Code and framework
#1
by nouranali - opened
Is the code to distill this model shared anywhere? I'm trying to distill a Qwen3 model but results are bad so I want to check another code
Yes i did the code but it is an kaggle notebook will share it i used qwen3-0.6B model as teacher model
https://www.kaggle.com/code/hdheklsd/notebookf4bae562f8 here yes but loss it littel hight around 32-30 , so yaa
Thank a lot
nouranali changed discussion status to closed