INDEX
Explanations
names of companies or products related to audio and technology
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
32
+0.18
1.0%
370
+0.14
0.8%
1296
+0.13
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1194
+0.18
0.04
1622
+0.14
0.03
1575
+0.13
0.03
Negative Logits
<bos>
-1.91
guma
-0.91
maging
-0.83
kahit
-0.80
akong
-0.78
itong
-0.77
maraming
-0.73
siyang
-0.70
buhay
-0.69
hanggang
-0.69
POSITIVE LOGITS
Na
1.48
naï
1.36
napoli
1.36
Châ
1.32
Bartholo
1.31
casio
1.29
accla
1.27
madonna
1.26
Na
1.26
suspic
1.26
Activations Density 0.264%