You should be able to first train it on generic text once, then duplicate the in... | Hacker News

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

vanviegen 74 days ago | parent | context | favorite | on: Claude mixes up who said what

You should be able to first train it on generic text once, then duplicate the input layer and fine-tune on conversation.

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