metadata
dataset_info:
features:
- name: problem
dtype: string
- name: r1_distill_70b_response
sequence:
sequence: string
- name: _majority_responses
sequence: string
- name: verified_r1_distill_70b_response
dtype: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 6524925052
num_examples: 24000
download_size: 2220345312
dataset_size: 6524925052
configs:
- config_name: default
data_files:
- split: train
path: data/train-*