INDEX
Explanations
names of political figures
references to specific individuals, particularly the name "Abdullah" and its variations
New Auto-Interp
Negative Logits
chnology
-0.94
intendent
-0.71
Seah
-0.71
nces
-0.68
mble
-0.66
Californ
-0.63
BOOK
-0.63
UNIVERS
-0.63
ngth
-0.62
Premium
-0.62
POSITIVE LOGITS
doms
1.18
Abdullah
1.05
hammad
0.92
dullah
0.89
ibn
0.88
iyah
0.84
Hassan
0.83
iya
0.82
istani
0.81
edIn
0.78
Activations Density 0.006%