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.

Sign up for GitHub

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

Jump to bottom

The base model and chat model have no difference when using generate_until, loglikelihood, loglikelihood_rolling,right? #2347

Open
belle9217 opened this issue Sep 25, 2024 · 1 comment
Open

The base model and chat model have no difference when using generate_until, loglikelihood, loglikelihood_rolling,right? #2347

belle9217 opened this issue Sep 25, 2024 · 1 comment
Labels
asking questions For asking for clarification / support on library usage.

Comments

@belle9217
Copy link

The base model and chat model have no difference when using generate_until, loglikelihood, loglikelihood_rolling,right?
generate_until, loglikelihood, loglikelihood_rolling,right is related to the dataset I use,right?
If i use a long answer dataset,i config it generate_until, while I use multi-choice dataset, i config it multi-choice,right?
@baberabb
Copy link
Contributor

yes, thats correct. You probably want to use --apply_chat_template (and optionally --fewshot_as_multiturn) for the instruct models to use the chat templates.

@baberabb baberabb added the asking questions For asking for clarification / support on library usage. label Sep 25, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
asking questions For asking for clarification / support on library usage.
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@belle9217 @baberabb