---
license: mit
language:
- pt
tags:
- gervasio-pt*
- gervasio-ptpt
- gervasio-ptbr
- gervasio-ptpt-base
- gervasio-ptbr-base
- portulan
- albertina-pt*
- albertina-ptpt
- albertina-ptbr
- albertina-ptbr-nobrwac
- albertina-ptpt-base
- albertina-ptbr-base
- clm
- gpt
- portuguese
- decoder
- foundation model
- instruct
datasets:
- PORTULAN/glue-ptpt
---
<img align="left" width="40" height="40" src="https://github.githubassets.com/images/icons/emoji/unicode/1f917.png">
<p style="text-align: center;">&nbsp;&nbsp;&nbsp;&nbsp;This is the model card for Gervásio 7B PT-PT Instruct Decoder 
  You may be interested in some of the other models in the <a href="https://huggingface.co/PORTULAN">Albertina (encoders) and Gervásio (decoders) families</a>.
</p>

# Gervásio 7B PT-PT Instruct


**Gervásio PT-*** is a ??, large language model for the **Portuguese language**.

+++++corrigir
It is a **decoder** of the GPT family, based on the neural architecture Transformer and 
developed over the Pythia model, with competitive performance for this language. 
It has different versions that were trained for different variants of Portuguese (PT), 
namely the European variant from Portugal (**PT-PT**) and the American variant from Brazil (**PT-BR**), 
and it is distributed free of charge and under a most permissible license.
++++++++

**Gervásio PT-PT 7B Instruct** is developed by NLX-Natural Language and Speech Group, at the University of Lisbon, Faculty of Sciences, Department of Informatics, Portugal.

For the record, its full name is **Gervásio Produz Textos em Português**, to which corresponds the natural acronym **GPT PT**, 
and which is know tough more shortly as **Gervásio PT-***, or even more briefly just as **Gervásio**, among his acquaintances.

For further details, check the respective [publication](https://arxiv.org/abs/?):

``` latex
@misc{albertina-pt,
      title={Advancing Generative AI for Portuguese with Open Decoder Gervásio~PT*}, 
      author={Rodrigo Santos, João Silva, Luís Gomes, João Rodrigues, António Branco},
      year={2024},
      eprint={?},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
```

Please use the above cannonical reference when using or citing this model.


<br>


# Model Description

**This model card is for Gervásio-7B-PTPT-Instruct-Decoder**, with 7 billion parameters, ? layers and a hidden size of ?.

Gervásio-PT-BR base is distributed under an [Apache 2.0 license](https://huggingface.co/PORTULAN/gervasio-ptpt-base/blob/main/LICENSE) (like Pythia).


<br>

# Training Data

[**Gervásio-PT-BR base**](https://huggingface.co/PORTULAN/gervasio-ptpt-base) was trained over a 3.7 billion token curated selection of documents from the [OSCAR](https://huggingface.co/datasets/oscar-corpus/OSCAR-2301) data set.
The OSCAR data set includes documents in more than one hundred languages, including Portuguese, and it is widely used in the literature. 
It is the result of a selection performed over the [Common Crawl](https://commoncrawl.org/) data set, crawled from the Web, that retains only pages whose metadata indicates permission to be crawled, that performs deduplication, and that removes some boilerplate, among other filters. 
Given that it does not discriminate between the Portuguese variants, we performed extra filtering by retaining only documents whose meta-data indicate the Internet country code top-level domain of Brazil. 
We used the January 2023 version of OSCAR, which is based on the November/December 2022 version of Common Crawl.



## Preprocessing

We filtered the PT-BR corpora using the [BLOOM pre-processing](https://github.com/bigscience-workshop/data-preparation) pipeline.
We skipped the default filtering of stopwords since it would disrupt the syntactic structure, and also the filtering for language identification given the corpus was pre-selected as Portuguese.


# Evaluation

The base model version was evaluated on downstream tasks, namely the translations into PT-PT of the English data sets used for a few of the tasks in the widely-used [GLUE benchmark](https://huggingface.co/datasets/glue).


## GLUE tasks translated


We resorted to [GLUE-PT](https://huggingface.co/datasets/PORTULAN/glue-ptpt), a **PT-PT version of the GLUE** benchmark. 
We automatically translated the same four tasks from GLUE using [DeepL Translate](https://www.deepl.com/), which specifically provides translation from English to PT-PT as an option.

| Model                    | RTE (Accuracy) | WNLI (Accuracy)| MRPC (F1) | STS-B (Pearson) |
|--------------------------|----------------|----------------|-----------|-----------------|
| **Albertina-PT-PT**      | **0.8339**     | 0.4225         | **0.9171**| **0.8801**      | 
| **Albertina-PT-PT base** |  0.6787        | **0.4507**     | 0.8829    | 0.8581          |

<br>

# How to use

You can use this model directly with a pipeline for causal language modeling (CLM):

```python3
>>> from transformers import pipeline
>>> generator = pipeline(model='PORTULAN/gervasio-ptbr-base')
>>> generator("A música brasileira é", max_new_tokens=10)
[{'generated_text': 'A música brasileira é uma das mais ricas do mundo. Ao'}]



```
<br>

# Acknowledgments

The research reported here was partially supported by: PORTULAN CLARIN—Research Infrastructure for the Science and Technology of Language,
funded by Lisboa 2020, Alentejo 2020 and FCT—Fundação para a Ciência e Tecnologia under the
grant PINFRA/22117/2016; research project GPT-PT - Transformer-based Decoder for the Portuguese Language, funded by FCT—Fundação para a Ciência e Tecnologia under the
grant CPCA-IAC/AV/478395/2022; innovation project 
ACCELERAT.AI - Multilingual Intelligent Contact Centers, funded by IAPMEI, I.P. - Agência para a Competitividade e Inovação 
under the grant C625734525-00462629, of Plano de Recuperação e Resiliência, 
call RE-C05-i01.01 – Agendas/Alianças Mobilizadoras para a Reindustrialização.