---
license: apache-2.0
dataset_info:
  features:
  - name: conversations
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
    - name: weight
      dtype: float64
  splits:
  - name: train
    num_bytes: 36626491
    num_examples: 20000
  download_size: 19631352
  dataset_size: 36626491
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---