Agent Skills: nanogpt

Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).

UncategorizedID: benchflow-ai/skillsbench/nanogpt

Author

benchflow-ai

https://github.com/benchflow-ai View all skills

Repository

benchflow-ai/skillsbench

benchflow-aiLicense: Apache-2.0

278174

Install this agent skill to your local

pnpm dlx add-skill https://github.com/benchflow-ai/skillsbench/nanogpt

Skill Files

Browse the full folder contents for nanogpt.

Download Skill

Loading file tree…

Select a file to preview its contents.