Update README.md
Browse files
README.md
CHANGED
|
@@ -45,6 +45,7 @@ The training process consisted of three distinct stages:
|
|
| 45 |
|
| 46 |
### Stage 3: Supervised Fine-Tuning (Using RWKV-LM-RLHF)
|
| 47 |
- Utilized a distillation dataset of 900K samples
|
|
|
|
| 48 |
- Implemented Variable Rank PEFT to enhance training efficiency
|
| 49 |
- Bone(Block Affine Transformation) r=512+
|
| 50 |
|
|
|
|
| 45 |
|
| 46 |
### Stage 3: Supervised Fine-Tuning (Using RWKV-LM-RLHF)
|
| 47 |
- Utilized a distillation dataset of 900K samples
|
| 48 |
+
- Smoothed Loss for faster convergence
|
| 49 |
- Implemented Variable Rank PEFT to enhance training efficiency
|
| 50 |
- Bone(Block Affine Transformation) r=512+
|
| 51 |
|