qwopqwop200/GPTQ-for-LLaMa

Does not support 3bit quantization?

foamliu opened this issue · 0 comments

3-bit quant of a 65B model, encoutered following error during pack stage:

image