Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -52,4 +52,17 @@ outputs = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
|
|
| 52 |
print(outputs)
|
| 53 |
```
|
| 54 |
|
| 55 |
-
Please find more details in our [GitHub repository](https://github.com/fe1ixxu/ALMA)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
print(outputs)
|
| 53 |
```
|
| 54 |
|
| 55 |
+
Please find more details in our [GitHub repository](https://github.com/fe1ixxu/ALMA)
|
| 56 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 57 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_haoranxu__ALMA-13B-Pretrain)
|
| 58 |
+
|
| 59 |
+
| Metric | Value |
|
| 60 |
+
|-----------------------|---------------------------|
|
| 61 |
+
| Avg. | 45.1 |
|
| 62 |
+
| ARC (25-shot) | 56.91 |
|
| 63 |
+
| HellaSwag (10-shot) | 80.15 |
|
| 64 |
+
| MMLU (5-shot) | 50.31 |
|
| 65 |
+
| TruthfulQA (0-shot) | 37.44 |
|
| 66 |
+
| Winogrande (5-shot) | 76.4 |
|
| 67 |
+
| GSM8K (5-shot) | 8.87 |
|
| 68 |
+
| DROP (3-shot) | 5.58 |
|