TheSequence is a no-BS newsletter focused on machine learning that takes only 5 minutes to read and aims to keep subscribers up-to-date with machine learning projects, research papers, and concepts. The current generation of large language models, such as the instruction following, is one of the cornerstones of these LLMs. Reinforcement learning with human preferences (RLHF) and techniques such as InstructGPT have been utilized to further improve these models. Databricks has recently fine-tuned a two-year-old LLM named Dolly to follow instructions like ChatGPT.

source update: How Databricks Finetuned a Two-Year-Old LLM to Follow… – Towards AI