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

OPT-30B #76

Open
Arthur-Ling opened this issue Apr 26, 2024 · 0 comments
Open

OPT-30B #76

Arthur-Ling opened this issue Apr 26, 2024 · 0 comments

Comments

@Arthur-Ling
Copy link

"
CUDA_VISIBLE_DEVICES=0 python main.py
--model /home/Projects/model_zoo/facebook/opt-30b
--epochs 20 --output_dir ./log/opt-30b-w6a6
--wbits 6 --abits 6 --lwc --let --alpha 0.75 --eval_ppl
--net opt-30b
"

When I use omniquant to quantizate OPT-30B to w6a6, an error happens in omniquant.py:
scale = (act.pow(args.alpha)/weight.pow(1-args.alpha)).clamp(min=1e-5)

"
RuntimeError: The size of tensor a (7168) must match the size of tensor b (5120) at non-singleton dimension 0
"

I find the shape of act is [7168], but the shape of weight is [5120].

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

1 participant