CodeLlama-13b-hf / README.md
leaderboard-pr-bot's picture
Adding Evaluation Results
793d2b2
|
raw
history blame
662 Bytes
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_NousResearch__CodeLlama-13b-hf)
| Metric | Value |
|-----------------------|---------------------------|
| Avg. | 37.91 |
| ARC (25-shot) | 40.87 |
| HellaSwag (10-shot) | 63.35 |
| MMLU (5-shot) | 32.81 |
| TruthfulQA (0-shot) | 43.79 |
| Winogrande (5-shot) | 67.17 |
| GSM8K (5-shot) | 12.13 |
| DROP (3-shot) | 5.25 |