INDEX
Explanations
references to nuclear weapons and the Iran nuclear program
New Auto-Interp
Negative Logits
oles
-0.16
dt
-0.14
lier
-0.14
dun
-0.14
erland
-0.14
دÙĨ
-0.14
esco
-0.14
inne
-0.14
dn
-0.14
isObject
-0.14
POSITIVE LOGITS
.synthetic
0.15
Herm
0.15
sdk
0.14
qh
0.14
ItemImage
0.14
Personnel
0.14
quivo
0.14
_CARD
0.14
asic
0.14
utomation
0.13
Activations Density 0.003%