Llama-3-Patronus-Lynx-8B-Instruct
This model evaluates answer fidelity to documents, ideal for researchers and professionals in fields requiring high accuracy like finance and academia.
What is Llama-3-Patronus-Lynx-8B-Instruct?
Llama-3-Patronus-Lynx-8B-Instruct is a model developed by Patronus AI based on Meta's Llama-3-8B-Instruct. It specializes in detecting hallucinations in RAG settings and evaluates if answers are faithful to given documents.