Hugging Face's logo

Datasets:
Axion1-0
/
axion-training-data

Tasks:
Text Generation
Question Answering
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
10K - 100K
ArXiv:
Tags:
agents
scientific-reasoning
tool-use
supervised-fine-tuning
reinforcement-learning
datasets
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
axion-training-data

Commit History

Today

  1. Add Hugging Face paper link (#1)⁠

    ChrisBullet +1: nielsr
    668831c

3 days ago

  1. Link the official Axion code repository from the dataset card

    ChrisBullet
    Verified signature
    eb1f569

  2. Axion 1.0 training data: SFT trajectories and RL tasks, Apache 2.0

    ChrisBullet
    95ee05b