Commit
•
a7783a2
1
Parent(s):
096e543
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (38d576dcd5836ca3ffd82af2c6da1197aeb6aed1)
Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>
README.md
CHANGED
|
@@ -79,4 +79,17 @@ A small 81M param (total) decoder model, enabled through tying the input/output
|
|
| 79 |
- slightly larger 101M param GQA pretrained version: [here](https://huggingface.co/BEE-spoke-data/smol_llama-101M-GQA)
|
| 80 |
- For the chat version of this model, please [see here](https://youtu.be/dQw4w9WgXcQ?si=3ePIqrY1dw94KMu4)
|
| 81 |
|
| 82 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
- slightly larger 101M param GQA pretrained version: [here](https://huggingface.co/BEE-spoke-data/smol_llama-101M-GQA)
|
| 80 |
- For the chat version of this model, please [see here](https://youtu.be/dQw4w9WgXcQ?si=3ePIqrY1dw94KMu4)
|
| 81 |
|
| 82 |
+
---
|
| 83 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 84 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_BEE-spoke-data__smol_llama-81M-tied)
|
| 85 |
+
|
| 86 |
+
| Metric | Value |
|
| 87 |
+
|-----------------------|---------------------------|
|
| 88 |
+
| Avg. | 24.52 |
|
| 89 |
+
| ARC (25-shot) | 22.18 |
|
| 90 |
+
| HellaSwag (10-shot) | 29.33 |
|
| 91 |
+
| MMLU (5-shot) | 24.06 |
|
| 92 |
+
| TruthfulQA (0-shot) | 43.97 |
|
| 93 |
+
| Winogrande (5-shot) | 49.25 |
|
| 94 |
+
| GSM8K (5-shot) | 0.23 |
|
| 95 |
+
| DROP (3-shot) | 2.64 |
|