wuyang/llm-atlas
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: add paired DeepSeek routing length control

Browse Source
This commit is contained in:
wuyang
2026-07-29 16:17:18 +08:00
parent 059458f8e3
commit 9ca08501a8
15 changed files with 318874 additions and 69 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/check-deepseek-browser.mjs
+31 -4
View File
@@ -305,21 +305,31 @@ const artifactCorpus = await evaluate(`(() => {
const read = () => ({
panel: root.querySelector("[data-artifact-panel]:not([hidden])").dataset.artifactPanel,
rows: root.querySelectorAll("[data-corpus-domain-rows] > div").length,
tokens: [...root.querySelectorAll("[data-corpus-domain-rows] > div > span:first-child > small")].map((node) => node.textContent.trim()),
heatRows: root.querySelectorAll("[data-corpus-heatmap] > div").length,
heatCells: root.querySelectorAll("[data-corpus-heatmap] > div > span").length,
jsdCells: root.querySelectorAll("[data-corpus-jsd] > *").length,
highest: root.querySelector("[data-corpus-highest-cv]").textContent.trim(),
largest: root.querySelector("[data-corpus-largest-jsd]").textContent.trim(),
heatTitle: root.querySelector("[data-corpus-heat-title]").textContent.trim(),
cohortTitle: root.querySelector("[data-corpus-cohort-title]").textContent.trim(),
exact: root.querySelector(".corpus-ledger .exact b").textContent.trim(),
modeNote: root.querySelector("[data-corpus-mode-note]").textContent.trim(),
deltaCards: root.querySelectorAll("[data-length-delta-grid] > article").length,
lengthJsd: root.querySelector("[data-length-jsd]").textContent.trim(),
lengthLargest: root.querySelector("[data-length-largest]").textContent.trim(),
});
const layer1 = read();
root.querySelector('[data-corpus-layer="4"]').click();
const layer4 = read();
root.querySelector('[data-corpus-cohort="matched16"]').click();
root.querySelector('[data-corpus-layer="2"]').click();
const matched16 = read();
root.querySelector('[data-corpus-cohort="matched24"]').click();
const matched24 = read();
root.querySelector('[data-corpus-mode="token_weighted"]').click();
const tokenWeighted = read();
return { layer1, layer4, tokenWeighted };
return { layer1, layer4, matched16, matched24, tokenWeighted };
})()`);
await evaluate(`(() => {
document.querySelector("[data-dsv2-lab]").scrollIntoView({ block: "start", behavior: "instant" });
@@ -327,6 +337,12 @@ await evaluate(`(() => {
})()`);
await pause(180);
await screenshot("/tmp/llm-atlas-deepseek-corpus-desktop.png");
await evaluate(`(() => {
document.querySelector(".length-sensitivity").scrollIntoView({ block: "start", behavior: "instant" });
window.scrollBy(0, -82);
})()`);
await pause(120);
await screenshot("/tmp/llm-atlas-deepseek-length-sensitivity-desktop.png");
const artifactEvidence = await evaluate(`(() => {
const root = document.querySelector("[data-dsv2-lab]");
@@ -399,6 +415,8 @@ const mobile = await evaluate(`(() => {
tabs: root.querySelectorAll("[data-ds-tab]").length,
artifactTabs: artifact.querySelectorAll("[data-artifact-tab]").length,
artifactHeatCells: artifact.querySelectorAll("[data-route-heatmap] > span").length,
corpusCohorts: artifact.querySelectorAll("[data-corpus-cohort]").length,
lengthDeltaCards: artifact.querySelectorAll("[data-length-delta-grid] > article").length,
offenders: [...document.querySelectorAll("body *")]
.filter((node) => !node.closest(".paper-chain, .advantage-table, .precision-table, .mapping-table, [data-deepseek-lab], [data-dsv2-lab]"))
.filter((node) => node.getBoundingClientRect().right > document.documentElement.clientWidth + 1)
@@ -420,6 +438,12 @@ await evaluate(`(() => {
})()`);
await pause(180);
await screenshot("/tmp/llm-atlas-deepseek-corpus-mobile.png");
await evaluate(`(() => {
document.querySelector(".length-sensitivity").scrollIntoView({ block: "start", behavior: "instant" });
window.scrollBy(0, -70);
})()`);
await pause(120);
await screenshot("/tmp/llm-atlas-deepseek-length-sensitivity-mobile.png");
const report = { overview, capacity, cache, codesign, rl, artifactRoute, artifactLoad, artifactCache, artifactAbsorb, artifactCorpus, artifactEvidence, home, papers, mobile, exceptions };
console.log(JSON.stringify(report, null, 2));
@@ -459,15 +483,18 @@ if (artifactAbsorb.panel !== "absorb" || artifactAbsorb.algebra !== 2 || artifac
if (artifactAbsorb.metrics[0] !== "8.8889×" || artifactAbsorb.metrics[1] !== "0.00390625" || artifactAbsorb.metrics[2] !== "1.19e-7" || artifactAbsorb.metrics[3] !== "BYTE-EXACT") failures.push("absorb 数值正确性或复跑闸门异常");
if (artifactAbsorb.precisionRows !== 3 || artifactAbsorb.matrixRows !== 5 || artifactAbsorb.localUnsupported !== 4 || artifactAbsorb.executionCards !== 3 || !artifactAbsorb.boundary.includes("不是 FlashMLA 性能")) failures.push("FlashMLA SM120 边界结构异常");
if (artifactCorpus.layer1.panel !== "corpus" || artifactCorpus.layer1.rows !== 4 || artifactCorpus.layer1.heatRows !== 4 || artifactCorpus.layer1.heatCells !== 256 || artifactCorpus.layer1.jsdCells !== 25) failures.push("128 样本路由区间结构异常");
if (!artifactCorpus.layer1.highest.includes("中文新闻 · 0.754") || !artifactCorpus.layer1.largest.includes("中文新闻 ↔ Python 代码 · 0.059") || artifactCorpus.layer1.exact !== "BYTE-EXACT") failures.push("Layer 1 多域统计或复跑闸门异常");
if (!artifactCorpus.layer1.highest.includes("中文新闻 · 0.754") || !artifactCorpus.layer1.largest.includes("中文新闻 ↔ Python 代码 · 0.059") || artifactCorpus.layer1.exact !== "3 / 3 EXACT") failures.push("Layer 1 多域统计或三 cohort 复跑闸门异常");
if (!artifactCorpus.layer4.highest.includes("中文新闻 · 0.910") || !artifactCorpus.layer4.largest.includes("中文新闻 ↔ Python 代码 · 0.150") || !artifactCorpus.layer4.heatTitle.includes("layer 4")) failures.push("Layer 4 多域路由切换异常");
if (!artifactCorpus.tokenWeighted.heatTitle.includes("按 token 加权") || !artifactCorpus.tokenWeighted.modeNote.includes("长 prompt 权重更高")) failures.push("公开语料聚合口径切换异常");
if (!artifactCorpus.matched16.highest.includes("Python 代码 · 0.969") || artifactCorpus.matched16.cohortTitle !== "同样本 · 16 tokens" || !artifactCorpus.matched16.tokens.every((value) => value.includes("512 tokens"))) failures.push("16-token 同源 cohort 切换异常");
if (!artifactCorpus.matched24.highest.includes("Python 代码 · 0.718") || artifactCorpus.matched24.cohortTitle !== "同样本 · 24 tokens" || !artifactCorpus.matched24.tokens.every((value) => value.includes("768 tokens"))) failures.push("24-token 同源 cohort 切换异常");
if (artifactCorpus.matched24.deltaCards !== 4 || artifactCorpus.matched24.lengthLargest !== "Python 代码 · Δ -0.251" || artifactCorpus.matched24.lengthJsd !== "0.091 → 0.065 · Δ -0.026") failures.push("16→24 token 成对敏感性结论异常");
if (!artifactCorpus.tokenWeighted.heatTitle.includes("token 加权") || !artifactCorpus.tokenWeighted.modeNote.includes("理论上重合")) failures.push("等长 cohort 聚合口径切换异常");
if (artifactEvidence.panel !== "evidence" || artifactEvidence.layers !== 27 || artifactEvidence.executed !== 7 || artifactEvidence.split !== 1 || artifactEvidence.unloaded !== 19 || artifactEvidence.exact !== "31 / 31") failures.push("真实工件执行边界或复跑闸门异常");
if (!artifactEvidence.dependency.includes("Transformers 5.5") || !artifactEvidence.dependency.includes("4.41.2") || !artifactEvidence.boundary.includes("完整 27 层生成")) failures.push("依赖版本或未覆盖边界异常");
if (artifactEvidence.keyboardSelected !== "load" || artifactEvidence.keyboardVisible !== "load") failures.push("真实工件实验键盘 tab 导航异常");
if (home.releaseCards !== 17 || !home.firstRelease.includes("47 页不再压成摘要") || home.firstHref !== "/k3/" || home.paperCount !== "486") failures.push("首页 DeepSeek 首发入口或论文数异常");
if (papers.total !== 486 || !papers.hasFilter || papers.visible < 20 || !papers.hasCoder || !papers.hasEngram) failures.push("论文库 DeepSeek 聚光异常");
if (!mobile.menuVisible || mobile.menuOpen !== "true" || mobile.tabs !== 4 || mobile.artifactTabs !== 6 || mobile.artifactHeatCells !== 64) failures.push("移动端导航或实验异常");
if (!mobile.menuVisible || mobile.menuOpen !== "true" || mobile.tabs !== 4 || mobile.artifactTabs !== 6 || mobile.artifactHeatCells !== 64 || mobile.corpusCohorts !== 3 || mobile.lengthDeltaCards !== 4) failures.push("移动端导航或实验异常");
if (mobile.offenders.length) failures.push(`移动端越界元素：${JSON.stringify(mobile.offenders)}`);
if (exceptions.length) failures.push(`浏览器异常：${exceptions.join(" | ")}`);