---
dataset_info:
  features:
  - name: trajectory_string
    dtype: string
  - name: reached_goal
    dtype: bool
  - name: reward
    dtype: float64
  - name: length
    dtype: int64
  - name: maze_spec
    struct:
    - name: goal
      struct:
      - name: x
        dtype: int64
      - name: y
        dtype: int64
    - name: height
      dtype: int64
    - name: start
      struct:
      - name: x
        dtype: int64
      - name: y
        dtype: int64
    - name: walls
      list:
      - name: x
        dtype: int64
      - name: y
        dtype: int64
    - name: width
      dtype: int64
  - name: start_pos
    struct:
    - name: x
      dtype: int64
    - name: y
      dtype: int64
  - name: goal_pos
    struct:
    - name: x
      dtype: int64
    - name: y
      dtype: int64
  - name: maze_index
    dtype: int64
  - name: language_prompt
    dtype: string
  - name: language_completion
    dtype: string
  - name: data_source
    dtype: string
  - name: prompt
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: ability
    dtype: string
  - name: reward_model
    struct:
    - name: ground_truth
      dtype: string
    - name: style
      dtype: string
  - name: extra_info
    struct:
    - name: index
      dtype: int64
    - name: length
      dtype: int64
    - name: maze_spec
      struct:
      - name: goal
        struct:
        - name: x
          dtype: int64
        - name: y
          dtype: int64
      - name: height
        dtype: int64
      - name: start
        struct:
        - name: x
          dtype: int64
        - name: y
          dtype: int64
      - name: walls
        list:
        - name: x
          dtype: int64
        - name: y
          dtype: int64
      - name: width
        dtype: int64
    - name: reached_goal
      dtype: bool
    - name: reward
      dtype: float64
    - name: split
      dtype: string
  splits:
  - name: train
    num_bytes: 8977684661
    num_examples: 237500
  - name: test
    num_bytes: 472647552
    num_examples: 12500
  download_size: 539696652
  dataset_size: 9450332213
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---