INDEX
Explanations
themes related to clothing and fashion choices
New Auto-Interp
Negative Logits
мÑı
-0.16
angan
-0.15
amiliar
-0.15
utters
-0.15
ixin
-0.14
prod
-0.14
ype
-0.14
rawer
-0.14
erral
-0.13
.generic
-0.13
POSITIVE LOGITS
wearing
0.27
wear
0.24
outfit
0.23
outfits
0.22
fashion
0.22
worn
0.22
wearable
0.22
wardrobe
0.21
dress
0.21
wears
0.21
Activations Density 0.304%