INDEX
Explanations
names of people and places
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1068
+0.13
0.5%
1350
+0.13
0.5%
406
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
981
+0.13
0.06
406
+0.13
0.03
662
+0.13
0.04
Negative Logits
jandra
-0.57
CppMethod
-0.53
ltä
-0.49
biet
-0.49
nhung
-0.49
toPromise
-0.49
Hasan
-0.48
nguoi
-0.47
ElementRef
-0.47
heo
-0.47
POSITIVE LOGITS
AY
0.89
ay
0.86
Ay
0.85
HAY
0.84
Ay
0.83
AY
0.80
ay
0.79
Hay
0.78
kay
0.77
gay
0.75
Activations Density 0.241%