---
dataset_info:
  features:
  - name: hash
    dtype: string
  - name: version
    dtype: string
  - name: truncation
    dtype: string
  - name: padding
    dtype: string
  - name: added_tokens
    dtype: string
  - name: normalizer
    dtype: string
  - name: pre_tokenizer
    dtype: string
  - name: post_processor
    dtype: string
  - name: decoder
    dtype: string
  - name: model
    dtype: string
  - name: tokenizer_id
    dtype: int64
  splits:
  - name: train
    num_bytes: 75248355676
    num_examples: 24798
  download_size: 40032749852
  dataset_size: 75248355676
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---