#Nemotron-H

2 stories · 0 in 7 days · 1 in 30 days

Stories, products and papers tagged "Nemotron-H", newest first.

Oct 16

训练 LLM 显存吃紧的可以看看：BitSparse 靠 ReLU 稀疏性无损压缩激活缓存，NanoGPT 和 Nemotron-H 上实测省三成显存，几乎不拖慢速度。

arXiv cs.LG@Max Zhu et al. (4)Source

Sep 3

PapersSelected73°11:51

This method improves efficiency for large language model training with quantization, potentially reducing costs for AI developers in the region.

arXiv cs.LG@Robert Hu et al. (3)Source