SFT training #301
Merged
SFT training #301
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
|
For this to support bigger models in any scenario that isn't a single gigantic GPU, I'd change the code to load the model with HF Accelerate and also use the 8-bit quantization from bitsandbytes. If ya'll are interested, I could open a PR for this simple change. |
|
That would be great, I am already looking into this but feel free to start with the PR. |
|
Cool, I'll work on this over the weekend. |
|
@mrcabbage972 see also the latest PR #452 and #456. Feel free to edit. |
|
Thanks, I see that @theblackcat102 already started working on 8-bit quantization in #456. |
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Outdated suggestions cannot be applied.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.
Base code for SFT pretraining. Currently, it is mostly a placeholder code. Will be adapted based on the format of the human generated data.