INDEX
Negative Logits
nya
4.29
ly
3.71
ting
3.35
th
3.11
to
3.08
nrow
2.90
𝘭
2.90
IZATION
2.77
ters
2.75
ta
2.74
POSITIVE LOGITS
an
4.04
न
3.37
۰
3.07
ar
3.04
ר
2.98
ed
2.86
it
2.77
ة
2.61
y
2.57
at
2.56
Activations Density 0.015%
nya
ly
ting
th
to
nrow
𝘭
IZATION
ters
ta
an
न
۰
ar
ר
ed
it
ة
y
at