Curious how big your dataset was if you used $1000 of GPU credits on DistilBERT.... | Hacker News

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

		Karrot_Kream 10 months ago \| parent \| context \| favorite \| on: When ChatGPT broke the field of NLP: An oral histo... Curious how big your dataset was if you used $1000 of GPU credits on DistilBERT. I've run BERT on CPU on moderate cloud instances no problem for datasets I've worked with, but which admittedly are not huge.

keyserj 10 months ago [–]

If I'm reading correctly, they used $1000 running a Llama model, not DistilBERT.

teruakohatu 10 months ago | [–]

You read it correctly. I obviously didn't explain myself well.

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