Adding Evaluation Results
#27
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -209,3 +209,17 @@ The 175B model was trained on 992 *80GB A100 GPUs*. The training duration was ro
|
|
| 209 |
primaryClass={cs.CL}
|
| 210 |
}
|
| 211 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 209 |
primaryClass={cs.CL}
|
| 210 |
}
|
| 211 |
```
|
| 212 |
+
|
| 213 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 214 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_facebook__opt-13b)
|
| 215 |
+
|
| 216 |
+
| Metric | Value |
|
| 217 |
+
|-----------------------|---------------------------|
|
| 218 |
+
| Avg. | 34.99 |
|
| 219 |
+
| ARC (25-shot) | 39.93 |
|
| 220 |
+
| HellaSwag (10-shot) | 71.2 |
|
| 221 |
+
| MMLU (5-shot) | 24.9 |
|
| 222 |
+
| TruthfulQA (0-shot) | 34.1 |
|
| 223 |
+
| Winogrande (5-shot) | 68.51 |
|
| 224 |
+
| GSM8K (5-shot) | 0.99 |
|
| 225 |
+
| DROP (3-shot) | 5.28 |
|