GRM-3.2-Sky / .eval_results /mmlu_pro.yaml
DedeProGames's picture
Create .eval_results/mmlu_pro.yaml
e5d7701 verified
Raw
History Blame Contribute Delete
196 Bytes
- dataset:
id: TIGER-Lab/MMLU-Pro
task_id: mmlu_pro
value: 89.5
source:
url: https://huggingface.co/OrionLLM/GRM-3.2-Sky
name: Official GRM-3.2 Benchmark
user: DedeProGames