INDEX
Explanations
information related to historical figures and events, possibly with a focus on political relationships and developments
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.31
1.4%
752
+0.10
0.4%
161
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
752
+0.31
0.06
687
+0.10
0.06
1654
+0.08
0.04
Negative Logits
<bos>
-3.38
ⓧ
-0.82
/**
-0.74
///**
-0.69
viewDidLoad
-0.66
PropertyGroup
-0.64
lateinit
-0.63
移
-0.62
moved
-0.62
<?
-0.62
POSITIVE LOGITS
Juf
1.69
délib
1.50
lele
1.48
affor
1.42
stockholm
1.42
bandung
1.41
Minang
1.39
pleins
1.37
increa
1.37
maneu
1.37
Activations Density 0.752%