“We keep on to discover hyperscaling of AI models leading to far better general performance, with seemingly no finish in sight,” a pair of Microsoft researchers wrote in Oct inside of a blog publish saying the company’s substantial Megatron-Turing NLG model, in-built colla