Add accuracy server configs for Mistral-Small-3.1-24B with mistral tokenizer#94
Draft
dbarbuzzi wants to merge 1 commit into
Draft
Add accuracy server configs for Mistral-Small-3.1-24B with mistral tokenizer#94dbarbuzzi wants to merge 1 commit into
dbarbuzzi wants to merge 1 commit into
Conversation
…kenizer The lm-eval accuracy test for Mistral-Small-3.1-24B-Instruct-2503 on HPU scored near-zero due to incorrect tokenization. The HuggingFace tokenizer has a known regex bug for this model, and without an accuracy/server.yml the test falls back to common/accuracy/server.yml which lacks tokenizer settings. Add per-model configs with tokenizer_mode/config_format/load_format set to mistral, matching the existing performance/server.yml and the pattern used by Voxtral models. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Domenic Barbuzzi <dbarbuzz@redhat.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
accuracy/server.ymlformistralai/Mistral-Small-3.1-24B-Instruct-2503andRedHatAI/Mistral-Small-3.1-24B-Instruct-2503-FP8-dynamictokenizer_mode: mistral,config_format: mistral,load_format: mistralmatching the existing performance configs and Voxtral precedentTest plan
mistralai/Mistral-Small-3.1-24B-Instruct-2503— confirm GSM8K score is within 10% of ground truth (0.89)RedHatAI/Mistral-Small-3.1-24B-Instruct-2503-FP8-dynamic— confirm GSM8K score is within 10% of ground truth🤖 Generated with Claude Code