Publish merged SETA v2 recovery selection

seta_v2/README.md

@@ -2,8 +2,8 @@
 
 ## Dataset stats
 
-- **Tasks:** 4,567 native tasks; 1,414 omitted upstream, 46 excluded in v1. **3,107 v1 tasks → 2,150 kept**, all previously patched; **957 temporarily excluded** in v2.
-- **Images:** reuse **10 of v1's 11 cached Apptainer environments**. No additional images or task repairs in v2; original unique image count was not recorded.
+- **Tasks:** 4,567 original SETA tasks; 3,107 kept in v1. The current v2 release keeps 2,205 and archives 902. Archived tasks have no recorded model success in the available results; some runs are incomplete.
+- **Environments:** v2 reuses the cached Apptainer environments from v1.
 
 ## Dataset characteristics
 
@@ -20,30 +20,28 @@
 
 Intended benchmark transfer: **Terminal-Bench 2.0, OpenThoughts-TBLite 2.0, DS-1000 mini-200 and BixBench CLI corrections @ Pi**, through terminal workflows, Python data analysis and executable verification.
 
-## Partial-result selection
+## Merged partial-result selection
 
-Snapshot: **15 October 2026, 20:57 UTC**. Keep a task as soon as either model records a successful attempt, even if other attempts are unfinished. **811 kept tasks have incomplete results; 1,339 have all 16 graded attempts per model.**
+The published task list combines the frozen 15 October 2026, 20:57 UTC snapshot with the completed `2026-10-15-pass16-recovery` run. A task is kept when either model has at least one verifier reward of 1 across those results. The original snapshot remains recorded in `metadata/selection.json`; recovery results are stored separately so overlapping attempt counts are not added or double-counted.
 
 | Model | Tasks solved at least once |
 | --- | ---: |
-| Qwen3-30B-A3B-Instruct-2507 | 1,991 |
-| Qwen3-30B-A3B-Thinking | 1,356 |
-| Either model | **2,150** |
+| Qwen3-30B-A3B-Instruct-2507 | 2,137 |
+| Qwen3-30B-A3B-Thinking | 1,459 |
+| Either model | **2,205** |
 
-| Temporary exclusion reason | Tasks |
+| Archived from this release | Tasks |
 | --- | ---: |
+| Incomplete results, no success recorded | 509 |
 | Complete 16 attempts per model, no success | 364 |
-| Incomplete results, no success recorded yet | 564 |
-| No scored results yet | 29 |
-| Total | **957** |
+| No scored results | 29 |
+| Total | **902** |
 
-Missing results remain unknown, not failures. The 593 exclusions with missing/incomplete results can be reconsidered as runs finish. A lack of success does not establish a task defect.
+The recovery newly retained 55 tasks: 50 first successes for Instruct and 5 for Thinking. Another 194 existing tasks gained a first observed success for one model; model totals count each task once. The 509 tasks with incomplete results and no recorded success are archived from this release; each recorded failed trajectory receives curation reward 0. For attempts where the verifier produced no reward, the raw `verifier_reward` remains null and the error message is preserved. The 29 tasks with no scored results remain unscored.
 
-Build, reference, no-op and rule-based checks carry over from [v1](../seta_v1/README.md). Evaluated hashes match v1; kept payloads are unchanged.
+## Successful attempts in the original snapshot
 
-## Successful attempts among kept tasks
-
-These are observed success counts from up to 16 attempts per model, not complete pass@16 estimates. Zero for one model can include missing results; the other model solved the task.
+The table below records the frozen 15 October snapshot only (2,150 kept tasks). It is retained for historical comparison and is not a distribution for the merged 2,205-task selection. Recovery attempts and raw rewards are available in `runs/2026-10-15-pass16-recovery/`; `tasks.parquet` has separate `recovery_*` columns for each task.
 
 | Successful attempts | Instruct tasks | Thinking tasks |
 | --- | ---: | ---: |
@@ -67,13 +65,15 @@ These are observed success counts from up to 16 attempts per model, not complete
 
 ## Files and selecting tasks
 
-`tasks.parquet` contains kept payloads and each model's successes, graded attempts and missing attempts. `metadata/selection.json` records every v1 task's results, selection reason and the pinned v1 revision. Excluded payloads remain in v1; no duplicate archive Parquet is uploaded. `patch.py` reproduces this snapshot.
+`tasks.parquet` contains the 2,205 kept payloads, frozen-snapshot metrics, and separate per-task recovery attempt, success and missing-reward counts. `metadata/selection.json` records every v1 task's results, selection reason and the pinned v1 revision. Excluded payloads remain in v1; no duplicate archive Parquet is uploaded. `patch.py` reproduces the merged task list from the pinned source tasks and `metadata/selection.json`.
 
 ```python
 import pyarrow.parquet as pq
-# Observed exactly one Instruct success, including incomplete runs.
-subset = pq.read_table("tasks.parquet", filters=[("instruct_successes", "=", 1)])
-# Add ("instruct_graded_attempts", "=", 16) to require a complete Instruct run.
+# At least one Instruct success in either the frozen snapshot or recovery tranche.
+t = pq.read_table("tasks.parquet")
+subset = t.filter(
+    (t["instruct_successes"] + t["recovery_instruct_successes"]) > 0
+)
 pq.write_table(subset, "selected.parquet", compression="zstd")
 ```
 

seta_v2/patch.py

@@ -1,4 +1,4 @@
-"""Reproduce the frozen SETA v2 partial-result selection. Requires pyarrow and huggingface_hub."""
+"""Reproduce the merged SETA v2 partial-result selection. Requires pyarrow and huggingface_hub."""
 import argparse,copy,hashlib,json
 from pathlib import Path
 import pyarrow as pa
@@ -14,6 +14,9 @@ for prefix in ['instruct','thinking']:
  for suffix in ['successes','graded_attempts','missing_attempts']:
   schema=schema.append(pa.field(prefix+'_'+suffix,pa.int64()))
 schema=schema.append(pa.field('target_attempts_per_model',pa.int64()))
+for prefix in ['recovery_instruct','recovery_thinking']:
+ for suffix in ['attempts','successes','missing_rewards']:
+  schema=schema.append(pa.field(prefix+'_'+suffix,pa.int64()))
 for r in t.to_pylist():
  s=m['tasks'][r['path']];assert r['content_sha256']==s['content_sha256']
  if not s['kept']:continue
@@ -22,9 +25,16 @@ for r in t.to_pylist():
   assert v['graded_attempts']==sum(x is not None for x in v['rewards'])
   assert v['missing_attempts']==16-v['graded_attempts']
   for suffix,key in [('successes','successful_attempts'),('graded_attempts','graded_attempts'),('missing_attempts','missing_attempts')]:r[prefix+'_'+suffix]=v[key]
- assert r['instruct_successes'] or r['thinking_successes']
+ recovery=s.get('recovery',{})
+ for prefix in ['instruct','thinking']:
+  v=recovery.get(prefix,{})
+  r['recovery_'+prefix+'_attempts']=v.get('attempts',0)
+  r['recovery_'+prefix+'_successes']=v.get('successful_attempts',0)
+  r['recovery_'+prefix+'_missing_rewards']=v.get('missing_verifier_rewards',0)
+ assert (r['instruct_successes'] or r['thinking_successes'] or
+         r['recovery_instruct_successes'] or r['recovery_thinking_successes'])
  r['target_attempts_per_model']=16;rows.append(r)
-assert len(rows)==2150
+assert len(rows)==2205
 names={r['path'] for r in rows};images=json.loads(t.schema.metadata[b'ot.images.v1']);images['tasks']={k:v for k,v in images['tasks'].items() if k in names};assert set(images['tasks'])==names
 bundles={e['bundle'] for v in images['tasks'].values() for e in v['environments']};images['bundles']={k:v for k,v in images['bundles'].items() if k in bundles};images['missing']=[]
 meta={b'ot.images.v1':json.dumps(images).encode(),b'ot.release.v2':json.dumps({k:v for k,v in m.items() if k!='tasks'}).encode()}

seta_v2/runs/2026-10-15-pass16-recovery/README.md

@@ -0,0 +1,7 @@
+# SETA pass@16 recovery results
+
+This folder contains the completed remaining-attempt recovery tranche for Qwen3-30B Instruct and Thinking. The two per-attempt Parquets contain 22,263 scheduled trials across 1,392 task/model pairs, with the saved trajectory and raw verifier result. Verifier rewards are preserved as recorded, including null when no reward was produced.
+
+The missing-reward audit contains 764 final trial attempts with their original exception messages: 524 harness failures, 63 supported agent failures, 32 verifier timeouts and 145 unresolved errors. The agent failures include terminal exits and terminal kills.
+
+This is the recovery tranche, not an independent full pass@16 estimate. The original SETA v2 selection remains available as the frozen 2026-10-15 20:57 UTC baseline in `metadata/selection.json`. The published task list now merges this recovery evidence under the existing rule and retains 55 additional tasks. Per-task recovery counts are separate from baseline counts to prevent double-counting overlapping attempts. See `run.json` for report hashes, job IDs, counts and provenance.

seta_v2/runs/2026-10-15-pass16-recovery/missing-reward-audit/retention-policy.json

@@ -0,0 +1,16 @@
+{
+  "saved_at": "2026-10-16T11:31:47.738383+02:00",
+  "attempts": 764,
+  "failure_status_counts": {
+    "harness_failure": 524,
+    "agent_failure": 63,
+    "unresolved_error": 145,
+    "verifier_timeout": 32
+  },
+  "verifier_reward": null,
+  "policy": "Retain original error messages and null verifier rewards. Agent-caused terminal exits/kills remain agent failures. Preserve attribution confidence and unresolved cases. No imputed reward, no retries launched, source archives unchanged.",
+  "jsonl": "retained-errors.jsonl",
+  "csv": "retained-errors.csv",
+  "csv_null_encoding": "literal null in verifier_reward column",
+  "validation": "All 764 unique attempt IDs and original exception messages checked against raw audit."
+}

seta_v2/runs/2026-10-15-pass16-recovery/missing-reward-audit/summary.json

@@ -0,0 +1,63 @@
+{
+  "created_at": "2026-10-16T11:23:32.406877+02:00",
+  "scope": "764 missing-reward attempts from the 22,263 attempts in the Oct 15 SETA recovery arrays only; not an audit of older preserved attempts.",
+  "raw_audit_sha256": "1ab10fb12a041709622280374e6017fa4a68ff00b2e13ed996a5867b3dede037",
+  "attempts": 764,
+  "finished_with_exception": 764,
+  "classifications": {
+    "harness_failure": 524,
+    "agent_action_supported": 63,
+    "unresolved": 145,
+    "verifier_timeout": 32
+  },
+  "subtypes": {
+    "output_truncation_lost_protocol_marker": 514,
+    "agent_heredoc_delimiter_collision": 44,
+    "TmuxBatchProtocolError": 104,
+    "TmuxSessionEndedError": 7,
+    "shell_startup_output_before_protocol_marker": 5,
+    "verifier_budget_exhausted": 32,
+    "AddTestsDirError": 6,
+    "bridge_output_decode_error": 4,
+    "RuntimeError": 8,
+    "pre_agent_task_setup_failure": 1,
+    "TmuxCommandError": 20,
+    "agent_killed_tmux_server": 15,
+    "agent_exited_shell": 4
+  },
+  "exception_types": {
+    "TmuxBatchProtocolError": 623,
+    "TmuxSessionEndedError": 70,
+    "VerifierTimeoutError": 32,
+    "AddTestsDirError": 6,
+    "RuntimeError": 13,
+    "TmuxCommandError": 20
+  },
+  "bridge_exec_return_codes": {
+    "0": 519,
+    "None": 137,
+    "-15": 52,
+    "-1": 12,
+    "255": 29,
+    "-9": 7,
+    "-11": 7,
+    "-3": 1
+  },
+  "by_model": {
+    "qwen3-30b-instruct-2507": {
+      "harness_failure": 296,
+      "agent_action_supported": 49,
+      "unresolved": 80,
+      "verifier_timeout": 19
+    },
+    "qwen3-30b-thinking": {
+      "verifier_timeout": 13,
+      "harness_failure": 228,
+      "agent_action_supported": 14,
+      "unresolved": 65
+    }
+  },
+  "scoring": "The frozen SETA task_metrics already retains these attempts in the pass@k denominator and counts no successful reward for them. Raw reward remains null. A derived failure score of 0 must be labelled as policy-derived, not verifier-observed.",
+  "retry_policy": "No retries launched. Harness failures are candidates after fixing their cause; retries must retain original evidence and not be appended as extra chances to solve an existing pass@16 sample.",
+  "limits": "Causal attribution is not always identifiable. Agent-action evidence is supported by saved commands and diagnostic state, not a replay. Numeric bridge return codes are not trial exit codes or rewards."
+}

seta_v2/runs/2026-10-15-pass16-recovery/run.json

@@ -0,0 +1,930 @@
+{
+  "run": "2026-10-15-pass16-recovery",
+  "kind": "pass16_remaining_attempts_recovery",
+  "created_at": "2026-10-16T09:59:39.828731+00:00",
+  "dataset_folder": "seta_v2",
+  "scope": "Completed remaining attempts for the SETA pass@16 teacher run. These are recovery attempts only, not a standalone pass@16 estimate.",
+  "models": {
+    "qwen3-30b-instruct-2507": {
+      "attempts": 13175,
+      "tasks": 824,
+      "verifier_rewards": 12731,
+      "missing_verifier_rewards": 444,
+      "successful_attempts": 4665,
+      "reward_counts": {
+        "0.0": 8066,
+        "1.0": 4665,
+        "None": 444
+      },
+      "trajectory_missing": 1,
+      "trace_file": "trajectories/qwen3-30b-instruct-2507.parquet",
+      "reports": [
+        {
+          "array_index": 0,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/000/submission/report/stage-6-514cb3b630cd.json",
+          "report_sha256": "026a6529be69cfe1eb2142b7a3cf09766b3fbc729b42944565f17f667c23ab3e",
+          "tasks": 9,
+          "attempts": 135
+        },
+        {
+          "array_index": 2,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/002/submission/report/stage-6-362e3ce8b020.json",
+          "report_sha256": "71671433e239999540b2879638732a979993cd4ba803e03cfdbb037d93b38c7f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 3,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/003/submission/report/stage-6-e0b9bbf09ceb.json",
+          "report_sha256": "a69b43c3e34ba76582220258b7b450631a1a71d8a5e680ee50875867dde1cfca",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 4,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/004/submission/report/stage-6-94fde44dbc91.json",
+          "report_sha256": "0ece2a78eaf3d359bdb956d7fbe531dfd5bd82c797c7325cf4459160ed57bf31",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 5,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/005/submission/report/stage-6-51b92ace578f.json",
+          "report_sha256": "f2884e3fbb6c9c9c39e2c5d996dff66f00fca747fcd3f2b3ab467d4d3e0c52c9",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 6,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/006/submission/report/stage-6-873521be2742.json",
+          "report_sha256": "214baaf37516df99ced3b3e2972d9faf6828748e4974ff43d2c65ecc7109dc92",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 7,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/007/submission/report/stage-6-98a7c7a2d56e.json",
+          "report_sha256": "4a4d3553198feb5ca8d5241b6e72403b886e634c4ba2cdc2567170a80cf78940",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 8,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/008/submission/report/stage-6-58cda3c69bea.json",
+          "report_sha256": "3edc22dad04f839f5bf9a8f0634559c7f56c0429cacc267298ee5e2dc6625b37",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 9,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/009/submission/report/stage-6-6918f891d5fe.json",
+          "report_sha256": "0033510c2e53587745ade1620855664e8206c33816ff6f93793377d8f4fdcfce",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 10,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/010/submission/report/stage-6-d2309fe4da1b.json",
+          "report_sha256": "b4e694ce74026807247ce2340b048e0da2833f26f74ceb360167dbd49a288b76",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 11,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/011/submission/report/stage-6-e0414d0b2e19.json",
+          "report_sha256": "b10fd62f0c073c72390f9586f35d5664486761707ac2141d97cc3197868b2d44",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 12,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/012/submission/report/stage-6-f687f398dcc4.json",
+          "report_sha256": "12dd5d919e0b8deff42cc42c5bf40a4edfba60037c61cb96722dabd8e6f3686b",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 13,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/013/submission/report/stage-6-54a441cdbd84.json",
+          "report_sha256": "308294018f0de433a1d7d23a68ad3226021d5236a3ddd7f665ad8d9329edd48b",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 14,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/014/submission/report/stage-6-59b0388415eb.json",
+          "report_sha256": "68d36d97ae4c650861b2a8885d424ec341afbbe6865088ac6e74311be43cb2b7",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 15,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/015/submission/report/stage-6-d02aac862bf3.json",
+          "report_sha256": "f7971cb513eed7bbf51c660fa243f972037714ce2198ddc293d21bb3b06b5475",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 16,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/016/submission/report/stage-6-f243c4d32988.json",
+          "report_sha256": "dbdde379e254109d6030ba8a8a3d61393e0a0e4970d6d6a59a989167fa1da5a5",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 17,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/017/submission/report/stage-6-e3f1a9a2b043.json",
+          "report_sha256": "7384814b3b952b388c20a25071946d4b3d39f6495f4ca04c0f909a457c044525",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 18,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/018/submission/report/stage-6-bced986b2434.json",
+          "report_sha256": "50f7796aa202a02dc5787c95bf572ac796bd24199b60d157fd0e2ab63c16fd6d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 19,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/019/submission/report/stage-6-238500475591.json",
+          "report_sha256": "658ca2165e7f7d650447ddf01dcebb9e3f72389a8c7a7f363ad923174efd56d3",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 20,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/020/submission/report/stage-6-9fce70c5419e.json",
+          "report_sha256": "b0a7c7067709d161832b2a4d873374d3e7c8f2c03bd60b7de2648a6cfc39d67c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 21,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/021/submission/report/stage-6-c734b2346718.json",
+          "report_sha256": "9996d08403d851685e97dbcf12c99af11362232683f9c4d538e5919e0b03e49e",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 22,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/022/submission/report/stage-6-adb443a3e48f.json",
+          "report_sha256": "15c26d5530889ada2fde960a8fae48d86ad7ac42a4a2079c2f37a00c9e27ae4f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 23,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/023/submission/report/stage-6-fe0bc1871f10.json",
+          "report_sha256": "27280ac5e576987aa9bf6b9ddfe282e90db6f1ce5b3652e161a719d602d38f28",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 24,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/024/submission/report/stage-6-fcd73b401223.json",
+          "report_sha256": "e6aac36829c3527ed0dd6d03d904a20b834bf5cba3a403be1a2b3bff26e28ec9",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 25,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/025/submission/report/stage-6-9328816f6151.json",
+          "report_sha256": "a4b95b8587ac4d750ab95eba75983f4d02874c03f014882815d03e66d3a1d618",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 26,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/026/submission/report/stage-6-e4bd19ecbf00.json",
+          "report_sha256": "6ad7760aecf08cd71aaf3e63b106eb3c45f5d2f59278e814e15160a0e9a6e26f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 27,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/027/submission/report/stage-6-0aa679ba8f30.json",
+          "report_sha256": "2e48678e8057397cdf940062e7966bead999cfb834d4601f792deaab116ddc0c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 28,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/028/submission/report/stage-6-fb424b8adf42.json",
+          "report_sha256": "e1e644f0a9a1f0fb7c8d999efc1f8742a3c7a0ba54ec67f1b71484e44949432e",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 29,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/029/submission/report/stage-6-94fdc26b5eb2.json",
+          "report_sha256": "387734404e7d54d54770bd76da77d8831f7294caf7db7d12d30be36de40357ad",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 30,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/030/submission/report/stage-6-5fbc1d9301b7.json",
+          "report_sha256": "825cd09e38e896858af2da00357be6a961745719fa6f2a3e4d9e4cb4218d8731",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 31,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/031/submission/report/stage-6-7139e1c8b0d4.json",
+          "report_sha256": "7bb06cb0f7c7b8a5db630b56c07fe650b872798fc26c8af980c6e8f0f37ee803",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 32,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/032/submission/report/stage-6-ad5e99413346.json",
+          "report_sha256": "c595aef6c68ef44f85053b7f40ae0db9a5ff8864363edb3eb45676d95bc9da04",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 33,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/033/submission/report/stage-6-b4f5223d2746.json",
+          "report_sha256": "d1a0e9ac7c60455fd6560d10f56d6cf74470b5d5136992db37ad98f4f7d9b1ce",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 34,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/034/submission/report/stage-6-a0419a2cb583.json",
+          "report_sha256": "ab4bc8c3e8ced973564341c00605e4d1660c6fdbb18e3a3b6e6df5f17d0ad55f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 35,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/035/submission/report/stage-6-77e847bcad6b.json",
+          "report_sha256": "f034bbf7b35d2d3d0d622b70ad805ad5dd22dba5e2ec9d6cd20ea18011823ac6",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 36,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/036/submission/report/stage-6-daa12419b4d0.json",
+          "report_sha256": "db2bb4ff0f6181991cfc0c7b60b2788a236141ef04ede5ac8d9d62a5a1d5251e",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 37,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/037/submission/report/stage-6-1aa038a087fa.json",
+          "report_sha256": "1c5c23be57ef933ddac28f6576aebd97e24c95ccc060452cd4e001ef64543ad6",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 38,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/038/submission/report/stage-6-800ef0397ad7.json",
+          "report_sha256": "93e1d17503e8fbb40413a12cb9f9e46a05070b01957752d3ca1e056065411379",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 39,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/039/submission/report/stage-6-14d0d59220a4.json",
+          "report_sha256": "a4bf6c5b546f6467a4e6a5a45aeedd2352e33c6e55c2d9b0d60252cdbf7df435",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 40,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/040/submission/report/stage-6-8fb42e588a24.json",
+          "report_sha256": "9a18529ddb8b3514cf73328b5fd732810b52e7ee6302f10f7245a92d388fd32f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 41,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/041/submission/report/stage-6-baeda5588db5.json",
+          "report_sha256": "b771f91334af5e2bde5fa94ed70f5731a47fb28d3173ee98c5c08215d92de9f2",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 42,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/042/submission/report/stage-6-3a5c0fb2eb5f.json",
+          "report_sha256": "5b5b9e06c43c779339c7f8d0f2fb1e20902a515aa818544cdfc74e27c6f48f88",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 43,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/043/submission/report/stage-6-f74429be8445.json",
+          "report_sha256": "592d79d6a825c4a26ae04e5d874f56f998d6932a86ae15b0185e48502ea51e90",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 44,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/044/submission/report/stage-6-c73e0f7a7069.json",
+          "report_sha256": "cf483eef144d106678201b4aeae3e63e771bb3367fa4c3bbd3642c4c939a2524",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 45,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/045/submission/report/stage-6-e99438a0f4e5.json",
+          "report_sha256": "63ee4aa081c030fb7f6a28f22feec7a3b3b950a88fd7aeb4166d81fa7d49d0a0",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 46,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/046/submission/report/stage-6-9f3d37b6269e.json",
+          "report_sha256": "939f7034ae556f62002220d61a77d7c80a503fdb35e6ce73a059284b850c2f61",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 47,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/047/submission/report/stage-6-938fd9a9630e.json",
+          "report_sha256": "21a009ffb48a4d5d32053dd8c077a4e0297cade3db1cc6fda4ef6e28a72a5f02",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 48,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/048/submission/report/stage-6-5df220dc66e5.json",
+          "report_sha256": "2089716ca6493116731697c03a0917ea636d214e4b8a9d539d6dc7ac294ac15a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 49,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/049/submission/report/stage-6-458360477d2c.json",
+          "report_sha256": "4f75d29da330aedce3cd5af71198a0c04e33093f21893509d7a1aa9a01e1e5d1",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 50,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/050/submission/report/stage-6-8f6e556b36a6.json",
+          "report_sha256": "19eaf9ebb6039d5094d8afe95d8b804cbee3ce926e569c6ce60080cc0dda1f59",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 51,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/051/submission/report/stage-6-5e52e2a4282e.json",
+          "report_sha256": "060b31cabf393874fc400721d59f79ca8abc40896a8e12e5cd13e06834557dbb",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 52,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/052/submission/report/stage-6-db0ea27cd51d.json",
+          "report_sha256": "95634497436a77eacf8a9028762a24f2fd29f192f6e0a73ed09831202014b38b",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 53,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/053/submission/report/stage-6-761a2fd8fea4.json",
+          "report_sha256": "112986fb0f9acd8a040047703b793f89554e08e951569f5cce8476e7195b6ae5",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 54,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/054/submission/report/stage-6-d4c271451092.json",
+          "report_sha256": "2ed0d854a8f041363d0bdd9eda7f63ae8ba390854f79652a4d5610882c924fbe",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 55,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/055/submission/report/stage-6-84f944aa83d5.json",
+          "report_sha256": "f8cec47f2ceee91ece7cc77f2bc2f0030020175014735bee56024c3d88b7f898",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 56,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/056/submission/report/stage-6-ae67e6df67bc.json",
+          "report_sha256": "74b11f5e4c34cdbb8332083fff73cb1ca575f46aa3a105c16587e6f3db59e257",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 57,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/057/submission/report/stage-6-44e4e82b9ccf.json",
+          "report_sha256": "28f0fb6666c50c510978795b0d38c7cbe68902f00dbca0f0cc1ff25dbcb3822d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 58,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/058/submission/report/stage-6-f097bc4f1c76.json",
+          "report_sha256": "da05e76bf71b0b649269f7e6d367d316f4126ef25d40f39c9b9e9dac87bcba26",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 59,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/059/submission/report/stage-6-ba68a291ea59.json",
+          "report_sha256": "80d6f649d9070352eda6c310dbf37dcfffe6392a850ed4139c18fcf5659afc56",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 60,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/060/submission/report/stage-6-1df4d0f5c5d5.json",
+          "report_sha256": "620b805843106006f984a99a33e0709f65ddf7cea3922e86558d25979b2cc1d3",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 61,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/061/submission/report/stage-6-0f698793b046.json",
+          "report_sha256": "787166545a1c8b142de3751f60170ec87fde905789474dbd78ca506c3c23f549",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 62,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/062/submission/report/stage-6-d99b28df64c0.json",
+          "report_sha256": "5409bb04710725d92d245557e7bdb66ba3b8289da774bc8add77885e0ce69105",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 63,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/063/submission/report/stage-6-160cae6ac28f.json",
+          "report_sha256": "12aafcda70a1815930f56c3c3512a360e5d5c53b8667d6e60454d5def63af1a7",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 64,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/064/submission/report/stage-6-a355c3cdc6e5.json",
+          "report_sha256": "db510c552456460965002d2e06d3ba1849fb6c6b9ddaaa6db0e47733eb68eb6c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 65,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/065/submission/report/stage-6-b79a0ab4d3b9.json",
+          "report_sha256": "8e9525f66fb7a4610c7d6194210b6c1a52491ecf86d9292ab801506f7120354c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 66,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/066/submission/report/stage-6-68d4c88032d8.json",
+          "report_sha256": "d9eb7eb0a08030f69b9cb49fa137ba254f97d331ea0668043cddce1eab247e8d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 67,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/067/submission/report/stage-6-73db6d7639df.json",
+          "report_sha256": "153d3966f981eba47b8964b238600ba10e1585a412a1eaabac37e988070017d6",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 68,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/068/submission/report/stage-6-f3f8aaf0f308.json",
+          "report_sha256": "cd8f7e2c2458b7d8ff5f1bdccdc5d631e5872e1e912fe166d608e4d2c009a952",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 69,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-instruct-2507/069/submission/report/stage-6-9e3a56a9868e.json",
+          "report_sha256": "db0b528bd48a48b1362a88a67d676c198bd7f68a36cf3ab5641baa3e7a65ae1e",
+          "tasks": 11,
+          "attempts": 176
+        }
+      ]
+    },
+    "qwen3-30b-thinking": {
+      "attempts": 9088,
+      "tasks": 568,
+      "verifier_rewards": 8768,
+      "missing_verifier_rewards": 320,
+      "successful_attempts": 1640,
+      "reward_counts": {
+        "0.0": 7128,
+        "1.0": 1640,
+        "None": 320
+      },
+      "trajectory_missing": 0,
+      "trace_file": "trajectories/qwen3-30b-thinking.parquet",
+      "reports": [
+        {
+          "array_index": 1,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/001/submission/report/stage-6-32cc1dcc1293.json",
+          "report_sha256": "c7d49e4801de0bdbc35fc211ee19e51b951112296480a55f080562c4a894c9f4",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 2,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/002/submission/report/stage-6-4298a9d2ac72.json",
+          "report_sha256": "1a3a87a675154f1e4fde4420852392f441b7932d090c9e91c9665f7d9ff6d945",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 3,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/003/submission/report/stage-6-6f2fe7db30ee.json",
+          "report_sha256": "ead1a472a17d38eca38870e70a26ee18651a0d356ccf7121c488ca8a6b9f603b",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 4,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/004/submission/report/stage-6-c114a10b679b.json",
+          "report_sha256": "7333f4996a1ea9fd3e9dcfc3d4f9d06c3ad005d14f5bfd30c0958914f0d8a631",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 5,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/005/submission/report/stage-6-da14f9d80b83.json",
+          "report_sha256": "092958e2bdbe3fe69c4f4c4b452e90a307dc0693f2b8c4b2d7041370e312360d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 6,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/006/submission/report/stage-6-a9b8b5fb80df.json",
+          "report_sha256": "36680d33c3a70f00aea204469ac301871dea470e6a9e531d43402e947449d549",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 7,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/007/submission/report/stage-6-de0b544a5289.json",
+          "report_sha256": "4c6acb3d0c83a2fe10aa4b78b1ca306fe6c778aded4238193cdf232fd45cc150",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 8,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/008/submission/report/stage-6-0891e55b2f84.json",
+          "report_sha256": "6eaaa10174a98b0452416e82ef839d14a305891512522da02ed1f20a9f0d11a5",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 9,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/009/submission/report/stage-6-a69885b3b17d.json",
+          "report_sha256": "6da2ea355565d7e9584b3b4a29a90f5e0a31ec1404b73953cce7ee5e9cc81749",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 10,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/010/submission/report/stage-6-89c7ccc9c19e.json",
+          "report_sha256": "23821f10fd4757d447835a423829a95654d6931bf1ac365e033b963c67dbd443",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 11,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/011/submission/report/stage-6-0a3bd3175bdf.json",
+          "report_sha256": "f48ff34d462fdd8681d75639fa4b13f84b0ca4851d18a1707f605298e8faf219",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 12,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/012/submission/report/stage-6-e7ce7ca50118.json",
+          "report_sha256": "8526da55ab3e4bf2504a5f0ea9acd440da787715a521d88b3f57cf9cd18da2ef",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 13,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/013/submission/report/stage-6-a08bff4a2d66.json",
+          "report_sha256": "8aeb3c8ed09f9af3169b3a8c1f99661469ed6a176012314795d8b8084d1aca4a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 14,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/014/submission/report/stage-6-a1bc7d30fce7.json",
+          "report_sha256": "b91f654f067b4c84ea67b3790a46fd6246163883794558d64f7eb657475cf85a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 15,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/015/submission/report/stage-6-2c3243ff4a39.json",
+          "report_sha256": "25694280a54ee790aed252010be1095bf8fb85519e68fb870f9062d031a50368",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 16,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/016/submission/report/stage-6-63b308a25e06.json",
+          "report_sha256": "02825b5be9441ee4169c30abc8c750acf4858d6bc3f28f91f2c5922a6d03fcd0",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 17,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/017/submission/report/stage-6-1aa3fec0de66.json",
+          "report_sha256": "a061ef978de51464daa0ffcb69da7ce943a63bce892ec88ee6e48066aa491487",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 18,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/018/submission/report/stage-6-cd94ac9608d0.json",
+          "report_sha256": "86d5f4f9936f8228f73e78d86dcf432adc0fd360ed2cc65df539521a785ce0e7",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 19,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/019/submission/report/stage-6-9e97b4b17e59.json",
+          "report_sha256": "4bdd94275feb21c7dc566a305a7702f2ac5689284ac2f1651d77ed84a55addc4",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 20,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/020/submission/report/stage-6-c242492140e4.json",
+          "report_sha256": "45ef2d7f4006e2c9f937f27be2400d2ed06d9d5517a87ee030f8e98e6463d158",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 21,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/021/submission/report/stage-6-dfcdf4307aab.json",
+          "report_sha256": "7d652c72085686a95eee227a12b159cff44733454f8010d159142acdbf50f444",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 22,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/022/submission/report/stage-6-4a30ac631c82.json",
+          "report_sha256": "c5a59aad62719df00ec7c67966707bf9a2616d957398275f54b0647e8f47116f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 23,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/023/submission/report/stage-6-1bcabfeeca2b.json",
+          "report_sha256": "c46d6a7c8efb30a94f81881fa1fe24ad76c9cb7d22a28f81f9032a50d9730d5a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 24,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/024/submission/report/stage-6-5f9a5df004de.json",
+          "report_sha256": "8f8a1dcce1b84aaf0a083d96fe73d8a1792bdf80f8ba5545a7ed3dec437950dd",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 25,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/025/submission/report/stage-6-64f32ff5ebec.json",
+          "report_sha256": "a546c617fa935e94c41c47c841473ed12d98f22f68d3ffe3e795798af4165a7c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 26,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/026/submission/report/stage-6-2a8223ffeb7d.json",
+          "report_sha256": "62815d3d86115fcf5609739381f706b1270116f558ff4782ecbb1a7fb8cfd2f2",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 27,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/027/submission/report/stage-6-9898fffcda96.json",
+          "report_sha256": "7c076a76c6bd7a1cc7028f445346ab361aedfa174d70acfceffff73753c5eabc",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 28,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/028/submission/report/stage-6-8a56f3985e3a.json",
+          "report_sha256": "005a0c0658a4085961da4ddf2a23e3484dedbdd3c81aa4ae9d2f500dc128843d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 29,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/029/submission/report/stage-6-7e30a218bcf5.json",
+          "report_sha256": "dcd58a5a3698b9247f052f7cdd87173f8d017c3c64e10e3c45b63cb0396f7b5c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 30,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/030/submission/report/stage-6-ac8fefd218da.json",
+          "report_sha256": "e342e1a212c8309a4eae1fed869f8408b17a7df7ff81a875c643b8224bc0d8fc",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 31,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/031/submission/report/stage-6-26fc39cb3bcd.json",
+          "report_sha256": "00e76a1711a054342340b7c2ef91676e6482a9ab277f7dbffe1fcd957eac3dce",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 32,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/032/submission/report/stage-6-67cd6c4dd223.json",
+          "report_sha256": "5b2da5ec150cfa71a981e61e47c93b87ba51e8a4c08a52d4e4896c2628fdfa69",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 33,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/033/submission/report/stage-6-27106a44f8f7.json",
+          "report_sha256": "7401b0fe535a3cf06279025011647b23dedb0257d7422e2c700ed866c3141ad4",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 34,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/034/submission/report/stage-6-a3a598d0acf8.json",
+          "report_sha256": "4f27de3b8206d6d1e2efcb46106b4948c4251bf359dcc71b4b145175821d615a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 35,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/035/submission/report/stage-6-0f4f407534ad.json",
+          "report_sha256": "c704779d103451fdc766c7888ffe34d2cc2e521c8e9e20115f3e640d065cc5e8",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 36,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/036/submission/report/stage-6-2719223ddf70.json",
+          "report_sha256": "0b5e9f814d700255d2ea0faaedcb28a2a024437ae75a50b1fcb16ed07201b96e",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 37,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/037/submission/report/stage-6-6ce1fc61d7c0.json",
+          "report_sha256": "9d53501b6cf9837903d4f8ad3099b08808acf43b0599f05a9343c8a49f2ffdc9",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 38,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/038/submission/report/stage-6-6b98f1eddf5e.json",
+          "report_sha256": "6638ca9daef1616af11efa615f7cc479ddd5469bd9f04e5efc5968e7a0c9386f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 39,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/039/submission/report/stage-6-0c533ca51b80.json",
+          "report_sha256": "ce7e6c6bde6a2fa119c2ac5e20d6db35da44072b533551f2f6eaf71376776b0b",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 40,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/040/submission/report/stage-6-42cee88034e9.json",
+          "report_sha256": "cff9cb03193855c8b09439328cb2637c43588ac21db5ec351145475c773c143d",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 41,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/041/submission/report/stage-6-42c38dea6d4c.json",
+          "report_sha256": "fe3bab4826ce9997cd9fb365ea5a032826e61a3b4322c6f9adcf287ca362c83f",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 42,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/042/submission/report/stage-6-c74c9db29fac.json",
+          "report_sha256": "6ecde8114aaa424b545a1344f09a7f0f2257cb59e3221e7b7ec6e29d2f13748e",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 43,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/043/submission/report/stage-6-bac886a72e5b.json",
+          "report_sha256": "69edc1e7d81a3027d83eee44d545bb5b2c7d9f5a2fbaa0ee8025ca442d49b701",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 44,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/044/submission/report/stage-6-0489110db96d.json",
+          "report_sha256": "1422f6bdc0bfaeea459f8fcb4a2fda6ad885ec7bcbb546cb90f3341f656f281c",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 45,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/045/submission/report/stage-6-b08ffcda28d6.json",
+          "report_sha256": "4d83520d7b0ab320af93115430aacb0984cdb9c864617e7067f188e4a150bc9a",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 46,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/046/submission/report/stage-6-115b77ab59a6.json",
+          "report_sha256": "d9a3189889d852e5b553e61280a1a20f36fd43f69ea976ba04de626ba5fa51f8",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 47,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/047/submission/report/stage-6-4a64070776ee.json",
+          "report_sha256": "a4f54eb9c96b5acd824b54c629fd15144b861666211a53ec6b9e61aa600d9c32",
+          "tasks": 12,
+          "attempts": 192
+        },
+        {
+          "array_index": 48,
+          "report": "/hnvme/workspace/y500bb12-seta-validation/experiments/teachers-recovery-20261015-remaining/qwen3-30b-thinking/048/submission/report/stage-6-ddf4e82a0f48.json",
+          "report_sha256": "e4cfe480263ae1a117c389b6f74f6609dcee4467c1eb338dff52d044a45fecc2",
+          "tasks": 4,
+          "attempts": 64
+        }
+      ]
+    }
+  },
+  "scheduled_attempts": 22263,
+  "missing_verifier_rewards": 764,
+  "missing_reward_classifications": {
+    "harness_failure": 524,
+    "agent_failure": 63,
+    "verifier_timeout": 32,
+    "unresolved_error": 145
+  },
+  "reward_policy": "The reward and verifier_reward fields contain the raw verifier reward. Both remain null when grading produced no reward. No failure score is imputed.",
+  "error_audit": {
+    "jsonl": "missing-reward-audit/retained-errors.jsonl",
+    "csv": "missing-reward-audit/retained-errors.csv",
+    "summary": "missing-reward-audit/summary.json"
+  },
+  "source_jobs": {
+    "qwen3-30b-instruct-2507": "1024392",
+    "qwen3-30b-thinking": "1024393"
+  },
+  "source_code": {
+    "archive": "code.tar.gz",
+    "sha256": "083e70d5dae3f98a8761c6fabedc91fd3e0e3ed67bdfaf4567daaf926646ad3e",
+    "source_tree": "/home/hpc/y500bb/y500bb12/pass16-recovery-20261015/code"
+  },
+  "snapshot_context": "The frozen SETA v2 partial selection is dated 2026-10-15 20:57 UTC. This complete recovery tranche overlaps that snapshot for shards already collected before the cutoff; the original selection remains a reproducible snapshot.",
+  "attempt_definition": "One row per trial using the exact attempt path recorded by its completed stage 7 report. Internal Harbor retries are not counted as separate pass@16 trials.",
+  "files": [
+    {
+      "path": "trajectories/qwen3-30b-instruct-2507.parquet",
+      "bytes": 106926263,
+      "sha256": "6f8dcbcf02010c5122524577bf4d1dcf70b1e255426b28763adbf032f5e1c19f"
+    },
+    {
+      "path": "trajectories/qwen3-30b-thinking.parquet",
+      "bytes": 289777842,
+      "sha256": "146da6ccc0490ad60c54aa57d00b60fd7d36bf3e9efc6fe17cc88f4a96bd0bd8"
+    },
+    {
+      "path": "missing-reward-audit/retained-errors.csv",
+      "bytes": 737162,
+      "sha256": "bde89bcab34632842037164b3bebc4703897333618bbbfe39ab07b8ee14fef59"
+    },
+    {
+      "path": "missing-reward-audit/retained-errors.jsonl",
+      "bytes": 1370391,
+      "sha256": "97cd4f6dcbee29b2d9bd972f7f88b051852b041fc3f8fef16e0bec092dd17155"
+    },
+    {
+      "path": "missing-reward-audit/retention-policy.json",
+      "bytes": 731,
+      "sha256": "67d6e1f948521ec0d1648789d0a256f1f40ddc024569955fd5fca09cd100769f"
+    },
+    {
+      "path": "missing-reward-audit/summary.json",
+      "bytes": 2356,
+      "sha256": "f91ae6b6d872e76c1724d8bea2d410f9c932c9c145a86e6cc4f1ef441e0758e1"
+    }
+  ],
+  "selection_merge": {
+    "as_of_utc": "2026-10-16T10:10:10+00:00",
+    "rule": "Keep task if either model has verifier reward 1 in frozen snapshot or this recovery run.",
+    "newly_retained_tasks": 55,
+    "published_kept_tasks": 2205,
+    "published_excluded_tasks": 902,
+    "solved_at_least_once": {
+      "qwen3-30b-instruct-2507": 2137,
+      "qwen3-30b-thinking": 1459
+    },
+    "baseline_attempt_counts_preserved": true
+  },
+  "curation_policy": "For incomplete task results with recorded failed trajectories, archive the task from the current release and treat each recorded failed trajectory as curation reward 0. Raw verifier_reward remains null when no verifier output exists; preserve the original error message."
+}

seta_v2/runs/2026-10-15-pass16-recovery/trajectories/qwen3-30b-instruct-2507.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f8dcbcf02010c5122524577bf4d1dcf70b1e255426b28763adbf032f5e1c19f
+size 106926263

seta_v2/runs/2026-10-15-pass16-recovery/trajectories/qwen3-30b-thinking.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:146da6ccc0490ad60c54aa57d00b60fd7d36bf3e9efc6fe17cc88f4a96bd0bd8
+size 289777842

seta_v2/tasks.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d447ca273c69e00b92abb677e6ae710d8e13bf345808b96f44c849ace5aa505
-size 224982161
+oid sha256:927df4ad0790782f500575a560a792e13a2ca3a31b52ff846408e0709c040653
+size 249061328