A from-scratch, dependency-free GPT implementation in pure NumPy with a byte-level tokenizer, multi-head self-attention, Adam trainer, and top-k sampler.
Read the original at github.com→★ 51 stars
Original headline: "prabeakarve/microgpt: A from-scratch, dependency-free GPT implementation in pure NumPy: byte-level tokenizer, multi-head self-attention, Adam trainer, and top-k sampler."
Coverage timeline
- Nov 5, 12:21 UTC GitHub Trending (AI) lead source prabeakarve/microgpt: A from-scratch, dependency-free GPT implementation in pure NumPy: byte-level tokenizer, multi-head self-attention, Adam trainer, and top-k sampler.