Fine-Tuning Nous-Hermes2 with GradientAI

The goal of this repo is to get experience fine-tuning a Large Language Model (LLM) in this case Nous-Hermes2 to see how it performs on a specific task or domain. Then, use the fine-tuned LLM for inference via a chatbot user interface.

I chose the mental health domain. The dataset for this mental health conversational AI can be found here: heliosbrahma/mental_health_chatbot_dataset.

The model was fine-tuned in Python and deployed via a Flask API to a React frontend.