Hacker News new | ask | show | jobs
by cherioo 191 days ago
GPT4.5 was allegedly such a pre-train. It just didn’t perform good enough to announce and product it as such.
1 comments
htrp 191 days ago
it wasn't economical to deploy but i expect it wasn't wasted, expect the openai team to pick that back up at some point
link
mips_avatar 191 days ago
The scoop Dylan Patel got was that part way through the gpt4.5 pretraining run the results were very very good, but it leveled off and they ended up with a huge base model that really wasn't any better on their evals.
link