TOP ACTIVATIONS
MAX = 0.081

sm elly ! Tim my wrink led his nose and said
it smelled disgusting ! She wrink led her nose and said
you are a girl . Girls can 't play with balls
borrowed her mom 's magn ifying glass to take a closer
. She noticed the same math book in the same place
, she accidentally dropped the chalk and it broke into many
. Grac ie saw the makeup and wanted to buy some
She falls off the tr amp oline . She lands on
. Em ma wrink led her nose . "
She looked at the heavy makeup box and realized it was
. Jacob whispered very quietly , " Please judge
surprised . She saw the makeup bag on the floor and
. She saw the broken makeup , the broken egg and
you ," Ben said , wrink ling his nose . "
teach me how to clean like you ?" Sally said ,
to eat squash . Squ ash is yellow and soft and
It smelled disgusting . Sandra wrink led her nose and shook
princess . She saw the makeup bag on the dress er
Lily wanted to look pretty like mom my , so she
she decided to throw the math book away . She threw

MIN ACTIVATIONS
MIN = -0.092

knows how to get things unst uck ." So , they
they didn 't get it unst uck . Just
before all the fur was unst uck ! The
what if she was not able to make it ?
books . Far mer David smiled . He
loo and pretended to be Esk im os . After a
as now he wasn 't able to go back to the
say and not be too quick to ignore their opinion .
he never would have been able to do without his own
tasty snack . Far ah and the bunny were
wall . Far ley watched as his mom
scared . Far mer Brown smiled and said
cool . Far ley was so happy to
of cheese and sauce and topp ings . The people who
carrot each ?" Far mer Joe looks at them
it with their many feat hered friends . Bob
could help get the car unst uck . The
and relax . Far mer Joe smiled and said
! But it wasn 't able to move much because the
to grow ! Far mer John smiled . He

INTERVAL 0.038 - 0.081
CONTAINS 1.043%

. Lily told him her ink was frozen . Tom was
oranges . She was a very cheerful girl who always smiled
. He was so amazed by her that he followed her
he went up to the cob web . He saw a
. The next night , Lily didn 't want

INTERVAL -0.005 - 0.038
CONTAINS 64.043%

liked to paint . He liked to use ink to paint
. " What was that ?" asked Lily . " I
the zoo . It was not happy to be in the
whist ling and was sure he was ready to win the
radio . Sara and Ben loved to drive the car and

INTERVAL -0.049 - -0.005
CONTAINS 34.794%

of yours . You can trust me ." The
that she started spl ashing it with water .
joy as he grabbed the biggest and most amazing ice cream
When Tim my woke up , he remembered his dream
sniff ing and w agging his tail . Finally , he

INTERVAL -0.092 - -0.049
CONTAINS 0.120%

shrug . " Don 't worry ," Sally said . "
up until the laser finally let her catch it . El
'm sure you 'll be able to see many more flowers
said , " Don 't worry , I will help you
of himself that he was able to create something tasty .
Token sm
Token ID895
Feature activation-2.89e-03

Pos logit contributions
 chip+0.010
 Twe+0.010
el+0.010
 Cro+0.010
 Tit+0.009
Neg logit contributions
 backstage-0.022
 projector-0.020
arium-0.020
 dining-0.020
dream-0.020
Tokenelly
Token ID6148
Feature activation+4.32e-03

Pos logit contributions
Fine+0.005
ipt+0.005
Flo+0.005
 Tit+0.005
 Cro+0.005
Neg logit contributions
book-0.003
 backstage-0.003
 chess-0.003
dream-0.003
 online-0.003
Token!
Token ID0
Feature activation+2.73e-03

Pos logit contributions
 backstage+0.006
dream+0.006
 projector+0.006
book+0.006
 upon+0.006
Neg logit contributions
Fine-0.007
ipt-0.007
el-0.006
 Cro-0.006
 Din-0.006
Token Tim
Token ID5045
Feature activation+4.83e-03

Pos logit contributions
 chess+0.004
 upon+0.004
dream+0.004
book+0.004
 online+0.004
Neg logit contributions
Fine-0.004
ipt-0.004
 lam-0.004
Flo-0.004
Bl-0.004
Tokenmy
Token ID1820
Feature activation+2.20e-02

Pos logit contributions
 upon+0.004
book+0.003
 chess+0.003
pron+0.003
 coloring+0.003
Neg logit contributions
Fine-0.011
ipt-0.011
 Din-0.011
 Cro-0.011
Flo-0.010
Token wrink
Token ID34866
Feature activation+8.06e-02

Pos logit contributions
 chess+0.023
pron+0.023
dream+0.023
 upon+0.022
book+0.021
Neg logit contributions
ipt-0.042
Fine-0.040
 Din-0.039
 Cro-0.039
Flo-0.038
Tokenled
Token ID992
Feature activation+1.35e-02

Pos logit contributions
book+0.091
 about+0.082
 upon+0.081
dream+0.080
 online+0.078
Neg logit contributions
 Cro-0.147
Bl-0.144
ipt-0.144
Flo-0.142
 Din-0.141
Token his
Token ID465
Feature activation+9.29e-03

Pos logit contributions
dream+0.019
 backstage+0.018
 chess+0.018
book+0.018
 projector+0.018
Neg logit contributions
Fine-0.023
ipt-0.022
Flo-0.019
 Cro-0.019
el-0.019
Token nose
Token ID9686
Feature activation+3.35e-03

Pos logit contributions
 chess+0.015
pron+0.015
 tablet+0.015
book+0.014
 projector+0.014
Neg logit contributions
Fine-0.015
ipt-0.014
Bl-0.012
E-0.012
 Din-0.012
Token and
Token ID290
Feature activation+1.06e-02

Pos logit contributions
 backstage+0.005
 projector+0.004
dream+0.004
book+0.004
 dining+0.004
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
am-0.005
Flo-0.005
Token said
Token ID531
Feature activation+1.88e-02

Pos logit contributions
 chess+0.016
dream+0.016
 backstage+0.015
 projector+0.015
book+0.015
Neg logit contributions
Fine-0.017
ipt-0.017
el-0.014
Flo-0.014
E-0.014
Token it
Token ID340
Feature activation+1.87e-02

Pos logit contributions
 backstage+0.037
 projector+0.035
dream+0.035
book+0.035
 dining+0.035
Neg logit contributions
Fine-0.019
ipt-0.019
el-0.017
 Cro-0.016
 Din-0.015
Token smelled
Token ID37609
Feature activation+6.79e-03

Pos logit contributions
book+0.025
 projector+0.024
 backstage+0.024
 online+0.024
 chess+0.024
Neg logit contributions
Fine-0.034
ipt-0.034
 Cro-0.029
am-0.028
 Din-0.028
Token disgusting
Token ID23374
Feature activation+2.55e-02

Pos logit contributions
 backstage+0.011
 projector+0.010
dream+0.010
book+0.010
 chess+0.010
Neg logit contributions
Fine-0.010
ipt-0.009
E-0.009
el-0.009
Flo-0.009
Token!
Token ID0
Feature activation+6.89e-03

Pos logit contributions
 backstage+0.040
 projector+0.038
dream+0.038
 upon+0.037
 dining+0.037
Neg logit contributions
Fine-0.039
ipt-0.036
el-0.035
 Din-0.032
 Cro-0.032
Token She
Token ID1375
Feature activation+1.20e-03

Pos logit contributions
 upon+0.009
dream+0.009
 chess+0.009
 backstage+0.009
book+0.009
Neg logit contributions
Fine-0.012
ipt-0.011
el-0.011
 ounces-0.011
 Cro-0.010
Token wrink
Token ID34866
Feature activation+7.84e-02

Pos logit contributions
 backstage+0.002
 projector+0.002
dream+0.002
 chess+0.002
 upon+0.002
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.002
 Cro-0.002
Flo-0.002
Tokenled
Token ID992
Feature activation+1.13e-02

Pos logit contributions
book+0.097
dream+0.091
 backstage+0.089
 upon+0.088
 online+0.087
Neg logit contributions
 Cro-0.138
ipt-0.136
Fine-0.135
 Din-0.130
Flo-0.130
Token her
Token ID607
Feature activation+1.98e-02

Pos logit contributions
 backstage+0.018
dream+0.017
book+0.017
 projector+0.017
 chess+0.017
Neg logit contributions
Fine-0.018
ipt-0.017
el-0.015
 Cro-0.014
Flo-0.014
Token nose
Token ID9686
Feature activation+2.51e-03

Pos logit contributions
 backstage+0.039
arium+0.037
 projector+0.037
dream+0.037
epad+0.037
Neg logit contributions
ipt-0.021
Fine-0.021
el-0.020
 ow-0.018
 Tit-0.017
Token and
Token ID290
Feature activation+3.75e-03

Pos logit contributions
 backstage+0.003
 projector+0.003
arium+0.003
 chess+0.003
 dining+0.003
Neg logit contributions
Fine-0.004
ipt-0.004
el-0.004
 Tit-0.004
am-0.004
Token said
Token ID531
Feature activation+1.71e-02

Pos logit contributions
 backstage+0.006
dream+0.006
 chess+0.006
 projector+0.006
book+0.006
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
Flo-0.005
 Cro-0.005
Token you
Token ID345
Feature activation+3.40e-02

Pos logit contributions
Fine+0.018
ipt+0.017
el+0.014
 Cro+0.014
Flo+0.014
Neg logit contributions
 backstage-0.031
dream-0.030
 projector-0.030
book-0.030
 chess-0.029
Token are
Token ID389
Feature activation-1.69e-02

Pos logit contributions
 dining+0.066
 backstage+0.065
book+0.064
 projector+0.063
dream+0.063
Neg logit contributions
Fine-0.040
el-0.036
Flo-0.035
am-0.034
 Cro-0.034
Token a
Token ID257
Feature activation-8.70e-03

Pos logit contributions
Fine+0.020
 Twe+0.019
el+0.018
 Din+0.018
 Cro+0.018
Neg logit contributions
book-0.030
 dining-0.030
 backstage-0.030
dream-0.030
 projector-0.030
Token girl
Token ID2576
Feature activation-2.59e-02

Pos logit contributions
Fine+0.011
ipt+0.010
E+0.009
Flo+0.009
 Cro+0.009
Neg logit contributions
 backstage-0.016
book-0.016
dream-0.015
 projector-0.015
arium-0.015
Token.
Token ID13
Feature activation-5.92e-03

Pos logit contributions
Fine+0.032
ipt+0.030
el+0.029
 Cro+0.027
Flo+0.027
Neg logit contributions
 backstage-0.046
 projector-0.045
dream-0.045
 dining-0.045
book-0.044
Token Girls
Token ID12002
Feature activation+7.70e-02

Pos logit contributions
ipt+0.008
Fine+0.008
ic+0.007
el+0.007
am+0.007
Neg logit contributions
 projector-0.009
 chess-0.009
book-0.009
 backstage-0.009
dream-0.009
Token can
Token ID460
Feature activation+1.50e-03

Pos logit contributions
 dining+0.148
 backstage+0.142
book+0.140
 projector+0.140
dream+0.139
Neg logit contributions
Fine-0.097
el-0.084
 Twe-0.081
 Din-0.079
 Cro-0.078
Token't
Token ID470
Feature activation+2.17e-02

Pos logit contributions
 backstage+0.002
dream+0.002
 projector+0.002
book+0.002
 chess+0.002
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.002
Flo-0.002
 Cro-0.002
Token play
Token ID711
Feature activation+1.83e-02

Pos logit contributions
 backstage+0.026
dream+0.024
 chess+0.024
 online+0.024
 upon+0.023
Neg logit contributions
ipt-0.042
Fine-0.042
 Tit-0.037
el-0.036
 Cro-0.036
Token with
Token ID351
Feature activation+1.24e-02

Pos logit contributions
 backstage+0.023
 chess+0.022
 projector+0.022
dream+0.022
book+0.022
Neg logit contributions
Fine-0.034
ipt-0.033
el-0.030
 Tit-0.029
 Cro-0.029
Token balls
Token ID11333
Feature activation+1.49e-02

Pos logit contributions
 backstage+0.022
 chess+0.021
dream+0.021
 online+0.021
 projector+0.020
Neg logit contributions
Fine-0.017
ipt-0.017
 Cro-0.014
 Tit-0.013
am-0.013
Token borrowed
Token ID22546
Feature activation+1.55e-02

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
 Din+0.001
 Cro+0.001
Neg logit contributions
 backstage-0.001
dream-0.001
 upon-0.001
 chess-0.001
 projector-0.001
Token her
Token ID607
Feature activation+1.03e-02

Pos logit contributions
 backstage+0.023
dream+0.023
 chess+0.022
 online+0.022
 projector+0.022
Neg logit contributions
Fine-0.026
ipt-0.025
am-0.021
el-0.021
Flo-0.020
Token mom
Token ID1995
Feature activation+8.96e-03

Pos logit contributions
 projector+0.020
book+0.020
 backstage+0.020
dream+0.019
epad+0.019
Neg logit contributions
el-0.011
ipt-0.011
 Tit-0.010
 branches-0.010
 Cro-0.010
Token's
Token ID338
Feature activation+9.81e-03

Pos logit contributions
 backstage+0.016
 projector+0.015
dream+0.015
 dining+0.015
epad+0.015
Neg logit contributions
el-0.012
ipt-0.011
Fine-0.011
 Tit-0.010
 chip-0.010
Token magn
Token ID7842
Feature activation+3.56e-02

Pos logit contributions
 backstage+0.019
 projector+0.019
book+0.019
dream+0.019
pron+0.019
Neg logit contributions
ipt-0.010
el-0.010
 branches-0.010
Fine-0.010
 feathers-0.009
Tokenifying
Token ID4035
Feature activation+7.68e-02

Pos logit contributions
 dining+0.029
 backstage+0.029
 projector+0.027
book+0.027
 online+0.026
Neg logit contributions
el-0.082
ipt-0.080
Fine-0.075
Flo-0.071
am-0.071
Token glass
Token ID5405
Feature activation+2.80e-02

Pos logit contributions
 backstage+0.142
dream+0.138
epad+0.135
book+0.133
 online+0.133
Neg logit contributions
ipt-0.095
el-0.090
 Cro-0.090
Fine-0.089
 Tit-0.084
Token to
Token ID284
Feature activation-1.12e-02

Pos logit contributions
 backstage+0.047
dream+0.046
 projector+0.045
 dining+0.045
epad+0.044
Neg logit contributions
el-0.040
ipt-0.039
 branches-0.035
Fine-0.035
 Cro-0.035
Token take
Token ID1011
Feature activation-1.24e-02

Pos logit contributions
Fine+0.014
el+0.014
 Cro+0.013
 Din+0.013
 Tit+0.013
Neg logit contributions
 backstage-0.020
 projector-0.020
 dining-0.019
book-0.019
dream-0.019
Token a
Token ID257
Feature activation-6.58e-03

Pos logit contributions
Fine+0.019
ipt+0.019
el+0.017
 Cro+0.016
Flo+0.016
Neg logit contributions
 backstage-0.019
 projector-0.018
dream-0.018
book-0.018
 chess-0.018
Token closer
Token ID5699
Feature activation-3.90e-02

Pos logit contributions
Flo+0.009
Fine+0.008
E+0.008
ipt+0.008
 Tit+0.008
Neg logit contributions
book-0.010
 dining-0.010
 backstage-0.010
 projector-0.010
 celebrating-0.009
Token.
Token ID13
Feature activation-1.11e-02

Pos logit contributions
 projector+0.037
epad+0.037
pron+0.035
 tablet+0.035
 menu+0.034
Neg logit contributions
Fine-0.013
el-0.013
ipt-0.011
 branches-0.011
 Twe-0.011
Token She
Token ID1375
Feature activation-8.22e-03

Pos logit contributions
Fine+0.011
ipt+0.010
E+0.008
 Cro+0.008
 Din+0.008
Neg logit contributions
 backstage-0.024
 projector-0.023
dream-0.023
 chess-0.023
book-0.022
Token noticed
Token ID6810
Feature activation+2.34e-02

Pos logit contributions
Fine+0.012
ipt+0.012
am+0.010
el+0.010
 Cro+0.010
Neg logit contributions
 backstage-0.014
book-0.014
 projector-0.014
epad-0.014
dream-0.013
Token the
Token ID262
Feature activation+2.71e-02

Pos logit contributions
 projector+0.041
book+0.041
 backstage+0.040
pron+0.040
 dining+0.040
Neg logit contributions
Fine-0.030
ipt-0.029
el-0.027
 Twe-0.025
 Din-0.024
Token same
Token ID976
Feature activation+2.38e-02

Pos logit contributions
uits+0.049
 projector+0.049
arium+0.048
 backstage+0.048
book+0.046
Neg logit contributions
Fine-0.035
 branches-0.029
 lowest-0.029
 rats-0.029
 knots-0.028
Token math
Token ID10688
Feature activation+7.62e-02

Pos logit contributions
 backstage+0.044
 projector+0.044
uits+0.043
arium+0.042
 online+0.042
Neg logit contributions
Fine-0.027
 branches-0.024
 rats-0.023
 brace-0.023
 bird-0.023
Token book
Token ID1492
Feature activation+2.17e-02

Pos logit contributions
 backstage+0.116
 projector+0.116
 upon+0.114
dream+0.113
uits+0.112
Neg logit contributions
Fine-0.113
 brace-0.096
 flies-0.094
 rats-0.092
ipt-0.090
Token in
Token ID287
Feature activation+1.41e-02

Pos logit contributions
 dining+0.037
 projector+0.037
 menus+0.037
 tablet+0.036
 backstage+0.035
Neg logit contributions
Fine-0.023
el-0.023
 flies-0.023
 branches-0.023
 brace-0.022
Token the
Token ID262
Feature activation-1.15e-02

Pos logit contributions
 backstage+0.027
 projector+0.027
dream+0.027
 online+0.026
pron+0.026
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.015
E-0.013
 Din-0.013
Token same
Token ID976
Feature activation-5.60e-03

Pos logit contributions
Fine+0.015
 branches+0.015
el+0.015
 Cro+0.014
ipt+0.014
Neg logit contributions
uits-0.019
dream-0.019
book-0.019
epad-0.019
 backstage-0.019
Token place
Token ID1295
Feature activation-2.73e-03

Pos logit contributions
el+0.008
Fine+0.008
ipt+0.007
 branches+0.007
 nests+0.007
Neg logit contributions
 backstage-0.009
 online-0.009
 projector-0.009
dream-0.008
 celebrating-0.008
Token,
Token ID11
Feature activation+9.91e-03

Pos logit contributions
 backstage+0.000
 projector+0.000
book+0.000
dream+0.000
 dining+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
el-0.000
 Din-0.000
 Cro-0.000
Token she
Token ID673
Feature activation+1.81e-02

Pos logit contributions
 dining+0.021
 menu+0.020
 projector+0.020
 telescope+0.020
 menus+0.020
Neg logit contributions
Fine-0.010
 Twe-0.008
 Din-0.008
E-0.007
 brace-0.007
Token accidentally
Token ID14716
Feature activation+4.25e-02

Pos logit contributions
 backstage+0.032
book+0.030
 projector+0.030
 dining+0.030
 menu+0.030
Neg logit contributions
Fine-0.026
ipt-0.024
am-0.022
el-0.022
E-0.019
Token dropped
Token ID5710
Feature activation+2.08e-02

Pos logit contributions
arium+0.080
 backstage+0.080
 dining+0.078
 projector+0.077
 menu+0.076
Neg logit contributions
Fine-0.047
ipt-0.045
el-0.044
 Din-0.042
am-0.041
Token the
Token ID262
Feature activation+2.05e-02

Pos logit contributions
 dining+0.044
arium+0.043
 backstage+0.043
 projector+0.043
pron+0.043
Neg logit contributions
el-0.020
ipt-0.018
Fine-0.018
 Din-0.017
 Twe-0.017
Token chalk
Token ID30860
Feature activation+7.57e-02

Pos logit contributions
arium+0.044
 projector+0.042
 backstage+0.042
 celebrating+0.042
 dining+0.041
Neg logit contributions
el-0.016
Fine-0.015
 ow-0.015
 spikes-0.015
 Din-0.015
Token and
Token ID290
Feature activation+9.46e-03

Pos logit contributions
 backstage+0.106
 dining+0.105
 projector+0.103
arium+0.102
 upon+0.101
Neg logit contributions
ipt-0.119
el-0.117
Fine-0.115
 Din-0.105
 chip-0.102
Token it
Token ID340
Feature activation-8.50e-04

Pos logit contributions
 backstage+0.018
 projector+0.017
dream+0.017
book+0.017
 chess+0.017
Neg logit contributions
Fine-0.012
ipt-0.011
el-0.010
 Cro-0.009
 Din-0.009
Token broke
Token ID6265
Feature activation+8.32e-03

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
am+0.001
 Cro+0.001
Neg logit contributions
 backstage-0.001
 projector-0.001
dream-0.001
 dining-0.001
book-0.001
Token into
Token ID656
Feature activation+2.68e-02

Pos logit contributions
 backstage+0.010
 chess+0.009
dream+0.009
book+0.009
 projector+0.009
Neg logit contributions
Fine-0.017
ipt-0.016
el-0.014
 Cro-0.014
 Tit-0.014
Token many
Token ID867
Feature activation-5.22e-03

Pos logit contributions
 backstage+0.047
 chess+0.046
book+0.045
dream+0.045
 projector+0.045
Neg logit contributions
Fine-0.038
ipt-0.035
am-0.031
 Cro-0.030
Flo-0.030
Token.
Token ID13
Feature activation+1.86e-02

Pos logit contributions
el+0.004
ipt+0.003
am+0.003
 branches+0.003
 So+0.003
Neg logit contributions
 projector-0.014
 backstage-0.013
 telescope-0.013
 tablet-0.012
epad-0.012
Token Grac
Token ID38099
Feature activation+1.55e-02

Pos logit contributions
 backstage+0.034
book+0.032
 projector+0.032
dream+0.032
epad+0.032
Neg logit contributions
Fine-0.024
ipt-0.024
 Din-0.020
E-0.020
Flo-0.020
Tokenie
Token ID494
Feature activation+1.46e-02

Pos logit contributions
 backstage+0.035
dream+0.035
 projector+0.035
 chess+0.034
book+0.034
Neg logit contributions
Fine-0.013
ipt-0.012
el-0.010
am-0.009
E-0.009
Token saw
Token ID2497
Feature activation+2.14e-02

Pos logit contributions
 backstage+0.027
 projector+0.026
book+0.026
dream+0.026
 chess+0.025
Neg logit contributions
Fine-0.020
ipt-0.019
el-0.016
am-0.016
 Cro-0.015
Token the
Token ID262
Feature activation+2.10e-02

Pos logit contributions
 backstage+0.037
 projector+0.036
book+0.036
dream+0.035
epad+0.035
Neg logit contributions
Fine-0.029
ipt-0.027
el-0.025
E-0.025
 Twe-0.025
Token makeup
Token ID16029
Feature activation+7.51e-02

Pos logit contributions
arium+0.042
 backstage+0.040
book+0.039
 projector+0.039
dream+0.038
Neg logit contributions
 chip-0.022
 tires-0.022
 k-0.022
 bird-0.021
 birds-0.020
Token and
Token ID290
Feature activation+5.35e-03

Pos logit contributions
 backstage+0.113
 projector+0.112
arium+0.110
 upon+0.108
dream+0.107
Neg logit contributions
Fine-0.109
el-0.108
ipt-0.104
 chip-0.098
 Din-0.095
Token wanted
Token ID2227
Feature activation+9.06e-03

Pos logit contributions
 backstage+0.010
 projector+0.010
book+0.010
dream+0.010
 chess+0.010
Neg logit contributions
Fine-0.007
ipt-0.006
el-0.005
 Din-0.005
 Tit-0.005
Token to
Token ID284
Feature activation+4.86e-03

Pos logit contributions
 projector+0.014
 backstage+0.014
 dining+0.013
book+0.013
arium+0.013
Neg logit contributions
Fine-0.013
ipt-0.013
el-0.012
 Cro-0.012
 Din-0.012
Token buy
Token ID2822
Feature activation+1.32e-04

Pos logit contributions
 backstage+0.008
 projector+0.008
dream+0.008
book+0.008
arium+0.008
Neg logit contributions
Fine-0.007
ipt-0.006
el-0.006
E-0.006
 Din-0.006
Token some
Token ID617
Feature activation-2.10e-02

Pos logit contributions
 backstage+0.000
 projector+0.000
dream+0.000
book+0.000
 chess+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
el-0.000
E-0.000
 Din-0.000
Token She
Token ID1375
Feature activation-6.71e-03

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.002
am+0.002
Flo+0.002
Neg logit contributions
dream-0.003
 backstage-0.003
book-0.003
 chess-0.003
 upon-0.003
Token falls
Token ID8953
Feature activation+2.59e-02

Pos logit contributions
Fine+0.008
ipt+0.008
el+0.006
E+0.006
 Din+0.006
Neg logit contributions
 backstage-0.013
dream-0.013
 projector-0.012
 chess-0.012
 upon-0.012
Token off
Token ID572
Feature activation+3.37e-02

Pos logit contributions
 projector+0.035
 backstage+0.035
 dining+0.035
book+0.034
dream+0.034
Neg logit contributions
Fine-0.045
ipt-0.044
el-0.040
 Cro-0.038
Flo-0.038
Token the
Token ID262
Feature activation+1.36e-02

Pos logit contributions
 backstage+0.058
 projector+0.056
dream+0.055
book+0.055
 chess+0.054
Neg logit contributions
Fine-0.047
ipt-0.045
el-0.040
 Cro-0.039
Flo-0.038
Token tr
Token ID491
Feature activation+5.08e-02

Pos logit contributions
 backstage+0.024
dream+0.024
book+0.024
 projector+0.024
 online+0.023
Neg logit contributions
Fine-0.018
ipt-0.017
el-0.014
 Cro-0.014
E-0.014
Tokenamp
Token ID696
Feature activation+7.42e-02

Pos logit contributions
 backstage+0.055
 chess+0.051
book+0.051
 projector+0.051
dream+0.050
Neg logit contributions
Fine-0.103
ipt-0.096
Flo-0.091
 Din-0.091
 Cro-0.090
Tokenoline
Token ID14453
Feature activation+3.42e-02

Pos logit contributions
 backstage+0.098
dream+0.098
 projector+0.095
epad+0.092
 dining+0.091
Neg logit contributions
Fine-0.133
ipt-0.132
el-0.121
am-0.113
 Cro-0.113
Token.
Token ID13
Feature activation-1.21e-03

Pos logit contributions
 backstage+0.053
 projector+0.051
dream+0.051
book+0.050
 dining+0.050
Neg logit contributions
Fine-0.052
ipt-0.050
el-0.046
 Cro-0.044
am-0.043
Token She
Token ID1375
Feature activation-9.57e-03

Pos logit contributions
Fine+0.002
ipt+0.002
el+0.001
am+0.001
Flo+0.001
Neg logit contributions
dream-0.002
 backstage-0.002
book-0.002
 chess-0.002
 upon-0.002
Token lands
Token ID8604
Feature activation+3.82e-02

Pos logit contributions
Fine+0.011
ipt+0.011
E+0.009
 Din+0.009
 Tit+0.008
Neg logit contributions
 backstage-0.018
dream-0.018
 chess-0.018
 upon-0.018
 projector-0.018
Token on
Token ID319
Feature activation+2.94e-02

Pos logit contributions
 projector+0.048
 dining+0.046
 chess+0.046
arium+0.045
pron+0.045
Neg logit contributions
Fine-0.071
ipt-0.069
el-0.063
Flo-0.062
 Cro-0.061
Token.
Token ID13
Feature activation+3.47e-03

Pos logit contributions
 backstage+0.002
dream+0.002
 projector+0.002
book+0.002
epad+0.001
Neg logit contributions
Fine-0.001
ipt-0.001
el-0.001
 Tit-0.001
 So-0.001
Token
Token ID198
Feature activation+9.00e-03

Pos logit contributions
 backstage+0.006
epad+0.006
 projector+0.006
book+0.006
dream+0.006
Neg logit contributions
Fine-0.004
ipt-0.004
E-0.004
 So-0.004
 Twe-0.004
Token
Token ID198
Feature activation+6.01e-03

Pos logit contributions
 backstage+0.016
 projector+0.016
epad+0.015
dream+0.015
 chess+0.015
Neg logit contributions
Fine-0.012
ipt-0.011
E-0.010
el-0.010
Flo-0.009
TokenEm
Token ID10161
Feature activation+2.67e-02

Pos logit contributions
 backstage+0.011
 projector+0.011
epad+0.011
dream+0.011
 menu+0.011
Neg logit contributions
Fine-0.008
E-0.007
ipt-0.007
 So-0.006
Flo-0.006
Tokenma
Token ID2611
Feature activation+1.74e-02

Pos logit contributions
 backstage+0.035
book+0.034
 projector+0.033
dream+0.033
 chess+0.032
Neg logit contributions
Fine-0.048
ipt-0.046
 Cro-0.042
 Din-0.042
Flo-0.042
Token wrink
Token ID34866
Feature activation+7.41e-02

Pos logit contributions
 backstage+0.028
dream+0.027
 projector+0.027
book+0.027
 chess+0.026
Neg logit contributions
Fine-0.027
ipt-0.025
el-0.023
 Cro-0.022
am-0.022
Tokenled
Token ID992
Feature activation+1.07e-02

Pos logit contributions
book+0.091
dream+0.084
 upon+0.084
 backstage+0.082
 online+0.080
Neg logit contributions
 Cro-0.130
Fine-0.127
Bl-0.125
ipt-0.124
 Din-0.124
Token her
Token ID607
Feature activation+1.89e-02

Pos logit contributions
 backstage+0.017
dream+0.016
book+0.016
 projector+0.016
 chess+0.016
Neg logit contributions
Fine-0.017
ipt-0.016
el-0.014
 Cro-0.014
Flo-0.014
Token nose
Token ID9686
Feature activation+1.96e-03

Pos logit contributions
 backstage+0.031
dream+0.030
 chess+0.030
book+0.029
 upon+0.029
Neg logit contributions
Fine-0.030
ipt-0.027
E-0.023
am-0.023
 Cro-0.023
Token.
Token ID13
Feature activation+2.21e-04

Pos logit contributions
 backstage+0.003
 projector+0.003
 chess+0.002
book+0.002
dream+0.002
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.003
am-0.003
Flo-0.003
Token "
Token ID366
Feature activation-3.50e-03

Pos logit contributions
 backstage+0.000
 projector+0.000
book+0.000
epad+0.000
dream+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
E-0.000
 Din-0.000
 So-0.000
Token She
Token ID1375
Feature activation-1.15e-03

Pos logit contributions
Fine+0.006
ipt+0.005
 Din+0.004
 Cro+0.004
E+0.004
Neg logit contributions
 backstage-0.013
 projector-0.013
pron-0.013
 upon-0.013
dream-0.013
Token looked
Token ID3114
Feature activation+2.67e-02

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
am+0.001
E+0.001
Neg logit contributions
 backstage-0.002
 projector-0.002
book-0.002
 dining-0.002
dream-0.002
Token at
Token ID379
Feature activation+1.02e-02

Pos logit contributions
book+0.059
 ago+0.055
epad+0.055
reetings+0.053
 dining+0.053
Neg logit contributions
el-0.023
 flies-0.020
 Twe-0.019
 ac-0.018
 cries-0.017
Token the
Token ID262
Feature activation+1.60e-02

Pos logit contributions
 projector+0.018
 backstage+0.018
dream+0.018
book+0.018
 dining+0.018
Neg logit contributions
Fine-0.013
ipt-0.013
el-0.012
E-0.011
 Tit-0.010
Token heavy
Token ID4334
Feature activation+2.79e-02

Pos logit contributions
arium+0.029
 projector+0.027
uits+0.027
ores+0.026
 online+0.025
Neg logit contributions
Fine-0.021
 brace-0.020
 bird-0.019
 birds-0.019
el-0.019
Token makeup
Token ID16029
Feature activation+7.41e-02

Pos logit contributions
 projector+0.042
 backstage+0.041
arium+0.041
 celebrating+0.040
 online+0.039
Neg logit contributions
el-0.042
Fine-0.041
ipt-0.038
 brace-0.038
 Din-0.035
Token box
Token ID3091
Feature activation-6.14e-03

Pos logit contributions
 backstage+0.112
 projector+0.109
 upon+0.108
arium+0.107
dream+0.107
Neg logit contributions
Fine-0.105
el-0.103
ipt-0.101
 brace-0.093
 chip-0.093
Token and
Token ID290
Feature activation-1.05e-02

Pos logit contributions
el+0.008
ipt+0.008
Fine+0.007
am+0.007
 brace+0.007
Neg logit contributions
 backstage-0.010
 projector-0.010
 menus-0.010
 dining-0.010
 tablet-0.009
Token realized
Token ID6939
Feature activation+7.87e-03

Pos logit contributions
Fine+0.013
ipt+0.011
 brace+0.010
el+0.010
 Twe+0.010
Neg logit contributions
 projector-0.019
book-0.019
 backstage-0.019
 tablet-0.019
 menu-0.018
Token it
Token ID340
Feature activation+3.95e-03

Pos logit contributions
 projector+0.017
 dining+0.016
book+0.016
 backstage+0.016
pron+0.015
Neg logit contributions
Fine-0.008
 Twe-0.007
ipt-0.007
 Din-0.007
el-0.007
Token was
Token ID373
Feature activation-1.07e-02

Pos logit contributions
 backstage+0.007
 projector+0.007
dream+0.007
book+0.007
 chess+0.007
Neg logit contributions
Fine-0.005
ipt-0.005
el-0.004
am-0.004
 Din-0.004
Token.
Token ID13
Feature activation+5.04e-03

Pos logit contributions
 dining+0.013
pron+0.013
 projector+0.012
 backstage+0.012
arium+0.012
Neg logit contributions
Fine-0.008
am-0.008
 tires-0.008
 Cro-0.008
el-0.008
Token
Token ID198
Feature activation+5.84e-03

Pos logit contributions
 backstage+0.007
 projector+0.007
book+0.007
dream+0.006
 upon+0.006
Neg logit contributions
Fine-0.009
ipt-0.008
el-0.008
 Din-0.008
 lam-0.007
Token
Token ID198
Feature activation+1.72e-03

Pos logit contributions
 backstage+0.008
book+0.008
dream+0.008
 projector+0.007
 upon+0.007
Neg logit contributions
Fine-0.010
ipt-0.010
el-0.009
 Din-0.009
 Cro-0.009
TokenJacob
Token ID46751
Feature activation-2.34e-03

Pos logit contributions
 backstage+0.003
book+0.003
 projector+0.003
dream+0.003
 upon+0.003
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.002
 Din-0.002
 Cro-0.002
Token whispered
Token ID25029
Feature activation+2.57e-02

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
am+0.003
 Cro+0.003
Neg logit contributions
 backstage-0.004
book-0.004
 chess-0.004
 projector-0.004
dream-0.004
Token very
Token ID845
Feature activation+7.36e-02

Pos logit contributions
 projector+0.040
 dining+0.039
book+0.038
arium+0.038
 backstage+0.038
Neg logit contributions
Fine-0.037
ipt-0.036
am-0.035
 Cro-0.033
 Tit-0.033
Token quietly
Token ID12703
Feature activation+2.68e-03

Pos logit contributions
 backstage+0.111
dream+0.109
 chess+0.107
book+0.105
 online+0.105
Neg logit contributions
Fine-0.118
ipt-0.116
el-0.101
 Din-0.096
am-0.095
Token,
Token ID11
Feature activation+2.90e-02

Pos logit contributions
pron+0.005
 projector+0.005
 dining+0.005
arium+0.005
 chess+0.005
Neg logit contributions
iously-0.003
el-0.003
am-0.002
Fine-0.002
 So-0.002
Token "
Token ID366
Feature activation-8.36e-03

Pos logit contributions
book+0.052
 backstage+0.052
arium+0.052
 projector+0.051
 dining+0.051
Neg logit contributions
Fine-0.036
ipt-0.032
 Cro-0.032
 Din-0.031
 Twe-0.031
TokenPlease
Token ID5492
Feature activation+1.86e-03

Pos logit contributions
ipt+0.011
Fine+0.011
am+0.010
el+0.010
 Tit+0.010
Neg logit contributions
dream-0.013
 backstage-0.013
 upon-0.013
book-0.013
epad-0.012
Token judge
Token ID5052
Feature activation+6.93e-03

Pos logit contributions
 backstage+0.003
book+0.003
 dining+0.003
 projector+0.003
dream+0.003
Neg logit contributions
Fine-0.002
 Cro-0.002
 Tit-0.002
el-0.002
Flo-0.002
Token surprised
Token ID6655
Feature activation-2.41e-03

Pos logit contributions
arium+0.015
pron+0.014
epad+0.014
book+0.014
 menu+0.014
Neg logit contributions
ipt-0.009
Fine-0.009
 Cro-0.009
 Tit-0.009
el-0.009
Token.
Token ID13
Feature activation+3.19e-03

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.004
Flo+0.004
 Cro+0.004
Neg logit contributions
 backstage-0.003
dream-0.003
book-0.003
 projector-0.003
 chess-0.003
Token She
Token ID1375
Feature activation-2.78e-03

Pos logit contributions
 backstage+0.007
 projector+0.007
dream+0.007
book+0.007
epad+0.007
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.002
 Cro-0.002
 Din-0.002
Token saw
Token ID2497
Feature activation+2.07e-02

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.003
 Cro+0.003
E+0.003
Neg logit contributions
 backstage-0.005
dream-0.005
 projector-0.005
book-0.004
 chess-0.004
Token the
Token ID262
Feature activation+2.63e-02

Pos logit contributions
 backstage+0.037
dream+0.036
 projector+0.036
book+0.036
 dining+0.035
Neg logit contributions
Fine-0.027
ipt-0.026
el-0.023
 Cro-0.022
E-0.022
Token makeup
Token ID16029
Feature activation+7.32e-02

Pos logit contributions
arium+0.048
 projector+0.044
 online+0.044
 backstage+0.042
book+0.042
Neg logit contributions
 brace-0.034
Fine-0.033
el-0.032
 ow-0.031
 chip-0.030
Token bag
Token ID6131
Feature activation-3.48e-03

Pos logit contributions
 backstage+0.094
 projector+0.091
dream+0.091
arium+0.090
 upon+0.090
Neg logit contributions
Fine-0.125
ipt-0.123
el-0.120
am-0.109
 Din-0.109
Token on
Token ID319
Feature activation+1.49e-02

Pos logit contributions
el+0.005
ipt+0.005
Fine+0.005
am+0.004
 flies+0.004
Neg logit contributions
 backstage-0.006
 projector-0.005
 dining-0.005
arium-0.005
dream-0.005
Token the
Token ID262
Feature activation-5.78e-03

Pos logit contributions
 backstage+0.026
dream+0.025
 projector+0.025
book+0.025
 online+0.024
Neg logit contributions
Fine-0.020
ipt-0.019
el-0.018
E-0.017
 Cro-0.017
Token floor
Token ID4314
Feature activation+1.55e-02

Pos logit contributions
Fine+0.006
 chip+0.006
 brace+0.006
 branches+0.006
 ow+0.006
Neg logit contributions
arium-0.011
dream-0.011
 upon-0.010
ores-0.010
epad-0.010
Token and
Token ID290
Feature activation-2.73e-03

Pos logit contributions
 projector+0.031
pron+0.030
 backstage+0.030
arium+0.030
 online+0.029
Neg logit contributions
Fine-0.015
ipt-0.015
el-0.015
am-0.013
 branches-0.011
Token.
Token ID13
Feature activation+5.64e-04

Pos logit contributions
pron+0.046
 projector+0.045
ifying+0.044
book+0.044
 telescope+0.043
Neg logit contributions
ipt-0.031
Fine-0.028
el-0.027
am-0.026
 So-0.026
Token She
Token ID1375
Feature activation-2.07e-03

Pos logit contributions
 backstage+0.001
epad+0.001
 projector+0.001
book+0.001
pron+0.001
Neg logit contributions
Fine-0.001
ipt-0.001
 Din-0.001
 Cro-0.001
 Tit-0.000
Token saw
Token ID2497
Feature activation+1.98e-02

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.002
 Cro+0.002
am+0.002
Neg logit contributions
 backstage-0.004
book-0.004
epad-0.004
 chess-0.004
 menu-0.004
Token the
Token ID262
Feature activation+1.84e-02

Pos logit contributions
 backstage+0.033
 projector+0.033
book+0.033
dream+0.032
 dining+0.032
Neg logit contributions
Fine-0.028
ipt-0.026
el-0.024
 Tit-0.024
 Cro-0.024
Token broken
Token ID5445
Feature activation+4.22e-02

Pos logit contributions
 backstage+0.036
arium+0.036
book+0.034
dream+0.034
uits+0.033
Neg logit contributions
el-0.020
Fine-0.020
 chip-0.019
 ow-0.019
ipt-0.018
Token makeup
Token ID16029
Feature activation+7.26e-02

Pos logit contributions
 backstage+0.094
dream+0.090
 projector+0.089
 online+0.088
book+0.088
Neg logit contributions
ipt-0.037
Fine-0.036
el-0.035
 Cro-0.029
 chip-0.028
Token,
Token ID11
Feature activation+9.40e-03

Pos logit contributions
 backstage+0.115
arium+0.113
 upon+0.110
 projector+0.109
 menus+0.106
Neg logit contributions
el-0.100
Fine-0.098
 chip-0.093
ipt-0.092
 Din-0.088
Token the
Token ID262
Feature activation+5.21e-03

Pos logit contributions
arium+0.018
 backstage+0.018
 projector+0.017
book+0.017
igsaw+0.017
Neg logit contributions
Fine-0.010
 brace-0.009
 Twe-0.009
 chip-0.009
 branches-0.009
Token broken
Token ID5445
Feature activation+2.92e-02

Pos logit contributions
arium+0.010
 backstage+0.010
uits+0.009
book+0.009
 projector+0.009
Neg logit contributions
Fine-0.006
 chip-0.006
 ow-0.006
 tires-0.005
 birds-0.005
Token egg
Token ID5935
Feature activation-9.45e-03

Pos logit contributions
 backstage+0.057
arium+0.053
 projector+0.052
 online+0.051
dream+0.051
Neg logit contributions
el-0.035
Fine-0.034
ipt-0.033
 Twe-0.031
 chip-0.031
Token and
Token ID290
Feature activation-2.27e-03

Pos logit contributions
el+0.013
Fine+0.011
ipt+0.011
 chip+0.011
am+0.011
Neg logit contributions
 backstage-0.017
 projector-0.016
arium-0.016
 chess-0.015
epad-0.015
Token you
Token ID345
Feature activation+4.29e-02

Pos logit contributions
ipt+0.021
Fine+0.020
am+0.016
 tires+0.015
iped+0.015
Neg logit contributions
 backstage-0.029
 chess-0.029
dream-0.028
 projector-0.028
book-0.027
Token,"
Token ID553
Feature activation-2.31e-02

Pos logit contributions
 backstage+0.074
 projector+0.071
dream+0.071
book+0.070
 chess+0.070
Neg logit contributions
Fine-0.059
ipt-0.057
el-0.051
 Cro-0.049
 Tit-0.048
Token Ben
Token ID3932
Feature activation-1.78e-02

Pos logit contributions
Fine+0.038
ipt+0.038
Flo+0.032
am+0.032
el+0.032
Neg logit contributions
 chess-0.034
 backstage-0.033
dream-0.033
 upon-0.032
 projector-0.032
Token said
Token ID531
Feature activation+2.20e-03

Pos logit contributions
Fine+0.027
ipt+0.026
el+0.023
Flo+0.022
 Cro+0.022
Neg logit contributions
 backstage-0.028
dream-0.028
book-0.027
 projector-0.027
 chess-0.027
Token,
Token ID11
Feature activation+1.99e-02

Pos logit contributions
 backstage+0.003
dream+0.003
book+0.003
 chess+0.003
 projector+0.003
Neg logit contributions
Fine-0.004
ipt-0.004
el-0.004
E-0.004
Flo-0.004
Token wrink
Token ID34866
Feature activation+7.25e-02

Pos logit contributions
 backstage+0.029
book+0.028
arium+0.028
 projector+0.028
 online+0.027
Neg logit contributions
Fine-0.033
ipt-0.031
 Cro-0.028
 Din-0.028
 Tit-0.027
Tokenling
Token ID1359
Feature activation+2.87e-02

Pos logit contributions
book+0.093
 backstage+0.088
dream+0.087
 upon+0.087
 online+0.083
Neg logit contributions
 Cro-0.122
Fine-0.120
ipt-0.119
Bl-0.117
 Tit-0.115
Token his
Token ID465
Feature activation+7.98e-03

Pos logit contributions
 backstage+0.041
dream+0.040
book+0.039
 projector+0.039
 chess+0.039
Neg logit contributions
Fine-0.049
ipt-0.047
el-0.041
 Cro-0.040
Flo-0.040
Token nose
Token ID9686
Feature activation-8.28e-03

Pos logit contributions
 chess+0.013
 backstage+0.013
book+0.012
 projector+0.012
dream+0.012
Neg logit contributions
Fine-0.013
ipt-0.012
E-0.010
Bl-0.010
Flo-0.010
Token.
Token ID13
Feature activation+3.02e-03

Pos logit contributions
Fine+0.013
ipt+0.012
el+0.011
am+0.011
 Cro+0.011
Neg logit contributions
 backstage-0.013
 projector-0.012
dream-0.012
 chess-0.012
book-0.012
Token "
Token ID366
Feature activation+1.71e-03

Pos logit contributions
 backstage+0.004
dream+0.004
 projector+0.004
book+0.004
 chess+0.004
Neg logit contributions
Fine-0.005
ipt-0.005
el-0.004
 Cro-0.004
Flo-0.004
Token teach
Token ID4545
Feature activation-1.31e-03

Pos logit contributions
book+0.082
 backstage+0.082
 menus+0.081
arium+0.081
 projector+0.081
Neg logit contributions
Fine-0.036
el-0.032
 So-0.027
E-0.027
 Din-0.026
Token me
Token ID502
Feature activation+4.03e-02

Pos logit contributions
Fine+0.002
ipt+0.002
Flo+0.002
 Cro+0.002
 Tit+0.002
Neg logit contributions
 backstage-0.002
 chess-0.002
dream-0.002
 online-0.002
book-0.002
Token how
Token ID703
Feature activation-8.62e-03

Pos logit contributions
dream+0.088
 projector+0.085
epad+0.083
book+0.083
 dining+0.082
Neg logit contributions
el-0.031
ipt-0.030
Fine-0.030
 myself-0.030
 flies-0.027
Token to
Token ID284
Feature activation+9.10e-03

Pos logit contributions
Fine+0.015
ipt+0.014
el+0.013
 Twe+0.013
am+0.012
Neg logit contributions
book-0.012
 projector-0.012
 backstage-0.012
dream-0.012
arium-0.011
Token clean
Token ID3424
Feature activation+1.47e-02

Pos logit contributions
 backstage+0.016
 projector+0.015
dream+0.015
book+0.015
arium+0.015
Neg logit contributions
Fine-0.013
el-0.011
ipt-0.011
E-0.010
Flo-0.010
Token like
Token ID588
Feature activation+7.23e-02

Pos logit contributions
dream+0.028
 backstage+0.027
 projector+0.027
book+0.027
epad+0.026
Neg logit contributions
Fine-0.017
ipt-0.017
el-0.016
 Twe-0.014
E-0.014
Token you
Token ID345
Feature activation+4.50e-02

Pos logit contributions
dream+0.157
 backstage+0.156
book+0.154
arium+0.151
 upon+0.150
Neg logit contributions
el-0.058
Fine-0.058
 Twe-0.056
ipt-0.054
E-0.050
Token?"
Token ID1701
Feature activation+6.10e-03

Pos logit contributions
 projector+0.089
 menus+0.087
uits+0.085
book+0.085
 dining+0.084
Neg logit contributions
Fine-0.038
 birds-0.035
el-0.035
 flies-0.034
 myself-0.034
Token Sally
Token ID25737
Feature activation+9.50e-03

Pos logit contributions
dream+0.008
 backstage+0.008
 chess+0.008
 online+0.008
 upon+0.008
Neg logit contributions
Fine-0.011
ipt-0.010
E-0.009
el-0.009
Flo-0.009
Token said
Token ID531
Feature activation+7.51e-03

Pos logit contributions
 backstage+0.015
dream+0.014
 chess+0.014
book+0.014
 projector+0.014
Neg logit contributions
Fine-0.015
ipt-0.014
el-0.013
 Cro-0.012
Flo-0.012
Token,
Token ID11
Feature activation+2.34e-02

Pos logit contributions
 backstage+0.009
dream+0.009
book+0.009
 projector+0.009
 chess+0.009
Neg logit contributions
Fine-0.014
ipt-0.014
el-0.013
 Cro-0.012
Flo-0.012
Token to
Token ID284
Feature activation-5.72e-03

Pos logit contributions
book+0.076
dream+0.075
 projector+0.073
 backstage+0.071
 dining+0.069
Neg logit contributions
Fine-0.074
ipt-0.072
 Cro-0.071
 Twe-0.070
 rats-0.070
Token eat
Token ID4483
Feature activation-1.66e-02

Pos logit contributions
Fine+0.010
ipt+0.010
el+0.009
 Cro+0.008
Flo+0.008
Neg logit contributions
 backstage-0.008
dream-0.007
 projector-0.007
book-0.007
 chess-0.007
Token squash
Token ID34613
Feature activation+1.92e-03

Pos logit contributions
Fine+0.023
ipt+0.022
el+0.021
E+0.019
Flo+0.019
Neg logit contributions
 backstage-0.028
dream-0.027
 projector-0.027
book-0.027
epad-0.026
Token.
Token ID13
Feature activation-2.33e-03

Pos logit contributions
 backstage+0.003
 projector+0.003
dream+0.003
book+0.003
 dining+0.003
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.003
 Cro-0.002
am-0.002
Token Squ
Token ID5056
Feature activation-3.96e-03

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
 Cro+0.003
E+0.003
Neg logit contributions
 backstage-0.004
dream-0.004
 projector-0.004
book-0.004
 chess-0.004
Tokenash
Token ID1077
Feature activation+7.22e-02

Pos logit contributions
Fine+0.006
ipt+0.006
el+0.005
am+0.005
 Cro+0.005
Neg logit contributions
 backstage-0.007
 projector-0.006
dream-0.006
book-0.006
 chess-0.006
Token is
Token ID318
Feature activation+9.62e-03

Pos logit contributions
 backstage+0.128
dream+0.123
 projector+0.123
book+0.120
 chess+0.120
Neg logit contributions
Fine-0.099
ipt-0.094
el-0.084
 Cro-0.079
Flo-0.078
Token yellow
Token ID7872
Feature activation+7.97e-03

Pos logit contributions
 backstage+0.021
arium+0.020
 telescope+0.020
 dining+0.019
 projector+0.019
Neg logit contributions
 chip-0.010
 Tit-0.009
 ow-0.009
 Cro-0.009
 rats-0.009
Token and
Token ID290
Feature activation+4.89e-03

Pos logit contributions
 backstage+0.011
dream+0.011
 projector+0.011
book+0.010
epad+0.010
Neg logit contributions
Fine-0.014
ipt-0.013
el-0.012
 Cro-0.012
am-0.012
Token soft
Token ID2705
Feature activation+4.45e-03

Pos logit contributions
 backstage+0.008
dream+0.008
 projector+0.008
book+0.008
 chess+0.008
Neg logit contributions
Fine-0.007
ipt-0.007
el-0.006
 Cro-0.006
am-0.006
Token and
Token ID290
Feature activation+2.96e-03

Pos logit contributions
 backstage+0.008
 projector+0.007
dream+0.007
epad+0.007
 upon+0.007
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
Flo-0.005
 chip-0.005
Token It
Token ID632
Feature activation-2.12e-02

Pos logit contributions
ipt+0.002
Fine+0.002
 Din+0.002
am+0.002
 Cro+0.002
Neg logit contributions
 backstage-0.004
 projector-0.004
pron-0.004
 chess-0.004
book-0.004
Token smelled
Token ID37609
Feature activation+3.74e-03

Pos logit contributions
Fine+0.029
ipt+0.028
el+0.024
 Cro+0.023
Flo+0.023
Neg logit contributions
 backstage-0.037
dream-0.036
 projector-0.036
book-0.036
 chess-0.035
Token disgusting
Token ID23374
Feature activation+2.69e-02

Pos logit contributions
 backstage+0.008
epad+0.007
arium+0.007
 projector+0.007
 telescope+0.007
Neg logit contributions
Fine-0.004
E-0.004
 Twe-0.004
el-0.004
 ow-0.004
Token.
Token ID13
Feature activation-1.23e-03

Pos logit contributions
 backstage+0.045
 projector+0.043
dream+0.043
epad+0.043
pron+0.042
Neg logit contributions
Fine-0.038
ipt-0.036
el-0.035
 chip-0.032
am-0.032
Token Sandra
Token ID32464
Feature activation-4.29e-03

Pos logit contributions
Fine+0.002
ipt+0.002
 Din+0.001
 Cro+0.001
am+0.001
Neg logit contributions
 backstage-0.002
 projector-0.002
 chess-0.002
book-0.002
pron-0.002
Token wrink
Token ID34866
Feature activation+7.19e-02

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.005
am+0.004
 Cro+0.004
Neg logit contributions
 backstage-0.008
 projector-0.008
 chess-0.008
book-0.008
dream-0.008
Tokenled
Token ID992
Feature activation+6.73e-03

Pos logit contributions
book+0.097
 backstage+0.095
dream+0.094
 online+0.089
 upon+0.088
Neg logit contributions
Fine-0.118
ipt-0.118
 Cro-0.112
Bl-0.109
 Din-0.108
Token her
Token ID607
Feature activation+1.59e-02

Pos logit contributions
 backstage+0.012
 projector+0.012
arium+0.011
book+0.011
 online+0.011
Neg logit contributions
Fine-0.009
ipt-0.008
el-0.008
 Din-0.007
E-0.007
Token nose
Token ID9686
Feature activation-2.43e-03

Pos logit contributions
 backstage+0.032
arium+0.031
 projector+0.031
 online+0.031
book+0.031
Neg logit contributions
Fine-0.016
ipt-0.016
el-0.015
am-0.014
 Tit-0.014
Token and
Token ID290
Feature activation+1.98e-03

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.003
 chip+0.003
am+0.003
Neg logit contributions
 backstage-0.004
 projector-0.004
arium-0.004
 chess-0.003
 online-0.003
Token shook
Token ID14682
Feature activation+1.52e-02

Pos logit contributions
 backstage+0.004
 projector+0.004
book+0.004
dream+0.003
 online+0.003
Neg logit contributions
Fine-0.003
ipt-0.002
el-0.002
 Tit-0.002
 Din-0.002
Token princess
Token ID21752
Feature activation+1.39e-02

Pos logit contributions
book+0.021
uits+0.021
 dining+0.020
side+0.020
arium+0.020
Neg logit contributions
 bird-0.011
 k-0.010
 chip-0.009
 brace-0.009
 Din-0.009
Token.
Token ID13
Feature activation+8.07e-03

Pos logit contributions
 dining+0.024
 projector+0.024
uits+0.024
dream+0.024
 backstage+0.023
Neg logit contributions
Fine-0.016
el-0.015
 brace-0.015
 bird-0.015
Flo-0.014
Token She
Token ID1375
Feature activation-3.26e-03

Pos logit contributions
 upon+0.010
 backstage+0.010
dream+0.010
 chess+0.010
book+0.010
Neg logit contributions
Fine-0.014
ipt-0.014
am-0.013
el-0.012
 winds-0.012
Token saw
Token ID2497
Feature activation+1.63e-02

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
 Cro+0.004
Flo+0.004
Neg logit contributions
 backstage-0.005
dream-0.005
 projector-0.005
book-0.005
 chess-0.005
Token the
Token ID262
Feature activation+2.04e-02

Pos logit contributions
 backstage+0.025
dream+0.024
 projector+0.024
book+0.024
 chess+0.024
Neg logit contributions
Fine-0.025
ipt-0.024
el-0.022
 Cro-0.021
 Din-0.021
Token makeup
Token ID16029
Feature activation+7.17e-02

Pos logit contributions
arium+0.041
 projector+0.039
book+0.039
ores+0.039
 online+0.038
Neg logit contributions
 brace-0.023
 birds-0.021
 bird-0.021
 spikes-0.019
el-0.019
Token bag
Token ID6131
Feature activation-2.80e-03

Pos logit contributions
dream+0.098
 upon+0.098
 backstage+0.098
 projector+0.097
arium+0.095
Neg logit contributions
el-0.110
Fine-0.107
ipt-0.104
 brace-0.099
 chip-0.098
Token on
Token ID319
Feature activation+6.04e-03

Pos logit contributions
el+0.004
ipt+0.003
Fine+0.003
am+0.003
 flies+0.003
Neg logit contributions
 backstage-0.005
 projector-0.005
 dining-0.005
dream-0.005
 chess-0.004
Token the
Token ID262
Feature activation-1.41e-02

Pos logit contributions
 backstage+0.011
dream+0.010
 projector+0.010
book+0.010
 online+0.010
Neg logit contributions
Fine-0.008
ipt-0.007
el-0.007
E-0.007
 Cro-0.006
Token dress
Token ID6576
Feature activation+2.66e-03

Pos logit contributions
 chip+0.016
 brace+0.015
Fine+0.015
 branches+0.015
el+0.015
Neg logit contributions
dream-0.026
 upon-0.025
arium-0.025
epad-0.023
ores-0.023
Tokener
Token ID263
Feature activation-7.81e-03

Pos logit contributions
 backstage+0.004
 projector+0.004
dream+0.004
 dining+0.004
 chess+0.004
Neg logit contributions
ipt-0.003
el-0.003
Fine-0.003
am-0.003
 Din-0.003
Token Lily
Token ID20037
Feature activation+1.16e-02

Pos logit contributions
 backstage+0.007
 projector+0.007
dream+0.007
book+0.007
 online+0.006
Neg logit contributions
Fine-0.005
ipt-0.005
el-0.004
E-0.004
 Din-0.004
Token wanted
Token ID2227
Feature activation+2.14e-02

Pos logit contributions
 backstage+0.018
 chess+0.017
dream+0.017
 projector+0.017
 upon+0.017
Neg logit contributions
Fine-0.018
ipt-0.017
el-0.015
 Cro-0.015
Flo-0.015
Token to
Token ID284
Feature activation+8.61e-03

Pos logit contributions
arium+0.034
 dining+0.033
 projector+0.033
 backstage+0.033
dream+0.032
Neg logit contributions
el-0.029
Fine-0.028
ipt-0.028
 Din-0.027
 Cro-0.027
Token look
Token ID804
Feature activation+1.01e-02

Pos logit contributions
 backstage+0.014
dream+0.014
 projector+0.014
book+0.013
arium+0.013
Neg logit contributions
Fine-0.013
ipt-0.012
el-0.011
E-0.011
 Din-0.011
Token pretty
Token ID2495
Feature activation+4.53e-02

Pos logit contributions
 dining+0.020
 ago+0.020
 projector+0.019
epad+0.019
book+0.019
Neg logit contributions
E-0.009
el-0.009
Fine-0.009
 So-0.009
 Twe-0.009
Token like
Token ID588
Feature activation+7.16e-02

Pos logit contributions
 projector+0.068
 backstage+0.067
book+0.066
 dining+0.065
dream+0.064
Neg logit contributions
Fine-0.071
el-0.066
ipt-0.065
E-0.061
 Din-0.059
Token mom
Token ID1995
Feature activation+4.84e-03

Pos logit contributions
 backstage+0.132
 projector+0.129
arium+0.127
book+0.126
dream+0.126
Neg logit contributions
Fine-0.084
ipt-0.080
 Din-0.077
el-0.074
 Twe-0.074
Tokenmy
Token ID1820
Feature activation+6.84e-03

Pos logit contributions
 backstage+0.007
 projector+0.007
dream+0.007
book+0.007
 chess+0.007
Neg logit contributions
Fine-0.008
ipt-0.008
el-0.007
 Cro-0.007
 Din-0.007
Token,
Token ID11
Feature activation+1.21e-02

Pos logit contributions
 backstage+0.011
 projector+0.011
dream+0.011
 dining+0.011
arium+0.011
Neg logit contributions
Fine-0.009
el-0.009
ipt-0.009
 Din-0.008
 chip-0.008
Token so
Token ID523
Feature activation+6.78e-03

Pos logit contributions
 backstage+0.019
 projector+0.019
book+0.019
 dining+0.018
dream+0.018
Neg logit contributions
Fine-0.018
ipt-0.017
 Cro-0.016
 Din-0.016
el-0.016
Token she
Token ID673
Feature activation+2.94e-02

Pos logit contributions
dream+0.009
 backstage+0.009
 upon+0.009
 chess+0.009
 online+0.009
Neg logit contributions
Fine-0.012
ipt-0.012
 Cro-0.010
am-0.010
el-0.010
Token she
Token ID673
Feature activation+3.55e-02

Pos logit contributions
 projector+0.062
 backstage+0.061
uits+0.060
pron+0.060
book+0.059
Neg logit contributions
Fine-0.019
el-0.017
ipt-0.016
 Din-0.015
 Twe-0.015
Token decided
Token ID3066
Feature activation+1.33e-02

Pos logit contributions
 backstage+0.056
dream+0.053
 projector+0.053
book+0.053
 chess+0.052
Neg logit contributions
Fine-0.055
ipt-0.053
el-0.047
 Cro-0.046
Flo-0.045
Token to
Token ID284
Feature activation+1.06e-02

Pos logit contributions
 projector+0.027
book+0.025
 dining+0.025
pron+0.023
 tablet+0.023
Neg logit contributions
Fine-0.017
 Twe-0.016
ipt-0.014
 Din-0.013
 Cro-0.013
Token throw
Token ID3714
Feature activation-5.50e-03

Pos logit contributions
 backstage+0.018
 projector+0.017
dream+0.017
book+0.017
 chess+0.017
Neg logit contributions
Fine-0.016
ipt-0.014
el-0.013
 Cro-0.013
 Tit-0.013
Token the
Token ID262
Feature activation+1.01e-02

Pos logit contributions
Fine+0.007
el+0.006
ipt+0.006
E+0.006
 Din+0.005
Neg logit contributions
 backstage-0.010
 dining-0.010
 projector-0.010
pron-0.010
dream-0.010
Token math
Token ID10688
Feature activation+7.13e-02

Pos logit contributions
arium+0.017
 dining+0.016
 projector+0.016
 celebrating+0.016
uits+0.016
Neg logit contributions
Fine-0.014
 lowest-0.012
 rats-0.011
 ounces-0.011
el-0.011
Token book
Token ID1492
Feature activation+1.78e-02

Pos logit contributions
 dining+0.107
 celebrating+0.105
dream+0.105
 upon+0.104
 projector+0.104
Neg logit contributions
Fine-0.103
ipt-0.093
 flies-0.091
 brace-0.090
el-0.090
Token away
Token ID1497
Feature activation+1.41e-02

Pos logit contributions
 dining+0.030
 projector+0.029
pron+0.029
 celebrating+0.029
 backstage+0.028
Neg logit contributions
ipt-0.023
Fine-0.023
el-0.022
am-0.021
 branches-0.021
Token.
Token ID13
Feature activation-5.52e-03

Pos logit contributions
 projector+0.029
pron+0.029
 backstage+0.028
 dining+0.028
dream+0.027
Neg logit contributions
Fine-0.013
ipt-0.012
el-0.012
 branches-0.012
 Cro-0.011
Token She
Token ID1375
Feature activation+1.96e-03

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
 Cro+0.005
Flo+0.005
Neg logit contributions
 backstage-0.010
dream-0.010
book-0.010
 projector-0.010
 chess-0.010
Token threw
Token ID9617
Feature activation+9.18e-03

Pos logit contributions
 backstage+0.003
dream+0.003
 projector+0.003
 chess+0.003
book+0.003
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.002
 Cro-0.002
Flo-0.002
Token knows
Token ID4206
Feature activation-5.97e-03

Pos logit contributions
Fine+0.018
ipt+0.018
el+0.015
E+0.015
 Cro+0.015
Neg logit contributions
 backstage-0.015
 projector-0.015
 online-0.015
 upon-0.014
dream-0.014
Token how
Token ID703
Feature activation-1.90e-02

Pos logit contributions
Fine+0.009
ipt+0.009
el+0.008
 Cro+0.008
 Tit+0.008
Neg logit contributions
dream-0.010
 backstage-0.010
book-0.009
 projector-0.009
 dining-0.009
Token to
Token ID284
Feature activation+2.70e-03

Pos logit contributions
Fine+0.032
ipt+0.030
 Cro+0.028
am+0.028
el+0.028
Neg logit contributions
 backstage-0.027
book-0.027
dream-0.026
 projector-0.026
 dining-0.025
Token get
Token ID651
Feature activation-9.54e-03

Pos logit contributions
 backstage+0.003
 chess+0.003
 online+0.003
 upon+0.003
 projector+0.003
Neg logit contributions
ipt-0.005
Fine-0.005
am-0.004
E-0.004
aw-0.004
Token things
Token ID1243
Feature activation-6.71e-03

Pos logit contributions
Fine+0.011
ipt+0.011
el+0.011
 Twe+0.011
 Cro+0.010
Neg logit contributions
book-0.019
dream-0.018
 backstage-0.018
 dining-0.018
 projector-0.017
Token unst
Token ID15014
Feature activation-9.16e-02

Pos logit contributions
Fine+0.008
 Twe+0.008
ipt+0.008
el+0.008
 k+0.007
Neg logit contributions
dream-0.012
book-0.012
 backstage-0.012
 projector-0.012
 dining-0.012
Tokenuck
Token ID1347
Feature activation-5.27e-02

Pos logit contributions
 ow+0.144
 Mir+0.139
 Tit+0.129
 ounces+0.128
Bl+0.127
Neg logit contributions
 online-0.126
 over-0.112
 downstairs-0.111
 inside-0.109
 forward-0.108
Token."
Token ID526
Feature activation+1.31e-04

Pos logit contributions
el+0.051
Fine+0.050
ipt+0.048
Flo+0.045
am+0.044
Neg logit contributions
dream-0.111
 backstage-0.108
 projector-0.107
 dining-0.106
epad-0.106
Token So
Token ID1406
Feature activation-3.88e-02

Pos logit contributions
 backstage+0.000
 projector+0.000
dream+0.000
book+0.000
epad+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
 Cro-0.000
el-0.000
am-0.000
Token,
Token ID11
Feature activation+5.95e-03

Pos logit contributions
Fine+0.055
ipt+0.051
el+0.049
 Din+0.048
 Twe+0.047
Neg logit contributions
 backstage-0.067
book-0.062
 projector-0.062
epad-0.062
dream-0.061
Token they
Token ID484
Feature activation+1.15e-02

Pos logit contributions
 backstage+0.012
epad+0.012
book+0.011
 dining+0.011
 ago+0.011
Neg logit contributions
Fine-0.006
 Twe-0.006
 Din-0.006
ipt-0.006
 chip-0.005
Token they
Token ID484
Feature activation+1.51e-02

Pos logit contributions
Fine+0.070
ipt+0.067
el+0.058
E+0.057
Flo+0.057
Neg logit contributions
 backstage-0.068
 chess-0.067
dream-0.066
 projector-0.066
 online-0.066
Token didn
Token ID1422
Feature activation+9.26e-04

Pos logit contributions
 backstage+0.023
dream+0.023
book+0.022
 projector+0.022
 chess+0.022
Neg logit contributions
Fine-0.024
ipt-0.023
el-0.020
Flo-0.019
E-0.019
Token't
Token ID470
Feature activation+1.02e-02

Pos logit contributions
 online+0.001
 chess+0.001
dream+0.001
book+0.001
 backstage+0.001
Neg logit contributions
Fine-0.002
ipt-0.002
 Tit-0.002
 lam-0.002
am-0.002
Token get
Token ID651
Feature activation-2.09e-02

Pos logit contributions
 backstage+0.014
dream+0.013
 online+0.013
 chess+0.013
 upon+0.013
Neg logit contributions
ipt-0.018
Fine-0.018
el-0.016
E-0.015
Flo-0.015
Token it
Token ID340
Feature activation-1.47e-02

Pos logit contributions
Fine+0.039
ipt+0.037
el+0.032
E+0.032
Flo+0.032
Neg logit contributions
 backstage-0.028
dream-0.027
 online-0.026
 chess-0.026
 upon-0.026
Token unst
Token ID15014
Feature activation-8.85e-02

Pos logit contributions
Fine+0.027
ipt+0.026
el+0.023
 Cro+0.023
am+0.022
Neg logit contributions
 backstage-0.019
 projector-0.018
dream-0.018
book-0.018
 chess-0.018
Tokenuck
Token ID1347
Feature activation-5.34e-02

Pos logit contributions
 Mir+0.145
 ow+0.142
 Tit+0.131
 ounces+0.129
Fine+0.128
Neg logit contributions
 online-0.108
 forward-0.108
 over-0.099
less-0.098
 downstairs-0.098
Token.
Token ID13
Feature activation+2.02e-04

Pos logit contributions
Fine+0.068
ipt+0.065
el+0.064
am+0.059
Flo+0.056
Neg logit contributions
 backstage-0.096
 projector-0.094
pron-0.093
 dining-0.092
book-0.091
Token
Token ID198
Feature activation-3.05e-03

Pos logit contributions
 upon+0.000
dream+0.000
 chess+0.000
 projector+0.000
book+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
el-0.000
 lam-0.000
 brace-0.000
Token
Token ID198
Feature activation-5.40e-03

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.005
 Din+0.005
 Cro+0.004
Neg logit contributions
 backstage-0.004
dream-0.004
book-0.004
 upon-0.004
 projector-0.004
TokenJust
Token ID5703
Feature activation-4.70e-03

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
 Cro+0.006
Flo+0.006
Neg logit contributions
 backstage-0.010
dream-0.010
 projector-0.010
book-0.009
 chess-0.009
Token before
Token ID878
Feature activation-2.06e-02

Pos logit contributions
Fine+0.033
ipt+0.032
am+0.032
el+0.031
 Cro+0.030
Neg logit contributions
 backstage-0.045
 projector-0.045
pron-0.043
 dining-0.043
 menu-0.043
Token all
Token ID477
Feature activation-1.15e-02

Pos logit contributions
Fine+0.025
ipt+0.024
el+0.022
 Cro+0.022
 Din+0.021
Neg logit contributions
 backstage-0.039
book-0.038
 projector-0.038
 dining-0.036
arium-0.036
Token the
Token ID262
Feature activation-3.12e-02

Pos logit contributions
Fine+0.018
ipt+0.017
el+0.015
 Cro+0.014
Flo+0.014
Neg logit contributions
 backstage-0.018
dream-0.018
 projector-0.018
book-0.018
 chess-0.017
Token fur
Token ID9230
Feature activation-1.27e-02

Pos logit contributions
Fine+0.037
ipt+0.036
el+0.030
 Cro+0.028
E+0.028
Neg logit contributions
 backstage-0.060
dream-0.058
 projector-0.057
book-0.057
 chess-0.057
Token was
Token ID373
Feature activation-2.54e-02

Pos logit contributions
Fine+0.017
ipt+0.016
el+0.015
am+0.014
 Cro+0.014
Neg logit contributions
 backstage-0.023
 projector-0.022
dream-0.021
 chess-0.021
book-0.021
Token unst
Token ID15014
Feature activation-8.74e-02

Pos logit contributions
 branches+0.029
 tires+0.029
 Twe+0.029
el+0.028
am+0.028
Neg logit contributions
 backstage-0.047
 projector-0.046
arium-0.046
 tablet-0.045
 menus-0.045
Tokenuck
Token ID1347
Feature activation-5.21e-02

Pos logit contributions
 ow+0.139
Fine+0.132
 Mir+0.128
 rats+0.127
 Lia+0.123
Neg logit contributions
less-0.116
 online-0.111
 forward-0.108
dream-0.105
 indoors-0.104
Token!
Token ID0
Feature activation-7.02e-03

Pos logit contributions
Fine+0.067
el+0.065
ipt+0.065
am+0.061
Flo+0.058
Neg logit contributions
 backstage-0.092
 projector-0.089
pron-0.088
book-0.087
 dining-0.086
Token
Token ID198
Feature activation-5.41e-03

Pos logit contributions
Fine+0.012
ipt+0.011
el+0.010
am+0.010
Flo+0.010
Neg logit contributions
 backstage-0.010
dream-0.010
 projector-0.010
book-0.010
 chess-0.010
Token
Token ID198
Feature activation-7.20e-03

Pos logit contributions
Fine+0.009
ipt+0.009
E+0.008
Flo+0.007
el+0.007
Neg logit contributions
 backstage-0.008
 projector-0.008
dream-0.008
book-0.008
 chess-0.008
TokenThe
Token ID464
Feature activation-1.36e-02

Pos logit contributions
Fine+0.008
ipt+0.007
E+0.006
Flo+0.006
Bl+0.006
Neg logit contributions
 backstage-0.016
 projector-0.015
 online-0.015
dream-0.015
 chess-0.015
Token what
Token ID644
Feature activation-3.13e-02

Pos logit contributions
Fine+0.003
el+0.002
ipt+0.002
 Din+0.002
 Cro+0.002
Neg logit contributions
 backstage-0.004
 projector-0.004
book-0.004
dream-0.004
pron-0.004
Token if
Token ID611
Feature activation-6.10e-02

Pos logit contributions
Fine+0.039
ipt+0.035
el+0.033
 Tit+0.031
 Cro+0.031
Neg logit contributions
 backstage-0.060
 projector-0.058
dream-0.056
 dining-0.055
 online-0.054
Token she
Token ID673
Feature activation+5.74e-03

Pos logit contributions
Fine+0.067
ipt+0.061
el+0.056
 Cro+0.054
 Din+0.053
Neg logit contributions
 backstage-0.121
 projector-0.119
book-0.117
dream-0.117
 dining-0.114
Token was
Token ID373
Feature activation-2.09e-02

Pos logit contributions
 backstage+0.011
 projector+0.010
dream+0.010
 dining+0.010
 chess+0.010
Neg logit contributions
Fine-0.008
ipt-0.007
el-0.006
am-0.006
E-0.006
Token not
Token ID407
Feature activation-7.85e-03

Pos logit contributions
Fine+0.022
 Din+0.020
 Twe+0.020
el+0.019
 Tit+0.018
Neg logit contributions
arium-0.040
 projector-0.039
 dining-0.038
uits-0.037
 menus-0.037
Token able
Token ID1498
Feature activation-8.46e-02

Pos logit contributions
Fine+0.009
el+0.008
 Din+0.008
ters+0.008
 Twe+0.008
Neg logit contributions
 projector-0.015
arium-0.014
 dining-0.014
 backstage-0.014
 tablet-0.013
Token to
Token ID284
Feature activation-1.95e-02

Pos logit contributions
Fine+0.090
am+0.090
el+0.088
 Cro+0.085
 rats+0.083
Neg logit contributions
 projector-0.167
book-0.152
pron-0.150
 dining-0.149
 backstage-0.146
Token make
Token ID787
Feature activation-2.52e-02

Pos logit contributions
Fine+0.030
ipt+0.027
el+0.026
E+0.024
 Din+0.023
Neg logit contributions
 backstage-0.031
 projector-0.031
book-0.030
dream-0.030
 online-0.030
Token it
Token ID340
Feature activation-1.35e-02

Pos logit contributions
Fine+0.033
ipt+0.031
el+0.029
 Din+0.027
 Cro+0.027
Neg logit contributions
 backstage-0.045
 projector-0.043
dream-0.043
book-0.043
 online-0.042
Token?
Token ID30
Feature activation+1.72e-03

Pos logit contributions
el+0.014
am+0.014
 flies+0.014
Fine+0.014
 Din+0.014
Neg logit contributions
 projector-0.025
book-0.024
 backstage-0.024
arium-0.024
dream-0.023
Token
Token ID198
Feature activation-5.30e-03

Pos logit contributions
 backstage+0.003
 projector+0.003
dream+0.003
book+0.003
 chess+0.003
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.002
 Cro-0.002
E-0.002
Token books
Token ID3835
Feature activation+1.94e-02

Pos logit contributions
Fine+0.015
el+0.012
 tires+0.012
ipt+0.012
 brace+0.012
Neg logit contributions
 backstage-0.042
 projector-0.041
pron-0.040
arium-0.039
dream-0.039
Token.
Token ID13
Feature activation-7.01e-03

Pos logit contributions
 backstage+0.037
 projector+0.037
 dining+0.037
 tablet+0.036
pron+0.036
Neg logit contributions
el-0.017
 cones-0.015
Fine-0.014
ipt-0.013
 Din-0.013
Token
Token ID220
Feature activation-8.60e-03

Pos logit contributions
Fine+0.011
ipt+0.010
el+0.009
 Cro+0.009
Flo+0.009
Neg logit contributions
 backstage-0.011
dream-0.011
 projector-0.011
book-0.011
 chess-0.011
Token
Token ID198
Feature activation-3.81e-03

Pos logit contributions
Fine+0.014
ipt+0.013
el+0.012
 Cro+0.011
Flo+0.011
Neg logit contributions
 backstage-0.013
dream-0.013
 projector-0.013
book-0.012
 chess-0.012
Token
Token ID198
Feature activation-6.13e-03

Pos logit contributions
Fine+0.006
ipt+0.006
E+0.005
el+0.005
 Cro+0.005
Neg logit contributions
 backstage-0.006
 projector-0.006
dream-0.006
 chess-0.006
book-0.006
TokenFar
Token ID21428
Feature activation-8.38e-02

Pos logit contributions
Fine+0.007
ipt+0.007
E+0.006
Flo+0.006
el+0.005
Neg logit contributions
 backstage-0.012
 projector-0.012
 chess-0.012
 online-0.011
book-0.011
Tokenmer
Token ID647
Feature activation+3.79e-03

Pos logit contributions
ipt+0.138
Fine+0.135
 Tit+0.124
 Din+0.121
 Cro+0.120
Neg logit contributions
book-0.109
 backstage-0.108
 online-0.105
dream-0.105
 projector-0.104
Token David
Token ID3271
Feature activation-6.57e-03

Pos logit contributions
 backstage+0.007
 chess+0.006
 online+0.006
 projector+0.006
dream+0.006
Neg logit contributions
Fine-0.006
ipt-0.005
el-0.005
 Cro-0.005
 Tit-0.005
Token smiled
Token ID13541
Feature activation+1.59e-02

Pos logit contributions
Fine+0.009
ipt+0.008
am+0.008
el+0.008
 Cro+0.007
Neg logit contributions
 backstage-0.012
 projector-0.011
dream-0.011
epad-0.011
book-0.011
Token.
Token ID13
Feature activation-7.12e-03

Pos logit contributions
book+0.028
 backstage+0.028
pron+0.027
 projector+0.027
arium+0.027
Neg logit contributions
ipt-0.020
el-0.019
am-0.019
Fine-0.018
 So-0.018
Token He
Token ID679
Feature activation-1.84e-02

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
 Cro+0.006
 Din+0.006
Neg logit contributions
 backstage-0.015
 projector-0.014
dream-0.014
book-0.014
 chess-0.014
Tokenloo
Token ID29680
Feature activation-3.59e-04

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
 Cro+0.001
Flo+0.001
Neg logit contributions
 backstage-0.001
dream-0.001
 projector-0.001
book-0.001
 chess-0.001
Token and
Token ID290
Feature activation-9.34e-03

Pos logit contributions
ipt+0.000
Fine+0.000
am+0.000
el+0.000
 chip+0.000
Neg logit contributions
 backstage-0.001
 projector-0.001
pron-0.001
 upon-0.001
book-0.001
Token pretended
Token ID36766
Feature activation-2.30e-03

Pos logit contributions
 Cro+0.011
 Tit+0.010
 Twe+0.010
Fine+0.010
am+0.010
Neg logit contributions
 backstage-0.016
pron-0.016
book-0.015
 menu-0.015
arium-0.015
Token to
Token ID284
Feature activation+1.27e-02

Pos logit contributions
 Cro+0.002
am+0.002
 Twe+0.002
ipt+0.002
 flies+0.002
Neg logit contributions
pron-0.004
 backstage-0.004
 dining-0.004
book-0.004
 projector-0.004
Token be
Token ID307
Feature activation+9.64e-03

Pos logit contributions
 backstage+0.024
dream+0.023
 projector+0.023
book+0.023
 chess+0.023
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.013
 Cro-0.012
Flo-0.012
Token Esk
Token ID47534
Feature activation-8.35e-02

Pos logit contributions
 backstage+0.018
book+0.017
pron+0.017
dream+0.017
arium+0.017
Neg logit contributions
 Cro-0.011
ipt-0.010
 Twe-0.010
 Din-0.010
 Tit-0.010
Tokenim
Token ID320
Feature activation-2.20e-02

Pos logit contributions
Fine+0.123
ipt+0.118
el+0.104
 Cro+0.100
Flo+0.099
Neg logit contributions
 backstage-0.138
dream-0.132
 projector-0.132
book-0.131
 chess-0.130
Tokenos
Token ID418
Feature activation-7.20e-03

Pos logit contributions
Fine+0.032
ipt+0.030
el+0.026
E+0.025
 Tit+0.025
Neg logit contributions
 backstage-0.037
dream-0.036
book-0.036
 projector-0.036
 chess-0.035
Token.
Token ID13
Feature activation-3.47e-03

Pos logit contributions
Fine+0.010
ipt+0.010
el+0.009
 Cro+0.009
Flo+0.009
Neg logit contributions
 backstage-0.012
dream-0.011
 projector-0.011
book-0.011
 dining-0.011
Token After
Token ID2293
Feature activation-1.15e-02

Pos logit contributions
ipt+0.003
Fine+0.003
 So+0.003
 Din+0.003
 Cro+0.003
Neg logit contributions
 backstage-0.008
epad-0.007
 dining-0.007
pron-0.007
 online-0.007
Token a
Token ID257
Feature activation-1.96e-02

Pos logit contributions
 Cro+0.013
 Din+0.012
Fine+0.012
 Twe+0.012
 Tit+0.012
Neg logit contributions
 backstage-0.021
book-0.021
epad-0.021
pron-0.020
 menu-0.020
Token as
Token ID355
Feature activation-3.01e-02

Pos logit contributions
Fine+0.004
 Cro+0.004
 Twe+0.004
 Din+0.004
ipt+0.003
Neg logit contributions
 backstage-0.007
 projector-0.007
book-0.007
arium-0.007
 dining-0.007
Token now
Token ID783
Feature activation-1.46e-02

Pos logit contributions
Fine+0.039
ipt+0.037
 Cro+0.035
 Tit+0.034
 Din+0.034
Neg logit contributions
 backstage-0.054
 projector-0.051
book-0.050
 dining-0.050
dream-0.049
Token he
Token ID339
Feature activation-3.85e-03

Pos logit contributions
Fine+0.017
ipt+0.015
el+0.014
 Din+0.014
 Cro+0.014
Neg logit contributions
 backstage-0.029
 projector-0.027
book-0.027
 dining-0.027
 upon-0.027
Token wasn
Token ID2492
Feature activation-6.20e-03

Pos logit contributions
Fine+0.006
ipt+0.005
el+0.005
am+0.005
 Cro+0.005
Neg logit contributions
 backstage-0.007
 projector-0.006
dream-0.006
book-0.006
 online-0.006
Token't
Token ID470
Feature activation-9.94e-03

Pos logit contributions
Fine+0.010
ipt+0.010
el+0.008
am+0.008
 Tit+0.008
Neg logit contributions
 backstage-0.009
dream-0.009
 upon-0.009
book-0.009
 projector-0.009
Token able
Token ID1498
Feature activation-8.34e-02

Pos logit contributions
Fine+0.015
ipt+0.014
 Cro+0.013
 Twe+0.013
 Din+0.013
Neg logit contributions
 projector-0.016
 backstage-0.015
arium-0.015
book-0.014
 tablet-0.014
Token to
Token ID284
Feature activation-1.78e-02

Pos logit contributions
Fine+0.127
ipt+0.124
am+0.124
 Cro+0.120
el+0.119
Neg logit contributions
 projector-0.124
 backstage-0.120
book-0.119
pron-0.111
arium-0.111
Token go
Token ID467
Feature activation-2.62e-02

Pos logit contributions
Fine+0.029
ipt+0.028
el+0.025
 Cro+0.024
Flo+0.024
Neg logit contributions
 backstage-0.027
 projector-0.026
dream-0.025
book-0.025
 chess-0.025
Token back
Token ID736
Feature activation-9.52e-03

Pos logit contributions
el+0.034
am+0.033
 Twe+0.032
 Cro+0.032
ipt+0.032
Neg logit contributions
pron-0.046
 projector-0.045
book-0.045
epad-0.042
 backstage-0.042
Token to
Token ID284
Feature activation+1.11e-03

Pos logit contributions
ipt+0.013
am+0.013
el+0.013
 Cro+0.012
Fine+0.012
Neg logit contributions
pron-0.017
 projector-0.016
book-0.015
dream-0.015
epad-0.015
Token the
Token ID262
Feature activation-1.40e-03

Pos logit contributions
book+0.002
pron+0.002
 projector+0.002
 backstage+0.002
dream+0.002
Neg logit contributions
Fine-0.001
 Cro-0.001
 Twe-0.001
el-0.001
ipt-0.001
Token say
Token ID910
Feature activation-4.26e-03

Pos logit contributions
Fine+0.000
el+0.000
 flies+0.000
am+0.000
 acts+0.000
Neg logit contributions
 projector-0.000
book-0.000
 dining-0.000
 backstage-0.000
pron-0.000
Token and
Token ID290
Feature activation+1.39e-03

Pos logit contributions
Fine+0.004
ters+0.003
 birds+0.003
 you+0.003
 winds+0.003
Neg logit contributions
pron-0.009
 dining-0.009
arium-0.008
 projector-0.008
 backstage-0.008
Token not
Token ID407
Feature activation+9.79e-03

Pos logit contributions
book+0.003
 projector+0.003
 backstage+0.003
dream+0.003
uits+0.003
Neg logit contributions
Fine-0.002
 Cro-0.001
 neither-0.001
am-0.001
 you-0.001
Token be
Token ID307
Feature activation-2.83e-02

Pos logit contributions
book+0.016
 projector+0.016
 backstage+0.015
pron+0.015
 dining+0.015
Neg logit contributions
Fine-0.015
 Cro-0.013
 flies-0.013
 Tit-0.013
am-0.013
Token too
Token ID1165
Feature activation-1.51e-02

Pos logit contributions
Fine+0.031
 Cro+0.025
 Twe+0.025
 rats+0.024
am+0.024
Neg logit contributions
 dining-0.056
 backstage-0.056
 projector-0.056
pron-0.055
book-0.055
Token quick
Token ID2068
Feature activation-8.23e-02

Pos logit contributions
Fine+0.016
 Tit+0.014
 rats+0.014
 Cro+0.013
el+0.013
Neg logit contributions
 projector-0.030
 backstage-0.029
 dining-0.029
book-0.029
arium-0.028
Token to
Token ID284
Feature activation-1.56e-02

Pos logit contributions
Fine+0.085
am+0.074
 winds+0.074
el+0.072
ipt+0.069
Neg logit contributions
pron-0.168
 backstage-0.164
 projector-0.162
 dining-0.160
book-0.159
Token ignore
Token ID8856
Feature activation-6.18e-03

Pos logit contributions
Fine+0.022
ipt+0.019
el+0.018
 Cro+0.017
Flo+0.017
Neg logit contributions
 projector-0.027
 backstage-0.027
book-0.027
 chess-0.026
 dining-0.026
Token their
Token ID511
Feature activation-2.00e-02

Pos logit contributions
Fine+0.006
el+0.006
ipt+0.005
 Twe+0.005
 lam+0.005
Neg logit contributions
book-0.013
pron-0.013
 backstage-0.013
 projector-0.013
dream-0.012
Token opinion
Token ID4459
Feature activation+3.02e-03

Pos logit contributions
Fine+0.020
ipt+0.017
el+0.017
am+0.016
 Cro+0.015
Neg logit contributions
book-0.043
 projector-0.042
 backstage-0.041
pron-0.041
dream-0.041
Token.
Token ID13
Feature activation-8.14e-03

Pos logit contributions
 dining+0.006
 projector+0.006
 backstage+0.006
pron+0.006
dream+0.006
Neg logit contributions
Fine-0.003
el-0.002
am-0.002
ters-0.002
 tires-0.002
Token he
Token ID339
Feature activation-9.30e-03

Pos logit contributions
Fine+0.021
ipt+0.020
el+0.019
 Cro+0.019
 Din+0.018
Neg logit contributions
 backstage-0.032
dream-0.029
pron-0.029
 projector-0.029
 dining-0.029
Token never
Token ID1239
Feature activation-9.38e-04

Pos logit contributions
Fine+0.014
ipt+0.013
am+0.012
 Cro+0.012
el+0.012
Neg logit contributions
 backstage-0.015
 projector-0.014
 dining-0.014
book-0.014
dream-0.014
Token would
Token ID561
Feature activation-2.61e-02

Pos logit contributions
Fine+0.002
ipt+0.001
am+0.001
 Cro+0.001
el+0.001
Neg logit contributions
 backstage-0.001
 projector-0.001
book-0.001
 dining-0.001
dream-0.001
Token have
Token ID423
Feature activation-3.34e-02

Pos logit contributions
Fine+0.026
 Cro+0.024
 Twe+0.022
 Din+0.022
el+0.022
Neg logit contributions
 projector-0.055
 backstage-0.053
book-0.052
 dining-0.052
pron-0.051
Token been
Token ID587
Feature activation-1.48e-02

Pos logit contributions
Fine+0.040
 Cro+0.038
am+0.037
ipt+0.036
 Twe+0.036
Neg logit contributions
 projector-0.060
 dining-0.059
 backstage-0.058
pron-0.058
 tablet-0.056
Token able
Token ID1498
Feature activation-8.16e-02

Pos logit contributions
 Cro+0.019
Fine+0.019
am+0.019
 Twe+0.018
 Din+0.018
Neg logit contributions
 projector-0.025
 dining-0.024
 tablet-0.023
 backstage-0.023
 telescope-0.022
Token to
Token ID284
Feature activation-2.53e-02

Pos logit contributions
 Cro+0.091
 Twe+0.087
am+0.086
ipt+0.083
el+0.082
Neg logit contributions
 projector-0.152
book-0.144
 backstage-0.141
pron-0.141
 dining-0.139
Token do
Token ID466
Feature activation-3.40e-02

Pos logit contributions
Fine+0.036
ipt+0.033
el+0.031
 Cro+0.030
 Din+0.029
Neg logit contributions
 backstage-0.043
 projector-0.042
book-0.041
dream-0.041
 dining-0.040
Token without
Token ID1231
Feature activation-2.48e-02

Pos logit contributions
Fine+0.034
am+0.031
el+0.028
 Twe+0.027
 ounces+0.027
Neg logit contributions
 projector-0.067
pron-0.065
 dining-0.064
 backstage-0.063
 telescope-0.062
Token his
Token ID465
Feature activation-3.09e-02

Pos logit contributions
Fine+0.035
ipt+0.032
el+0.031
 Cro+0.030
 Din+0.030
Neg logit contributions
 backstage-0.041
 projector-0.041
book-0.041
arium-0.039
pron-0.039
Token own
Token ID898
Feature activation+1.31e-02

Pos logit contributions
Fine+0.039
am+0.038
 tires+0.036
ipt+0.036
el+0.035
Neg logit contributions
book-0.055
 backstage-0.054
 projector-0.054
arium-0.053
 dining-0.050
Token tasty
Token ID25103
Feature activation-2.87e-02

Pos logit contributions
Flo+0.016
 chip+0.016
 Cro+0.016
E+0.016
Fine+0.016
Neg logit contributions
book-0.022
 projector-0.021
 backstage-0.021
 dining-0.021
 menus-0.020
Token snack
Token ID26906
Feature activation-1.87e-02

Pos logit contributions
Fine+0.047
ipt+0.045
el+0.043
 Cro+0.042
 chip+0.042
Neg logit contributions
 backstage-0.041
book-0.038
 projector-0.038
dream-0.037
epad-0.037
Token.
Token ID13
Feature activation-2.45e-03

Pos logit contributions
ipt+0.022
Fine+0.021
el+0.019
 Din+0.019
 chip+0.018
Neg logit contributions
 backstage-0.035
 projector-0.033
dream-0.032
 upon-0.032
 dining-0.032
Token
Token ID198
Feature activation-2.55e-03

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.004
E+0.004
am+0.003
Neg logit contributions
 backstage-0.003
dream-0.003
 projector-0.003
book-0.003
 chess-0.003
Token
Token ID198
Feature activation-5.40e-03

Pos logit contributions
Fine+0.005
ipt+0.005
Flo+0.005
E+0.004
el+0.004
Neg logit contributions
 backstage-0.003
 projector-0.003
epad-0.003
 chess-0.003
book-0.003
TokenFar
Token ID21428
Feature activation-8.15e-02

Pos logit contributions
Fine+0.007
ipt+0.007
Flo+0.006
el+0.006
E+0.006
Neg logit contributions
 backstage-0.010
 projector-0.010
book-0.010
 chess-0.010
dream-0.010
Tokenah
Token ID993
Feature activation-6.04e-03

Pos logit contributions
ipt+0.139
Fine+0.136
 ow+0.129
 Din+0.127
 Tit+0.122
Neg logit contributions
 backstage-0.097
 upon-0.094
book-0.091
arium-0.090
 projector-0.088
Token and
Token ID290
Feature activation-2.59e-03

Pos logit contributions
Fine+0.010
ipt+0.009
el+0.008
 Din+0.008
E+0.008
Neg logit contributions
 backstage-0.009
dream-0.009
book-0.009
 upon-0.009
 projector-0.009
Token the
Token ID262
Feature activation-8.95e-03

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.003
E+0.003
am+0.003
Neg logit contributions
 backstage-0.004
 chess-0.004
dream-0.004
 upon-0.004
book-0.004
Token bunny
Token ID44915
Feature activation-2.29e-02

Pos logit contributions
Fine+0.010
ipt+0.009
el+0.008
 Cro+0.007
Flo+0.007
Neg logit contributions
 backstage-0.019
dream-0.018
 projector-0.018
book-0.018
 chess-0.018
Token were
Token ID547
Feature activation-2.88e-02

Pos logit contributions
Fine+0.031
ipt+0.030
am+0.028
el+0.027
 Cro+0.027
Neg logit contributions
 backstage-0.041
book-0.039
epad-0.039
 projector-0.039
 chess-0.038
Token wall
Token ID3355
Feature activation+8.06e-03

Pos logit contributions
dream+0.000
arium+0.000
uits+0.000
 upon+0.000
ores+0.000
Neg logit contributions
el-0.000
 sounds-0.000
Fine-0.000
 chip-0.000
 boy-0.000
Token.
Token ID13
Feature activation+5.62e-03

Pos logit contributions
dream+0.015
 backstage+0.015
 projector+0.014
pron+0.014
 dining+0.014
Neg logit contributions
el-0.008
ipt-0.008
am-0.008
Fine-0.008
 tires-0.007
Token
Token ID220
Feature activation-6.08e-03

Pos logit contributions
dream+0.007
 upon+0.007
 backstage+0.007
 projector+0.007
 online+0.007
Neg logit contributions
Fine-0.010
ipt-0.009
el-0.008
am-0.008
mm-0.008
Token
Token ID198
Feature activation-8.49e-04

Pos logit contributions
Fine+0.011
ipt+0.010
el+0.009
 Cro+0.009
 Tit+0.009
Neg logit contributions
 backstage-0.008
dream-0.008
 upon-0.008
book-0.008
 chess-0.008
Token
Token ID198
Feature activation-3.55e-03

Pos logit contributions
Fine+0.002
ipt+0.001
el+0.001
am+0.001
 Din+0.001
Neg logit contributions
 backstage-0.001
dream-0.001
 upon-0.001
book-0.001
 projector-0.001
TokenFar
Token ID21428
Feature activation-8.15e-02

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
am+0.004
 Cro+0.004
Neg logit contributions
 backstage-0.006
dream-0.006
 projector-0.006
 upon-0.006
book-0.006
Tokenley
Token ID1636
Feature activation-8.60e-03

Pos logit contributions
ipt+0.136
Fine+0.130
 ow+0.125
 Tit+0.123
am+0.122
Neg logit contributions
 backstage-0.100
 online-0.094
book-0.094
 upon-0.093
dream-0.092
Token watched
Token ID7342
Feature activation+4.43e-03

Pos logit contributions
Fine+0.011
ipt+0.011
am+0.010
el+0.009
 Cro+0.009
Neg logit contributions
 backstage-0.016
dream-0.016
epad-0.015
book-0.015
 projector-0.015
Token as
Token ID355
Feature activation-9.92e-03

Pos logit contributions
arium+0.008
 backstage+0.008
 dining+0.008
book+0.008
 projector+0.008
Neg logit contributions
el-0.005
ipt-0.005
Fine-0.005
 Din-0.005
 Cro-0.005
Token his
Token ID465
Feature activation-2.81e-02

Pos logit contributions
Fine+0.015
ipt+0.015
el+0.013
am+0.013
Flo+0.012
Neg logit contributions
 backstage-0.016
dream-0.015
 chess-0.015
 projector-0.015
book-0.015
Token mom
Token ID1995
Feature activation-3.36e-03

Pos logit contributions
ipt+0.029
Fine+0.029
el+0.025
am+0.024
 Cro+0.023
Neg logit contributions
 backstage-0.056
dream-0.055
book-0.055
arium-0.054
epad-0.053
Token scared
Token ID12008
Feature activation-1.15e-02

Pos logit contributions
Fine+0.034
ipt+0.031
el+0.028
 Cro+0.027
E+0.027
Neg logit contributions
 backstage-0.036
book-0.035
 projector-0.035
 chess-0.035
dream-0.034
Token.
Token ID13
Feature activation-6.02e-03

Pos logit contributions
Fine+0.019
ipt+0.018
el+0.017
 Cro+0.016
 Din+0.016
Neg logit contributions
 backstage-0.017
 projector-0.016
book-0.016
dream-0.016
 chess-0.016
Token
Token ID220
Feature activation-1.08e-02

Pos logit contributions
Fine+0.010
ipt+0.009
el+0.008
E+0.008
am+0.008
Neg logit contributions
dream-0.009
 upon-0.009
 backstage-0.008
 projector-0.008
book-0.008
Token
Token ID198
Feature activation-2.69e-03

Pos logit contributions
Fine+0.017
ipt+0.017
el+0.015
 Cro+0.014
Flo+0.014
Neg logit contributions
 backstage-0.017
 projector-0.016
dream-0.016
book-0.016
 chess-0.016
Token
Token ID198
Feature activation-4.91e-03

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
am+0.004
 Cro+0.004
Neg logit contributions
dream-0.004
 backstage-0.004
book-0.003
 upon-0.003
 projector-0.003
TokenFar
Token ID21428
Feature activation-8.08e-02

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
am+0.006
 Cro+0.006
Neg logit contributions
dream-0.008
 backstage-0.008
book-0.008
 upon-0.008
 projector-0.008
Tokenmer
Token ID647
Feature activation+3.98e-03

Pos logit contributions
ipt+0.140
Fine+0.133
 ow+0.125
 Din+0.124
 Tit+0.122
Neg logit contributions
book-0.093
 backstage-0.092
 upon-0.091
 online-0.091
arium-0.090
Token Brown
Token ID4373
Feature activation-8.33e-03

Pos logit contributions
 backstage+0.007
dream+0.006
 projector+0.006
book+0.006
 chess+0.006
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
 Cro-0.005
Flo-0.005
Token smiled
Token ID13541
Feature activation+9.37e-03

Pos logit contributions
Fine+0.012
ipt+0.012
el+0.011
am+0.010
 Cro+0.010
Neg logit contributions
 backstage-0.014
dream-0.013
 projector-0.013
book-0.013
 chess-0.013
Token and
Token ID290
Feature activation+2.02e-03

Pos logit contributions
book+0.016
 backstage+0.016
 projector+0.016
pron+0.016
epad+0.015
Neg logit contributions
ipt-0.012
Fine-0.012
el-0.012
am-0.012
 Cro-0.011
Token said
Token ID531
Feature activation+4.52e-03

Pos logit contributions
 backstage+0.003
dream+0.003
 projector+0.003
 online+0.003
book+0.003
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.003
E-0.003
Flo-0.002
Token cool
Token ID3608
Feature activation-2.32e-02

Pos logit contributions
Fine+0.035
ipt+0.033
el+0.030
 Cro+0.029
 Din+0.029
Neg logit contributions
 backstage-0.034
 projector-0.032
dream-0.032
book-0.032
 chess-0.031
Token.
Token ID13
Feature activation-1.22e-03

Pos logit contributions
Fine+0.033
ipt+0.031
el+0.029
 Din+0.028
am+0.028
Neg logit contributions
 backstage-0.038
 projector-0.038
book-0.037
dream-0.037
 dining-0.036
Token
Token ID220
Feature activation-4.30e-03

Pos logit contributions
Fine+0.002
ipt+0.002
el+0.002
am+0.002
E+0.002
Neg logit contributions
dream-0.002
 backstage-0.002
 upon-0.002
 projector-0.002
 chess-0.002
Token
Token ID198
Feature activation+8.39e-04

Pos logit contributions
Fine+0.008
ipt+0.007
el+0.007
 Cro+0.006
 Tit+0.006
Neg logit contributions
 backstage-0.006
dream-0.005
 upon-0.005
book-0.005
 chess-0.005
Token
Token ID198
Feature activation-2.35e-03

Pos logit contributions
 backstage+0.001
dream+0.001
 projector+0.001
book+0.001
 chess+0.001
Neg logit contributions
Fine-0.001
ipt-0.001
el-0.001
 Cro-0.001
Flo-0.001
TokenFar
Token ID21428
Feature activation-8.05e-02

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
 Cro+0.003
am+0.003
Neg logit contributions
 backstage-0.004
dream-0.004
 projector-0.004
book-0.004
 chess-0.004
Tokenley
Token ID1636
Feature activation-6.09e-03

Pos logit contributions
ipt+0.135
Fine+0.128
 ow+0.124
 Tit+0.123
am+0.122
Neg logit contributions
 backstage-0.096
 online-0.092
book-0.091
dream-0.090
 projector-0.089
Token was
Token ID373
Feature activation-1.13e-03

Pos logit contributions
Fine+0.008
ipt+0.008
am+0.007
el+0.007
 Cro+0.006
Neg logit contributions
 backstage-0.011
dream-0.011
book-0.011
 online-0.011
epad-0.011
Token so
Token ID523
Feature activation-5.60e-03

Pos logit contributions
ipt+0.001
 Twe+0.001
 Cro+0.001
 Tit+0.001
Fine+0.001
Neg logit contributions
arium-0.002
 menu-0.002
 backstage-0.002
 tablet-0.002
 menus-0.002
Token happy
Token ID3772
Feature activation-3.96e-03

Pos logit contributions
Fine+0.009
ipt+0.009
el+0.008
 Cro+0.008
 Tit+0.008
Neg logit contributions
 projector-0.008
 backstage-0.008
arium-0.008
 chess-0.008
 tablet-0.008
Token to
Token ID284
Feature activation+1.64e-02

Pos logit contributions
Fine+0.005
ipt+0.005
 Cro+0.004
el+0.004
 Twe+0.004
Neg logit contributions
 projector-0.007
book-0.007
 backstage-0.007
pron-0.007
 dining-0.007
Token of
Token ID286
Feature activation-2.46e-02

Pos logit contributions
pron+0.029
 backstage+0.029
dream+0.028
 dining+0.028
epad+0.028
Neg logit contributions
el-0.010
 branches-0.010
 cones-0.010
Fine-0.009
am-0.009
Token cheese
Token ID9891
Feature activation-1.96e-02

Pos logit contributions
 tires+0.021
 branches+0.020
Fine+0.020
 flies+0.019
 nuts+0.019
Neg logit contributions
 backstage-0.054
pron-0.052
arium-0.051
ifying-0.050
side-0.049
Token and
Token ID290
Feature activation-2.51e-02

Pos logit contributions
 cones+0.019
 chip+0.018
 rats+0.018
am+0.018
ters+0.018
Neg logit contributions
dream-0.039
epad-0.038
 backstage-0.037
pron-0.036
 projector-0.036
Token sauce
Token ID10746
Feature activation-1.54e-02

Pos logit contributions
Fine+0.030
ipt+0.027
 Cro+0.026
el+0.025
 Tit+0.025
Neg logit contributions
 backstage-0.048
dream-0.047
book-0.046
 projector-0.045
 upon-0.044
Token and
Token ID290
Feature activation-2.50e-02

Pos logit contributions
am+0.013
Fine+0.013
 cones+0.013
ipt+0.013
el+0.012
Neg logit contributions
dream-0.031
epad-0.030
 telescope-0.029
 backstage-0.029
 projector-0.029
Token topp
Token ID23126
Feature activation-8.04e-02

Pos logit contributions
Fine+0.028
 Cro+0.024
 Tit+0.024
el+0.024
ipt+0.024
Neg logit contributions
 backstage-0.051
dream-0.049
book-0.047
 projector-0.047
arium-0.046
Tokenings
Token ID654
Feature activation-2.26e-02

Pos logit contributions
Fine+0.142
ipt+0.125
 Din+0.118
 Cro+0.114
 Twe+0.113
Neg logit contributions
book-0.116
 menu-0.112
 upon-0.110
 projector-0.108
 backstage-0.107
Token.
Token ID13
Feature activation+4.99e-03

Pos logit contributions
Fine+0.024
el+0.022
Flo+0.021
 Cro+0.021
ipt+0.021
Neg logit contributions
 backstage-0.044
pron-0.044
arium-0.043
 telescope-0.043
dream-0.043
Token The
Token ID383
Feature activation-6.57e-03

Pos logit contributions
 projector+0.008
 backstage+0.008
 chess+0.008
book+0.008
 online+0.008
Neg logit contributions
Fine-0.007
ipt-0.007
el-0.006
am-0.006
Flo-0.006
Token people
Token ID661
Feature activation+9.48e-03

Pos logit contributions
 Cro+0.007
 chip+0.007
 cones+0.006
 rats+0.006
 bird+0.006
Neg logit contributions
 backstage-0.013
dream-0.013
book-0.012
arium-0.012
 indoors-0.012
Token who
Token ID508
Feature activation+1.61e-02

Pos logit contributions
 chess+0.018
 dining+0.018
 backstage+0.018
dream+0.018
 projector+0.018
Neg logit contributions
el-0.010
Fine-0.009
 flies-0.009
am-0.008
 k-0.008
Token carrot
Token ID40562
Feature activation-3.01e-02

Pos logit contributions
Fine+0.021
ipt+0.021
el+0.019
 chip+0.018
Flo+0.018
Neg logit contributions
 backstage-0.020
dream-0.019
 projector-0.019
epad-0.019
book-0.018
Token each
Token ID1123
Feature activation-4.26e-02

Pos logit contributions
ipt+0.040
am+0.039
 chip+0.038
 Cro+0.038
el+0.038
Neg logit contributions
dream-0.053
 backstage-0.050
 projector-0.050
epad-0.050
book-0.048
Token?"
Token ID1701
Feature activation-1.99e-03

Pos logit contributions
Fine+0.054
el+0.049
ipt+0.049
am+0.047
 Cro+0.047
Neg logit contributions
 backstage-0.081
 projector-0.078
epad-0.077
dream-0.076
 dining-0.075
Token
Token ID198
Feature activation-2.55e-03

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.003
E+0.003
am+0.003
Neg logit contributions
dream-0.002
 backstage-0.002
 upon-0.002
 online-0.002
 chess-0.002
Token
Token ID198
Feature activation-6.04e-03

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
am+0.004
 brace+0.004
Neg logit contributions
dream-0.003
 backstage-0.003
book-0.003
 upon-0.003
 projector-0.003
TokenFar
Token ID21428
Feature activation-7.93e-02

Pos logit contributions
Fine+0.009
ipt+0.009
el+0.008
am+0.007
 brace+0.007
Neg logit contributions
 backstage-0.009
dream-0.009
 projector-0.009
book-0.009
 upon-0.009
Tokenmer
Token ID647
Feature activation+9.43e-03

Pos logit contributions
Fine+0.134
ipt+0.133
 Din+0.118
E+0.117
el+0.116
Neg logit contributions
 backstage-0.103
book-0.099
 online-0.098
 projector-0.098
arium-0.097
Token Joe
Token ID5689
Feature activation-9.10e-03

Pos logit contributions
 backstage+0.014
dream+0.013
 projector+0.013
 online+0.013
 dining+0.013
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.015
 Cro-0.014
am-0.014
Token looks
Token ID3073
Feature activation+6.57e-03

Pos logit contributions
Fine+0.014
ipt+0.012
el+0.011
am+0.011
 Cro+0.011
Neg logit contributions
 backstage-0.016
 projector-0.015
book-0.015
dream-0.015
epad-0.015
Token at
Token ID379
Feature activation+4.05e-03

Pos logit contributions
epad+0.014
book+0.013
pron+0.013
 dining+0.013
 backstage+0.013
Neg logit contributions
am-0.007
 So-0.007
Fine-0.006
 runs-0.006
 flies-0.006
Token them
Token ID606
Feature activation+7.28e-03

Pos logit contributions
 backstage+0.006
 upon+0.006
 chess+0.006
 online+0.006
 projector+0.006
Neg logit contributions
Fine-0.007
ipt-0.006
Flo-0.006
am-0.005
Bl-0.005
Token it
Token ID340
Feature activation-1.86e-02

Pos logit contributions
ipt+0.025
Fine+0.025
el+0.024
Flo+0.024
 Tit+0.024
Neg logit contributions
 backstage-0.036
 projector-0.034
dream-0.034
epad-0.034
pron-0.033
Token with
Token ID351
Feature activation-1.66e-02

Pos logit contributions
Fine+0.027
ipt+0.026
el+0.025
Flo+0.025
am+0.025
Neg logit contributions
 backstage-0.029
 projector-0.028
book-0.027
epad-0.027
dream-0.027
Token their
Token ID511
Feature activation-3.98e-02

Pos logit contributions
Fine+0.025
ipt+0.024
el+0.021
 Cro+0.021
 Tit+0.020
Neg logit contributions
 backstage-0.027
dream-0.026
 projector-0.026
book-0.026
 chess-0.025
Token many
Token ID867
Feature activation-2.19e-02

Pos logit contributions
Fine+0.049
ipt+0.047
el+0.044
 Cro+0.042
am+0.041
Neg logit contributions
 backstage-0.073
book-0.073
 projector-0.072
dream-0.072
 chess-0.070
Token feat
Token ID2218
Feature activation-6.54e-02

Pos logit contributions
Fine+0.027
ipt+0.026
el+0.023
am+0.021
Flo+0.021
Neg logit contributions
 backstage-0.041
 projector-0.040
dream-0.040
book-0.040
 chess-0.039
Tokenhered
Token ID6083
Feature activation-7.91e-02

Pos logit contributions
Fine+0.108
ipt+0.105
el+0.090
E+0.090
 Tit+0.090
Neg logit contributions
 backstage-0.095
dream-0.094
book-0.092
 projector-0.091
 chess-0.091
Token friends
Token ID2460
Feature activation-9.04e-03

Pos logit contributions
Fine+0.079
ipt+0.076
el+0.064
am+0.061
Flo+0.060
Neg logit contributions
 backstage-0.168
 projector-0.163
book-0.162
dream-0.161
 chess-0.160
Token.
Token ID13
Feature activation-4.81e-03

Pos logit contributions
el+0.007
am+0.006
ipt+0.006
 k+0.006
 chip+0.006
Neg logit contributions
 projector-0.019
 backstage-0.019
pron-0.019
epad-0.019
book-0.018
Token
Token ID198
Feature activation-4.06e-03

Pos logit contributions
Fine+0.008
ipt+0.007
el+0.007
 Cro+0.006
Flo+0.006
Neg logit contributions
 backstage-0.007
dream-0.007
 projector-0.007
book-0.007
 chess-0.007
Token
Token ID198
Feature activation-6.70e-03

Pos logit contributions
Fine+0.006
ipt+0.006
Flo+0.005
E+0.005
 Tit+0.005
Neg logit contributions
 backstage-0.007
epad-0.006
 projector-0.006
 chess-0.006
dream-0.006
TokenBob
Token ID18861
Feature activation-7.45e-03

Pos logit contributions
Fine+0.007
ipt+0.006
Flo+0.005
E+0.005
 Tit+0.005
Neg logit contributions
 backstage-0.015
 chess-0.014
 projector-0.014
epad-0.014
dream-0.014
Token could
Token ID714
Feature activation-2.09e-02

Pos logit contributions
 backstage+0.006
dream+0.005
book+0.005
epad+0.005
 dining+0.005
Neg logit contributions
Fine-0.004
ipt-0.004
 Cro-0.004
am-0.003
el-0.003
Token help
Token ID1037
Feature activation-4.86e-02

Pos logit contributions
Fine+0.027
ipt+0.026
 Cro+0.024
el+0.023
Flo+0.023
Neg logit contributions
 backstage-0.038
book-0.037
dream-0.036
 projector-0.036
 chess-0.035
Token get
Token ID651
Feature activation-1.70e-02

Pos logit contributions
Fine+0.083
ipt+0.080
el+0.066
am+0.066
 Tit+0.065
Neg logit contributions
 backstage-0.071
dream-0.069
 online-0.067
book-0.067
 chess-0.066
Token the
Token ID262
Feature activation-2.47e-03

Pos logit contributions
Fine+0.028
ipt+0.027
el+0.025
 Cro+0.024
Flo+0.024
Neg logit contributions
 backstage-0.025
dream-0.024
 projector-0.023
book-0.023
 chess-0.023
Token car
Token ID1097
Feature activation-4.23e-03

Pos logit contributions
Fine+0.003
ipt+0.002
 Cro+0.002
 tires+0.002
 chip+0.002
Neg logit contributions
 backstage-0.005
book-0.005
dream-0.005
arium-0.005
 projector-0.005
Token unst
Token ID15014
Feature activation-7.88e-02

Pos logit contributions
ipt+0.007
Fine+0.007
el+0.006
am+0.006
 Cro+0.006
Neg logit contributions
 backstage-0.006
dream-0.006
 projector-0.006
book-0.006
 dining-0.006
Tokenuck
Token ID1347
Feature activation-4.85e-02

Pos logit contributions
 ow+0.122
 Mir+0.121
Fine+0.112
 brace+0.110
 Lia+0.109
Neg logit contributions
 online-0.100
 forward-0.099
 indoors-0.096
 downstairs-0.096
 inside-0.096
Token.
Token ID13
Feature activation-5.52e-03

Pos logit contributions
ipt+0.068
el+0.067
Fine+0.066
am+0.063
Flo+0.061
Neg logit contributions
 backstage-0.079
dream-0.077
 projector-0.077
pron-0.076
 chess-0.075
Token
Token ID198
Feature activation-2.26e-03

Pos logit contributions
Fine+0.009
ipt+0.008
el+0.007
am+0.007
 Tit+0.007
Neg logit contributions
 backstage-0.008
dream-0.008
 projector-0.008
book-0.008
 chess-0.008
Token
Token ID198
Feature activation-4.89e-03

Pos logit contributions
Fine+0.003
ipt+0.003
E+0.003
Flo+0.003
el+0.003
Neg logit contributions
 backstage-0.004
 projector-0.004
 chess-0.004
book-0.004
dream-0.004
TokenThe
Token ID464
Feature activation-1.15e-02

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
 Cro+0.004
am+0.004
Neg logit contributions
 backstage-0.010
dream-0.010
 projector-0.009
book-0.009
 chess-0.009
Token and
Token ID290
Feature activation-1.91e-02

Pos logit contributions
 projector+0.014
 backstage+0.014
 dining+0.013
dream+0.013
book+0.013
Neg logit contributions
Fine-0.013
ipt-0.012
am-0.011
el-0.011
 Din-0.011
Token relax
Token ID8960
Feature activation-3.04e-02

Pos logit contributions
Fine+0.022
 Tit+0.021
 Cro+0.020
 Din+0.020
 Twe+0.020
Neg logit contributions
 backstage-0.036
book-0.035
dream-0.035
 projector-0.034
arium-0.033
Token.
Token ID13
Feature activation+2.48e-03

Pos logit contributions
el+0.031
 Twe+0.030
Fine+0.029
Flo+0.028
ipt+0.028
Neg logit contributions
 backstage-0.057
 projector-0.057
epad-0.055
arium-0.054
 telescope-0.053
Token
Token ID198
Feature activation-3.37e-03

Pos logit contributions
 backstage+0.004
dream+0.004
 projector+0.004
book+0.004
 chess+0.004
Neg logit contributions
Fine-0.004
ipt-0.004
el-0.003
 Cro-0.003
Flo-0.003
Token
Token ID198
Feature activation-5.27e-03

Pos logit contributions
Fine+0.005
ipt+0.005
E+0.004
Flo+0.004
el+0.004
Neg logit contributions
 backstage-0.006
 projector-0.005
 chess-0.005
 online-0.005
dream-0.005
TokenFar
Token ID21428
Feature activation-7.86e-02

Pos logit contributions
Fine+0.007
ipt+0.006
E+0.006
Flo+0.005
Bl+0.005
Neg logit contributions
 backstage-0.011
 projector-0.010
 online-0.010
epad-0.010
 menu-0.010
Tokenmer
Token ID647
Feature activation+8.99e-03

Pos logit contributions
ipt+0.137
Fine+0.133
 Din+0.122
 Tit+0.121
 ow+0.121
Neg logit contributions
 backstage-0.096
book-0.092
arium-0.091
 upon-0.091
dream-0.090
Token Joe
Token ID5689
Feature activation-5.53e-03

Pos logit contributions
 backstage+0.015
 online+0.014
dream+0.014
 projector+0.014
book+0.014
Neg logit contributions
Fine-0.013
ipt-0.013
el-0.011
 Cro-0.011
am-0.011
Token smiled
Token ID13541
Feature activation+1.34e-02

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
am+0.006
 Cro+0.006
Neg logit contributions
 backstage-0.010
book-0.010
 projector-0.010
 chess-0.010
epad-0.010
Token and
Token ID290
Feature activation+9.19e-03

Pos logit contributions
book+0.024
pron+0.023
 projector+0.023
 backstage+0.023
arium+0.022
Neg logit contributions
ipt-0.018
Fine-0.017
am-0.016
el-0.016
 Cro-0.015
Token said
Token ID531
Feature activation+9.08e-03

Pos logit contributions
 backstage+0.017
 projector+0.016
book+0.016
dream+0.016
 chess+0.016
Neg logit contributions
Fine-0.012
ipt-0.012
el-0.010
 Cro-0.010
 Tit-0.010
Token!
Token ID0
Feature activation-9.39e-03

Pos logit contributions
ipt+0.017
el+0.017
am+0.017
 cones+0.015
 Cro+0.015
Neg logit contributions
pron-0.026
 backstage-0.026
 dining-0.026
 projector-0.024
dream-0.024
Token But
Token ID887
Feature activation-7.24e-03

Pos logit contributions
Fine+0.011
ipt+0.011
el+0.009
 Cro+0.009
 Din+0.008
Neg logit contributions
 backstage-0.019
dream-0.018
book-0.018
 projector-0.018
 chess-0.017
Token it
Token ID340
Feature activation+3.12e-03

Pos logit contributions
 Cro+0.008
Fine+0.008
ipt+0.007
 Twe+0.007
el+0.007
Neg logit contributions
 backstage-0.015
book-0.014
epad-0.014
 menu-0.014
 dining-0.014
Token wasn
Token ID2492
Feature activation-3.90e-03

Pos logit contributions
 backstage+0.006
dream+0.005
book+0.005
 projector+0.005
 chess+0.005
Neg logit contributions
Fine-0.004
ipt-0.004
el-0.004
 Cro-0.003
Flo-0.003
Token't
Token ID470
Feature activation-5.70e-03

Pos logit contributions
Fine+0.007
ipt+0.006
el+0.005
am+0.005
 Tit+0.005
Neg logit contributions
 backstage-0.006
 upon-0.006
dream-0.005
 projector-0.005
 online-0.005
Token able
Token ID1498
Feature activation-7.81e-02

Pos logit contributions
Fine+0.008
ipt+0.008
 Cro+0.007
 Tit+0.007
el+0.007
Neg logit contributions
 backstage-0.009
 projector-0.009
 dining-0.009
book-0.009
dream-0.009
Token to
Token ID284
Feature activation-1.18e-02

Pos logit contributions
Fine+0.129
ipt+0.128
 Cro+0.124
am+0.124
el+0.121
Neg logit contributions
 projector-0.106
book-0.105
 backstage-0.104
pron-0.099
 dining-0.098
Token move
Token ID1445
Feature activation-2.59e-02

Pos logit contributions
Fine+0.021
ipt+0.021
E+0.018
el+0.018
 Din+0.017
Neg logit contributions
 backstage-0.016
dream-0.015
 chess-0.015
 projector-0.015
 online-0.014
Token much
Token ID881
Feature activation+1.47e-02

Pos logit contributions
el+0.031
Fine+0.031
 Cro+0.030
ipt+0.029
 Twe+0.029
Neg logit contributions
 dining-0.048
 projector-0.047
 backstage-0.047
pron-0.047
book-0.046
Token because
Token ID780
Feature activation+2.27e-02

Pos logit contributions
 backstage+0.025
book+0.025
 dining+0.024
 projector+0.024
pron+0.024
Neg logit contributions
Fine-0.020
ipt-0.019
el-0.018
am-0.017
 Cro-0.017
Token the
Token ID262
Feature activation+3.93e-03

Pos logit contributions
 backstage+0.048
 dining+0.045
dream+0.045
book+0.045
 projector+0.045
Neg logit contributions
ipt-0.022
Fine-0.022
el-0.021
 Cro-0.020
 Din-0.019
Token to
Token ID284
Feature activation-1.90e-02

Pos logit contributions
el+0.015
 Cro+0.015
am+0.014
ipt+0.014
 Tit+0.014
Neg logit contributions
 projector-0.015
book-0.015
 backstage-0.015
pron-0.015
 dining-0.014
Token grow
Token ID1663
Feature activation-2.78e-03

Pos logit contributions
Fine+0.020
el+0.020
Flo+0.019
 Cro+0.019
 Twe+0.019
Neg logit contributions
 backstage-0.036
 projector-0.036
book-0.036
 chess-0.035
dream-0.035
Token!
Token ID0
Feature activation-1.24e-02

Pos logit contributions
Fine+0.004
ipt+0.004
el+0.004
Flo+0.004
am+0.004
Neg logit contributions
 backstage-0.004
 projector-0.004
dream-0.004
book-0.004
 chess-0.004
Token
Token ID198
Feature activation-2.62e-03

Pos logit contributions
Fine+0.016
ipt+0.016
 Cro+0.013
el+0.013
 Tit+0.013
Neg logit contributions
 backstage-0.023
 projector-0.021
dream-0.021
book-0.021
 chess-0.021
Token
Token ID198
Feature activation-4.95e-03

Pos logit contributions
Fine+0.004
ipt+0.003
E+0.003
 Tit+0.003
Flo+0.003
Neg logit contributions
 backstage-0.005
epad-0.004
 projector-0.004
 chess-0.004
dream-0.004
TokenFar
Token ID21428
Feature activation-7.79e-02

Pos logit contributions
Fine+0.006
ipt+0.005
E+0.005
Flo+0.005
 Tit+0.004
Neg logit contributions
 backstage-0.010
 chess-0.010
 projector-0.010
epad-0.009
dream-0.009
Tokenmer
Token ID647
Feature activation+7.96e-03

Pos logit contributions
ipt+0.137
Fine+0.132
 Din+0.125
 Tit+0.124
 ow+0.120
Neg logit contributions
book-0.091
 backstage-0.090
 online-0.089
arium-0.088
 projector-0.088
Token John
Token ID1757
Feature activation-4.90e-04

Pos logit contributions
 backstage+0.013
dream+0.013
 chess+0.012
 projector+0.012
 online+0.012
Neg logit contributions
Fine-0.012
ipt-0.011
el-0.010
 Cro-0.010
am-0.010
Token smiled
Token ID13541
Feature activation+1.44e-02

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
am+0.001
 Cro+0.001
Neg logit contributions
 backstage-0.001
dream-0.001
 projector-0.001
 chess-0.001
book-0.001
Token.
Token ID13
Feature activation-7.45e-03

Pos logit contributions
book+0.025
 backstage+0.025
pron+0.025
 projector+0.024
arium+0.024
Neg logit contributions
ipt-0.019
Fine-0.018
el-0.018
am-0.017
 Cro-0.016
Token He
Token ID679
Feature activation-2.04e-02

Pos logit contributions
Fine+0.007
ipt+0.007
 Cro+0.006
el+0.006
 Tit+0.005
Neg logit contributions
 backstage-0.016
 projector-0.015
dream-0.015
book-0.015
 chess-0.015
Token.
Token ID13
Feature activation-1.55e-03

Pos logit contributions
Fine+0.010
ipt+0.010
el+0.009
am+0.008
 Cro+0.008
Neg logit contributions
 backstage-0.017
 projector-0.017
pron-0.016
 dining-0.016
arium-0.016
Token Lily
Token ID20037
Feature activation+3.99e-03

Pos logit contributions
Fine+0.002
ipt+0.002
el+0.002
Flo+0.002
 Cro+0.002
Neg logit contributions
 backstage-0.003
dream-0.003
book-0.003
 projector-0.003
 chess-0.003
Token told
Token ID1297
Feature activation+1.78e-02

Pos logit contributions
 backstage+0.006
dream+0.006
 projector+0.006
book+0.006
 chess+0.006
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
am-0.005
 Cro-0.005
Token him
Token ID683
Feature activation+1.72e-02

Pos logit contributions
 backstage+0.028
dream+0.028
 chess+0.028
 online+0.027
book+0.027
Neg logit contributions
ipt-0.028
Fine-0.028
Flo-0.024
E-0.023
 Tit-0.023
Token her
Token ID607
Feature activation+1.26e-02

Pos logit contributions
 projector+0.025
 backstage+0.024
 dining+0.024
pron+0.023
book+0.023
Neg logit contributions
Fine-0.027
am-0.025
ipt-0.025
 Cro-0.024
el-0.024
Token ink
Token ID16882
Feature activation+4.57e-02

Pos logit contributions
 backstage+0.022
 dining+0.020
 projector+0.020
ifying+0.019
 cinema+0.019
Neg logit contributions
el-0.016
am-0.016
 ow-0.015
Fine-0.015
ipt-0.015
Token was
Token ID373
Feature activation-3.18e-03

Pos logit contributions
 backstage+0.087
 projector+0.083
 dining+0.083
dream+0.082
arium+0.079
Neg logit contributions
el-0.052
ipt-0.051
Fine-0.050
am-0.050
 brace-0.046
Token frozen
Token ID12912
Feature activation+4.61e-03

Pos logit contributions
el+0.003
 ow+0.003
 Twe+0.003
 Din+0.003
 brace+0.003
Neg logit contributions
 backstage-0.007
 projector-0.007
arium-0.006
 dining-0.006
 cinema-0.006
Token.
Token ID13
Feature activation-2.24e-03

Pos logit contributions
 backstage+0.008
dream+0.008
 projector+0.008
 dining+0.008
pron+0.008
Neg logit contributions
el-0.006
Fine-0.005
ipt-0.005
 Cro-0.005
 Twe-0.005
Token Tom
Token ID4186
Feature activation-2.28e-03

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.002
Flo+0.002
E+0.002
Neg logit contributions
 backstage-0.004
dream-0.004
 chess-0.004
book-0.004
 projector-0.004
Token was
Token ID373
Feature activation+6.38e-03

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
 Cro+0.003
Flo+0.003
Neg logit contributions
 backstage-0.004
dream-0.004
 projector-0.004
book-0.004
 chess-0.004
Token oranges
Token ID48389
Feature activation+3.08e-02

Pos logit contributions
 backstage+0.021
 projector+0.019
dream+0.019
book+0.019
 chess+0.019
Neg logit contributions
Fine-0.042
ipt-0.041
el-0.038
 Cro-0.037
 Tit-0.037
Token.
Token ID13
Feature activation+6.26e-03

Pos logit contributions
 backstage+0.039
 online+0.039
book+0.039
dream+0.038
 chess+0.038
Neg logit contributions
Fine-0.058
ipt-0.055
E-0.048
el-0.048
Flo-0.047
Token She
Token ID1375
Feature activation+7.21e-03

Pos logit contributions
 chess+0.010
book+0.010
 online+0.010
 backstage+0.010
 upon+0.010
Neg logit contributions
Fine-0.010
ipt-0.009
am-0.008
el-0.008
 Tit-0.007
Token was
Token ID373
Feature activation+8.77e-03

Pos logit contributions
book+0.006
 online+0.006
 upon+0.006
pron+0.006
dream+0.006
Neg logit contributions
ipt-0.016
Fine-0.016
el-0.014
 heights-0.014
am-0.014
Token a
Token ID257
Feature activation-1.39e-03

Pos logit contributions
 online+0.010
 backstage+0.010
 chess+0.010
dream+0.010
 upon+0.010
Neg logit contributions
Fine-0.018
ipt-0.017
am-0.015
el-0.015
Flo-0.014
Token very
Token ID845
Feature activation+4.34e-02

Pos logit contributions
Fine+0.002
ipt+0.002
el+0.002
am+0.002
 Cro+0.002
Neg logit contributions
 backstage-0.002
dream-0.002
 projector-0.002
 chess-0.002
book-0.002
Token cheerful
Token ID37999
Feature activation+1.07e-02

Pos logit contributions
 backstage+0.065
dream+0.062
 projector+0.062
book+0.061
 chess+0.061
Neg logit contributions
Fine-0.071
ipt-0.068
el-0.061
 Cro-0.059
Flo-0.058
Token girl
Token ID2576
Feature activation-2.30e-03

Pos logit contributions
 backstage+0.019
 projector+0.019
book+0.018
dream+0.018
 online+0.018
Neg logit contributions
Fine-0.014
ipt-0.013
el-0.012
 Cro-0.012
 Din-0.012
Token who
Token ID508
Feature activation+1.55e-02

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
 Din+0.003
 Cro+0.003
Neg logit contributions
 backstage-0.004
 projector-0.004
dream-0.004
 dining-0.004
 menu-0.004
Token always
Token ID1464
Feature activation+1.02e-02

Pos logit contributions
 backstage+0.018
book+0.017
dream+0.017
 projector+0.017
 online+0.017
Neg logit contributions
Fine-0.030
ipt-0.030
el-0.027
am-0.026
 Cro-0.026
Token smiled
Token ID13541
Feature activation+1.14e-02

Pos logit contributions
 backstage+0.016
dream+0.015
 projector+0.015
book+0.015
 chess+0.015
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.014
 Cro-0.013
Flo-0.013
Token.
Token ID13
Feature activation-5.60e-03

Pos logit contributions
Fine+0.042
am+0.039
el+0.039
ipt+0.039
 Cro+0.036
Neg logit contributions
 projector-0.068
pron-0.066
 dining-0.065
book-0.065
 backstage-0.064
Token He
Token ID679
Feature activation-1.26e-02

Pos logit contributions
Fine+0.008
ipt+0.008
el+0.007
Flo+0.007
 lam+0.007
Neg logit contributions
 chess-0.009
 upon-0.009
dream-0.009
 online-0.009
 backstage-0.009
Token was
Token ID373
Feature activation+7.11e-03

Pos logit contributions
Fine+0.017
ipt+0.016
el+0.014
am+0.014
 Cro+0.014
Neg logit contributions
 backstage-0.023
book-0.022
dream-0.022
 projector-0.022
 dining-0.021
Token so
Token ID523
Feature activation+9.67e-04

Pos logit contributions
book+0.013
 backstage+0.013
pron+0.013
 dining+0.012
arium+0.012
Neg logit contributions
Fine-0.009
 Cro-0.008
 Tit-0.008
ipt-0.008
 Twe-0.008
Token amazed
Token ID24562
Feature activation+1.66e-03

Pos logit contributions
 backstage+0.001
 projector+0.001
book+0.001
 chess+0.001
dream+0.001
Neg logit contributions
Fine-0.002
ipt-0.001
el-0.001
 Cro-0.001
am-0.001
Token by
Token ID416
Feature activation+4.16e-02

Pos logit contributions
 backstage+0.002
dream+0.002
 projector+0.002
book+0.002
 chess+0.002
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.003
 Cro-0.003
Flo-0.003
Token her
Token ID607
Feature activation+1.49e-02

Pos logit contributions
 backstage+0.067
 projector+0.064
dream+0.064
book+0.064
 online+0.062
Neg logit contributions
Fine-0.063
ipt-0.061
el-0.055
 Cro-0.053
 Tit-0.053
Token that
Token ID326
Feature activation+2.54e-02

Pos logit contributions
book+0.033
 projector+0.031
pron+0.030
 dining+0.030
 backstage+0.030
Neg logit contributions
am-0.014
el-0.013
 Cro-0.013
ipt-0.013
Fine-0.013
Token he
Token ID339
Feature activation+2.62e-02

Pos logit contributions
 backstage+0.055
 projector+0.053
uits+0.051
book+0.051
pron+0.051
Neg logit contributions
Fine-0.023
 Cro-0.021
 Twe-0.020
 Din-0.020
 Tit-0.020
Token followed
Token ID3940
Feature activation-2.05e-02

Pos logit contributions
 backstage+0.045
book+0.044
dream+0.043
 projector+0.042
 dining+0.042
Neg logit contributions
Fine-0.039
ipt-0.036
am-0.032
el-0.032
 Cro-0.031
Token her
Token ID607
Feature activation-4.57e-03

Pos logit contributions
Fine+0.020
el+0.019
ipt+0.019
 Twe+0.019
 Cro+0.019
Neg logit contributions
 dining-0.041
 projector-0.041
 backstage-0.041
book-0.041
pron-0.041
Token he
Token ID339
Feature activation+2.60e-02

Pos logit contributions
dream+0.024
 upon+0.023
 backstage+0.023
 chess+0.023
 online+0.022
Neg logit contributions
Fine-0.029
ipt-0.028
el-0.025
Flo-0.024
bit-0.024
Token went
Token ID1816
Feature activation+2.82e-03

Pos logit contributions
 backstage+0.047
 projector+0.045
dream+0.045
book+0.045
 chess+0.044
Neg logit contributions
Fine-0.035
ipt-0.033
el-0.029
 Cro-0.028
Flo-0.028
Token up
Token ID510
Feature activation-4.59e-03

Pos logit contributions
book+0.005
 projector+0.005
 dining+0.005
pron+0.005
ifying+0.005
Neg logit contributions
 Cro-0.003
 branches-0.003
 birds-0.003
ipt-0.003
 Din-0.003
Token to
Token ID284
Feature activation+8.32e-03

Pos logit contributions
ipt+0.006
el+0.006
 Cro+0.006
Fine+0.006
 Tit+0.006
Neg logit contributions
 projector-0.008
pron-0.008
 dining-0.008
book-0.007
epad-0.007
Token the
Token ID262
Feature activation+5.79e-03

Pos logit contributions
 backstage+0.016
book+0.016
 projector+0.015
dream+0.015
 dining+0.015
Neg logit contributions
Fine-0.010
ipt-0.010
el-0.009
 Cro-0.008
am-0.008
Token cob
Token ID22843
Feature activation+3.83e-02

Pos logit contributions
 backstage+0.010
dream+0.009
 projector+0.009
book+0.009
 chess+0.009
Neg logit contributions
Fine-0.009
ipt-0.008
el-0.007
am-0.007
 Cro-0.007
Tokenweb
Token ID12384
Feature activation+2.67e-02

Pos logit contributions
 backstage+0.064
 projector+0.063
dream+0.063
 chess+0.063
epad+0.062
Neg logit contributions
Fine-0.055
ipt-0.053
el-0.047
 Tit-0.044
Flo-0.044
Token.
Token ID13
Feature activation+7.30e-03

Pos logit contributions
pron+0.051
 backstage+0.050
 projector+0.050
epad+0.050
book+0.049
Neg logit contributions
el-0.026
ipt-0.026
 branches-0.023
am-0.023
 So-0.022
Token He
Token ID679
Feature activation-7.73e-03

Pos logit contributions
 backstage+0.013
 projector+0.012
book+0.012
dream+0.012
 online+0.012
Neg logit contributions
Fine-0.010
ipt-0.010
E-0.008
el-0.008
 Cro-0.008
Token saw
Token ID2497
Feature activation+2.14e-02

Pos logit contributions
Fine+0.011
ipt+0.011
el+0.009
am+0.009
 Cro+0.009
Neg logit contributions
 backstage-0.013
 projector-0.013
book-0.013
dream-0.013
 chess-0.013
Token a
Token ID257
Feature activation+2.29e-02

Pos logit contributions
 backstage+0.034
dream+0.032
book+0.032
 projector+0.032
 online+0.031
Neg logit contributions
Fine-0.033
ipt-0.032
el-0.029
 Cro-0.028
 Tit-0.028
Token.
Token ID13
Feature activation+4.27e-03

Pos logit contributions
Fine+0.035
ipt+0.033
el+0.032
 Cro+0.030
 Din+0.030
Neg logit contributions
 projector-0.027
 backstage-0.027
 dining-0.026
book-0.026
pron-0.025
Token
Token ID198
Feature activation+4.67e-03

Pos logit contributions
 backstage+0.006
dream+0.006
 upon+0.006
 chess+0.006
book+0.006
Neg logit contributions
Fine-0.007
ipt-0.006
el-0.006
Flo-0.006
 Cro-0.006
Token
Token ID198
Feature activation+1.16e-03

Pos logit contributions
 backstage+0.007
dream+0.006
book+0.006
 projector+0.006
 chess+0.006
Neg logit contributions
Fine-0.008
ipt-0.008
el-0.007
 Cro-0.007
 Din-0.006
TokenThe
Token ID464
Feature activation-6.71e-03

Pos logit contributions
 backstage+0.002
 projector+0.002
dream+0.002
book+0.002
 chess+0.002
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.001
E-0.001
Flo-0.001
Token next
Token ID1306
Feature activation-9.52e-03

Pos logit contributions
Fine+0.009
ipt+0.008
E+0.007
el+0.007
 ow+0.007
Neg logit contributions
book-0.012
 backstage-0.011
dream-0.011
 online-0.011
epad-0.011
Token night
Token ID1755
Feature activation+4.51e-02

Pos logit contributions
Fine+0.021
ipt+0.020
el+0.018
 Cro+0.018
am+0.018
Neg logit contributions
 backstage-0.009
 projector-0.009
dream-0.009
book-0.009
 chess-0.008
Token,
Token ID11
Feature activation+2.01e-03

Pos logit contributions
 backstage+0.080
 online+0.076
epad+0.074
 projector+0.073
 dining+0.073
Neg logit contributions
Fine-0.057
ipt-0.054
 Cro-0.052
el-0.051
 Twe-0.051
Token Lily
Token ID20037
Feature activation-3.37e-03

Pos logit contributions
 backstage+0.004
 projector+0.004
book+0.004
epad+0.004
 menu+0.004
Neg logit contributions
Fine-0.002
ipt-0.002
 Twe-0.002
E-0.002
 Din-0.002
Token didn
Token ID1422
Feature activation+5.28e-03

Pos logit contributions
Fine+0.005
ipt+0.004
el+0.004
am+0.004
Flo+0.004
Neg logit contributions
 backstage-0.006
book-0.006
 projector-0.006
 online-0.005
 chess-0.005
Token't
Token ID470
Feature activation+1.82e-02

Pos logit contributions
 backstage+0.007
dream+0.007
 chess+0.007
 online+0.007
book+0.007
Neg logit contributions
Fine-0.009
ipt-0.009
 Tit-0.008
el-0.008
am-0.008
Token want
Token ID765
Feature activation+4.19e-04

Pos logit contributions
 backstage+0.032
 projector+0.031
 dining+0.031
book+0.030
 chess+0.030
Neg logit contributions
Fine-0.025
Flo-0.021
 Cro-0.021
 Tit-0.021
el-0.020
Token liked
Token ID8288
Feature activation+1.04e-02

Pos logit contributions
Fine+0.009
ipt+0.009
el+0.008
E+0.007
 Cro+0.007
Neg logit contributions
 backstage-0.007
dream-0.007
 projector-0.007
book-0.007
 chess-0.006
Token to
Token ID284
Feature activation-2.22e-02

Pos logit contributions
 backstage+0.013
book+0.012
 projector+0.012
dream+0.011
epad+0.011
Neg logit contributions
ipt-0.019
Fine-0.019
 Cro-0.018
 Tit-0.017
 Twe-0.017
Token paint
Token ID7521
Feature activation+1.86e-02

Pos logit contributions
Fine+0.034
ipt+0.032
el+0.029
 Cro+0.028
Flo+0.028
Neg logit contributions
 backstage-0.036
 projector-0.034
dream-0.034
book-0.034
 chess-0.033
Token.
Token ID13
Feature activation-8.18e-03

Pos logit contributions
 backstage+0.030
dream+0.028
 projector+0.028
 dining+0.027
 upon+0.027
Neg logit contributions
Fine-0.027
ipt-0.027
el-0.026
 Cro-0.024
 Din-0.024
Token He
Token ID679
Feature activation-7.99e-03

Pos logit contributions
Fine+0.014
ipt+0.013
am+0.011
el+0.011
 lam+0.011
Neg logit contributions
 chess-0.012
 projector-0.011
dream-0.011
 online-0.011
book-0.011
Token liked
Token ID8288
Feature activation+2.65e-02

Pos logit contributions
Fine+0.016
ipt+0.015
el+0.014
 Cro+0.013
Flo+0.013
Neg logit contributions
 backstage-0.009
dream-0.009
book-0.009
 projector-0.009
 chess-0.009
Token to
Token ID284
Feature activation-9.20e-03

Pos logit contributions
 backstage+0.032
 projector+0.032
book+0.032
dream+0.031
 dining+0.030
Neg logit contributions
ipt-0.048
Fine-0.048
 Cro-0.045
el-0.044
 Tit-0.044
Token use
Token ID779
Feature activation+1.39e-03

Pos logit contributions
Fine+0.014
ipt+0.013
el+0.012
 Din+0.012
 Cro+0.012
Neg logit contributions
 backstage-0.014
 projector-0.014
dream-0.014
book-0.014
 upon-0.013
Token ink
Token ID16882
Feature activation+3.63e-02

Pos logit contributions
 backstage+0.003
 dining+0.003
arium+0.003
 projector+0.003
book+0.003
Neg logit contributions
el-0.001
 tires-0.001
ipt-0.001
am-0.001
 branches-0.001
Token to
Token ID284
Feature activation-1.03e-02

Pos logit contributions
 dining+0.057
 projector+0.056
 backstage+0.056
arium+0.052
pron+0.052
Neg logit contributions
am-0.052
el-0.052
 brace-0.049
ipt-0.049
Fine-0.049
Token paint
Token ID7521
Feature activation+2.97e-02

Pos logit contributions
Fine+0.013
el+0.012
ipt+0.011
 Din+0.011
 Cro+0.011
Neg logit contributions
 backstage-0.018
 projector-0.018
dream-0.018
arium-0.018
 dining-0.018
Token.
Token ID13
Feature activation+4.20e-05

Pos logit contributions
Fine+0.005
ipt+0.005
el+0.004
 Cro+0.004
 Din+0.004
Neg logit contributions
 dining-0.005
 backstage-0.005
pron-0.005
book-0.005
 chess-0.005
Token "
Token ID366
Feature activation-9.02e-03

Pos logit contributions
 backstage+0.000
 projector+0.000
dream+0.000
book+0.000
 online+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
el-0.000
E-0.000
 Cro-0.000
TokenWhat
Token ID2061
Feature activation-2.20e-02

Pos logit contributions
Fine+0.012
ipt+0.012
am+0.012
el+0.011
 Tit+0.011
Neg logit contributions
 backstage-0.014
dream-0.014
 projector-0.014
book-0.014
 upon-0.013
Token was
Token ID373
Feature activation-8.96e-03

Pos logit contributions
Fine+0.033
ipt+0.032
Flo+0.027
el+0.027
E+0.026
Neg logit contributions
 chess-0.036
 backstage-0.035
book-0.035
 upon-0.035
 projector-0.035
Token that
Token ID326
Feature activation-1.05e-02

Pos logit contributions
Fine+0.015
ipt+0.014
el+0.013
 Cro+0.013
 Din+0.012
Neg logit contributions
 backstage-0.013
book-0.012
dream-0.012
 projector-0.012
 chess-0.012
Token?"
Token ID1701
Feature activation+1.37e-02

Pos logit contributions
Fine+0.011
ipt+0.010
 Cro+0.009
 Tit+0.009
el+0.009
Neg logit contributions
 backstage-0.021
dream-0.021
epad-0.020
 projector-0.020
book-0.020
Token asked
Token ID1965
Feature activation-1.49e-02

Pos logit contributions
dream+0.021
 backstage+0.021
book+0.021
 chess+0.021
 projector+0.020
Neg logit contributions
Fine-0.021
ipt-0.021
el-0.018
am-0.018
 Cro-0.018
Token Lily
Token ID20037
Feature activation-1.08e-02

Pos logit contributions
Fine+0.024
ipt+0.023
el+0.019
am+0.019
Flo+0.019
Neg logit contributions
 backstage-0.023
book-0.023
 chess-0.022
 online-0.022
 projector-0.022
Token.
Token ID13
Feature activation+9.46e-03

Pos logit contributions
Fine+0.017
ipt+0.017
el+0.015
 Cro+0.014
am+0.014
Neg logit contributions
 backstage-0.016
 projector-0.016
dream-0.016
book-0.016
 chess-0.015
Token "
Token ID366
Feature activation-6.42e-03

Pos logit contributions
 backstage+0.014
dream+0.013
 projector+0.013
book+0.013
 chess+0.013
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.014
 Cro-0.013
Flo-0.013
TokenI
Token ID40
Feature activation+1.83e-02

Pos logit contributions
Fine+0.008
ipt+0.008
am+0.007
el+0.007
 Cro+0.006
Neg logit contributions
 backstage-0.012
dream-0.011
book-0.011
 projector-0.011
 upon-0.011
Token the
Token ID262
Feature activation-1.10e-02

Pos logit contributions
book+0.015
dream+0.014
 projector+0.014
 backstage+0.014
pron+0.014
Neg logit contributions
Fine-0.009
ipt-0.009
el-0.008
 Cro-0.008
 Twe-0.008
Token zoo
Token ID26626
Feature activation-4.26e-03

Pos logit contributions
Fine+0.013
 Cro+0.012
 branches+0.012
 ducks+0.012
 birds+0.012
Neg logit contributions
book-0.022
dream-0.021
epad-0.021
pron-0.021
 backstage-0.020
Token.
Token ID13
Feature activation-4.46e-03

Pos logit contributions
el+0.004
am+0.004
 nests+0.004
ipt+0.004
 Cro+0.004
Neg logit contributions
pron-0.008
dream-0.008
 backstage-0.008
book-0.008
 projector-0.008
Token It
Token ID632
Feature activation-3.04e-02

Pos logit contributions
Fine+0.006
ipt+0.006
el+0.005
 Cro+0.005
am+0.005
Neg logit contributions
 backstage-0.008
dream-0.008
 projector-0.008
book-0.008
 chess-0.008
Token was
Token ID373
Feature activation-1.62e-02

Pos logit contributions
Fine+0.041
ipt+0.037
 Cro+0.036
Flo+0.035
el+0.035
Neg logit contributions
 backstage-0.056
epad-0.054
book-0.054
dream-0.053
 chess-0.052
Token not
Token ID407
Feature activation+1.36e-03

Pos logit contributions
 Cro+0.016
 Twe+0.016
 ow+0.015
 Tit+0.015
 Din+0.015
Neg logit contributions
 backstage-0.033
arium-0.033
book-0.032
igsaw-0.032
 dining-0.031
Token happy
Token ID3772
Feature activation-2.65e-02

Pos logit contributions
 backstage+0.002
 dining+0.002
 projector+0.002
arium+0.002
dream+0.002
Neg logit contributions
 Twe-0.002
 Cro-0.002
 Din-0.002
Fine-0.002
 Tit-0.002
Token to
Token ID284
Feature activation+1.20e-02

Pos logit contributions
Fine+0.029
ipt+0.028
am+0.028
el+0.027
 Cro+0.026
Neg logit contributions
pron-0.052
 projector-0.051
 backstage-0.051
dream-0.049
book-0.049
Token be
Token ID307
Feature activation+5.26e-03

Pos logit contributions
 backstage+0.019
dream+0.018
 projector+0.018
book+0.018
 chess+0.018
Neg logit contributions
Fine-0.019
ipt-0.018
el-0.016
 Cro-0.015
E-0.015
Token in
Token ID287
Feature activation+1.78e-02

Pos logit contributions
 dining+0.009
pron+0.009
 projector+0.009
 backstage+0.009
book+0.009
Neg logit contributions
am-0.007
 Twe-0.006
 Cro-0.006
 ducks-0.006
 feathers-0.006
Token the
Token ID262
Feature activation-1.44e-02

Pos logit contributions
 backstage+0.035
pron+0.035
 projector+0.035
book+0.034
 dining+0.034
Neg logit contributions
el-0.018
ipt-0.018
Fine-0.018
 Cro-0.017
 rats-0.016
Token whist
Token ID32262
Feature activation+1.99e-02

Pos logit contributions
 backstage+0.011
dream+0.011
 projector+0.011
book+0.011
 online+0.010
Neg logit contributions
Fine-0.010
ipt-0.010
el-0.009
 Cro-0.008
 Tit-0.008
Tokenling
Token ID1359
Feature activation+2.08e-03

Pos logit contributions
 backstage+0.039
 dining+0.039
 projector+0.038
 online+0.038
arium+0.038
Neg logit contributions
Fine-0.022
el-0.021
ipt-0.020
am-0.020
bit-0.017
Token and
Token ID290
Feature activation+2.31e-03

Pos logit contributions
 backstage+0.004
pron+0.004
arium+0.003
 dining+0.003
 projector+0.003
Neg logit contributions
Fine-0.003
el-0.002
am-0.002
 winds-0.002
 tires-0.002
Token was
Token ID373
Feature activation-1.46e-02

Pos logit contributions
book+0.004
 backstage+0.004
arium+0.004
 menu+0.004
 menus+0.004
Neg logit contributions
Fine-0.003
ipt-0.003
 Twe-0.003
am-0.003
 Cro-0.003
Token sure
Token ID1654
Feature activation-4.24e-02

Pos logit contributions
Fine+0.015
am+0.015
 Twe+0.015
 Tit+0.014
ipt+0.014
Neg logit contributions
arium-0.028
 backstage-0.026
 menu-0.026
 tablet-0.026
 folder-0.026
Token he
Token ID339
Feature activation+9.62e-03

Pos logit contributions
Fine+0.060
ipt+0.057
el+0.050
am+0.050
 Cro+0.049
Neg logit contributions
 backstage-0.074
 projector-0.072
book-0.070
dream-0.070
 dining-0.068
Token was
Token ID373
Feature activation-4.71e-03

Pos logit contributions
 backstage+0.018
 dining+0.017
 projector+0.016
dream+0.016
book+0.016
Neg logit contributions
Fine-0.013
ipt-0.012
el-0.011
am-0.011
 Cro-0.011
Token ready
Token ID3492
Feature activation-1.97e-02

Pos logit contributions
Fine+0.005
am+0.005
 Twe+0.005
 tires+0.005
el+0.005
Neg logit contributions
arium-0.008
 menu-0.008
book-0.008
 tablet-0.008
 folder-0.008
Token to
Token ID284
Feature activation-5.60e-03

Pos logit contributions
el+0.019
am+0.019
ipt+0.017
iously+0.017
 Din+0.017
Neg logit contributions
arium-0.035
 projector-0.035
 tablet-0.033
 laboratory-0.033
book-0.033
Token win
Token ID1592
Feature activation-1.16e-03

Pos logit contributions
Fine+0.007
el+0.006
ipt+0.006
am+0.006
 Cro+0.006
Neg logit contributions
book-0.010
 backstage-0.010
 online-0.010
dream-0.010
 upon-0.010
Token the
Token ID262
Feature activation-1.18e-02

Pos logit contributions
Fine+0.001
ipt+0.001
el+0.001
 Cro+0.001
am+0.001
Neg logit contributions
 projector-0.002
pron-0.002
 dining-0.002
 backstage-0.002
book-0.002
Token radio
Token ID5243
Feature activation+9.22e-03

Pos logit contributions
 k+0.001
 tires+0.001
 brakes+0.001
 chip+0.001
 bird+0.001
Neg logit contributions
arium-0.002
 backstage-0.002
 online-0.002
book-0.002
ores-0.002
Token.
Token ID13
Feature activation-7.40e-03

Pos logit contributions
 dining+0.018
arium+0.017
pron+0.017
book+0.017
dream+0.017
Neg logit contributions
el-0.009
 k-0.008
 tires-0.008
am-0.008
 chip-0.008
Token Sara
Token ID24799
Feature activation+5.38e-03

Pos logit contributions
Fine+0.011
ipt+0.010
el+0.010
am+0.009
 brace+0.009
Neg logit contributions
 upon-0.011
dream-0.011
 backstage-0.011
 menu-0.011
book-0.011
Token and
Token ID290
Feature activation+8.32e-03

Pos logit contributions
 backstage+0.009
 projector+0.009
dream+0.009
book+0.009
 dining+0.009
Neg logit contributions
Fine-0.008
ipt-0.007
am-0.007
el-0.007
 Cro-0.006
Token Ben
Token ID3932
Feature activation-1.74e-02

Pos logit contributions
 backstage+0.013
dream+0.013
 chess+0.013
 projector+0.012
 upon+0.012
Neg logit contributions
Fine-0.013
ipt-0.013
Flo-0.011
E-0.011
el-0.011
Token loved
Token ID6151
Feature activation+2.85e-02

Pos logit contributions
Fine+0.026
ipt+0.024
el+0.022
am+0.021
 Cro+0.020
Neg logit contributions
 backstage-0.029
 projector-0.028
book-0.028
 chess-0.028
dream-0.028
Token to
Token ID284
Feature activation-6.40e-03

Pos logit contributions
book+0.040
 backstage+0.039
 projector+0.038
dream+0.038
 dining+0.037
Neg logit contributions
Fine-0.048
ipt-0.047
 Cro-0.043
el-0.042
 Tit-0.042
Token drive
Token ID3708
Feature activation-6.97e-03

Pos logit contributions
Fine+0.010
ipt+0.010
el+0.009
 Cro+0.008
am+0.008
Neg logit contributions
 backstage-0.010
book-0.010
dream-0.009
 projector-0.009
 chess-0.009
Token the
Token ID262
Feature activation+5.23e-03

Pos logit contributions
Fine+0.008
ipt+0.008
el+0.008
 tires+0.007
am+0.007
Neg logit contributions
 dining-0.013
pron-0.013
 backstage-0.012
dream-0.012
 projector-0.012
Token car
Token ID1097
Feature activation+9.14e-03

Pos logit contributions
 backstage+0.011
dream+0.011
book+0.011
 dining+0.010
 projector+0.010
Neg logit contributions
Fine-0.005
ipt-0.005
el-0.004
am-0.004
 Cro-0.004
Token and
Token ID290
Feature activation-1.19e-02

Pos logit contributions
 dining+0.013
pron+0.013
 projector+0.013
 backstage+0.013
dream+0.012
Neg logit contributions
ipt-0.014
Fine-0.014
am-0.013
el-0.013
 tires-0.013
Token of
Token ID286
Feature activation-8.74e-03

Pos logit contributions
Fine+0.014
ipt+0.013
el+0.011
Flo+0.011
 Cro+0.011
Neg logit contributions
 backstage-0.028
dream-0.028
 projector-0.028
 dining-0.027
book-0.027
Token yours
Token ID12431
Feature activation+1.76e-02

Pos logit contributions
Fine+0.009
el+0.008
ipt+0.008
 Cro+0.008
 Twe+0.007
Neg logit contributions
book-0.018
 dining-0.018
dream-0.018
 backstage-0.018
 projector-0.017
Token.
Token ID13
Feature activation+6.02e-03

Pos logit contributions
dream+0.039
 projector+0.039
pron+0.038
book+0.038
epad+0.037
Neg logit contributions
Fine-0.014
ipt-0.012
Flo-0.010
am-0.010
 Cro-0.010
Token You
Token ID921
Feature activation+2.92e-02

Pos logit contributions
 backstage+0.012
dream+0.012
book+0.012
 projector+0.012
 chess+0.011
Neg logit contributions
Fine-0.007
ipt-0.006
el-0.005
 Cro-0.005
Flo-0.005
Token can
Token ID460
Feature activation-1.99e-03

Pos logit contributions
 backstage+0.051
book+0.049
 dining+0.049
dream+0.048
 projector+0.048
Neg logit contributions
Fine-0.041
ipt-0.036
el-0.035
am-0.034
Flo-0.033
Token trust
Token ID3774
Feature activation-1.56e-02

Pos logit contributions
Fine+0.003
ipt+0.003
el+0.003
Flo+0.003
 Cro+0.003
Neg logit contributions
 backstage-0.003
book-0.003
 projector-0.003
dream-0.003
 chess-0.003
Token me
Token ID502
Feature activation+2.65e-02

Pos logit contributions
Fine+0.015
ipt+0.013
el+0.012
 Cro+0.011
Flo+0.011
Neg logit contributions
 projector-0.034
 backstage-0.034
book-0.034
 dining-0.034
dream-0.034
Token."
Token ID526
Feature activation+4.38e-03

Pos logit contributions
 projector+0.055
 dining+0.053
dream+0.052
book+0.052
 backstage+0.052
Neg logit contributions
Flo-0.025
Fine-0.024
ipt-0.022
el-0.022
 myself-0.022
Token
Token ID198
Feature activation+2.90e-03

Pos logit contributions
 backstage+0.007
 chess+0.007
dream+0.007
 projector+0.007
book+0.007
Neg logit contributions
Fine-0.007
ipt-0.006
el-0.006
Flo-0.006
E-0.006
Token
Token ID198
Feature activation-2.27e-04

Pos logit contributions
 backstage+0.004
dream+0.004
book+0.004
 upon+0.004
 chess+0.004
Neg logit contributions
Fine-0.005
ipt-0.005
el-0.005
 Din-0.004
 Cro-0.004
TokenThe
Token ID464
Feature activation-8.18e-03

Pos logit contributions
Fine+0.000
ipt+0.000
el+0.000
 Din+0.000
 Cro+0.000
Neg logit contributions
 backstage-0.000
 chess-0.000
dream-0.000
book-0.000
 projector-0.000
Token that
Token ID326
Feature activation+1.59e-03

Pos logit contributions
 backstage+0.012
 projector+0.012
dream+0.012
 chess+0.011
book+0.011
Neg logit contributions
Fine-0.010
ipt-0.010
el-0.008
am-0.008
 Cro-0.008
Token she
Token ID673
Feature activation+1.79e-02

Pos logit contributions
 backstage+0.003
 projector+0.003
epad+0.003
dream+0.003
 online+0.003
Neg logit contributions
Fine-0.002
ipt-0.002
 Din-0.002
 Twe-0.002
 Cro-0.002
Token started
Token ID2067
Feature activation-6.68e-03

Pos logit contributions
 backstage+0.034
book+0.032
 projector+0.032
dream+0.032
epad+0.031
Neg logit contributions
Fine-0.023
ipt-0.020
am-0.019
Flo-0.018
el-0.018
Token spl
Token ID4328
Feature activation-4.37e-03

Pos logit contributions
ipt+0.010
el+0.010
am+0.009
 Cro+0.009
 Din+0.009
Neg logit contributions
 projector-0.011
book-0.010
arium-0.010
 backstage-0.010
pron-0.009
Tokenashing
Token ID2140
Feature activation+2.49e-02

Pos logit contributions
Fine+0.010
ipt+0.009
am+0.009
el+0.009
Flo+0.008
Neg logit contributions
 backstage-0.004
 projector-0.004
 dining-0.004
book-0.004
 chess-0.004
Token it
Token ID340
Feature activation-1.04e-02

Pos logit contributions
pron+0.040
 projector+0.039
epad+0.038
 dining+0.037
 backstage+0.037
Neg logit contributions
Fine-0.038
ipt-0.036
el-0.035
 flies-0.034
 branches-0.034
Token with
Token ID351
Feature activation-1.72e-02

Pos logit contributions
Fine+0.012
el+0.012
 flies+0.012
 branches+0.012
am+0.012
Neg logit contributions
 projector-0.019
 dining-0.018
 menus-0.018
 backstage-0.017
epad-0.017
Token water
Token ID1660
Feature activation-7.45e-03

Pos logit contributions
Fine+0.022
ipt+0.021
el+0.019
 Din+0.018
E+0.018
Neg logit contributions
 backstage-0.031
 projector-0.031
book-0.030
pron-0.030
arium-0.030
Token.
Token ID13
Feature activation+5.83e-05

Pos logit contributions
el+0.008
ipt+0.008
am+0.007
 chip+0.007
 Cro+0.007
Neg logit contributions
 projector-0.015
book-0.014
pron-0.014
 dining-0.014
epad-0.014
Token
Token ID198
Feature activation+8.59e-04

Pos logit contributions
 backstage+0.000
 projector+0.000
book+0.000
 chess+0.000
epad+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
 Din-0.000
E-0.000
 Cro-0.000
Token
Token ID198
Feature activation-1.16e-03

Pos logit contributions
 backstage+0.002
 projector+0.002
 chess+0.001
epad+0.001
 menu+0.001
Neg logit contributions
Fine-0.001
ipt-0.001
E-0.001
Flo-0.001
el-0.001
Token joy
Token ID8716
Feature activation+4.61e-03

Pos logit contributions
 backstage+0.007
book+0.007
pron+0.007
 dining+0.007
 projector+0.007
Neg logit contributions
Fine-0.006
ipt-0.006
el-0.005
 Cro-0.005
 Din-0.005
Token as
Token ID355
Feature activation-2.83e-02

Pos logit contributions
 projector+0.008
arium+0.008
 backstage+0.008
book+0.008
 dining+0.008
Neg logit contributions
ipt-0.005
el-0.005
Fine-0.005
am-0.005
 Din-0.005
Token he
Token ID339
Feature activation-1.42e-02

Pos logit contributions
Fine+0.038
ipt+0.036
el+0.032
 Cro+0.031
 Tit+0.031
Neg logit contributions
 backstage-0.051
 projector-0.049
book-0.049
dream-0.048
 chess-0.048
Token grabbed
Token ID13176
Feature activation-5.85e-03

Pos logit contributions
Fine+0.019
ipt+0.019
el+0.016
 Cro+0.016
E+0.016
Neg logit contributions
 backstage-0.025
dream-0.024
book-0.024
 projector-0.023
 chess-0.023
Token the
Token ID262
Feature activation-1.33e-02

Pos logit contributions
Fine+0.006
ipt+0.006
el+0.006
 Din+0.005
 Cro+0.005
Neg logit contributions
 backstage-0.011
 dining-0.011
 projector-0.011
arium-0.011
pron-0.011
Token biggest
Token ID4094
Feature activation-8.02e-03

Pos logit contributions
Fine+0.016
ipt+0.016
el+0.014
 Cro+0.013
 Din+0.013
Neg logit contributions
 backstage-0.025
book-0.024
 projector-0.024
arium-0.024
dream-0.023
Token and
Token ID290
Feature activation+1.67e-04

Pos logit contributions
Fine+0.011
ipt+0.010
el+0.010
 Cro+0.009
 tires+0.009
Neg logit contributions
 backstage-0.014
 projector-0.013
book-0.013
arium-0.013
 online-0.013
Token most
Token ID749
Feature activation+2.96e-03

Pos logit contributions
 backstage+0.000
dream+0.000
 chess+0.000
 projector+0.000
 online+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
el-0.000
E-0.000
am-0.000
Token amazing
Token ID4998
Feature activation+1.64e-03

Pos logit contributions
 backstage+0.005
book+0.005
 projector+0.005
arium+0.004
 dining+0.004
Neg logit contributions
Fine-0.005
ipt-0.004
el-0.004
 Din-0.004
E-0.004
Token ice
Token ID4771
Feature activation-1.80e-02

Pos logit contributions
 backstage+0.003
book+0.003
 projector+0.003
 online+0.003
dream+0.003
Neg logit contributions
Fine-0.002
ipt-0.002
el-0.002
 Cro-0.002
am-0.002
Token cream
Token ID8566
Feature activation-1.95e-02

Pos logit contributions
Fine+0.025
ipt+0.025
el+0.022
 Cro+0.021
am+0.021
Neg logit contributions
 backstage-0.031
 projector-0.030
dream-0.029
book-0.029
 chess-0.029
Token
Token ID198
Feature activation-2.25e-03

Pos logit contributions
Fine+0.002
ipt+0.001
E+0.001
Flo+0.001
el+0.001
Neg logit contributions
 backstage-0.002
 projector-0.002
dream-0.002
 online-0.002
book-0.002
TokenWhen
Token ID2215
Feature activation-1.90e-02

Pos logit contributions
Fine+0.003
ipt+0.002
E+0.002
Flo+0.002
Bl+0.002
Neg logit contributions
 backstage-0.005
 online-0.005
 projector-0.005
epad-0.004
 menu-0.004
Token Tim
Token ID5045
Feature activation-1.56e-02

Pos logit contributions
Fine+0.024
ipt+0.023
 Cro+0.021
el+0.021
 Din+0.020
Neg logit contributions
 backstage-0.035
book-0.034
 projector-0.033
dream-0.033
 online-0.033
Tokenmy
Token ID1820
Feature activation+5.44e-04

Pos logit contributions
Fine+0.026
ipt+0.025
el+0.022
am+0.021
 Cro+0.021
Neg logit contributions
 backstage-0.023
dream-0.022
 projector-0.022
book-0.022
 chess-0.022
Token woke
Token ID19092
Feature activation-2.04e-02

Pos logit contributions
 backstage+0.001
dream+0.001
 projector+0.001
book+0.001
 chess+0.001
Neg logit contributions
Fine-0.001
ipt-0.001
el-0.001
 Cro-0.001
am-0.001
Token up
Token ID510
Feature activation-1.28e-02

Pos logit contributions
Fine+0.032
ipt+0.031
el+0.029
am+0.028
 Cro+0.028
Neg logit contributions
book-0.031
 backstage-0.030
 projector-0.030
dream-0.029
epad-0.028
Token,
Token ID11
Feature activation+1.60e-02

Pos logit contributions
 Cro+0.015
 Twe+0.015
el+0.014
 Din+0.013
 chip+0.013
Neg logit contributions
 projector-0.025
book-0.024
pron-0.024
 backstage-0.024
epad-0.023
Token he
Token ID339
Feature activation+1.05e-02

Pos logit contributions
 backstage+0.036
 projector+0.035
book+0.035
pron+0.034
epad+0.034
Neg logit contributions
Fine-0.014
 Twe-0.013
 Cro-0.013
ipt-0.012
 Din-0.012
Token remembered
Token ID12086
Feature activation-7.26e-03

Pos logit contributions
 backstage+0.021
book+0.021
epad+0.020
 online+0.020
 projector+0.020
Neg logit contributions
Fine-0.011
ipt-0.011
am-0.011
 Cro-0.010
el-0.010
Token his
Token ID465
Feature activation-6.74e-03

Pos logit contributions
 Twe+0.005
 Cro+0.005
 Tit+0.005
 Pe+0.004
 ac+0.004
Neg logit contributions
 projector-0.017
book-0.017
 backstage-0.016
 tablet-0.016
pron-0.016
Token dream
Token ID4320
Feature activation+1.03e-02

Pos logit contributions
Fine+0.007
 chip+0.007
am+0.007
ipt+0.007
 tires+0.006
Neg logit contributions
 backstage-0.013
book-0.013
 projector-0.012
epad-0.012
arium-0.012
Token sniff
Token ID26300
Feature activation+8.08e-03

Pos logit contributions
book+0.018
ifying+0.017
 projector+0.017
igsaw+0.016
 dining+0.016
Neg logit contributions
Fine-0.014
 Twe-0.014
ipt-0.014
 Cro-0.014
 Tit-0.013
Tokening
Token ID278
Feature activation-2.42e-03

Pos logit contributions
 backstage+0.012
book+0.012
dream+0.012
 online+0.012
 upon+0.012
Neg logit contributions
Fine-0.013
ipt-0.012
el-0.010
 Din-0.010
 Tit-0.010
Token and
Token ID290
Feature activation+5.92e-03

Pos logit contributions
Fine+0.004
el+0.003
ipt+0.003
 Cro+0.003
am+0.003
Neg logit contributions
 backstage-0.004
pron-0.004
 dining-0.004
 projector-0.004
book-0.003
Token w
Token ID266
Feature activation+2.13e-03

Pos logit contributions
book+0.009
 projector+0.008
 backstage+0.008
pron+0.008
arium+0.008
Neg logit contributions
Fine-0.009
 Cro-0.009
ipt-0.009
 Tit-0.009
 Twe-0.009
Tokenagging
Token ID16406
Feature activation+2.47e-03

Pos logit contributions
dream+0.003
 backstage+0.003
book+0.003
 online+0.003
 chess+0.003
Neg logit contributions
ipt-0.004
Fine-0.004
 Din-0.003
 Cro-0.003
 Tit-0.003
Token his
Token ID465
Feature activation-2.57e-02

Pos logit contributions
 backstage+0.005
 projector+0.005
 dining+0.005
pron+0.005
 online+0.005
Neg logit contributions
ipt-0.003
Fine-0.003
el-0.002
 Cro-0.002
 Tit-0.002
Token tail
Token ID7894
Feature activation-1.33e-02

Pos logit contributions
Fine+0.037
ipt+0.035
el+0.031
 Cro+0.030
Flo+0.029
Neg logit contributions
 backstage-0.044
dream-0.042
 projector-0.042
book-0.042
 chess-0.042
Token.
Token ID13
Feature activation-1.04e-02

Pos logit contributions
Fine+0.013
ipt+0.013
el+0.012
am+0.012
 branches+0.012
Neg logit contributions
 backstage-0.026
 projector-0.026
pron-0.025
arium-0.025
book-0.025
Token Finally
Token ID9461
Feature activation-8.71e-03

Pos logit contributions
Fine+0.013
ipt+0.012
 So+0.010
 Din+0.010
 Cro+0.010
Neg logit contributions
 backstage-0.020
book-0.019
 dining-0.019
dream-0.018
 upon-0.018
Token,
Token ID11
Feature activation+2.28e-03

Pos logit contributions
Fine+0.014
ipt+0.013
el+0.012
 Cro+0.012
 Din+0.012
Neg logit contributions
 backstage-0.013
 projector-0.013
book-0.013
 dining-0.012
epad-0.012
Token he
Token ID339
Feature activation+4.33e-03

Pos logit contributions
 dining+0.005
 backstage+0.005
book+0.005
 ago+0.005
epad+0.005
Neg logit contributions
Fine-0.002
 Twe-0.002
ipt-0.002
 neither-0.002
 Cro-0.002
Token shrug
Token ID32545
Feature activation+9.39e-03

Pos logit contributions
 backstage+0.022
 online+0.021
 projector+0.021
book+0.021
 dining+0.021
Neg logit contributions
Fine-0.018
ipt-0.017
 Cro-0.016
el-0.016
 Tit-0.016
Token.
Token ID13
Feature activation+1.54e-03

Pos logit contributions
 backstage+0.014
 projector+0.013
dream+0.013
book+0.013
 chess+0.013
Neg logit contributions
Fine-0.016
ipt-0.015
el-0.014
 Cro-0.013
Flo-0.013
Token "
Token ID366
Feature activation-1.12e-02

Pos logit contributions
 upon+0.002
dream+0.002
 chess+0.002
 online+0.002
 backstage+0.002
Neg logit contributions
Fine-0.003
ipt-0.003
el-0.003
Flo-0.002
Bl-0.002
TokenDon
Token ID3987
Feature activation-1.71e-02

Pos logit contributions
ipt+0.016
am+0.016
el+0.015
Fine+0.015
ries+0.015
Neg logit contributions
dream-0.017
 upon-0.016
 backstage-0.016
book-0.015
pron-0.015
Token't
Token ID470
Feature activation+2.04e-02

Pos logit contributions
Fine+0.041
ipt+0.039
 ounces+0.038
 lam+0.037
 heights+0.037
Neg logit contributions
 over-0.011
 upon-0.011
:-0.010
 online-0.010
 about-0.010
Token worry
Token ID5490
Feature activation-5.03e-02

Pos logit contributions
 upon+0.033
 backstage+0.032
dream+0.032
 chess+0.031
 celebrating+0.030
Neg logit contributions
ipt-0.032
Fine-0.029
sn-0.024
E-0.024
am-0.024
Token,"
Token ID553
Feature activation-5.51e-03

Pos logit contributions
Fine+0.122
ipt+0.122
Bl+0.108
sn+0.102
el+0.102
Neg logit contributions
 upon-0.041
 about-0.040
 chess-0.035
 menu-0.032
 online-0.032
Token Sally
Token ID25737
Feature activation+1.75e-05

Pos logit contributions
Fine+0.010
ipt+0.010
 ounces+0.009
Bl+0.009
Flo+0.009
Neg logit contributions
 chess-0.006
 upon-0.006
 online-0.006
dream-0.006
 early-0.006
Token said
Token ID531
Feature activation+1.14e-03

Pos logit contributions
dream+0.000
 upon+0.000
 backstage+0.000
book+0.000
 chess+0.000
Neg logit contributions
Fine-0.000
ipt-0.000
E-0.000
el-0.000
Flo-0.000
Token.
Token ID13
Feature activation+9.02e-03

Pos logit contributions
dream+0.001
 backstage+0.001
 online+0.001
 satisfied+0.001
 chess+0.001
Neg logit contributions
Fine-0.003
ipt-0.003
E-0.002
el-0.002
Bl-0.002
Token "
Token ID366
Feature activation-7.03e-03

Pos logit contributions
 chess+0.010
 backstage+0.010
 upon+0.010
dream+0.010
 online+0.010
Neg logit contributions
Fine-0.018
ipt-0.016
el-0.016
Flo-0.015
E-0.015
Token up
Token ID510
Feature activation-2.29e-02

Pos logit contributions
Fine+0.065
ipt+0.061
el+0.059
Flo+0.057
 Tit+0.056
Neg logit contributions
 projector-0.080
book-0.079
 dining-0.078
dream-0.076
 backstage-0.076
Token until
Token ID1566
Feature activation-1.55e-02

Pos logit contributions
Fine+0.022
el+0.019
 tires+0.018
 heights+0.018
 rats+0.018
Neg logit contributions
 projector-0.050
pron-0.047
arium-0.046
 dining-0.046
book-0.045
Token the
Token ID262
Feature activation-9.82e-03

Pos logit contributions
Fine+0.014
el+0.012
 Din+0.011
 Twe+0.011
ipt+0.011
Neg logit contributions
 projector-0.034
 tablet-0.032
 backstage-0.032
book-0.032
 dining-0.032
Token laser
Token ID12855
Feature activation+2.80e-02

Pos logit contributions
Fine+0.011
 k+0.010
 chip+0.010
 bird+0.010
 birds+0.010
Neg logit contributions
book-0.018
arium-0.017
 backstage-0.017
 online-0.017
dream-0.017
Token finally
Token ID3443
Feature activation-4.04e-02

Pos logit contributions
 backstage+0.052
book+0.050
 dining+0.050
 projector+0.050
dream+0.049
Neg logit contributions
Fine-0.035
ipt-0.032
el-0.029
am-0.029
 Cro-0.028
Token let
Token ID1309
Feature activation-4.95e-02

Pos logit contributions
Fine+0.042
 flew+0.040
am+0.040
 flies+0.039
ipt+0.037
Neg logit contributions
 tablet-0.080
 dining-0.079
 projector-0.079
 backstage-0.078
 menu-0.076
Token her
Token ID607
Feature activation-2.29e-02

Pos logit contributions
Fine+0.063
ipt+0.060
el+0.057
 Din+0.054
 Cro+0.054
Neg logit contributions
 backstage-0.088
 projector-0.085
dream-0.084
 dining-0.083
 chess-0.083
Token catch
Token ID4929
Feature activation-3.33e-02

Pos logit contributions
Fine+0.028
am+0.027
el+0.027
ipt+0.026
 Cro+0.025
Neg logit contributions
 dining-0.039
book-0.039
 backstage-0.039
 projector-0.038
dream-0.038
Token it
Token ID340
Feature activation-9.01e-03

Pos logit contributions
el+0.036
ipt+0.036
Fine+0.035
 Twe+0.034
am+0.033
Neg logit contributions
 projector-0.062
 backstage-0.061
pron-0.061
 dining-0.061
book-0.061
Token.
Token ID13
Feature activation+1.03e-03

Pos logit contributions
el+0.010
am+0.010
ipt+0.009
Fine+0.009
Flo+0.009
Neg logit contributions
 projector-0.017
pron-0.016
 dining-0.016
 backstage-0.016
dream-0.016
Token El
Token ID2574
Feature activation-1.02e-02

Pos logit contributions
 backstage+0.002
 projector+0.002
dream+0.002
book+0.002
 chess+0.002
Neg logit contributions
Fine-0.001
ipt-0.001
el-0.001
E-0.001
 Cro-0.001
Token'm
Token ID1101
Feature activation+2.03e-02

Pos logit contributions
 backstage+0.054
 chess+0.054
 projector+0.053
book+0.053
 tablet+0.053
Neg logit contributions
Fine-0.025
Flo-0.024
el-0.022
E-0.021
 So-0.021
Token sure
Token ID1654
Feature activation-3.47e-02

Pos logit contributions
 menu+0.037
book+0.037
 tablet+0.037
 menus+0.037
 backstage+0.036
Neg logit contributions
 Twe-0.024
el-0.022
 ow-0.021
 Din-0.021
Flo-0.021
Token you
Token ID345
Feature activation+4.60e-02

Pos logit contributions
Fine+0.034
ipt+0.031
el+0.026
 Cro+0.025
 Tit+0.024
Neg logit contributions
 backstage-0.075
 projector-0.073
dream-0.073
book-0.073
 chess-0.071
Token'll
Token ID1183
Feature activation+5.52e-03

Pos logit contributions
 projector+0.091
 dining+0.091
 backstage+0.091
 chess+0.090
dream+0.089
Neg logit contributions
Fine-0.046
el-0.043
am-0.042
 flies-0.041
Flo-0.040
Token be
Token ID307
Feature activation-2.51e-03

Pos logit contributions
 projector+0.012
book+0.011
 menus+0.011
 chess+0.011
 backstage+0.011
Neg logit contributions
Flo-0.005
el-0.005
Fine-0.005
 Twe-0.005
 k-0.004
Token able
Token ID1498
Feature activation-5.68e-02

Pos logit contributions
 Twe+0.003
Flo+0.003
Fine+0.003
el+0.003
am+0.002
Neg logit contributions
 menu-0.005
book-0.005
 projector-0.005
 tablet-0.004
 backstage-0.004
Token to
Token ID284
Feature activation-5.50e-03

Pos logit contributions
Fine+0.079
 Twe+0.077
ipt+0.077
el+0.074
 Cro+0.074
Neg logit contributions
book-0.093
 projector-0.092
 backstage-0.086
dream-0.083
 tablet-0.082
Token see
Token ID766
Feature activation+1.52e-02

Pos logit contributions
Fine+0.007
el+0.007
ipt+0.007
Flo+0.006
 Cro+0.006
Neg logit contributions
book-0.010
 projector-0.009
 backstage-0.009
dream-0.009
 chess-0.009
Token many
Token ID867
Feature activation-1.71e-03

Pos logit contributions
 projector+0.029
book+0.029
dream+0.029
 backstage+0.029
 dining+0.029
Neg logit contributions
Fine-0.016
ipt-0.016
el-0.015
 Twe-0.014
Flo-0.014
Token more
Token ID517
Feature activation+7.57e-03

Pos logit contributions
Fine+0.002
ipt+0.002
el+0.002
E+0.002
Flo+0.002
Neg logit contributions
 projector-0.003
 backstage-0.003
book-0.003
dream-0.003
 chess-0.003
Token flowers
Token ID12734
Feature activation+1.43e-02

Pos logit contributions
 projector+0.015
 chess+0.014
book+0.014
dream+0.014
 backstage+0.014
Neg logit contributions
Fine-0.008
el-0.008
 flies-0.008
ipt-0.008
 lam-0.007
Token said
Token ID531
Feature activation+2.28e-02

Pos logit contributions
Fine+0.008
ipt+0.007
Flo+0.007
 Cro+0.007
am+0.007
Neg logit contributions
 backstage-0.011
 projector-0.011
 dining-0.011
book-0.011
dream-0.011
Token,
Token ID11
Feature activation+3.53e-02

Pos logit contributions
 backstage+0.029
dream+0.028
 projector+0.028
book+0.028
 chess+0.028
Neg logit contributions
Fine-0.042
ipt-0.041
el-0.037
 Cro-0.035
E-0.035
Token "
Token ID366
Feature activation-5.28e-03

Pos logit contributions
 backstage+0.057
 projector+0.054
 dining+0.053
book+0.053
dream+0.052
Neg logit contributions
Fine-0.053
ipt-0.050
 Cro-0.047
 Tit-0.046
 Din-0.046
TokenDon
Token ID3987
Feature activation-1.68e-02

Pos logit contributions
ipt+0.008
 ow+0.008
am+0.008
el+0.008
 Tit+0.007
Neg logit contributions
 upon-0.007
dream-0.007
pron-0.007
book-0.006
 backstage-0.006
Token't
Token ID470
Feature activation+2.32e-02

Pos logit contributions
ipt+0.037
Fine+0.037
 ounces+0.034
 wrists+0.034
 lam+0.033
Neg logit contributions
 upon-0.014
 over-0.012
 everywhere-0.012
 online-0.011
book-0.011
Token worry
Token ID5490
Feature activation-5.01e-02

Pos logit contributions
 upon+0.037
 backstage+0.035
dream+0.035
 chess+0.034
 celebrating+0.032
Neg logit contributions
ipt-0.038
Fine-0.034
sn-0.029
E-0.029
mm-0.028
Token,
Token ID11
Feature activation+1.89e-02

Pos logit contributions
ipt+0.133
Fine+0.130
el+0.113
Pe+0.112
Bl+0.111
Neg logit contributions
 about-0.034
 upon-0.033
 chess-0.026
 for-0.025
 everywhere-0.023
Token I
Token ID314
Feature activation+3.17e-02

Pos logit contributions
 chess+0.032
 upon+0.032
 backstage+0.031
 celebrating+0.031
 projector+0.030
Neg logit contributions
ipt-0.027
Fine-0.026
el-0.023
am-0.022
E-0.020
Token will
Token ID481
Feature activation-5.19e-03

Pos logit contributions
 backstage+0.070
 dining+0.068
 projector+0.067
book+0.067
pron+0.066
Neg logit contributions
Fine-0.030
am-0.028
 Cro-0.025
Flo-0.025
 So-0.024
Token help
Token ID1037
Feature activation-3.62e-02

Pos logit contributions
Fine+0.006
Flo+0.005
el+0.005
 Twe+0.005
 Cro+0.005
Neg logit contributions
 backstage-0.010
book-0.010
 projector-0.010
 dining-0.010
dream-0.010
Token you
Token ID345
Feature activation+4.02e-02

Pos logit contributions
ipt+0.051
Fine+0.046
am+0.038
el+0.037
mm+0.037
Neg logit contributions
 backstage-0.064
 chess-0.063
 upon-0.062
 online-0.061
 celebrating-0.060
Token of
Token ID286
Feature activation+2.16e-02

Pos logit contributions
 backstage+0.020
book+0.019
 projector+0.019
epad+0.018
 dining+0.018
Neg logit contributions
ipt-0.014
Fine-0.014
el-0.012
 chip-0.012
 Cro-0.012
Token himself
Token ID2241
Feature activation-7.93e-03

Pos logit contributions
 backstage+0.047
 projector+0.046
dream+0.046
book+0.045
 online+0.044
Neg logit contributions
Fine-0.021
ipt-0.020
el-0.017
 Cro-0.016
 Tit-0.016
Token that
Token ID326
Feature activation+1.95e-02

Pos logit contributions
Fine+0.015
ipt+0.015
el+0.015
am+0.014
 Din+0.014
Neg logit contributions
 backstage-0.008
 projector-0.008
dream-0.007
book-0.007
pron-0.007
Token he
Token ID339
Feature activation+1.65e-02

Pos logit contributions
 backstage+0.043
 projector+0.040
dream+0.039
uits+0.038
 dining+0.038
Neg logit contributions
Fine-0.018
 Cro-0.017
 Din-0.017
ipt-0.017
el-0.016
Token was
Token ID373
Feature activation+1.88e-03

Pos logit contributions
 backstage+0.023
dream+0.022
 projector+0.022
book+0.021
 chess+0.021
Neg logit contributions
Fine-0.029
ipt-0.027
el-0.025
 Cro-0.024
am-0.024
Token able
Token ID1498
Feature activation-5.51e-02

Pos logit contributions
 backstage+0.003
 projector+0.003
arium+0.003
book+0.003
dream+0.003
Neg logit contributions
Fine-0.003
ipt-0.003
 Din-0.002
 Cro-0.002
 Twe-0.002
Token to
Token ID284
Feature activation-7.86e-03

Pos logit contributions
Fine+0.099
ipt+0.098
 Cro+0.091
am+0.090
el+0.090
Neg logit contributions
 backstage-0.071
 projector-0.069
book-0.068
dream-0.064
arium-0.062
Token create
Token ID2251
Feature activation-5.29e-03

Pos logit contributions
Fine+0.011
ipt+0.010
el+0.010
 Cro+0.009
Flo+0.009
Neg logit contributions
 backstage-0.014
 projector-0.013
dream-0.013
book-0.013
 online-0.013
Token something
Token ID1223
Feature activation-1.01e-02

Pos logit contributions
Fine+0.007
ipt+0.007
el+0.006
 Cro+0.006
am+0.006
Neg logit contributions
 backstage-0.009
 projector-0.009
dream-0.008
book-0.008
epad-0.008
Token tasty
Token ID25103
Feature activation-4.35e-02

Pos logit contributions
Fine+0.012
 Din+0.012
 Cro+0.012
 Twe+0.012
 chip+0.012
Neg logit contributions
 projector-0.017
book-0.017
 backstage-0.017
dream-0.016
 dining-0.016
Token.
Token ID13
Feature activation+2.94e-03

Pos logit contributions
Fine+0.051
 chip+0.051
el+0.050
 Cro+0.049
ipt+0.049
Neg logit contributions
 backstage-0.077
 projector-0.075
dream-0.074
pron-0.073
 dining-0.071