INDEX
Explanations
technical descriptions including mention of technical specifications, dual-channel functionality, and processing units
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
897
+0.13
0.5%
1870
+0.13
0.5%
90
+0.12
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
90
+0.13
0.02
1328
+0.13
0.02
143
+0.12
0.02
Negative Logits
رشف
-0.45
bulunabilir
-0.45
ElementRef
-0.42
TType
-0.41
梗
-0.40
Shirts
-0.40
chiff
-0.39
inence
-0.39
볍
-0.39
Camp
-0.39
POSITIVE LOGITS
Dual
1.12
Dual
1.07
dual
1.07
dual
0.99
DUAL
0.93
pymysql
0.91
twin
0.86
alberto
0.80
conftest
0.79
twin
0.77
Activations Density 0.071%