> The biggest lever to achieving 42% accuracy was fine-tuning a Llama 2 (7B) mod... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		pants2 on Aug 23, 2024 \| parent \| context \| favorite \| on: Leveraging AI for efficient incident response > The biggest lever to achieving 42% accuracy was fine-tuning a Llama 2 (7B) model 42% accuracy on a tiny, outdated model - surely it would improve significantly by fine-tuning Llama 3.1 405B!

teleforce on Aug 25, 2024 [–]

Yes very interesting potential, it looks like it can be increased in accuracy considerably because Llama 3.1 with 405B parameters has very similar performance with the latest GPT-4o.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact