INDEX
Explanations
mentions of the color purple
references to the color purple and related concepts
New Auto-Interp
Negative Logits
elong
-0.82
Afee
-0.80
yer
-0.78
ographed
-0.78
opic
-0.76
ablished
-0.76
76561
-0.76
INESS
-0.74
rina
-0.73
otomy
-0.72
POSITIVE LOGITS
velvet
0.81
ink
0.75
Kush
0.71
veyard
0.71
prose
0.70
Purple
0.68
ande
0.68
pill
0.67
shif
0.66
tail
0.65
Activations Density 0.034%