INDEX
Explanations
technical terminology related to software development and specific tools or applications
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1870
+0.11
0.4%
50
+0.09
0.3%
468
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2010
+0.11
0.07
1380
+0.09
0.03
264
+0.08
0.04
Negative Logits
rè
-0.74
Bourgoin
-0.72
dè
-0.71
jaro
-0.70
ordina
-0.69
WQS
-0.67
elit
-0.65
laude
-0.64
puc
-0.64
lele
-0.64
POSITIVE LOGITS
pamph
0.90
disagre
0.78
shenan
0.77
unspeak
0.73
disreg
0.72
ineffec
0.69
exagger
0.68
maneu
0.67
apprehen
0.67
unwarran
0.67
Activations Density 1.014%