INDEX
Explanations
names and titles related to a specific political figure, likely Kim Jong-un
mentions of Kim Jong Il and Kim Jong Un
New Auto-Interp
Negative Logits
ATTLE
-0.67
arians
-0.65
Results
-0.61
aneous
-0.61
houses
-0.60
RGB
-0.60
ources
-0.60
ANT
-0.60
Cantor
-0.60
Italians
-0.59
POSITIVE LOGITS
sung
1.06
Jong
1.02
ongyang
0.95
enei
0.92
Sung
0.92
jong
0.91
Nam
0.89
Il
0.88
Suk
0.86
ishi
0.83
Activations Density 0.024%