main
[shardformer] fix the moe (#5883)
[llama] fix dataloader for hybrid parallel (#5358)
[hotfix] fix lora load (#6231)
[Distributed RLHF] Integration of PP (#6257)
[shardformer] support pipeline for deepseek v3 and optimize lora save (#6188)
[shardformer] support ep for deepseek v3 (#6185)
[lora] add lora APIs for booster, support lora for TorchDDP (#4981)
[misc] update pre-commit and run all files (#4752)