Which is good for SFT? Using pre-trained peft_model or init new peft model? #417
Unanswered
yusufcakmakk
asked this question in
Q&A
Replies: 1 comment
-
I think I found the solution in 3rd row here. Is it right? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi everyone 👋🏻
Let's assume we have a model trained with this method. For the SFT training, there is a parameter called "peft_model" as mentioned here and here. Should I leave it empty to train the instruction model or give the PEFT model pre-trained in the first step? What do you think I should do?
Thanks for your guidance.
Beta Was this translation helpful? Give feedback.
All reactions