INDEX
Explanations
references to ancient civilizations and their cultural or historical significance
New Auto-Interp
Negative Logits
aps
-0.14
.ng
-0.13
nic
-0.13
rezent
-0.13
ahlen
-0.13
kees
-0.13
719
-0.13
aksi
-0.13
otten
-0.13
getter
-0.13
POSITIVE LOGITS
/original
0.18
usan
0.15
-old
0.15
andalone
0.15
imes
0.14
arges
0.14
ly
0.14
-language
0.14
SPA
0.14
/current
0.13
Activations Density 0.029%