关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
OpenI
/
aiforge
Not watched
33
71
Fork 76
Code
Releases 61 Wiki Activity
Issues 271 Pull Requests 1 Datasets Cloudbrain
Pull Requests
/
Detail
Labels Milestones
New Pull Request

#4740 update mind page

Merged
ychao_1983 merged 3 commits from dev-csh into V20230912 7 months ago
Conversation 0 Commits 3 Files Changed 1
1 changed files with 1 additions and 1 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. +1
    -1
      web_src/vuepages/pages/model/mind/index.vue

+ 1
- 1
web_src/vuepages/pages/model/mind/index.vue View File

@@ -232,7 +232,7 @@ export default {
tabList: ['PengCheng-Mind 200B', 'PengCheng-Mind 7B'],
tabContent: [{
descr: [
`PengCheng-Mind 200B 脑海大模型是基于Transformer 架构的2010亿参数自回归式语言模型。模型全流程基于“鹏城云脑Ⅱ”（中国算力网枢纽节点）的全自主安全可控国产软硬件平台进行开发和训练，采用MindSpore框架实现在大规模集群上长期稳定的多维分布式并行训练。训练模型保存为3456个切片，容量约7.6TB，推理模型保存为16个切片，容量约375GB。支持4096个token的上下文窗口长度，主要聚焦中文核心能力，兼顾英文和部分多语言能力。当前模型已完成训练1T Tokens数据量，仍在持续训练迭代中。`,
`PengCheng-Mind 200B 脑海大模型是基于Transformer 架构的2010亿参数自回归式语言模型。模型全流程基于中国算力网枢纽节点（鹏城云脑Ⅱ）的全自主安全可控国产软硬件平台进行开发和训练，采用MindSpore框架实现在大规模集群上长期稳定的多维分布式并行训练。训练模型保存为3456个切片，容量约7.6TB，推理模型保存为16个切片，容量约375GB。支持4096个token的上下文窗口长度，主要聚焦中文核心能力，兼顾英文和部分多语言能力。当前模型已完成训练1T Tokens数据量，仍在持续训练迭代中。`,
`大模型的开发充满未知与挑战，随着中国算力网的不断发展壮大，希望通过模型训练全流程开源开放的方式，将鹏城·脑海200B 模型培育为算力网上一个种子应用，吸引产学研界更多人参与到通用大语言模型基座的研究和应用中，一起助力国产智能软硬件生态的发展。`,
],
stage: [


Write Preview
Loading…
Cancel
Save
Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit