Adding Evaluation Results
#12
by leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -150,3 +150,17 @@ If you found OpenLLaMA useful in your research or applications, please cite usin
|
|
| 150 |
|
| 151 |
|
| 152 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 150 |
|
| 151 |
|
| 152 |
|
| 153 |
+
|
| 154 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 155 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_openlm-research__open_llama_3b)
|
| 156 |
+
|
| 157 |
+
| Metric | Value |
|
| 158 |
+
|-----------------------|---------------------------|
|
| 159 |
+
| Avg. | 33.52 |
|
| 160 |
+
| ARC (25-shot) | 39.85 |
|
| 161 |
+
| HellaSwag (10-shot) | 62.65 |
|
| 162 |
+
| MMLU (5-shot) | 26.94 |
|
| 163 |
+
| TruthfulQA (0-shot) | 34.97 |
|
| 164 |
+
| Winogrande (5-shot) | 64.72 |
|
| 165 |
+
| GSM8K (5-shot) | 0.45 |
|
| 166 |
+
| DROP (3-shot) | 5.06 |
|