Adding Evaluation Results
#19
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -97,4 +97,17 @@ The Yi series models are fully open for academic research and free commercial
|
|
| 97 |
usage with permission via applications. All usage must adhere to the [Model
|
| 98 |
License Agreement 2.0](https://huggingface.co/01-ai/Yi-6B/blob/main/LICENSE). To
|
| 99 |
apply for the official commercial license, please contact us
|
| 100 |
-
([yi@01.ai](mailto:yi@01.ai)).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
usage with permission via applications. All usage must adhere to the [Model
|
| 98 |
License Agreement 2.0](https://huggingface.co/01-ai/Yi-6B/blob/main/LICENSE). To
|
| 99 |
apply for the official commercial license, please contact us
|
| 100 |
+
([yi@01.ai](mailto:yi@01.ai)).
|
| 101 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 102 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_01-ai__Yi-6B)
|
| 103 |
+
|
| 104 |
+
| Metric | Value |
|
| 105 |
+
|-----------------------|---------------------------|
|
| 106 |
+
| Avg. | 53.06 |
|
| 107 |
+
| ARC (25-shot) | 55.55 |
|
| 108 |
+
| HellaSwag (10-shot) | 76.42 |
|
| 109 |
+
| MMLU (5-shot) | 63.85 |
|
| 110 |
+
| TruthfulQA (0-shot) | 41.86 |
|
| 111 |
+
| Winogrande (5-shot) | 73.8 |
|
| 112 |
+
| GSM8K (5-shot) | 12.66 |
|
| 113 |
+
| DROP (3-shot) | 47.32 |
|