evals
/ˈivælz/
repeatable tests that measure model output quality
“The team ran evals to check the model's accuracy.”
เพิ่มลงใน My Deck
inoh
inoh
/ˈivælz/
repeatable tests that measure model output quality
“The team ran evals to check the model's accuracy.”
เพิ่มลงใน My Deck