Add RoPE scaling feature to SFT trainer #1152
Open
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
| override_config_kwargs = {} | ||
| if 'rope_scaling' in self.config.model and self.config.model.rope_scaling is not None: | ||
| override_config_kwargs['rope_scaling'] = dict(self.config.model.rope_scaling) | ||
| print(f'rope_scaling setted. rope_scaling={override_config_kwargs["rope_scaling"]}') |
Collaborator
|
@openhands-agent could you merge with main and fix potential pre-commit errors? |
Contributor
Author
|
@eric-haibin-lin i think you need to "@OpenHands" for OpenHands Cloud :) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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Outdated suggestions cannot be applied.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.
Add RoPE Scaling Feature to SFT Trainer
Description
This PR adds support for RoPE (Rotary Position Embedding) scaling in the SFT trainer. RoPE scaling is a technique that allows models to handle longer context lengths than they were originally trained on by scaling the position embeddings.
Changes
Usage
To use this feature, add a
rope_scalingconfiguration in your model config:Testing
Tested with various models that support RoPE scaling, including Llama and Qwen models.