trunk/cef26d1e97fcb9dd61b4471f9bd7fa9a32bd42b9: [MPS] faster norms (#186076)
🔒
https://github.com
«Routes norm(1/2, dim=-1) through the efficient sum_reduction_inner kernel via abs/square load + sqrt finalize, instead of the slow per-output-threadgroup norm kernel with per-element pow.
Example:
import torch
import tor...»
Automatische Weiterleitung...
1.5s