During Training, Train Loss Decreases Significantly, but Eval Loss Decreases Only Slightly #1745
Unanswered
SevenTailCat
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
During the training process, the train loss decreases from 0.13 to 0.08, while the eval loss decreases from 0.22 to around 0.20. In other words, the train loss has a much larger decrease compared to the eval loss.
Is this normal? If I want to further optimize this model, what should I do?
Beta Was this translation helpful? Give feedback.
All reactions