INDEX
Explanations
descriptions of clothing and appearances
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1385
+0.17
0.5%
1343
+0.09
0.3%
198
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
736
+0.17
0.06
1188
+0.09
0.02
1821
+0.08
0.04
Negative Logits
bandeau
-1.05
hairc
-0.93
ecru
-0.92
legging
-0.78
hoody
-0.75
Cfr
-0.71
pettico
-0.69
diffusi
-0.68
indiri
-0.66
swarovski
-0.66
POSITIVE LOGITS
wearing
0.57
dress
0.57
attire
0.56
triomphe
0.52
costume
0.52
shoes
0.50
luigi
0.50
zucht
0.50
shirt
0.50
worn
0.49
Activations Density 0.256%