Q-bert/MetaMath-Cybertron-Starling

1---2license: cc-by-nc-4.03datasets:4- meta-math/MetaMathQA5language:6- en7pipeline_tag: text-generation8tags:9- Math10- merge11base_model:12- Q-bert/MetaMath-Cybertron13- berkeley-nest/Starling-LM-7B-alpha14---15 16## MetaMath-Cybertron-Starling17 18 19Merge [Q-bert/MetaMath-Cybertron](https://huggingface.co/Q-bert/MetaMath-Cybertron) and [berkeley-nest/Starling-LM-7B-alpha](https://huggingface.co/berkeley-nest/Starling-LM-7B-alpha) using slerp merge.20 21You can use ChatML format.22 23# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)24Detailed results can be found [Here](https://huggingface.co/datasets/open-llm-leaderboard/results/blob/main/Q-bert/MetaMath-Cybertron-Starling/results_2023-12-07T21-59-56.458563.json)25 26| Metric                | Value                     |27|-----------------------|---------------------------|28| Avg.                  | 71.35               |29| ARC (25-shot)         | 67.75               |30| HellaSwag (10-shot)   | 86.23               |31| MMLU (5-shot)         | 65.24               |32| TruthfulQA (0-shot)   | 55.94               |33| Winogrande (5-shot)   | 81.45               |34| GSM8K (5-shot)        | 71.49               |