Manacá-1B: An Open, Reproducible Brazilian-Portuguese Language Model and a Tokenizer-Aware, Paired Evaluation
Bruno Leonardo Santos Menezes, Carlos Leonardo Souza Cardoso, Fabio Andre Machado Porto
Abstract
Brazilian Portuguese remains under-served by open language models, and the few that exist are difficult to reproduce and are often compared without measures of uncertainty. We release Manacá-1B, an open decoder-only model of 1.72 billion parameters trained from scratch for Brazilian Portuguese with a fully containerized, reproducible pipeline. The pretraining is stable, with zero skipped or NaN steps and self-recovering loss spikes, and we release its full log and dynamics. We evaluate the model against nine open baselines on four Portuguese benchmarks under a single harness. Every comparison …
Who it is from
LNCCnome no título: manaca
Other papers from the same organizations
- On the salient limitations of the methods of assembly theory and their classification of molecular biosignaturesLNCC · 34 citations
- Minimal algorithmic information loss methods for dimension reduction, feature selection and network sparsificationLNCC · 27 citations
- Assembly Theory is an approximation to algorithmic complexity based on LZ compression that does not explain selection or evolutionLNCC · 23 citations
- SuperARC: a test for artificial superintelligence based on compressed modelling, recursive prediction and problem complexityLNCC · 10 citations
- Neuro-Spectral Architectures for Causal Physics-Informed NetworksLNCC · 10 citations
- Regression and Classification with Single-Qubit Quantum Neural NetworksLNCC · 9 citations