---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 55438571833.0
    num_examples: 1250000
  download_size: 29385394454
  dataset_size: 55438571833.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

5.12B tokens Pre Train Data