INDEX
Explanations
words related to medical conditions and vaccinations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
394
+0.44
1.8%
1842
+0.25
1.0%
163
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
394
+0.44
0.08
1842
+0.25
0.07
609
+0.13
0.04
Negative Logits
impractica
-0.74
IUrlHelper
-0.58
philanth
-0.56
incarcer
-0.55
StoryboardSegue
-0.54
unlaw
-0.53
EEU
-0.52
SharedDtor
-0.51
<bos>
-0.51
prerog
-0.51
POSITIVE LOGITS
navire
0.47
Secrétaire
0.47
skimage
0.46
règlement
0.43
Sakit
0.43
Gön
0.42
°;
0.42
soyez
0.41
trésor
0.41
ille
0.40
Activations Density 0.578%