INDEX
Explanations
phrases related to economic and political events and discussions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1445
+0.10
0.3%
344
+0.08
0.2%
946
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1068
+0.10
0.05
1677
+0.08
0.04
1445
+0.08
0.04
Negative Logits
bandai
-0.71
trouva
-0.65
tricot
-0.64
thermomix
-0.61
fordable
-0.61
aquare
-0.58
loroethene
-0.58
Roskov
-0.56
artamento
-0.56
cellence
-0.55
POSITIVE LOGITS
Izvori
0.61
Manbalar
0.59
Referencoj
0.57
critics
0.56
others
0.56
livré
0.55
panneau
0.53
typique
0.51
Sklici
0.51
cassert
0.51
Activations Density 0.202%