Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -132,3 +132,17 @@ NOTE: The **WizardLM-30B-V1.0** & **WizardLM-13B-V1.0** use different prompt wit
|
|
| 132 |
2. For **WizardLM-7B-V1.0** , the Prompt should be as following:
|
| 133 |
|
| 134 |
"{instruction}\n\n### Response:"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 132 |
2. For **WizardLM-7B-V1.0** , the Prompt should be as following:
|
| 133 |
|
| 134 |
"{instruction}\n\n### Response:"
|
| 135 |
+
|
| 136 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 137 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_TheBloke__WizardLM-30B-GPTQ)
|
| 138 |
+
|
| 139 |
+
| Metric | Value |
|
| 140 |
+
|-----------------------|---------------------------|
|
| 141 |
+
| Avg. | 38.24 |
|
| 142 |
+
| ARC (25-shot) | 28.84 |
|
| 143 |
+
| HellaSwag (10-shot) | 26.08 |
|
| 144 |
+
| MMLU (5-shot) | 24.62 |
|
| 145 |
+
| TruthfulQA (0-shot) | 49.14 |
|
| 146 |
+
| Winogrande (5-shot) | 76.32 |
|
| 147 |
+
| GSM8K (5-shot) | 34.42 |
|
| 148 |
+
| DROP (3-shot) | 28.29 |
|