Adding Evaluation Results
#7
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -140,4 +140,17 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 140 |
感谢：
|
| 141 |
- LLaMA2
|
| 142 |
- Firefly项目
|
| 143 |
-
- shareGPT中文数据集的建设者们
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 140 |
感谢：
|
| 141 |
- LLaMA2
|
| 142 |
- Firefly项目
|
| 143 |
+
- shareGPT中文数据集的建设者们
|
| 144 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 145 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_shareAI__llama2-13b-Chinese-chat)
|
| 146 |
+
|
| 147 |
+
| Metric | Value |
|
| 148 |
+
|-----------------------|---------------------------|
|
| 149 |
+
| Avg. | 48.23 |
|
| 150 |
+
| ARC (25-shot) | 60.58 |
|
| 151 |
+
| HellaSwag (10-shot) | 82.19 |
|
| 152 |
+
| MMLU (5-shot) | 55.45 |
|
| 153 |
+
| TruthfulQA (0-shot) | 45.11 |
|
| 154 |
+
| Winogrande (5-shot) | 76.64 |
|
| 155 |
+
| GSM8K (5-shot) | 11.37 |
|
| 156 |
+
| DROP (3-shot) | 6.24 |
|