Wenwen Qu 21624f6f81 fix(moe): remove norm&gate force sync (#448) ... * add zero broadcast_sync * delete old sync logic * fix merged error * refactor code * remove some unused function (is norm/gate group)		2023-11-01 11:29:55 +08:00
..
__init__.py	feat(moe): add moe module (#182)	2023-09-27 15:54:53 +08:00
embedding.py	fix(model/embedding.py): ci lint check error (#345)	2023-09-21 14:46:22 +08:00
linear.py	feat(linear): optimize mlp by using jit (#321)	2023-09-19 14:57:43 +08:00
loss.py	…
metrics.py	feat(train): add fsdp training option (#293)	2023-10-09 18:59:31 +08:00
modeling_internlm.py	fix(*)/all-reduce for norm in sequence parallel (#443)	2023-10-25 14:16:32 +08:00
modeling_moe.py	fix(moe): remove norm&gate force sync (#448)	2023-11-01 11:29:55 +08:00
moe.py	Doc(moe): add documentation for moe training (#411)	2023-10-19 10:01:12 +08:00
multi_head_attention.py	fix(model): fix errant inference_forward (#396)	2023-10-09 08:29:11 -05:00
norm.py	Merge develop to main (#233)	2023-08-24 22:03:04 +08:00
utils.py	fix(moe): remove norm&gate force sync (#448)	2023-11-01 11:29:55 +08:00