INDEX
Explanations
mentions of specific names related to North Korean politics
references to Kim Jong Un and related political figures or events
New Auto-Interp
Negative Logits
ANT
-0.70
rawdownloadcloneembedreportprint
-0.69
FANTASY
-0.67
ICAN
-0.67
Proced
-0.67
Results
-0.66
ATTLE
-0.66
BACK
-0.66
HER
-0.66
RGB
-0.65
POSITIVE LOGITS
sung
1.02
Jong
0.96
ongyang
0.94
Suk
0.90
enei
0.85
unta
0.83
jong
0.80
lasses
0.80
resy
0.80
Il
0.79
Activations Density 0.027%