The Swiss Federal Institute of Technology in Lausanne (EPFL), the Federal Institute of Technology in Zurich (ETH Zurich), and the Swiss National Supercomputing Center (CSCS) have teamed up to release a large-scale open-source language model called Apertus. Apertus means "open" in Latin, and it does take the word "open" to the extreme. While large American models such as OpenAI's GPT series, Meta's Llama, and Anthropic's Claude were still playing "data black box" magic, Apertus learned from model...
瑞士联邦理工学院洛桑分校 (EPFL)、苏黎世联邦理工学院 (ETH Zurich) 与瑞士国家超级计算中心 (CSCS) 联手发布一款名为Apertus的大规模开源语言模型。Apertus在拉丁语里是“开放”的意思,而它也确实把“开放”这两个字做到了极致。在OpenAI的GPT系列、Meta的Llama、Anthropic的Claude这些美国大模型还在玩“数据黑箱”魔术的时候,Apertus从模型权重、架构、训练...