INDEX
Explanations
references to fairy tales and princesses
references to fairytales and princesses
New Auto-Interp
Negative Logits
paio
-0.79
ahan
-0.79
iago
-0.73
sych
-0.72
oday
-0.71
enegger
-0.71
icion
-0.70
emporary
-0.69
roit
-0.69
bsp
-0.69
POSITIVE LOGITS
princess
1.26
Princess
1.15
Sparkle
1.06
tale
1.05
Celest
1.04
Elsa
1.03
Leia
1.03
Elsa
0.98
Bride
0.95
Belle
0.93
Activations Density 0.082%