Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
I Trained a Small Language Model from Scratch (nwosunneoma.medium.com)
8 points by Ada-Ihueze 4 months ago | hide | past | favorite | 3 comments


So, where is evaluation? How often are the answers nonsensical? What fraction of common customer questions can it answer? How does LLM conpare on that?

Without those answerw, the article is meaningless.


The title in the article is 'How I trained...' yet there is no explanation of how it was done, just that it was.


smells




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

Search: