INDEX
Negative Logits
бө
-0.09
builtin
-0.08
bott
-0.08
paragraphs
-0.08
_encode
-0.07
clarified
-0.07
rea
-0.07
Builtin
-0.07
recommending
-0.07
bl
-0.07
POSITIVE LOGITS
criminals
0.09
frowned
0.08
delinc
0.08
crimin
0.08
щиков
0.08
savvy
0.08
antics
0.08
наруш
0.08
LOGY
0.07
caric
0.07
Activations Density 0.012%