INDEX
Explanations
Slavic names or words with specific emphasis patterns
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1967
+0.18
0.6%
1343
+0.17
0.6%
1839
+0.14
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
227
+0.18
0.09
981
+0.17
0.09
394
+0.14
0.06
Negative Logits
affez
-1.01
tupperware
-0.94
motorola
-0.90
philips
-0.90
fluo
-0.89
scrat
-0.89
gopro
-0.85
panasonic
-0.84
logitech
-0.83
eyel
-0.83
POSITIVE LOGITS
cs
0.50
EMBLY
0.47
ss
0.47
hash
0.46
see
0.45
uebas
0.45
ACTERS
0.45
us
0.44
örgy
0.44
Galería
0.43
Activations Density 0.473%