[training] show how metadata stuff should be incorporated in training scripts. #11707
+70
−5
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
|
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. |
linoytsaban
approved these changes
Jun 16, 2025
There was a problem hiding this comment.
LGTM! Only thing is I wonder if we should add some check to the user input for alpha? perhaps adding a warning/info that when alpha != rank, if alpha < rank, the effect of the lora weights is downscaled and vise versa if alpha > rank the effect is upscaled and if they want to maintain a 1 scale factor they should set alpha==rank?
|
Good point. I believe that could be documented in the READMEs. |
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Outdated suggestions cannot be applied.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.
What does this PR do?
Since #11324 is merged now, we can safely inject metadata into safetensor files when training LoRAs.
Linoy, let's propagate this to other important scripts?
@a-r-r-o-w for awareness.