NousResearch/Nous-Hermes-Llama2-13b

Nous-Hermes-Llama2-13b is a 13 billion parameter language model developed by Nous Research, fine-tuned on over 300,000 instructions using the Llama 2 architecture with a 4096 token context length. This model is distinguished by its long response generation, reduced hallucination rates, and absence of OpenAI's censorship mechanisms. It was primarily trained on high-quality synthetic GPT-4 outputs, making it suitable for complex instruction following and knowledge-intensive tasks. The model shows strong performance in reasoning and common sense benchmarks, including top rankings on ARC-c, ARC-e, Hellaswag, and OpenBookQA.