INDEX
Negative Logits
as
1.18
↵
0.98
is
0.97
↵↵
0.95
in
0.95
0.95
i
0.93
'
0.91
of
0.90
to
0.86
POSITIVE LOGITS
ي
1.61
at
1.45
ב
1.29
يلا
1.26
ле
1.22
يل
1.14
ين
1.13
تي
1.13
ви
1.11
서
1.08
Activations Density 0.042%
as
↵
is
↵↵
in
i
'
of
to
ي
at
ב
يلا
ле
يل
ين
تي
ви
서