Yu Ju, Lin

Results 3 comments of Yu Ju, Lin

I encounter the issue today, not quite into the details of all these script. But I here proposed a method at least you can try before you can go in...

我也遇到类似的问题training loss 变成zero,请问为什么改成padding to right可以解决问题

having same issue for custom model (i build from GGUF file) while work without problems with library models