Update README.md
Browse files
README.md
CHANGED
|
@@ -31,7 +31,7 @@ At 500 steps, the loss was plateauing so I decided to stop training to prevent e
|
|
| 31 |
#### Training Details
|
| 32 |
|
| 33 |
- **Base Model**: Qwen 2.5-14B
|
| 34 |
-
- **Fine-Tuning Dataset**: Verified subset of **NuminaMathCoT** using Qwen 2.5 3B Instruct as a judge.
|
| 35 |
- **QLoRA Configuration**:
|
| 36 |
- **Rank**: 32
|
| 37 |
- **Rank Stabilization**: Enabled
|
|
|
|
| 31 |
#### Training Details
|
| 32 |
|
| 33 |
- **Base Model**: Qwen 2.5-14B
|
| 34 |
+
- **Fine-Tuning Dataset**: Verified subset of **NuminaMathCoT** using Qwen 2.5 3B Instruct as a judge. (the `sharegpt-verified-cleaned` subset from my dataset).
|
| 35 |
- **QLoRA Configuration**:
|
| 36 |
- **Rank**: 32
|
| 37 |
- **Rank Stabilization**: Enabled
|