Commit
•
d5feda7
1
Parent(s):
b8fbe09
Adding Evaluation Results (#3)
Browse files- Adding Evaluation Results (2c710085e9b05b20df23acc1b53b2cc45c79cdf2)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -139,3 +139,17 @@ The finetuning scripts will be available in our [RAIL Github Repository](https:/
|
|
| 139 |
## Evaluation
|
| 140 |
|
| 141 |
**TODO**
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 139 |
## Evaluation
|
| 140 |
|
| 141 |
**TODO**
|
| 142 |
+
|
| 143 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 144 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Vmware__open-llama-7b-v2-open-instruct)
|
| 145 |
+
|
| 146 |
+
| Metric | Value |
|
| 147 |
+
|-----------------------|---------------------------|
|
| 148 |
+
| Avg. | 40.34 |
|
| 149 |
+
| ARC (25-shot) | 39.76 |
|
| 150 |
+
| HellaSwag (10-shot) | 70.31 |
|
| 151 |
+
| MMLU (5-shot) | 35.16 |
|
| 152 |
+
| TruthfulQA (0-shot) | 39.53 |
|
| 153 |
+
| Winogrande (5-shot) | 64.33 |
|
| 154 |
+
| GSM8K (5-shot) | 7.43 |
|
| 155 |
+
| DROP (3-shot) | 25.88 |
|