README.md

metadata

dataset_info:
  features:
    - name: row_num
      dtype: int64
    - name: id
      dtype: string
    - name: eng_system_prompt
      dtype: string
    - name: mar_system_prompt
      dtype: string
    - name: eng_question
      dtype: string
    - name: mar_question
      dtype: string
    - name: eng_response
      dtype: string
    - name: mar_response
      dtype: string
  splits:
    - name: train
      num_bytes: 418882307
      num_examples: 100000
  download_size: 159689428
  dataset_size: 418882307
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: mit
task_categories:
  - text-generation
  - question-answering
language:
  - mr
  - en

Dataset card for Marathi OpenOrca

Translated subset of Open-Orca/1million-gpt-4 to marathi language.