INDEX
Explanations
instances of the word "variety."
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1921
+0.16
0.6%
871
+0.14
0.5%
1865
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1921
+0.16
0.03
1865
+0.14
0.02
871
+0.13
0.03
Negative Logits
Román
-0.58
foton
-0.58
Belén
-0.57
Áng
-0.56
häm
-0.53
Héctor
-0.53
Kaip
-0.52
Nuorodos
-0.52
Cuen
-0.51
silikon
-0.51
POSITIVE LOGITS
actionTypes
0.93
Variety
0.84
variety
0.84
variety
0.82
varieties
0.81
pymysql
0.78
variation
0.78
vary
0.77
VARI
0.77
variations
0.75
Activations Density 0.076%