Closed taehyunzzz closed 1 month ago
If you refer to https://github.com/EleutherAI/lm-evaluation-harness then pretty much any task should be supported I think? It may not perform as well on non-English task.
I think you'd want to use the base model there.
First of all, thank you for sharing your work. I was wondering which generative tasks are supported by the open-sourced pretrained weights? I want to test them out using LLM Harness. Also, to use in LLM Harness generative tasks, which one of the base/Instruct/SFT versions should I use?
Taehyun