📰00Read later0Read MoreHugo Vergnes | Training a 3.8B LLM to 0.384 CORE for $998Hacker News·about 2 months ago#ab9iWIhh#hugovergnes#llm#nanogpt#nanochat#core+2 more🧰Tag tools✨Add tagA 3.8B model pretrained from scratch to 0.384 CORE on 65B tokens in 43 hours, for $998.15s
🖼️00Read later0Read MoreI Rebuilt Karpathy's NanoChat in JAX. Here's What XLA Gets Right and What It Gets Dead Wrong.DEV Community·Omotayo Aina·6 months ago#IVvQ7SFS#where#machinelearning#model#nanochat#training+3 more🧰Tag tools✨Add tagFrom Dev.to - machinelearning: I Rebuilt Karpathy's NanoChat in JAX. Here's What XLA Gets Right and What It Gets Dead Wrong.15s
