INDEX
Negative Logits
the
0.44
$
0.44
an
0.43
ال
0.41
\
0.40
it
0.39
は
0.39
\
0.36
as
0.32
ization
0.32
POSITIVE LOGITS
at
0.48
AT
0.44
OM
0.43
TS
0.43
BS
0.42
νο
0.39
ల
0.37
AB
0.37
UR
0.36
OL
0.36
Activations Density 0.365%
the
$
an
ال
\
it
は
\
as
ization
at
AT
OM
TS
BS
νο
ల
AB
UR
OL