Sampling parameters

Claude Text Watermarking Doesn’t Always Work: 3 Sampling Parameters and 5 Types of API Configurations Developers Should Check
ByFourteen 2026年 11月 6日

Author’s note: Anthropic has updated its official documentation to warn that Sonnet 4.6 may not reliably carry text watermarks when non-default sampling parameters are used. This article explains when Claude text watermarking applies and provides an API configuration checklist for developers. Not all Claude requests produce output with a reliable text watermark. In API documentation…

Read More Claude Text Watermarking Doesn’t Always Work: 3 Sampling Parameters and 5 Types of API Configurations Developers Should Check
ChatGPT and OpenAI

A Score Surge Doesn’t Necessarily Mean Greater Capability: Truncation and Repetition Pitfalls in Reasoning Model Evaluation, as Seen in Astra’s Modification of Qwen Weights, Plus a 5-Point Checklist
ByFourteen 2026年 11月 6日

Author’s note: Using a retrospective on Astra’s experiment rewriting Qwen3.5-0.8B’s weights, this article explains the difference between getting an answer wrong and failing to finish it when evaluating reasoning models—and how to break down results by truncation. On November 2, four MIT CSAIL researchers published a blog post titled “Sparks of Weight-Space Programming.” They reported…

Read More A Score Surge Doesn’t Necessarily Mean Greater Capability: Truncation and Repetition Pitfalls in Reasoning Model Evaluation, as Seen in Astra’s Modification of Qwen Weights, Plus a 5-Point Checklist