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

The convert.py may not fit the rwkv-4-world-169m. #8

Open
kip234 opened this issue Mar 1, 2024 · 1 comment
Open

The convert.py may not fit the rwkv-4-world-169m. #8

kip234 opened this issue Mar 1, 2024 · 1 comment

Comments

@kip234
Copy link

kip234 commented Mar 1, 2024

I have tried to convert the weight file which i download from https://huggingface.co/RWKV/rwkv-4-world-169m with convert.py. I found the following problems.

  • All of the keys of weight file begin with 'rwkv.' compared to convert.py.
  • The .att. in convert.py should be replaced with .attention..
  • The value of rwkv.blocks.0.attention.time_decay is a 1d tensor. The headsnume and headsize in convert.py can't unpacke from w[f"rwkv.blocks.0.attention.time_decay"].shape

I don't know how to fix the third problem.

@harrisonvanderbyl
Copy link
Collaborator

Hi @kip234 I am glad to tell you that the reason you cannot convert the v4 world model is due to the fact it has been deprecated in favor of the v5 world model (Also known as the Eagle models)!
You can find the checkpoints here: https://huggingface.co/BlinkDL/rwkv-5-world/tree/main

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