INDEX
Explanations
expressions of requests or conditions related to individuals and their actions or statuses
New Auto-Interp
Negative Logits
azo
-0.17
said
-0.15
說
-0.14
said
-0.14
Twilight
-0.14
nab
-0.14
get
-0.14
adoras
-0.14
says
-0.13
says
-0.13
POSITIVE LOGITS
wish
0.71
Wish
0.67
wish
0.65
wishes
0.64
wished
0.59
wishing
0.54
desire
0.31
require
0.30
souha
0.29
wishlist
0.28
Activations Density 0.223%