Adding Evaluation Results
#4
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -108,3 +108,17 @@ The following are benchmarks we checked for contamination for:
|
|
| 108 |
- MMLU
|
| 109 |
|
| 110 |
- GPT4All
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
- MMLU
|
| 109 |
|
| 110 |
- GPT4All
|
| 111 |
+
|
| 112 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 113 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_NousResearch__Nous-Capybara-7B)
|
| 114 |
+
|
| 115 |
+
| Metric | Value |
|
| 116 |
+
|-----------------------|---------------------------|
|
| 117 |
+
| Avg. | 49.94 |
|
| 118 |
+
| ARC (25-shot) | 55.29 |
|
| 119 |
+
| HellaSwag (10-shot) | 80.73 |
|
| 120 |
+
| MMLU (5-shot) | 48.72 |
|
| 121 |
+
| TruthfulQA (0-shot) | 51.13 |
|
| 122 |
+
| Winogrande (5-shot) | 73.32 |
|
| 123 |
+
| GSM8K (5-shot) | 6.97 |
|
| 124 |
+
| DROP (3-shot) | 33.44 |
|