Adding Evaluation Results
#1
by
leaderboard-pr-bot
- opened
README.md
CHANGED
|
@@ -60,3 +60,17 @@ Star Wars Dataset Eval (Rogue One Novel)
|
|
| 60 |
```
|
| 61 |
|
| 62 |
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to Concedo on the KoboldAI discord.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
```
|
| 61 |
|
| 62 |
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to Concedo on the KoboldAI discord.
|
| 63 |
+
|
| 64 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 65 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_KoboldAI__OPT-2.7B-Nerybus-Mix)
|
| 66 |
+
|
| 67 |
+
| Metric | Value |
|
| 68 |
+
|-----------------------|---------------------------|
|
| 69 |
+
| Avg. | 32.29 |
|
| 70 |
+
| ARC (25-shot) | 33.7 |
|
| 71 |
+
| HellaSwag (10-shot) | 61.21 |
|
| 72 |
+
| MMLU (5-shot) | 26.6 |
|
| 73 |
+
| TruthfulQA (0-shot) | 37.57 |
|
| 74 |
+
| Winogrande (5-shot) | 62.04 |
|
| 75 |
+
| GSM8K (5-shot) | 0.15 |
|
| 76 |
+
| DROP (3-shot) | 4.79 |
|