This project focuses on developing a customer support chatbot using the LLaMA-2 7B (7 billion parameters) model, fine-tuned to enhance its performance in understanding and responding to customer inquiries. The chatbot aims to provide efficient and informative assistance, effectively addressing customer needs and resolving issues promptly.

Model Used: LLaMA-2 7B (7 Billion Parameters)

Fine-tuning: The LLaMA-2 7B model was fine-tuned using the QLoRA technique on the tsp1718/customer-support-llama2-1k-only-text dataset. This fine-tuning significantly improves the model's ability to handle customer support queries with accuracy and relevance. The resulting fine-tuned model, tsp1718/llama-2-customer-support, is now available on Hugging Face.