Commit
•
7fb67dc
1
Parent(s):
88775ea
Update README.md
Browse files
README.md
CHANGED
|
@@ -75,6 +75,8 @@ The used dataset does not allow commercial usage.
|
|
| 75 |
|
| 76 |
The training notebook is available at the following link: [Qwen1_5_7B_Dutch_Chat_DPO](https://github.com/RobinSmits/Dutch-LLMs/blob/main/Qwen1_5_7B_Dutch_Chat_DPO.ipynb)
|
| 77 |
|
|
|
|
|
|
|
| 78 |
It achieves the following results on the evaluation set:
|
| 79 |
- Loss: 0.2610
|
| 80 |
- Rewards/chosen: -0.7248
|
|
|
|
| 75 |
|
| 76 |
The training notebook is available at the following link: [Qwen1_5_7B_Dutch_Chat_DPO](https://github.com/RobinSmits/Dutch-LLMs/blob/main/Qwen1_5_7B_Dutch_Chat_DPO.ipynb)
|
| 77 |
|
| 78 |
+
Training was performed with Google Colab PRO on a A100 - 40GB and lasted around 4 hours.
|
| 79 |
+
|
| 80 |
It achieves the following results on the evaluation set:
|
| 81 |
- Loss: 0.2610
|
| 82 |
- Rewards/chosen: -0.7248
|