README.md

metadata

dataset_info:
  - config_name: long
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 10508746018.593323
        num_examples: 2781867
    download_size: 35147587836
    dataset_size: 10508746018.593323
  - config_name: medium
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 42699699704.44106
        num_examples: 11303431
    download_size: 75050119308
    dataset_size: 42699699704.44106
  - config_name: short
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 718888576987.6477
        num_examples: 190303620
    download_size: 320882355051
    dataset_size: 718888576987.6477
  - config_name: xlong
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 2244430950.3190165
        num_examples: 594144
    download_size: 14663872589
    dataset_size: 2244430950.3190165
  - config_name: xxlong
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 255686010.24725088
        num_examples: 67685
    download_size: 3176336379
    dataset_size: 255686010.24725088
  - config_name: xxxlong
    features:
      - name: text
        dtype: string
      - name: id
        dtype: string
      - name: dump
        dtype: string
      - name: url
        dtype: string
      - name: date
        dtype: string
      - name: file_path
        dtype: string
      - name: language
        dtype: string
      - name: language_score
        dtype: float64
      - name: token_count
        dtype: int64
      - name: num_tokens
        dtype: int64
    splits:
      - name: train
        num_bytes: 21962893.751606952
        num_examples: 5814
    download_size: 558825509
    dataset_size: 21962893.751606952
configs:
  - config_name: long
    data_files:
      - split: train
        path: long/train-*
  - config_name: medium
    data_files:
      - split: train
        path: medium/train-*
  - config_name: short
    data_files:
      - split: train
        path: short/train-*
  - config_name: xlong
    data_files:
      - split: train
        path: xlong/train-*
  - config_name: xxlong
    data_files:
      - split: train
        path: xxlong/train-*
  - config_name: xxxlong
    data_files:
      - split: train
        path: xxxlong/train-*

length categories

Classification	Token Length Range	Description
short	< 2,048	Less than 2K tokens
medium	2,048 - 4,095	2K to 4K tokens
long	4,096 - 8,191	4K to 8K tokens
xlong	8,192 - 16,383	8K to 16K tokens
xxlong	16,384 - 32,767	16K to 32K tokens
xxxlong	≥ 32,768	32K tokens or more