Set keep_trailing_newline=True to preserve the trailing newline when …#39
Open
abse4411 wants to merge 1 commit intoGen-Verse:mainfrom
Open
Set keep_trailing_newline=True to preserve the trailing newline when …#39abse4411 wants to merge 1 commit intoGen-Verse:mainfrom
abse4411 wants to merge 1 commit intoGen-Verse:mainfrom
Conversation
…rendering prompt templates
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Set keep_trailing_newline=True to preserve the trailing newline when rendering prompt templates
I noticed that the prompt Template of the paper contains a line break at the end,

but the actual inference code will remove the line break by default when using the Jinja Template:
This can make a difference in the reproducibility of the experiment.
So I set
keep_trailing_newline=Truein the code to prevent this problem:Here's my experimental results on the MATH500 dataset (using TraDo-8B-Instruct) :
The experimental results with end-of-line breaks (75.46...) are very close to the reported value (75.5) ! I hope this PR can help others.