Commit
•
5cd6adb
1
Parent(s):
64dd787
Adding Evaluation Results (#1)
Browse files- Adding Evaluation Results (3d8c0265b603e8656012def7e1468c036d197c2d)
Co-authored-by: Open LLM Leaderboard PR Bot <[email protected]>
README.md
CHANGED
|
@@ -51,4 +51,17 @@ Training took about 1 hour using QLoRa on 1xA100, so this model can be recreated
|
|
| 51 |
- Regenerating using gpt-4.
|
| 52 |
- A bit more data with more diversity.
|
| 53 |
- Refactor dataset generation script.
|
| 54 |
-
- Test some model merges using this model.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
- Regenerating using gpt-4.
|
| 52 |
- A bit more data with more diversity.
|
| 53 |
- Refactor dataset generation script.
|
| 54 |
+
- Test some model merges using this model.
|
| 55 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 56 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_totally-not-an-llm__EverythingLM-13b-16k)
|
| 57 |
+
|
| 58 |
+
| Metric | Value |
|
| 59 |
+
|-----------------------|---------------------------|
|
| 60 |
+
| Avg. | 45.7 |
|
| 61 |
+
| ARC (25-shot) | 56.57 |
|
| 62 |
+
| HellaSwag (10-shot) | 80.58 |
|
| 63 |
+
| MMLU (5-shot) | 50.18 |
|
| 64 |
+
| TruthfulQA (0-shot) | 47.46 |
|
| 65 |
+
| Winogrande (5-shot) | 72.77 |
|
| 66 |
+
| GSM8K (5-shot) | 6.44 |
|
| 67 |
+
| DROP (3-shot) | 5.89 |
|