INDEX
Negative Logits
deserted
0.50
forested
0.50
barren
0.47
suffers
0.45
convuls
0.44
decades
0.43
malignant
0.43
accuses
0.42
peror
0.42
SELECTOR
0.42
POSITIVE LOGITS
unglaublich
0.44
Arbeit
0.43
靔
0.41
πραγμα
0.41
রেজ
0.40
Komm
0.40
اکی
0.39
Prizes
0.39
Gras
0.39
κάν
0.39
Activations Density 0.006%