Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Investigate low accuracy in albert/albert-large-v2 #761

Open
ayerofieiev-tt opened this issue Feb 12, 2025 · 2 comments
Open

Investigate low accuracy in albert/albert-large-v2 #761

ayerofieiev-tt opened this issue Feb 12, 2025 · 2 comments
Assignees
@amalbasaTT
Labels
accuracy

Comments

@ayerofieiev-tt
Copy link
Member

No description provided.

@ayerofieiev-tt
Copy link
Member Author 

	1. low pcc at ttnn_tanh (0.994849691145142 in model and 0.9960485128140038 in unit test provided in created ticket) (created ticket: https://github.com/tenstorrent/tt-metal/issues/17840)
	2. Accuracy unusually small (0.236), when compared to large and base (investigating further!)

@amalbasaTT
Copy link

amalbasaTT commented Feb 19, 2025 •
edited
Loading

@ayerofieiev-tt #760 (comment). Same applies here. Accuracy went to 0.9650104636691526 after replacing ttnn_tanh with aten_tanh

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@amalbasaTT amalbasaTT

Labels
accuracy
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@ayerofieiev-tt @amalbasaTT