Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- aligner/aligner-20K
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
tags:
|
| 7 |
+
- safety
|
| 8 |
+
- ai-safety
|
| 9 |
+
- aligner
|
| 10 |
+
- llama
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
# Aligner Model Card
|
| 15 |
+
|
| 16 |
+
## Model details
|
| 17 |
+
|
| 18 |
+
Aligner is a model-agnostic plug-and-play module that works on open-source and API-based models. It is trained based on [Llama2-Base](https://huggingface.co/meta-llama), using the *residual correction* strategy.
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
## Model Sources
|
| 22 |
+
|
| 23 |
+
- **Repository:** <https://github.com/Aligner2024/aligner>
|
| 24 |
+
- **Dataset:**<https://huggingface.co/datasets/aligner/aligner-20K>
|
| 25 |
+
|
| 26 |
+
## More Details
|
| 27 |
+
|
| 28 |
+
- **Website:**<https://aligner2024.github.io/>
|
| 29 |
+
|
| 30 |
+
*More aligners (7B,13B,70B) trained across different datasets (20K,30K,40K,50K) will come soon...*
|
| 31 |
+
|