metadata
dataset_info:
features:
- name: chatml
list:
- name: content
dtype: string
- name: name
dtype: string
- name: role
dtype: string
- name: token_count
dtype: int64
splits:
- name: train
num_bytes: 7017585
num_examples: 3246
- name: test
num_bytes: 219478
num_examples: 361
download_size: 1355714
dataset_size: 7237063
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*