runtime error

Exit code: 1. Reason: 0/4.93G [00:00<?, ?B/s] model-00003-of-00004.safetensors: 10%|▉ | 493M/4.93G [00:01<00:09, 465MB/s] model-00003-of-00004.safetensors: 40%|███▉ | 1.95G/4.93G [00:02<00:02, 1.03GB/s] model-00003-of-00004.safetensors: 71%|███████▏ | 3.52G/4.93G [00:03<00:01, 1.27GB/s] model-00003-of-00004.safetensors: 100%|█████████▉| 4.93G/4.93G [00:03<00:00, 1.25GB/s] Downloading shards: 75%|███████▌ | 3/4 [00:12<00:04, 4.22s/it] model-00004-of-00004.safetensors: 0%| | 0.00/1.69G [00:00<?, ?B/s] model-00004-of-00004.safetensors: 32%|███▏ | 538M/1.69G [00:01<00:02, 532MB/s] model-00004-of-00004.safetensors: 100%|█████████▉| 1.69G/1.69G [00:01<00:00, 1.02GB/s] Downloading shards: 100%|██████████| 4/4 [00:14<00:00, 3.32s/it] Downloading shards: 100%|██████████| 4/4 [00:14<00:00, 3.66s/it] `Qwen2VLRotaryEmbedding` can now be fully parameterized by passing the model config through the `config` argument. All other arguments will be removed in v4.46 Traceback (most recent call last): File "/home/user/app/app.py", line 13, in <module> "OS-Copilot/OS-Atlas-Base-7B": Qwen2VLForConditionalGeneration.from_pretrained("OS-Copilot/OS-Atlas-Base-7B", torch_dtype="auto", device_map="auto"), File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 4189, in from_pretrained max_memory = get_balanced_memory( File "/usr/local/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 957, in get_balanced_memory max_memory = get_max_memory(max_memory) File "/usr/local/lib/python3.10/site-packages/accelerate/utils/modeling.py", line 825, in get_max_memory _ = torch.tensor([0], device=i) File "/usr/local/lib/python3.10/site-packages/spaces/zero/torch/patching.py", line 146, in __torch_function__ device = torch.device(device) RuntimeError: Cannot access accelerator device when none is available.

Container logs: