metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: token_count
dtype: int64
- name: chatml
list:
- name: content
dtype: string
- name: name
dtype: 'null'
- name: role
dtype: string
splits:
- name: train
num_bytes: 457245282
num_examples: 200035
download_size: 166642712
dataset_size: 457245282
configs:
- config_name: default
data_files:
- split: train
path: data/train-*