Commit
•
947d447
1
Parent(s):
f9c26d2
Adding Evaluation Results (#2)
Browse files- Adding Evaluation Results (de996f2b20b840596d6c51e5351746fd530b0a40)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -36,4 +36,17 @@ tags:
|
|
| 36 |
## Limitations
|
| 37 |
|
| 38 |
- May hallucinate or generate incorrect information
|
| 39 |
-
- Large model size leads to high compute requirements
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
## Limitations
|
| 37 |
|
| 38 |
- May hallucinate or generate incorrect information
|
| 39 |
+
- Large model size leads to high compute requirements
|
| 40 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 41 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_NewstaR__Starlight-13B)
|
| 42 |
+
|
| 43 |
+
| Metric | Value |
|
| 44 |
+
|-----------------------|---------------------------|
|
| 45 |
+
| Avg. | 46.87 |
|
| 46 |
+
| ARC (25-shot) | 59.3 |
|
| 47 |
+
| HellaSwag (10-shot) | 82.15 |
|
| 48 |
+
| MMLU (5-shot) | 55.67 |
|
| 49 |
+
| TruthfulQA (0-shot) | 37.39 |
|
| 50 |
+
| Winogrande (5-shot) | 76.64 |
|
| 51 |
+
| GSM8K (5-shot) | 10.84 |
|
| 52 |
+
| DROP (3-shot) | 6.08 |
|