FP8 training enhancements #3496
Merged
+73
−16
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
Datta0
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
For FP8 models, make it supported for non fast inference.
Also fix for when weight shape is not multiple of 8. Read here.
Also move hf quantizer patch to unsloth from unsloth zoo to make it run for non fast inference.
Patch Fbgemmfp8linear and fp8linear classes' forward methods here to make them work for compiled models (which don't use
mamtul_loraexplicitly)depends on #unslothai/unsloth-zoo#337
Sample Qwen 2.5 VL 7B on FP8 GRPO :)
