Complete Comparison of BERT and ModernBERT
Nov 24, 2025
In my blog post "Code-Aware Tokenization: ModernBERT" I have explored the code awareness in tokenization techniques used in ModernBERT comparted to the traditional BERT tokenizer...
Hello! I'm Mohammed Sbaihi
I'm an AI Engineering apprentice at Onepoint and a final-year student at CentraleSupélec. Welcome to my personal space, where I share blog posts about LLMs, Machine Learning, and hands-on technical guides. You can also explore my projects. Feel free to contact me on LinkedIn for collaborations.
Recent Blogs
Creating your custom Ghibli Text-to-Image model
May 1, 2025
While Ghibli-style AI-generated images have become a global trend, we're taking a unique approach by creating something more culturally meaningful...
Fine-Tuning ModernBERT with LoRA for Classification — Single or Multi-GPU with PyTorch & Accelerate
Feb 10, 2025
This is a straightforward guide to help you fine-tune ModernBERT on any classification task. Any number of labels. Any type of input (code, text, etc.). We’ll walk through the full code...
Code-Aware Tokenization: ModernBERT
Jan 13, 2025
ModernBERT(2024) is the latest encoder-only model at the time of this blog's creation. The authors present it as a Smarter, Better, Faster, and Longer version of BERT, designed to impr...
Research Contributions
Diffguard: Text-Based Safety Checker for Diffusion Models
November 2024
Accepted in poster session at CAID (Conference of AI for Defence) 2025.
Professional & Academic Affiliations
CentraleSupélec
AI Engineering Student
Onepoint
AI Engineer Apprentice
AtlasIA
Data Team Member