Then Unsloth and Axolotl are your best options imo. There's no difference in terms of PEFT/FFT. I think Spectrum is only supported by Axolotl at the moment if you're interested.
@maximelabonne
-
Unsloth vs Axolotl: Choosing the Right Fine-Tuning Framework
By
–
In general, I'd recommend Unsloth for single-GPU settings and Axolotl for multi-GPU settings. For research, you might want to have more fine-grained control though. TRL works there, but reimplementing it in PyTorch is also a good option for full control.
-
Top 10 Hugging Face Creator Celebrates Milestone with Community
By
–
I almost missed it but I broke into the top 10 most followed accounts on @huggingface It's been a wild journey and I'm really happy where I am right now 🙂 Thanks everyone, you made it possible! And no worries: more articles, models, and datasets to come.
-
BigLlama-3.1-1T Upload Triggers Community Flashbacks
By
–
This gave me flashbacks of BigLlama-3.1-1T's upload
-
Distillation Emerges as Superior Model Improvement Technique Over DPO
By
–
Because model merging and abliteration are destructive, DPO has been essential to healing networks. These days are now over: distillation is the new powerhouse that not only heals models but also improves them.
-
Tokenizer and Chat Template Issues Affect Merged Model Performance
By
–
Issues related to the tokenizer and chat template are the main reasons why merged models underperformed on the Open LLM Leaderboard v2. The evals don't change anything. A model merged to maximize MMLU will also perform well on these new evals.
-
NeuralDaredevil LLM Outperforms Llama 3.1 Models
By
–
Haha NeuralDaredevil is a beast, outperforming most Llama 3.1 models. All of that despite being based on Llama 3.0 and the performance drop due to the abliteration process. Thanks to Dampfinchen for fixing an issue with the tokenizer that messed up the previous IFEval score.
-
Fine-tuning and Model Merging Techniques Explained
By
–
Excellent breakdown of my fine-tuning and model merging talk by @TheTuringPost 🙂
-
Fine-tuning and Merging LLMs: Techniques and Best Practices
By
–
If you're interested in fine-tuning and merging LLMs, here's a well-edited video of a talk I gave in July. We talk about: – How to create a dataset
– SFT techniques
– How to merge models Thanks to @Tunehq_ai for the invitation! https://
youtu.be/8aFqLVQjHTw?si
=K8KpM55G_93wwZmM
… -
Fine-Tuning Your Own Llama 3 Model Series
By
–
I'm giving a new & improved series of talks about fine-tuning – @DataCamp
: live coding session, today at 11 AM ET (
https://
datacamp.com/webinars/fine-
tuning-your-own-llama-3-model
…) – @_odsc
: Open Data Science Conference, September 5-6 (
https://
odsc.com/europe/?utm_ca
mpaign=ODSCEurope&utm_content=304738063&utm_medium=social&utm_source=twitter&hss_channel=tw-3018841323
…) – @dctanner
: AI Engineer London Meetup, September 12