deepseek scratch

Pick anything below and it plays in the bar at the foot of the window — and keeps playing while you go on browsing the directory.

1. 1▶All about Sinusoidal Positional Encodings | What’s with the weird sin-cos formula?Build DeepSeek from ScratchNotes
   +Watch later
2. 2▶Integer and Binary Positional Encodings | Journey towards Rotary Positional Encodings (RoPE)Build DeepSeek from ScratchNotes
   +Watch later
3. 3▶Multi-Head Latent Attention Coded from Scratch in PythonBuild DeepSeek from ScratchNotes
   +Watch later
4. 4▶Multi-Head Latent Attention From Scratch | One of the major DeepSeek innovationBuild DeepSeek from ScratchNotes
   +Watch later
5. 5▶Understand Grouped Query Attention (GQA) | The final frontier before latent attentionBuild DeepSeek from ScratchNotes
   +Watch later
6. 6▶Multi-Query Attention Explained | Dealing with KV Cache Memory Issues Part 1Build DeepSeek from ScratchNotes
   +Watch later
7. 7▶Key Value Cache from Scratch: The good side and the bad sideBuild DeepSeek from ScratchNotes
   +Watch later
8. 8▶Multi-Head Attention Handwritten from ScratchBuild DeepSeek from ScratchNotes
   +Watch later
9. 9▶Multi-Head Attention Visually ExplainedBuild DeepSeek from ScratchNotes
   +Watch later
10. 10▶Causal Attention Explained: Don't Peek into the Future!Build DeepSeek from ScratchNotes
    +Watch later
11. 11▶Self Attention Mechanism - Handwritten from scratchBuild DeepSeek from ScratchNotes
    +Watch later
12. 12▶The Attention Mechanism 1 hour explanationBuild DeepSeek from ScratchNotes
    +Watch later
13. 13▶LLM Architecture in 1 hour | Journey of token through the LLM ArchitectureBuild DeepSeek from ScratchNotes
    +Watch later
14. 14▶DeepSeek BasicsBuild DeepSeek from ScratchNotes
    +Watch later
15. 15▶Build DeepSeek from Scratch: Series IntroductionBuild DeepSeek from ScratchNotes
    +Watch later