INDEX
Negative Logits
ك
1.15
ە
1.13
ן
1.13
ς
1.10
лі
1.07
ся
1.03
ธ์
1.03
റ്റർ
1.02
માં
1.02
ے
1.01
POSITIVE LOGITS
n
1.60
A
1.42
W
1.37
be
1.31
ت
1.28
X
1.27
was
1.24
at
1.23
i
1.23
\
1.22
Activations Density 0.028%
ك
ە
ן
ς
лі
ся
ธ์
റ്റർ
માં
ے
n
A
W
be
ت
X
was
at
i
\