benchmark
f4acd1a7 - Reduce nanogpt gen sizes (#1659)

Commit
2 years ago
Reduce nanogpt gen sizes (#1659) Summary: Inductor is timing out on compilation so this is used to reduce the number of gen steps Pull Request resolved: https://github.com/pytorch/benchmark/pull/1659 Reviewed By: xuzhao9 Differential Revision: D45956640 Pulled By: drisspg fbshipit-source-id: e14505d6a0e35d4efebaae277cb872ce9072d05c
Author
Parents
Loading