Stack_Tokenized / README.md
Trevor Dohm
Tokenized clarion
345f2f8 verified
|
raw
history blame
371 Bytes
metadata
dataset_info:
  features:
    - name: content
      dtype: string
    - name: input_ids
      sequence: int32
    - name: attention_mask
      sequence: int8
  splits:
    - name: train
      num_bytes: 18180705
      num_examples: 1237
  download_size: 5206550
  dataset_size: 18180705
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/clarion/train-*