Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
lewtunΒ 
posted an update Apr 11
Post
5021
Introducing Zephyr 141B-A35B πŸͺ:

HuggingFaceH4/zephyr-orpo-141b-A35b-v0.1

Yesterday, Mistral released their latest base model (via magnet link of course πŸ˜…) and the community quickly converted it to transformers format and pushed it to the Hub: mistral-community/Mixtral-8x22B-v0.1

Early evals of this model looked extremely strong, so we teamed up with Argilla and KAIST AI to cook up a Zephyr recipe with a few new alignment techniques that came out recently:

πŸ§‘β€πŸ³ Align the base model with Odds Ratio Preference Optimisation (ORPO). This novel algorithm developed by @JW17 and @nlee-208 and @j6mes and does not require an SFT step to achieve high performance and is thus much more computationally efficient than methods like DPO and PPO.

🦫 Use a brand new dataset of 7k high-quality, multi-turn preferences that has been developed by our friends at Argilla. To create this dataset, they took the excellent Capybara SFT dataset from @LDJnr LDJnr/Capybara and converted it into a preference dataset by augmenting the final turn with responses from new LLMs that were then ranked by GPT-4.

What we find especially neat about this approach is that training on 7k samples only takes ~1.3h on 4 H100 nodes, yet produces a model that is very strong on chat benchmarks like IFEval and BBH.

Kudos to @alvarobartt @JW17 and @nlee-208 for this very nice and fast-paced collab!

For more details on the paper and dataset, checkout our collection: HuggingFaceH4/zephyr-orpo-6617eba2c5c0e2cc3c151524
In this post