Label PR

Workflow runs for Label PR

• feat: reuse a source across notebooks from a library

  completed successfully:
  Label PR#136:Pull request#217openedbymrsibe
  Oct 2, 202610s
  More actions for workflow run 136

• feat: move notes tools into studio-style content cards

  completed successfully:
  Label PR#135:Pull request#216openedbymrsibe
  Oct 2, 20269s
  More actions for workflow run 135

• chore(eval): integrate Retrieval Eval v2 and close Phase 1

  completed successfully:
  Label PR#134:Pull request#215openedbymrsibe
  Oct 1, 202610s
  More actions for workflow run 134

• fix: make interrupted indexing retryable and reset cancelled dialogs

  completed successfully:
  Label PR#133:Pull request#212openedbymrsibe
  Oct 1, 202614s
  More actions for workflow run 133

• fix: surface generation errors and guide initial model setup

  completed successfully:
  Label PR#132:Pull request#211openedbymrsibe
  Oct 1, 202613s
  More actions for workflow run 132

• fix(chat): preserve reasoning and simplify streaming indicators

  completed successfully:
  Label PR#131:Pull request#210editedbymrsibe
  Sep 30, 20268s
  More actions for workflow run 131

• fix(chat): preserve reasoning and simplify streaming indicators

  completed successfully:
  Label PR#130:Pull request#210openedbymrsibe
  Sep 30, 202611s
  More actions for workflow run 130

• fix(ui): restore workspace shortcuts and refine model settings

  completed successfully:
  Label PR#129:Pull request#209editedbymrsibe
  Sep 30, 20269s
  More actions for workflow run 129

• fix(ui): restore workspace shortcuts and refine model settings

  completed successfully:
  Label PR#128:Pull request#209openedbymrsibe
  Sep 30, 202611s
  More actions for workflow run 128

• feat(eval): separate candidateK from contextK, and run the production config

  completed successfully:
  Label PR#127:Pull request#194editedbymrsibe
  Sep 30, 202611s
  More actions for workflow run 127

• feat(eval): fifteen near-miss unanswerable questions — closes Phase 1

  completed successfully:
  Label PR#126:Pull request#208openedbymrsibe
  Sep 30, 202611s
  More actions for workflow run 126

• feat(eval): per-question paired deltas, and hybrid advantage is five questions

  completed successfully:
  Label PR#125:Pull request#207openedbymrsibe
  Sep 30, 202612s
  More actions for workflow run 125

• feat(eval): dense score diagnostics, and the answer is not a threshold

  completed successfully:
  Label PR#124:Pull request#206openedbymrsibe
  Sep 30, 20267s
  More actions for workflow run 124

• fix(eval): select the strategy on validation, and stop calling candidates passages

  completed successfully:
  Label PR#123:Pull request#204editedbymrsibe
  Sep 30, 20268s
  More actions for workflow run 123

• feat(eval): confusion clusters, and the corpus reaches 53 chunks

  completed successfully:
  Label PR#122:Pull request#205openedbymrsibe
  Sep 30, 202613s
  More actions for workflow run 122

• fix(eval): select the strategy on validation, and stop calling candidates passages

  completed successfully:
  Label PR#121:Pull request#204openedbymrsibe
  Sep 30, 20269s
  More actions for workflow run 121

• fix(retrieval): trace the dense channel threshold for hybrid too

  completed successfully:
  Label PR#120:Pull request#200editedbymrsibe
  Sep 30, 202610s
  More actions for workflow run 120

• feat(eval): nine cross-lingual questions, and the corpus stops being saturated

  completed successfully:
  Label PR#119:Pull request#203openedbymrsibe
  Sep 30, 20269s
  More actions for workflow run 119

• feat(eval): unanswerable questions, and an explicit split manifest

  completed successfully:
  Label PR#118:Pull request#202openedbymrsibe
  Sep 30, 20269s
  More actions for workflow run 118

• fix(eval): refuse a context window wider than the retrieval depth

  completed successfully:
  Label PR#117:Pull request#201openedbymrsibe
  Sep 30, 202610s
  More actions for workflow run 117

• fix(retrieval): trace the dense channel threshold for hybrid too

  completed successfully:
  Label PR#116:Pull request#200openedbymrsibe
  Sep 30, 20269s
  More actions for workflow run 116

• feat(eval): bounded parameter sweep with a trade-off dashboard

  completed successfully:
  Label PR#115:Pull request#199openedbymrsibe
  Sep 30, 20268s
  More actions for workflow run 115

• feat(eval): measure the context window, and name what still needs a model

  completed successfully:
  Label PR#114:Pull request#198openedbymrsibe
  Sep 30, 202612s
  More actions for workflow run 114

• feat(eval): adopt on the first metric with headroom, not a saturated one

  completed successfully:
  Label PR#113:Pull request#197openedbymrsibe
  Sep 30, 202610s
  More actions for workflow run 113

• feat(eval): derive the similarity threshold on validation, report on test

  completed successfully:
  Label PR#112:Pull request#196openedbymrsibe
  Sep 30, 202610s
  More actions for workflow run 112