INDEX
Explanations
references to a specific individual named "Mousavi"
names of individuals involved in a political or public narrative
New Auto-Interp
Negative Logits
isu
-0.84
ORD
-0.76
2048
-0.74
cession
-0.72
robe
-0.69
âĶĢâĶĢ
-0.69
nee
-0.68
arty
-0.68
Borders
-0.68
ords
-0.68
POSITIVE LOGITS
Mous
1.08
hammad
0.88
Cage
0.86
xual
0.83
eters
0.81
ovych
0.79
pread
0.77
uit
0.75
millenn
0.74
ongh
0.74
Activations Density 0.014%