RAG-DSE-PAST-PAPER-2012-ICT

Gemma 3 4B model fine-tuned for answering questions about DSE ICT 2012 past paper using RAG

Model Description

This model is based on Gemma 3 4B and has been optimized for answering questions about PDF documents using Retrieval Augmented Generation (RAG).

Usage

You can use this model with the Transformers library:

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("khysam2022/RAG-DSE-PAST-PAPER-2012-ICT")
model = AutoModelForCausalLM.from_pretrained("khysam2022/RAG-DSE-PAST-PAPER-2012-ICT")

# Example usage with RAG
prompt = "What is the main topic of this document?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=100)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Training Details

This model was adapted from Google's Gemma 3 4B model and fine-tuned for PDF question-answering.

Limitations and Biases

This model inherits the limitations and biases of the base Gemma 3 model.