Fine-tuning CodeLlama-7B on Synthetic Training Data for Fortran Code Generation using PEFT A Stanford CS 224N Final Project submission. Selected as one of the outstanding custom projects. Final project report