INDEX
Explanations
descriptions of items and situations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
872
+0.08
0.2%
1535
+0.08
0.2%
1055
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
458
+0.08
0.05
1882
+0.08
0.04
2032
+0.08
0.03
Negative Logits
goTo
-0.53
ecru
-0.52
ughter
-0.51
getTarget
-0.51
getCity
-0.50
addItem
-0.50
getAddress
-0.50
getFirst
-0.50
minValue
-0.49
itemList
-0.49
POSITIVE LOGITS
nasel
0.60
potrivit
0.60
primit
0.60
onViewCreated
0.59
destul
0.59
Autoritní
0.58
herre
0.56
strona
0.55
SUDOC
0.55
información
0.54
Activations Density 0.334%