metadata

language:
  - en
license: apache-2.0
size_categories:
  - 10K<n<100K
task_categories:
  - text-generation
dataset_info:
  features:
    - name: prompt
      dtype: string
  splits:
    - name: train
      num_bytes: 39374567
      num_examples: 50488
  download_size: 18735701
  dataset_size: 39374567
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Arxiv Math

This is a cleansed version of ArtifactAI/arxiv-math-instruct-50k

Usage

from datasets import load_dataset
dataset = load_dataset("Sharathhebbar24/arxiv-math-instruct-50k", split="train")