Tucano 2b4

🚀 Tucano-2b4

Tucano is a series of decoder-transformers natively pretrained in Portuguese. It offers a powerful solution for Portuguese text generation tasks, trained on a large - scale Portuguese text corpus.

🚀 Quick Start

Basic Usage

from transformers import pipeline

generator = pipeline("text-generation", model="TucanoBR/Tucano-2b4")

completions  = generator("A floresta da Amazônia é conhecida por sua", num_return_sequences=2, max_new_tokens=100)

for comp in completions:
  print(f"🤖 {comp['generated_text']}")

Advanced Usage

from transformers import GenerationConfig, TextGenerationPipeline, AutoTokenizer, AutoModelForCausalLM
import torch

# Specify the model and tokenizer
model_id = "TucanoBR/Tucano-2b4"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# Specify the generation parameters as you like
generation_config = GenerationConfig(
    **{
    "do_sample": True,
    "max_new_tokens": 2048,
    "renormalize_logits": True,
    "repetition_penalty": 1.2,
    "temperature": 0.1,
    "top_k": 50,
    "top_p": 1.0,
    "use_cache": True, 
  }
)

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
generator = TextGenerationPipeline(model=model, task="text-generation", tokenizer=tokenizer, device=device)

# Generate text
prompt = "A floresta da Amazônia é conhecida por sua"
completion = generator(prompt, generation_config=generation_config)
print(completion[0]['generated_text'])

✨ Features

• Native Portuguese Training: Tucano models are natively pretrained in Portuguese, making them well - suited for Portuguese language tasks.
• Large - Scale Dataset: Trained on GigaVerbo, a concatenation of deduplicated Portuguese text corpora amounting to 200 billion tokens.
• Long Context Support: With a context length of 4096 tokens, it can handle longer text inputs.

📦 Installation

The README does not provide specific installation steps, so this section is skipped.

📚 Documentation

Model Summary

Tucano is a series of decoder - transformers natively pretrained in Portuguese. All Tucano models were trained on GigaVerbo, a concatenation of deduplicated Portuguese text corpora amounting to 200 billion tokens.

Read our preprint here.

Details

This repository has the [source code](https://github.com/Nkluge - correa/Tucano) used to train this model. The main libraries used are:

• PyTorch
• Transformers
• Datasets
• Tokenizers
• Sentencepiece
• Accelerate
• [FlashAttention](https://github.com/Dao - AILab/flash - attention)
• [Liger Kernel](https://github.com/linkedin/Liger - Kernel)
• Codecarbon
• TRL

Intended Uses

The primary intended use of the Tucano models is to serve as foundations for research and development involving native Portuguese language modeling. Checkpoints saved during training are designed to provide a controlled setting for performing comparative experiments, specifically regarding the effects of active pretraining on the performance of currently available benchmarks. You may also fine - tune and adapt Tucano models for deployment if your use follows the Apache 2.0 license. If you decide to use the Tucano models as a basis for your fine - tuned model, please conduct your own risk and bias assessment.

Out - of - scope Use

• Tucano models are not intended for deployment. They are not an out - of - the - box product and should not be used for human - facing interactions.
• Tucano models are for the Portuguese language only and are unsuitable for text generation tasks in other languages.
• Tucano models have not been fine - tuned for downstream tasks.

Limitations

Like almost all other language models trained on large text datasets scraped from the web, the Tucano models show behavior that does not make them an out - of - the - box solution to many real - world applications, especially those requiring factual, reliable, and nontoxic text generation. Tucano models are all subject to the following:

• Hallucinations: Tucano models can produce content that can be mistaken as true facts, but are misleading or entirely false, i.e., hallucination.
• Biases and Toxicity: Tucano models inherit the social and historical stereotypes from the data used to train them. Given these biases, the model can produce toxic content, i.e., harmful, offensive, or detrimental to individuals, groups, or communities.
• Unreliable Code: Tucano models may produce incorrect code snippets and statements. These code generations should not be treated as suggestions or accurate solutions.
• Language Limitations: Tucano models are primarily designed to interact with Portuguese. Other languages might challenge its comprehension, leading to potential misinterpretations or errors in response.
• Repetition and Verbosity: Tucano models may get stuck on repetition loops (especially if the repetition penalty during generations is set to a meager value) or produce verbose responses unrelated to the prompt it was given.

Hence, even though our models are released with a permissive license, we urge users to perform their risk analysis on them if they intend to use them for real - world applications.

Evaluations

The table below compares our models against several Portuguese and multilingual language models on the evaluation harness used in our study. More information on it can be found [here](https://github.com/Nkluge - correa/Tucano/tree/main/evaluations/README.md). To learn more about our evaluation harness selection, read our preprint.

🔧 Technical Details

The README does not provide specific technical details that meet the requirement of more than 50 - word description, so this section is skipped.

📄 License

Tucano is licensed under the Apache License, Version 2.0. For more details, see the LICENSE file.

Cite as 🤗

@misc{correa2024tucanoadvancingneuraltext,
      title={{Tucano: Advancing Neural Text Generation for Portuguese}}, 
      author={Corr{\^e}a, Nicholas Kluge and Sen, Aniket and Falk, Sophia and Fatimah, Shiza},
      year={2024},
      eprint={2411.07854},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2411.07854}, 
}

Aknowlegments

We gratefully acknowledge the granted access to the [Marvin cluster](https://www.hpc.uni - bonn.de/en/systems/marvin) hosted by [University of Bonn](https://www.uni - bonn.de/en) along with the support provided by its High Performance Computing & Analytics Lab.