Wenwen Qu e2b7a7fa89 set default expert parallel size		2023-09-27 17:51:58 +08:00
..
legacy	merge upstream/develop into feature_add_moe	2023-09-11 16:20:08 +08:00
__init__.py	feat(numa): bind numa if possible (#320)	2023-09-25 19:34:52 +08:00
initialize_tensor.py	merge upstream/develop into feature_add_moe	2023-09-11 16:20:08 +08:00
initialize_trainer.py	merge upstream/develop into feature_add_moe	2023-09-11 16:20:08 +08:00
launch.py	set default expert parallel size	2023-09-27 17:51:58 +08:00