Chinna-0112/distilgpt2

Chinna Nayaki K Fine-Tuned Model

This repository contains the training files for a small personal-details question-answering model based on DistilGPT-2 and LoRA/PEFT.

Model information

• Name: Chinna Nayaki K
• College: Dayanada Sagar University
• Course: MSc Data Science
• Base model: distilgpt2
• Fine-tuning: LoRA using PEFT
• Task: Causal language modeling / text generation

Repository files

• dataset.jsonl — training examples
• train.py — LoRA fine-tuning script
• inference.py — load the trained adapter and ask questions
• requirements.txt — Python dependencies
• README.md — repository documentation

Train

pip install -r requirements.txt
python train.py

The trained LoRA adapter is saved in:

./chinna_nayaki_finetuned/

Run inference

After training:

python inference.py

Example questions:

What is your name?
Who is Chinna Nayaki K?
Which college do you study at?
What are you studying?
Tell me about Chinna Nayaki K.

Important

This is a small demonstration model trained on a very small dataset. Its answers are limited to the information included in the training examples. The Hugging Face repository should contain the generated adapter files (adapter_config.json, adapter_model.safetensors, and tokenizer files) after training/uploading the trained model.

Hugging Face upload

After creating a model repository on Hugging Face, upload the generated files from:

chinna_nayaki_finetuned/

together with this README if you want the repository to contain the trained model.