INDEX
Explanations
words related to news reporting and current events
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
572
+0.09
0.3%
227
+0.09
0.2%
1445
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
572
+0.09
0.03
339
+0.09
0.03
656
+0.07
0.04
Negative Logits
Subtype
-0.49
Aucune
-0.48
.='
-0.46
Longueur
-0.46
sub
-0.43
.'</
-0.43
inemann
-0.43
=$("#-0.43
Toutefois
-0.42
basic
-0.42
POSITIVE LOGITS
others
1.08
others
0.92
countless
0.85
other
0.84
Others
0.84
OTHERS
0.83
Others
0.82
many
0.82
thousands
0.75
millions
0.75
Activations Density 0.366%