1
GigaGPT: GPT-3 sized models in 565 lines of code
GigaGPT is Cerebras’ implementation of Andrei Karpathy’s nanoGPT – the simplest and most compact code base to train and fine-tune GPT models.
Cerebras powers the world's fastest AI inference on the biggest wafer chip. Cerebras CS-4 delivers up to 30x faster inference than GPUs.
Visit cerebras.net → On Hacker News Mar 2023 – Dec 2023
GigaGPT is Cerebras’ implementation of Andrei Karpathy’s nanoGPT – the simplest and most compact code base to train and fine-tune GPT models.