Overview

SlovenianGPT is a 7 billion parameter base Large Language Model (LLM) developed by Aleksa Gordić, specifically designed and trained for the Slovenian language. Built on the robust Mistral 7B architecture, this model has been trained on tens of billions of Slovenian language tokens, establishing it as the premier open-source base LLM for Slovenian.

Key Capabilities

• Slovenian Language Generation: Optimized for generating high-quality text in Slovenian.
• Base Model Functionality: Operates as a powerful autocomplete engine, suitable for various text completion tasks.
• Performance: Evaluation results demonstrate strong performance compared to other general-purpose models like Mistral 7B, LLaMA 2 7B, and Gemma on Slovenian language benchmarks.

Important Notes

• Base Model: As a base model, SlovenianGPT does not include built-in moderation mechanisms.
• Instruction Following: It functions as a powerful autocomplete engine and is not designed to follow complex instructions directly.

Citation

Users can cite SlovenianGPT using the provided BibTeX entry:

@article{SlovenianGPT,
  author    = "Gordić Aleksa",
  title     = "SlovenianGPT - an open-source LLM for Slovenian language",
  year      = "2024"
  howpublished = {\url{https://huggingface.co/gordicaleksa/SlovenianGPT}},
}