Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -1899,4 +1899,20 @@ tests/

- campaign: `continuous-loop-20260808-continuous-openui-schedu-02340fde-c2`
- checkpoints: `runs/c20260808-continuous-openui-schedu-02340fde-c2-component-plan/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-02340fde-c2-control/checkpoints/last.pt`
## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c1-bounds/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c1-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.

## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c2-component-plan/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c2-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.

## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c3-bounds/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c3-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.
16 changes: 16 additions & 0 deletions docs/MODEL_CARD.md
Original file line number Diff line number Diff line change
Expand Up @@ -2555,4 +2555,20 @@ Date (UTC) column). Do not delete history.

- campaign: `continuous-loop-20260808-continuous-openui-schedu-02340fde-c2`
- checkpoints: `runs/c20260808-continuous-openui-schedu-02340fde-c2-component-plan/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-02340fde-c2-control/checkpoints/last.pt`
## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c1-bounds/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c1-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.

## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c2-component-plan/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c2-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.

## Continuous autotrain note (2026-08-08, continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3)

- campaign: `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3`
- checkpoints: `runs/c20260808-continuous-openui-schedu-a8e59439-c3-bounds/checkpoints/last.pt`, `runs/c20260808-continuous-openui-schedu-a8e59439-c3-control/checkpoints/last.pt`
- honesty: fixture/scratch continuous cycle — **not** a ship promotion.
Original file line number Diff line number Diff line change
@@ -0,0 +1,45 @@
{
"auto": true,
"campaign_id": "continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1",
"candidate_metrics": {
"binder_reference_f1": 0.6333333333333333,
"latency_ms_p50": 5082.96,
"meaningful_program_rate": 0.0,
"parse_rate": 1.0,
"smoke.binder_reference_f1": 0.6333333333333333,
"smoke.latency_ms_p50": 5082.96,
"smoke.meaningful_program_rate": 0.0,
"smoke.parse_rate": 1.0,
"smoke.structural_similarity": 0.057499999999999996,
"structural_similarity": 0.057499999999999996
},
"control_metrics": {
"binder_reference_f1": 0.6333333333333333,
"latency_ms_p50": 5605.15,
"meaningful_program_rate": 0.0,
"parse_rate": 1.0,
"smoke.binder_reference_f1": 0.6333333333333333,
"smoke.latency_ms_p50": 5605.15,
"smoke.meaningful_program_rate": 0.0,
"smoke.parse_rate": 1.0,
"smoke.structural_similarity": 0.057499999999999996,
"structural_similarity": 0.057499999999999996
},
"cycle_index": 1,
"cycle_intent": "screening",
"cycle_role": "screening",
"evidence_class": "fixture",
"honesty": "fixture_screening_only_not_ship",
"loop_id": "continuous-openui-scheduled-r7m3xz",
"measurement_complete": true,
"positive": false,
"primary_metric": "smoke.structural_similarity",
"reasons": [
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c1-bounds",
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c1-control",
"primary_metric_null_or_worse:smoke.structural_similarity:control=0.057499999999999996 candidate=0.057499999999999996 improvement=0.0",
"fixture_insufficient_n_alone"
],
"schema": "continuous_cycle_results/v1",
"stack_layer": false
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Continuous cycle `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1`

- loop_id: `continuous-openui-scheduled-r7m3xz`
- cycle_index: `1`
- role/intent: `screening` / `screening`
- primary_metric: `smoke.structural_similarity`
- positive: **False**
- stack_layer: **False**
- measurement_complete: `True`
- evidence_class: `fixture`
- reasons: fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c1-bounds, fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c1-control, primary_metric_null_or_worse:smoke.structural_similarity:control=0.057499999999999996 candidate=0.057499999999999996 improvement=0.0, fixture_insufficient_n_alone
- control_metrics: `{'latency_ms_p50': 5605.15, 'parse_rate': 1.0, 'meaningful_program_rate': 0.0, 'structural_similarity': 0.057499999999999996, 'binder_reference_f1': 0.6333333333333333, 'smoke.latency_ms_p50': 5605.15, 'smoke.parse_rate': 1.0, 'smoke.meaningful_program_rate': 0.0, 'smoke.structural_similarity': 0.057499999999999996, 'smoke.binder_reference_f1': 0.6333333333333333}`
- candidate_metrics: `{'latency_ms_p50': 5082.96, 'parse_rate': 1.0, 'meaningful_program_rate': 0.0, 'structural_similarity': 0.057499999999999996, 'binder_reference_f1': 0.6333333333333333, 'smoke.latency_ms_p50': 5082.96, 'smoke.parse_rate': 1.0, 'smoke.meaningful_program_rate': 0.0, 'smoke.structural_similarity': 0.057499999999999996, 'smoke.binder_reference_f1': 0.6333333333333333}`

Auto-documented by the continuous driver self-heal closeout. Fixture screening only — not a ship claim.
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
{
"auto": true,
"campaign_id": "continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2",
"candidate_metrics": {
"binder_reference_f1": null,
"latency_ms_p50": null,
"meaningful_program_rate": null,
"parse_rate": null,
"structural_similarity": null
},
"control_metrics": {
"binder_reference_f1": null,
"latency_ms_p50": null,
"meaningful_program_rate": null,
"parse_rate": null,
"structural_similarity": null
},
"cycle_index": 2,
"cycle_intent": "screening",
"cycle_role": "screening",
"evidence_class": "fixture",
"honesty": "fixture_screening_only_not_ship",
"loop_id": "continuous-openui-scheduled-r7m3xz",
"measurement_complete": false,
"positive": false,
"primary_metric": "smoke.structural_similarity",
"reasons": [
"measurement_incomplete:c20260808-continuous-openui-schedu-a8e59439-c2-control:smoke:incomplete_document_n=3:decode_timeout_count=3",
"measurement_incomplete:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan:smoke:incomplete_document_n=3:decode_timeout_count=3",
"harness_failure:c20260808-continuous-openui-schedu-a8e59439-c2-control:experiment_failed",
"harness_failure:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan:experiment_failed",
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan",
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c2-control",
"primary_metric_unavailable",
"fixture_insufficient_n_alone"
],
"schema": "continuous_cycle_results/v1",
"stack_layer": false
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Continuous cycle `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2`

- loop_id: `continuous-openui-scheduled-r7m3xz`
- cycle_index: `2`
- role/intent: `screening` / `screening`
- primary_metric: `smoke.structural_similarity`
- positive: **False**
- stack_layer: **False**
- measurement_complete: `False`
- evidence_class: `fixture`
- reasons: measurement_incomplete:c20260808-continuous-openui-schedu-a8e59439-c2-control:smoke:incomplete_document_n=3:decode_timeout_count=3, measurement_incomplete:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan:smoke:incomplete_document_n=3:decode_timeout_count=3, harness_failure:c20260808-continuous-openui-schedu-a8e59439-c2-control:experiment_failed, harness_failure:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan:experiment_failed, fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c2-component-plan, fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c2-control, primary_metric_unavailable, fixture_insufficient_n_alone
- control_metrics: `{'latency_ms_p50': None, 'parse_rate': None, 'meaningful_program_rate': None, 'structural_similarity': None, 'binder_reference_f1': None}`
- candidate_metrics: `{'latency_ms_p50': None, 'parse_rate': None, 'meaningful_program_rate': None, 'structural_similarity': None, 'binder_reference_f1': None}`

Auto-documented by the continuous driver self-heal closeout. Fixture screening only — not a ship claim.
Original file line number Diff line number Diff line change
@@ -0,0 +1,47 @@
{
"auto": true,
"campaign_id": "continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3",
"candidate_metrics": {
"binder_reference_f1": 0.7333333333333334,
"latency_ms_p50": 10058.72,
"meaningful_program_rate": 0.3333333333333333,
"parse_rate": 1.0,
"smoke.binder_reference_f1": 0.7333333333333334,
"smoke.latency_ms_p50": 10058.72,
"smoke.meaningful_program_rate": 0.3333333333333333,
"smoke.parse_rate": 1.0,
"smoke.structural_similarity": 0.19083333333333333,
"structural_similarity": 0.19083333333333333
},
"control_metrics": {
"binder_reference_f1": 0.7333333333333334,
"latency_ms_p50": 11408.3,
"meaningful_program_rate": 0.3333333333333333,
"parse_rate": 1.0,
"smoke.binder_reference_f1": 0.7333333333333334,
"smoke.latency_ms_p50": 11408.3,
"smoke.meaningful_program_rate": 0.3333333333333333,
"smoke.parse_rate": 1.0,
"smoke.structural_similarity": 0.19083333333333333,
"structural_similarity": 0.19083333333333333
},
"cycle_index": 3,
"cycle_intent": "screening",
"cycle_role": "screening",
"evidence_class": "fixture",
"honesty": "fixture_screening_only_not_ship",
"loop_id": "continuous-openui-scheduled-r7m3xz",
"measurement_complete": true,
"positive": false,
"primary_metric": "smoke.structural_similarity",
"reasons": [
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c3-bounds",
"fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c3-control",
"efficiency_win:mpr_per_ms:2.9218493e-05->3.3138743e-05:gain_fraction=0.13417015:minimum=0.05",
"quality_held:parse=1.0 mpr=0.3333333333333333",
"primary_metric_null_or_worse:smoke.structural_similarity:control=0.19083333333333333 candidate=0.19083333333333333 improvement=0.0",
"fixture_insufficient_n_alone"
],
"schema": "continuous_cycle_results/v1",
"stack_layer": false
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Continuous cycle `continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3`

- loop_id: `continuous-openui-scheduled-r7m3xz`
- cycle_index: `3`
- role/intent: `screening` / `screening`
- primary_metric: `smoke.structural_similarity`
- positive: **False**
- stack_layer: **False**
- measurement_complete: `True`
- evidence_class: `fixture`
- reasons: fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c3-bounds, fixture_insufficient_n:c20260808-continuous-openui-schedu-a8e59439-c3-control, efficiency_win:mpr_per_ms:2.9218493e-05->3.3138743e-05:gain_fraction=0.13417015:minimum=0.05, quality_held:parse=1.0 mpr=0.3333333333333333, primary_metric_null_or_worse:smoke.structural_similarity:control=0.19083333333333333 candidate=0.19083333333333333 improvement=0.0, fixture_insufficient_n_alone
- control_metrics: `{'latency_ms_p50': 11408.3, 'parse_rate': 1.0, 'meaningful_program_rate': 0.3333333333333333, 'structural_similarity': 0.19083333333333333, 'binder_reference_f1': 0.7333333333333334, 'smoke.latency_ms_p50': 11408.3, 'smoke.parse_rate': 1.0, 'smoke.meaningful_program_rate': 0.3333333333333333, 'smoke.structural_similarity': 0.19083333333333333, 'smoke.binder_reference_f1': 0.7333333333333334}`
- candidate_metrics: `{'latency_ms_p50': 10058.72, 'parse_rate': 1.0, 'meaningful_program_rate': 0.3333333333333333, 'structural_similarity': 0.19083333333333333, 'binder_reference_f1': 0.7333333333333334, 'smoke.latency_ms_p50': 10058.72, 'smoke.parse_rate': 1.0, 'smoke.meaningful_program_rate': 0.3333333333333333, 'smoke.structural_similarity': 0.19083333333333333, 'smoke.binder_reference_f1': 0.7333333333333334}`

Auto-documented by the continuous driver self-heal closeout. Fixture screening only — not a ship claim.
15 changes: 15 additions & 0 deletions src/slm_training/resources/versions.json
Original file line number Diff line number Diff line change
Expand Up @@ -8825,6 +8825,21 @@
"date": "2026-08-08",
"note": "no-bump: record continuous-loop-20260808-continuous-openui-local-8c0b60dd-c8's control-only fixture screen (component-edge arm not executed, non-positive, checkpoint provenance) in MODEL_CARD and README; SLM-228 behavior is unchanged."
},
{
"version": "v3",
"date": "2026-08-08",
"note": "no-bump: record scheduled loop continuous-openui-scheduled-r7m3xz continuous-loop-20260808-continuous-openui-schedu-a8e59439-c3's checkpoint-note-only doc update (fixture/scratch continuous cycle honesty stub); behavior unchanged."
},
{
"version": "v3",
"date": "2026-08-08",
"note": "no-bump: record scheduled loop continuous-openui-scheduled-r7m3xz continuous-loop-20260808-continuous-openui-schedu-a8e59439-c2's checkpoint-note-only doc update (fixture/scratch continuous cycle honesty stub); behavior unchanged."
},
{
"version": "v3",
"date": "2026-08-08",
"note": "no-bump: record scheduled loop continuous-openui-scheduled-r7m3xz continuous-loop-20260808-continuous-openui-schedu-a8e59439-c1's checkpoint-note-only doc update (fixture/scratch continuous cycle honesty stub); behavior unchanged."
},
{
"version": "v3",
"date": "2026-08-08",
Expand Down