Adding Evaluation Results
#34
by leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -161,3 +161,17 @@ The 175B model was trained on 992 *80GB A100 GPUs*. The training duration was ro
|
|
| 161 |
}
|
| 162 |
```
|
| 163 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
}
|
| 162 |
```
|
| 163 |
|
| 164 |
+
|
| 165 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 166 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_facebook__opt-350m)
|
| 167 |
+
|
| 168 |
+
| Metric | Value |
|
| 169 |
+
|-----------------------|---------------------------|
|
| 170 |
+
| Avg. | 26.32 |
|
| 171 |
+
| ARC (25-shot) | 23.55 |
|
| 172 |
+
| HellaSwag (10-shot) | 36.73 |
|
| 173 |
+
| MMLU (5-shot) | 26.02 |
|
| 174 |
+
| TruthfulQA (0-shot) | 40.83 |
|
| 175 |
+
| Winogrande (5-shot) | 52.64 |
|
| 176 |
+
| GSM8K (5-shot) | 0.3 |
|
| 177 |
+
| DROP (3-shot) | 4.16 |
|