INDEX
Explanations
words related to weightlifting and physical exercise
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.14
0.5%
1385
+0.12
0.4%
1870
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1385
+0.14
0.07
1705
+0.12
0.07
1380
+0.10
0.01
Negative Logits
<bos>
-0.53
Atsauces
-0.51
Unmarshaller
-0.50
apont
-0.47
RunAsync
-0.47
idosis
-0.45
SequentialGroup
-0.44
atualizar
-0.44
TagMode
-0.43
Definición
-0.43
POSITIVE LOGITS
swarovski
1.30
bordeaux
1.17
ecru
1.16
wien
1.16
oleo
1.12
broderie
1.12
tricot
1.11
ananas
1.10
matel
1.10
effe
1.10
Activations Density 1.527%