INDEX
Explanations
mentions and descriptions of dresses
references to dresses and dress codes
New Auto-Interp
Negative Logits
ntil
-0.82
ocalyptic
-0.73
è¦ļéĨĴ
-0.71
untu
-0.71
uilt
-0.67
raltar
-0.65
interrupted
-0.63
irlf
-0.63
emonic
-0.63
ategory
-0.62
POSITIVE LOGITS
maker
1.07
makers
1.01
glers
0.95
rehearsal
0.91
gown
0.90
ings
0.89
dresses
0.88
cases
0.87
bag
0.86
shoes
0.85
Activations Density 0.010%