Clean up codes and comments for SGLang rollout #1
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
zyzshishui
reviewed
May 24, 2025
| # TODO(chenyang): is `max_turns` the max number of tool call rounds? | ||
| # If so, I think it should be a small number like 3 or 5. | ||
| # self.config.max_model_len // 3 is a large number. | ||
|
|
There was a problem hiding this comment.
Yes. But if max_model_len is reached or the request state becomes complete, the rollout could also stop. So generally, self.config.max_model_len //3 would not come to effect. it's a kind of a placeholder in case max_turn is not specified in the configuration.
zyzshishui
reviewed
May 24, 2025
| non_pad_index = torch.nonzero(prompt_token_ids != pad_token_id, as_tuple=False)[0][0] | ||
| non_pad_index = torch.nonzero(prompt_token_ids != pad_token_id, as_tuple=False)[0][ | ||
| 0 | ||
| ] |
There was a problem hiding this comment.
just fix lint of verl. I don't know why a lot of lint failed but verl's CI passed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Checklist Before Starting
What does this PR do?
I cleaned up the codes used in
verl/workers/rollout/sglang_rollout/sglang_rollout.pyand added more comments for users. There are three questions left:# TODO(chenyang): Strange way to calculate nnodes.SGLangRolloutbe used? I am just pondering why inSGLangRollout, SGLang'sVerlEngineis not used. But intests/workers/rollout/test_sglang_spmd.py, theVerlEngineis used.