INDEX
Negative Logits
checksum
0.47
round
0.44
stationary
0.44
Round
0.43
/
0.43
spline
0.42
rocks
0.42
footer
0.41
spaces
0.41
Classifier
0.40
POSITIVE LOGITS
Despite
0.40
Помимо
0.39
справедливо
0.35
финансовых
0.35
״
0.34
мимо
0.34
зая
0.34
quán
0.34
निभाया
0.34
Deshalb
0.34
Activations Density 0.002%