Question about HuggingFaceH4/zephyr-7b-alpha and its base model

#43
by dqdw - opened

Dear [Developer/Team],

Thanks for releasing HuggingFaceH4/zephyr-7b-alpha. I've been experimenting with it and it works well.

Since I plan to extend this model, I want to confirm how it relates to mistralai/Mistral-7B-v0.1 according to Hugging Face:

Direct Fine-tuning: Was HuggingFaceH4/zephyr-7b-alpha directly fine-tuned from mistralai/Mistral-7B-v0.1, or are there intermediate steps?

Inheritance: Was there any merging or distillation involved?

This would help me ensure I use the model correctly.

I really appreciate your help.

All the best

Sign up or log in to comment