We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
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
Linux
CUDA 12.x
3.11
N/A
https://huggingface.co/FrenzyBiscuit/Skyfall-39B-v1-1.0bpw-h6-exl2
I just did a 1.0 bpw conversion, but the size ended up being the exact same as the 2.0bpw.
Is it not possible to do a 1.0bpw quant?
CUDA_VISIBLE_DEVICES=1 python3 convert.py -i ~/weights/Skyfall-39B-v1 -o ~/mytemp2 -m measurement.json -cf ~/Skyfall-39B-v1-1.0bpw-h6-exl2 -b 1.0 -hb 6
works
No response
The text was updated successfully, but these errors were encountered:
Note that its possible I messed up here, as some things got a little confusing/hectic while doing 14 quants of the same model.
Sorry, something went wrong.
No, it's not possible.
#675 (comment)
okay, thank you kindly!
No branches or pull requests
OS
Linux
GPU Library
CUDA 12.x
Python version
3.11
Pytorch version
N/A
Model
https://huggingface.co/FrenzyBiscuit/Skyfall-39B-v1-1.0bpw-h6-exl2
Describe the bug
I just did a 1.0 bpw conversion, but the size ended up being the exact same as the 2.0bpw.
Is it not possible to do a 1.0bpw quant?
Reproduction steps
CUDA_VISIBLE_DEVICES=1 python3 convert.py -i ~/weights/Skyfall-39B-v1 -o ~/mytemp2 -m measurement.json -cf ~/Skyfall-39B-v1-1.0bpw-h6-exl2 -b 1.0 -hb 6
Expected behavior
works
Logs
No response
Additional context
No response
Acknowledgements
The text was updated successfully, but these errors were encountered: