INDEX
Negative Logits
valid
-0.78
valid
-0.68
exploiting
-0.66
Appropriate
-0.66
exploit
-0.65
exploited
-0.63
betweenstory
-0.61
tagHelperRunner
-0.60
válida
-0.60
válido
-0.59
POSITIVE LOGITS
ating
0.58
ated
0.57
Pristupljeno
0.56
kuiten
0.55
BeginContext
0.52
abb
0.51
aton
0.49
aber
0.49
]")]
0.49
agrid
0.49
Activations Density 0.056%