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.

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

Add linear wrapper and kan feedback #189

Merged

Conversation

klei22
Copy link
Collaborator

@klei22 klei22 commented Jun 18, 2024

No description provided.

SenmiaoORZ and others added 8 commits June 4, 2024 21:52
This change prevents division by zero.

This appears to be the reason for the initial instability of inference,
so now we can explore various KAN applications.

Note:
We should convert this clamp into an argparse param and config for
experimentation.
This allows us to set the mean and standard deviation for random weights
for both linear modules and for the embedding table.
After experimentation, we don't need to include these after including
the dictionary.
Utilize inheritance based interface adaptation.
@klei22 klei22 requested a review from gkielian June 18, 2024 21:28
@klei22 klei22 mentioned this pull request Jun 18, 2024
@klei22
Copy link
Collaborator Author

klei22 commented Jun 18, 2024

Previous discussions here:

After these changes wpe is not set if we're using a different strategy
for position embedding.
@gkielian gkielian merged commit 0e9ac28 into ReaLLMASIC:master Jun 18, 2024
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants