Name |
Last commit
|
Last update |
---|---|---|
.github | ||
docker | ||
docs | ||
examples | ||
patches | ||
scripts | ||
tests | ||
verl | ||
.gitignore | ||
.readthedocs.yaml | ||
.style.yapf | ||
LICENSE | ||
Notice.txt | ||
README.md | ||
pyproject.toml | ||
requirements.txt | ||
setup.py |
Verl's megatron core_r0.11.0 backend successfully tested with 3D parallelism with multiple bug fixed (#495) This PR combines multiple modifications. # QWen2.5 checkpoint saver bug fix Thanks for the efforts @uygnef contributed to #368 , we use the new saver for model loader and saver for 3D parallelism support. # Megatron backend 3D-parallelism test benches We modify the scripts in `examples/ppo_trainer` and `tests/e2e`, as well as the CI workflows, all tested. # Bug Fix for 3D-parallelism Including configuration bugs as well as the module packing. Original TP VocabParallelEntropy can lead to CUDA OOM, we refactor the implementation with `torch.bmm`. # Fully migration to Megatron Core Now we only use Megatron core in verl, fully get rid of calling other components. If they are in need, please integrate them into `utils/megatron`. --------- Co-authored-by: uygnef <admin@fengyu.org>
Name |
Last commit
|
Last update |
---|---|---|
.github | Loading commit data... | |
docker | Loading commit data... | |
docs | Loading commit data... | |
examples | Loading commit data... | |
patches | Loading commit data... | |
scripts | Loading commit data... | |
tests | Loading commit data... | |
verl | Loading commit data... | |
.gitignore | Loading commit data... | |
.readthedocs.yaml | Loading commit data... | |
.style.yapf | Loading commit data... | |
LICENSE | Loading commit data... | |
Notice.txt | Loading commit data... | |
README.md | Loading commit data... | |
pyproject.toml | Loading commit data... | |
requirements.txt | Loading commit data... | |
setup.py | Loading commit data... |