INDEX
Explanations
mathematical notation and expressions related to equations and inequalities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
156
+0.32
1.9%
444
+0.14
0.8%
210
+0.13
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
210
+0.32
0.01
444
+0.14
0.01
30
+0.13
0.01
Negative Logits
ãģĹãģ¾ãģĻ
-1.64
ãģķãĤĮ
-1.53
ãģķãĤĮãģŁ
-1.53
MERCHANTABILITY
-1.52
lique
-1.52
ãģĹãģŁ
-1.45
(**
-1.43
completion
-1.42
discovery
-1.40
ãģĹãģ¦ãģĦãĤĭ
-1.39
POSITIVE LOGITS
Ĩ
5.55
ħ
5.45
Ŀ
5.44
Ļª
5.41
Ĵ
5.28
ij
5.27
ļ
5.20
IJ
5.20
¿
5.19
Ń
5.06
Activations Density 0.078%