INDEX
Explanations
This neuron detects the “Share this:” share-prompt header that appears at the start of social-sharing lines.
New Auto-Interp
Negative Logits
quanh
-0.07
FullName
-0.06
sleeve
-0.06
불
-0.06
dioxide
-0.06
Gee
-0.06
sat
-0.06
холод
-0.06
unami
-0.06
效
-0.06
POSITIVE LOGITS
suburbs
0.07
Courier
0.07
Fasc
0.06
address
0.06
barring
0.06
Brit
0.06
RefPtr
0.06
getElement
0.06
0.06
ROUTE
0.06
Activations Density 0.027%