Adding Evaluation Results
#2
by
leaderboard-pr-bot
- opened
README.md
CHANGED
@@ -12,4 +12,17 @@ Pretty sure I did not use merged models, just finetuned ones so it's not stupid
|
|
12 |
|
13 |
Maybe.
|
14 |
|
15 |
-
My friendo tested it and said S1 was smartest but he liked A1 the most so u choose hehe
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
|
13 |
Maybe.
|
14 |
|
15 |
+
My friendo tested it and said S1 was smartest but he liked A1 the most so u choose hehe
|
16 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
17 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_gradientputri__MegaMix-S1-13B)
|
18 |
+
|
19 |
+
| Metric | Value |
|
20 |
+
|-----------------------|---------------------------|
|
21 |
+
| Avg. | 49.89 |
|
22 |
+
| ARC (25-shot) | 62.46 |
|
23 |
+
| HellaSwag (10-shot) | 83.65 |
|
24 |
+
| MMLU (5-shot) | 57.88 |
|
25 |
+
| TruthfulQA (0-shot) | 44.52 |
|
26 |
+
| Winogrande (5-shot) | 75.85 |
|
27 |
+
| GSM8K (5-shot) | 18.35 |
|
28 |
+
| DROP (3-shot) | 6.51 |
|