INDEX
Explanations
information related to technical support, copyright, and company contact details
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.21
1.0%
227
+0.09
0.4%
2034
+0.09
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
227
+0.21
0.06
460
+0.09
0.04
137
+0.09
0.04
Negative Logits
<bos>
-3.25
/***
-0.71
<>
-0.65
<?
-0.64
//};
-0.64
وتسجيلات
-0.64
propOrder
-0.64
StoreMessageInfo
-0.63
disbur
-0.63
bewerken
-0.61
POSITIVE LOGITS
véhic
1.05
délib
0.98
pleins
0.90
fameux
0.89
soulign
0.88
Traité
0.85
considér
0.85
↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
0.85
déri
0.84
soigneusement
0.84
Activations Density 0.178%