INDEX
Negative Logits
द्र
0.40
屡
0.39
डल
0.38
होईल
0.38
0.37
ausreiche
0.36
complejidad
0.36
主流
0.35
ይ
0.35
Complexity
0.35
POSITIVE LOGITS
intended
0.77
somehow
0.75
meant
0.71
someone
0.71
somebody
0.67
nějak
0.67
irgende
0.65
intended
0.64
Someone
0.63
algum
0.63
Activations Density 0.125%