Datasets:

Brendan
/

bookcorpus_processed

dataset_info:
  features:
    - name: input_ids
      sequence: int32
    - name: attention_mask
      sequence: int8
    - name: special_tokens_mask
      sequence: int8
  splits:
    - name: train
      num_bytes: 5587060752
      num_examples: 1811628
    - name: test
      num_bytes: 1396660332
      num_examples: 452873
  download_size: 2308641872
  dataset_size: 6983721084

Dataset Card for "bookcorpus_processed"

More Information needed