INDEX
Explanations
formal titles and greetings
New Auto-Interp
Negative Logits
걔
0.50
っぽい
0.46
grunge
0.45
みんな
0.43
killer
0.43
ゲー
0.42
kids
0.42
funky
0.41
vibe
0.41
nerd
0.41
POSITIVE LOGITS
sir
1.88
madam
1.73
monsieur
1.73
Madam
1.66
先生
1.66
госпо
1.63
Sir
1.62
senhor
1.61
Sir
1.59
gentlemen
1.59
Activations Density 0.109%