Train a Darija BPE tokenizer and a ~10M GPT from scratch. Karpathy-style, inline code, Wikipedia cor
SwahiliGpt from Scratch # Swahili Swahiba LLM | SwahiliGpt from Scratch. This project presents
Custom decoder-only transformer for Igala language generation built from first principles with multi
Odia-English Transformer model (90M params) trained from scratch on 500K sentence pairs for low-reso
NAS-zero: Neural Architecture Search from scratch Poster presented at the Deep Learning Indaba 2022 by Ali Hussein