Re:Nvidia Hopper Sweeps AI Inference Benchmarks in MLPerf Debut
by riku_iki on Friday September 09, 2022 @02:37AM
(#32776021)
> - AlphaCode (from February, reasonably high success rate on solving competitive programming problems)
reasonably high was 50% on 10 attempts, meaning success rate on first attempt can be as low as 5%, out of which who knows how many were leaked to training data.
reasonably high was 50% on 10 attempts, meaning success rate on first attempt can be as low as 5%, out of which who knows how many were leaked to training data.