metadata
dataset_info:
features:
- name: query
dtype: string
- name: completion
dtype: string
- name: total_reward
dtype: int64
splits:
- name: train
num_bytes: 789375349
num_examples: 779653
download_size: 38748974
dataset_size: 789375349
configs:
- config_name: default
data_files:
- split: train
path: data/train-*