INDEX
Explanations
The neuron activates on words used when describing or referring to building (“app,” “application,” “developing,” “building,” etc.) in a software‐development context.
New Auto-Interp
Negative Logits
IAM
-0.07
uestra
-0.07
Erie
-0.07
CActive
-0.07
921
-0.06
uang
-0.06
_max
-0.06
DEFAULT
-0.06
ournament
-0.06
_scaling
-0.06
POSITIVE LOGITS
/*******************************************************************************↵
0.08
Closing
0.06
мот
0.06
arkers
0.06
Intelli
0.06
consolid
0.06
]])↵↵
0.06
disg
0.06
tps
0.06
grad
0.06
Activations Density 0.040%