INDEX
Negative Logits
尖
-0.08
ugh
-0.08
occhi
-0.08
tenor
-0.07
片
-0.07
Cad
-0.07
Tem
-0.07
abusive
-0.07
ikh
-0.07
lady
-0.07
POSITIVE LOGITS
concur
0.08
propagated
0.08
totes
0.08
propagation
0.08
₪
0.08
woont
0.08
welcomed
0.08
алин
0.07
վեն
0.07
아니
0.07
Activations Density 0.003%