Update README.md
Browse files
README.md
CHANGED
|
@@ -23,7 +23,7 @@ The model is designed for natural language processing tasks, including text gene
|
|
| 23 |
|
| 24 |
## Training
|
| 25 |
|
| 26 |
-
This model is obtained through full-parameter fine-tuning. For each dataset, only the corresponding experts are fine-tuned.
|
| 27 |
|
| 28 |
- **Coding**: First 20k rows of `nvidia/OpenCodeReasoning` dataset for 1 epoch.
|
| 29 |
- **Math**: Entire `unsloth/OpenMathReasoning-mini` dataset for 1 epoch.
|
|
|
|
| 23 |
|
| 24 |
## Training
|
| 25 |
|
| 26 |
+
This model is obtained through full-parameter fine-tuning. For each dataset(max_length=1024), only the corresponding experts are fine-tuned.
|
| 27 |
|
| 28 |
- **Coding**: First 20k rows of `nvidia/OpenCodeReasoning` dataset for 1 epoch.
|
| 29 |
- **Math**: Entire `unsloth/OpenMathReasoning-mini` dataset for 1 epoch.
|