INDEX
Explanations
names of a specific individual, possibly in a legal or controversial context
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1961
+0.17
0.7%
1276
+0.15
0.7%
871
+0.15
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1276
+0.17
0.03
1961
+0.15
0.03
981
+0.15
0.03
Negative Logits
Pringle
-0.56
Dunlap
-0.55
McLaugh
-0.54
Beauchamp
-0.51
Burnet
-0.49
McInt
-0.49
Maloney
-0.48
McCarty
-0.48
Moulton
-0.48
Kinney
-0.46
POSITIVE LOGITS
Howard
1.55
Howard
1.46
HOWARD
1.22
Portail
0.80
Février
0.79
Décembre
0.72
pavillon
0.67
Hô
0.67
Août
0.65
Novembre
0.65
Activations Density 0.172%