ldwang 36ac14a566 Add patch_mixtral_replace_moe_impl for full training Mitral using DeepSpeed Zero3.
Signed-off-by: ldwang <ftgreat@gmail.com>

Former-commit-id: d1413dcec8a3b1d671f240b82a689c72b54d7b93
2024-01-24 14:43:16 +08:00
..
2024-01-18 09:54:23 +08:00
2024-01-20 20:15:56 +08:00
2024-01-22 16:01:58 +08:00
2024-01-20 20:15:56 +08:00