PPO Dtensor value model support sequence packing + CP
Open
enhancement
- Dominant language
- Python
- Stars
- 2k
- Forks
- 561
- Avg merge
- 4d 5h
- Merged PRs (30d)
- 145
Description
Tracking missing features in https://github.com/NVIDIA-NeMo/RL/pull/2837.
- [ ] sequence packing
- [ ] CP
Contributor guide
Assessment
This issue has not been assessed yet.