INDEX
Explanations
phrases that indicate the presence of armed individuals or groups
New Auto-Interp
Negative Logits
usz
-0.08
atcher
-0.07
ople
-0.07
á»iji
-0.07
us
-0.07
adem
-0.07
eker
-0.07
er
-0.07
izer
-0.07
iates
-0.07
POSITIVE LOGITS
lord
0.06
247
0.06
McDon
0.06
ament
0.06
illos
0.06
fone
0.06
-for
0.06
/software
0.06
éijij
0.05
McCl
0.05
Activations Density 0.006%