INDEX
Explanations
financial figures and technical details, possibly related to a specific industry or domain
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
555
+0.20
1.0%
1624
+0.14
0.7%
50
+0.14
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
605
+0.20
0.02
900
+0.14
0.04
1343
+0.14
0.04
Negative Logits
<bos>
-1.97
///**
-0.73
<!--
-0.64
/***
-0.63
<?
-0.63
rehabilitate
-0.63
ratify
-0.63
//{
-0.58
/*!
-0.57
netinet
-0.56
POSITIVE LOGITS
karton
0.91
silikon
0.83
optik
0.80
kristal
0.78
mentale
0.77
+
0.74
keramik
0.73
alkoh
0.73
Heeren
0.71
keram
0.70
Activations Density 0.229%