INDEX
Explanations
references to specific architects and their works
New Auto-Interp
Negative Logits
forder
-0.15
orthand
-0.15
uale
-0.14
tim
-0.14
Basement
-0.14
ocket
-0.14
RG
-0.13
olicy
-0.13
ollision
-0.13
Cute
-0.13
POSITIVE LOGITS
Modern
0.22
Bau
0.20
Brut
0.19
Wright
0.19
Modern
0.18
Designed
0.18
Moder
0.17
Falling
0.17
brutal
0.17
Designed
0.17
Activations Density 0.059%