INDEX
Negative Logits
?
1.08
are
0.83
6
0.82
of
0.80
prodigious
0.79
ни
0.77
erical
0.77
schreibt
0.76
9
0.76
enggak
0.75
POSITIVE LOGITS
(
1.01
م
1.01
ار
0.98
מ
0.96
에
0.93
k
0.92
h
0.89
m
0.86
に
0.85
ور
0.83
Activations Density 0.006%
?
are
6
of
prodigious
ни
erical
schreibt
9
enggak
(
م
ار
מ
에
k
h
m
に
ور