Advanced32 min
Evaluations and Red-Teaming
Designing evals that measure dangerous capability rather than benchmark skill: capability versus propensity, elicitation, and why contamination makes most numbers unreliable.
Assumes you know
Not yet written
This lesson is on the syllabus but has no text yet
The full curriculum is published up front so you can see the whole route and its dependencies. Lessons are being written in curriculum order.
What it will cover
- evals
- red-teaming
- dangerous capability
- elicitation
- contamination