Sebastian Raschka
Sebastian Raschka
Thanks for the note. I just gave it a try and it worked: I believe the culprit may be that you are using watermark version 2.3 whereas the newest version...
Huh, this is really strange as scipy 1.10.1 seems to work for me as well:
I wish I could give you pointers as to why this happens. If you happen to find out what causes it, I'd be happy to hear. I'd also appreciate anyone's...
I wonder if it is possible to skip tests more elegantly than via the pytest CLI. I.e., I see we have the following `@RunIf` selector in the tests, like ```python...
> Maybe it's better to keep it in a Draft mode till all the testing is completed? Sure, I can move that to draft mode. But like I said in...
In case we want to pursue this, some [findings from Daniel Han](https://x.com/danielhanchen/status/1814317286389666094): My findings for Mistral NeMo 12b: 1. EOS token is untrained in base - a bug? 2. EOS...
Interesting, thanks for the anAlysis @Andrei-Aksionov . It's quite weird that QLoRA becomes worse for large microbatch sizes. I think this may potentially be related to #477 where a similar...
I think this is fixed now via #1899
So weird, I wasn't able to reproduce!
Thanks, I was out for 2 weeks and am just reading this. I may not have the bandwidth to address this immediately due to other issues on my list, but...