README.md

---
dataset_info:
  features:
  - name: query_id
    dtype: string
  - name: query
    dtype: string
  - name: positive_passages
    list:
    - name: docid
      dtype: string
    - name: explanation
      dtype: string
    - name: followir_score
      dtype: float64
    - name: joint_id
      dtype: string
    - name: text
      dtype: string
    - name: title
      dtype: string
  - name: negative_passages
    list:
    - name: docid
      dtype: string
    - name: text
      dtype: string
    - name: title
      dtype: string
  - name: only_instruction
    dtype: string
  - name: only_query
    dtype: string
  - name: has_instruction
    dtype: bool
  - name: new_negatives
    list:
    - name: docid
      dtype: string
    - name: explanation
      dtype: string
    - name: followir_score
      dtype: float64
    - name: joint_id
      dtype: string
    - name: text
      dtype: string
    - name: title
      dtype: string
  - name: origin_flag
    dtype: bool
  splits:
  - name: train
    num_bytes: 318112731
    num_examples: 24320
  download_size: 181059131
  dataset_size: 318112731
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---