INDEX
Explanations
mentions of clothing items
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
964
+0.11
0.3%
1547
+0.10
0.3%
1385
+0.09
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1547
+0.11
0.04
736
+0.10
0.06
1018
+0.09
0.03
Negative Logits
bandeau
-0.85
hairc
-0.74
allarg
-0.74
autorytatywna
-0.71
legging
-0.70
ecru
-0.68
hoody
-0.67
notor
-0.67
Atsauces
-0.67
burbu
-0.66
POSITIVE LOGITS
wearing
0.74
wear
0.66
worn
0.65
dress
0.64
wears
0.62
Wearing
0.62
attire
0.62
costume
0.61
wearable
0.60
wearing
0.60
Activations Density 0.615%