INDEX
Explanations
mentions of diversity and related terms in various contexts
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
555
+0.13
0.5%
1870
+0.12
0.4%
478
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
555
+0.13
0.02
869
+0.12
0.02
143
+0.11
0.02
Negative Logits
Équipe
-0.59
suon
-0.59
adesso
-0.58
Rodrig
-0.58
sappi
-0.57
Molto
-0.56
riman
-0.55
surpl
-0.54
frambo
-0.54
textil
-0.54
POSITIVE LOGITS
diversity
1.14
Diversity
1.05
Diversity
0.99
diversity
0.97
diverse
0.80
diverse
0.80
tanong
0.79
sarili
0.78
diversify
0.78
maraming
0.76
Activations Density 0.080%