Adding Evaluation Results
#31
by leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -160,4 +160,17 @@ The 175B model was trained on 992 *80GB A100 GPUs*. The training duration was ro
|
|
| 160 |
archivePrefix={arXiv},
|
| 161 |
primaryClass={cs.CL}
|
| 162 |
}
|
| 163 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 160 |
archivePrefix={arXiv},
|
| 161 |
primaryClass={cs.CL}
|
| 162 |
}
|
| 163 |
+
```
|
| 164 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 165 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_facebook__opt-1.3b)
|
| 166 |
+
|
| 167 |
+
| Metric | Value |
|
| 168 |
+
|-----------------------|---------------------------|
|
| 169 |
+
| Avg. | 30.38 |
|
| 170 |
+
| ARC (25-shot) | 29.52 |
|
| 171 |
+
| HellaSwag (10-shot) | 54.53 |
|
| 172 |
+
| MMLU (5-shot) | 24.96 |
|
| 173 |
+
| TruthfulQA (0-shot) | 38.71 |
|
| 174 |
+
| Winogrande (5-shot) | 59.75 |
|
| 175 |
+
| GSM8K (5-shot) | 0.15 |
|
| 176 |
+
| DROP (3-shot) | 5.02 |
|