Benjamin Bossan

> Yes, I'll fix the comments tonight Thanks. No need to rush, I just wanted to inquire if you're still on it :)

@moghadas76 LMK once you're finished with your changes and want me to do another review.

Gentle ping @moghadas76

Thanks for investigating this issue. Could you provide a bit more information so that we can reproduce this issue (ideally with a smaller model, so that we can create a...

Sorry for the delay, I was at a conference, will review soon.

> Thanks! I'm using bf16. Any broadly Llama-like model should work, eg. I've run tests with Meta-Llama-3-8B and Mistral-7B-v0.1 (the original, non-MoE Mistral). Thanks for the info. I still couldn't...

@rationalism gentle ping, do you have updates?

At the risk of repeating myself the 5th time, could you please make the following change: > Great, thanks. On top of what I mentioned, let's also move this to...

That would be great, thanks a lot!

Hmm, code quality checks are still failing with: > tests/test_loraplus_helper.py:1:1: I001 [*] Import block is un-sorted or un-formatted Is it possible that your local ruff version differs? CI uses v0.2.2.