Update README.md
Browse files
README.md
CHANGED
|
@@ -7,4 +7,30 @@ sdk: static
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# 🚀 Caca Transformers by Lyon
|
| 11 |
+
|
| 12 |
+
Custom Transformer architecture with modern features for efficient language modeling.
|
| 13 |
+
|
| 14 |
+
## 🎯 Features
|
| 15 |
+
|
| 16 |
+
- **30+ Model Variants** (10M - 1T parameters)
|
| 17 |
+
- **Grouped Query Attention (GQA)**
|
| 18 |
+
- **RoPE Positional Encoding**
|
| 19 |
+
- **SwiGLU Activation**
|
| 20 |
+
- **Flash Attention Support**
|
| 21 |
+
|
| 22 |
+
## 📦 Installation
|
| 23 |
+
|
| 24 |
+
```bash
|
| 25 |
+
pip install caca-transformers
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
## 🤗 Available Models
|
| 29 |
+
|
| 30 |
+
Browse all models below ↓
|
| 31 |
+
|
| 32 |
+
## 🔗 Links
|
| 33 |
+
|
| 34 |
+
- PyPI: https://pypi.org/project/caca-transformers/
|
| 35 |
+
- GitHub: https://github.com/Lyon-28/caca-transformers
|
| 36 |
+
- Email: cacatransformers@gmail.com
|