Commit
•
c9ed0ac
1
Parent(s):
cac68b3
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (e8dacf17e4b7ed7052d46f823ee4c131d1686224)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -83,3 +83,17 @@ A small 101M param (total) decoder model. This is the first version of the model
|
|
| 83 |
|
| 84 |
---
|
| 85 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 83 |
|
| 84 |
---
|
| 85 |
|
| 86 |
+
|
| 87 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 88 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_BEE-spoke-data__smol_llama-101M-GQA)
|
| 89 |
+
|
| 90 |
+
| Metric | Value |
|
| 91 |
+
|-----------------------|---------------------------|
|
| 92 |
+
| Avg. | 25.32 |
|
| 93 |
+
| ARC (25-shot) | 23.55 |
|
| 94 |
+
| HellaSwag (10-shot) | 28.77 |
|
| 95 |
+
| MMLU (5-shot) | 24.24 |
|
| 96 |
+
| TruthfulQA (0-shot) | 45.76 |
|
| 97 |
+
| Winogrande (5-shot) | 50.67 |
|
| 98 |
+
| GSM8K (5-shot) | 0.83 |
|
| 99 |
+
| DROP (3-shot) | 3.39 |
|