convert : fix context length for nomic-embed-text-v2-moe #13216
Merged
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
ggerganov
approved these changes
Apr 30, 2025
gabe-l-hart
added a commit
to gabe-l-hart/llama.cpp
that referenced
this pull request
May 2, 2025
* GraniteMoEShared: fix: Fix the input to the shared experts fix: Cleaner (maybe more correct?) splitting for gate/up feat: First WIP cut at model arch in cpp fix: Split MoE fused tensors for shared experts in conversion feat: hparam and arch plumbing for granitemoeshared feat: Add GGUF conversion for granitemoeshared llama-model : support Qwen2 embedding models and pooling_mode_lasttoken (ggml-org#13245) convert : use correct context length for nomic-embed-text-v2 (ggml-org#13216)
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Outdated suggestions cannot be applied.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.
As noted by ggerganov, nomic-embed-text-v2-moe is correctly documented to be trained with up to 512 tokens of context, so the hardcoded value of 2048 used in the convert script is not accurate.
With this change, nomic-embed-text-v1 and v1.5 still convert with context_length=2048, and nomic-embed-text-v2-moe now converts with context_length=512.