Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

System prompt data #224

Merged
merged 3 commits into from
Jun 27, 2023
Merged

System prompt data #224

merged 3 commits into from
Jun 27, 2023

Conversation

winglian
Copy link
Collaborator

adds support for pulling the system prompt from the data files. Useful for some CoT datasets.

Also adds more test coverage

Copy link
Collaborator

@NanoCode012 NanoCode012 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It looks alright. I couldn't test it myself, but seem like the change is small for people not using system prompts.

I'm just worried for increased complexity. Optional system prompts, optional inputs etc.

@@ -128,9 +132,6 @@ def build_prompt(
) -> Generator[str, None, None]:
yield instruction

def get_response(self, output: str) -> str:
return output.strip()
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are these get_response removed because they are unused?

@winglian
Copy link
Collaborator Author

It looks alright. I couldn't test it myself, but seem like the change is small for people not using system prompts.

I'm just worried for increased complexity. Optional system prompts, optional inputs etc.

Yeah. That was dead code that was copied from alpaca repo.

@winglian winglian merged commit f150c02 into main Jun 27, 2023
6 checks passed
@winglian winglian deleted the system-prompt-data branch June 27, 2023 21:57
mkeoliya pushed a commit to mkeoliya/axolotl that referenced this pull request Dec 15, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

support alpaca datasets with the system prompt baked into the dataset
2 participants