INDEX
Explanations
contact information and sales-related phrases
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.22
0.9%
381
+0.08
0.3%
1150
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1426
+0.22
0.09
1995
+0.08
0.06
1434
+0.08
0.06
Negative Logits
<bos>
-2.71
-0.89
ⓧ
-0.86
/**
-0.83
/*
-0.76
<?
-0.74
<?
-0.57
#![
-0.56
HasColumnType
-0.54
themselves
-0.54
POSITIVE LOGITS
ananas
0.96
siyah
0.93
nuoc
0.93
pylab
0.92
tyn
0.92
pymysql
0.90
maneu
0.90
thuy
0.90
alpes
0.88
gmbh
0.88
Activations Density 1.716%