INDEX
Negative Logits
ference
-0.07
ochran
-0.07
ekler
-0.07
regions
-0.06
è
-0.06
XIV
-0.06
Perl
-0.06
institutional
-0.06
parable
-0.06
policy
-0.06
POSITIVE LOGITS
while
0.10
хотя
0.08
though
0.08
虽然
0.08
withstanding
0.07
whilst
0.07
While
0.07
ارت
0.06
"While
0.06
Although
0.06
Activations Density 0.030%