Llama-3-Patronus-Lynx-8B-Instruct-v1.1
Open-source hallucination evaluation model
CommonProductProgrammingText GenerationHallucination Evaluation
Patronus-Lynx-8B-Instruct-v1.1 is a fine-tuned version of the meta-llama/Meta-Llama-3.1-8B-Instruct model, specifically designed to detect hallucinations in RAG setups. The model has been trained on multiple datasets, including CovidQA, PubmedQA, DROP, and RAGTruth, incorporating both manually annotated and synthetic data. It assesses whether the given document, question, and answer are faithful to the document content, without providing new information beyond the document scope or contradicting the document information.
Llama-3-Patronus-Lynx-8B-Instruct-v1.1 Visit Over Time
Monthly Visits
20899836
Bounce Rate
46.04%
Page per Visit
5.2
Visit Duration
00:04:57