Commit
•
ab5a315
1
Parent(s):
9c14057
Adding Evaluation Results (#7)
Browse files- Adding Evaluation Results (69354a15e9488e98c64c92a4bfbd0848f9b054a1)
Co-authored-by: Open LLM Leaderboard PR Bot <leaderboard-pr-bot@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -49,4 +49,17 @@ model-index:
|
|
| 49 |
value: 46.5
|
| 50 |
---
|
| 51 |
|
| 52 |
-
Pretrained GPT2 with expanded n_ctx up to 2048(also with expanded embedding dimension to 1536) in Korean.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
value: 46.5
|
| 50 |
---
|
| 51 |
|
| 52 |
+
Pretrained GPT2 with expanded n_ctx up to 2048(also with expanded embedding dimension to 1536) in Korean.
|
| 53 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 54 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_psyche__kogpt)
|
| 55 |
+
|
| 56 |
+
| Metric | Value |
|
| 57 |
+
|-----------------------|---------------------------|
|
| 58 |
+
| Avg. | 24.27 |
|
| 59 |
+
| ARC (25-shot) | 21.16 |
|
| 60 |
+
| HellaSwag (10-shot) | 28.11 |
|
| 61 |
+
| MMLU (5-shot) | 26.56 |
|
| 62 |
+
| TruthfulQA (0-shot) | 42.06 |
|
| 63 |
+
| Winogrande (5-shot) | 49.09 |
|
| 64 |
+
| GSM8K (5-shot) | 0.0 |
|
| 65 |
+
| DROP (3-shot) | 2.89 |
|