INDEX
Explanations
geographical names and proper nouns
New Auto-Interp
Negative Logits
-League
-0.18
onis
-0.15
ÑģÑĤво
-0.15
agus
-0.15
elow
-0.15
olis
-0.14
olf
-0.14
uset
-0.14
318
-0.14
eman
-0.14
POSITIVE LOGITS
ensis
0.31
-based
0.20
-born
0.19
ská
0.18
iese
0.17
ský
0.17
-region
0.16
δια
0.16
umlu
0.15
ian
0.15
Activations Density 0.236%