additional load_in_4bit removed

artidoro / qlora

QLoRA: Efficient Finetuning of Quantized LLMs

https://arxiv.org/abs/2305.14314

MIT License

9.74k stars 800 forks source link

additional load_in_4bit removed #295

Open shirinyamani opened 1 month ago

shirinyamani commented 1 month ago

In this PR the additional load_in_4bit=True is removed because in the quantization_config the load_in_4bit is already set to True, it cannot be duplicated!