INDEX
Explanations
proper nouns, particularly associated with authors and their works
New Auto-Interp
Negative Logits
osate
-0.15
ekil
-0.14
who
-0.14
eca
-0.14
posium
-0.14
icari
-0.13
quina
-0.13
abbo
-0.13
whole
-0.13
ATIC
-0.13
POSITIVE LOGITS
fore
0.18
fore
0.18
et
0.17
alli
0.15
Fore
0.15
interviewed
0.14
(auth
0.13
èģŀ
0.13
jspx
0.13
641
0.13
Activations Density 0.112%