INDEX
Explanations
specific references to clothing items
specific proper nouns, particularly names and brands
New Auto-Interp
Negative Logits
anwhile
-0.82
totality
-0.77
terday
-0.75
quickShipAvailable
-0.69
SPD
-0.64
parity
-0.64
TextColor
-0.64
cknow
-0.63
Vald
-0.63
swers
-0.62
POSITIVE LOGITS
ussian
0.94
uggle
0.93
onian
0.84
ussie
0.83
istine
0.80
ollywood
0.79
adian
0.76
anian
0.75
ilian
0.73
oodle
0.73
Activations Density 0.487%