INDEX
Explanations
code snippets involving method chaining and function callbacks
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1343
+0.17
0.5%
1577
+0.12
0.4%
876
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1678
+0.17
0.03
597
+0.12
0.02
210
+0.11
0.03
Negative Logits
<bos>
-0.62
osoba
-0.58
Contribu
-0.56
Eksterne
-0.55
Perú
-0.54
Estat
-0.53
Trat
-0.53
fromnode
-0.50
Precis
-0.49
Atra
-0.48
POSITIVE LOGITS
cannes
1.08
thermomix
1.07
doman
1.05
lende
1.05
provence
1.05
bordeaux
1.04
milano
1.03
quarelle
1.03
bandung
1.03
vinci
1.03
Activations Density 0.125%