Ask what should change when the equation is manipulated, then let the visualization test that expectation.
Foundation Lab
Model-Graded Evaluations
Enables scalable safety testing without human bottleneck
#92LLM-as-JudgeScaling & Alignment
key equation
\hat{\mu} = \frac{1}{n} \sum_i E(x_i, y_i; R)Selected Foundation Object
Keep the equation fixed; move through the evidence.
Concept 92 of 100LLM-as-JudgeScaling & Alignment / Phase 12: Advanced alignment & safety research
Current question
Rubric defines what "good" means: helpfulness, truthfulness, safety
\hat{\mu} = \frac{1}{n} \sum_i E(x_i, y_i; R)Use the runnable panel, the key equation, and canonical papers as separate forms of evidence for the same object.
The useful learning product is the reusable mechanism you can carry into another model, paper, or engineering tradeoff.
Use prerequisites, dependents, and semantic links to repair the next gap without leaving the object behind.
Why It Matters for Modern Models
- Enables scalable safety testing without human bottleneck
- Fast iteration loops for alignment research
- Powers modern benchmarks: Chatbot Arena, AlpacaEval
What Tutorials Skip
What is still poorly explained in textbooks and papers:
- Rubric defines what "good" means: helpfulness, truthfulness, safety
- Calibration: does model-graded score match human judgment?
- Position bias: models prefer first option—randomize order
Interactive Visualization
Core Math (Optional Deep Dive)
If you want intuition first, start with the key equation and the visualization. Come back here for the full walkthrough.
Key Equation
Evaluator model scores outputs against rubric :
Aggregate:
Validate by correlating with human ratings. Track regressions across model versions.