What is a Large Language Model (LLM) and how does it work
The day I am writing this blogpost ChatGPT is celebrating its first anniversary (Nov 30th 2023). ChatGPT has taken the world by storm and has made artificial intelligence (AI) available to the masses. A lot has been said and written since the launch but in reality not many people really know how ChatGPT works. ChatGPT is a chatbot that let’s you talk to a large language model (LLM) that acts as the smart engine under the hood. You type your prompt in the textbox and the LLM returns an answer. But how does this work, what are the opportunities and challenges for the future for this technology? Andrej Karpathy, the former head of AI at Tesla, has recorded a great video in which he explains all these questions. I would highty recommend to watch this video since it is a comprehensive primer on LLMs.