Skip to content

Conversation

@Hzfengsy
Copy link
Member

Usage:

python benchmark/eval/mmlu.py --model dist/Meta-Llama-3-8B-Instruct-q4f16_1-MLC --dataset /path/to/dataset --device cuda --log-dir debug/mmlu

Note that chat mode is problematic, which needs to be fixed.

@Hzfengsy Hzfengsy marked this pull request as ready for review June 16, 2024 06:58
@Hzfengsy Hzfengsy force-pushed the mmlu_eval branch 4 times, most recently from 3d8232d to d015821 Compare June 18, 2024 03:14
Usage:
```
python python/bench/eval/mmlu.py --model dist/Meta-Llama-3-8B-Instruct-q4f16_1-MLC --dataset /path/to/dataset --device cuda --log-dir debug/mmlu
```

Note that chat mode is problematic, which needs to be fixed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant