About how to use

[RIKI0913]

I want to use human-eval to get the correct code and program code generated by LLM.
I am able to build the environment, but do not know how to execute it (including selecting the LLM).
If you could please tell me how to do this?

[sarakh1999]

you can identify which model you want to use and the logic you want to follow in the generate_one_completion module indicated in the readme.