Adding Evaluation Results
#2
by leaderboard-pr-bot - opened
README.md
CHANGED
|
@@ -11,4 +11,17 @@ This repo contains the merged fp16 model.
|
|
| 11 |
- language:
|
| 12 |
- en <br>
|
| 13 |
- reference: https://github.com/hiyouga/LLaMA-Efficient-Tuning/tree/main
|
| 14 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
- language:
|
| 12 |
- en <br>
|
| 13 |
- reference: https://github.com/hiyouga/LLaMA-Efficient-Tuning/tree/main
|
| 14 |
+
---
|
| 15 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 16 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_TheTravellingEngineer__bloom-560m-RLHF)
|
| 17 |
+
|
| 18 |
+
| Metric | Value |
|
| 19 |
+
|-----------------------|---------------------------|
|
| 20 |
+
| Avg. | 26.17 |
|
| 21 |
+
| ARC (25-shot) | 24.4 |
|
| 22 |
+
| HellaSwag (10-shot) | 36.96 |
|
| 23 |
+
| MMLU (5-shot) | 23.63 |
|
| 24 |
+
| TruthfulQA (0-shot) | 40.76 |
|
| 25 |
+
| Winogrande (5-shot) | 53.12 |
|
| 26 |
+
| GSM8K (5-shot) | 0.3 |
|
| 27 |
+
| DROP (3-shot) | 3.99 |
|