INDEX
Explanations
references to architectural landmarks and their historical significance
New Auto-Interp
Negative Logits
alian
-0.16
Samar
-0.15
nds
-0.15
abyss
-0.15
291
-0.14
hom
-0.14
ushman
-0.14
ixed
-0.13
wnd
-0.13
anol
-0.13
POSITIVE LOGITS
Complex
0.16
omite
0.16
Complexity
0.16
uplic
0.15
yth
0.15
stone
0.14
ваÑı
0.14
complex
0.14
Ballard
0.14
/site
0.14
Activations Density 0.108%