FareedKhan-dev/train-llm-from-scratch

A straightforward method for training your LLM, from downloading data to generating text.

Jupyter Notebook9.3K stars1.3K forks+316 today

Topics

#gemini#large-language-models#llm#openai#training#transformers

What it does

This repository provides a comprehensive guide and code for training large language models from scratch using PyTorch, making it accessible to researchers and developers interested in AI.

Star history

Not enough history yet — 1 day(s) recorded. The daily snapshot builds this up.

Tracking

Last trending: 2026-05-30

Creator kit

Hook

Train your own billion or million parameter LLM with a single GPU! Dive into FareedKhan-dev's guide and codebase for an in-depth look at transformer models.

Content angles

Step-by-step tutorial on training large language models
Comparison of different datasets used for training
Exploring the impact of hyperparameters on model performance

Who should care

Researchers, developers, and data scientists interested in AI and natural language processing.