From 6c84dc91d7b21ec15e8fe56335c6875f793198ed Mon Sep 17 00:00:00 2001 From: Gahow Wang Date: Mon, 29 Jun 2026 02:34:12 +0800 Subject: [PATCH] Document hardened topology feedback --- .../prefill-scheduler-normalized-design-20260629.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/harness-ablation/prefill-scheduler-normalized-design-20260629.md b/docs/harness-ablation/prefill-scheduler-normalized-design-20260629.md index fba5b3f..ff52351 100644 --- a/docs/harness-ablation/prefill-scheduler-normalized-design-20260629.md +++ b/docs/harness-ablation/prefill-scheduler-normalized-design-20260629.md @@ -320,10 +320,12 @@ session = aituner-prefill-hardened-6b25d56 | 0004 | `gpu-memory-utilization=0.9` | 3.258 | GMU=0.9 是当前 best，达到已知 no-harness 水平 | | 0005 | GMU=0.9 + scheduler seed | 2.025 | GMU 与 scheduler seed 的组合被 falsify | | 0006 | `gpu-memory-utilization=0.92` | 3.258 | 与 GMU=0.9 打平，没有继续提升 | +| 0007 | `tensor-parallel-size=4, data-parallel-size=2` | 1.000 | DP/topology probe 被 falsify | candidate-set-0007 没有继续提出 `gpu-memory-utilization=0.94`，而是转向 `tensor-parallel-size=4, data-parallel-size=2` topology probe。这验证了 measurement-gated GMU climb：GMU=0.92 只是打平时，不再继续向更高 GMU 盲目爬升。 +candidate-set-0008 在 TP4/DP2 被 falsify 后继续测试 `tensor-parallel-size=8`。 当前最重要的机制结论：