• Megatron-Turing NLG 530B, the World’s Largest Generative Language Model https://www.microsoft.com/en-us/research/blog/using-deepspeed-and-megatron-to-train-megatron-turing-nlg-530b-the-worlds-largest-and-most-powerful-generative-language-model/ 2 comments 17/10/2021
• Microsoft and Nvidia have created a 530B parameter language model https://www.microsoft.com/en-us/research/blog/using-deepspeed-and-megatron-to-train-megatron-turing-nlg-530b-the-worlds-largest-and-most-powerful-generative-language-model/ 2 comments 12/10/2021

• [D] Where did MT-NLG go wrong with their scaling experiments, comparing its capabilities to PaLM? https://www.microsoft.com/en-us/research/blog/using-deepspeed-and-megatron-to-train-megatron-turing-nlg-530b-the-worlds-largest-and-most-powerful-generative-language-model/ 6 comments 2/5/2022 machinelearning

Linking pages

Related searches:

Search whole site: site:microsoft.com

Search title: Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, the World’s Largest and Most Powerful Generative Language Model - Microsoft Research

See how to search.