Question: How to fine tune an already finetuned model like NuExtract as a fine tune of Phi-3.5

unslothai / unsloth

Finetune Llama 3.2, Mistral, Phi, Qwen 2.5 & Gemma LLMs 2-5x faster with 80% less memory

https://unsloth.ai

Apache License 2.0

18.41k stars 1.29k forks source link

Question: How to fine tune an already finetuned model like NuExtract as a fine tune of Phi-3.5 #1202

Open KIC opened 3 weeks ago

KIC commented 3 weeks ago

I have seen that we can finetune the Phi-3.5-mini. Now I see that the NuExtract is also based on the Phi-3.5-mini. And I really would like to further fine tune the nuextract model could you please guide me in the right direction how I can do this?

danielhanchen commented 3 weeks ago

Apologies on the delay - this will work when all model support is added in - for now I had to manually mistral-fy Phi

KIC commented 1 week ago

Thank you! So should I just wait then? Or can I reproduce this manual procedure of mistral-fy Phi? Is there a timeline for this eventually?