INDEX
Explanations
dates or time-related phrases
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.24
0.9%
1806
+0.09
0.3%
281
+0.07
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1869
+0.24
0.07
175
+0.09
0.08
927
+0.07
0.06
Negative Logits
<bos>
-2.09
ⓧ
-0.94
uesia
-0.63
/**
-0.62
/*
-0.61
-0.61
<?
-0.61
protoimpl
-0.59
<?
-0.58
queryInterface
-0.58
POSITIVE LOGITS
maneu
1.32
Minang
1.31
seksi
1.31
saar
1.27
chrysler
1.27
maksi
1.27
bandung
1.26
kafe
1.25
Meksi
1.22
toledo
1.21
Activations Density 0.725%