INDEX
Explanations
expressions of emotions and their intensity
New Auto-Interp
Negative Logits
ãĥ³ãĥij
-0.16
counts
-0.14
ucchini
-0.14
GENERIC
-0.14
ú
-0.14
raz
-0.14
otland
-0.14
counts
-0.14
ancial
-0.13
discord
-0.13
POSITIVE LOGITS
ippo
0.17
apus
0.16
ÛĮÙĩ
0.15
ÅĻÃŃm
0.14
GRA
0.14
ائÙĤ
0.14
kova
0.14
nil
0.13
obel
0.13
awah
0.13
Activations Density 0.062%