How Chat GPT and Other LLMs Work and Where They Could Go Next

Chat GPT, Google Bard and other bots like them are examples of large language models or LLMs. Most LLMs use a specific neural network architecture called a transformer which has some tricks particularly suited to language processing. One of the key innovations of these transformers is the self-attention mechanism. It allows for a greater level of comprehension than would otherwise be possible. You might also notice generated text being rather generic or cliched from a bot that's trying to synthesize responses from giant repositories of text.