Datasets:

chanind
/

logiclm_bookcorpus_dataset

dataset_info:
  features:
    - name: input_ids
      sequence: int32
    - name: token_type_ids
      sequence: int8
    - name: attention_mask
      sequence: int8
    - name: special_tokens_mask
      sequence: int8
  splits:
    - name: train
      num_bytes: 8483436000
      num_examples: 2356510
  download_size: 2276245237
  dataset_size: 8483436000

Dataset Card for "logiclm_bookcorpus_dataset"

More Information needed