README.md

---
dataset_info:
  features:
  - name: world
    dtype: string
  - name: difficulty
    dtype: string
  - name: solution turns
    dtype: int64
  - name: status (model=gpt-4.1, prompt=explore, legend=verbose)
    dtype: string
  - name: turns (model=gpt-4.1, prompt=explore, legend=verbose)
    dtype: int64
  - name: messages (model=gpt-4.1, prompt=explore, legend=verbose)
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: status (model=o3, effort=high, prompt=explore, legend=verbose)
    dtype: string
  - name: turns (model=o3, effort=high, prompt=explore, legend=verbose)
    dtype: int64
  - name: messages (model=o3, effort=high, prompt=explore, legend=verbose)
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: status (model=o4-mini, effort=high, prompt=explore, legend=verbose)
    dtype: string
  - name: turns (model=o4-mini, effort=high, prompt=explore, legend=verbose)
    dtype: int64
  - name: messages (model=o4-mini, effort=high, prompt=explore, legend=verbose)
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: status (model=o4-mini, effort=low, prompt=explore, legend=verbose)
    dtype: string
  - name: turns (model=o4-mini, effort=low, prompt=explore, legend=verbose)
    dtype: int64
  - name: messages (model=o4-mini, effort=low, prompt=explore, legend=verbose)
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  splits:
  - name: train
    num_bytes: 856869
    num_examples: 10
  download_size: 273824
  dataset_size: 856869
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---