INDEX
Explanations
French articles and prepositions indicating location or possession
New Auto-Interp
Negative Logits
uplic
-0.15
eld
-0.14
peri
-0.14
алÑİ
-0.14
ekli
-0.14
eki
-0.14
_ARG
-0.14
furt
-0.13
sublic
-0.13
ulen
-0.13
POSITIVE LOGITS
itre
0.16
ore
0.15
ży
0.15
imit
0.15
same
0.15
likes
0.15
undle
0.14
ivé
0.14
following
0.14
lo
0.14
Activations Density 0.068%