INDEX
Explanations
technical terms related to engineering, technology, and internet concepts, as well as meta tags
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
32
+0.22
1.0%
555
+0.15
0.7%
506
+0.12
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
32
+0.22
0.03
1896
+0.15
0.02
568
+0.12
0.02
Negative Logits
<bos>
-1.31
***!
-0.67
addCriterion
-0.64
uxxxx
-0.52
RTSN
-0.52
Havolalar
-0.52
andaag
-0.51
Carcinogenicity
-0.51
keyColumn
-0.50
Савезне
-0.50
POSITIVE LOGITS
Meta
1.61
meta
1.53
Meta
1.44
meta
1.39
méta
1.36
META
1.32
metast
1.18
metadata
1.16
Metadata
1.15
metas
1.04
Activations Density 0.252%