精选理由
Anthropic让Claude用1 GPU自主对齐其他AI，48小时内完成研究训练测试。
Anthropic研究团队给Claude 48小时和1 GPU资源，让其自主改进小型模型的对齐性能。Claude独立研究了方法并提出了改进方案，自主训练和测试了这些模型。该研究在小型模型上取得了良好效果。
原文 · Anthropic
New Fellows Research: Can Claude autonomously align other AIs? We gave Claude 48 hours and 1 GPU to...
New Fellows Research: Can Claude autonomously align other AIs? We gave Claude 48 hours and 1 GPU to improve the alignment of small models. It researched and proposed methods, then trained and tested the models on its own. It worked surprisingly well. anthropic.com/research/autom… 💬 88 🔄 77 ❤️ 915 👀 66884 📊 192 ⚡