INDEX
Explanations
The neuron activates on occurrences of code.google.com URLs (i.e. Google Code project links).
New Auto-Interp
Negative Logits
улыб
-0.06
أغ
-0.06
اش
-0.06
BS
-0.06
PARK
-0.06
.angular
-0.06
蘇
-0.06
arbon
-0.06
mio
-0.06
_get
-0.06
POSITIVE LOGITS
Code
0.08
code
0.08
Code
0.08
/code
0.07
qualified
0.07
code
0.07
Claudia
0.07
.Country
0.07
grade
0.07
cane
0.07
Activations Density 0.006%