INDEX
Explanations
various types of quotes and spoken dialogue in the text
New Auto-Interp
Negative Logits
{}",-0.59
曖昧さ回避
-0.53
surla
-0.51
couz
-0.50
}>;
-0.50
++;
-0.49
Postup
-0.49
"}";
-0.49
apunov
-0.47
hyo
-0.47
POSITIVE LOGITS
}{*}{0.90
eradish
0.68
Almighty
0.60
wpi
0.59
durs
0.59
fraid
0.59
Tinggi
0.58
vettoriale
0.58
تانيه
0.57
للمعارف
0.57
Activations Density 0.164%