metadata
language:
- en
dataset_info:
features:
- name: target
sequence: string
- name: pubmed_id
dtype: int64
- name: text_combined
dtype: string
- name: text
sequence: string
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 1065756674
num_examples: 190654
download_size: 513375508
dataset_size: 1065756674
configs:
- config_name: default
data_files:
- split: train
path: data/train-*