INDEX
Negative Logits
chré
-0.93
ArrowToggle
-0.90
Cæsar
-0.87
disambiguazione
-0.87
nakalista
-0.87
tendenza
-0.86
specchio
-0.85
hematical
-0.85
pitié
-0.85
blaze
-0.83
POSITIVE LOGITS
time
0.63
-
0.63
for
0.60
able
0.60
out
0.59
off
0.59
co
0.55
up
0.55
net
0.55
ette
0.53
Activations Density 0.040%