INDEX
Explanations
words related to specific phonetic sequences or patterns
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1677
+0.22
0.9%
1068
+0.17
0.7%
1328
+0.17
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1677
+0.22
0.07
981
+0.17
0.07
662
+0.17
0.06
Negative Logits
ärna
-0.50
ürzen
-0.50
rales
-0.43
برابوك
-0.42
zköz
-0.42
tarko
-0.42
päivä
-0.41
ända
-0.41
#--
-0.40
etmeye
-0.40
POSITIVE LOGITS
Kot
0.78
Sot
0.76
Kot
0.75
protokol
0.75
akut
0.73
Strukt
0.71
kosme
0.70
kot
0.69
keramik
0.69
SOT
0.68
Activations Density 0.397%