INDEX
Negative Logits
ти
1.17
م
1.08
يد
1.05
ल
1.04
اف
1.02
أب
1.02
is
1.00
ఇ
0.99
𝙖
0.96
"
0.95
POSITIVE LOGITS
that
1.34
(
1.30
that
1.23
G
1.11
Y
1.10
de
1.07
X
1.07
F
1.02
که
1.01
Co
0.96
Activations Density 0.010%
ти
م
يد
ल
اف
أب
is
ఇ
𝙖
"
that
(
that
G
Y
de
X
F
که
Co