runtime error

Exit code: 1. Reason: del-00005-of-00005.safetensors: 9%|▉ | 409M/4.60G [00:02<00:19, 214MB/s]  model-00005-of-00005.safetensors: 19%|█▊ | 860M/4.60G [00:03<00:11, 315MB/s] model-00005-of-00005.safetensors: 31%|███ | 1.43G/4.60G [00:04<00:07, 405MB/s] model-00005-of-00005.safetensors: 40%|████ | 1.85G/4.60G [00:05<00:06, 401MB/s] model-00005-of-00005.safetensors: 55%|█████▍ | 2.53G/4.60G [00:06<00:04, 491MB/s] model-00005-of-00005.safetensors: 66%|██████▌ | 3.03G/4.60G [00:07<00:03, 459MB/s] model-00005-of-00005.safetensors: 79%|███████▉ | 3.63G/4.60G [00:08<00:01, 499MB/s] model-00005-of-00005.safetensors: 90%|█████████ | 4.16G/4.60G [00:09<00:00, 508MB/s] model-00005-of-00005.safetensors: 100%|█████████▉| 4.60G/4.60G [00:10<00:00, 445MB/s] Traceback (most recent call last): File "/home/user/app/app.py", line 312, in <module> main() File "/home/user/app/app.py", line 307, in main model, processor = _load_model_processor(args) File "/home/user/app/app.py", line 65, in _load_model_processor model = Gemma3ForConditionalGeneration.from_pretrained(args.checkpoint_path, File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 311, in _wrapper return func(*args, **kwargs) File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 4752, in from_pretrained config = cls._autoset_attn_implementation( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2315, in _autoset_attn_implementation cls._check_and_enable_flash_attn_2( File "/usr/local/lib/python3.10/site-packages/transformers/modeling_utils.py", line 2466, in _check_and_enable_flash_attn_2 raise ValueError( ValueError: FlashAttention2 has been toggled on, but it cannot be used due to the following error: Flash Attention 2 is not available on CPU. Please make sure torch can access a CUDA device.

Container logs: