I am testing AutoTrain and after finetuning meta-llama/Llama-2-13b-hf model on a very small set, I select to publish it to hub. It is published, but there is no “Deploy” or “Inference” option when I look at the model. It only has a general Model Trained Using AutoTrain description and no further options.
I am quite new to HF, an I missing something? I was expecting to have an option to quickly deploy and test finetuned model. Are there some extra steps before I can do this?
Thanks!