INDEX
Explanations
statistics and numbers related to various topics and categories
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
776
+0.13
0.4%
1978
+0.12
0.3%
1013
+0.12
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
776
+0.13
0.05
382
+0.12
0.04
415
+0.12
0.03
Negative Logits
intersper
-1.19
depic
-1.07
indescri
-1.00
shenan
-1.00
intrigu
-0.99
sophistic
-0.95
impra
-0.94
unve
-0.94
snoopy
-0.94
reluct
-0.93
POSITIVE LOGITS
karton
0.67
kristal
0.62
kredi
0.62
silikon
0.61
kask
0.61
Banten
0.59
kafe
0.59
hunde
0.58
kado
0.58
%,
0.57
Activations Density 0.144%