This also means local fine-tuning is possible. Expect to see an explosion of new... | Hacker News

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

		benlivengood on Feb 20, 2023 \| parent \| context \| favorite \| on: Running large language models like ChatGPT on a si... This also means local fine-tuning is possible. Expect to see an explosion of new things like we did with Stable Diffusion, limited to some extent by the ~0.7 order of magnitude more VRAM required.

bioemerl on Feb 20, 2023 [–]

Does it? I would have expected compression losses to make training really hard.

Miraste on Feb 20, 2023 | [–]

The compression is optional.

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