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

How to adjust the hyperparameter r in the Lora modules? #4

Open
pzp-dzd opened this issue Sep 2, 2023 · 1 comment
Open

How to adjust the hyperparameter r in the Lora modules? #4

pzp-dzd opened this issue Sep 2, 2023 · 1 comment

Comments

@pzp-dzd
Copy link

pzp-dzd commented Sep 2, 2023

Thank you very much for contributing the source code of the paper.
When using Lora for fine-tuning, what is the exact value of the hyperparameter r in Lora modules, and how should it be adjusted when running the code?

@vishaal27
Copy link

It seems like they hard-code r=4 in their code, see here (self.lora_attn_dim is the rank I think):

'''
LoRA setting
'''
self.lora_moe_lambda = 1.0
self.lora_moe_act = 'linear'
self.lora_r_dropout = None
self.lora_attn_dim = 4
self.lora_moe = 0
self.lora_attn_alpha=128

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

No branches or pull requests

2 participants