INDEX
Explanations
references to locations and geographical directions
New Auto-Interp
Negative Logits
itat
-0.16
599
-0.15
iec
-0.15
oni
-0.15
itous
-0.15
ni
-0.15
adir
-0.14
uti
-0.14
isti
-0.14
ilha
-0.14
POSITIVE LOGITS
eyn
0.16
Eag
0.16
leck
0.15
Banco
0.14
cobra
0.14
962
0.14
alus
0.14
anker
0.14
_UNKNOWN
0.14
GLOSS
0.14
Activations Density 0.365%