metadata
dataset_info:
features:
- name: bff_contained_ngram_count_before_dedupe
dtype: int64
- name: language_id_whole_page_fasttext
struct:
- name: en
dtype: float64
- name: metadata
struct:
- name: Content-Length
dtype: string
- name: Content-Type
dtype: string
- name: WARC-Block-Digest
dtype: string
- name: WARC-Concurrent-To
dtype: string
- name: WARC-Date
dtype: string
- name: WARC-IP-Address
dtype: string
- name: WARC-Payload-Digest
dtype: string
- name: WARC-Record-ID
dtype: string
- name: WARC-Target-URI
dtype: string
- name: WARC-Truncated
dtype: string
- name: WARC-Type
dtype: string
- name: WARC-Warcinfo-ID
dtype: string
- name: previous_word_count
dtype: int64
- name: text
dtype: string
- name: url
dtype: string
- name: warcinfo
dtype: string
- name: cleaned_text
dtype: string
splits:
- name: train
num_bytes: 109345140
num_examples: 16000
download_size: 65485786
dataset_size: 109345140
configs:
- config_name: default
data_files:
- split: train
path: data/train-*