Training NanoGPT in 39.9 Seconds — Hyperstition
hyperstition.cc · saved by 4 readers
How I brought NanoGPT training from 73.889 to 39.914 seconds with ANVIL II, sampled softmax, and sparse updates to the bigram and trigram tables.
How I brought NanoGPT training from 73.889 to 39.914 seconds with ANVIL II, sampled softmax, and sparse updates to the bigram and trigram tables.