INDEX
Explanations
information related to health risks, specifically regarding processed meat consumption and its comparison to the risk of smoking
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1253
+0.14
0.4%
764
+0.09
0.3%
1443
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
41
+0.14
0.05
1788
+0.09
0.03
183
+0.08
0.02
Negative Logits
%";
-0.45
romptu
-0.44
rajz
-0.43
Viited
-0.42
$#
-0.42
frescoes
-0.41
Биография
-0.41
geschreven
-0.41
Vanjske
-0.41
PreExecute
-0.41
POSITIVE LOGITS
tille
0.89
makro
0.84
scrat
0.84
eyel
0.83
drap
0.81
mef
0.80
increa
0.79
ciga
0.79
plak
0.79
alip
0.77
Activations Density 0.368%