INDEX
Negative Logits
adlı
0.44
adrenalin
0.42
disinformation
0.42
config
0.42
friendship
0.40
disadvantage
0.40
platform
0.40
'
0.40
hangi
0.39
dodgy
0.39
POSITIVE LOGITS
таксама
0.52
початку
0.50
начиная
0.46
Throughout
0.44
Pregnant
0.44
थी
0.44
толькі
0.43
ปกติ
0.42
чках
0.42
भी
0.41
Activations Density 0.010%