INDEX
Explanations
musical genres and their characteristics
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.23
0.9%
1870
+0.16
0.6%
1013
+0.10
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1870
+0.23
0.04
678
+0.16
0.05
74
+0.10
0.04
Negative Logits
<bos>
-2.28
principalTable
-0.95
principalColumn
-0.93
Paglinawan
-0.93
GeoNames
-0.88
censiti
-0.87
Wikimédia
-0.85
HasIndex
-0.85
consultato
-0.85
***!
-0.83
POSITIVE LOGITS
increa
2.01
affor
1.92
impra
1.92
maneu
1.85
reluct
1.82
Juf
1.79
emphat
1.78
unspeak
1.78
unden
1.77
disagre
1.77
Activations Density 0.321%