INDEX
Explanations
references to the French culture or items associated with French heritage
New Auto-Interp
Negative Logits
rame
-0.17
emean
-0.16
jang
-0.15
.osgi
-0.14
Õ¡
-0.13
reamble
-0.13
ÙģØ§Øª
-0.13
eid
-0.13
swagen
-0.13
ismu
-0.13
POSITIVE LOGITS
ily
0.18
EXEMPLARY
0.15
spe
0.14
Greene
0.14
born
0.14
zel
0.14
BO
0.14
es
0.14
nyder
0.14
ing
0.13
Activations Density 0.009%