CryptoWorld News: Sakana AI has partnered with NVIDIA to open-source a sparse data format called twell and its supporting acceleration kernel, successfully enabling GPUs to skip "near-zero" ineffective calculations when running large models. This solution, without sacrificing model accuracy, boosts inference speed on H100 by up to 30%, accelerates training by up to 24%, and significantly reduces peak video memory. Data also reveals a pattern: the larger the model parameters, the more dormant neurons there are (the non-zero ratio of a 2 billion parameter model is 38% lower than that of a 500 million parameter model). This means that as future efforts pursue larger-scale models, this hardware-level optimization will unlock more substantial performance benefits.