ConceptAI & ML
AlpacaEval
An LLM-as-a-judge evaluation tool comparing a candidate model's single-turn response against a baseline model on a standard set of prompts.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Instruction tuning1LLM-as-a-judge1
Lecturers
1Connections
2The lectures that define it or assume it, and the concepts next to it.