metadata
language:
- en
dataset_info:
features:
- name: id
dtype: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: weight
dtype: 'null'
- name: text
dtype: string
splits:
- name: train
num_bytes: 3630771342
num_examples: 994896
download_size: 1980759415
dataset_size: 3630771342
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
Open-Orca/1million-gpt-4 converted to sharegpt and llama chat format