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

Refactor QA Formatter / Configs for Phi & Llama #35

Merged
merged 23 commits into from
Aug 7, 2024
Merged

Conversation

jack89roberts
Copy link
Contributor

@jack89roberts jack89roberts commented Jul 29, 2024

WIP PR for trying to get Phi working.

  • QAFormatter adapted so that no. of tokens in the question (without the answer) can be computed with more complex prompt templates.
  • Model loading code edited to account for possibility of loading PEFT models
  • Added prompt templates for GPT2 (just a blank template), Phi-3 mini, Phi-3 small, Llama 3.1 8b.

Nothing tested yet and no hyperparameters.

@jack89roberts jack89roberts changed the base branch from develop to 2-new-dataset July 29, 2024 23:02
@jack89roberts jack89roberts changed the title Refactor QA Formatter / Configs for Phi Refactor QA Formatter / Configs for Phi & Llama Jul 31, 2024
@jack89roberts jack89roberts force-pushed the 27-phi branch 4 times, most recently from 4d90179 to 6e1162c Compare August 1, 2024 18:56
@jack89roberts jack89roberts changed the base branch from 2-new-dataset to experiment1 August 2, 2024 14:41
@jack89roberts jack89roberts marked this pull request as ready for review August 7, 2024 09:14
@jack89roberts jack89roberts merged commit f44e801 into experiment1 Aug 7, 2024
@jack89roberts jack89roberts deleted the 27-phi branch August 7, 2024 10:42
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.

1 participant