Model weight issue

[Albert2X]

Hi,
Thank you for your valuable work on this project. I am currently working on reproducing your code, and I have encountered a couple of challenges that I was hoping you could help clarify:

1. It seems that the repository does not include the learnable weights for the <task> tokens, which is resulting in random images being generated. Could you kindly provide the necessary weights or let me know if there's an alternative approach to obtain them?

2. Additionally, I noticed that the generation script appears to be incomplete. Specifically, the path to the LLM weights and other potential configuration details seem to be missing. Would it be possible to share the full script or provide guidance on the required setup for successful generation?

I greatly appreciate your time and assistance in addressing these concerns. Looking forward to your response!