Skip to content

Seed-Coder-8B-Instruct,在几个评测集上效果很差。。。 #16

@dog-qiuqiu

Description

@dog-qiuqiu

hello,感谢开源模型,我在几个公开的评测集上进行自评测,保证采用每个benchmark官方的评测框架,发现效果一般,是模型的某些参数导致的吗(例如温度,一般的benchmark默认采用0.0),这是我评测的结果:

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