INDEX
Explanations
mentions of fashion-related terms
mentions of fashion and related concepts
New Auto-Interp
Negative Logits
terness
-0.71
lehem
-0.69
escription
-0.69
yip
-0.68
nen
-0.67
ilver
-0.66
minster
-0.66
ulhu
-0.66
venants
-0.65
nikov
-0.64
POSITIVE LOGITS
ably
1.05
boutique
0.91
ables
0.85
apparel
0.85
istas
0.83
ista
0.79
fashion
0.79
Fashion
0.79
clothing
0.79
faux
0.78
Activations Density 0.035%