INDEX
Explanations
biographical facts related to a specific person
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
2034
+0.16
0.5%
1699
+0.13
0.4%
906
+0.12
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1716
+0.16
0.02
1966
+0.13
0.01
1737
+0.12
0.01
Negative Logits
Ży
-0.67
Przyp
-0.66
Dlaczego
-0.61
Jakie
-0.61
Czym
-0.61
Hermoso
-0.60
Kto
-0.60
Flere
-0.60
scienced
-0.59
pandémie
-0.58
POSITIVE LOGITS
maroc
0.92
thuy
0.88
toscana
0.88
vinci
0.86
jaya
0.86
guir
0.85
peluche
0.85
bandung
0.84
ventus
0.84
torba
0.84
Activations Density 0.029%