freeze LM

[puririshi98]

goal^

trying the frozen pathway
when i try the PR it causes:

  File "/opt/pyg/pytorch_geometric/examples/llm/tech_qa.py", line 325, in <module>
    model, test_loader = train(args, data_lists)
                         ^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/pyg/pytorch_geometric/examples/llm/tech_qa.py", line 270, in train
    loss.backward()
  File "/usr/local/lib/python3.12/dist-packages/torch/_tensor.py", line 626, in backward
    torch.autograd.backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/__init__.py", line 347, in backward
    _engine_run_backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/graph.py", line 823, in _engine_run_backward
    return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn

I want the gradients to flow through the LLM but i dont want its params updated (and any memory savings possible). what is the easiest/recommended way to do this.
initiating the model w/:
llm = LLM(model_name=args.llm_generator_name, freeze=True).eval()