INDEX
Explanations
The neuron fires on words and phrases related to software installation steps (e.g. forms of “install” or “instalar” and “installation”).
New Auto-Interp
Negative Logits
ple
-0.08
برای
-0.07
ingenious
-0.07
BIND
-0.06
Popular
-0.06
ariate
-0.06
Hits
-0.06
PBS
-0.06
�
-0.06
anı
-0.06
POSITIVE LOGITS
ctica
0.07
Kuzey
0.06
Brush
0.06
(dest
0.06
cath
0.06
(nullptr
0.06
�
0.06
(btn
0.06
documentation
0.06
згод
0.06
Activations Density 0.013%