INDEX
Explanations
names of people or characters
New Auto-Interp
Negative Logits
ingleton
-0.17
ertz
-0.17
esign
-0.15
istrovstvÃŃ
-0.15
ulace
-0.14
ichten
-0.14
.glide
-0.14
achuset
-0.14
aukee
-0.14
mojom
-0.14
POSITIVE LOGITS
hottest
0.19
Cum
0.19
Cum
0.16
cum
0.15
Latin
0.15
latin
0.15
_lex
0.14
.
0.14
Coul
0.14
hot
0.14
Activations Density 0.040%