---
dataset_info:
  features:
  - name: anchor
    dtype: string
  - name: positive
    dtype: string
  - name: negative
    dtype: string
  splits:
  - name: train
    num_bytes: 20550568977
    num_examples: 25863649
  - name: validation
    num_bytes: 419477179
    num_examples: 527832
  download_size: 13638899915
  dataset_size: 20970046156
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
license: apache-2.0
language:
- pt
pretty_name: msmarco-triplets-ptbr
size_categories:
- 10M<n<100M
---

Need a huge dataset translated? Connect with me!