INDEX
Explanations
locations or addresses, such as street names and cities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.14
0.5%
481
+0.07
0.3%
1092
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1487
+0.14
0.02
1498
+0.07
0.02
702
+0.07
0.02
Negative Logits
<bos>
-1.97
///**
-0.58
ždý
-0.56
intios
-0.56
},[])
-0.55
glBind
-0.52
/***
-0.52
lateinit
-0.52
//{
-0.51
TestBed
-0.50
POSITIVE LOGITS
affor
1.08
stockholm
1.05
perfet
1.02
maneu
1.00
nece
0.96
guarante
0.96
budapest
0.95
effe
0.93
fortn
0.93
nomine
0.92
Activations Density 0.030%