⏯️ Fix logging when resuming from checkpoint GRPO #3185
+3
−4
Conversation
This file contains 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
…ngface/trl into fix_resume_from_checkpoint
qgallouedec
commented
Apr 1, 2025
| @@ -481,7 +480,7 @@ def data_collator(features): # No data collation is needed in GRPO | |||
| # vLLM specific sampling arguments | |||
| self.guided_decoding_regex = args.vllm_guided_decoding_regex | |||
|
|
|||
| self._last_loaded_step = 0 # tag to avoid useless loading during grad accumulation | |||
| self._last_loaded_step = -1 # tag to avoid useless loading during grad accumulation | |||
There was a problem hiding this comment.
Make sure that for the very first generation, the model loaded is the same as that of the model running on the server.
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
qgallouedec
changed the title
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Outdated suggestions cannot be applied.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.
First training, (orange) from 0 to 100 timestep
Second training (purple), resume from timestep 40: curves overlap.