启用fp16, load_in_8bit设置为False, 报错： RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! - Githubissues

mymusise / ChatGLM-Tuning

基于ChatGLM-6B + LoRA的Fintune方案

MIT License

3.71k stars 443 forks source link

启用fp16, load_in_8bit设置为False, 报错： RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! #244

Open guotong1988 opened 1 year ago

yuanzhiyong1999 commented 1 year ago

请问解决了吗