Reference: Qwen2.5-1.5B (exact lineage). Probes: 199. Candidate pool: r1_orig, qwen3_32b, gemini, r1, gpt_oss_120b, llama33_70b.
Best reasoning-family mean alignment: -0.005 (gemini) vs gate +0.05. Not passed: no reasoning-teacher signal above threshold. Stop.
tier 1 gate not passed: no reasoning-teacher signal above threshold
Candidate pool: r1_orig, qwen3_32b, gemini, r1, gpt_oss_120b, llama33_70b; style classifier pool: r1_orig, qwen3_32b, gemini, r1, gpt_oss_120b, llama33_70b. A strong preference for a wrong near-relative is what a missing true teacher looks like; the pool must include the suspect's plausible teachers and their pipelines.