Inference using fine-tuned model
1. Inference using fine-tuned model
In the last video, we took our generally trained LLM and sent it off to school, fine-tune school. Now, it's time to put the model to work and see how it does. If you're not logged into your Snowflake account, please pause the video now to log into your account. Navigate to the Projects tab on the left panel and select Notebooks. Click on the Fine-tuning Mistral 7B notebook and select the Start button at the top right to initiate the notebook session. We had already run the notebook until the fine-tuning section, but the notebook session ended, so we need to run the import statements again. And we'll also run the second cell to use this database and schema for the notebook. And we'll also run the prompt and ticket categories definition. Next up, we will invoke the fine-tuned Mistral 7B model to generate custom responses to support tickets based on our customer's contact preference. Look at the snippet of code in Inference Using Fine-Tuned Model section. We invoke the cortex complete function with the name of the fine-tuned model instead of Mistral 7B or Mistral Large. Looking at the inference output, it is clear that the fine-tuned model is able to generate custom responses in the desired style for our customer support tickets. Well done! In this video, we ran inference on the fine-tuned Mistral 7B model using cortex complete. We reviewed that the model responses are in line with our customer's preferred mode of contact. Fantastic! In the next video, we will be looking at how to build a Streamlit UI interface for the AI app we built so far. That is, we will learn how to auto-generate custom responses using LLMs and share that information using a Streamlit app with our marketing team. See you then!2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.