INDEX
Explanations
words related to political figures or leaders
references to the Kim Jong family
New Auto-Interp
Negative Logits
rawdownloadcloneembedreportprint
-0.80
ATTLE
-0.72
BACK
-0.71
WORK
-0.71
Proced
-0.68
ENC
-0.68
Results
-0.68
ICAN
-0.68
RGB
-0.67
HER
-0.65
POSITIVE LOGITS
ongyang
1.04
sung
0.95
Jong
0.88
lasses
0.88
stress
0.81
ishi
0.80
jri
0.79
Suk
0.78
enei
0.77
etsu
0.76
Activations Density 0.018%