INDEX
Negative Logits
_repr
-0.07
Allow
-0.06
pistol
-0.06
лекс
-0.06
Version
-0.06
이루
-0.06
woods
-0.06
"?
-0.06
алом
-0.06
";
-0.06
POSITIVE LOGITS
ogui
0.09
_TUN
0.07
figuring
0.06
transportation
0.06
generalized
0.06
اکتبر
0.06
ندية
0.06
Specialty
0.06
action
0.06
freshmen
0.06
Activations Density 0.001%