激活是真实存在的，且存在于读出层：循环深度推理器中的潜在组成、难度驱动的激活以及由界面构成的提交
The Ignition Is Real, and It Lives at the Readout: Latent composition, difficulty-clocked ignition, and the interface-constituted commit in a recurrent-depth reasoner
浏览论文内容
中文总结 AI 辅助
本研究验证潜在推理模型的“组成激活”为真实计算，复现30M参数循环深度推理器，发现激活存在于读出层，决策时边际跃升，隐藏状态方向突变后冻结，撤回早期速度低谷声明。
中文摘要 AI 辅助
我们测试在潜在推理模型中报告的“组成激活”是真实计算、仪器伪影还是从语言训练数据继承而来。我们从头开始复现了已发表的30M参数循环深度推理器的独立实现（采用相同的配方和随机种子），记录其发展过程，通过预注册的全签名门验证保真度，并同时在词汇读出和隐藏状态两个通道测量分辨率。激活是真实存在的，且存在于读出层：到达时间随问题深度呈规律性上升，分辨率清晰且保持稳定，且该签名在两条拥有相同随机种子但训练轨迹不同的实现中可复现。在提交决策时，决策边际在一次迭代中跃升5.8至8.0 logits，96%的案例中超过近阈值非事件步骤的第90百分位；此处带符号边际的零交叉是定义性的，不具有证据权重，因此证据为条件幅度。隐藏状态方向在原始几何中突然改变，满足预注册标准（在解码器的LayerNorm坐标中，其衰减略低于我们的阈值，因此复合解码器坐标声明未得到确认），随后在两个维度中冻结（在解码器坐标中为描述性冻结；8次迭代内角度步长从52.9度变为1.2度），而后续位移主要为径向（平方范数的0.961），且在测量范围内对读出无影响（径向logit效应≤5.7e-6）。早期的速度低谷声明被撤回：预注册的归一化控制显示其与坐标相关。中间状态从未通过绑定读出恢复（中继值为0.00）。所有标准均在获取数据前冻结；预测台账（包括本文撤回的标题）随配套仓库发布。
英文摘要
We test whether the "compositional ignition" reported in latent-reasoning models is real computation, an instrument artifact, or inherited from verbal training data. We grow an independent realization of a published 30M-parameter recurrent-depth reasoner from scratch (same recipe and seed), film its development, certify fidelity through a pre-registered whole-signature gate, and measure resolution in two channels at once: the vocabulary readout and the hidden state. The ignition is real and lives at the readout: arrival time rises lawfully with problem depth, resolution is sharp and holds, and the signature reproduces across two same-seed realizations with divergent training trajectories. At commitment the decision margin jumps 5.8-8.0 logits in one iteration, exceeding the 90th percentile of near-threshold non-event steps in 96% of cases; the signed margin's zero-crossing there is definitional and carries no evidential weight, so the evidence is that conditioned magnitude. The hidden-state direction snaps in raw geometry, meeting its pre-registered criterion (in the decoder's LayerNorm coordinates it attenuates just below our bar, so the composite decoder-coordinate claim is not confirmed), and then freezes in both (descriptively so in decoder coordinates; angular steps 52.9 to 1.2 degrees over eight iterations), while subsequent displacement is predominantly radial (0.961 of squared-norm) and readout-null to a measured bound (radial logit effect <=5.7e-6). An earlier velocity-trough claim is withdrawn: pre-registered normalization controls showed it coordinate-dependent. Intermediates were never recoverable through the tied readout (relay 0.00). All criteria were frozen before their data; the predictions ledger, including this paper's own withdrawn headline, ships in the companion repository.
