not at all comparable. it's just a scaled up GPT2, no new ideas deep learning wi... | Hacker News

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

make3 on Feb 10, 2020 | parent | context | favorite | on: Turing-NLG: A 17B-parameter language model

not at all comparable. it's just a scaled up GPT2, no new ideas deep learning wise.

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