---
dataset_info:
  features:
  - name: codes
    sequence:
      sequence:
        sequence: int64
  - name: embd
    sequence:
      sequence:
        sequence: float32
  - name: prompt
    dtype: string
  - name: chunk_id
    dtype: int64
  - name: nb_chunks
    dtype: int64
  splits:
  - name: train
    num_bytes: 515572807375
    num_examples: 263063
  - name: test
  download_size: 429498575612
  dataset_size: 515572807375
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---