AutoModel failed with empty tensor error

[jiqing-feng]

System Info

Copy-and-paste the text below in your GitHub issue and FILL OUT the two last points.

• transformers version: 4.50.0.dev0
• Platform: Linux-4.18.0-553.16.1.el8_10.x86_64-x86_64-with-glibc2.35
• Python version: 3.10.12
• Huggingface_hub version: 0.28.1
• Safetensors version: 0.5.2
• Accelerate version: 1.4.0.dev0
• Accelerate config: - compute_environment: LOCAL_MACHINE
  - distributed_type: MULTI_CPU
  - mixed_precision: bf16
  - use_cpu: True
  - debug: False
  - num_processes: 4
  - machine_rank: 0
  - num_machines: 4
  - main_process_ip: 127.0.0.1
  - main_process_port: 29500
  - rdzv_backend: static
  - same_network: True
  - main_training_function: main
  - enable_cpu_affinity: False
  - ipex_config: {'ipex': False}
  - mpirun_config: {'mpirun_ccl': '1', 'mpirun_hostfile': '/home/jiqingfe/jiqing_hf/HuggingFace/tests/workloads/fine-tune/hostfile'}
  - downcast_bf16: no
  - tpu_use_cluster: False
  - tpu_use_sudo: False
  - tpu_env: []
• DeepSpeed version: not installed
• PyTorch version (GPU?): 2.6.0+cpu (False)
• Tensorflow version (GPU?): not installed (NA)
• Flax version (CPU?/GPU?/TPU?): not installed (NA)
• Jax version: not installed
• JaxLib version: not installed
• Using distributed or parallel set-up in script?:

Who can help?

@SunMarc @ArthurZucker @Rocketknight1

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

Run the following codes:

from transformers import AutoModel

model = AutoModel.from_pretrained("meta-llama/Llama-3.1-8B-Instruct", device_map="auto")

Error:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/jiqingfe/transformers/src/transformers/models/auto/auto_factory.py", line 564, in from_pretrained
    return model_class.from_pretrained(
  File "/home/jiqingfe/transformers/src/transformers/modeling_utils.py", line 271, in _wrapper
    return func(*args, **kwargs)
  File "/home/jiqingfe/transformers/src/transformers/modeling_utils.py", line 4535, in from_pretrained
    dispatch_model(model, **device_map_kwargs)
  File "/home/jiqingfe/accelerate/src/accelerate/big_modeling.py", line 496, in dispatch_model
    model.to(device)
  File "/home/jiqingfe/transformers/src/transformers/modeling_utils.py", line 3262, in to
    return super().to(*args, **kwargs)
  File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1343, in to
    return self._apply(convert)
  File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 903, in _apply
    module._apply(fn)
  File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 930, in _apply
    param_applied = fn(param)
  File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1336, in convert
    raise NotImplementedError(
NotImplementedError: Cannot copy out of meta tensor; no data! Please use torch.nn.Module.to_empty() instead of torch.nn.Module.to() when moving module from meta to a different device.

Expected behavior

Expect got a base model.

[SunMarc]

Fixed above. thanks for the report