Commit
•
4f9f2bc
1
Parent(s):
2335683
Adding Evaluation Results (#2)
Browse files- Adding Evaluation Results (271cee23f1d7656cc1edca0ac8591a2ae01bb3f0)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -15,4 +15,17 @@ Recipe:
|
|
| 15 |
* WIP storytelling LORA
|
| 16 |
|
| 17 |
|
| 18 |
-
Responds well to the Alpaca prompt format.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
* WIP storytelling LORA
|
| 16 |
|
| 17 |
|
| 18 |
+
Responds well to the Alpaca prompt format.
|
| 19 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 20 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chargoddard__storytime-13b)
|
| 21 |
+
|
| 22 |
+
| Metric | Value |
|
| 23 |
+
|-----------------------|---------------------------|
|
| 24 |
+
| Avg. | 50.55 |
|
| 25 |
+
| ARC (25-shot) | 62.03 |
|
| 26 |
+
| HellaSwag (10-shot) | 83.96 |
|
| 27 |
+
| MMLU (5-shot) | 57.48 |
|
| 28 |
+
| TruthfulQA (0-shot) | 52.5 |
|
| 29 |
+
| Winogrande (5-shot) | 75.53 |
|
| 30 |
+
| GSM8K (5-shot) | 8.34 |
|
| 31 |
+
| DROP (3-shot) | 14.0 |
|