ParentFull threadEgoIncarnate·not "an H200", "In the table above, tensor parallelism is compared to pipeline parallelism with each across eight GPUs"View on HN