metadata
dataset_info:
features:
- name: uid
dtype: string
- name: contest_id
dtype: int64
- name: contest_name
dtype: string
- name: contest_type
dtype: string
- name: index
dtype: string
- name: time_limit
dtype: float64
- name: memory_limit
dtype: int64
- name: rating
dtype: float64
- name: description
dtype: string
- name: reward_model
struct:
- name: ground_truth
dtype: string
- name: output_text
list: string
- name: eval-detail
list:
list: string
- name: best_eval
list: string
splits:
- name: train
num_bytes: 329539040
num_examples: 65
download_size: 201366526
dataset_size: 329539040
configs:
- config_name: default
data_files:
- split: train
path: data/train-*