INDEX
Explanations
This neuron detects parts of web URLs (especially the “https://” prefix and domain name tokens).
New Auto-Interp
Negative Logits
undecided
-0.07
.act
-0.07
composer
-0.06
Amazing
-0.06
379
-0.06
�
-0.06
sen
-0.06
gal
-0.06
'id
-0.06
SSI
-0.06
POSITIVE LOGITS
/>,
0.07
AngularFire
0.07
منابع
0.06
anale
0.06
лы
0.06
visualization
0.06
-ended
0.06
.nextElement
0.06
udicots
0.06
}elseif
0.06
Activations Density 0.019%