long-proof-pile / README.md
vikhyatk's picture
Update README.md
d709bc5 verified
metadata
dataset_info:
  features:
    - name: text
      dtype: string
    - name: meta
      dtype: string
  splits:
    - name: train
      num_bytes: 52428748
      num_examples: 100
  download_size: 20492289
  dataset_size: 52428748
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Documents from Proof Pile with over 128K tokens, as measured by the starmie-v1 tokenizer.

Intended use is to measure long context cross-entropy.