IronMan19/Fine-tune-science-tutor-mistral-7b-lora

🧠 AI Science Tutor (Mistral-7B LoRA)

This repository contains a fine-tuned AI tutoring model built using:

• Base model: mistralai/Mistral-7B-Instruct-v0.2
• Fine-tuning method: LoRA (PEFT)
• Task: Educational tutoring (step-by-step explanations)

---

What’s inside?

• LoRA adapter weights (adapter_model.safetensors)
• Adapter config (adapter_config.json)
• Tokenizer (with embedded chat template)
• Ready-to-use inference pipeline

---

How to Use

Load Model + Tokenizer

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

model_id = "IronMan19/Fine-tune-science-tutor-mistral-7b-lora"

tokenizer = AutoTokenizer.from_pretrained(model_id)

base_model = AutoModelForCausalLM.from_pretrained(
    "mistralai/Mistral-7B-Instruct-v0.2",
    device_map="auto",
    torch_dtype="auto"
)

model = PeftModel.from_pretrained(base_model, model_id)
model.eval()