INDEX
Negative Logits
affles
0.49
Debugging
0.49
yacute
0.47
ji
0.46
arthy
0.46
ırs
0.44
nette
0.43
이지만
0.43
ucci
0.43
propag
0.43
POSITIVE LOGITS
singlet
0.52
fifties
0.46
impetus
0.45
intricately
0.45
Clarion
0.44
interl
0.44
;
0.42
industrious
0.42
ingles
0.41
っいて
0.41
Activations Density 0.004%