iq3_xxs: guards for the no-imatrix situation #5334
Merged
Conversation
This file contains 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
ggerganov
approved these changes
Feb 5, 2024
ggerganov
changed the title
jordankanter
pushed a commit
to jordankanter/llama.cpp
that referenced
this pull request
Mar 13, 2024
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
hodlen
pushed a commit
to hodlen/llama.cpp
that referenced
this pull request
Apr 1, 2024
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
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Outdated suggestions cannot be applied.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.
IQ3_XXScan give a very bad quantization when used without an importance matrix (imatrix), see #5332.Instead of adding a warning or even disallowing
IQ3_XXSquantization without an imatrix, this PR prevents a bad outcome by usingQ3_Kfor theattn_vtensors, and a mix ofQ4_KandQ3_Kfor theffn_downtensors when no imatrix has been supplied. This results in a somewhat larger quantized model (e.g., 2.61 GiB vs 2.5 GiB for 7B LLaMAs) but a more reasonable PPL (e.g.,5.4923for LLaMA-v2-7B and a context of 4096 vs100+).