---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: data_source
    dtype: large_string
  - name: prompt
    dtype: large_string
  - name: ability
    dtype: large_string
  - name: reward_model
    dtype: large_string
  - name: extra_info
    dtype: large_string
  - name: ground_truth
    dtype: int64
  - name: trajectory_data
    dtype: large_string
  - name: transcript
    dtype: large_string
  splits:
  - name: train
    num_bytes: 469160304
    num_examples: 2656
  download_size: 172718198
  dataset_size: 469160304
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---