metadata
dataset_info:
features:
- name: query
dtype: string
- name: completion
dtype: string
- name: total_reward
dtype: int64
- name: num_tokens
dtype: int64
splits:
- name: train
num_bytes: 716050601.3709316
num_examples: 701687
- name: test
num_bytes: 79561971.62906832
num_examples: 77966
download_size: 93365214
dataset_size: 795612573
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*