Updated rule of thumb: a billion parameters * trillion tokens is $5,000 so gemma-2b was 2b parameters @ 6t tokens=$60,000. llama-3.1-405b (405b params * 15T tokens) cost ~$30,000,000.…
Posters: cis_female. cis_female
2 likes recorded by the source corpus.
View Post: https://x.com/cis_female/status/1820305397821112726