metadata
dataset_info:
features:
- name: env
dtype: string
- name: prompt
dtype: string
- name: extra
struct:
- name: game_data
struct:
- name: answer
dtype: string
- name: difficulty
dtype: int64
- name: gpt_response
dtype: string
- name: metadata
struct:
- name: closing_sequence
dtype: string
- name: fill_length
dtype: int64
- name: full_sequence
dtype: string
- name: n_types
dtype: int64
- name: nesting_depth
dtype: int64
- name: question_sequence
dtype: string
- name: seed
dtype: int64
- name: total_length
dtype: int64
- name: trace_id
dtype: string
- name: question
dtype: string
- name: metadata
struct:
- name: closing_sequence
dtype: string
- name: fill_length
dtype: int64
- name: full_sequence
dtype: string
- name: n_types
dtype: int64
- name: nesting_depth
dtype: int64
- name: question_sequence
dtype: string
- name: seed
dtype: int64
- name: total_length
dtype: int64
- name: trace_id
dtype: string
- name: seed
dtype: int64
- name: task_id
dtype: int64
- name: task_type
dtype: string
splits:
- name: train
num_bytes: 5762898
num_examples: 5093
download_size: 4216720
dataset_size: 5762898
configs:
- config_name: default
data_files:
- split: train
path: data/train-*