metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: title
dtype: string
- name: content_idx
dtype: int64
- name: article_url
dtype: string
- name: section_idx
dtype: int64
- name: subsection_idx
dtype: int64
- name: keyword
dtype: string
- name: document_idx
dtype: int64
- name: author
dtype: string
- name: author_url
dtype: string
- name: context
dtype: string
- name: generated_qa_pairs
list:
- name: answer
dtype: string
- name: question
dtype: string
- name: status
sequence: int64
- name: generated_question
dtype: string
- name: generated_answer
dtype: string
- name: generated_text
dtype: string
- name: inputs_embeds
sequence:
sequence: float32
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: query_ids
sequence: int64
- name: query_attention_mask
sequence: int64
splits:
- name: train
num_bytes: 477293046
num_examples: 52386
- name: eval
num_bytes: 4211036
num_examples: 400
- name: test
num_bytes: 3627241
num_examples: 400
download_size: 302773529
dataset_size: 485131323
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
- split: test
path: data/test-*