Fix GLM4 alignment issue #2723
Open
Fix GLM4 alignment issue #2723
+50
−15
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
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Outdated suggestions cannot be applied.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 PR primarily addresses the generation issue in GLM4, specifically the missing
rope_ratioin the construction of the cosine-sine cache. Additionally, it enables the configuration to be loaded from a JSON file and allows weights to be loaded from a local path, facilitated by a new utility function namedhub_load_local_safetensors. Furthermore, it resolves a compilation issue caused by the updatedhf-hubcrate in the previous #2691, which now depends on a newer version of thetokiopackage.Tested case:
cargo run --release --example glm4 --features cuda -- --weight-path /home/weights/glm-4-9b-chat/ --prompt "Please talk about deep learning."It now generates better answers (align with the official results).