Skip to content

Potential inconsistency in the example evaluation #33

Description

@powidla

Hi! Thank you for your great work! I apologize for bothering you, I just wanted to flag a potential issue with either IFBench_test.jsonl or sample_output.jsonl. It occurred to me that these files have different length: IFBench_test.jsonl has 300 while sample_output.jsonl contains 294. On top of that, running the following command from readme: python3 -m run_eval --input_data=IFBench_test.jsonl --input_response_data=sample_output.jsonl --output_dir=eval for me results in key errors. I believe it happens because of the fact that sample_output.jsonl contains some prompts with extra trailing whitespaces. For me those are 0,1,14 and 20 keys, but I might have missed some. After fixing those, the command works just fine. Thank you in advance for reading this issue.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions