INDEX
Explanations
entities related to artistic creation and cultural expression
New Auto-Interp
Negative Logits
macho
-0.61
BoxFit
-0.60
sexist
-0.57
potty
-0.55
bulldog
-0.55
sexism
-0.54
Kombat
-0.54
Bully
-0.53
Nasty
-0.52
disgusting
-0.52
POSITIVE LOGITS
sunset
0.81
moonlight
0.78
dusk
0.77
autumn
0.74
twilight
0.73
Autumn
0.73
wind
0.72
moon
0.71
clouds
0.71
Dusk
0.69
Activations Density 0.423%