INDEX
Explanations
terms related to sports teams, competition, and divisions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1842
+0.27
1.1%
50
+0.23
0.9%
1967
+0.15
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1842
+0.27
0.06
1385
+0.23
0.08
1984
+0.15
0.08
Negative Logits
<bos>
-2.73
ⓧ
-0.84
"..\..\..\
-0.76
<?
-0.73
Aholisi
-0.71
GEBURTSDATUM
-0.70
Referències
-0.69
-0.68
//});
-0.67
createState
-0.66
POSITIVE LOGITS
wien
1.51
stockholm
1.48
maroc
1.48
madonna
1.47
affor
1.46
eiffel
1.46
lamborghini
1.31
frankfurt
1.30
strick
1.29
chrysler
1.28
Activations Density 0.937%