ArtificialGuyBR

Home / Projects / OpenHermesV2 PTBR - Portuguese Mistral 7B Fine-tune

LLM · Text Generation

OpenHermesV2 PTBR - Portuguese Mistral 7B Fine-tune

Portuguese fine-tune of OpenHermes 2 Mistral 7B for ChatML instruction-following chat, coding and roleplay.

About this model

OpenHermesV2 PTBR is a Brazilian Portuguese fine-tune of OpenHermes 2 Mistral 7B, the Mistral-7B-v0.1 fine-tune by Teknium known for strong instruction-following performance. The original OpenHermes model was trained on about 900,000 entries of primarily GPT-4 generated data sourced from open datasets, extensively filtered and converted to ShareGPT format, then transformed to ChatML by axolotl. This release adapts that model for speakers of Brazilian Portuguese, covering chat, programming Q&A and roleplay. The author describes it as a personal fine-tune made for fun and testing, not a production release.

Learn more

Read the in-depth guide about OpenHermesV2 PTBR - Portuguese Mistral 7B Fine-tune

Project signals

  • 23 Hugging Face downloads
  • 3 Hugging Face likes

Topics

transformers · pytorch · safetensors · mistral · text-generation · instruct · finetune · chatml · gpt4 · synthetic data

Explore the source