INDEX
Explanations
references to tobacco and related products
New Auto-Interp
Negative Logits
asma
-0.17
derec
-0.15
unci
-0.15
ulings
-0.15
bul
-0.14
:CGRect
-0.14
smoking
-0.14
-flash
-0.14
blo
-0.14
acente
-0.14
POSITIVE LOGITS
harm
0.16
antz
0.15
pouch
0.15
Nic
0.15
chewing
0.15
leaf
0.15
chew
0.15
nit
0.15
Giov
0.14
graphic
0.14
Activations Density 0.006%