---
dataset_info:
  features:
  - name: material_id
    dtype: string
  - name: formula_reduced
    dtype: string
  - name: CIF
    dtype: string
  - name: n_tokens
    dtype: string
  splits:
  - name: train
    num_bytes: 645139036.9742815
    num_examples: 607675
  - name: validation
    num_bytes: 71682705.0257185
    num_examples: 67520
  download_size: 185922472
  dataset_size: 716821742.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---