Benjamin Bossan

Thanks for the PR @ved1beta. Unfortunately, there is an issue with rendering many of the notebooks in the browser, including [`main.ipynb`](https://github.com/huggingface/peft/blob/dd0c3f93ed6c5caa4c36afe89904dd420608907d/Zmybenchmarks/main.ipynb). Could you please check if this can be fixed?...

> i dont know what is the issue with ipynb files , here is the google collab link for it [link](https://colab.research.google.com/drive/113IQw8ORbMzUN1e0GwtJ422bJp2JLDzs?usp=sharing) Thanks for the link. I see that there is...

Thank you for investigating the runtime performance of LoRA. Regarding your results, I'm not sure if this is an apples to apples comparison, especially since you didn't show how exactly...

@leexcape did you have time to look into this further?

Thanks for the report. Merging works for me, it's most likely an issue with a package version. Which versions of torch, bitsandbytes and transformers do you use? If not the...

I agree, it looks like it could make sense to consider the scaling factor when applying LoftQ. @fxmeng could you comment on that?

Thanks a lot for reporting. OLoRA should work with quantization, but as your reproducer shows, there is something amiss. Interestingly, when using the [OLoRA example script](https://github.com/huggingface/peft/blob/main/examples/olora_finetuning/olora_finetuning.py) (which is not all...

Gentle ping again @fxmeng

I did some further research, though unfortunately I still haven't found the solution. One difference that I spotted between the scripts is the usage of `bnb_4bit_quant_storage`, but even when removed,...

Another ping @fxmeng.