Very basic framework for parameterized large language model (Q)LoRa fine-tuning using mlx, mlx_lm, and OgbujiPT. Architecture for systematic running of easily parameterized fine-tunes - View it on GitHub