INDEX
Explanations
geographic locations and place names
New Auto-Interp
Negative Logits
ova
-0.16
uran
-0.15
ung
-0.15
опаÑģ
-0.15
804
-0.15
apan
-0.14
flip
-0.14
cka
-0.14
chem
-0.14
ape
-0.14
POSITIVE LOGITS
.portal
0.16
umlu
0.15
ernal
0.15
Syn
0.14
essler
0.14
ghi
0.14
apas
0.14
.syn
0.14
ácil
0.14
Barton
0.13
Activations Density 0.364%