1 个 GPU 不能工作, 2 个 GPU 无法使用内存

how to deal with the error below, 1*A100 PCIe 80gb. Followed the instruction with error below. 2*A100 80gb working fine but out of memory. I guess the code default to multiple GPUs. the only workable solution is 2*a100 80gb for Qwen/Qwen2.5-1.5B. For my training, Qwen/Qwen2.5-1.5B does not have very good training result, Qwen/Qwen2.5-3B with 2*H200 has very good training result.