INDEX
Explanations
instances of violence and intimidation, particularly in a political context
New Auto-Interp
Negative Logits
ingleton
-0.19
hostages
-0.17
isoner
-0.16
victim
-0.16
657
-0.16
alborg
-0.16
caut
-0.15
podrob
-0.15
ocha
-0.15
unic
-0.15
POSITIVE LOGITS
forces
0.23
groups
0.22
masked
0.22
hired
0.21
gangs
0.20
forces
0.20
bands
0.19
mobs
0.19
bounty
0.19
hood
0.19
Activations Density 0.242%