README.md

---
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 competitive **fully open** decoder for the **Portuguese language** language. 


It is a **decoder** of the GPT family, based on the neural architecture Transformer and developed over the LLaMA~2 7B model.
Its further improvement through additional training was done over language resources that include new instruction data sets of Portuguese prepared for this purpose.

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**).

All versions of Gervásio are **distributed for free and under a fully open license**, including for either research or commercial usage, and can 
be run on consumer-grade hardware, thus seeking to contribute to the advancement of research and innovation in language technology for Portuguese.

**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, a hidden size of 4096 units, an intermediate size of 11,008 units, 32 attention heads, 32 hidden layers, and a tokenizer obtained using the Byte-Pair Encoding (BPE) algorithm implemented with SentencePiece, featuring a vocabulary size of 32,000.
Gervásio-7B-PTPT-Instruct-Decoder is distributed under an [MIT license](https://huggingface.co/PORTULAN/albertina-ptpt/blob/main/LICENSE).


<br>

# Training Data

**Gervásio-7B-PTPT-Instruct-Decoder** over standard supervised fine-tuning, and to keep some alignment with mainstream benchmarks for English, we resorted to tasks and respective datasets in the GLUE and the SuperGLUE collections.


We selected those datasets where the outcome of their machine translation into Portuguese could preserve, in the target language, the linguistic properties at stake.

From GLUE, we resorted to the following four tasks:
- MRPC (paraphrase Detection).
- RTE (recognizing Textual Entailment).
- STS-B (semantic textual similarity).
- WNLI (coreference and natural language inference).

And from SuperGLUE, we included these other four tasks: 
- BoolQ (yes/no question answering).
- CB (inference with 3 labels).
- COPA (reasoning)
- MultiRC (question answering).


Instruction templates have been manually crafted for each task.
These take the various fields in the dataset and arrange them into a prompt.
For instance, appending ``Frase 1:'' (Eng.~``Sentence 1:'') before the first sentence of an example in the RTE dataset.
These templates are listed in full detail in TODO.

## 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.