Qwen3.8 27B 用英文单词做大数加法的实验

AI 摘要

Simon Willison 在本地 DGX Spark 硬件上复现了 Colin Frasier 两年前用 GPT-4o 做过的实验，测试模型能否在计算大数加法后以英文单词形式输出答案。他使用 Qwen3.8-27B-Q4_K_M.gguf 模型，在关闭推理时每组合运行 30 次，开启推理时每组合仅运行一次，结果 169 次尝试中答对 167 次。报告还展示了部分大数计算的推理轨迹。

正文节选

4th October 2026 Colin Frasier posted on Bluesky about an experiment he ran over two years ago using GPT-4o to see how well it could "compute the sum but return the answer in words" across increasingly large numbers. Here's the chart he shared of those results: I'm confident GPT-4o didn't cheat and use a calculator, especially since it got so many of the calculations wrong, but I was inspired to run the experiment again on local hardware (a DGX Spark) to explore the effect in a fully controlled

发布时间：2026-10-05 07:34

抓取时间：2026-10-06 00:47

来源机构：Simon Willison