Re:Google's First Tensor Processing Unit: Architecture
by HarHarVeryFunny on Tuesday March 26, 2024 @11:28AM
(#39829133)
There's also Amazon's AWS "Trainium" chips, which is what Anthropic will be using going forward.
If you're talking about training LLMs, involving 10's of thousands of processors, then the specifics of one processor vs another isn't the most important thing - it's the overall architecture and infrastructure in place to manage it.
If you're talking about training LLMs, involving 10's of thousands of processors, then the specifics of one processor vs another isn't the most important thing - it's the overall architecture and infrastructure in place to manage it.