INDEX
Explanations
names of individuals and institutions related to legal or organizational matters
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
137
+0.13
0.4%
1177
+0.12
0.4%
453
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
137
+0.13
0.05
981
+0.12
0.05
227
+0.11
0.05
Negative Logits
<bos>
-0.98
Paglinawan
-0.66
propOrder
-0.61
Administrativna
-0.59
Италијани
-0.57
jubile
-0.57
republi
-0.54
AssemblyCompany
-0.54
disambiguazione
-0.54
rosserie
-0.53
POSITIVE LOGITS
unsplash
0.80
considér
0.72
shutterstock
0.72
embodi
0.72
fabriqué
0.71
contex
0.71
compréhen
0.71
Cfr
0.70
Illus
0.70
Gorb
0.70
Activations Density 0.222%