Error while deserializing header: HeaderTooLarge

[jack161641]

跑ReadMe的Demo code都报错，该怎么解决？

SafetensorError Traceback (most recent call last)
Cell In[4], line 5
3 model_path = '/home/admin/Atom-7B-Chat'
4 device_map = "cuda:0" if torch.cuda.is_available() else "auto"
----> 5 model = AutoModelForCausalLM.from_pretrained(model_path,device_map=device_map,torch_dtype=torch.float16,load_in_8bit=True,trust_remote_code=True,use_flash_attention_2=True)
6 model =model.eval()
7 tokenizer = AutoTokenizer.from_pretrained(model_path,use_fast=False)

File ~/anaconda3/envs/py3.8/lib/python3.8/site-packages/transformers/models/auto/auto_factory.py:556, in _BaseAutoModelClass.from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs)
554 else:
555 cls.register(config.class, model_class, exist_ok=True)
--> 556 return model_class.from_pretrained(
557 pretrained_model_name_or_path, *model_args, config=config, **hub_kwargs, **kwargs
558 )
559 elif type(config) in cls._model_mapping.keys():
560 model_class = _get_model_class(config, cls._model_mapping)

File ~/anaconda3/envs/py3.8/lib/python3.8/site-packages/transformers/modeling_utils.py:3502, in PreTrainedModel.from_pretrained(cls, pretrained_model_name_or_path, config, cache_dir, ignore_mismatched_sizes, force_download, local_files_only, token, revision, use_safetensors, *model_args, **kwargs)
3493 if dtype_orig is not None:
3494 torch.set_default_dtype(dtype_orig)
3495 (
3496 model,
3497 missing_keys,
3498 unexpected_keys,
3499 mismatched_keys,
3500 offload_index,
3501 error_msgs,
-> 3502 ) = cls._load_pretrained_model(
3503 model,
3504 state_dict,
3505 loaded_state_dict_keys, # XXX: rename?
3506 resolved_archive_file,
3507 pretrained_model_name_or_path,
3508 ignore_mismatched_sizes=ignore_mismatched_sizes,
3509 sharded_metadata=sharded_metadata,
3510 _fast_init=_fast_init,
3511 low_cpu_mem_usage=low_cpu_mem_usage,
3512 device_map=device_map,
3513 offload_folder=offload_folder,
3514 offload_state_dict=offload_state_dict,
3515 dtype=torch_dtype,
3516 hf_quantizer=hf_quantizer,
3517 keep_in_fp32_modules=keep_in_fp32_modules,
3518 )
3520 # make sure token embedding weights are still tied if needed
3521 model.tie_weights()

File ~/anaconda3/envs/py3.8/lib/python3.8/site-packages/transformers/modeling_utils.py:3903, in PreTrainedModel._load_pretrained_model(cls, model, state_dict, loaded_keys, resolved_archive_file, pretrained_model_name_or_path, ignore_mismatched_sizes, sharded_metadata, _fast_init, low_cpu_mem_usage, device_map, offload_folder, offload_state_dict, dtype, hf_quantizer, keep_in_fp32_modules)
3901 if shard_file in disk_only_shard_files:
3902 continue
-> 3903 state_dict = load_state_dict(shard_file)
3905 # Mistmatched keys contains tuples key/shape1/shape2 of weights in the checkpoint that have a shape not
3906 # matching the weights in the model.
3907 mismatched_keys += _find_mismatched_keys(
3908 state_dict,
3909 model_state_dict,
(...)
3913 ignore_mismatched_sizes,
3914 )

File ~/anaconda3/envs/py3.8/lib/python3.8/site-packages/transformers/modeling_utils.py:505, in load_state_dict(checkpoint_file)
500 """
501 Reads a PyTorch checkpoint file, returning properly formatted errors if they arise.
502 """
503 if checkpoint_file.endswith(".safetensors") and is_safetensors_available():
504 # Check format of the archive
--> 505 with safe_open(checkpoint_file, framework="pt") as f:
506 metadata = f.metadata()
507 if metadata.get("format") not in ["pt", "tf", "flax"]:

SafetensorError: Error while deserializing header: HeaderTooLarge

[zhushuaiCoding]

解决了吗，我也有这个问题