INDEX
Explanations
references to socio-economic classes, particularly the middle class
New Auto-Interp
Negative Logits
radi
-0.18
ide
-0.18
rapped
-0.17
tiv
-0.16
ieren
-0.15
epad
-0.15
xac
-0.15
leo
-0.15
ptime
-0.15
tsky
-0.15
POSITIVE LOGITS
-aged
0.31
sex
0.25
aged
0.24
Ages
0.24
SEX
0.23
wares
0.22
weight
0.22
tons
0.21
finger
0.21
bury
0.21
Activations Density 0.018%