INDEX
Explanations
instances of existential constructs indicating existence or presence
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.35
1.5%
381
+0.18
0.7%
2019
+0.11
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
381
+0.35
0.07
573
+0.18
0.08
289
+0.11
0.06
Negative Logits
<bos>
-2.79
/***
-0.68
/**
-0.68
///**
-0.67
/*!
-0.58
//};
-0.56
demografica
-0.53
dè
-0.52
rousel
-0.52
//...
-0.51
POSITIVE LOGITS
accla
0.99
maneu
0.99
reluct
0.97
increa
0.96
affor
0.95
impractica
0.95
fortn
0.93
inev
0.91
vry
0.90
Sklici
0.90
Activations Density 0.823%