Running
on
A100
886
🪁
Zephyr 7B
Models, datasets, and demos associated with Zephyr 7B. For code to train the models, see: https://github.com/huggingface/alignment-handbook
Note Chat with our Zephyr 7B models!
-
Zephyr: Direct Distillation of LM Alignment
Paper • 2310.16944 • Published • 121
HuggingFaceH4/zephyr-7b-beta
Text Generation • Updated • 782k • • 1.6kNote A state-of-the-art chat model at the 7B parameter scale. Trained on synthetic data with a mix of SFT and DPO.
HuggingFaceH4/zephyr-7b-alpha
Text Generation • Updated • 34.4k • • 1.1kNote The precursor to Zephyr-7B-β. Trained on synthetic data with a mix of SFT and DPO.
HuggingFaceH4/mistral-7b-sft-beta
Text Generation • Updated • 13k • 24Note The SFT model used for the DPO training of Zephyr-7B-β
HuggingFaceH4/mistral-7b-sft-alpha
Text Generation • Updated • 603 • 3Note The SFT model used for the DPO training of Zephyr-7B-α
HuggingFaceH4/ultrachat_200k
Viewer • Updated • 515k • 13.1k • 473Note The SFT dataset used to train Zephyr-7B-β
HuggingFaceH4/ultrafeedback_binarized
Viewer • Updated • 187k • 5.87k • 238Note The dataset of AI preferences used to train Zephyr-7B-β with DPO