this repo support 2bit finetuning the llama model？ Is there any case to show how to run the scripts?

[zlh1992]

By the way is there any 2-bit model weights of llama7b,13b,33b,65b?

[johnsmith0031]

You can use load_llama_model_4bit_low_ram and set bits=2. And I haven't seen any 2 bits weight for llama model as well, also curious about it.