INDEX
Negative Logits
illegal
-0.09
legalized
-0.08
iping
-0.08
illegal
-0.08
座
-0.07
idza
-0.07
path
-0.07
unexpected
-0.07
COPYRIGHT
-0.07
needed
-0.07
POSITIVE LOGITS
split
0.13
Split
0.13
Split
0.11
_split
0.11
splits
0.11
(split
0.10
splitted
0.10
dividir
0.10
تقس
0.10
split
0.10
Activations Density 0.004%