INDEX
Explanations
technology-related terms and product descriptions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
227
+0.20
0.6%
1870
+0.13
0.4%
1438
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
227
+0.20
0.08
1819
+0.13
0.08
936
+0.11
0.04
Negative Logits
fte
-1.10
Hv
-1.09
daf
-1.07
squa
-1.07
Strö
-1.06
mef
-1.05
secon
-1.03
fep
-1.03
Öster
-1.02
dispen
-1.00
POSITIVE LOGITS
Informações
0.52
featuring
0.50
imiz
0.49
sold
0.49
designed
0.48
Construcción
0.48
manufactured
0.48
marketed
0.47
xae
0.47
series
0.45
Activations Density 0.499%