1. Mapping Model Behavior to Internal Representations
4 leccionesLearn to observe and measure what happens inside a model between input and output.
2. Attribution Methods for Input Influence
5 leccionesIdentify which parts of an input most strongly affect model outputs.
3. Probing and Feature Detection
5 leccionesTest whether internal activations encode specific concepts, patterns, or behaviors.
4. Causal Intervention and Circuit Analysis
5 leccionesChange internal model states to test whether identified features actually control behavior.
5. Diagnosing Failure Modes in Production Systems
5 leccionesApply interpretability to real debugging scenarios where models behave unexpectedly.
6. Building Auditable Explanations for Governance
5 leccionesCreate evidence-based explanations that support risk assessment and compliance requirements.
7. Integrating Interpretability into Development Workflows
5 leccionesEmbed interpretability practices into model evaluation, fine-tuning, and deployment pipelines.
Recibe un correo cuando abra este curso