INDEX
Explanations
mentions of dining events and meal-related activities
New Auto-Interp
Negative Logits
ors
-0.19
erer
-0.18
exp
-0.16
oll
-0.16
.za
-0.15
breakfast
-0.15
æ
-0.15
e
-0.15
bis
-0.15
è¨Ń
-0.15
POSITIVE LOGITS
/sn
0.23
ware
0.22
time
0.20
-time
0.19
/movie
0.18
룸
0.16
preparation
0.16
ertime
0.16
STIT
0.15
swagger
0.15
Activations Density 0.017%