Cert Notes/ Commute Study Notes
Roadmap
KOEN
CLF-C02 · FoundationalCloud Practitioner - Foundational
DVA-C02 · AssociateDeveloper - Associate
SAA-C03 · AssociateSolutions Architect - Associate
SOA-C02 · AssociateCloudOps Engineer - Associate
SAP-C02 · ProfessionalSolutions Architect - Professional
DOP-C02 · ProfessionalDevOps Engineer - Professional
SCS-C03 · SpecialtySecurity - Specialty
MLA-C01 · AssociateMachine Learning Engineer - Associate
AIF-C01 · FoundationalAI Practitioner - Foundational
  • Week 1
    • 1.The Relationship and Differences Between AI, Machine Learning, and Deep Learning
    • 2.Learning Types: Supervised, Unsupervised, and Reinforcement Learning
    • 3.Problems Where ML Fits vs. Problems Where It Doesn't
    • 4.Key Terms: Model, Training, Inference, Feature, Label, Overfitting
    • 5.Week 1 Wrap-Up: AI/ML Fundamentals Review
  • Week 2
    • 1.ML Lifecycle Overview: One Complete Cycle from Data to Operations
    • 2.Data: Structured vs Unstructured, Data Splitting, and the Power of Quality
    • 3.Model Evaluation Basics: Accuracy, Precision, Recall, and Overfitting/Underfitting
    • 4.The Human Role in ML Development: Labeling, Feedback, and Iterative Improvement
    • 5.Week 2 Summary: ML Lifecycle and Data at a Glance
  • Week 3
    • 1.What Is Generative AI: Difference from Traditional ML, and Foundation Models
    • 2.How LLMs Work: Tokens, Embeddings, Context Window, and Inference
    • 3.Prompt Engineering Basics: Good Prompts, Zero/Few-shot, and Limitations
    • 4.Limitations and Risks of Generative AI: Hallucinations, Bias, Non-determinism, and Appropriate Use Cases
    • 5.Week 3 Comprehensive Review: Wrapping Up Generative AI Fundamentals at a Glance
  • Week 4
    • 1.Amazon Bedrock: Fully Managed Service for Renting Foundation Models
    • 2.Amazon SageMaker: A Platform for Directly Training and Deploying ML Models
    • 3.AWS AI Services (1): Managed APIs Handling Images, Documents, Text, and Speech
    • 4.AWS AI Services (2) + Amazon Q: Chatbots, Search, Recommendations, Forecasting, and Generative Assistant
    • 5.Week 4 Comprehensive Review: Complete Map of AWS AI/ML Services
  • Week 5
    • 1.Principles of Responsible AI: Fairness, Bias, Transparency, Explainability, Robustness, Privacy
    • 2.AWS's Responsible AI Tools: SageMaker Clarify, Model Monitor, Bedrock Guardrails, AI Service Cards
    • 3.AI Security: Least Privilege IAM, Data Encryption, PII Protection, PrivateLink, Shared Responsibility Model
    • 4.Data Governance and Compliance: Data Origin·Quality, Model Governance, Audit·Logging, Legal and Ethical Considerations for Generative AI
    • 5.Week 5 Comprehensive Review: Binding Responsible AI·Security·Governance into One
  • Week 6
    • 1.Domain Review 1: AI/ML Fundamentals + Generative AI Fundamentals: Critical Summary
    • 2.Domain Review 2: Foundation Model Applications (AWS AI Services) Critical Summary
    • 3.Domain Review 3: Responsible AI + Security·Governance Critical Summary
    • 4.Full Mock Exam Pace: Five Domains Comprehensive Questions
    • 5.D-Day Wrap-Up: Exam Structure, Keyword → Service Translation Table, Frequently-Missed Traps
DEA-C01 · AssociateData Engineer - Associate
MLS-C01 · SpecialtyMachine Learning - Specialty
← AIF-C01/Week 3/Day 2
AIF-C01· AssociateWeek 3 · Day 2~21 min read

Day 2 - How LLMs Work: Tokens, Embeddings, Context Window, and Inference

Yesterday we learned that LLMs "probabilistically predict the next token." But a model is a computer. Computers don't understand letters. They only understand numbers. So how does a sentence like "The weather is nice today" become numbers that go into the model, and how does it come back out as letters?

Today we'll trace how text is processed inside an LLM using four core concepts: Token → Embedding → Context Window → Inference. These four terms appear directly on the AIF exam and are also the keys to understanding cost and performance. Our goal is to understand without formulas—like drawing a picture.

Tokens: The Unit by Which Models Divide Text

LLMs don't receive sentences all at once

The rest is Pro only

Week 1 is free for everyone. Week 2 onwards — plus mock exams and unlimited review — is included in the Pro plan.

See ProLogin
PreviousWhat Is Generative AI: Difference from Traditional ML, and Foundation ModelsWeek 3 · Day 1Next Prompt Engineering Basics: Good Prompts, Zero/Few-shot, and LimitationsWeek 3 · Day 3