In this day and age it’s generally not a good idea to train a Deep Neural Network (DNN) from scratch without first trying to find an existing neural network that accomplishes a similar task to the one you are trying to tackle.

Intro

Organisation’s are always quick to jump on the bandwagon when it comes to AI/ML. As a result many will end up falling into common pitfalls when actually trying to get things up and running.

Intro

Useful Resources

Disclaimer: There has been a large amount of work performed by people when it comes to understanding of LLMs and the like. As a result this post isn’t made with the intentions to re-work the efforts of many but rather guide it’s exploration, focusing on some of the key, personally biased, interesting aspects. Also, at the end of the day ChatGPT is a proprietary model, meaning unless you work at OpenAI you don’t acutally know the specifics beyond the information that has been publically released or empirically evaluated. Always ensure you read multiple sources when trying to verify your own thoughts.