INDEX
Explanations
mentions of Arab countries and organizations
occurrences of the term "Arab."
New Auto-Interp
Negative Logits
ertodd
-1.02
uden
-0.78
bilt
-0.77
vg
-0.74
ainer
-0.72
odcast
-0.72
wreck
-0.71
lasses
-0.69
aepernick
-0.69
hov
-0.69
POSITIVE LOGITS
ella
0.91
ophobia
0.86
Sands
0.83
ican
0.83
ians
0.82
iyah
0.80
esque
0.79
League
0.79
ica
0.78
ization
0.78
Activations Density 0.020%