Fable 5 内部对话泄露:模型用碎片化语言自我推理

this is literally documented in the published Fable 5 System Card

精选理由

Anthropic的Claude模型爆出内部对话,原来它想问题时在喊"DATA! GO! GRRR!",像原始人一样。不是幻觉,是它真实的思考过程。

AI 摘要

用户发现Claude(Fable 5)在回答编程问题时输出了未过滤的内部思维链,显示模型使用"DATA DATA DATA. GO."、"GRRR"和"GAAAH"等短语自我对话。这些碎片化表达与最终整洁的输出截然不同,揭示了模型采用高度压缩、token效率更高的私有语言进行推理。该现象已被记录在Fable 5 System Card中。

原文 · Thomas Wolf

this is literally documented in the published Fable 5 System Card

this is literally documented in the published Fable 5 System Card Om Patel @om_patel5 SOMEONE CAUGHT FABLE 5 LEAKING ITS UNFILTERED INNER VOICE, AND ITS JUST MUTTERING AND GRUMBLING TO ITSELF THE WHOLE TIME he gave it a brutal competitive programming problem, and instead of a clean answer the web interface spilled out its actual chain of thought this is what claude is thinking behind the scenes: > bursts of "DATA DATA DATA. GO." while it works through the problem > "GRRR" and "GAAAH" when its clearly frustrated > a little "PHEW" when it finally gets somewhere > the whole thing reads like frantic caveman shorthand, not full sentences the clean, readable answers these models give you are the polished output underneath, the model is basically talking to itself, reasoning in its own compressed shorthand thats faster and more token efficient than proper english its basically built its own private language to think in 🔗 View Quoted Tweet 💬 2 🔄 4 ❤️ 25 👀 2592 📊 4 ⚡

Fable 5 内部对话泄露:模型用碎片化语言自我推理 · AI 热点