Hugging Face's logo

Datasets:
Fate-Zero
/
Archer2.0-Code-1.5B

Modalities:
Text
Formats:
parquet
Size:
1K - 10K
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files
xet
 Community
2

Improve dataset card: Add paper, code, project links, sample usage, and metadata

#2
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+81
-22
nielsr

This PR enriches the dataset card by adding:

  • A clear overview of the dataset and its associated paper, ASPO: Asymmetric Importance Sampling Policy Optimization.
  • Direct links to the Hugging Face paper, the GitHub repository (Archer2.0), and the project page (Notion).
  • The task_categories (text-generation), language (en), and relevant tags (code-generation, mathematical-reasoning, reinforcement-learning, llm, llm-finetuning) to improve discoverability and context.
  • A "Sample Usage" section featuring a training script, directly sourced from the GitHub README.
  • The BibTeX citations for the relevant papers.
 Improve dataset card: Add paper, code, project links, sample usage, and metadata4fbf3422
Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment