---
dataset_info:
  features:
  - name: meta
    dtype: string
  - name: task_type
    dtype: string
  - name: task_subtype
    dtype: string
  - name: task_subsubtype
    dtype: string
  - name: difficulty
    dtype: string
  - name: domain
    dtype: string
  - name: instruction
    dtype: string
  - name: criteria
    list:
    - name: criteria_description
      dtype: string
    - name: criteria_name
      dtype: string
    - name: rubrics
      dtype: string
    - name: rubrics_example
      dtype: string
  - name: prompt_id
    dtype: int64
  - name: reference_answer
    dtype: string
  splits:
  - name: train
    num_bytes: 60785826
    num_examples: 2115
  download_size: 2456200
  dataset_size: 60785826
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---