Hugging Face's logo

Datasets:
sungyub
/
deepmath-103k-verl

Tasks:
Text Generation
Question Answering
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
100K - 1M
ArXiv:
Tags:
math
reasoning
verl
reinforcement-learning
math-reasoning
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
deepmath-103k-verl

Commit History

  1. Update README with unified format (v3.0)

    sungyub
    Verified signature
    32653c5

  2. Update README with unified format (v3.0)

    sungyub
    Verified signature
    8c09673

  3. Update README with unified format (v3.0)

    sungyub
    Verified signature
    6e699ab

  1. Upload DeepMath-103K cleaned dataset: 101,844 samples (1,147 duplicates removed, orz-math cleaning)

    sungyub
    Verified signature
    3f632b9

  2. Add comprehensive README with DeepMath-103K metadata and cleaning statistics

    sungyub
    Verified signature
    8d1f242

  3. initial commit

    sungyub
    Verified signature
    b350266