convert_hf_to_gguf.py : conversion from hf weights to Q6_0 #483
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This quantization script is obtained by making a sort of "cross multiplication" with the python code for q5_0, and the C code for q5_0 and q6_0 in order to get through trial and error the code for the q6_0 conversion script, this with the help of a 7xB parameters AI model.
It was an interesting experiment!
Tested on Llama 3.2 instruct 1B and Qwen 2.5 instruct 1.5B.
Bitrate of this q6_0 conversion is 6.50BPW straight.
PPL equivalent (+/-0.5%) to a regular q6_0 quant from a fp16 gguf.
Inference is working as intended in my Croco.cpp.