INDEX
Explanations
words indicating totality or completeness
New Auto-Interp
Negative Logits
oba
-0.19
obic
-0.17
ovich
-0.16
-AA
-0.15
ync
-0.15
rians
-0.14
incidental
-0.14
he
-0.14
elli
-0.14
ocks
-0.13
POSITIVE LOGITS
jis
0.18
igator
0.17
jedn
0.14
ptest
0.14
_initializer
0.14
idos
0.14
hiba
0.14
raya
0.14
otte
0.14
Âı
0.14
Activations Density 0.022%