+0.262
 policy+0.261
Neg logit contributions
 com-0.089
 icy-0.081
 warm-0.081
 grass-0.071
 soft-0.070
Token had
Token ID550
Feature activation-5.43e-03

Pos logit contributions
 com+0.191
 icy+0.181
 warm+0.175
 grass+0.163
 soft+0.162
Neg logit contributions
mate-0.324
orman-0.310
anish-0.303
 policy-0.298
uing-0.297
Token a
Token ID257
Feature activation+1.93e-02

Pos logit contributions
orman+0.011
mate+0.011
date+0.011
anish+0.011
 occasions+0.011
Neg logit contributions
 com-0.004
 icy-0.004
 clouds-0.003
 vel-0.003
 parked-0.003
Token fun
Token ID1257
Feature activation-5.12e-02

Pos logit contributions
 grass+0.025
 clouds+0.025
 icy+0.024
 parked+0.024
 com+0.023
Neg logit contributions
mate-0.031
date-0.030
orman-0.029
 guest-0.028
oker-0.028
Token play
Token ID711
Feature activation+2.52e-02

Pos logit contributions
mate+0.090
orman+0.085
anish+0.084
date+0.083
 policy+0.082
Neg logit contributions
 com-0.071
 icy-0.067
 warm-0.065
 grass-0.061
 soft-0.061
Tokendate
Token ID4475
Feature activation-2.10e-01

Pos logit contributions
 warm+0.035
 very+0.034
 icy+0.034
 com+0.034
 all+0.033
Neg logit contributions
 policy-0.042
 client-0.042
 purposes-0.042
reetings-0.042
 Majesty-0.042
Token.
Token ID13
Feature activation-7.51e-03

Pos logit contributions
 Troll+0.453
 Majesty+0.445
 avail+0.443
orman+0.443
rum+0.437
Neg logit contributions
 very-0.204
 even-0.201
 all-0.200
 warm-0.193
 so-0.191
Token<|endoftext|>
Token ID50256
Feature activation-1.28e-02

Pos logit contributions
mate+0.012
orman+0.012
anish+0.011
 Troll+0.011
date+0.011
Neg logit contributions
 com-0.011
 icy-0.011
 warm-0.010
 clouds-0.009
 vel-0.009
TokenOnce
Token ID7454
Feature activation-1.66e-01

Pos logit contributions
mate+0.026
orman+0.025
anish+0.025
uing+0.025
uer+0.025
Neg logit contributions
 com-0.014
 warm-0.013
 icy-0.013
 grass-0.012
 soft-0.012
Token upon
Token ID2402
Feature activation+2.25e-02

Pos logit contributions
mate+0.312
anish+0.304
 occasions+0.303
fer+0.300
uer+0.298
Neg logit contributions
 com-0.222
 warm-0.213
 icy-0.207
 very-0.205
 soft-0.200
Token a
Token ID257
Feature activation+1.82e-02

Pos logit contributions
 refreshing+0.021
 com+0.020
 vel+0.019
 icy+0.019
 prett+0.018
Neg logit contributions
orman-0.045
mate-0.043
anish-0.041
date-0.041
 aid-0.040
Tokenmy
Token ID1820
Feature activation-1.79e-02

Pos logit contributions
mate+0.077
date+0.074
uing+0.073
orman+0.073
anish+0.072
Neg logit contributions
 com-0.056
 icy-0.053
 warm-0.050
 soft-0.049
 vel-0.048
Token were
Token ID547
Feature activation+5.71e-02

Pos logit contributions
mate+0.035
orman+0.033
date+0.032
anish+0.032
uing+0.032
Neg logit contributions
 com-0.020
 icy-0.019
 warm-0.018
 soft-0.017
 grass-0.017
Token on
Token ID319
Feature activation-5.93e-02

Pos logit contributions
 com+0.065
 icy+0.061
 warm+0.060
 soft+0.055
 grass+0.055
Neg logit contributions
mate-0.115
orman-0.111
anish-0.108
 policy-0.106
 Troll-0.106
Token a
Token ID257
Feature activation+1.81e-02

Pos logit contributions
orman+0.117
date+0.116
mate+0.115
 avail+0.115
 occasions+0.111
Neg logit contributions
 com-0.054
 crunch-0.049
 refreshing-0.049
 icy-0.046
 vel-0.045
Token play
Token ID711
Feature activation+2.50e-02

Pos logit contributions
 com+0.022
 icy+0.021
 clouds+0.021
 warm+0.020
 grass+0.020
Neg logit contributions
mate-0.031
orman-0.030
date-0.030
 Troll-0.029
 password-0.029
Tokendate
Token ID4475
Feature activation-2.10e-01

Pos logit contributions
 warm+0.026
 com+0.025
 icy+0.025
 grass+0.024
 very+0.024
Neg logit contributions
 policy-0.052
 Majesty-0.052
uer-0.051
orman-0.051
reetings-0.051
Token.
Token ID13
Feature activation-8.04e-03

Pos logit contributions
 Majesty+0.488
uer+0.475
reetings+0.473
mate+0.470
 Troll+0.468
Neg logit contributions
 warm-0.188
 very-0.187
 all-0.179
 com-0.175
 so-0.174
Token Alice
Token ID14862
Feature activation-9.98e-02

Pos logit contributions
mate+0.011
orman+0.011
anish+0.011
 begs+0.011
date+0.010
Neg logit contributions
 com-0.013
 icy-0.013
 warm-0.012
 soft-0.012
 grass-0.012
Token was
Token ID373
Feature activation+7.73e-02

Pos logit contributions
mate+0.203
orman+0.195
anish+0.191
 policy+0.187
 Troll+0.187
Neg logit contributions
 com-0.111
 icy-0.104
 warm-0.103
 grass-0.095
 soft-0.095
Token playing
Token ID2712
Feature activation+4.65e-03

Pos logit contributions
 com+0.073
 icy+0.068
 warm+0.066
 grass+0.063
 soft+0.060
Neg logit contributions
mate-0.167
orman-0.162
date-0.156
anish-0.156
 Troll-0.155
Token with
Token ID351
Feature activation-1.81e-02

Pos logit contributions
 com+0.004
 warm+0.004
 icy+0.004
 grass+0.004
 very+0.004
Neg logit contributions
mate-0.010
orman-0.010
 policy-0.010
uer-0.010
 Majesty-0.010
Token can
Token ID460
Feature activation-9.86e-04

Pos logit contributions
mate+0.047
orman+0.044
anish+0.043
 Troll+0.042
uing+0.042
Neg logit contributions
 com-0.043
 icy-0.041
 warm-0.041
 grass-0.039
 soft-0.039
Token have
Token ID423
Feature activation+2.85e-02

Pos logit contributions
mate+0.002
orman+0.001
anish+0.001
 Troll+0.001
 policy+0.001
Neg logit contributions
 com-0.002
 icy-0.001
 warm-0.001
 grass-0.001
 soft-0.001
Token your
Token ID534
Feature activation-5.55e-02

Pos logit contributions
 com+0.027
 icy+0.025
 warm+0.024
 grass+0.022
 soft+0.022
Neg logit contributions
mate-0.061
orman-0.059
anish-0.058
 Troll-0.057
 policy-0.057
Token play
Token ID711
Feature activation+2.81e-02

Pos logit contributions
mate+0.091
orman+0.086
anish+0.083
 policy+0.082
date+0.082
Neg logit contributions
 com-0.083
 icy-0.079
 warm-0.078
 grass-0.074
 soft-0.074
Token-
Token ID12
Feature activation+1.38e-01

Pos logit contributions
 com+0.043
 warm+0.042
 icy+0.041
 very+0.041
 soft+0.039
Neg logit contributions
 policy-0.044
 Troll-0.043
orman-0.043
 purposes-0.043
 begs-0.042
Tokendate
Token ID4475
Feature activation-2.10e-01

Pos logit contributions
 com+0.250
 icy+0.238
 grass+0.233
 warm+0.232
 refreshing+0.227
Neg logit contributions
mate-0.184
orman-0.174
date-0.171
or-0.163
oker-0.163
Token with
Token ID351
Feature activation-1.46e-02

Pos logit contributions
mate+0.400
orman+0.398
 Troll+0.394
 policy+0.387
anish+0.386
Neg logit contributions
 com-0.252
 warm-0.244
 very-0.238
 icy-0.233
 so-0.223
Token R
Token ID371
Feature activation+3.44e-02

Pos logit contributions
mate+0.027
orman+0.026
anish+0.025
 Troll+0.025
 policy+0.025
Neg logit contributions
 com-0.019
 icy-0.018
 warm-0.018
 grass-0.017
 soft-0.017
Tokenuf
Token ID3046
Feature activation+7.18e-02

Pos logit contributions
 com+0.064
 icy+0.061
 warm+0.060
 grass+0.057
 soft+0.057
Neg logit contributions
mate-0.045
anish-0.040
orman-0.040
date-0.039
 password-0.039
Tokenus
Token ID385
Feature activation-5.81e-03

Pos logit contributions
 com+0.120
 icy+0.113
 very+0.110
 warm+0.108
 clouds+0.104
Neg logit contributions
mate-0.103
date-0.101
anish-0.099
 avail-0.098
 begs-0.095
Token now
Token ID783
Feature activation-7.21e-02

Pos logit contributions
mate+0.009
 avail+0.009
orman+0.009
 policy+0.009
date+0.009
Neg logit contributions
 com-0.009
 very-0.009
 warm-0.009
 icy-0.008
 soft-0.008
Token him
Token ID683
Feature activation-4.95e-02

Pos logit contributions
mate+0.032
orman+0.031
anish+0.030
 policy+0.030
 Troll+0.030
Neg logit contributions
 com-0.015
 icy-0.014
 warm-0.014
 grass-0.012
 soft-0.012
Token.
Token ID13
Feature activation-4.96e-03

Pos logit contributions
mate+0.093
orman+0.091
anish+0.089
oker+0.088
date+0.087
Neg logit contributions
 com-0.063
 warm-0.060
 icy-0.060
 very-0.058
 soft-0.056
Token But
Token ID887
Feature activation-4.32e-02

Pos logit contributions
mate+0.009
orman+0.009
anish+0.009
 policy+0.009
uing+0.009
Neg logit contributions
 com-0.006
 icy-0.006
 warm-0.006
 grass-0.006
 very-0.006
Token to
Token ID284
Feature activation-9.07e-02

Pos logit contributions
mate+0.092
orman+0.091
anish+0.089
uer+0.088
oker+0.088
Neg logit contributions
 com-0.042
 icy-0.041
 warm-0.040
 very-0.039
 even-0.036
Token no
Token ID645
Feature activation-4.28e-02

Pos logit contributions
mate+0.190
orman+0.183
anish+0.179
 policy+0.176
uer+0.175
Neg logit contributions
 com-0.096
 icy-0.090
 warm-0.089
 grass-0.082
 soft-0.081
Token avail
Token ID29107
Feature activation-2.09e-01

Pos logit contributions
orman+0.077
mate+0.075
anish+0.075
uer+0.074
uers+0.073
Neg logit contributions
 com-0.061
 icy-0.060
 warm-0.058
 grass-0.056
 orange-0.055
Token,
Token ID11
Feature activation-4.50e-02

Pos logit contributions
oker+0.489
 jury+0.477
uers+0.473
 avail+0.470
 Majesty+0.467
Neg logit contributions
 very-0.217
 even-0.198
 all-0.189
 and-0.184
 so-0.183
Token they
Token ID484
Feature activation-7.18e-02

Pos logit contributions
mate+0.098
orman+0.095
anish+0.093
uing+0.091
uer+0.091
Neg logit contributions
 com-0.043
 icy-0.041
 warm-0.040
 grass-0.036
 soft-0.036
Token were
Token ID547
Feature activation+5.56e-02

Pos logit contributions
mate+0.151
orman+0.147
anish+0.145
 policy+0.144
 occasions+0.143
Neg logit contributions
 com-0.072
 icy-0.070
 warm-0.069
 very-0.066
 even-0.062
Token too
Token ID1165
Feature activation+5.33e-02

Pos logit contributions
 com+0.046
 icy+0.042
 warm+0.042
 grass+0.038
 soft+0.037
Neg logit contributions
mate-0.128
orman-0.124
anish-0.120
 Troll-0.119
date-0.118
Token scared
Token ID12008
Feature activation-3.89e-02

Pos logit contributions
 com+0.047
 grass+0.044
 icy+0.042
 warm+0.042
 vel+0.041
Neg logit contributions
orman-0.112
mate-0.109
uer-0.107
uers-0.106
oker-0.106
Token for
Token ID329
Feature activation-7.41e-02

Pos logit contributions
mate+0.200
orman+0.194
anish+0.191
date+0.189
uer+0.187
Neg logit contributions
 com-0.119
 warm-0.112
 icy-0.112
 very-0.105
 soft-0.105
Token using
Token ID1262
Feature activation-3.52e-03

Pos logit contributions
mate+0.148
orman+0.146
uer+0.143
anish+0.143
oker+0.142
Neg logit contributions
 com-0.084
 warm-0.081
 icy-0.080
 soft-0.074
 very-0.074
Token the
Token ID262
Feature activation+4.16e-02

Pos logit contributions
mate+0.007
orman+0.007
anish+0.007
 policy+0.007
 Troll+0.007
Neg logit contributions
 com-0.004
 icy-0.003
 warm-0.003
 grass-0.003
 soft-0.003
Token past
Token ID1613
Feature activation-5.85e-02

Pos logit contributions
 com+0.048
 icy+0.045
 warm+0.045
 grass+0.041
 soft+0.041
Neg logit contributions
mate-0.083
orman-0.080
anish-0.078
 Troll-0.077
 policy-0.077
Token-
Token ID12
Feature activation+1.40e-01

Pos logit contributions
mate+0.100
uers+0.098
uer+0.097
orman+0.096
uing+0.094
Neg logit contributions
 com-0.083
 icy-0.081
 warm-0.079
 grass-0.077
 clouds-0.076
Tokendate
Token ID4475
Feature activation-2.09e-01

Pos logit contributions
 com+0.212
 icy+0.197
 warm+0.194
 grass+0.184
 soft+0.183
Neg logit contributions
mate-0.230
orman-0.220
anish-0.211
date-0.207
oker-0.207
Token cheese
Token ID9891
Feature activation+5.38e-02

Pos logit contributions
 avail+0.447
 Majesty+0.433
uing+0.433
mate+0.433
anish+0.431
Neg logit contributions
 very-0.224
 so-0.209
 warm-0.205
 com-0.199
 even-0.198
Token.
Token ID13
Feature activation-6.16e-03

Pos logit contributions
 com+0.061
 warm+0.060
 icy+0.060
 very+0.058
 grass+0.057
Neg logit contributions
orman-0.104
mate-0.104
 avail-0.102
 Majesty-0.101
 Troll-0.100
Token Sam
Token ID3409
Feature activation-6.81e-02

Pos logit contributions
mate+0.012
uing+0.012
 policy+0.012
uer+0.012
fer+0.012
Neg logit contributions
 com-0.007
 very-0.006
 grass-0.006
 warm-0.006
 icy-0.006
Token learned
Token ID4499
Feature activation+1.62e-02

Pos logit contributions
mate+0.126
orman+0.122
anish+0.117
 Troll+0.116
uer+0.115
Neg logit contributions
 com-0.085
 icy-0.081
 warm-0.080
 grass-0.075
 soft-0.075
Token an
Token ID281
Feature activation+1.39e-02

Pos logit contributions
 com+0.020
 icy+0.019
 warm+0.018
 clouds+0.018
 grass+0.017
Neg logit contributions
orman-0.030
mate-0.030
 policy-0.028
 client-0.027
 begs-0.027
Token times
Token ID1661
Feature activation-2.02e-02

Pos logit contributions
 com+0.023
 icy+0.021
 warm+0.021
 grass+0.019
 soft+0.019
Neg logit contributions
mate-0.039
orman-0.037
anish-0.036
 Troll-0.035
 policy-0.035
Token,
Token ID11
Feature activation-4.82e-02

Pos logit contributions
mate+0.037
orman+0.036
anish+0.035
 Majesty+0.035
 Troll+0.035
Neg logit contributions
 com-0.026
 warm-0.025
 icy-0.025
 very-0.024
 soft-0.023
Token but
Token ID475
Feature activation+9.64e-04

Pos logit contributions
mate+0.070
anish+0.063
 begs+0.063
 policy+0.062
 purposes+0.061
Neg logit contributions
 com-0.073
 icy-0.070
 warm-0.069
 clouds-0.066
 vel-0.066
Token to
Token ID284
Feature activation-9.26e-02

Pos logit contributions
 com+0.001
 icy+0.001
 warm+0.001
 grass+0.001
 soft+0.001
Neg logit contributions
mate-0.002
orman-0.002
anish-0.002
 policy-0.002
 Troll-0.002
Token no
Token ID645
Feature activation-4.41e-02

Pos logit contributions
mate+0.164
orman+0.155
 avail+0.148
anish+0.148
 Troll+0.146
Neg logit contributions
 com-0.125
 icy-0.118
 warm-0.114
 grass-0.110
 clouds-0.109
Token avail
Token ID29107
Feature activation-2.09e-01

Pos logit contributions
mate+0.057
orman+0.057
uers+0.056
uer+0.055
anish+0.055
Neg logit contributions
 com-0.083
 warm-0.081
 icy-0.081
 grass-0.077
 orange-0.076
Token.
Token ID13
Feature activation-7.00e-03

Pos logit contributions
uers+0.541
oker+0.536
 Majesty+0.534
 jury+0.518
fer+0.515
Neg logit contributions
 very-0.188
 even-0.165
 all-0.145
 so-0.142
 because-0.137
Token
Token ID220
Feature activation+9.56e-03

Pos logit contributions
mate+0.014
fer+0.013
orman+0.013
oker+0.013
uing+0.013
Neg logit contributions
 com-0.008
 warm-0.008
 grass-0.008
 icy-0.008
 very-0.008
Token
Token ID628
Feature activation-1.10e-02

Pos logit contributions
 com+0.008
 warm+0.007
 very+0.007
 icy+0.007
 soft+0.007
Neg logit contributions
mate-0.022
fer-0.021
 occasions-0.021
oker-0.021
 Troll-0.021
Token Jane
Token ID12091
Feature activation-7.03e-02

Pos logit contributions
mate+0.024
orman+0.022
anish+0.022
uing+0.022
 Troll+0.022
Neg logit contributions
 com-0.011
 icy-0.011
 warm-0.011
 very-0.010
 soft-0.010
Token's
Token ID338
Feature activation-1.40e-02

Pos logit contributions
mate+0.141
 policy+0.135
oker+0.133
anish+0.133
orman+0.131
Neg logit contributions
 com-0.083
 icy-0.079
 warm-0.078
 very-0.076
 grass-0.073
Token almost
Token ID2048
Feature activation-1.86e-02

Pos logit contributions
 com+0.045
 icy+0.039
 grass+0.039
 warm+0.037
 clouds+0.034
Neg logit contributions
mate-0.193
orman-0.192
 Majesty-0.184
date-0.183
uing-0.180
Token ready
Token ID3492
Feature activation-1.58e-01

Pos logit contributions
mate+0.035
orman+0.034
anish+0.033
 Troll+0.032
 policy+0.032
Neg logit contributions
 com-0.023
 icy-0.022
 warm-0.022
 grass-0.020
 soft-0.020
Token for
Token ID329
Feature activation-7.32e-02

Pos logit contributions
mate+0.346
orman+0.340
anish+0.337
 Troll+0.335
uer+0.332
Neg logit contributions
 com-0.159
 warm-0.155
 icy-0.149
 very-0.146
 soft-0.137
Token the
Token ID262
Feature activation+3.94e-02

Pos logit contributions
mate+0.162
orman+0.156
anish+0.152
 policy+0.150
 Troll+0.150
Neg logit contributions
 com-0.068
 icy-0.063
 warm-0.062
 grass-0.056
 soft-0.056
Token play
Token ID711
Feature activation+2.91e-02

Pos logit contributions
 com+0.053
 warm+0.051
 icy+0.051
 soft+0.047
 grass+0.047
Neg logit contributions
mate-0.073
orman-0.070
anish-0.068
uing-0.067
rum-0.067
Tokendate
Token ID4475
Feature activation-2.09e-01

Pos logit contributions
 very+0.052
 warm+0.051
 icy+0.051
 com+0.050
 all+0.049
Neg logit contributions
rums-0.039
 Majesty-0.038
 purposes-0.037
rum-0.037
reetings-0.037
Token!
Token ID0
Feature activation-5.22e-03

Pos logit contributions
anish+0.445
uing+0.439
 Majesty+0.436
rums+0.435
mate+0.435
Neg logit contributions
 very-0.215
 so-0.210
 warm-0.206
 com-0.204
 even-0.203
Token
Token ID198
Feature activation+1.63e-02

Pos logit contributions
mate+0.010
orman+0.009
anish+0.009
date+0.009
 policy+0.009
Neg logit contributions
 com-0.007
 icy-0.007
 warm-0.006
 soft-0.006
 grass-0.006
Token
Token ID198
Feature activation+1.75e-02

Pos logit contributions
 com+0.018
 icy+0.017
 warm+0.017
 grass+0.016
 soft+0.016
Neg logit contributions
mate-0.033
orman-0.032
anish-0.031
date-0.030
 policy-0.030
TokenBilly
Token ID49640
Feature activation-4.10e-02

Pos logit contributions
 com+0.021
 icy+0.020
 warm+0.019
 grass+0.018
 soft+0.018
Neg logit contributions
mate-0.035
orman-0.033
anish-0.032
date-0.032
uing-0.031
Token opened
Token ID4721
Feature activation+1.83e-02

Pos logit contributions
mate+0.080
orman+0.077
anish+0.075
 Troll+0.073
uing+0.073
Neg logit contributions
 com-0.048
 icy-0.045
 warm-0.045
 soft-0.041
 grass-0.041
Token friends
Token ID2460
Feature activation-1.43e-02

Pos logit contributions
mate+0.073
anish+0.070
orman+0.069
rum+0.068
oker+0.067
Neg logit contributions
 com-0.071
 warm-0.069
 icy-0.066
 soft-0.065
 very-0.064
Token over
Token ID625
Feature activation-1.89e-02

Pos logit contributions
uer+0.032
 Majesty+0.031
 avail+0.031
orman+0.031
uers+0.031
Neg logit contributions
 very-0.013
 warm-0.013
 com-0.013
 all-0.013
 even-0.012
Token for
Token ID329
Feature activation-7.50e-02

Pos logit contributions
mate+0.037
orman+0.037
anish+0.036
uer+0.036
 Majesty+0.036
Neg logit contributions
 com-0.022
 warm-0.021
 icy-0.020
 very-0.020
 soft-0.019
Token a
Token ID257
Feature activation+2.07e-02

Pos logit contributions
mate+0.184
orman+0.177
anish+0.172
date+0.171
 policy+0.170
Neg logit contributions
 com-0.052
 icy-0.046
 warm-0.044
 grass-0.039
 soft-0.038
Token play
Token ID711
Feature activation+2.58e-02

Pos logit contributions
 com+0.030
 icy+0.029
 warm+0.028
 grass+0.027
 soft+0.026
Neg logit contributions
mate-0.035
orman-0.033
anish-0.032
date-0.032
 Troll-0.031
Tokendate
Token ID4475
Feature activation-2.09e-01

Pos logit contributions
 warm+0.050
 very+0.049
 all+0.048
 grass+0.047
 icy+0.047
Neg logit contributions
 Majesty-0.034
 purposes-0.031
rum-0.031
 inspection-0.030
reetings-0.030
Token.
Token ID13
Feature activation-2.86e-03

Pos logit contributions
 Majesty+0.509
uer+0.482
 avail+0.476
opter+0.476
reetings+0.474
Neg logit contributions
 very-0.180
 warm-0.175
 all-0.172
 a-0.169
 so-0.167
Token She
Token ID1375
Feature activation+6.39e-03

Pos logit contributions
fer+0.007
rum+0.007
mate+0.007
eman+0.006
 avail+0.006
Neg logit contributions
 grass-0.003
 com-0.003
 warm-0.002
 very-0.002
 soft-0.002
Token showed
Token ID3751
Feature activation-6.66e-02

Pos logit contributions
 com+0.008
 icy+0.008
 warm+0.008
 grass+0.007
 soft+0.007
Neg logit contributions
mate-0.012
orman-0.011
anish-0.011
 policy-0.011
oker-0.011
Token them
Token ID606
Feature activation-9.56e-03

Pos logit contributions
mate+0.125
uer+0.121
orman+0.121
anish+0.120
 policy+0.120
Neg logit contributions
 com-0.083
 warm-0.081
 icy-0.078
 soft-0.076
 grass-0.075
Token her
Token ID607
Feature activation-4.50e-02

Pos logit contributions
mate+0.021
orman+0.020
anish+0.020
 Troll+0.020
uing+0.020
Neg logit contributions
 com-0.008
 icy-0.008
 warm-0.007
 clouds-0.007
 grass-0.007
Token deep
Token ID2769
Feature activation-1.84e-02

Pos logit contributions
mate+0.087
orman+0.084
anish+0.082
oker+0.082
uing+0.082
Neg logit contributions
 com-0.063
 warm-0.060
 icy-0.059
 soft-0.057
 very-0.056
Token paws
Token ID49193
Feature activation+9.19e-02

Pos logit contributions
mate+0.036
orman+0.035
date+0.035
oker+0.035
oke+0.034
Neg logit contributions
 com-0.025
 icy-0.023
 warm-0.023
 soft-0.022
 grass-0.022
Token,
Token ID11
Feature activation-4.45e-02

Pos logit contributions
 com+0.120
 warm+0.119
 icy+0.118
 very+0.113
 soft+0.110
Neg logit contributions
mate-0.168
orman-0.164
anish-0.161
oker-0.160
 Troll-0.158
Token to
Token ID284
Feature activation-9.17e-02

Pos logit contributions
mate+0.069
orman+0.065
 Troll+0.064
 policy+0.063
anish+0.062
Neg logit contributions
 com-0.062
 icy-0.058
 warm-0.058
 clouds-0.055
 soft-0.054
Token no
Token ID645
Feature activation-4.50e-02

Pos logit contributions
mate+0.158
orman+0.149
anish+0.144
 Troll+0.142
date+0.141
Neg logit contributions
 com-0.131
 icy-0.124
 warm-0.122
 grass-0.115
 soft-0.115
Token avail
Token ID29107
Feature activation-2.08e-01

Pos logit contributions
orman+0.080
mate+0.078
anish+0.077
uers+0.077
uer+0.076
Neg logit contributions
 com-0.067
 icy-0.065
 warm-0.064
 grass-0.062
 orange-0.061
Token.
Token ID13
Feature activation-2.05e-03

Pos logit contributions
oker+0.530
 jury+0.512
uers+0.508
 Majesty+0.497
date+0.491
Neg logit contributions
 very-0.182
 even-0.158
 all-0.150
 so-0.142
 a-0.141
Token He
Token ID679
Feature activation-7.79e-03

Pos logit contributions
mate+0.004
uing+0.004
orman+0.004
 policy+0.004
oker+0.004
Neg logit contributions
 com-0.002
 warm-0.002
 icy-0.002
 grass-0.002
 very-0.002
Token finally
Token ID3443
Feature activation-9.18e-02

Pos logit contributions
mate+0.015
orman+0.014
anish+0.013
 Troll+0.013
uer+0.013
Neg logit contributions
 com-0.010
 icy-0.009
 warm-0.009
 soft-0.009
 grass-0.009
Token thought
Token ID1807
Feature activation-2.92e-02

Pos logit contributions
mate+0.161
orman+0.150
anish+0.150
 policy+0.147
date+0.147
Neg logit contributions
 com-0.131
 icy-0.125
 warm-0.123
 grass-0.116
 very-0.114
Token of
Token ID286
Feature activation-4.68e-02

Pos logit contributions
mate+0.058
orman+0.055
anish+0.054
 policy+0.053
uing+0.053
Neg logit contributions
 com-0.034
 icy-0.032
 warm-0.032
 grass-0.029
 soft-0.029
Token came
Token ID1625
Feature activation-8.56e-03

Pos logit contributions
uer+0.178
uing+0.172
rum+0.172
anish+0.171
 occasions+0.171
Neg logit contributions
 warm-0.082
 very-0.081
 a-0.081
 com-0.078
 so-0.077
Token over
Token ID625
Feature activation-1.51e-02

Pos logit contributions
mate+0.019
orman+0.018
anish+0.017
 Troll+0.017
date+0.017
Neg logit contributions
 com-0.008
 icy-0.008
 warm-0.008
 grass-0.007
 soft-0.007
Token for
Token ID329
Feature activation-7.37e-02

Pos logit contributions
uer+0.033
 Majesty+0.033
uers+0.033
mate+0.033
uing+0.032
Neg logit contributions
 com-0.015
 warm-0.014
 very-0.014
 icy-0.013
 soft-0.012
Token a
Token ID257
Feature activation+2.41e-02

Pos logit contributions
mate+0.156
orman+0.150
anish+0.147
 Troll+0.144
date+0.144
Neg logit contributions
 com-0.075
 icy-0.070
 warm-0.069
 grass-0.063
 soft-0.063
Token play
Token ID711
Feature activation+2.52e-02

Pos logit contributions
 com+0.032
 warm+0.031
 icy+0.031
 soft+0.029
 grass+0.028
Neg logit contributions
mate-0.044
orman-0.042
anish-0.041
 Troll-0.040
 policy-0.040
Tokendate
Token ID4475
Feature activation-2.07e-01

Pos logit contributions
 warm+0.044
 icy+0.042
 very+0.042
 com+0.041
 grass+0.040
Neg logit contributions
 Majesty-0.038
uer-0.036
 purposes-0.035
rum-0.035
reetings-0.035
Token.
Token ID13
Feature activation+9.39e-04

Pos logit contributions
 Majesty+0.486
uer+0.469
uing+0.464
reetings+0.461
 avail+0.460
Neg logit contributions
 very-0.183
 warm-0.180
 so-0.178
 too-0.171
 a-0.168
Token Her
Token ID2332
Feature activation+8.48e-03

Pos logit contributions
 com+0.001
 warm+0.001
 icy+0.001
 soft+0.001
 grass+0.001
Neg logit contributions
mate-0.002
orman-0.002
uing-0.002
uer-0.002
anish-0.002
Token friend
Token ID1545
Feature activation-7.57e-02

Pos logit contributions
 warm+0.014
 com+0.013
 soft+0.013
 icy+0.013
 very+0.012
Neg logit contributions
mate-0.014
rum-0.014
 occasions-0.014
anish-0.014
oker-0.014
Token asked
Token ID1965
Feature activation-1.29e-01

Pos logit contributions
mate+0.166
 occasions+0.166
uer+0.163
 policy+0.163
anish+0.162
Neg logit contributions
 very-0.079
 com-0.077
 even-0.076
 warm-0.076
 all-0.073
Token for
Token ID329
Feature activation-7.40e-02

Pos logit contributions
uer+0.370
uers+0.353
fer+0.351
 Majesty+0.345
opter+0.336
Neg logit contributions
 very-0.087
 a-0.082
 all-0.073
 warm-0.072
 even-0.070
Token my
Token ID616
Feature activation-5.79e-02

Pos logit contributions
mate+0.169
orman+0.163
anish+0.159
 policy+0.156
uing+0.156
Neg logit contributions
 com-0.108
 icy-0.102
 warm-0.102
 grass-0.094
 soft-0.094
Token house
Token ID2156
Feature activation-1.28e-02

Pos logit contributions
mate+0.099
anish+0.098
oker+0.098
uing+0.098
orman+0.097
Neg logit contributions
 warm-0.088
 com-0.086
 icy-0.086
 soft-0.082
 very-0.081
Token for
Token ID329
Feature activation-7.34e-02

Pos logit contributions
mate+0.024
orman+0.024
 Troll+0.023
anish+0.023
 policy+0.023
Neg logit contributions
 com-0.015
 icy-0.015
 warm-0.015
 very-0.014
 soft-0.014
Token a
Token ID257
Feature activation+2.37e-02

Pos logit contributions
mate+0.171
orman+0.165
anish+0.161
date+0.159
 Troll+0.159
Neg logit contributions
 com-0.060
 icy-0.054
 warm-0.053
 grass-0.048
 soft-0.047
Token play
Token ID711
Feature activation+2.87e-02

Pos logit contributions
 com+0.030
 warm+0.029
 icy+0.028
 soft+0.027
 very+0.026
Neg logit contributions
mate-0.045
orman-0.044
anish-0.044
rum-0.043
uing-0.043
Tokendate
Token ID4475
Feature activation-2.07e-01

Pos logit contributions
 warm+0.051
 very+0.050
 icy+0.050
 all+0.049
 too+0.048
Neg logit contributions
 Majesty-0.039
 purposes-0.039
 inspection-0.038
uer-0.038
 policy-0.038
Token.
Token ID13
Feature activation+5.48e-03

Pos logit contributions
uing+0.408
 avail+0.400
orman+0.400
uer+0.400
 jury+0.398
Neg logit contributions
 so-0.241
 very-0.239
 warm-0.238
 too-0.237
 all-0.233
Token I
Token ID314
Feature activation+7.11e-03

Pos logit contributions
 com+0.010
 icy+0.010
 warm+0.009
 grass+0.009
 soft+0.009
Neg logit contributions
mate-0.007
anish-0.007
orman-0.007
 Troll-0.006
date-0.006
Token invite
Token ID14037
Feature activation-7.44e-02

Pos logit contributions
 com+0.010
 warm+0.010
 icy+0.010
 very+0.009
 soft+0.009
Neg logit contributions
mate-0.012
orman-0.012
anish-0.012
 policy-0.011
 Troll-0.011
Token you
Token ID345
Feature activation-5.79e-02

Pos logit contributions
mate+0.140
orman+0.136
anish+0.132
uer+0.130
oker+0.130
Neg logit contributions
 com-0.093
 warm-0.088
 icy-0.088
 grass-0.082
 soft-0.082
Token."
Token ID526
Feature activation-1.18e-02

Pos logit contributions
mate+0.111
orman+0.107
anish+0.104
 policy+0.103
 Troll+0.102
Neg logit contributions
 com-0.071
 icy-0.067
 warm-0.067
 grass-0.062
 soft-0.062
Token out
Token ID503
Feature activation-7.71e-02

Pos logit contributions
mate+0.122
orman+0.116
anish+0.113
 policy+0.111
uing+0.110
Neg logit contributions
 com-0.095
 icy-0.091
 warm-0.090
 grass-0.084
 soft-0.084
Token,
Token ID11
Feature activation-4.53e-02

Pos logit contributions
orman+0.146
mate+0.140
anish+0.139
 client+0.139
 Majesty+0.138
Neg logit contributions
 com-0.103
 icy-0.099
 warm-0.099
 very-0.096
 soft-0.092
Token but
Token ID475
Feature activation+2.90e-03

Pos logit contributions
mate+0.073
orman+0.067
 Troll+0.065
anish+0.065
 begs+0.064
Neg logit contributions
 com-0.065
 icy-0.061
 warm-0.060
 clouds-0.058
 soft-0.057
Token to
Token ID284
Feature activation-9.19e-02

Pos logit contributions
 com+0.004
 icy+0.004
 warm+0.003
 soft+0.003
 grass+0.003
Neg logit contributions
mate-0.005
orman-0.005
anish-0.005
 policy-0.005
 Troll-0.005
Token no
Token ID645
Feature activation-4.25e-02

Pos logit contributions
mate+0.164
orman+0.153
 avail+0.146
 Troll+0.143
 Majesty+0.143
Neg logit contributions
 com-0.122
 icy-0.112
 warm-0.111
 clouds-0.107
 soft-0.106
Token avail
Token ID29107
Feature activation-2.07e-01

Pos logit contributions
mate+0.058
orman+0.057
anish+0.056
uer+0.054
uers+0.053
Neg logit contributions
 com-0.077
 icy-0.075
 warm-0.075
 grass-0.071
 soft-0.070
Token.
Token ID13
Feature activation-5.55e-03

Pos logit contributions
oker+0.516
 jury+0.511
uers+0.510
 Majesty+0.500
 occasions+0.491
Neg logit contributions
 very-0.186
 even-0.168
 all-0.153
 so-0.147
 because-0.136
Token He
Token ID679
Feature activation-9.28e-03

Pos logit contributions
mate+0.011
uing+0.010
orman+0.010
oker+0.010
 policy+0.010
Neg logit contributions
 com-0.007
 icy-0.007
 warm-0.006
 grass-0.006
 very-0.006
Token felt
Token ID2936
Feature activation+1.20e-01

Pos logit contributions
mate+0.017
orman+0.016
uer+0.015
 Troll+0.015
anish+0.015
Neg logit contributions
 com-0.012
 warm-0.011
 icy-0.011
 fresh-0.011
 soft-0.011
Token sad
Token ID6507
Feature activation-1.43e-02

Pos logit contributions
 com+0.109
 grass+0.096
 stars+0.094
 clouds+0.092
 vel+0.092
Neg logit contributions
mate-0.229
orman-0.223
 begs-0.210
rum-0.210
ings-0.207
Token and
Token ID290
Feature activation+1.66e-01

Pos logit contributions
mate+0.034
orman+0.033
anish+0.033
date+0.033
oker+0.032
Neg logit contributions
 com-0.012
 icy-0.011
 warm-0.011
 very-0.010
 soft-0.009
Token would
Token ID561
Feature activation+1.05e-02

Pos logit contributions
mate+0.149
orman+0.144
anish+0.142
 Troll+0.140
 Majesty+0.140
Neg logit contributions
 com-0.069
 warm-0.066
 icy-0.065
 very-0.062
 soft-0.059
Token have
Token ID423
Feature activation+2.84e-02

Pos logit contributions
 warm+0.013
 com+0.013
 icy+0.012
 very+0.012
 even+0.011
Neg logit contributions
mate-0.020
orman-0.020
anish-0.019
uer-0.019
oke-0.019
Token a
Token ID257
Feature activation+1.92e-02

Pos logit contributions
 com+0.027
 icy+0.024
 warm+0.023
 grass+0.021
 soft+0.021
Neg logit contributions
orman-0.059
mate-0.059
date-0.056
 occasions-0.056
anish-0.056
Token play
Token ID711
Feature activation+2.52e-02

Pos logit contributions
 com+0.025
 icy+0.024
 clouds+0.023
 grass+0.023
 warm+0.023
Neg logit contributions
mate-0.033
orman-0.032
date-0.032
 password-0.031
 Troll-0.030
Token-
Token ID12
Feature activation+1.40e-01

Pos logit contributions
 warm+0.049
 icy+0.049
 very+0.049
 com+0.048
 grass+0.047
Neg logit contributions
 Majesty-0.028
 purposes-0.028
rum-0.028
reetings-0.028
 policy-0.027
Tokendate
Token ID4475
Feature activation-2.07e-01

Pos logit contributions
 com+0.254
 icy+0.243
 warm+0.240
 grass+0.231
 soft+0.229
Neg logit contributions
mate-0.188
orman-0.180
anish-0.169
date-0.168
oker-0.166
Token and
Token ID290
Feature activation+1.67e-01

Pos logit contributions
 Majesty+0.444
orman+0.434
 Troll+0.434
mate+0.433
uer+0.433
Neg logit contributions
 warm-0.205
 very-0.201
 com-0.191
 all-0.190
 even-0.188
Token wear
Token ID5806
Feature activation+5.39e-02

Pos logit contributions
 com+0.223
 warm+0.216
 icy+0.211
 grass+0.200
 very+0.199
Neg logit contributions
mate-0.306
orman-0.292
anish-0.291
uer-0.284
uing-0.284
Token makeup
Token ID16029
Feature activation+4.56e-02

Pos logit contributions
 com+0.061
 icy+0.059
 warm+0.052
 clouds+0.052
 vel+0.050
Neg logit contributions
mate-0.100
orman-0.097
date-0.094
 password-0.093
 occasions-0.092
Token.
Token ID13
Feature activation-3.87e-03

Pos logit contributions
 com+0.060
 warm+0.057
 icy+0.057
 very+0.053
 soft+0.053
Neg logit contributions
mate-0.084
orman-0.081
anish-0.079
 Troll-0.078
 policy-0.078
Token
Token ID198
Feature activation+1.50e-02

Pos logit contributions
mate+0.009
uer+0.008
anish+0.008
orman+0.008
uing+0.008
Neg logit contributions
 com-0.004
 warm-0.004
 grass-0.004
 icy-0.003
 very-0.003
Token let
Token ID1309
Feature activation-4.62e-02

Pos logit contributions
mate+0.081
orman+0.079
uer+0.077
uing+0.076
anish+0.076
Neg logit contributions
 com-0.055
 warm-0.054
 icy-0.053
 very-0.050
 grass-0.049
Token's
Token ID338
Feature activation-1.27e-02

Pos logit contributions
uer+0.115
 policy+0.110
ider+0.109
uers+0.108
uing+0.108
Neg logit contributions
 warm-0.039
 icy-0.036
 very-0.035
 all-0.035
 grass-0.035
Token have
Token ID423
Feature activation+2.63e-02

Pos logit contributions
mate+0.025
orman+0.024
 policy+0.024
anish+0.024
uer+0.024
Neg logit contributions
 warm-0.015
 com-0.015
 icy-0.014
 very-0.013
 soft-0.013
Token a
Token ID257
Feature activation+1.97e-02

Pos logit contributions
 com+0.021
 icy+0.019
 warm+0.019
 grass+0.017
 soft+0.017
Neg logit contributions
mate-0.061
orman-0.059
anish-0.057
date-0.056
 policy-0.056
Token play
Token ID711
Feature activation+2.44e-02

Pos logit contributions
 com+0.024
 warm+0.023
 icy+0.023
 soft+0.021
 very+0.021
Neg logit contributions
mate-0.038
orman-0.037
anish-0.036
 Troll-0.035
 policy-0.035
Tokendate
Token ID4475
Feature activation-2.06e-01

Pos logit contributions
 warm+0.027
 very+0.027
 icy+0.026
 com+0.026
 so+0.024
Neg logit contributions
 Troll-0.048
 purposes-0.048
 Majesty-0.047
orman-0.047
 client-0.046
Token!"
Token ID2474
Feature activation-3.19e-02

Pos logit contributions
 Troll+0.417
orman+0.415
mate+0.409
 avail+0.407
uer+0.405
Neg logit contributions
 very-0.222
 warm-0.222
 so-0.221
 even-0.212
 all-0.211
Token
Token ID198
Feature activation+1.01e-02

Pos logit contributions
uing+0.070
mate+0.070
uer+0.069
 Majesty+0.068
 password+0.068
Neg logit contributions
 com-0.029
 very-0.029
 so-0.028
 grass-0.028
 warm-0.027
TokenThe
Token ID464
Feature activation+6.31e-02

Pos logit contributions
 com+0.010
 warm+0.009
 icy+0.009
 soft+0.009
 grass+0.009
Neg logit contributions
mate-0.021
 Majesty-0.021
uing-0.021
uer-0.021
anish-0.020
Token steak
Token ID26320
Feature activation+7.09e-03

Pos logit contributions
 com+0.095
 icy+0.090
 warm+0.089
 grass+0.084
 soft+0.084
Neg logit contributions
mate-0.104
orman-0.099
anish-0.096
 Troll-0.094
 policy-0.093
Token was
Token ID373
Feature activation+8.03e-02

Pos logit contributions
 com+0.008
 icy+0.008
 warm+0.008
 grass+0.007
 very+0.007
Neg logit contributions
mate-0.014
orman-0.013
 policy-0.013
anish-0.013
date-0.013
Token.
Token ID13
Feature activation-7.40e-03

Pos logit contributions
mate+0.022
orman+0.020
anish+0.020
date+0.020
 policy+0.019
Neg logit contributions
 com-0.022
 icy-0.022
 warm-0.021
 grass-0.020
 soft-0.020
Token One
Token ID1881
Feature activation-3.33e-02

Pos logit contributions
mate+0.016
uer+0.016
orman+0.015
fer+0.015
uers+0.015
Neg logit contributions
 com-0.007
 grass-0.007
 orange-0.007
 warm-0.007
 icy-0.007
Token day
Token ID1110
Feature activation-1.13e-01

Pos logit contributions
orman+0.048
mate+0.048
date+0.047
oker+0.047
 policy+0.046
Neg logit contributions
 com-0.050
 vel-0.046
 crunch-0.044
 icy-0.044
 clouds-0.043
Token,
Token ID11
Feature activation-4.61e-02

Pos logit contributions
uer+0.259
mate+0.249
rum+0.247
 Majesty+0.247
 client+0.245
Neg logit contributions
 a-0.123
 very-0.112
 warm-0.112
 lots-0.110
 all-0.110
Token Jill
Token ID22772
Feature activation-6.19e-02

Pos logit contributions
mate+0.081
uer+0.078
orman+0.077
rum+0.077
 Majesty+0.076
Neg logit contributions
 com-0.064
 warm-0.064
 icy-0.063
 very-0.062
 bright-0.060
Token suggested
Token ID5220
Feature activation-2.05e-01

Pos logit contributions
mate+0.124
orman+0.118
anish+0.116
uing+0.114
 policy+0.114
Neg logit contributions
 com-0.071
 icy-0.067
 warm-0.067
 grass-0.062
 soft-0.061
Token that
Token ID326
Feature activation-1.04e-01

Pos logit contributions
mate+0.469
uer+0.436
oker+0.434
 Troll+0.434
anish+0.433
Neg logit contributions
 warm-0.193
 very-0.189
 com-0.186
 icy-0.182
 soft-0.175
Token they
Token ID484
Feature activation-6.94e-02

Pos logit contributions
mate+0.201
orman+0.193
anish+0.188
uer+0.188
uing+0.185
Neg logit contributions
 com-0.128
 warm-0.120
 icy-0.120
 grass-0.111
 soft-0.110
Token go
Token ID467
Feature activation+3.22e-02

Pos logit contributions
mate+0.133
orman+0.129
anish+0.125
 Troll+0.123
 policy+0.123
Neg logit contributions
 com-0.084
 warm-0.080
 icy-0.080
 soft-0.074
 grass-0.074
Token and
Token ID290
Feature activation+1.67e-01

Pos logit contributions
 com+0.041
 icy+0.039
 warm+0.038
 grass+0.036
 soft+0.036
Neg logit contributions
mate-0.061
orman-0.058
anish-0.057
 Troll-0.056
 policy-0.055
Token pick
Token ID2298
Feature activation+5.19e-02

Pos logit contributions
 com+0.195
 warm+0.194
 icy+0.175
 very+0.173
 grass+0.168
Neg logit contributions
mate-0.334
orman-0.327
anish-0.327
uer-0.326
rum-0.325
Token The
Token ID383
Feature activation+6.46e-02

Pos logit contributions
mate+0.002
 policy+0.002
uing+0.002
orman+0.002
anish+0.002
Neg logit contributions
 com-0.001
 icy-0.001
 grass-0.001
 warm-0.001
 soft-0.001
Token fish
Token ID5916
Feature activation+8.05e-02

Pos logit contributions
 com+0.114
 icy+0.102
 vel+0.102
 warm+0.102
 grass+0.099
Neg logit contributions
orman-0.081
mate-0.078
 Troll-0.078
oker-0.075
anish-0.075
Token listened
Token ID16399
Feature activation-2.94e-03

Pos logit contributions
 com+0.103
 icy+0.099
 warm+0.099
 very+0.093
 soft+0.091
Neg logit contributions
mate-0.150
orman-0.143
 policy-0.139
anish-0.138
oker-0.138
Token carefully
Token ID7773
Feature activation+1.52e-02

Pos logit contributions
mate+0.006
orman+0.006
anish+0.006
uing+0.006
uer+0.006
Neg logit contributions
 com-0.003
 icy-0.003
 warm-0.003
 grass-0.003
 soft-0.003
Token and
Token ID290
Feature activation+1.64e-01

Pos logit contributions
 com+0.007
 warm+0.007
 icy+0.007
 very+0.006
 grass+0.006
Neg logit contributions
mate-0.040
orman-0.040
oker-0.039
 Majesty-0.038
uer-0.038
Token suggested
Token ID5220
Feature activation-2.04e-01

Pos logit contributions
 com+0.198
 icy+0.191
 warm+0.189
 grass+0.178
 very+0.175
Neg logit contributions
mate-0.319
orman-0.303
anish-0.298
 policy-0.297
oker-0.294
Token that
Token ID326
Feature activation-1.07e-01

Pos logit contributions
mate+0.446
orman+0.419
anish+0.419
 policy+0.417
date+0.414
Neg logit contributions
 com-0.202
 icy-0.196
 warm-0.194
 very-0.185
 soft-0.178
Token the
Token ID262
Feature activation+3.76e-02

Pos logit contributions
mate+0.214
orman+0.206
anish+0.200
 Troll+0.198
 policy+0.197
Neg logit contributions
 com-0.121
 icy-0.113
 warm-0.111
 soft-0.103
 grass-0.103
Token crab
Token ID32202
Feature activation+5.60e-02

Pos logit contributions
 com+0.050
 icy+0.047
 warm+0.047
 grass+0.044
 soft+0.043
Neg logit contributions
mate-0.069
orman-0.066
anish-0.065
 policy-0.063
uing-0.063
Token should
Token ID815
Feature activation-3.93e-02

Pos logit contributions
 com+0.056
 icy+0.054
 very+0.054
 warm+0.054
 even+0.050
Neg logit contributions
mate-0.122
anish-0.116
 avail-0.115
orman-0.115
oker-0.114
Token explore
Token ID7301
Feature activation+1.84e-02

Pos logit contributions
mate+0.068
orman+0.065
anish+0.064
uer+0.063
 Troll+0.063
Neg logit contributions
 com-0.055
 warm-0.054
 icy-0.053
 soft-0.050
 grass-0.049
Token is
Token ID318
Feature activation+7.60e-02

Pos logit contributions
 com+0.114
 warm+0.109
 icy+0.109
 grass+0.104
 soft+0.102
Neg logit contributions
mate-0.103
orman-0.100
 Troll-0.097
anish-0.097
 begs-0.096
Token not
Token ID407
Feature activation+1.13e-02

Pos logit contributions
 com+0.085
 clouds+0.077
 grass+0.074
 vel+0.072
 warm+0.071
Neg logit contributions
mate-0.131
orman-0.126
 policy-0.126
 password-0.125
 Troll-0.125
Token big
Token ID1263
Feature activation-9.38e-03

Pos logit contributions
 com+0.009
 icy+0.008
 warm+0.008
 grass+0.007
 clouds+0.007
Neg logit contributions
mate-0.026
orman-0.025
anish-0.024
 policy-0.024
 Troll-0.024
Token and
Token ID290
Feature activation+1.79e-01

Pos logit contributions
orman+0.022
mate+0.021
uers+0.021
uer+0.020
date+0.020
Neg logit contributions
 icy-0.009
 orange-0.009
 even-0.009
 warm-0.009
 very-0.009
Token yellow
Token ID7872
Feature activation+8.53e-02

Pos logit contributions
 com+0.170
 warm+0.151
 icy+0.150
 clouds+0.147
 grass+0.146
Neg logit contributions
mate-0.340
oker-0.330
 policy-0.325
orman-0.322
 Troll-0.322
Token and
Token ID290
Feature activation+1.78e-01

Pos logit contributions
 icy+0.101
 com+0.099
 grass+0.097
 warm+0.097
 clouds+0.094
Neg logit contributions
orman-0.169
mate-0.165
uers-0.163
uer-0.162
uing-0.158
Token happy
Token ID3772
Feature activation-3.80e-02

Pos logit contributions
 com+0.197
 icy+0.182
 warm+0.178
 especially+0.174
 parked+0.174
Neg logit contributions
mate-0.286
 Troll-0.276
oker-0.276
 policy-0.276
anish-0.274
Token.
Token ID13
Feature activation+6.16e-03

Pos logit contributions
mate+0.075
orman+0.075
uer+0.073
uers+0.072
date+0.071
Neg logit contributions
 com-0.047
 icy-0.045
 warm-0.045
 very-0.043
 grass-0.042
Token It
Token ID632
Feature activation+6.97e-02

Pos logit contributions
 com+0.010
 icy+0.010
 warm+0.009
 grass+0.009
 soft+0.009
Neg logit contributions
mate-0.009
orman-0.009
anish-0.009
 Troll-0.008
 policy-0.008
Token is
Token ID318
Feature activation+7.37e-02

Pos logit contributions
 com+0.113
 warm+0.109
 icy+0.108
 fresh+0.104
 grass+0.103
Neg logit contributions
mate-0.092
orman-0.091
 Troll-0.091
 begs-0.091
anish-0.088
Token brown
Token ID7586
Feature activation+3.22e-02

Pos logit contributions
 com+0.080
 clouds+0.072
 grass+0.070
 icy+0.069
 vel+0.068
Neg logit contributions
mate-0.131
 policy-0.126
orman-0.126
 password-0.126
 Troll-0.124
Token laugh
Token ID6487
Feature activation+8.41e-02

Pos logit contributions
 com+0.284
 icy+0.276
 warm+0.261
 clouds+0.256
 grass+0.254
Neg logit contributions
mate-0.251
 policy-0.238
oker-0.238
anish-0.236
orman-0.234
Token.
Token ID13
Feature activation+2.52e-03

Pos logit contributions
 com+0.094
 icy+0.088
 warm+0.086
 grass+0.080
 soft+0.079
Neg logit contributions
mate-0.170
orman-0.163
anish-0.159
date-0.156
uing-0.156
Token<|endoftext|>
Token ID50256
Feature activation-1.08e-02

Pos logit contributions
 com+0.004
 icy+0.004
 warm+0.003
 vel+0.003
 refreshing+0.003
Neg logit contributions
mate-0.004
orman-0.004
anish-0.004
 policy-0.004
 Troll-0.004
TokenL
Token ID43
Feature activation+7.07e-03

Pos logit contributions
mate+0.016
orman+0.015
anish+0.014
 Troll+0.014
date+0.014
Neg logit contributions
 com-0.019
 icy-0.018
 warm-0.017
 soft-0.017
 grass-0.016
Tokenila
Token ID10102
Feature activation+2.46e-02

Pos logit contributions
 com+0.013
 icy+0.012
 warm+0.012
 grass+0.012
 soft+0.012
Neg logit contributions
mate-0.009
orman-0.009
uer-0.009
anish-0.009
 Troll-0.008
Token and
Token ID290
Feature activation+1.68e-01

Pos logit contributions
 com+0.022
 icy+0.020
 warm+0.020
 soft+0.018
 vel+0.018
Neg logit contributions
mate-0.052
orman-0.052
anish-0.051
 Troll-0.049
oker-0.049
Token True
Token ID6407
Feature activation-8.11e-02

Pos logit contributions
 com+0.242
 icy+0.232
 warm+0.214
 grass+0.211
 refreshing+0.208
Neg logit contributions
mate-0.281
orman-0.269
anish-0.257
date-0.256
 client-0.255
Token were
Token ID547
Feature activation+5.94e-02

Pos logit contributions
mate+0.160
orman+0.152
anish+0.148
 policy+0.146
uing+0.146
Neg logit contributions
 com-0.097
 warm-0.092
 icy-0.091
 grass-0.085
 soft-0.084
Token best
Token ID1266
Feature activation-7.10e-02

Pos logit contributions
 com+0.082
 icy+0.078
 warm+0.077
 grass+0.074
 clouds+0.072
Neg logit contributions
mate-0.099
orman-0.096
anish-0.093
date-0.093
 policy-0.091
Token friends
Token ID2460
Feature activation-1.57e-02

Pos logit contributions
mate+0.107
orman+0.101
anish+0.098
date+0.097
 policy+0.096
Neg logit contributions
 com-0.113
 icy-0.109
 warm-0.107
 grass-0.103
 soft-0.100
Token.
Token ID13
Feature activation-4.45e-03

Pos logit contributions
mate+0.022
orman+0.019
anish+0.019
 policy+0.019
date+0.019
Neg logit contributions
 com-0.027
 icy-0.026
 warm-0.025
 grass-0.025
 refreshing-0.024
Token liked
Token ID8288
Feature activation+5.65e-02

Pos logit contributions
 com+0.021
 icy+0.019
 warm+0.019
 fresh+0.018
 soft+0.018
Neg logit contributions
mate-0.026
orman-0.025
 Troll-0.024
 begs-0.024
 client-0.024
Token to
Token ID284
Feature activation-8.84e-02

Pos logit contributions
 com+0.085
 icy+0.082
 warm+0.080
 grass+0.075
 clouds+0.074
Neg logit contributions
mate-0.083
orman-0.078
 occasions-0.076
 policy-0.075
 client-0.074
Token feel
Token ID1254
Feature activation+1.51e-01

Pos logit contributions
mate+0.153
orman+0.146
anish+0.142
 policy+0.139
 Troll+0.139
Neg logit contributions
 com-0.125
 warm-0.120
 icy-0.119
 grass-0.112
 soft-0.112
Token their
Token ID511
Feature activation-5.82e-02

Pos logit contributions
 com+0.157
 icy+0.144
 vel+0.140
 orange+0.135
 clouds+0.135
Neg logit contributions
mate-0.271
orman-0.253
oker-0.252
ings-0.251
 policy-0.250
Token wool
Token ID25749
Feature activation+8.48e-02

Pos logit contributions
mate+0.093
 policy+0.089
anish+0.086
 password+0.086
 Majesty+0.086
Neg logit contributions
 com-0.077
 icy-0.072
 even-0.069
 especially-0.068
 warm-0.068
Token and
Token ID290
Feature activation+1.76e-01

Pos logit contributions
 com+0.102
 icy+0.096
 warm+0.093
 grass+0.087
 soft+0.087
Neg logit contributions
mate-0.161
orman-0.150
anish-0.150
uing-0.148
date-0.147
Token to
Token ID284
Feature activation-8.92e-02

Pos logit contributions
 com+0.235
 icy+0.226
 warm+0.207
 clouds+0.206
 orange+0.203
Neg logit contributions
mate-0.288
 purposes-0.275
 policy-0.274
 occasions-0.271
 password-0.270
Token jump
Token ID4391
Feature activation+8.37e-02

Pos logit contributions
mate+0.147
orman+0.135
anish+0.133
uing+0.130
 Majesty+0.130
Neg logit contributions
 com-0.130
 icy-0.124
 warm-0.117
 grass-0.113
 clouds-0.113
Token and
Token ID290
Feature activation+1.76e-01

Pos logit contributions
 com+0.097
 icy+0.089
 warm+0.085
 vel+0.080
 grass+0.079
Neg logit contributions
mate-0.167
orman-0.151
anish-0.151
date-0.150
ings-0.149
Token run
Token ID1057
Feature activation+7.07e-02

Pos logit contributions
 com+0.245
 icy+0.235
 warm+0.222
 clouds+0.211
 soft+0.211
Neg logit contributions
mate-0.289
orman-0.268
anish-0.264
 Troll-0.261
 occasions-0.260
Token on
Token ID319
Feature activation-5.74e-02

Pos logit contributions
 com+0.090
 icy+0.086
 warm+0.086
 grass+0.080
 soft+0.080
Neg logit contributions
mate-0.132
orman-0.127
anish-0.124
 Troll-0.122
 policy-0.121
Token We
Token ID775
Feature activation-3.52e-02

Pos logit contributions
 com+0.001
 icy+0.001
 warm+0.001
 grass+0.001
 soft+0.001
Neg logit contributions
mate-0.001
orman-0.001
anish-0.001
 Troll-0.001
 policy-0.001
Token have
Token ID423
Feature activation+2.88e-02

Pos logit contributions
mate+0.064
orman+0.062
anish+0.060
 Troll+0.060
 policy+0.059
Neg logit contributions
 com-0.046
 warm-0.045
 icy-0.044
 very-0.042
 soft-0.041
Token to
Token ID284
Feature activation-8.89e-02

Pos logit contributions
 com+0.042
 icy+0.040
 warm+0.039
 grass+0.037
 soft+0.036
Neg logit contributions
mate-0.047
orman-0.046
anish-0.044
 policy-0.044
 purposes-0.043
Token be
Token ID307
Feature activation+7.90e-02

Pos logit contributions
mate+0.146
orman+0.140
anish+0.136
 Troll+0.134
oker+0.133
Neg logit contributions
 com-0.134
 warm-0.131
 icy-0.129
 soft-0.121
 grass-0.120
Token patient
Token ID5827
Feature activation-2.55e-02

Pos logit contributions
 com+0.093
 icy+0.088
 warm+0.087
 grass+0.081
 soft+0.080
Neg logit contributions
mate-0.156
orman-0.149
anish-0.145
 Troll-0.143
 policy-0.143
Token and
Token ID290
Feature activation+1.66e-01

Pos logit contributions
mate+0.067
orman+0.065
 Troll+0.064
oker+0.064
anish+0.064
Neg logit contributions
 com-0.013
 warm-0.012
 icy-0.012
 very-0.011
 soft-0.010
Token wait
Token ID4043
Feature activation-7.23e-02

Pos logit contributions
 com+0.267
 warm+0.257
 icy+0.256
 soft+0.241
 grass+0.240
Neg logit contributions
mate-0.260
orman-0.246
anish-0.240
 Troll-0.233
date-0.232
Token for
Token ID329
Feature activation-7.42e-02

Pos logit contributions
mate+0.137
orman+0.132
anish+0.129
 Troll+0.127
date+0.125
Neg logit contributions
 com-0.091
 icy-0.086
 warm-0.086
 soft-0.079
 grass-0.079
Token the
Token ID262
Feature activation+4.17e-02

Pos logit contributions
mate+0.153
orman+0.145
 policy+0.141
 occasions+0.140
anish+0.140
Neg logit contributions
 com-0.078
 icy-0.073
 warm-0.070
 grass-0.065
 soft-0.065
Token right
Token ID826
Feature activation-1.33e-01

Pos logit contributions
 com+0.054
 icy+0.049
 warm+0.049
 grass+0.048
 vel+0.048
Neg logit contributions
mate-0.071
orman-0.070
 password-0.067
 client-0.066
 policy-0.066
Token time
Token ID640
Feature activation-8.86e-02

Pos logit contributions
mate+0.229
uer+0.215
uers+0.214
anish+0.212
uing+0.212
Neg logit contributions
 com-0.195
 icy-0.193
 warm-0.188
 clouds-0.179
 very-0.178
Token put
Token ID1234
Feature activation+8.81e-02

Pos logit contributions
 com+0.017
 icy+0.017
 warm+0.016
 grass+0.015
 soft+0.015
Neg logit contributions
mate-0.024
orman-0.023
anish-0.022
 Troll-0.022
uing-0.022
Token them
Token ID606
Feature activation-5.30e-03

Pos logit contributions
 com+0.137
 icy+0.130
 warm+0.124
 soft+0.118
 grass+0.117
Neg logit contributions
mate-0.137
orman-0.126
anish-0.121
 Troll-0.120
 password-0.119
Token in
Token ID287
Feature activation-1.04e-02

Pos logit contributions
mate+0.009
orman+0.009
uing+0.009
 password+0.009
anish+0.008
Neg logit contributions
 com-0.007
 icy-0.006
 refreshing-0.006
 warm-0.006
 grass-0.006
Token a
Token ID257
Feature activation+2.23e-02

Pos logit contributions
mate+0.023
orman+0.022
 password+0.021
 policy+0.021
ings+0.021
Neg logit contributions
 com-0.009
 icy-0.008
 warm-0.007
 refreshing-0.007
 clouds-0.007
Token bag
Token ID6131
Feature activation+4.39e-02

Pos logit contributions
 com+0.031
 warm+0.031
 soft+0.029
 icy+0.029
 very+0.028
Neg logit contributions
mate-0.042
anish-0.042
rums-0.041
orman-0.041
rum-0.041
Token and
Token ID290
Feature activation+1.76e-01

Pos logit contributions
 com+0.035
 icy+0.033
 warm+0.032
 grass+0.029
 soft+0.028
Neg logit contributions
mate-0.102
orman-0.099
anish-0.097
 Troll-0.095
 policy-0.095
Token pretended
Token ID36766
Feature activation-1.70e-02

Pos logit contributions
 com+0.212
 icy+0.200
 warm+0.195
 soft+0.182
 grass+0.179
Neg logit contributions
mate-0.330
orman-0.323
 Troll-0.307
anish-0.307
 password-0.306
Token it
Token ID340
Feature activation+1.20e-02

Pos logit contributions
mate+0.035
anish+0.033
uer+0.032
oker+0.032
orman+0.032
Neg logit contributions
 com-0.019
 warm-0.018
 icy-0.018
 very-0.017
 grass-0.017
Token was
Token ID373
Feature activation+8.07e-02

Pos logit contributions
 com+0.013
 icy+0.013
 warm+0.013
 grass+0.012
 soft+0.012
Neg logit contributions
mate-0.024
orman-0.023
anish-0.023
oker-0.023
 policy-0.023
Token their
Token ID511
Feature activation-6.28e-02

Pos logit contributions
 com+0.065
 icy+0.060
 warm+0.058
 grass+0.053
 soft+0.051
Neg logit contributions
mate-0.187
orman-0.181
anish-0.175
 policy-0.173
 Troll-0.173
Token loot
Token ID16702
Feature activation-4.77e-02

Pos logit contributions
mate+0.108
orman+0.104
anish+0.103
rum+0.101
uing+0.100
Neg logit contributions
 com-0.092
 warm-0.089
 icy-0.088
 soft-0.083
 very-0.083
Token the
Token ID262
Feature activation+3.99e-02

Pos logit contributions
mate+0.169
orman+0.164
anish+0.162
uer+0.160
uing+0.159
Neg logit contributions
 com-0.115
 warm-0.112
 icy-0.110
 grass-0.104
 soft-0.102
Token squirrel
Token ID33039
Feature activation+8.48e-02

Pos logit contributions
 com+0.047
 icy+0.046
 warm+0.046
 grass+0.043
 soft+0.042
Neg logit contributions
mate-0.079
orman-0.076
uing-0.075
anish-0.074
oker-0.074
Token.
Token ID13
Feature activation-1.55e-03

Pos logit contributions
 com+0.117
 warm+0.111
 icy+0.111
 grass+0.105
 very+0.104
Neg logit contributions
mate-0.152
orman-0.147
anish-0.144
 policy-0.141
oker-0.140
Token The
Token ID383
Feature activation+6.46e-02

Pos logit contributions
mate+0.003
orman+0.003
anish+0.003
 policy+0.003
uing+0.003
Neg logit contributions
 com-0.002
 warm-0.002
 icy-0.002
 grass-0.002
 soft-0.002
Token squirrel
Token ID33039
Feature activation+8.54e-02

Pos logit contributions
 com+0.076
 icy+0.067
 warm+0.064
 refreshing+0.060
 fresh+0.059
Neg logit contributions
mate-0.122
 Troll-0.116
orman-0.116
anish-0.113
 client-0.112
Token didn
Token ID1422
Feature activation+5.26e-02

Pos logit contributions
 com+0.109
 icy+0.103
 warm+0.102
 grass+0.095
 soft+0.095
Neg logit contributions
mate-0.157
orman-0.150
anish-0.146
 Troll-0.144
uer-0.144
Token't
Token ID470
Feature activation+6.99e-03

Pos logit contributions
 com+0.106
 icy+0.102
 warm+0.101
 grass+0.100
 refreshing+0.097
Neg logit contributions
mate-0.060
orman-0.055
anish-0.050
uing-0.050
 begs-0.049
Token run
Token ID1057
Feature activation+6.87e-02

Pos logit contributions
 warm+0.010
 com+0.010
 icy+0.009
 very+0.009
 even+0.009
Neg logit contributions
mate-0.012
anish-0.012
orman-0.012
 Majesty-0.012
date-0.012
Token away
Token ID1497
Feature activation-1.38e-01

Pos logit contributions
 very+0.108
 warm+0.108
 com+0.107
 icy+0.106
 soft+0.101
Neg logit contributions
 Majesty-0.103
orman-0.102
anish-0.101
mate-0.101
 occasions-0.099
Token and
Token ID290
Feature activation+1.68e-01

Pos logit contributions
orman+0.287
mate+0.285
date+0.279
oker+0.279
anish+0.279
Neg logit contributions
 com-0.142
 very-0.138
 warm-0.138
 icy-0.134
 even-0.127
Token instead
Token ID2427
Feature activation-1.37e-01

Pos logit contributions
 com+0.191
 warm+0.181
 icy+0.178
 very+0.168
 soft+0.165
Neg logit contributions
mate-0.340
orman-0.333
anish-0.331
 policy-0.326
 purposes-0.321
Token Grand
Token ID5675
Feature activation-1.64e-02

Pos logit contributions
 com+0.046
 warm+0.044
 icy+0.043
 soft+0.040
 very+0.040
Neg logit contributions
mate-0.074
orman-0.070
anish-0.070
oker-0.069
uer-0.069
Tokenma
Token ID2611
Feature activation-4.46e-02

Pos logit contributions
mate+0.025
orman+0.024
uer+0.023
anish+0.023
oker+0.023
Neg logit contributions
 com-0.027
 warm-0.026
 icy-0.026
 very-0.024
 orange-0.024
Token.
Token ID13
Feature activation+2.59e-03

Pos logit contributions
mate+0.068
orman+0.066
 policy+0.062
uing+0.062
uer+0.061
Neg logit contributions
 com-0.067
 icy-0.065
 warm-0.062
 grass-0.061
 refreshing-0.060
Token Grand
Token ID5675
Feature activation-1.63e-02

Pos logit contributions
 com+0.003
 warm+0.003
 icy+0.003
 soft+0.003
 grass+0.003
Neg logit contributions
mate-0.005
anish-0.005
orman-0.005
oker-0.005
uer-0.005
Tokenma
Token ID2611
Feature activation-4.43e-02

Pos logit contributions
mate+0.023
orman+0.022
anish+0.021
uer+0.021
oker+0.021
Neg logit contributions
 com-0.029
 icy-0.028
 warm-0.028
 very-0.026
 soft-0.026
Token smiled
Token ID13541
Feature activation+9.80e-02

Pos logit contributions
orman+0.077
mate+0.077
 Troll+0.071
uing+0.071
 begs+0.071
Neg logit contributions
 com-0.056
 icy-0.056
 warm-0.053
 fresh-0.051
 grass-0.051
Token and
Token ID290
Feature activation+1.71e-01

Pos logit contributions
 com+0.092
 icy+0.090
 grass+0.085
 refreshing+0.085
 warm+0.079
Neg logit contributions
mate-0.199
orman-0.193
uer-0.186
 begs-0.186
uing-0.186
Token kissed
Token ID27946
Feature activation+3.47e-02

Pos logit contributions
 icy+0.258
 clouds+0.240
 com+0.240
 grass+0.233
 warm+0.226
Neg logit contributions
orman-0.237
 begs-0.236
mate-0.234
 Troll-0.227
uers-0.223
Token them
Token ID606
Feature activation-5.70e-03

Pos logit contributions
 com+0.042
 warm+0.041
 icy+0.038
 very+0.038
 soft+0.038
Neg logit contributions
mate-0.067
orman-0.066
uer-0.066
 Troll-0.065
 Majesty-0.065
Token.
Token ID13
Feature activation+1.63e-03

Pos logit contributions
mate+0.009
orman+0.009
anish+0.009
date+0.009
uing+0.009
Neg logit contributions
 com-0.008
 icy-0.008
 warm-0.008
 grass-0.007
 soft-0.007
Token
Token ID198
Feature activation+2.44e-02

Pos logit contributions
 com+0.002
 icy+0.002
 warm+0.002
 grass+0.002
 soft+0.002
Neg logit contributions
mate-0.003
orman-0.003
anish-0.003
 Troll-0.003
 policy-0.003
Token looked
Token ID3114
Feature activation+5.89e-02

Pos logit contributions
mate+0.013
orman+0.013
anish+0.012
 Troll+0.012
date+0.012
Neg logit contributions
 com-0.010
 icy-0.010
 warm-0.010
 soft-0.009
 grass-0.009
Token at
Token ID379
Feature activation-1.06e-01

Pos logit contributions
 com+0.093
 icy+0.088
 warm+0.085
 clouds+0.084
 grass+0.084
Neg logit contributions
mate-0.082
orman-0.079
 begs-0.078
uing-0.077
uer-0.075
Token the
Token ID262
Feature activation+4.34e-02

Pos logit contributions
mate+0.171
 avail+0.152
ings+0.151
orman+0.147
anish+0.147
Neg logit contributions
 com-0.162
 vel-0.136
 energ-0.132
 crunch-0.130
 icy-0.128
Token car
Token ID1097
Feature activation+3.71e-02

Pos logit contributions
 com+0.068
 icy+0.063
 warm+0.063
 grass+0.059
 soft+0.059
Neg logit contributions
mate-0.068
orman-0.064
anish-0.062
 policy-0.062
 password-0.061
Token.
Token ID13
Feature activation+4.22e-04

Pos logit contributions
 com+0.036
 icy+0.035
 very+0.035
 warm+0.034
 grass+0.032
Neg logit contributions
mate-0.081
orman-0.079
 Majesty-0.079
anish-0.078
oker-0.077
Token It
Token ID632
Feature activation+6.43e-02

Pos logit contributions
 com+0.000
 warm+0.000
 icy+0.000
 grass+0.000
 clouds+0.000
Neg logit contributions
mate-0.001
orman-0.001
uing-0.001
uer-0.001
 occasions-0.001
Token was
Token ID373
Feature activation+7.90e-02

Pos logit contributions
 com+0.095
 warm+0.088
 icy+0.087
 fresh+0.086
 grass+0.084
Neg logit contributions
mate-0.101
orman-0.100
 password-0.091
 Troll-0.089
anish-0.089
Token ruined
Token ID20484
Feature activation-8.84e-02

Pos logit contributions
 com+0.083
 grass+0.077
 clouds+0.072
 icy+0.072
 vel+0.072
Neg logit contributions
mate-0.145
orman-0.145
 password-0.139
 policy-0.138
ings-0.134
Token.
Token ID13
Feature activation+6.97e-04

Pos logit contributions
mate+0.143
orman+0.134
anish+0.131
 policy+0.129
uing+0.127
Neg logit contributions
 com-0.136
 icy-0.128
 warm-0.127
 grass-0.121
 soft-0.120
Token They
Token ID1119
Feature activation+1.21e-02

Pos logit contributions
 com+0.001
 warm+0.001
 icy+0.001
 grass+0.001
 soft+0.001
Neg logit contributions
mate-0.001
orman-0.001
uing-0.001
uer-0.001
 occasions-0.001
Token felt
Token ID2936
Feature activation+1.23e-01

Pos logit contributions
 com+0.017
 icy+0.016
 warm+0.015
 fresh+0.015
 grass+0.014
Neg logit contributions
mate-0.020
orman-0.020
anish-0.019
 jury-0.019
 Troll-0.018
Token with
Token ID351
Feature activation-7.86e-03

Pos logit contributions
 com+0.014
 icy+0.013
 clouds+0.013
 vel+0.013
 grass+0.013
Neg logit contributions
uing-0.018
mate-0.018
ings-0.018
orman-0.017
anish-0.017
Token lots
Token ID6041
Feature activation+5.71e-02

Pos logit contributions
mate+0.014
 policy+0.014
 purposes+0.014
ings+0.014
 password+0.014
Neg logit contributions
 com-0.007
 icy-0.006
 parked-0.006
 especially-0.006
ocked-0.006
Token of
Token ID286
Feature activation-4.24e-02

Pos logit contributions
 com+0.088
 icy+0.087
 vel+0.084
 prett+0.081
 warm+0.081
Neg logit contributions
mate-0.075
 policy-0.071
uing-0.070
ings-0.069
oker-0.069
Token leaves
Token ID5667
Feature activation+1.49e-01

Pos logit contributions
 occasions+0.083
 purposes+0.080
 policy+0.080
mate+0.077
 password+0.075
Neg logit contributions
 parked-0.033
 com-0.032
 icy-0.030
 vel-0.030
 clouds-0.028
Token to
Token ID284
Feature activation-8.70e-02

Pos logit contributions
 com+0.163
 icy+0.150
 warm+0.144
 grass+0.134
 soft+0.133
Neg logit contributions
mate-0.294
uing-0.278
orman-0.276
ings-0.275
anish-0.273
Token keep
Token ID1394
Feature activation+8.58e-02

Pos logit contributions
mate+0.133
orman+0.121
date+0.117
uing+0.117
 Troll+0.116
Neg logit contributions
 com-0.138
 icy-0.130
 warm-0.125
 clouds-0.122
 soft-0.120
Token them
Token ID606
Feature activation-5.65e-03

Pos logit contributions
 com+0.107
 icy+0.099
 warm+0.093
 grass+0.091
 clouds+0.090
Neg logit contributions
mate-0.158
orman-0.147
uing-0.145
ings-0.143
 policy-0.143
Token warm
Token ID5814
Feature activation+4.66e-02

Pos logit contributions
mate+0.012
uing+0.012
uer+0.011
orman+0.011
ings+0.011
Neg logit contributions
 com-0.005
 icy-0.004
 clouds-0.004
 grass-0.004
 vel-0.004
Token and
Token ID290
Feature activation+1.68e-01

Pos logit contributions
 com+0.058
 icy+0.054
 warm+0.052
 grass+0.050
 soft+0.049
Neg logit contributions
mate-0.085
anish-0.079
orman-0.079
 policy-0.079
date-0.078
Token a
Token ID257
Feature activation+2.36e-02

Pos logit contributions
 com+0.098
 clouds+0.096
 icy+0.096
 vel+0.081
 grass+0.080
Neg logit contributions
mate-0.351
 occasions-0.350
 policy-0.348
 Majesty-0.347
uing-0.346
Token big
Token ID1263
Feature activation-1.11e-02

Pos logit contributions
 com+0.017
 icy+0.016
 clouds+0.014
 warm+0.014
 grass+0.014
Neg logit contributions
mate-0.054
orman-0.052
uer-0.052
uers-0.051
date-0.051
Token day
Token ID1110
Feature activation-1.08e-01

Pos logit contributions
mate+0.041
orman+0.040
oker+0.039
date+0.038
 policy+0.038
Neg logit contributions
 com-0.035
 grass-0.032
 icy-0.032
 warm-0.031
 vel-0.031
Token,
Token ID11
Feature activation-3.96e-02

Pos logit contributions
uer+0.210
uing+0.209
mate+0.204
 purposes+0.202
anish+0.202
Neg logit contributions
 warm-0.132
 com-0.132
 a-0.129
 very-0.125
 icy-0.125
Token Charlie
Token ID11526
Feature activation-1.14e-01

Pos logit contributions
mate+0.084
anish+0.082
uing+0.082
orman+0.081
uer+0.080
Neg logit contributions
 com-0.040
 warm-0.038
 icy-0.038
 very-0.035
 grass-0.035
Token was
Token ID373
Feature activation+8.56e-02

Pos logit contributions
mate+0.254
anish+0.241
uing+0.237
oker+0.236
 policy+0.234
Neg logit contributions
 com-0.120
 warm-0.112
 icy-0.110
 very-0.106
 even-0.104
Token in
Token ID287
Feature activation-1.25e-02

Pos logit contributions
 com+0.085
 icy+0.079
 warm+0.079
 very+0.074
 soft+0.071
Neg logit contributions
mate-0.188
anish-0.181
orman-0.178
 policy-0.177
oker-0.175
Token the
Token ID262
Feature activation+4.10e-02

Pos logit contributions
mate+0.031
orman+0.030
anish+0.030
uing+0.030
uer+0.030
Neg logit contributions
 com-0.009
 icy-0.008
 warm-0.007
 very-0.006
 soft-0.006
Token kitchen
Token ID9592
Feature activation-3.50e-02

Pos logit contributions
 com+0.051
 icy+0.048
 warm+0.048
 grass+0.045
 soft+0.044
Neg logit contributions
mate-0.077
orman-0.074
anish-0.072
 Troll-0.071
date-0.071
Token with
Token ID351
Feature activation-9.54e-03

Pos logit contributions
mate+0.074
orman+0.070
anish+0.070
uer+0.068
oker+0.068
Neg logit contributions
 com-0.039
 icy-0.038
 warm-0.037
 grass-0.034
 very-0.034
Token his
Token ID465
Feature activation-4.15e-02

Pos logit contributions
mate+0.020
orman+0.019
anish+0.019
date+0.018
 Troll+0.018
Neg logit contributions
 com-0.010
 icy-0.009
 warm-0.009
 grass-0.008
 soft-0.008
Token mom
Token ID1995
Feature activation-4.08e-02

Pos logit contributions
mate+0.080
orman+0.078
anish+0.077
uing+0.075
oker+0.075
Neg logit contributions
 com-0.053
 warm-0.051
 icy-0.050
 soft-0.048
 very-0.047
Token.
Token ID13
Feature activation-2.39e-03

Pos logit contributions
mate+0.091
orman+0.089
anish+0.088
uer+0.086
uers+0.086
Neg logit contributions
 com-0.040
 warm-0.039
 icy-0.038
 very-0.036
 soft-0.034
Token park
Token ID3952
Feature activation+4.28e-02

Pos logit contributions
 com+0.047
 vel+0.041
 icy+0.041
 warm+0.041
 even+0.040
Neg logit contributions
mate-0.066
orman-0.065
date-0.064
 Troll-0.063
 password-0.063
Token.
Token ID13
Feature activation-9.56e-04

Pos logit contributions
 very+0.046
 icy+0.045
 warm+0.045
 com+0.045
 so+0.042
Neg logit contributions
mate-0.088
anish-0.087
uing-0.086
orman-0.086
rums-0.085
Token She
Token ID1375
Feature activation+3.13e-03

Pos logit contributions
mate+0.002
uing+0.002
orman+0.002
uer+0.002
anish+0.002
Neg logit contributions
 com-0.001
 warm-0.001
 icy-0.001
 grass-0.001
 soft-0.001
Token did
Token ID750
Feature activation-3.62e-02

Pos logit contributions
 com+0.004
 icy+0.004
 warm+0.004
 soft+0.004
 grass+0.004
Neg logit contributions
mate-0.005
orman-0.005
anish-0.005
 Troll-0.005
 begs-0.005
Token not
Token ID407
Feature activation+8.51e-03

Pos logit contributions
mate+0.073
orman+0.071
anish+0.067
uing+0.067
 policy+0.067
Neg logit contributions
 com-0.040
 icy-0.037
 warm-0.037
 grass-0.035
 clouds-0.034
Token want
Token ID765
Feature activation-2.03e-02

Pos logit contributions
 warm+0.011
 com+0.011
 icy+0.011
 very+0.010
 soft+0.010
Neg logit contributions
mate-0.016
anish-0.016
orman-0.015
uing-0.015
 Troll-0.015
Token to
Token ID284
Feature activation-9.06e-02

Pos logit contributions
mate+0.033
orman+0.031
 policy+0.030
 Troll+0.030
 occasions+0.029
Neg logit contributions
 com-0.030
 icy-0.029
 warm-0.028
 soft-0.026
 grass-0.026
Token make
Token ID787
Feature activation+2.61e-02

Pos logit contributions
mate+0.177
anish+0.172
orman+0.168
oke+0.167
uer+0.166
Neg logit contributions
 warm-0.117
 com-0.112
 icy-0.109
 soft-0.103
 very-0.102
Token Tom
Token ID4186
Feature activation-7.73e-02

Pos logit contributions
 com+0.026
 warm+0.025
 icy+0.024
 soft+0.023
 very+0.022
Neg logit contributions
mate-0.059
anish-0.057
uer-0.056
orman-0.055
oker-0.055
Token unhappy
Token ID19283
Feature activation-6.27e-02

Pos logit contributions
mate+0.158
orman+0.150
anish+0.148
 policy+0.147
 Troll+0.146
Neg logit contributions
 com-0.093
 warm-0.091
 icy-0.089
 very-0.085
 soft-0.083
Token.
Token ID13
Feature activation-1.54e-03

Pos logit contributions
uer+0.142
uers+0.142
orman+0.140
mate+0.140
uing+0.137
Neg logit contributions
 very-0.058
 too-0.056
 com-0.055
 so-0.055
 even-0.054
Token grass
Token ID8701
Feature activation+1.51e-01

Pos logit contributions
 com+0.013
 warm+0.012
 icy+0.012
 refreshing+0.011
 vel+0.011
Neg logit contributions
mate-0.023
anish-0.022
 occasions-0.021
orman-0.021
date-0.021
Token in
Token ID287
Feature activation-1.33e-02

Pos logit contributions
 com+0.194
 warm+0.176
 icy+0.176
 soft+0.163
 fresh+0.162
Neg logit contributions
mate-0.285
anish-0.263
orman-0.262
ings-0.255
date-0.255
Token the
Token ID262
Feature activation+4.02e-02

Pos logit contributions
mate+0.025
orman+0.023
anish+0.023
rum+0.022
 policy+0.022
Neg logit contributions
 com-0.017
 icy-0.016
 warm-0.015
 soft-0.014
 clouds-0.014
Token lawn
Token ID21133
Feature activation+5.99e-02

Pos logit contributions
 com+0.040
 especially+0.034
 refreshing+0.033
 vel+0.033
 warm+0.032
Neg logit contributions
mate-0.073
 ha-0.071
date-0.070
anish-0.069
 aid-0.068
Token and
Token ID290
Feature activation+1.68e-01

Pos logit contributions
 com+0.086
 icy+0.082
 warm+0.081
 grass+0.077
 soft+0.076
Neg logit contributions
mate-0.102
orman-0.097
anish-0.094
 Troll-0.092
 policy-0.092
Token fell
Token ID3214
Feature activation+2.18e-02

Pos logit contributions
 com+0.227
 icy+0.214
 warm+0.211
 soft+0.198
 fresh+0.197
Neg logit contributions
mate-0.285
orman-0.274
anish-0.260
 begs-0.259
 policy-0.254
Token down
Token ID866
Feature activation-5.30e-02

Pos logit contributions
 com+0.034
 refreshing+0.033
 icy+0.032
 warm+0.031
 vel+0.031
Neg logit contributions
mate-0.031
rum-0.028
ings-0.028
anish-0.026
 begs-0.026
Token.
Token ID13
Feature activation-1.87e-03

Pos logit contributions
mate+0.081
orman+0.074
anish+0.072
 policy+0.071
date+0.070
Neg logit contributions
 com-0.084
 warm-0.079
 icy-0.079
 soft-0.075
 grass-0.075
Token She
Token ID1375
Feature activation+3.16e-03

Pos logit contributions
mate+0.003
orman+0.003
anish+0.003
date+0.002
 Troll+0.002
Neg logit contributions
 com-0.003
 icy-0.003
 warm-0.003
 grass-0.003
 soft-0.003
Token started
Token ID2067
Feature activation+1.17e-02

Pos logit contributions
 com+0.005
 fresh+0.005
 warm+0.005
 icy+0.005
 refreshing+0.005
Neg logit contributions
orman-0.004
mate-0.004
 begs-0.004
 ha-0.004
rums-0.004
Token to
Token ID284
Feature activation-9.15e-02

Pos logit contributions
 com+0.019
 icy+0.018
 warm+0.018
 clouds+0.018
 refreshing+0.018
Neg logit contributions
mate-0.015
orman-0.014
 begs-0.014
ings-0.014
 aid-0.013
Token Billy
Token ID15890
Feature activation-6.16e-02

Pos logit contributions
mate+0.254
orman+0.249
anish+0.243
uer+0.238
 Troll+0.238
Neg logit contributions
 com-0.147
 warm-0.142
 icy-0.139
 soft-0.129
 very-0.128
Token.
Token ID13
Feature activation-1.58e-03

Pos logit contributions
mate+0.121
anish+0.115
orman+0.113
 policy+0.113
date+0.112
Neg logit contributions
 com-0.078
 very-0.077
 warm-0.077
 soft-0.072
 icy-0.072
Token "
Token ID366
Feature activation+3.31e-02

Pos logit contributions
mate+0.002
anish+0.002
orman+0.002
uers+0.002
 policy+0.002
Neg logit contributions
 com-0.002
 icy-0.002
 warm-0.002
 clouds-0.002
 vel-0.002
TokenLet
Token ID5756
Feature activation-1.40e-02

Pos logit contributions
 icy+0.049
 com+0.049
 warm+0.046
 cold+0.044
 grass+0.044
Neg logit contributions
mate-0.059
anish-0.054
orman-0.052
date-0.052
fer-0.050
Token's
Token ID338
Feature activation-1.11e-02

Pos logit contributions
mate+0.022
orman+0.021
anish+0.020
date+0.020
 Troll+0.020
Neg logit contributions
 com-0.022
 icy-0.021
 warm-0.021
 grass-0.020
 soft-0.020
Token play
Token ID711
Feature activation+2.56e-02

Pos logit contributions
mate+0.019
orman+0.018
date+0.018
anish+0.018
 Troll+0.017
Neg logit contributions
 com-0.016
 icy-0.015
 warm-0.014
 grass-0.014
 soft-0.014
Token again
Token ID757
Feature activation-3.78e-03

Pos logit contributions
 warm+0.027
 com+0.027
 icy+0.027
 very+0.026
 soft+0.024
Neg logit contributions
mate-0.051
 policy-0.051
orman-0.050
 Troll-0.050
oker-0.050
Token soon
Token ID2582
Feature activation-6.60e-02

Pos logit contributions
mate+0.007
orman+0.007
anish+0.007
oker+0.007
date+0.007
Neg logit contributions
 com-0.005
 icy-0.005
 warm-0.004
 grass-0.004
 very-0.004
Token."
Token ID526
Feature activation-1.24e-02

Pos logit contributions
mate+0.123
orman+0.117
 Troll+0.114
anish+0.112
date+0.109
Neg logit contributions
 com-0.093
 icy-0.091
 warm-0.091
 very-0.087
 so-0.087
Token And
Token ID843
Feature activation+3.54e-02

Pos logit contributions
anish+0.015
mate+0.015
 begs+0.015
uers+0.015
 policy+0.015
Neg logit contributions
 com-0.021
 icy-0.021
 vel-0.020
 clouds-0.020
 warm-0.019
Token with
Token ID351
Feature activation-1.66e-02

Pos logit contributions
 com+0.048
 icy+0.046
 warm+0.044
 grass+0.043
 clouds+0.043
Neg logit contributions
mate-0.062
orman-0.058
 policy-0.057
anish-0.056
date-0.056
Token it
Token ID340
Feature activation+1.76e-02

Pos logit contributions
 com+0.004
 icy+0.003
 warm+0.003
 grass+0.003
 soft+0.003
Neg logit contributions
mate-0.006
orman-0.006
anish-0.006
date-0.006
uer-0.006
Token with
Token ID351
Feature activation-3.98e-03

Pos logit contributions
 com+0.022
 icy+0.021
 warm+0.021
 very+0.020
 soft+0.019
Neg logit contributions
mate-0.033
orman-0.033
 policy-0.032
 avail-0.032
 Troll-0.032
Token the
Token ID262
Feature activation+4.57e-02

Pos logit contributions
mate+0.007
anish+0.007
orman+0.007
 occasions+0.007
 policy+0.007
Neg logit contributions
 com-0.005
 icy-0.004
 warm-0.004
 grass-0.004
 soft-0.004
Token big
Token ID1263
Feature activation-8.93e-03

Pos logit contributions
 com+0.058
 icy+0.055
 warm+0.054
 grass+0.051
 soft+0.051
Neg logit contributions
mate-0.085
orman-0.082
anish-0.080
 Troll-0.078
uing-0.078
Token B
Token ID347
Feature activation+8.91e-02

Pos logit contributions
mate+0.018
orman+0.018
uing+0.017
uer+0.017
date+0.017
Neg logit contributions
 com-0.010
 icy-0.010
 warm-0.009
 grass-0.009
 orange-0.009
Token!"
Token ID2474
Feature activation-1.77e-02

Pos logit contributions
 com+0.170
 icy+0.160
 warm+0.160
 soft+0.153
 very+0.152
Neg logit contributions
mate-0.112
orman-0.104
anish-0.100
uer-0.099
oker-0.099
Token He
Token ID679
Feature activation+1.02e-04

Pos logit contributions
mate+0.035
orman+0.033
anish+0.032
 Troll+0.032
 policy+0.032
Neg logit contributions
 com-0.021
 icy-0.020
 warm-0.020
 grass-0.018
 soft-0.018
Token picks
Token ID11103
Feature activation+4.80e-02

Pos logit contributions
 icy+0.000
 warm+0.000
 com+0.000
 soft+0.000
 fresh+0.000
Neg logit contributions
mate-0.000
orman-0.000
 begs-0.000
uing-0.000
anish-0.000
Token up
Token ID510
Feature activation-8.55e-03

Pos logit contributions
 com+0.050
 icy+0.046
 warm+0.045
 grass+0.041
 soft+0.041
Neg logit contributions
mate-0.100
orman-0.097
anish-0.093
 Troll-0.092
uing-0.092
Token the
Token ID262
Feature activation+4.53e-02

Pos logit contributions
mate+0.016
orman+0.016
ings+0.016
 Troll+0.015
or+0.015
Neg logit contributions
 com-0.009
 icy-0.008
 warm-0.007
 vel-0.007
 clouds-0.007
Token card
Token ID2657
Feature activation+1.98e-02

Pos logit contributions
 com+0.051
 icy+0.048
 warm+0.047
 grass+0.043
 soft+0.043
Neg logit contributions
mate-0.091
orman-0.087
anish-0.085
 Troll-0.084
 policy-0.083
Token.
Token ID13
Feature activation-6.19e-03

Pos logit contributions
mate+0.072
orman+0.068
 policy+0.066
rum+0.064
 Troll+0.062
Neg logit contributions
 com-0.105
 refreshing-0.101
 icy-0.101
 warm-0.095
 fresh-0.095
Token She
Token ID1375
Feature activation+1.66e-03

Pos logit contributions
mate+0.010
orman+0.010
anish+0.010
date+0.009
 policy+0.009
Neg logit contributions
 com-0.009
 icy-0.009
 warm-0.009
 grass-0.008
 soft-0.008
Token loved
Token ID6151
Feature activation+6.63e-02

Pos logit contributions
 com+0.002
 refreshing+0.002
 icy+0.002
 prett+0.002
 sunscreen+0.002
Neg logit contributions
orman-0.003
mate-0.003
gh-0.003
date-0.003
 Troll-0.003
Token to
Token ID284
Feature activation-9.15e-02

Pos logit contributions
 refreshing+0.116
 com+0.114
 icy+0.112
 warm+0.108
 colder+0.107
Neg logit contributions
 occasions-0.069
 client-0.063
 jury-0.061
 guest-0.058
 clients-0.058
Token play
Token ID711
Feature activation+2.57e-02

Pos logit contributions
 occasions+0.109
mate+0.108
date+0.108
 volunteer+0.099
 password+0.099
Neg logit contributions
 icy-0.150
 com-0.150
 refreshing-0.145
 clouds-0.143
 prett-0.140
Token with
Token ID351
Feature activation-1.27e-02

Pos logit contributions
 com+0.028
 icy+0.026
 warm+0.026
 grass+0.024
 soft+0.024
Neg logit contributions
mate-0.053
orman-0.051
anish-0.050
 policy-0.050
 Troll-0.049
Token her
Token ID607
Feature activation-4.61e-02

Pos logit contributions
mate+0.016
 occasions+0.015
 client+0.014
orman+0.014
date+0.014
Neg logit contributions
 refreshing-0.020
 com-0.019
 icy-0.019
 energ-0.019
 warm-0.018
Token toys
Token ID14958
Feature activation+3.75e-02

Pos logit contributions
mate+0.069
 policy+0.063
 client+0.062
orman+0.061
date+0.061
Neg logit contributions
 com-0.069
 icy-0.067
 refreshing-0.063
 warm-0.063
 clouds-0.063
Token and
Token ID290
Feature activation+1.67e-01

Pos logit contributions
 com+0.053
 refreshing+0.051
 icy+0.048
 vel+0.047
 warm+0.045
Neg logit contributions
mate-0.063
date-0.059
uing-0.055
or-0.055
anish-0.054
Token go
Token ID467
Feature activation+3.33e-02

Pos logit contributions
 com+0.240
 icy+0.234
 warm+0.223
 clouds+0.212
 refreshing+0.209
Neg logit contributions
mate-0.272
orman-0.253
 occasions-0.250
anish-0.247
 policy-0.246
Token on
Token ID319
Feature activation-5.81e-02

Pos logit contributions
 com+0.044
 icy+0.042
 refreshing+0.040
 vel+0.040
 clouds+0.039
Neg logit contributions
mate-0.056
date-0.052
orman-0.051
or-0.051
 password-0.050
Token.
Token ID13
Feature activation-4.90e-03

Pos logit contributions
 com+0.062
 icy+0.060
 warm+0.058
 grass+0.054
 soft+0.053
Neg logit contributions
mate-0.112
orman-0.109
anish-0.106
uer-0.105
uing-0.104
Token Something
Token ID13742
Feature activation+9.00e-02

Pos logit contributions
mate+0.009
orman+0.009
uing+0.009
uer+0.008
 occasions+0.008
Neg logit contributions
 com-0.006
 warm-0.006
 icy-0.006
 soft-0.005
 grass-0.005
Token even
Token ID772
Feature activation+9.50e-02

Pos logit contributions
 com+0.108
 icy+0.102
 warm+0.100
 grass+0.094
 soft+0.092
Neg logit contributions
mate-0.173
orman-0.166
anish-0.160
 policy-0.159
 Troll-0.158
Token more
Token ID517
Feature activation+2.17e-02

Pos logit contributions
 com+0.119
 icy+0.112
 warm+0.108
 clouds+0.108
 grass+0.104
Neg logit contributions
mate-0.170
orman-0.162
uing-0.156
 purposes-0.156
anish-0.155
Token unusual
Token ID8468
Feature activation-2.99e-02

Pos logit contributions
 com+0.023
 icy+0.022
 warm+0.021
 grass+0.020
 soft+0.020
Neg logit contributions
mate-0.045
orman-0.044
anish-0.043
date-0.042
 Troll-0.042
Token happened
Token ID3022
Feature activation-1.11e-01

Pos logit contributions
mate+0.060
orman+0.059
uer+0.058
uers+0.058
uing+0.057
Neg logit contributions
 com-0.033
 icy-0.032
 grass-0.032
 warm-0.031
 very-0.031
Token:
Token ID25
Feature activation+2.30e-02

Pos logit contributions
mate+0.226
orman+0.221
 avail+0.221
oker+0.220
uer+0.217
Neg logit contributions
 very-0.117
 com-0.111
 warm-0.110
 icy-0.109
 even-0.106
Token a
Token ID257
Feature activation+2.17e-02

Pos logit contributions
 com+0.023
 icy+0.021
 warm+0.021
 grass+0.019
 soft+0.019
Neg logit contributions
mate-0.048
orman-0.047
anish-0.045
 Troll-0.045
 policy-0.044
Token big
Token ID1263
Feature activation-1.43e-02

Pos logit contributions
 com+0.027
 icy+0.026
 warm+0.026
 soft+0.024
 very+0.024
Neg logit contributions
mate-0.041
orman-0.040
anish-0.039
uing-0.039
 begs-0.038
Token flying
Token ID7348
Feature activation+6.94e-02

Pos logit contributions
mate+0.025
orman+0.025
uing+0.024
anish+0.024
uer+0.023
Neg logit contributions
 com-0.020
 icy-0.019
 warm-0.019
 grass-0.018
 soft-0.018
Token elephant
Token ID20950
Feature activation+7.77e-03

Pos logit contributions
 com+0.102
 icy+0.098
 warm+0.097
 grass+0.091
 soft+0.091
Neg logit contributions
mate-0.115
orman-0.110
anish-0.107
 Troll-0.104
 policy-0.104
Token,
Token ID11
Feature activation-4.41e-02

Pos logit contributions
mate+0.074
orman+0.071
anish+0.070
 Troll+0.069
 policy+0.069
Neg logit contributions
 com-0.031
 icy-0.029
 warm-0.029
 grass-0.026
 soft-0.026
Token but
Token ID475
Feature activation+3.14e-03

Pos logit contributions
mate+0.066
orman+0.062
 begs+0.062
 Troll+0.060
uing+0.060
Neg logit contributions
 com-0.065
 icy-0.063
 warm-0.062
 clouds-0.059
 vel-0.059
Token it
Token ID340
Feature activation+6.74e-03

Pos logit contributions
 com+0.004
 warm+0.004
 icy+0.004
 soft+0.004
 grass+0.004
Neg logit contributions
mate-0.006
orman-0.005
 policy-0.005
anish-0.005
 Troll-0.005
Token was
Token ID373
Feature activation+7.72e-02

Pos logit contributions
 com+0.009
 warm+0.009
 icy+0.008
 fresh+0.008
 grass+0.008
Neg logit contributions
mate-0.012
orman-0.011
anish-0.011
 Troll-0.011
date-0.011
Token too
Token ID1165
Feature activation+5.29e-02

Pos logit contributions
 com+0.070
 grass+0.069
 warm+0.063
 sunshine+0.061
 clouds+0.061
Neg logit contributions
mate-0.146
orman-0.135
 policy-0.134
 problem-0.133
rum-0.132
Token late
Token ID2739
Feature activation-1.34e-01

Pos logit contributions
 vel+0.063
 grass+0.059
 com+0.056
 feathers+0.056
 kisses+0.055
Neg logit contributions
orman-0.082
à-0.075
ings-0.075
 problem-0.074
oker-0.074
Token.
Token ID13
Feature activation-4.91e-03

Pos logit contributions
mate+0.266
date+0.260
oker+0.259
anish+0.259
 Majesty+0.259
Neg logit contributions
 com-0.149
 very-0.145
 icy-0.145
 warm-0.144
 grass-0.134
Token The
Token ID383
Feature activation+6.43e-02

Pos logit contributions
mate+0.009
orman+0.008
anish+0.008
 Troll+0.008
 policy+0.008
Neg logit contributions
 com-0.007
 icy-0.006
 warm-0.006
 grass-0.006
 soft-0.006
Token bridge
Token ID7696
Feature activation+9.07e-02

Pos logit contributions
 com+0.100
 warm+0.091
 icy+0.088
 vel+0.087
 soft+0.086
Neg logit contributions
mate-0.097
orman-0.095
 Troll-0.091
 password-0.091
uers-0.091
Token broke
Token ID6265
Feature activation-4.01e-02

Pos logit contributions
 com+0.105
 icy+0.102
 warm+0.099
 grass+0.092
 very+0.092
Neg logit contributions
mate-0.178
orman-0.170
anish-0.167
 Troll-0.165
 policy-0.164
Token,
Token ID11
Feature activation-4.38e-02

Pos logit contributions
mate+0.080
orman+0.076
anish+0.074
ings+0.073
 policy+0.073
Neg logit contributions
 com-0.045
 warm-0.041
 icy-0.041
 grass-0.038
 soft-0.037
Token,
Token ID11
Feature activation-4.53e-02

Pos logit contributions
 very+0.033
 warm+0.032
 icy+0.031
 com+0.031
 all+0.030
Neg logit contributions
fer-0.055
orman-0.055
oker-0.054
mate-0.054
 avail-0.053
Token but
Token ID475
Feature activation+6.45e-03

Pos logit contributions
 begs+0.077
 inspection+0.069
mate+0.068
 purposes+0.068
 excuse+0.066
Neg logit contributions
 com-0.050
 clouds-0.047
 vel-0.047
 refreshing-0.046
 sunset-0.045
Token he
Token ID339
Feature activation-6.67e-02

Pos logit contributions
 com+0.010
 clouds+0.009
 sprink+0.009
 stars+0.009
 prett+0.009
Neg logit contributions
mate-0.010
orman-0.009
 guy-0.008
 policy-0.008
 ha-0.008
Token didn
Token ID1422
Feature activation+5.32e-02

Pos logit contributions
mate+0.118
orman+0.114
anish+0.107
 Troll+0.107
uer+0.106
Neg logit contributions
 com-0.092
 icy-0.086
 warm-0.086
 grass-0.081
 soft-0.081
Token't
Token ID470
Feature activation+7.17e-03

Pos logit contributions
 com+0.116
 icy+0.111
 grass+0.110
 warm+0.110
 refreshing+0.110
Neg logit contributions
mate-0.052
orman-0.048
anish-0.041
 begs-0.041
 policy-0.040
Token know
Token ID760
Feature activation-8.73e-02

Pos logit contributions
 com+0.012
 warm+0.012
 icy+0.012
 very+0.012
 soft+0.011
Neg logit contributions
mate-0.010
orman-0.010
anish-0.009
 Majesty-0.009
date-0.009
Token how
Token ID703
Feature activation+5.00e-02

Pos logit contributions
mate+0.146
rum+0.134
 policy+0.131
rums+0.128
 purposes+0.126
Neg logit contributions
 com-0.122
 refreshing-0.113
 warm-0.113
 icy-0.113
 vel-0.107
Token.
Token ID13
Feature activation-2.70e-03

Pos logit contributions
 com+0.070
 icy+0.066
 warm+0.066
 grass+0.062
 soft+0.061
Neg logit contributions
mate-0.088
orman-0.084
anish-0.081
 policy-0.080
 Troll-0.080
Token He
Token ID679
Feature activation-4.63e-03

Pos logit contributions
mate+0.004
orman+0.004
anish+0.004
 Troll+0.004
uer+0.004
Neg logit contributions
 com-0.004
 icy-0.004
 warm-0.004
 soft-0.004
 grass-0.004
Token tried
Token ID3088
Feature activation-9.14e-03

Pos logit contributions
mate+0.007
orman+0.007
 Troll+0.007
 begs+0.007
uer+0.007
Neg logit contributions
 com-0.007
 icy-0.007
 warm-0.007
 fresh-0.006
 clouds-0.006
Token to
Token ID284
Feature activation-9.12e-02

Pos logit contributions
mate+0.013
orman+0.011
 policy+0.011
rum+0.011
 Troll+0.011
Neg logit contributions
 com-0.015
 refreshing-0.015
 icy-0.014
 clouds-0.014
 warm-0.014
Token.
Token ID13
Feature activation-3.50e-03

Pos logit contributions
mate+0.080
orman+0.076
anish+0.076
oker+0.073
 policy+0.073
Neg logit contributions
 com-0.060
 warm-0.057
 icy-0.055
 very-0.055
 soft-0.052
Token Now
Token ID2735
Feature activation-4.61e-02

Pos logit contributions
mate+0.006
orman+0.006
anish+0.006
uing+0.006
 policy+0.006
Neg logit contributions
 com-0.005
 warm-0.004
 icy-0.004
 grass-0.004
 soft-0.004
Token,
Token ID11
Feature activation-4.26e-02

Pos logit contributions
mate+0.094
orman+0.093
anish+0.091
 policy+0.091
uer+0.091
Neg logit contributions
 com-0.055
 warm-0.051
 even-0.049
 all-0.048
 so-0.047
Token I
Token ID314
Feature activation-1.12e-04

Pos logit contributions
orman+0.080
mate+0.080
uer+0.078
rum+0.077
anish+0.077
Neg logit contributions
 com-0.054
 warm-0.053
 icy-0.049
 soft-0.048
 grass-0.047
Token want
Token ID765
Feature activation-2.04e-02

Pos logit contributions
mate+0.000
orman+0.000
anish+0.000
 policy+0.000
 avail+0.000
Neg logit contributions
 com-0.000
 warm-0.000
 icy-0.000
 very-0.000
 even-0.000
Token to
Token ID284
Feature activation-8.80e-02

Pos logit contributions
mate+0.036
orman+0.034
anish+0.033
 Troll+0.032
 policy+0.032
Neg logit contributions
 com-0.028
 icy-0.027
 warm-0.026
 grass-0.024
 soft-0.024
Token repay
Token ID20152
Feature activation-6.99e-03

Pos logit contributions
mate+0.153
orman+0.148
anish+0.144
 policy+0.142
oker+0.142
Neg logit contributions
 com-0.125
 warm-0.122
 icy-0.119
 soft-0.112
 grass-0.111
Token her
Token ID607
Feature activation-4.43e-02

Pos logit contributions
mate+0.013
orman+0.013
uer+0.013
uing+0.012
uers+0.012
Neg logit contributions
 com-0.009
 warm-0.009
 icy-0.008
 grass-0.008
 soft-0.008
Token kindness
Token ID23887
Feature activation+2.00e-02

Pos logit contributions
mate+0.073
orman+0.072
anish+0.069
oker+0.068
 policy+0.068
Neg logit contributions
 com-0.067
 warm-0.063
 icy-0.063
 very-0.059
 soft-0.059
Token and
Token ID290
Feature activation+1.69e-01

Pos logit contributions
 com+0.025
 warm+0.024
 icy+0.024
 grass+0.023
 very+0.023
Neg logit contributions
mate-0.037
orman-0.037
 Troll-0.035
uers-0.035
uer-0.035
Token show
Token ID905
Feature activation-5.58e-02

Pos logit contributions
 com+0.247
 warm+0.237
 icy+0.230
 soft+0.222
 grass+0.220
Neg logit contributions
mate-0.291
orman-0.281
anish-0.271
uing-0.267
date-0.265
Token shared
Token ID4888
Feature activation-3.04e-02

Pos logit contributions
mate+0.131
orman+0.120
date+0.118
anish+0.116
uer+0.115
Neg logit contributions
 com-0.098
 icy-0.091
 grass-0.087
 warm-0.085
 soft-0.084
Token this
Token ID428
Feature activation-7.31e-02

Pos logit contributions
mate+0.064
orman+0.061
anish+0.060
 policy+0.059
 Troll+0.059
Neg logit contributions
 com-0.032
 icy-0.030
 warm-0.029
 grass-0.027
 soft-0.027
Token moment
Token ID2589
Feature activation-6.24e-02

Pos logit contributions
mate+0.113
orman+0.109
 policy+0.109
 password+0.108
uer+0.104
Neg logit contributions
 com-0.104
 icy-0.100
 vel-0.100
 refreshing-0.097
 warm-0.097
Token.
Token ID13
Feature activation+1.59e-04

Pos logit contributions
mate+0.111
orman+0.106
anish+0.103
 policy+0.101
uing+0.101
Neg logit contributions
 com-0.085
 icy-0.081
 warm-0.080
 grass-0.075
 soft-0.074
Token He
Token ID679
Feature activation-6.39e-03

Pos logit contributions
 com+0.000
 icy+0.000
 warm+0.000
 grass+0.000
 soft+0.000
Neg logit contributions
mate-0.000
orman-0.000
anish-0.000
 policy-0.000
 Troll-0.000
Token said
Token ID531
Feature activation-1.30e-01

Pos logit contributions
orman+0.007
mate+0.007
 begs+0.007
oke+0.007
uer+0.006
Neg logit contributions
 icy-0.011
 warm-0.011
 com-0.011
 fresh-0.011
 clouds-0.011
Token they
Token ID484
Feature activation-7.14e-02

Pos logit contributions
mate+0.246
orman+0.235
anish+0.230
date+0.226
 Troll+0.226
Neg logit contributions
 com-0.168
 warm-0.157
 icy-0.157
 soft-0.147
 grass-0.144
Token had
Token ID550
Feature activation-2.71e-03

Pos logit contributions
mate+0.110
orman+0.100
uing+0.100
uer+0.097
or+0.097
Neg logit contributions
 com-0.110
 icy-0.108
 grass-0.103
 refreshing-0.101
 warm-0.097
Token to
Token ID284
Feature activation-9.00e-02

Pos logit contributions
orman+0.005
mate+0.005
anish+0.005
 purposes+0.005
 occasions+0.005
Neg logit contributions
 com-0.003
 icy-0.003
 warm-0.003
 grass-0.002
 soft-0.002
Token go
Token ID467
Feature activation+3.65e-02

Pos logit contributions
mate+0.140
orman+0.131
anish+0.127
date+0.125
 password+0.124
Neg logit contributions
 com-0.142
 icy-0.135
 warm-0.130
 grass-0.126
 soft-0.125
Token back
Token ID736
Feature activation-4.61e-02

Pos logit contributions
 com+0.047
 icy+0.044
 warm+0.044
 grass+0.041
 soft+0.041
Neg logit contributions
mate-0.068
orman-0.065
anish-0.063
 Troll-0.062
 policy-0.062