---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 61846407893
    num_examples: 10888966
  download_size: 28809168123
  dataset_size: 61846407893
task_categories:
- text-generation
language:
- pt
pretty_name: OSCAR-2301-PT (deduplicated)
size_categories:
- 10B<n<100B
source_datasets:
  - extended|oscar-corpus/OSCAR-2301
---
# OSCAR-2301-PT (deduplicated)

OSCAR-2301-PT (deduplicated) is a curation from OSCAR-2301 in the Portuguese language, depuplicated by [MinHash algorithm](https://dl.acm.org/doi/abs/10.5555/647819.736184) and [Locality Sensitive Hashing](https://dspace.mit.edu/bitstream/handle/1721.1/134231/v008a014.pdf?sequence=2&isAllowed=y), following the approach of [Lee et al. (2022)](http://arxiv.org/abs/2107.06499).

This version of the dataset is the portuguese subset from [OSCAR-2301](https://huggingface.co/datasets/oscar-corpus/OSCAR-2301).
[OSCAR-2301](http://arxiv.org/abs/2201.06642) is a multilingual corpus extracted from the November/December 2022 dump of Common Crawl.

## Statistics
We used 5-grams and a signature of size 256, considering two documents to be identical if their Jaccard Similarity exceeded 0.7. 

- Number of docs. - before deduplication: 18,031,400
- Number of docs. - after deduplication: 10,888,966
- Duplicates: 39,61%