metadata
dataset_info:
features:
- name: added
dtype: string
- name: created
dtype: timestamp[s]
- name: id
dtype: string
- name: metadata
struct:
- name: bucket
dtype: string
- name: cc_segment
dtype: string
- name: date_download
dtype: timestamp[s]
- name: digest
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: length
dtype: int64
- name: line_ids
sequence: int64
- name: nlines
dtype: int64
- name: original_length
dtype: int64
- name: original_nlines
dtype: int64
- name: perplexity
dtype: float64
- name: provenance
dtype: string
- name: source_domain
dtype: string
- name: title
dtype: string
- name: url
dtype: string
- name: source
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 438217413
num_examples: 127250
download_size: 240405343
dataset_size: 438217413
configs:
- config_name: default
data_files:
- split: train
path: data/train-*