Commit
•
521ea99
1
Parent(s):
5f80b37
Adding Evaluation Results (#2)
Browse files- Adding Evaluation Results (d6cf3873f761cb5609eff4f513793e80843d488c)
Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>
README.md
CHANGED
|
@@ -36,4 +36,17 @@ CARL:
|
|
| 36 |
```
|
| 37 |
|
| 38 |
Note:
|
| 39 |
-
This is just a research experiment, and the model should NOT be used as a human therapist. Use "cat" command to join all pytorch_model.bin parts.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
```
|
| 37 |
|
| 38 |
Note:
|
| 39 |
+
This is just a research experiment, and the model should NOT be used as a human therapist. Use "cat" command to join all pytorch_model.bin parts.
|
| 40 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 41 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ajibawa-2023__carl-33b)
|
| 42 |
+
|
| 43 |
+
| Metric | Value |
|
| 44 |
+
|-----------------------|---------------------------|
|
| 45 |
+
| Avg. | 55.01 |
|
| 46 |
+
| ARC (25-shot) | 64.59 |
|
| 47 |
+
| HellaSwag (10-shot) | 85.27 |
|
| 48 |
+
| MMLU (5-shot) | 58.38 |
|
| 49 |
+
| TruthfulQA (0-shot) | 45.32 |
|
| 50 |
+
| Winogrande (5-shot) | 76.24 |
|
| 51 |
+
| GSM8K (5-shot) | 6.37 |
|
| 52 |
+
| DROP (3-shot) | 48.92 |
|