INDEX
Negative Logits
ن
1.36
the
1.33
н
1.29
น
1.12
to
1.10
ت
1.01
ه
0.99
न
0.98
ق
0.95
was
0.91
POSITIVE LOGITS
며
0.96
行う
0.89
З
0.82
기를
0.82
세요
0.81
gång
0.79
atures
0.78
ഡ്
0.77
.…
0.77
g
0.76
Activations Density 0.002%
ن
the
н
น
to
ت
ه
न
ق
was
며
行う
З
기를
세요
gång
atures
ഡ്
.…
g