Kushal

@minstar Did you find the reason for this?

Blog link: https://kushalj001.github.io/black-box-ml/language%20modeling/text-generation/vaes/posterior%20collapse/nlp/elbo/2021/05/05/Generating-Text-with-VAEs-and-Posterior-Collapse.html Repo link: https://github.com/kushalj001/black-box-ml Thank you!

Thanks a lot, @saiamrit. Will try it out.

Hi @sathsaraRasantha. I have not been able to look at your issue due to other commitments. I'll try to take a look this weekend. It would be better if you...

Hi @sathsaraRasantha @Marwa-1995 The error is most likely related to some wrong dimension/axis in a tensor being accessed. In order to get a more readable error message, turn of your...

You first need to get the dimensions of your PDF document's page. This can be done as follows: ``` from camelot.utils import get_page_layout _, dim = get_page_layout(pdf_path) ``` Then you...

This implementation has broken down the learnable parameters. Instead of using a 6d trainable weight after concatenating the 3 tensors [h; u; h ◦u], he has used 3 different trainable...

Can the 7B model fit on a single A100 80GB GPU with inference.py given in the recipes?

Hi @HuangOwen, do you have the script you used to reproduce llama-1 results on GSM8k? If you have done something similar for llama-2, it would be great! Thanks

Hi, I am finetuning llama-2-7b with a custom RL algorithm. It basically involves generating samples from the model given a prompt and some reward shaping. I've been trying to debug...