Paper page - Soaring from 4K to 400K: Extending LLM's Context with Activation Beacon

Vector embeddings model for medical literature

Best 7B LLM on leaderboards made by an amateur following a medium tutorial

LiteLlama-460M-1T has 460M parameters trained with 1T tokens

Microsoft Phi-2 model changes licence to MIT

LLaMA-Pro-8B

Open LLM Leaderboard

Local vector database with arXiv metadata

Local vector database with all of Wikipedia

Gemini vs. GPT-4V: A Preliminary Comparison Through Qualitative Cases