INDEX
Explanations
terms related to a specific brand, particularly Intel
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
204
+0.16
0.6%
1404
+0.14
0.5%
1323
+0.14
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
204
+0.16
0.02
1404
+0.14
0.02
1103
+0.14
0.02
Negative Logits
unspeak
-1.36
gaily
-1.26
vainly
-1.26
apprehen
-1.25
unwarran
-1.21
effectually
-1.19
nobly
-1.18
disagre
-1.17
ingrat
-1.16
tolerably
-1.16
POSITIVE LOGITS
Intel
1.54
Intel
1.52
intel
1.25
intel
1.14
INTEL
0.95
INTEL
0.94
Intelligence
0.87
Intelligence
0.81
intelligence
0.79
INTELL
0.77
Activations Density 0.077%