gemma-3n-vneid-merged

🇻🇳 Vietnamese Fine-tuned Gemma Model

This is a Vietnamese fine-tuned version of Google's Gemma 2B model using Unsloth and LoRA adapters, optimized for Vietnamese text generation.

📊 Model Details

• Base Model: google/gemma-2-2b
• Language: Vietnamese (vi)
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Framework: Unsloth
• Model Type: Causal Language Model
• License: Apache 2.0

🚀 Quick Start

Using Transformers

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Load model and tokenizer
model_name = "anhtuan15082023/gemma-3n-vneid-merged"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto",
    trust_remote_code=True
)

# Generate Vietnamese text
def generate_vietnamese_text(prompt, max_length=100):
    inputs = tokenizer(prompt, return_tensors="pt")
    
    with torch.no_grad():
        outputs = model.generate(
            **inputs,
            max_length=max_length,
            temperature=0.7,
            do_sample=True,
            top_p=0.9,
            pad_token_id=tokenizer.eos_token_id
        )
    
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response[len(prompt):].strip()

# Example usage
prompt = "Xin chào, tôi là"
result = generate_vietnamese_text(prompt)
print(f"Input: {prompt}")
print(f"Output: {result}")

Using Inference API

import requests

API_URL = "https://api-inference.huggingface.co/models/anhtuan15082023/gemma-3n-vneid-merged"
headers = {"Authorization": f"Bearer {YOUR_HF_TOKEN}"}

def query(payload):
    response = requests.post(API_URL, headers=headers, json=payload)
    return response.json()

# Generate text
output = query({
    "inputs": "Việt Nam là",
    "parameters": {
        "max_length": 100,
        "temperature": 0.7
    }
})
print(output)

🎯 Use Cases

• Vietnamese text completion
• Creative writing in Vietnamese
• Chatbot responses in Vietnamese
• Content generation for Vietnamese applications

⚙️ Training Details

• Dataset: Vietnamese text corpus
• Training Framework: Unsloth (optimized training)
• Fine-tuning Method: LoRA adapters merged into base model
• Base Model: Google Gemma 2B

🏷️ Model Tags

• Vietnamese language model
• Text generation
• Fine-tuned Gemma
• LoRA adaptation

📜 License

This model inherits the Apache 2.0 license from the base Gemma model.

🤝 Citation

If you use this model, please consider citing:

@model{vietnamese-gemma-finetuned,
  title={Vietnamese Fine-tuned Gemma Model},
  author={anhtuan15082023},
  year={2024},
  url={https://huggingface.co/anhtuan15082023/gemma-3n-vneid-merged}
}

📞 Contact

For questions or issues, please open an issue on the model's repository page.