CompletedPythonTensorFlowGoogle Colab+2 more
Fine-tuning LLM with Google Colab
Step-by-step walkthrough of fine-tuning Tiny-LLaMA inside Google Colab — covering tokenization, training setup, model export, and local deployment.
Timeline
November 2025 - December 2025
Role
AI/ML Engineer
Status
CompletedTechnology Stack
Python
TensorFlow
Google Colab
Hugging Face
PyTorch
Fine-tuning LLM with Google Colab
Overview
A comprehensive project demonstrating how to fine-tune the Tiny-LLaMA model using Google Colab, making AI model customization accessible to everyone. Uses a dataset that converts HTML to JSON, with detailed explanations of each step.
What's Covered
- What each import and library does
- How tokenization and training setup works
- LoRA (Parameter-Efficient Fine-Tuning) configuration
- How to fine-tune and export your model
- How to download and run it locally
Tech Stack
- Language: Python
- Platform: Google Colab
- Model: Tiny-LLaMA 1.1B
- Libraries: Hugging Face Transformers, PyTorch, PEFT, TRL, bitsandbytes