Technical question: Lineage of Qwen/Qwen2-7B-Instruct

#28
by dqdw - opened

Dear [Developer/Team],

I really appreciate your work on Qwen/Qwen2-7B-Instruct. I've been using it recently and it's been very helpful.

I'm planning to use it as a base, so I need to confirm its relation with Qwen/Qwen2-7B:

Direct Fine-tuning: Was Qwen/Qwen2-7B the direct starting point for training Qwen/Qwen2-7B-Instruct?

Inheritance: Does it strictly inherit the architecture and weights of Qwen/Qwen2-7B without merging or distilling from other models?

Understanding this will help me ensure compatibility with the Qwen/Qwen2-7B ecosystem.

Appreciate your support!

Sign up or log in to comment