README.md

---
dataset_info:
  features:
  - name: item_id
    dtype: string
  - name: question
    dtype: string
  - name: answer_value
    dtype: string
  - name: prediction
    dtype: string
  - name: extracted_code
    dtype: string
  - name: predicted_answer
    dtype: string
  - name: correctness
    dtype: bool
  - name: correctness_from_feedback
    dtype: bool
  - name: system
    dtype: string
  - name: user
    dtype: string
  - name: assistant
    dtype: string
  - name: feedback_based_actor_prediction
    dtype: string
  - name: feedback_based_actor_correctness
    dtype: bool
  - name: feedback_based_actor_extracted_code
    dtype: string
  - name: feedback_based_actor_trajectory
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  splits:
  - name: train
    num_bytes: 51484572
    num_examples: 6196
  download_size: 17578817
  dataset_size: 51484572
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---