INDEX
Explanations
technical instructions and terms related to devices and technology
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1842
+0.15
0.5%
1819
+0.13
0.4%
845
+0.12
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1819
+0.15
0.10
845
+0.13
0.06
1870
+0.12
0.04
Negative Logits
AndEndTag
-0.65
kaynağından
-0.64
ftagPool
-0.60
клопе
-0.59
aarrggbb
-0.57
שוליים
-0.56
surla
-0.56
Espèce
-0.55
Савезне
-0.54
Према
-0.54
POSITIVE LOGITS
disagre
1.52
encomp
1.46
inev
1.43
shenan
1.42
increa
1.38
reluct
1.38
uninten
1.36
accla
1.35
depic
1.33
excru
1.32
Activations Density 0.915%