AI Systems · core
Language Modeling
Next-token prediction, context windows, perplexity.
Mental model
A language model does one thing: predict the next token given the ones before it. Chat, code, reasoning — all of it is built on top of this single objective.
How to study Language Modeling
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 Alisa's Book of LLMs (Alisa Wuffles), Stanford CS336 — Language Modeling from Scratch (course), Neural Networks: Zero to Hero (Karpathy) to check details, but close the source before writing your explanation. Retrieval is the learning step; rereading is only preparation.
Next, compare Language Modeling with Sampling & Decoding. Ask what changes in correctness, latency, resource use, operability, and failure recovery. Complete Perplexity from average NLL 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
Perplexity from average NLL
Average negative log-likelihood per token = 2.3 nats. Compute perplexity = exp(NLL).
Expected evidence: PPX ≈ e^2.3 ≈ 9.97.
Open the interactive drill →Review prompts
- What does perplexity actually measure, and why can two models' perplexities be incomparable?
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