INDEX
Negative Logits
aute
-0.09
Доп
-0.09
suppos
-0.08
acclaimed
-0.08
-nous
-0.07
Sér
-0.07
undeniably
-0.07
preferably
-0.07
arguably
-0.07
konst
-0.07
POSITIVE LOGITS
BOX
0.08
cal
0.08
delight
0.07
$',
0.07
simmer
0.07
IN
0.07
Warnings
0.07
PACK
0.07
grandmother
0.07
Mon
0.07
Activations Density 0.125%