Then why not read their paper? "The largest Llama 2-Chat model is competitive wi... | Hacker News

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

jmiskovic on July 18, 2023 | parent | context | favorite | on: Llama 2

Then why not read their paper?

"The largest Llama 2-Chat model is competitive with ChatGPT. Llama 2-Chat 70B model has a win rate of 36% and a tie rate of 31.5% relative to ChatGPT."

capableweb on July 18, 2023 [–]

Do they specify which GPT version they used? Could Llama 2 really beat GPT-4?

jmiskovic on July 18, 2023 | | [–]

The 70B Llama2 model ties in with 173B ChatGPT-0301 model. The GPT-4 still stands unchallenged.

sebzim4500 on July 18, 2023 | | | [–]

Source on the 173B parameters?

exradr on July 19, 2023 | | | [–]

The wikipedia article for GPT-4 has this as its source: https://the-decoder.com/gpt-4-architecture-datasets-costs-an...

jmiskovic on July 20, 2023 | | | | [–]

It's actually 175B. https://arxiv.org/pdf/2005.14165.pdf

davidkunz on July 18, 2023 | | [–]

They used ChatGPT-0301, it can't beat GPT-4.

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