INDEX
Explanations
words related to fashion
references to fashion
New Auto-Interp
Negative Logits
lehem
-0.73
escription
-0.71
ramid
-0.69
minster
-0.68
venants
-0.67
nikov
-0.67
20439
-0.66
ilver
-0.65
yip
-0.65
ulhu
-0.65
POSITIVE LOGITS
ably
1.06
ables
0.90
fashion
0.86
apparel
0.85
Fashion
0.82
istas
0.78
ista
0.78
fashion
0.77
boutique
0.77
clothing
0.76
Activations Density 0.022%