Merge LoRA with base model using 4-bit quantization and upload to HF 8826122 rwitz commited on 24 days ago