manav·vor 7 Monaten·discussLast time I had this problem, I got it fixed after applying for and accepting a job at Apple.
manav·vor 10 Monaten·discussWe're all training similarly large base++; near same data, just pricing it differently... with grok removing a few filters and maybe some safeguards? For that matter, many of the benchmarks are flawed and are just easily gamed and whatnot. iykyk.