Adding Evaluation Results
#44
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -296,4 +296,17 @@ The model is licensed under the BigCode OpenRAIL-M v1 license agreement. You can
|
|
| 296 |
journal={arXiv preprint arXiv:2301.03988},
|
| 297 |
year={2023}
|
| 298 |
}
|
| 299 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 296 |
journal={arXiv preprint arXiv:2301.03988},
|
| 297 |
year={2023}
|
| 298 |
}
|
| 299 |
+
```
|
| 300 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 301 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_bigcode__santacoder)
|
| 302 |
+
|
| 303 |
+
| Metric | Value |
|
| 304 |
+
|-----------------------|---------------------------|
|
| 305 |
+
| Avg. | 25.33 |
|
| 306 |
+
| ARC (25-shot) | 26.28 |
|
| 307 |
+
| HellaSwag (10-shot) | 25.6 |
|
| 308 |
+
| MMLU (5-shot) | 25.89 |
|
| 309 |
+
| TruthfulQA (0-shot) | 51.24 |
|
| 310 |
+
| Winogrande (5-shot) | 48.07 |
|
| 311 |
+
| GSM8K (5-shot) | 0.0 |
|
| 312 |
+
| DROP (3-shot) | 0.21 |
|