INDEX
Explanations
government or legal terms and phrases
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1499
+0.18
0.6%
184
+0.16
0.5%
1842
+0.16
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1499
+0.18
0.06
184
+0.16
0.03
344
+0.16
0.02
Negative Logits
swarovski
-1.54
tricot
-1.42
embodi
-1.36
bonbons
-1.33
fluo
-1.31
indestru
-1.29
bordeaux
-1.29
nutella
-1.28
broderie
-1.23
strass
-1.22
POSITIVE LOGITS
Soorten
0.70
Aiheesta
0.65
reportedly
0.65
IsContent
0.64
detallada
0.62
rrggbb
0.62
intptr
0.62
Cyfarwyddwr
0.62
sumpay
0.61
ImageContext
0.59
Activations Density 0.231%