Stack_TokLlama / README.md
Trevor Dohm
Tokenized bison
2f4b8b7 verified
|
raw
history blame
329 Bytes
metadata
dataset_info:
  features:
    - name: input_ids
      sequence: int32
    - name: attention_mask
      sequence: int8
  splits:
    - name: train
      num_bytes: 8709278
      num_examples: 176
  download_size: 954382
  dataset_size: 8709278
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/bison/train-*