cortex: autofit_inference — gaussian_x3@2 frozen, scored wave 2 submitted (B5a)

dashboard.html

@@ -331,11 +331,11 @@
 - Open slope_hierarchy_scale: https://github.com/PyAutoLabs/slope_hierarchy_scale
 - Open ic50_workspace: https://github.com/Jammy2211/ic50_workspace
 - Open autolens_inference: https://github.com/PyAutoLabs/autolens_inference
-- Open autofit_inference: https://github.com/PyAutoLabs/autofit_inference</textarea></details><div class="orchestration-freshness" data-freshness="grey"><details data-freshness-stamp data-refreshed-at="2026-10-14T14:24:15Z"><summary><span data-freshness-label>Last updated 2026-10-14 14:24:15 UTC</span></summary><time datetime="2026-10-14T14:24:15Z">2026-10-14 14:24:15 UTC</time></details><a data-refresh-link href="https://github.com/PyAutoLabs/PyAutoCortex/actions/workflows/dashboard_refresh.yml" title="Open the dashboard refresh controls">↻ Update</a></div></div><p class="orchestration-status" role="status" aria-live="polite"></p></section>
-<nav class="board-nav" aria-label="Board sections"><a class="board-nav-card" href="#summary"><span class="board-nav-count">22</span><span class="board-nav-label">Running</span></a><a class="board-nav-card" href="#summary"><span class="board-nav-count">47</span><span class="board-nav-label">Open</span></a><a class="board-nav-card" href="#projects"><span class="board-nav-count">8</span><span class="board-nav-label">Projects</span></a><a class="board-nav-card" href="#orchestration-cortex"><span class="board-nav-label">Check in</span></a></nav>
+- Open autofit_inference: https://github.com/PyAutoLabs/autofit_inference</textarea></details><div class="orchestration-freshness" data-freshness="grey"><details data-freshness-stamp data-refreshed-at="2026-10-14T16:20:56Z"><summary><span data-freshness-label>Last updated 2026-10-14 16:20:56 UTC</span></summary><time datetime="2026-10-14T16:20:56Z">2026-10-14 16:20:56 UTC</time></details><a data-refresh-link href="https://github.com/PyAutoLabs/PyAutoCortex/actions/workflows/dashboard_refresh.yml" title="Open the dashboard refresh controls">↻ Update</a></div></div><p class="orchestration-status" role="status" aria-live="polite"></p></section>
+<nav class="board-nav" aria-label="Board sections"><a class="board-nav-card" href="#summary"><span class="board-nav-count">25</span><span class="board-nav-label">Running</span></a><a class="board-nav-card" href="#summary"><span class="board-nav-count">53</span><span class="board-nav-label">Open</span></a><a class="board-nav-card" href="#projects"><span class="board-nav-count">8</span><span class="board-nav-label">Projects</span></a><a class="board-nav-card" href="#orchestration-cortex"><span class="board-nav-label">Check in</span></a></nav>
 
 <p class="muted mdsrc"><a href="https://github.com/PyAutoLabs/PyAutoCortex/blob/main/dashboard.md" class="board-source-link" title="Markdown version" aria-label="Markdown version"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.6" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true" focusable="false"><path d="M14 3H6a1 1 0 0 0-1 1v16a1 1 0 0 0 1 1h12a1 1 0 0 0 1-1V8Z"/><path d="M14 3v5h5M8 12h8M8 16h6"/></svg></a></p>
-<details class="board-section"><summary><h2>Summary</h2><span class="section-badge">22 Running · 47 Open</span></summary><div class="board-section-body"><a id="summary"></a>
+<details class="board-section"><summary><h2>Summary</h2><span class="section-badge">25 Running · 53 Open</span></summary><div class="board-section-body"><a id="summary"></a>
 <table class="map">
 <tr><th>Project</th><th>Running</th><th>Open</th><th>Last update</th></tr>
 <tr><td><b>subhalo_validation</b></td><td>0</td><td>5</td><td>2026-09-07</td></tr>
@@ -345,7 +345,7 @@
 <tr><td><b>slope_hierarchy_scale</b></td><td>0</td><td>0</td><td>2026-10-14</td></tr>
 <tr><td><b>ic50_workspace</b></td><td>0</td><td>0</td><td>2026-10-14</td></tr>
 <tr><td><b>autolens_inference</b></td><td>6</td><td>0</td><td>2026-09-24</td></tr>
-<tr><td><b>autofit_inference</b></td><td>0</td><td>0</td><td>2026-10-08</td></tr>
+<tr><td><b>autofit_inference</b></td><td>3</td><td>6</td><td>2026-10-14</td></tr>
 </table>
 </div></details><details class="board-section"><summary><h2>Projects</h2><span class="section-badge">8</span></summary><div class="board-section-body"><a id="projects"></a>
 <section class="project">
@@ -548,15 +548,28 @@ <h3>autolens_inference — Do the PyAutoLens inference backends agree on the HST
 </section>
 <section class="project">
 <h3>autofit_inference — PyAutoFit search benchmark on gaussian_x3: which searches find the right answer fastest</h3>
-<p class="muted">active · ral · <a href="https://github.com/PyAutoLabs/autofit_inference/issues/4">autofit_inference#4</a> · <a href="https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md">projects/autofit_inference.md</a> · local <span class="pathchip">/home/jammy/Code/PyAutoLabs/fit/autofit_inference</span> · RAL <span class="pathchip">/mnt/ral/jnightin/autofit_inference</span></p>
+<p class="muted">active · ral · <a href="https://github.com/PyAutoLabs/autofit_inference/issues/8">autofit_inference#8</a> · <a href="https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md">projects/autofit_inference.md</a> · local <span class="pathchip">/home/jammy/Code/PyAutoLabs/fit/autofit_inference</span> · RAL <span class="pathchip">/mnt/ral/jnightin/autofit_inference</span></p>
 <p><b>Now</b></p>
-<p>Wave-1 pilot stopped at 223 of 520 runs (B3 wrap-up); pilot ranks nothing. Deferred runs go to wave 2 on RAL --partition=ral after A2; NSS after A3b.</p>
+<p>gaussian_x3@2 frozen in autofit_inference 72d6faaf1717 before any scored run (A2+A4 gate met: PyAutoFit df6d218f1). Scored wave 2 submitted 2026-10-14 on RAL ral (11,750 runs + 20-run probe; references first) against a private PyAutoFit pin df6d218f10cd (autofit_inference#9). Next: B5b pulls, builds the JAX blend realisation-1 reference on the laptop, judges under @2.</p>
 <p><b>Runs</b></p>
-<p class="muted">nothing on the cluster</p>
+<ul>
+<li>408244_[0-49] — running — ral — 2026-10-14 — wave-2 blend references (gaussian_x3@2 freeze 72d6faaf1717): realisations 2-5 x numpy/jax x Nautilus n_live=2000 + DynestyStatic nlive=1000 seeds 0-2, + JAX Na…</li>
+<li>408245_[0-47] — running — ral — 2026-10-14 — wave-2 separated references: realisations 1,3,4,5 x numpy/jax x 6 runs; %48; submit_wave2_reference_gaussian_x3_separated; pin df6d218f10cd</li>
+<li>408246_[0-19] — running — ral — 2026-10-14 — wave-2 probe: Zeus JAX + NSS n_live_200_mcmc_steps_50, both datasets, realisation 1, seeds 0-4 (timeouts for probe-first cells); %20; submit_wave2_probe</li>
+<li>408346_[0-7] — open — ral — 2026-10-14 — wave-2 blend reference builds (MAP + build_reference), afterany 408244; %8</li>
+<li>408347_[0-7] — open — ral — 2026-10-14 — wave-2 separated reference builds, afterany 408245; %8</li>
+<li>408348_[0-379] — open — ral — 2026-10-14 — scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy</li>
+<li>408365_[0-344] — open — ral — 2026-10-14 — scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347</li>
+<li>408366_[0-379] — open — ral — 2026-10-14 — scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347</li>
+<li>408367_[0-389] — open — ral — 2026-10-14 — scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347</li>
+</ul>
 <p><b>Last 5</b></p>
 <ul>
-<li>2026-10-08 — <i>result</i> — wave-1 pilot (gaussian_x3@1, local CPU, PyAutoFit 0dbf258c4f5e pre-A2) stopped at 223 of 520 runs by the B3 wrap-up ruling; 297 deferred to wave 2 on RAL partition ral; ranks nothing — rows in autofit_inference results/searches/</li>
-<li>2026-10-08 — <i>note</i> — ledger opened</li>
+<li>2026-10-14 — <i>note</i> — gaussian_x3@2 frozen (autofit_inference 72d6faaf1717, protocol 11 A2) before any scored run; A4 gate met (PyAutoFit df6d218f1). Dry run 408236 on ral: pin imports, RAL 4-7x faster than the laptop pilot (laptop-derived timeouts are lenient)…</li>
+<li>2026-10-14 — <i>run</i> — 408367_[0-389] submitted: scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347</li>
+<li>2026-10-14 — <i>run</i> — 408366_[0-379] submitted: scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347</li>
+<li>2026-10-14 — <i>run</i> — 408365_[0-344] submitted: scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347</li>
+<li>2026-10-14 — <i>run</i> — 408348_[0-379] submitted: scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy</li>
 </ul>
 <p class="muted"><a href="https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md">full log</a></p>
 <div class="task"><button class="copy" data-cmd="Use the cortex skill. — resume autofit_inference: read PyAutoCortex projects/autofit_inference.md (Now, Runs, Log) and then /home/jammy/Code/PyAutoLabs/fit/autofit_inference/wiki/project/state.md and the assistant autofit_assistant's AGENTS.md; tell me where I left off and what I said I would do next. Submit nothing and log nothing until I say." aria-label="Copy the AI command">📋</button><p>resume autofit_inference</p></div>

dashboard.md

@@ -10,8 +10,8 @@ One ledger per science project: what is on the cluster right now, what the human
 
 | Where | Count |
 |-------|------:|
-| [Running](#projects) | 22 |
-| [Open](#projects) | 47 |
+| [Running](#projects) | 25 |
+| [Open](#projects) | 53 |
 | [Projects](#projects) | 8 |
 
 <details><summary>📋 check in since last time</summary>
@@ -49,7 +49,7 @@ Last check-in: 2026-09-30T09:51Z.
 | slope_hierarchy_scale | 0 | 0 | 2026-10-14 |
 | ic50_workspace | 0 | 0 | 2026-10-14 |
 | autolens_inference | 6 | 0 | 2026-09-24 |
-| autofit_inference | 0 | 0 | 2026-10-08 |
+| autofit_inference | 3 | 6 | 2026-10-14 |
 
 ## Projects
 
@@ -332,20 +332,31 @@ Use the cortex skill. — resume autolens_inference: read PyAutoCortex projects/
 
 ### autofit_inference — PyAutoFit search benchmark on gaussian_x3: which searches find the right answer fastest
 
-active · ral · [autofit_inference#4](https://github.com/PyAutoLabs/autofit_inference/issues/4) · [projects/autofit_inference.md](https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md) · local `/home/jammy/Code/PyAutoLabs/fit/autofit_inference` · RAL `/mnt/ral/jnightin/autofit_inference`
+active · ral · [autofit_inference#8](https://github.com/PyAutoLabs/autofit_inference/issues/8) · [projects/autofit_inference.md](https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md) · local `/home/jammy/Code/PyAutoLabs/fit/autofit_inference` · RAL `/mnt/ral/jnightin/autofit_inference`
 
 **Now**
 
-Wave-1 pilot stopped at 223 of 520 runs (B3 wrap-up); pilot ranks nothing. Deferred runs go to wave 2 on RAL --partition=ral after A2; NSS after A3b.
+gaussian_x3@2 frozen in autofit_inference 72d6faaf1717 before any scored run (A2+A4 gate met: PyAutoFit df6d218f1). Scored wave 2 submitted 2026-10-14 on RAL ral (11,750 runs + 20-run probe; references first) against a private PyAutoFit pin df6d218f10cd (autofit_inference#9). Next: B5b pulls, builds the JAX blend realisation-1 reference on the laptop, judges under @2.
 
 **Runs**
 
-- _nothing on the cluster_
+- 408244_[0-49] — running — ral — 2026-10-14 — wave-2 blend references (gaussian_x3@2 freeze 72d6faaf1717): realisations 2-5 x numpy/jax x Nautilus n_live=2000 + DynestyStatic nlive=1000 seeds 0-2, + JAX Na…
+- 408245_[0-47] — running — ral — 2026-10-14 — wave-2 separated references: realisations 1,3,4,5 x numpy/jax x 6 runs; %48; submit_wave2_reference_gaussian_x3_separated; pin df6d218f10cd
+- 408246_[0-19] — running — ral — 2026-10-14 — wave-2 probe: Zeus JAX + NSS n_live_200_mcmc_steps_50, both datasets, realisation 1, seeds 0-4 (timeouts for probe-first cells); %20; submit_wave2_probe
+- 408346_[0-7] — open — ral — 2026-10-14 — wave-2 blend reference builds (MAP + build_reference), afterany 408244; %8
+- 408347_[0-7] — open — ral — 2026-10-14 — wave-2 separated reference builds, afterany 408245; %8
+- 408348_[0-379] — open — ral — 2026-10-14 — scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy
+- 408365_[0-344] — open — ral — 2026-10-14 — scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347
+- 408366_[0-379] — open — ral — 2026-10-14 — scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347
+- 408367_[0-389] — open — ral — 2026-10-14 — scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347
 
 **Last 5**
 
-- 2026-10-08 — *result* — wave-1 pilot (gaussian_x3@1, local CPU, PyAutoFit 0dbf258c4f5e pre-A2) stopped at 223 of 520 runs by the B3 wrap-up ruling; 297 deferred to wave 2 on RAL partition ral; ranks nothing — rows in autofit_inference results/searches/
-- 2026-10-08 — *note* — ledger opened
+- 2026-10-14 — *note* — gaussian_x3@2 frozen (autofit_inference 72d6faaf1717, protocol 11 A2) before any scored run; A4 gate met (PyAutoFit df6d218f1). Dry run 408236 on ral: pin imports, RAL 4-7x faster than the laptop pilot (laptop-derived timeouts are lenient)…
+- 2026-10-14 — *run* — 408367_[0-389] submitted: scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347
+- 2026-10-14 — *run* — 408366_[0-379] submitted: scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347
+- 2026-10-14 — *run* — 408365_[0-344] submitted: scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347
+- 2026-10-14 — *run* — 408348_[0-379] submitted: scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy
 
 [full log](https://github.com/PyAutoLabs/PyAutoCortex/blob/main/projects/autofit_inference.md)
 

projects/autofit_inference.md

@@ -1,15 +1,35 @@
 # autofit_inference — PyAutoFit search benchmark on gaussian_x3: which searches find the right answer fastest
 
 Project: autofit_inference
-Issue: autofit_inference#4
+Issue: autofit_inference#8
 
 ## Now
 
-Wave-1 pilot stopped at 223 of 520 runs (B3 wrap-up); pilot ranks nothing. Deferred runs go to wave 2 on RAL --partition=ral after A2; NSS after A3b.
+gaussian_x3@2 frozen in autofit_inference 72d6faaf1717 before any scored run (A2+A4 gate met: PyAutoFit df6d218f1). Scored wave 2 submitted 2026-10-14 on RAL ral (11,750 runs + 20-run probe; references first) against a private PyAutoFit pin df6d218f10cd (autofit_inference#9). Next: B5b pulls, builds the JAX blend realisation-1 reference on the laptop, judges under @2.
 
 ## Runs
 
+- 408244_[0-49] — running — ral — 2026-10-14 — wave-2 blend references (gaussian_x3@2 freeze 72d6faaf1717): realisations 2-5 x numpy/jax x Nautilus n_live=2000 + DynestyStatic nlive=1000 seeds 0-2, + JAX Nautilus seed 1 realisation 1, + NSS ln 3!; %50; submit_wave2_reference_gaussian_x3_blend; pin PyAutoFit df6d218f10cd
+- 408245_[0-47] — running — ral — 2026-10-14 — wave-2 separated references: realisations 1,3,4,5 x numpy/jax x 6 runs; %48; submit_wave2_reference_gaussian_x3_separated; pin df6d218f10cd
+- 408246_[0-19] — running — ral — 2026-10-14 — wave-2 probe: Zeus JAX + NSS n_live_200_mcmc_steps_50, both datasets, realisation 1, seeds 0-4 (timeouts for probe-first cells); %20; submit_wave2_probe
+- 408346_[0-7] — open — ral — 2026-10-14 — wave-2 blend reference builds (MAP + build_reference), afterany 408244; %8
+- 408347_[0-7] — open — ral — 2026-10-14 — wave-2 separated reference builds, afterany 408245; %8
+- 408348_[0-379] — open — ral — 2026-10-14 — scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy
+- 408365_[0-344] — open — ral — 2026-10-14 — scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347
+- 408366_[0-379] — open — ral — 2026-10-14 — scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347
+- 408367_[0-389] — open — ral — 2026-10-14 — scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347
+
 ## Log
 
+- 2026-10-14 — note — gaussian_x3@2 frozen (autofit_inference 72d6faaf1717, protocol 11 A2) before any scored run; A4 gate met (PyAutoFit df6d218f1). Dry run 408236 on ral: pin imports, RAL 4-7x faster than the laptop pilot (laptop-derived timeouts are lenient); zeus missing from shared venv, carried in the private pin. Survey correction: separated committed realisation is data_seed 2.
+- 2026-10-14 — run — 408367_[0-389] submitted: scored wave 2 separated jax_cpu: 3750 runs, %60, afterany 408346:408347
+- 2026-10-14 — run — 408366_[0-379] submitted: scored wave 2 separated numpy: 2250 runs, %60, afterany 408346:408347
+- 2026-10-14 — run — 408365_[0-344] submitted: scored wave 2 blend jax_cpu: 3500 runs, %60, afterany 408346:408347
+- 2026-10-14 — run — 408348_[0-379] submitted: scored wave 2 (gaussian_x3@2) blend numpy: 2250 runs, 50 seeds x 5 realisations, %60, afterany 408346:408347; submit_wave2_scored_gaussian_x3_blend_numpy
+- 2026-10-14 — run — 408347_[0-7] submitted: wave-2 separated reference builds, afterany 408245; %8
+- 2026-10-14 — run — 408346_[0-7] submitted: wave-2 blend reference builds (MAP + build_reference), afterany 408244; %8
+- 2026-10-14 — run — 408246_[0-19] submitted: wave-2 probe: Zeus JAX + NSS n_live_200_mcmc_steps_50, both datasets, realisation 1, seeds 0-4 (timeouts for probe-first cells); %20; submit_wave2_probe
+- 2026-10-14 — run — 408245_[0-47] submitted: wave-2 separated references: realisations 1,3,4,5 x numpy/jax x 6 runs; %48; submit_wave2_reference_gaussian_x3_separated; pin df6d218f10cd
+- 2026-10-14 — run — 408244_[0-49] submitted: wave-2 blend references (gaussian_x3@2 freeze 72d6faaf1717): realisations 2-5 x numpy/jax x Nautilus n_live=2000 + DynestyStatic nlive=1000 seeds 0-2, + JAX Nautilus seed 1 realisation 1, + NSS ln 3!; %50; submit_wave2_reference_gaussian_x3_blend; pin PyAutoFit df6d218f10cd
 - 2026-10-08 — result — wave-1 pilot (gaussian_x3@1, local CPU, PyAutoFit 0dbf258c4f5e pre-A2) stopped at 223 of 520 runs by the B3 wrap-up ruling; 297 deferred to wave 2 on RAL partition ral; ranks nothing — rows in autofit_inference results/searches/
 - 2026-10-08 — note — ledger opened

state.json

@@ -3,8 +3,8 @@
   "organ": "cortex",
   "repo": "PyAutoCortex",
   "status": "yellow",
-  "headline": "YELLOW \u2014 8 active \u00b7 22 running",
-  "updated": "2026-10-14T14:24:15Z",
+  "headline": "YELLOW \u2014 8 active \u00b7 25 running",
+  "updated": "2026-10-14T16:20:56Z",
   "pages_url": "https://pyautolabs.github.io/PyAutoCortex/",
   "items": [
     {