Olaf LenzmannMastering LLMs for Complex Classification TasksTesting top LLMs’ ability to find answers in text: prompt structure, logprobs, and fine-tuning. Quite a few surprises there!Jan 15Jan 15
Olaf LenzmannAchieving high-precision RAG in enterprise realityRetrieval Augmented Generation (RAG) is the go-to approach for dynamically infusing external information into LLMs for generative tasks…Nov 17, 20231Nov 17, 20231
Olaf LenzmannKicking a tire of Cohere Embed v3Having read the announcement of Embed v3, especially its promise to reflect content quality in rank of similarities, I was intrigued. Even…Nov 7, 2023Nov 7, 2023