INDEX
Explanations
names of newspapers and publications
New Auto-Interp
Negative Logits
miss
-0.17
ucch
-0.16
split
-0.15
ç½®
-0.15
bounty
-0.15
accept
-0.14
UGHT
-0.14
anja
-0.14
rapper
-0.14
Accept
-0.14
POSITIVE LOGITS
ctxt
0.18
ipar
0.16
Wein
0.15
templ
0.15
ypress
0.15
æı®
0.14
oved
0.14
plode
0.14
Jamal
0.13
ÑĤеÑĢн
0.13
Activations Density 0.058%