INDEX
Explanations
information about news, events, and community updates
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.24
0.9%
1343
+0.13
0.5%
690
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
924
+0.24
0.12
1403
+0.13
0.06
684
+0.08
0.08
Negative Logits
<bos>
-2.44
/*
-0.90
/**
-0.88
<?
-0.86
propOrder
-0.80
Walkover
-0.71
новништво
-0.68
GeoNames
-0.67
ⓧ
-0.66
lateinit
-0.66
POSITIVE LOGITS
affor
1.95
impra
1.76
maneu
1.74
unspeak
1.70
practition
1.69
resear
1.68
accla
1.67
volunte
1.66
philanth
1.63
reluct
1.60
Activations Density 1.613%