Fix nvfp4/mxfp8 formats silently becoming fp8 in --layer-config - #56
Open
rockerBOO wants to merge 1 commit into
Open
Fix nvfp4/mxfp8 formats silently becoming fp8 in --layer-config#56rockerBOO wants to merge 1 commit into
rockerBOO wants to merge 1 commit into
Conversation
The fmt -> layer_format narrowing switch in convert_to_fp8_scaled only handled float8*/int8* prefixes and fell back to "fp8" for anything else, so --layer-config entries with "format": "nvfp4" or "mxfp8" quantized as plain FP8 with no warning. Extracted the mapping into resolve_layer_config_format() and added the missing branches. AI Assisted
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
nvfp4 specifically for me was defaulting silently back to fp8 when using the --layer-config for setting the format. This fixes it to be the correct format but new formats will still fall through to fp8. Might be better as an error so it doesn't get all the way through quantizing before noticing it is not correct.