ai-medical-model-32bit: Fine-Tuned Llama3 for Technical Medical Questions

This repository provides a fine-tuned version of the powerful Llama3 8B Instruct model, specifically designed to answer medical questions in an informative way. It leverages the rich knowledge contained in the AI Medical Dataset (ruslanmv/ai-medical-dataset).

Model & Development

• Developed by: ruslanmv
• License: Apache-2.0
• Finetuned from model: meta-llama/Meta-Llama-3-8B-Instruct

Key Features

• Medical Focus: Optimized to address health-related inquiries.
• Knowledge Base: Trained on a comprehensive medical dataset.
• Text Generation: Generates informative and potentially helpful responses.

Installation

This model is accessible through the Hugging Face Transformers library. Install it using pip:

!python -m pip install --upgrade pip
!pip3 install torch==2.2.1  torchvision torchaudio xformers --index-url https://download.pytorch.org/whl/cu121
!pip install  bitsandbytes  accelerate

Usage Example

Here's a Python code snippet demonstrating how to interact with the ai-medical-model-32bit model and generate answers to your medical questions:

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig 
import torch
model_name = "ruslanmv/ai-medical-model-32bit"
device_map = 'auto' 
bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.float16,
)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    quantization_config=bnb_config,
    trust_remote_code=True,
    use_cache=False,
    device_map=device_map
)
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
tokenizer.pad_token = tokenizer.eos_token

def askme(question):
  prompt = f"<|start_header_id|>system<|end_header_id|> You are a Medical AI chatbot assistant. <|eot_id|><|start_header_id|>User: <|end_header_id|>This is the question: {question}<|eot_id|>"
  # Tokenizing the input and generating the output
  #prompt = f"{question}"
  # Tokenizing the input and generating the output
  inputs = tokenizer([prompt], return_tensors="pt").to("cuda")
  outputs = model.generate(**inputs, max_new_tokens=256, use_cache=True)
  answer = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
  # Try Remove the prompt
  try:
      # Split the answer at the first line break, assuming system intro and question are on separate lines
      answer_parts = answer.split("\n", 1)
      # If there are multiple parts, consider the second part as the answer
      if len(answer_parts) > 1:
        answers = answer_parts[1].strip()  # Remove leading/trailing whitespaces
      else:
        answers = ""  # If no split possible, set answer to empty string
      print(f"Answer: {answers}")   
  except:
      print(answer)  

# Example usage
# - Question:  Make the question.
question="What was the main cause of the inflammatory CD4+ T cells?"
askme(question)

the type of answer is :

Answer: I'm happy to help!

The main cause of inflammatory CD4+ T cells is a complex process that involves multiple factors. However, some of the key triggers include:

1. Activation of CD4+ T cells: CD4+ T cells are activated by antigens, cytokines, and other signals, leading to their proliferation and differentiation into effector cells.
2. Cytokine production: Activated CD4+ T cells produce cytokines such as interleukin-2 (IL-2), interferon-gamma (IFN-γ), and tumor necrosis factor-alpha (TNF-α), which promote inflammation and immune responses.
3. Chemokine production: CD4+ T cells also produce chemokines, such as CCL3, CCL4, and CCL5, which attract other immune cells to the site of inflammation.
4. Toll-like receptor (TLR) activation: TLRs are pattern recognition receptors that recognize pathogen-associated molecular patterns (PAMPs) and activate CD4+ T cells.
5. Bacterial or viral infections: Infections caused by bacteria, viruses, or fungi can trigger the activation of CD4+ T cells and the production of cytokines and chemokines

Important Note

This model is intended for informational purposes only and should not be used as a substitute for professional medical advice. Always consult with a qualified healthcare provider for any medical concerns.

License

This model is distributed under the Apache License 2.0 (see LICENSE file for details).

Contributing

We welcome contributions to this repository! If you have improvements or suggestions, feel free to create a pull request.

Disclaimer

While we strive to provide informative responses, the accuracy of the model's outputs cannot be guaranteed.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	67.67
AI2 Reasoning Challenge (25-Shot)	61.43
HellaSwag (10-Shot)	78.69
MMLU (5-Shot)	68.10
TruthfulQA (0-shot)	51.99
Winogrande (5-shot)	75.77
GSM8k (5-shot)	70.05