The Transformer Architecture
In 2017, a team at Google DeepMind published Attention Is All You Need. At the time, nobody outside the research community paid much attention (pun aggressively intended). RNNs were the establishment; this paper was an upstart. We know now that this paper redirected the field, kicked off the LLM era, and — if you're reading this — quite possibly inspired you to study AI!
✦
Go deeper: The Annotated Transformer
For a line-by-line walkthrough of the original paper with working PyTorch code, check out The Annotated Transformer. It's one of the best resources available for understanding how the architecture translates into actual implementation.