who's fine-tuning LLMs for reasoning? This dataset has been trending for a few weeks and there's a list of models trained on it.

- It has SFT formatted reasoning sequences, like those in o1.
- You could incorporate these into post training to boost reasoning abilities.

came across this example in agent-as-a-judge from Meta. It uses agent-as-a-judge to evaluate the effectiveness of a DevAI app.

- It's based on an open dataset.
- It's more accurate than LLM as a judge
- It explains its evaluation based on preferences, and requirements.

https://buff.ly/49tN6CQ

Learn local and private LLMs with Hugging Face 🤗 : Participation is free so join now!

Even better, there are minimal GPU requirements and no paid services.

Start with, Instruction Tuning, Preference Alignment, Parameter-efficient Fine-tuning.

GitHub: https://buff.ly/3ZCMKX2

I've just contributed 10 examples to this dataset:

data-is-better-together-fineweb-c.hf.space/share-your-p...

The great @benburtenshaw.bsky.social is running an open course on fine-tuning smol LLMs, and it’s seriously worth checking out.

If you’re into AI or just curious about how these small language models work, this could be right up your alley. Don’t miss it—it’s super interesting!

#AI #LLMs #Learning

👐 Open Image Preferences is an Apache 2.0 licensed dataset for text-to-image generation by the @hf.co community. This dataset contains 10K text-to-image preference pairs across image generation categories, using different model families and prompt complexities.

Blog: huggingface.co/blog/image-p...

🔗 See it in action: huggingface.co/spaces/argil...

✨ Argilla 2.5.0 is live and it comes with webhook listener support to supercharge your workflows! 🚀

#AI #MachineLearning #Webhooks #TechUpdate