INDEX
Negative Logits
ل
1.85
د
1.85
ع
1.60
في
1.58
ر
1.53
л
1.49
خ
1.49
ش
1.46
在
1.45
at
1.43
POSITIVE LOGITS
are
1.66
were
1.22
đều
1.11
ate
1.08
1.00
WERE
0.98
-
0.95
ᆹ
0.95
ARE
0.95
eli
0.94
Activations Density 0.371%
ل
د
ع
في
ر
л
خ
ش
在
at
are
were
đều
ate
WERE
-
ᆹ
ARE
eli