but even then it is not so trivial. Yesterday I gave DeepSeek a simple diophanti... | Hacker News

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

		kavalg on Feb 7, 2025 \| parent \| context \| favorite \| on: Understanding Reasoning LLMs but even then it is not so trivial. Yesterday I gave DeepSeek a simple diophantine equation and it got it wrong 3 times, tried to correct itself and didn't end on a correct solution, but rather lied that the final solution is correct.

wolfgangK on Feb 7, 2025 | [–]

DeepSeek is not a model.Which model did you use (v3 ? R1 ? a distillation ?) at which quantization ?

Synaesthesia on Feb 7, 2025 | [–]

Did you use the full version? And did you try R1?

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