Adding Evaluation Results
#10
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -120,4 +120,17 @@ It uses the Alpaca format (see [FastChat implementation example](https://github.
|
|
| 120 |
journal={arXiv preprint arXiv:2302.13971},
|
| 121 |
year={2023}
|
| 122 |
}
|
| 123 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
journal={arXiv preprint arXiv:2302.13971},
|
| 121 |
year={2023}
|
| 122 |
}
|
| 123 |
+
```
|
| 124 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 125 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Open-Orca__OpenOrca-Preview1-13B)
|
| 126 |
+
|
| 127 |
+
| Metric | Value |
|
| 128 |
+
|-----------------------|---------------------------|
|
| 129 |
+
| Avg. | 49.33 |
|
| 130 |
+
| ARC (25-shot) | 54.95 |
|
| 131 |
+
| HellaSwag (10-shot) | 78.19 |
|
| 132 |
+
| MMLU (5-shot) | 50.12 |
|
| 133 |
+
| TruthfulQA (0-shot) | 49.05 |
|
| 134 |
+
| Winogrande (5-shot) | 71.03 |
|
| 135 |
+
| GSM8K (5-shot) | 4.93 |
|
| 136 |
+
| DROP (3-shot) | 37.04 |
|