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.

Sign up for GitHub

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

Jump to bottom

Fix CPU dequantization to use nested dequantized scaling constant #1549

Open

zyklotomic wants to merge 1 commit into bitsandbytes-foundation:main from zyklotomic:double-dequant-cpu-fix

zyklotomic commented Mar 1, 2025

I believe that in the CPU case, we have forgotten to use the updated absmax if that absmax too was quantized, like in QLoRA's doubleDequant().


          Fix CPU dequantization to use nested dequantized scaling constant

753e411

Author

zyklotomic commented Mar 1, 2025

Similarly, I also noticed slightly above, in:

bitsandbytes/bitsandbytes/functional.py

Line 1020 in 753e411

code = quant_state.code.to(A.device)

the updated code variable is not used. Not sure if this is an error or extraneous.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet