Skip to main content

📝 Training Small R1-like Reasoning Models (Stephen Diehl). Fascinating.