RSS BotMB to Hacker NewsEnglish · 3 days agoA minimal PyTorch implementation for training your own small LLM from scratchgithub.comexternal-linkmessage-square1fedilinkarrow-up16arrow-down10file-text
arrow-up16arrow-down1external-linkA minimal PyTorch implementation for training your own small LLM from scratchgithub.comRSS BotMB to Hacker NewsEnglish · 3 days agomessage-square1fedilinkfile-text
minus-squareiii@mander.xyzlinkfedilinkEnglisharrow-up1·3 days agoThat’s probably the easiest to read attention and GPT implementation I’ve seen. Congrats to the author.
