LLM · Text Generation
Llama 3 8B Alpaca Data PT-BR
Portuguese fine-tune of Llama 3 8B on 51k Alpaca examples, built for Brazilian Portuguese chat and instruction following.
About this model
This model is a fine-tuned version of NousResearch/Meta-Llama-3-8B trained on the dominguesm/alpaca-data-pt-br dataset — a cleaned and translated Portuguese (Brazil) version of the original Stanford Alpaca dataset containing 51k examples. The training used Axolotl with 2 epochs, cosine learning rate schedule (2e-5), 8-bit paged AdamW optimizer, flash attention, and 8192 sequence length with sample packing. It achieves a validation loss of 1.1227. The model is intended for Portuguese-language chatbots, virtual assistants, and translation systems targeting the Brazilian market. License is listed as "other" on Hugging Face; verify before commercial use. No trigger words or special prompt format are documented — use standard Llama 3 chat template.
Project signals
- 20 Hugging Face downloads
- 4 Hugging Face likes
Topics
transformers · pytorch · safetensors · llama · text-generation · conversational · base_model:nousresearch/meta-llama-3-8b · base_model:finetune:nousresearch/meta-llama-3-8b · text-generation-inference · endpoints_compatible