Show HN: How LLMs Work – Interactive visual guide based on Karpathy's lecture
5 hours ago
- #LLMs
- #AI
- #ChatGPT
- Large language models such as ChatGPT are built from raw internet text to conversational assistants.
- The process involves training with 15 trillion tokens, 405 billion parameters, 44 terabytes of text data, and a token vocabulary size of 100,000.
- This walkthrough is based on Andrej Karpathy's technical deep dive into the model construction.