X · @teortaxesTex
· X / Twitter
Population genetics research test from Sasha I don't like his takes but I'll trust him on building a rigorous validation dataset. My guess is that gro…
Population genetics research test from SashaI don't like his takes but I'll trust him on building a rigorous validation dataset. My guess is that grok 4.5 is genuinely very good, but underperformance of Sol and Opus is more about sandbagging. Elon doesn't careSasha Gusev: @how_now00 Just finished: