INDEX
Negative Logits
religion
-0.08
former
-0.08
ലം
-0.08
جميل
-0.08
virtue
-0.08
enberg
-0.07
=train
-0.07
nei
-0.07
willing
-0.07
силы
-0.07
POSITIVE LOGITS
Guards
0.09
Rhino
0.09
_INCLUDED
0.08
Sir
0.08
antioxid
0.08
Morning
0.08
_QUOTES
0.08
Sunrise
0.08
#ifndef
0.08
_ALREADY
0.08
Activations Density 0.001%