Datasets:

ohsuz
/

sft-koen-v1-with-math

dataset_info:
  features:
    - name: conversations
      list:
        - name: from
          dtype: string
        - name: value
          dtype: string
    - name: language
      dtype: string
  splits:
    - name: train
      num_bytes: 2559490897
      num_examples: 1170532
  download_size: 1223229307
  dataset_size: 2559490897
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*