---
dataset_info:
  config_name: mmarco_train10M_preprossesd_for_AraBERT
  features:
  - name: input_ids
    sequence: int32
  - name: token_type_ids
    sequence: int8
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: float64
  splits:
  - name: train
    num_bytes: 21841398000
    num_examples: 10000000
  download_size: 2197994660
  dataset_size: 21841398000
configs:
- config_name: mmarco_train10M_preprossesd_for_AraBERT
  data_files:
  - split: train
    path: mmarco_train10M_preprossesd_for_AraBERT/train-*
---