INDEX
Explanations
strings related to computer code
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.19
1.2%
1870
+0.11
0.7%
90
+0.10
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
90
+0.19
0.03
1296
+0.11
0.03
281
+0.10
0.03
Negative Logits
<bos>
-3.29
ⓧ
-0.86
<?
-0.80
-0.80
/***
-0.78
/**
-0.77
springfox
-0.72
/*++
-0.67
<?
-0.63
Kontrola
-0.61
POSITIVE LOGITS
Code
1.18
code
1.17
Code
1.10
code
1.09
CODE
1.08
codes
1.07
maroc
1.05
getCode
1.04
CODE
1.04
Codes
1.03
Activations Density 0.060%