INDEX
Explanations
photographs and images of individuals, such as politicians, royalty, and celebrities
references to photographs or images related to individuals and events
New Auto-Interp
Negative Logits
imester
-0.74
cipled
-0.72
issan
-0.69
izons
-0.69
ItemTracker
-0.68
politics
-0.67
ainer
-0.67
ente
-0.65
Berry
-0.64
Parameter
-0.64
POSITIVE LOGITS
smiling
1.15
grinning
1.07
kneeling
1.02
adorned
0.99
interacting
0.99
behaving
0.94
surrounded
0.94
undergoing
0.93
hugging
0.91
enjoying
0.91
Activations Density 0.490%