INDEX
Explanations
numerical values or references that relate to legal cases and their citations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.17
0.8%
1964
+0.13
0.6%
1575
+0.12
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1343
+0.17
0.05
667
+0.13
0.04
1978
+0.12
0.04
Negative Logits
<bos>
-1.56
Poznám
-0.65
Fordítás
-0.63
дописавши
-0.61
<?
-0.58
solicited
-0.56
userInput
-0.54
yfikacja
-0.54
benhavn
-0.54
AssemblyProduct
-0.54
POSITIVE LOGITS
muna
1.01
guma
0.97
jaya
0.95
jati
0.88
bayan
0.87
alip
0.85
hina
0.85
silang
0.83
kasa
0.83
kani
0.81
Activations Density 0.087%