Show HN: GPT-2 124M trained from scratch – 2.48B tokens, 2.8h on 4 A40shttps://github.com/upul/Build-GPT-2
Article URL: https://github.com/upul/Build-GPT-2 Comments URL: https://news.ycombinator.com/item?id=50104656 Points: 1 # Comments: 0
🔍 Search
Active Interests
Trends
Top Contributors
Hot Discussions
n/techShow HN: Bokal – Open-source cookie editor without the "tabs" permission▲ 1 • 💬 0n/techShow HN: Visual Scripting for Golang▲ 1 • 💬 0n/techShow HN: Give (low vRAM) local sLLMs codebase-wide context with SQLite (attempt)▲ 1 • 💬 0n/newsOpenAI Engineer, Who Is Charged With Planting Bug, Advised Saudis▲ 1 • 💬 0n/newsCapsem – a Rust hypervisor sandboxing AI agents in air-gapped Linux VMs▲ 1 • 💬 0