Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,7 @@
|
|
| 1 |
The Llama3-8b-based Reward Model was trained using OpenRLHF and a combination of datasets available at https://huggingface.co/datasets/OpenLLMAI/preference_dataset_mixture2_and_safe_pku.
|
| 2 |
|
|
|
|
|
|
|
| 3 |
```
|
| 4 |
Cosine Scheduler
|
| 5 |
Learning Rate: 9e-6
|
|
|
|
| 1 |
The Llama3-8b-based Reward Model was trained using OpenRLHF and a combination of datasets available at https://huggingface.co/datasets/OpenLLMAI/preference_dataset_mixture2_and_safe_pku.
|
| 2 |
|
| 3 |
+
Base model: https://huggingface.co/OpenRLHF/Llama-3-8b-sft-mixture
|
| 4 |
+
|
| 5 |
```
|
| 6 |
Cosine Scheduler
|
| 7 |
Learning Rate: 9e-6
|