INDEX
Explanations
information related to news articles, editorials, and political events
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1577
+0.32
1.2%
394
+0.24
0.9%
453
+0.16
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1499
+0.32
0.19
453
+0.24
0.17
1048
+0.16
0.02
Negative Logits
hairc
-1.32
ecru
-1.23
tupperware
-1.23
cushi
-1.15
swarovski
-1.06
Whence
-1.04
unspeak
-1.03
embodi
-1.03
tolerably
-1.02
gaily
-0.97
POSITIVE LOGITS
ویکیپدی
0.59
Viitteet
0.58
smithy
0.54
Palmar
0.53
├
0.51
Географиясе
0.51
Erreferentziak
0.50
RenderAtEndOf
0.50
Solución
0.50
+#+#
0.49
Activations Density 5.166%