I am running an ONNX model through TensorRT.
I can verify that inference is running on the GPU through the results and nvsys profile logs.
However, I would like to see the corresponding PTX binary that TensorRT generates for my input model.
Is there a specific flag or argument to do so?
