Better device_map and max_memory for loading vicuna model

Hi! Thank you for this wonderful repo.
When I was trying to load vicuna model with limited VRAM across different GPUs. I discovered that your "max_memory" part would cause the loading to fail:
in model_adapter.py line 219 to 231.

        if num_gpus != 1:
            kwargs["device_map"] = "auto"
            if max_gpu_memory is None:
                kwargs[
                    "device_map"
                ] = "sequential"  # This is important for not the same VRAM sizes
                available_gpu_memory = get_gpu_memory(num_gpus)
                kwargs["max_memory"] = {
                    i: str(int(available_gpu_memory[i] * 0.85)) + "GiB"
                    for i in range(num_gpus)
                }
            else:
                kwargs["max_memory"] = {i: max_gpu_memory for i in range(num_gpus)}

If comment "available_gpu_memory" and "kwargs["max_memory"]", the loading would succeed.
I wonder why you compute max_memory here and use it as one of kwargs to load the model, since this could cause the failure of loading.

Thanks again for your time and effort!

Start in model_adapter.py around lines 219-231 and reproduce Vicuna loading with limited VRAM across GPUs. Compare the loading behavior with and without the available_gpu_memory and max_memory handling, then establish a configuration that loads successfully without breaking the intended device_map behavior.