main
[Feature] Support Distributed LogProb for GRPO Training (#6247)
[Sharderformer] Support zbv in Sharderformer Policy (#6150)
[Ring Attention] Add more detailed references (#6294)
[misc] update pre-commit and run all files (#4752)
[fp8] support hybrid parallel plugin (#5982)
[CI] Cleanup Dist Optim tests with shared helper funcs (#6125)
fix logprob, add filtering, temperature annealing, lr descent
[Hotfix] hotfix normalization (#6163)
[shardformer] refactor embedding resize (#5603)