Update README.md
Browse files
README.md
CHANGED
|
@@ -1,4 +1,25 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
pipeline_tag: text-generation
|
| 4 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- nlp
|
| 5 |
+
- text-generation
|
| 6 |
+
- bitnet
|
| 7 |
+
- pytorch
|
| 8 |
+
- tensorflow
|
| 9 |
pipeline_tag: text-generation
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# PlayWeBit (BitNet b1.58-2B-4T)
|
| 13 |
+
|
| 14 |
+
This is an implementation of the `microsoft/bitnet-b1.58-2B-4T` model, optimized for text generation. It supports PyTorch and TensorFlow and can run on CPU or GPU. A custom model class (`custom_bitnet.py`) is included for compatibility with the official Hugging Face `transformers` library.
|
| 15 |
+
|
| 16 |
+
## Model Description
|
| 17 |
+
- **Architecture**: BitNet b1.58-2B with 4-bit quantization
|
| 18 |
+
- **Parameters**: 2 billion
|
| 19 |
+
- **Training Data**: Not disclosed (refer to original Microsoft repository)
|
| 20 |
+
- **Intended Use**: Text generation, chat applications
|
| 21 |
+
|
| 22 |
+
## Installation
|
| 23 |
+
Install the Hugging Face `transformers` library:
|
| 24 |
+
```bash
|
| 25 |
+
pip install transformers
|