AI Systems · core
Embeddings (Transformer)
Token & position embeddings, tied weights.
Mental model
Embeddings turn discrete things (words, items, users) into vectors of numbers so that similar things end up close together in space. Distance and direction in that space carry meaning — that geometry is what the model learns from.
How to study Embeddings (Transformer)
Begin by restating the mental model in your own words, then connect it to a concrete system you have built or operated. Name the mechanism, the constraint it addresses, and the trade-off it introduces. Use Stanford CS336 — L3 Architectures & hyperparameters (Spring 2025), The Illustrated GPT-2 — token & position embeddings, The Illustrated Transformer to check details, but close the source before writing your explanation. Retrieval is the learning step; rereading is only preparation.
Next, compare Embeddings (Transformer) with Self-Attention, Embeddings. Ask what changes in correctness, latency, resource use, operability, and failure recovery. Complete Sinusoidal position encoding and preserve the command, input, output, and one failed attempt as evidence. Finish by explaining the idea without jargon to someone who has not studied the track.
Proof of understanding
- Explain the mechanism from first principles and identify the state it reads or changes.
- Give one situation where the concept is the right choice and one where it is not.
- Predict a realistic failure mode before running the drill, then compare the prediction with evidence.
- Connect the result to a roadmap or build artifact instead of treating the concept as isolated trivia.
Learn from primary sources
Practice and explain it back
Sinusoidal position encoding
pos=1, i=0, d=4: PE(pos,2i)=sin(pos/10000^(2i/d)). Compute the first two dims for pos=1.
Expected evidence: sin(1), cos(1) pattern for even/odd dims.
Open the interactive drill →Review prompts
- Self-attention is permutation-equivariant — shuffle the tokens and the output shuffles with them. What does that force the embedding layer to do, and why?
Build evidence
Synthesize: AI Models & Training
Move from transformer foundations through pre-training, fine-tuning, post-training, compression, and evaluation. Produce one working system, benchmark, or evidence-backed design that integrates the path.
- Implements or precisely models the core mechanisms from all three milestones
- Includes at least one injected failure or adversarial case and demonstrates recovery
- Reports quality, latency, resource, reliability, or usability measurements relevant to the domain
- Ships a concise architecture note explaining decisions, trade-offs, and remaining risks