Commit
·
8425fdc
1
Parent(s):
339ddac
Update README.md
Browse files
README.md
CHANGED
|
@@ -10,12 +10,6 @@ tags:
|
|
| 10 |
- rlaif
|
| 11 |
- rlhf
|
| 12 |
---
|
| 13 |
-
<p align="right">
|
| 14 |
-
<a href="https://github.com/argilla-io/distilabel">
|
| 15 |
-
<img src="https://raw.githubusercontent.com/argilla-io/distilabel/main/docs/assets/distilabel-badge-light.png" alt="Built with Distilabel" width="200" height="32"/>
|
| 16 |
-
</a>
|
| 17 |
-
</p>
|
| 18 |
-
|
| 19 |
# ⚗️ distilabeled OpenHermes 2.5 Mistral 7B
|
| 20 |
|
| 21 |
> A Half Neural DPO of OpenHermes 2.5, less is more for DPO!
|
|
@@ -24,6 +18,11 @@ tags:
|
|
| 24 |
<img src="https://cdn-uploads.huggingface.co/production/uploads/60420dccc15e823a685f2b03/yWdvBtKKfJdpdnPiSlNb9.png">
|
| 25 |
</div>
|
| 26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
## Introduction
|
| 29 |
This model is the launching partner of our new open dataset [argilla/distilabel-intel-orca-dpo-pairs](https://huggingface.co/datasets/argilla/distilabel-intel-orca-dpo-pairs). It outperforms the awesome `mlabonne/NeuralHermes-2.5-Mistral-7B` with the **exact same DPO recipe but 54% less data**.
|
|
|
|
| 10 |
- rlaif
|
| 11 |
- rlhf
|
| 12 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
# ⚗️ distilabeled OpenHermes 2.5 Mistral 7B
|
| 14 |
|
| 15 |
> A Half Neural DPO of OpenHermes 2.5, less is more for DPO!
|
|
|
|
| 18 |
<img src="https://cdn-uploads.huggingface.co/production/uploads/60420dccc15e823a685f2b03/yWdvBtKKfJdpdnPiSlNb9.png">
|
| 19 |
</div>
|
| 20 |
|
| 21 |
+
<p align="center">
|
| 22 |
+
<a href="https://github.com/argilla-io/distilabel">
|
| 23 |
+
<img src="https://raw.githubusercontent.com/argilla-io/distilabel/main/docs/assets/distilabel-badge-light.png" alt="Built with Distilabel" width="200" height="32"/>
|
| 24 |
+
</a>
|
| 25 |
+
</p>
|
| 26 |
|
| 27 |
## Introduction
|
| 28 |
This model is the launching partner of our new open dataset [argilla/distilabel-intel-orca-dpo-pairs](https://huggingface.co/datasets/argilla/distilabel-intel-orca-dpo-pairs). It outperforms the awesome `mlabonne/NeuralHermes-2.5-Mistral-7B` with the **exact same DPO recipe but 54% less data**.
|