Fine-Tuning a model to build an AI assistant for hosting

This project fine-tunes the Falcon 7B language model for FAQs in the travel and accommodation domain. Major steps include downloading the model and tokenizer, quantizing to optimize memory usage, freezing model parameters, and fine-tuning using LoRA to specialize in domain-specific responses. The model will be trained on preprocessed data and checked for performance using the BLEU metric. The resulting chatbot provides quick and accurate answers, and therefore is one great tool to automate customer support.