Adding Evaluation Results
#1
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -22,4 +22,17 @@ Mainly intended to be utilized with Open
|
|
| 22 |
Source KoboldAI software. Optimal sampler
|
| 23 |
and settings not determined. Feedback Welcome!
|
| 24 |
|
| 25 |
-
https://github.com/KoboldAI/KoboldAI-Client
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
Source KoboldAI software. Optimal sampler
|
| 23 |
and settings not determined. Feedback Welcome!
|
| 24 |
|
| 25 |
+
https://github.com/KoboldAI/KoboldAI-Client
|
| 26 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 27 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_digitous__Javelin-R)
|
| 28 |
+
|
| 29 |
+
| Metric | Value |
|
| 30 |
+
|-----------------------|---------------------------|
|
| 31 |
+
| Avg. | 35.33 |
|
| 32 |
+
| ARC (25-shot) | 41.64 |
|
| 33 |
+
| HellaSwag (10-shot) | 69.01 |
|
| 34 |
+
| MMLU (5-shot) | 30.7 |
|
| 35 |
+
| TruthfulQA (0-shot) | 34.5 |
|
| 36 |
+
| Winogrande (5-shot) | 64.8 |
|
| 37 |
+
| GSM8K (5-shot) | 1.67 |
|
| 38 |
+
| DROP (3-shot) | 5.01 |
|