do not quantize embedding weight #840
Open
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
wbruna
reviewed
Sep 17, 2025
| if (!force_clip_f32) { | ||
| auto tensor_type = tensor_types.find(prefix + "token_embedding.weight"); | ||
| if (tensor_type != tensor_types.end()) | ||
| if (tensor_type != tensor_types.end() && tensor_type->second == GGML_TYPE_F16) { |
There was a problem hiding this comment.
Maybe lower quants could be kept at F16 too (or even Q8)? The main reason for quantization is reducing memory usage, so it'd be surprising if a Q8_0 or Q4_0 quant ended up using more VRAM than F16.
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Outdated suggestions cannot be applied.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.
Try to fix #837.