tinylamb

Results 2 comments of tinylamb

这个项目很好啊

> Getting this error while using single A100 8G0GB while loading llama-7b > > I tried reducing the batch size also changes the **--gradient_accumulation_steps** but not able to work it...