INDEX
Negative Logits
'
0.89
р
0.88
’
0.86
-
0.85
н
0.85
й
0.82
在
0.78
га
0.78
У
0.76
Ο
0.75
POSITIVE LOGITS
or
1.16
but
1.04
are
1.02
for
0.98
is
0.96
as
0.94
of
0.91
are
0.86
ک
0.83
re
0.82
Activations Density 0.008%
'
р
’
-
н
й
在
га
У
Ο
or
but
are
for
is
as
of
are
ک
re