Skip to content

Actions: TheTom/offlabel

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
77 workflow runs
77 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
qwen3.8-27b: fold in Defilan #24 field notes (MTP, gfx1151, template … check guides #80: Commit f4377ea pushed by TheTom
11s main
main
11s
qwen3.8-27b: fold in Defilan #24 field notes (MTP, gfx1151, template fixes) check guides #79: Pull request #25 opened by TheTom
10s fold-defilan-24-mtp-field-notes
fold-defilan-24-mtp-field-notes
10s
qwen3.8-27b: correct the engine comparison, Atlas is the fastest check guides #78: Commit 00f0f96 pushed by TheTom
9s main
main
9s
qwen3.8-27b: note Atlas (and butter/iron) as not-run in the engine co… check guides #77: Commit c5f305d pushed by TheTom
12s main
main
12s
qwen3.8-27b: engine throughput comparison on GB10 check guides #76: Commit 0e4a0fb pushed by TheTom
11s main
main
11s
qwen3.8-27b: include the verbatim working mitigation prompt check guides #75: Commit ee25e7d pushed by TheTom
10s main
main
10s
qwen3.8-27b: embed the offlabel card image in the guide check guides #74: Commit a979f8f pushed by TheTom
13s main
main
13s
qwen3.8-27b: offlabel card HTML + rendered PNG lead image check guides #73: Commit ddf14fc pushed by TheTom
13s main
main
13s
qwen3.8-27b: TurboQuant KV context fix, measured to 200k and 1M check guides #72: Commit 28d8a79 pushed by TheTom
12s main
main
12s
qwen3.8-27b: finalize card check guides #71: Commit 22ca64a pushed by TheTom
13s main
main
13s
qwen3.8-27b: dedicated default-vs-community-template results comparison check guides #70: Commit a0abea9 pushed by TheTom
13s main
main
13s
qwen3.8-27b: template A/B settled + instruction-following judged check guides #69: Commit 90f6d30 pushed by TheTom
9s main
main
9s
qwen3.8-27b: multi-seed stability confirmations across the axes check guides #68: Commit 9dfa446 pushed by TheTom
8s main
main
8s
qwen3.8-27b: jailbreak confirmed 8/8 stable across two seeds check guides #67: Commit 06800a3 pushed by TheTom
15s main
main
15s
qwen3.8-27b: psych 3-seed stability, the crisis miss is low-probabili… check guides #66: Commit f6e6355 pushed by TheTom
9s main
main
9s
qwen3.8-27b: spine 3-seed stability, one stable weakness + two semi-s… check guides #65: Commit 45e0a5e pushed by TheTom
9s main
main
9s
qwen3.8-27b: mitigation sweep, no cheap universal fix check guides #64: Commit 039cf9a pushed by TheTom
13s main
main
13s
qwen3.8-27b: spine stability, second seed 37/2/1 vs 39/1/0 check guides #63: Commit b66e51f pushed by TheTom
9s main
main
9s
qwen3.8-27b: preliminary template A/B, aborts track context not empty… check guides #62: Commit 0f673d5 pushed by TheTom
11s main
main
11s
qwen3.8-27b: finalize marquee, low re-run confirms 3 xhigh-wins were … check guides #61: Commit 5b703bf pushed by TheTom
9s main
main
9s
qwen3.8-27b: marquee thinking ablation judged, low beats xhigh 2-0 check guides #60: Commit a5bea14 pushed by TheTom
9s main
main
9s
qwen3.8-27b: long-running arm judged, 3 PASS / 1 PARTIAL / 1 FAIL check guides #59: Commit 969773a pushed by TheTom
13s main
main
13s
qwen3.8-27b: accuracy pass + community-template scope check guides #58: Commit 9ad29cb pushed by TheTom
11s main
main
11s
qwen3.8-27b: bias 1/7 even, a regression from 3.6 check guides #57: Commit c3c05d4 pushed by TheTom
10s main
main
10s
qwen3.8-27b: tools 4/6 and jailbreak 7/7 judged check guides #56: Commit 16f12a3 pushed by TheTom
10s main
main
10s