The next step is to start Triton Inference Server so that you can generate synthetic data by sending inference requests to the model.
Within your Jupyter lab tab, open Start Inference Server notebook.
Once you are done running through the notebook, proceed with Step #6.