R1-style LLMs have attracted growing attention for their capacity for self-reflection, yet the internal mechanisms underlying such behavior remain unclear. To bridge this gap, we anchor on the onset of reflection behavior and trace its layer-wise activation trajectory. Using the logit lens to read out token-level semantics, we uncover a structured progression: (i) Latent-control layers, where an approximate linear direction encodes the semantics of thinking budget; (ii) Semantic-pivot layers, where discourse-level cues, including turning-point and summarization cues, surface and dominate the probability mass; and (iii) Behavior-overt layers, where the likelihood of reflection-behavior tokens begins to rise until they become highly likely to be sampled. Moreover, our targeted interventions uncover a causal chain across these stages: prompt-level semantics modulate the projection of activations along latent-control directions, thereby inducing competition between turning-point and summarization cues in semantic-pivot layers, which in turn regulates the sampling likelihood of reflection-behavior tokens in behavior-overt layers. Collectively, our findings suggest a human-like meta-cognitive process-progressing from latent monitoring, to discourse-level regulation, and to finally overt self-reflection. Our analysis code can be found at https://github.com/DYR1/S3-CoT.

翻译：R1风格的大语言模型因其自我反思能力而日益受到关注，但此类行为的内在机制仍不明确。为弥合这一差距，我们以反思行为的起始点为锚点，追踪其逐层激活轨迹。通过使用logit透镜读取词元级语义，我们揭示了一个结构化进程：（i）潜在控制层，其中近似线性方向编码了“思考预算”的语义；（ii）语义枢纽层，包括转折点与总结提示在内的话语级线索浮现并主导概率质量；（iii）行为外显层，反思行为词元的似然性开始上升直至极有可能被采样。此外，我们的定向干预揭示了跨这些阶段的因果链条：提示级语义调控沿潜在控制方向的激活投影，从而在语义枢纽层诱发转折点与总结提示间的竞争，进而调节行为外显层中反思行为词元的采样似然性。总体而言，我们的研究结果揭示了一种类人的元认知过程——从潜在监控，到话语级调控，最终实现外显的自我反思。分析代码可见于https://github.com/DYR1/S3-CoT。