JH-Qwen3-8B: LoRA-Tuned Qwen3-8B Model

This repository contains a fine-tuned version of Qwen/Qwen3-8B using LoRA (Low-Rank Adaptation). The model has been merged with its LoRA weights for easier deployment.

Model Details

• Base Model: Qwen/Qwen3-8B
• Architecture: qwen3
• Size: 8B parameters (8,190,735,360)
• Framework: PyTorch
• Task: text-generation
• License: Apache 2.0

Usage

You can use this model directly with the Hugging Face Transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model_name = "jhghar/jh-qwen3-8b"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

# Example usage
prompt = "Hello, how are you?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=100)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Model Architecture

This model uses the Qwen3 architecture, which is based on a transformer decoder-only model. The base model has been fine-tuned using LoRA and the weights have been merged for production use.

Files in this Repository

• Model weights split into 4 safetensors files
• Configuration files (config.json, tokenizer_config.json)
• Tokenizer files (tokenizer.json, vocab.json, merges.txt, added_tokens.json)
• Special tokens mapping

License

This model is licensed under the Apache License 2.0. Please see the LICENSE file for details.

Citation

If you use this model, please cite both the original Qwen3 model and this adaptation:

@misc{qwen3-8b,
  author = {Qwen Team},
  title = {Qwen3-8B Language Model},
  year = {2024},
  publisher = {Hugging Face},
  journal = {Hugging Face repository},
  howpublished = {\url{https://huggingface.co/Qwen/Qwen3-8B}}
}