INDEX
Explanations
high-quality images
references to images
New Auto-Interp
Negative Logits
cffff
-0.93
aptic
-0.85
hurst
-0.84
ategory
-0.81
itte
-0.78
aido
-0.78
merce
-0.77
laws
-0.76
odcast
-0.76
odo
-0.75
POSITIVE LOGITS
caption
0.93
galleries
0.93
depicting
0.91
gallery
0.89
gallery
0.87
image
0.86
images
0.86
img
0.85
photographed
0.85
macros
0.81
Activations Density 0.045%