Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -125,4 +125,17 @@ Beluga is a new technology that carries risks with use. Testing conducted to dat
|
|
| 125 |
journal = {GitHub repository},
|
| 126 |
howpublished = {\url{https://github.com/tatsu-lab/stanford_alpaca}},
|
| 127 |
}
|
| 128 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
journal = {GitHub repository},
|
| 126 |
howpublished = {\url{https://github.com/tatsu-lab/stanford_alpaca}},
|
| 127 |
}
|
| 128 |
+
```
|
| 129 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 130 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_stabilityai__StableBeluga1-Delta)
|
| 131 |
+
|
| 132 |
+
| Metric | Value |
|
| 133 |
+
|-----------------------|---------------------------|
|
| 134 |
+
| Avg. | 46.36 |
|
| 135 |
+
| ARC (25-shot) | 68.17 |
|
| 136 |
+
| HellaSwag (10-shot) | 85.88 |
|
| 137 |
+
| MMLU (5-shot) | 64.83 |
|
| 138 |
+
| TruthfulQA (0-shot) | 55.81 |
|
| 139 |
+
| Winogrande (5-shot) | 49.8 |
|
| 140 |
+
| GSM8K (5-shot) | 0.0 |
|
| 141 |
+
| DROP (3-shot) | 0.0 |
|