Understanding Transformers Architecture

In this series of posts, we’ll explore each and every component of Transformers architecture that powers latest LLMs like ChatGPT and vision models like Stable Diffiusion.

• Multi-Head Attention From Scratch
• Masking in Transformer Encoder/Decoder Models
• Implementing Transformer Encoder Layer From Scratch
• Implementing Transformer Decoder Layer From Scratch
• Decoding strategies in Decoder models (LLMs)