vsingh10/bitnet-medical-chat

🩺 Hi! I am Dr. Bitsy, your personal healthcare assistant, how can I assist you ?

This is a fine-tuned version of microsoft/bitnet-b1.58-2B-4T-bf16 using LoRA adapters on a medical chatbot dataset.
It is designed to act as a helpful and knowledgeable medical assistant for answering patient queries with medically accurate and detailed explanations.

---

📊 Model Details

• Base model: microsoft/bitnet-b1.58-2B-4T-bf16
• Fine-tuning method: LoRA (Low-Rank Adaptation)
• Merged model size: ~2B parameters
• Framework: Transformers, PEFT, TRL
• Dataset: ruslanmv/ai-medical-chatbot
  ~5,000 samples of patient–doctor dialogues.

---

🧑‍⚕️ Intended Use

• Research on medical dialogue generation
• Experimentation with domain adaptation of BitNet
• Exploration of LoRA fine-tuning for healthcare-related tasks

Not intended for:

• Real clinical use
• Emergency healthcare guidance

---

🚀 Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "vsingh10/bitnet-medical-chat"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

messages = [
    {"role": "system", "content": "You are a helpful and knowledgeable medical doctor. Always provide detailed, medically accurate explanations."},
    {"role": "user", "content": "Hello doctor, I have bad acne. How do I get rid of it?"}
]

prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")

outputs = model.generate(
    **inputs,
    max_new_tokens=300,
    temperature=0.7,
    top_p=0.9,
    do_sample=True
)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))