INDEX
Explanations
work-related terms or information
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1967
+0.21
0.7%
453
+0.16
0.5%
1870
+0.13
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
453
+0.21
0.05
1618
+0.16
0.03
736
+0.13
0.04
Negative Logits
transQ
-0.59
жалуйста
-0.57
этому
-0.56
asgi
-0.55
cloudf
-0.55
SharedDtor
-0.55
WithIOException
-0.54
Демографія
-0.54
gruß
-0.53
CoreData
-0.52
POSITIVE LOGITS
keramik
1.27
antik
1.26
silikon
1.26
alkoh
1.24
panik
1.24
kafe
1.22
ekos
1.18
optik
1.16
teras
1.15
kosme
1.14
Activations Density 0.179%