INDEX
Explanations
mentions of Afghanistan and related terms
New Auto-Interp
Negative Logits
ery
-0.66
Browne
-0.64
Bartlett
-0.63
Parry
-0.60
Divinity
-0.58
pire
-0.57
ceta
-0.57
Drummond
-0.57
căng
-0.57
ester
-0.57
POSITIVE LOGITS
Afghanistan
1.66
Afghanistan
1.56
Afghan
1.54
Afghans
1.39
Afghan
1.38
afghan
1.37
Taliban
1.33
ABUL
1.32
Afgan
1.27
Kabul
1.25
Activations Density 0.002%