How to run inference after SFT tuning ? #2459
-
Hi, this is how I load the model: I am using Qwen1.5 for the finetune. thanks |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 3 replies
-
use CLI demo to infer trained models |
Beta Was this translation helpful? Give feedback.
-
I tried increasing the n_epochs, from 3 to eventually 800, finally getting some meaningful response. However, I worry that this will have model overfitting. The loss is pretty low now : appreciate your work and support very much, @hiyouga ! |
Beta Was this translation helpful? Give feedback.
use CLI demo to infer trained models