INDEX
Explanations
references to databases and resources for literary and academic articles
New Auto-Interp
Negative Logits
klad
-0.17
llib
-0.16
ubbo
-0.15
rell
-0.14
hers
-0.14
afil
-0.14
textbook
-0.14
oÄŁlu
-0.14
orsi
-0.14
orf
-0.14
POSITIVE LOGITS
Coverage
0.17
ivals
0.17
oli
0.15
alan
0.15
coverage
0.15
rawn
0.15
olen
0.14
Western
0.14
Austral
0.14
Strength
0.14
Activations Density 0.018%