INDEX
Explanations
information related to technical specifications or requirements
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1052
+0.12
0.4%
382
+0.10
0.3%
1343
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
382
+0.12
0.03
1052
+0.10
0.02
415
+0.10
0.02
Negative Logits
erad
-0.76
Tole
-0.73
Knud
-0.73
porno
-0.72
depic
-0.72
Bartholo
-0.71
Rine
-0.71
sputnik
-0.70
Hn
-0.70
Gorb
-0.70
POSITIVE LOGITS
/+
0.69
/-
0.64
=+
0.61
+
0.60
+%
0.57
ruly
0.57
|+
0.57
＋
0.55
}+
0.55
plus
0.55
Activations Density 0.081%