[BugFix] Correct max_model_len derivation from config.json for Mistral format #17937
+48
−3
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Outdated suggestions cannot be applied.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.
FIX #17747
@tjohnson31415
Thank you for your patience and suggestions.
I've reverted the code to its original version first before applying modifications.
I have removed all logic related to max_seq_len.
The subsequent logic (to fetch from standard Hugging Face config) will only be triggered if max_position_embeddings cannot be obtained from params.json.
I am referencing the configuration loaded from config.json as it's done for ConfigFormat.HF (i.e., using AutoConfig.from_pretrained).
Something went wrong, and I accidentally caused issues in the original branch, so I've opened a new PR to merge the code.