metadata

dataset_info:
  features:
    - name: prompt
      dtype: string
    - name: prompt_id
      dtype: string
    - name: messages
      list:
        - name: content
          dtype: string
        - name: role
          dtype: string
    - name: category
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 30628039
      num_examples: 9500
    - name: test
      num_bytes: 1644450
      num_examples: 500
  download_size: 19873853
  dataset_size: 32272489
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
task_categories:
  - conversational
  - text-generation
language:
  - pt
pretty_name: Portuguese Chat
license: cc-by-nc-4.0

Dataset Card for Portuguese Chat

We know that current English-first LLMs don’t work well for many other languages, both in terms of performance, latency, and speed. Building instruction datasets for non-English languages is an important challenge that needs to be solved.

Dedicated towards addressing this problem, I release 3 new datasets rishiraj/portuguesechat, rishiraj/bengalichat & rishiraj/hindichat of 10,000 instructions and demonstrations each. This data can be used for supervised fine-tuning (SFT) to make language multilingual models follow instructions better.

Dataset Summary

rishiraj/portuguesechat was modelled after the instruction dataset described in OpenAI's InstructGPT paper, and is translated from HuggingFaceH4/no_robots which comprised mostly of single-turn instructions across the following categories:

Category	Count
Generation	4560
Open QA	1240
Brainstorm	1120
Chat	850
Rewrite	660
Summarize	420
Coding	350
Classify	350
Closed QA	260
Extract	190

Languages

The data in rishiraj/portuguesechat are in Portuguese (BCP-47 pt).

Data Fields

The data fields are as follows:

prompt: Describes the task the model should perform.
prompt_id: A unique ID for the prompt.
messages: An array of messages, where each message indicates the role (system, user, assistant) and the content.
category: Which category the example belongs to (e.g. Chat or Coding).
text: Content of messages in a format that is compatible with dataset_text_field of SFTTrainer.

Data Splits

	train_sft	test_sft
portuguesechat	9500	500

Licensing Information

The dataset is available under the Creative Commons NonCommercial (CC BY-NC 4.0).

Citation Information

@misc{portuguesechat,
  author = {Rishiraj Acharya},
  title = {Portuguese Chat},
  year = {2023},
  publisher = {Hugging Face},
  journal = {Hugging Face repository},
  howpublished = {\url{https://huggingface.co/datasets/rishiraj/portuguesechat}}
}