Is there any plan to retry this with facebook’s LLaMA models? They claim it outperforms GPT-3. Since it’s open for research, it might give more precise answers to some of the questions.
Current theme: default
Less Wrong (text)
Less Wrong (link)
Is there any plan to retry this with facebook’s LLaMA models? They claim it outperforms GPT-3. Since it’s open for research, it might give more precise answers to some of the questions.