Medical_Homework2

Medical_Homework2 — Fine-Tuned SmolLM2-1.7B for Medical Reasoning

Medical_Homework2 is a fine-tuned version of SmolAI/SmolLM2-1.7B, trained specifically on structured medical question-answer data and short reasoning tasks.
The model aims to provide concise, accurate, and educational medical explanations suitable for students and basic learning purposes.

---

Model Overview

This model is optimized for medical comprehension tasks such as:

• Short medical answers
• Step-by-step reasoning
• Explanations of conditions, symptoms, and basic physiology
• Educational or homework-style responses

It is not designed for professional medical diagnosis or treatment decisions.

---

Intended Use

Recommended Use Cases

• Medical homework and assignment assistance
• Explanation of medical concepts in simple language
• Introductory physiology and pathology topics
• Basic reasoning about medical questions

Not Recommended

• Real-world clinical decision-making
• Emergency or diagnostic use
• Any situation requiring professional medical judgement

---

Training Data

The model was fine-tuned using:

• Synthetic medical question-answer pairs
• Simplified educational medical explanations
• Instruction-answer examples
• Homework-style reasoning data

No real patient data or clinical records were used.

---

Training Details

• Base model: SmolAI/SmolLM2-1.7B
• Fine-tuning objective: Causal language modeling
• Method: Full or LoRA fine-tuning (depending on your actual setup)
• Optimizer: AdamW
• Typical epochs: 1–3

If you want, a full training script section can be added.

---

Usage Example

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "Abeersherif/Medical_Homework2"

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

prompt = "Explain what type 2 diabetes is in simple terms."

inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(
    **inputs,
    max_new_tokens=150,
    temperature=0.7,
    top_p=0.9,
)

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