Finetune bigScience-bloom-1B7 model with Persian News Summary You can check all the steps in the notebook or you can run the notebook on kaggle if you want to test it out Because this model is big enough to make most of GPUs out-of-memory, I used LoRA and 4-bit Quantization
