INDEX
Negative Logits
ك
1.86
can
1.66
an
1.58
be
1.53
are
1.33
ка
1.24
isn
1.17
on
1.14
is
1.13
ה
1.13
POSITIVE LOGITS
n
1.34
and
1.33
is
1.24
categories
1.23
at
1.19
re
1.13
as
1.10
ig
1.07
en
1.06
u
1.06
Activations Density 0.082%
ك
can
an
be
are
ка
isn
on
is
ה
n
and
is
categories
at
re
as
ig
en
u