Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
datasets:
|
| 4 |
+
- BlossomsAI/reduced_vietnamese_instruction_dataset
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
- vi
|
| 8 |
+
base_model:
|
| 9 |
+
- sail/Sailor2-8B
|
| 10 |
+
---
|
| 11 |
+
<div align="center">
|
| 12 |
+
<img src="https://github.com/bloomifycafe/blossomsAI/blob/main/assets/logo.png?raw=true" alt="Logo"/>
|
| 13 |
+
</div>
|
| 14 |
+
</br>
|
| 15 |
+
<div align="center">
|
| 16 |
+
|
| 17 |
+
# 🌟 BloomVN-8B-instruct
|
| 18 |
+
|
| 19 |
+
</div>
|
| 20 |
+
|
| 21 |
+
### A fine-tuned multilingual model for Vietnamese language
|
| 22 |
+
|
| 23 |
+
## 📋 Overview
|
| 24 |
+
|
| 25 |
+
- This model serves as an experimental implementation of instruction fine-tuning for Vietnamese language processing. Built primarily as a proof-of-concept to test unsloth's training capabilities, it was trained on a limited dataset.
|
| 26 |
+
|
| 27 |
+
Key characteristics:
|
| 28 |
+
- Experimental model focused on Vietnamese language
|
| 29 |
+
- Limited training data
|
| 30 |
+
- Primary purpose: Testing unsloth's training performance
|
| 31 |
+
- Not intended for production use
|
| 32 |
+
|
| 33 |
+
Note: This is a research prototype meant for evaluating training methodologies rather than practical applications.
|
| 34 |
+
|
| 35 |
+
## 🔧 Method
|
| 36 |
+
|
| 37 |
+
- The fine-tuning process was conducted using unsloth, with the primary focus on running sample experiments to test the framework's capabilities.
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
The experimentation involved:
|
| 41 |
+
- Using unsloth for model fine-tuning
|
| 42 |
+
- Running several basic test samples
|
| 43 |
+
- Evaluating training behavior
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
## 🤝 Contributors
|
| 47 |
+
|
| 48 |
+
Developed with ❤️ by [BlossomAI](https://github.com/BlossomAI)
|
| 49 |
+
|
| 50 |
+
---
|
| 51 |
+
<div align="center">
|
| 52 |
+
<sub>Star ⭐️ this repo if you find it valuable!</sub>
|
| 53 |
+
</div>
|