INDEX
Explanations
numbers combined with specific legal or bureaucratic terms
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.20
0.6%
1967
+0.14
0.4%
1343
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1363
+0.20
0.05
678
+0.14
0.05
1710
+0.10
0.05
Negative Logits
управління
-0.52
elite
-0.51
katkı
-0.50
facteur
-0.49
sw
-0.49
lüğü
-0.49
Profesor
-0.49
rmse
-0.48
estaw
-0.48
takimi
-0.48
POSITIVE LOGITS
solidar
1.35
cyr
1.34
dispen
1.33
immen
1.32
abbra
1.32
lapto
1.30
utop
1.30
reger
1.29
emble
1.28
ristor
1.27
Activations Density 0.208%