这个是因为显存不够还是什么？

[af-74413592]

│ /home/ubuntu/.cache/huggingface/modules/transformers_modules/fnlp/moss-moon- │ │ 003-sft/dc39155402a323c3ad85ac9c4cb02556c61594f1/modeling_moss.py:53 in │ │ apply_rotary_pos_emb │ │ │ │ 50 │ │ 51 # Copied from transformers.models.gptj.modeling_gptj.apply_rotarypos │ │ 52 def apply_rotary_pos_emb(tensor: torch.Tensor, sin: torch.Tensor, cos: │ │ ❱ 53 │ sin = torch.repeat_interleave(sin[:, :, None, :], 2, 3) │ │ 54 │ cos = torch.repeat_interleave(cos[:, :, None, :], 2, 3) │ │ 55 │ return (tensor cos) + (rotate_every_two(tensor) sin) │ │ 56 │ ╰──────────────────────────────────────────────────────────────────────────────╯ RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

[WanJuWuGo]

版本不对。安装新版本就行 pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117

[piglaker]

请问您解决了吗，

[wktdwktd]

请问您解决了吗，

我按照这个 解决了 pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117

[baigang666]

请问您解决了吗，

我按照这个 解决了 pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117

请问我就是图中这个版本，在A100上尝试，几次问答后，依然报同样的错误。请问怎么解决