ColossalAI
ColossalAI copied to clipboard
[BUG]: RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.HalfTensor [1, 1, 512, 128]] is at version 3; expected version 2 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).
๐ Describe the bug
no
Environment
No response
ไฝ ่งฃๅณ่ฟไธช้่ฏฏไบๅ
Bot detected the issue body's language is not English, translate it automatically. ๐ฏ๐ญ๐ป๐งโ๐คโ๐ง๐ซ๐ง๐ฟโ๐คโ๐ง๐ป๐ฉ๐พโ๐คโ๐จ๐ฟ๐ฌ๐ฟ
did you fix this error
ๆบๅจไบบๆฃๆตๅฐ้ฎ้ขๆญฃๆ็่ฏญ่จไธๆฏ่ฑ่ฏญ๏ผ่ฏท่ชๅจ็ฟป่ฏใ๐ฏ๐ญ๐ป๐งโ๐คโ๐ง๐ซ๐ง๐ฟโ๐คโ๐ง๐ป๐ฉ๐พโ๐คโ๐จ๐ฟ๐ฌ๐ฟ
ๆจๆฏๅฆไฟฎๅคไบๆญค้่ฏฏ
ๆฒกๆ
Bot detected the issue body's language is not English, translate it automatically. ๐ฏ๐ญ๐ป๐งโ๐คโ๐ง๐ซ๐ง๐ฟโ๐คโ๐ง๐ป๐ฉ๐พโ๐คโ๐จ๐ฟ๐ฌ๐ฟ
The robot has detected that the language of the question text is not English, please translate it automatically. ๐ฏ๐ญ๐ป๐งโ๐คโ๐ง๐ซ๐ง๐ฟโ๐คโ๐ง๐ป๐ฉ๐พโ๐คโ๐จ๐ฟ๐ฌ๐ฟ
Did you fix this bug
No
Hi @liulihui123 Could you please provide more details to help us to reproduce this issue? The information is now too limited to locate. Thanks.