模型过大，尝试卸载一部分模型但是失败了

Traceback (most recent call last):
File "/mnt/e/code/code_xuexi/github/MOSS-main/moss_cli_demo.py", line 33, in
model = load_checkpoint_and_dispatch(
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/accelerate/big_modeling.py", line 479, in load_checkpoint_and_dispatch
load_checkpoint_in_model(
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/accelerate/utils/modeling.py", line 924, in load_checkpoint_in_model
checkpoint = load_state_dict(checkpoint_file, device_map=device_map)
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/accelerate/utils/modeling.py", line 826, in load_state_dict
return torch.load(checkpoint_file)
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/torch/serialization.py", line 607, in load
return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/torch/serialization.py", line 882, in _load
result = unpickler.load()
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/torch/serialization.py", line 857, in persistent_load
load_tensor(data_type, size, key, _maybe_decode_ascii(location))
File "/root/anaconda3/envs/moss/lib/python3.8/site-packages/torch/serialization.py", line 845, in load_tensor
storage = zip_file.get_storage_from_record(name, size, dtype).storage()
OSError: [Errno 14] Bad address
ERROR conda.cli.main_run:execute(49): conda run python /mnt/e/code/code_xuexi/github/MOSS-main/moss_cli_demo.py failed. (See above for error)

Start with moss_cli_demo.py at line 33 and the load_checkpoint_and_dispatch call shown in the traceback. Reproduce it with the displayed conda-run command and inspect checkpoint loading through accelerate and torch; done means the model-loading failure is explained and a verified resolution is documented.