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

Fix supervised pretraining bugs and add new datasets #699

Merged
merged 6 commits into from
Jan 14, 2023
Merged

Conversation

theblackcat102
Copy link
Collaborator

Code change

Mainly typo and logic error by previous PR merge.

Code refactor in datasets

Dataset change

Added these datasets into training lists

  • SODA
  • Joke Explaination
  • GSM8k

@theblackcat102
Copy link
Collaborator Author

And one big change was changing the question answer tokens to <human> and <bot> for better clarity. Since some of the datasets are about multi turns dialogue (SODA)

Copy link
Collaborator

@sanagno sanagno left a comment

Choose a reason for hiding this comment

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

Thanks a lot for taking care of this! This was a pain.

@theblackcat102
Copy link
Collaborator Author

Thanks a lot for taking care of this! This was a pain.

We should only allow PR with tests case written.

@theblackcat102
Copy link
Collaborator Author

Going to merge now

@theblackcat102 theblackcat102 merged commit dbf8f77 into main Jan 14, 2023
@theblackcat102 theblackcat102 deleted the sft-fixes branch January 14, 2023 12:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants