INDEX
Negative Logits
ا
1.02
다
0.91
to
0.84
the
0.79
ين
0.76
я
0.76
ta
0.75
可能な
0.73
ва
0.73
ून
0.73
POSITIVE LOGITS
for
1.27
i
0.99
are
0.78
for
0.77
0.75
ak
0.73
ado
0.73
es
0.71
iest
0.68
ade
0.67
Activations Density 0.022%
ا
다
to
the
ين
я
ta
可能な
ва
ून
for
i
are
for
ak
ado
es
iest
ade