10-19-2026, 02:15 AM
Quote:[quote]I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. A lot of these
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.
Mostly CUDA, sometimes wrong, usually measured