Commit
•
2ba1246
1
Parent(s):
7147483
Update README.md
Browse files
README.md
CHANGED
|
@@ -13,6 +13,21 @@ This model is further pre-trained model based on [01-ai/Yi-6B](https://huggingfa
|
|
| 13 |
|
| 14 |
This is a preview version, for experimental use only, we will use it to fine-tune on downstream tasks and evaluate the performance.
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
## Usage
|
| 17 |
|
| 18 |
```python
|
|
|
|
| 13 |
|
| 14 |
This is a preview version, for experimental use only, we will use it to fine-tune on downstream tasks and evaluate the performance.
|
| 15 |
|
| 16 |
+
|
| 17 |
+
### [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 18 |
+
|
| 19 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_hon9kon9ize__CantoneseLLM-6B-preview202402)
|
| 20 |
+
| Metric | Value |
|
| 21 |
+
|-----------------------|---------------------------|
|
| 22 |
+
| Avg. | 56.93 |
|
| 23 |
+
| ARC (25-shot) | 55.63 |
|
| 24 |
+
| HellaSwag (10-shot) | 75.8 |
|
| 25 |
+
| MMLU (5-shot) | 63.07 |
|
| 26 |
+
| TruthfulQA (0-shot) | 42.26 |
|
| 27 |
+
| Winogrande (5-shot) | 74.11 |
|
| 28 |
+
| GSM8K (5-shot) | 30.71 |
|
| 29 |
+
|
| 30 |
+
|
| 31 |
## Usage
|
| 32 |
|
| 33 |
```python
|