---
license: apache-2.0
language:
- en
---

# zephyr_0.2

The DPO-trained model from `alignment-handbook/zephyr-7b-sft-full` using 20% data of `HuggingFaceH4/ultrafeedback_binarized`, as in the "[Weak-to-Strong Extrapolation Expedites Alignment](https://arxiv.org/abs/2404.16792)" paper.