Commit
•
9f457c2
1
Parent(s):
40a5134
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (50c6534cfbabdd5639fa03beef81d8fa5a3898c7)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -8,4 +8,17 @@ tags:
|
|
| 8 |
|
| 9 |
Similar to llama2-22b, but with BLOCK_DIAGONAL=false in the merge and twice the fine-tuning tokens.
|
| 10 |
|
| 11 |
-
Again, not intended for direct use - meant as a base for further tuning and merging.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
|
| 9 |
Similar to llama2-22b, but with BLOCK_DIAGONAL=false in the merge and twice the fine-tuning tokens.
|
| 10 |
|
| 11 |
+
Again, not intended for direct use - meant as a base for further tuning and merging.
|
| 12 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 13 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chargoddard__llama2-22b-blocktriangular)
|
| 14 |
+
|
| 15 |
+
| Metric | Value |
|
| 16 |
+
|-----------------------|---------------------------|
|
| 17 |
+
| Avg. | 46.86 |
|
| 18 |
+
| ARC (25-shot) | 58.28 |
|
| 19 |
+
| HellaSwag (10-shot) | 82.69 |
|
| 20 |
+
| MMLU (5-shot) | 54.53 |
|
| 21 |
+
| TruthfulQA (0-shot) | 39.23 |
|
| 22 |
+
| Winogrande (5-shot) | 75.93 |
|
| 23 |
+
| GSM8K (5-shot) | 11.22 |
|
| 24 |
+
| DROP (3-shot) | 6.17 |
|