INDEX
Explanations
mentions of the name "Barack Obama."
New Auto-Interp
Negative Logits
rray
-0.16
egov
-0.16
soft
-0.15
Trick
-0.15
orrent
-0.14
اÙĦبر
-0.14
Gen
-0.14
Všech
-0.14
izontal
-0.14
ادÙĦ
-0.14
POSITIVE LOGITS
Hussein
0.20
abella
0.16
Huss
0.15
hus
0.15
μι
0.15
igner
0.14
Moss
0.14
owers
0.14
ymology
0.14
romo
0.14
Activations Density 0.009%