How to save the lora weight and config as the huggingface format?

[XuMayi]

How to save the lora weight and config as the following format: adapter_config.json adapter_model.bin

https://huggingface.co/LLMs/AlpacaGPT4-LoRA-13B-elina/tree/main

[XuMayi]

Leveraging the vllm to import lora requires these files and formats.

[bhimrazy]

Hi @XuMayi, this guide might be helpful for your use case: 👉 https://github.com/Lightning-AI/litgpt/blob/main/tutorials/convert_lit_models.md

It also includes steps for merging LoRA weights, which might be relevant for your scenario. Let us know if this helps!