Skip to content

Question about eval results on Qwen2-7B-Instruct #21

@LUOMU17

Description

@LUOMU17

When I evaluated Qwen2-7B-Instruct on the data/test/GSM8K_test_data.jsonl and data/test/MATH_test_data.jsonl datasets using eval_math.py, the accuracy I obtained exceeded the values reported in the paper. Could there be an issue with one of my operations?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions