Files

Gas Town→Term -- Boot the Dog→having LLM judges check for discrete signals→scaled evaluation

⇩ sources

Search this bundle

Enter at least two characters.

scaled evaluation

diff - scaled evaluation vs. binary pass or fail evaluation

Like Likert scale which is 1-5

Backlinks

diff - scaled evaluation vs. binary pass or fail evaluation

^ diff - scaled evaluation vs. binary pass or fail evaluation

see in context
having LLM judges check for discrete signals

This is kind of the opposite of trying to grade based on generic evals, or even scaled evaluation like Likert scale.

see in context