Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -39,6 +39,8 @@ Fine-tuned [Qwen/Qwen3.5-27B](https://huggingface.co/Qwen/Qwen3.5-27B) for **Tex
|
|
| 39 |
| **MMLU (social sciences)** | 92.0% | 92.0% | 0% (no regression) |
|
| 40 |
| **MMLU (other)** | 87.5% | 87.5% | 0% (no regression) |
|
| 41 |
| **GSM8K (math, strict)** | 60.4% | 35.4% | **-25.0% (regression)** |
|
|
|
|
|
|
|
| 42 |
|
| 43 |
## Usage
|
| 44 |
|
|
|
|
| 39 |
| **MMLU (social sciences)** | 92.0% | 92.0% | 0% (no regression) |
|
| 40 |
| **MMLU (other)** | 87.5% | 87.5% | 0% (no regression) |
|
| 41 |
| **GSM8K (math, strict)** | 60.4% | 35.4% | **-25.0% (regression)** |
|
| 42 |
+
| **HellaSwag (common sense)** | 79.6% | 84.1% | +4.5% (improved) |
|
| 43 |
+
| **ARC-Challenge (reasoning)** | 69.3% | 71.3% | +2.0% (improved) |
|
| 44 |
|
| 45 |
## Usage
|
| 46 |
|