LLM evaluation & fine-tuning

Question 1 of 8

What is a golden dataset in LLM work?

Question 2 of 8

What is "LLM-as-a-judge"?

Question 3 of 8

What does LoRA do in fine-tuning?

Question 4 of 8

When does fine-tuning beat prompting?

Question 5 of 8

During fine-tuning, training loss keeps falling while validation loss starts rising. This is:

Question 6 of 8

What does 4-bit quantisation do?

Question 7 of 8

Perplexity measures:

Question 8 of 8

What is benchmark contamination?