INDEX
Explanations
mentions of the keyword "face"
references to facial recognition or facial appearance
New Auto-Interp
Negative Logits
Serv
-0.65
distant
-0.64
red
-0.63
outgoing
-0.62
contro
-0.62
reliable
-0.62
aku
-0.61
Wright
-0.60
central
-0.60
rie
-0.60
POSITIVE LOGITS
face
1.38
faces
1.12
lihood
1.11
faced
0.98
liest
0.93
Face
0.91
nces
0.89
ername
0.86
xual
0.84
BOOK
0.84
Activations Density 0.011%