Weyaxi/Orca-Nova-13B
Weyaxi/Orca-Nova-13B is a 13 billion parameter language model, likely based on the OpenOrca/OpenChat architecture, designed for general language understanding and generation tasks. It demonstrates competitive performance across various benchmarks, including ARC, HellaSwag, MMLU, and Winogrande, making it suitable for applications requiring strong reasoning and common sense. With a 4096-token context length, it can process moderately long inputs for diverse NLP applications.
Loading preview...
Overview
Weyaxi/Orca-Nova-13B is a 13 billion parameter language model, likely derived from the OpenOrca/OpenChat architecture, as indicated by the upcoming Hugging Face link. This model is evaluated on the Open LLM Leaderboard, showcasing its capabilities across a range of tasks.
Key Capabilities & Performance
The model achieves an average score of 49.69 on the Open LLM Leaderboard. Specific benchmark results highlight its strengths in:
- ARC (25-shot): 62.37
- HellaSwag (10-shot): 82.47
- MMLU (5-shot): 57.44
- Winogrande (5-shot): 77.58
While performing well in general reasoning and common sense tasks, its scores on more complex reasoning benchmarks like GSM8K (14.48) and DROP (7.52) suggest areas for further development.
Good For
- General text generation and understanding tasks.
- Applications requiring strong performance in common sense reasoning and factual recall, as indicated by its HellaSwag and Winogrande scores.
- Use cases where a 13B parameter model with a 4096-token context window is suitable for balancing performance and computational resources.