---
dataset_info:
  features:
  - name: id
    dtype: large_string
  - name: query
    dtype: large_string
  - name: data
    dtype: large_string
  - name: output
    dtype: large_string
  - name: timestamp
    dtype: large_string
  - name: query_type
    dtype: large_string
  - name: chunks
    list: string
  - name: num_chunks
    dtype: int64
  - name: top_chunks
    list: string
  - name: top_scores
    list: float64
  - name: prompt
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: num_top_chunks
    dtype: int64
  - name: generation_time
    dtype: float64
  - name: output_Qwen-Qwen3-0.6B
    dtype: string
  - name: reasoning_Qwen-Qwen3-0.6B
    dtype: 'null'
  - name: generation_success
    dtype: bool
  splits:
  - name: train
    num_bytes: 1196336
    num_examples: 8
  download_size: 1041534
  dataset_size: 1196336
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---