INDEX
Explanations
phrases related to legal proceedings and decision-making
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
678
+0.10
0.3%
599
+0.08
0.2%
453
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1208
+0.10
0.03
1831
+0.08
0.02
734
+0.07
0.03
Negative Logits
inconce
-1.66
disagre
-1.60
emphat
-1.58
indestru
-1.57
impra
-1.55
reluct
-1.54
suscep
-1.54
perfet
-1.52
maneu
-1.52
increa
-1.51
POSITIVE LOGITS
soon
0.78
announcement
0.77
Cyfarwyddwr
0.73
@[+][
0.71
>=",
0.70
mphony
0.69
Personensuche
0.69
announcements
0.69
announce
0.69
ujednoznacz
0.68
Activations Density 0.316%