Commit
•
e0660af
1
Parent(s):
45ba2f6
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (b573826118ff52ab8d50c21732b9c0f68b2c7e5d)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -6,4 +6,17 @@ language:
|
|
| 6 |
|
| 7 |
A Gradient Merge of Stheno-P1 and Stheno-P2, using [BlockMerge_Gradient](https://github.com/Gryphe/BlockMerge_Gradient) using a script modified by @Vali to replace the tensor calculations with SLERP instead.
|
| 8 |
|
| 9 |
-
So far its pretty good in personal tests.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
A Gradient Merge of Stheno-P1 and Stheno-P2, using [BlockMerge_Gradient](https://github.com/Gryphe/BlockMerge_Gradient) using a script modified by @Vali to replace the tensor calculations with SLERP instead.
|
| 8 |
|
| 9 |
+
So far its pretty good in personal tests.
|
| 10 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 11 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Sao10K__Stheno-1.3-L2-13B)
|
| 12 |
+
|
| 13 |
+
| Metric | Value |
|
| 14 |
+
|-----------------------|---------------------------|
|
| 15 |
+
| Avg. | 49.32 |
|
| 16 |
+
| ARC (25-shot) | 56.83 |
|
| 17 |
+
| HellaSwag (10-shot) | 81.7 |
|
| 18 |
+
| MMLU (5-shot) | 52.79 |
|
| 19 |
+
| TruthfulQA (0-shot) | 50.23 |
|
| 20 |
+
| Winogrande (5-shot) | 71.11 |
|
| 21 |
+
| GSM8K (5-shot) | 0.23 |
|
| 22 |
+
| DROP (3-shot) | 32.34 |
|