Datasets:

llm-crafter
/

l3-04-five-task-data

dataset_info:
  features:
    - name: conversations
      list:
        - name: content
          dtype: string
        - name: role
          dtype: string
  splits:
    - name: train
      num_bytes: 740401
      num_examples: 900
    - name: validation
      num_bytes: 77465
      num_examples: 100
  download_size: 407599
  dataset_size: 817866
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*