Superset: every prompt-reviewed task, 12 splits, tmax_clean 6086, tmax_repair 7453, tmax_killed 1055, calibforge_clean 3299, calibforge_repair 1993, calibforge_killed 139

README.md

@@ -31,12 +31,12 @@ size_categories:
 **Every TMAX and CalibForge task we have reviewed with one of our prompt rubrics, in one dataset.** Each task appears
 exactly once, with its most recent verdict, where it came from (`origin`), every earlier verdict
 (`verdict_history`), and, where one exists, a solution verified the way the line that produced it verified it.
-19,951 tasks in 6 splits: clean, repair and killed for each corpus.
+20,025 tasks in 6 splits: clean, repair and killed for each corpus.
 
 | split | rows | with a solution | security-domain rows |
 |---|---:|---:|---:|
 | `tmax_clean` | **6,086** | 6,086 | 542 |
-| `tmax_repair` | **7,379** | 2,755 | 874 |
+| `tmax_repair` | **7,453** | 2,777 | 874 |
 | `tmax_killed` | **1,055** | 3 | 106 |
 | `calibforge_clean` | **3,299** | 3,299 | 285 |
 | `calibforge_repair` | **1,993** | 1,261 | 175 |
@@ -55,7 +55,7 @@ ds = load_dataset("Fzz1/tb-full-data")   # ds["tmax_clean"], ds["tmax_repair"],
   TMAX 1,055 (first removed by: v1r2 555, v13 tb-training-mix-agentpick 203, v13 tb-training-mix 129, gate2 129, gate1 23, v12 8, v8 8), CalibForge
   139 (answer leaks in `Fzz1/tb-training-mix-agentpick`'s review). The stage, the label it
   recorded and the reason are in `origin`, `verdict_history` and the split file.
-- **Answer leaks found by the latest review are in repair**, with the verdict visible (TMAX 3,333,
+- **Answer leaks found by the latest review are in repair**, with the verdict visible (TMAX 3,332,
   CalibForge 557). **Filter on `verdict` before training on a repair split**:
   `ds["tmax_repair"].filter(lambda r: r["verdict"] != "ANSWER-LEAK")`.
 - **One verdict per task, the most recent of ours.** Earlier verdicts and kills are in `verdict_history` and in the
@@ -77,7 +77,7 @@ ds = load_dataset("Fzz1/tb-full-data")   # ds["tmax_clean"], ds["tmax_repair"],
 
 ## TMAX
 
-`allenai/tmax-15k-open-instruct` at `7b090eca`: 14,520 of its
+`allenai/tmax-15k-open-instruct` at `7b090eca`: 14,594 of its
 14,601 tasks.
 - The tasks of this dataset's TMAX line: judged by Kimi K3 under rubric v13 (some earlier v13 verdicts), rolled out by
   DeepSeek V4.1 Flash (effort high, retries at max), replayed once.
@@ -90,15 +90,19 @@ ds = load_dataset("Fzz1/tb-full-data")   # ds["tmax_clean"], ds["tmax_repair"],
 - The tasks earlier lines had already reviewed or released: in `Fzz1/Tmax-Tasks-Clean`,
   `Fzz1/tb-training-mix-agentpick` or `Fzz1/tb-training-mix`, killed by an earlier review or gate, or reviewed under an
   earlier rubric. Where a v13 verdict exists it decides. Otherwise the task's latest pre-v13 verdict is kept **in its
-  own vocabulary** (v8 1,107, v2 167, v1r1_review 164, v3 79, v1r1_rejudge 51, v7 5, v1r2_fable 4, v1r2_opus 4, v4_disp 3, v6_b1 1): `rubric` and `tier_source` name the
+  own vocabulary** (v8 1,107, v1r1_review 163, v2 160, audit452 97, v3 70, v1r1_rejudge 49, v1r2_fable 4, v1r2_opus 4, v4_disp 3, v7 1, v6_b1 1): `rubric` and `tier_source` name the
   stage, `verdict` holds its label, `tier` is empty, and the row is in repair (`pre_v13_verdict_only`), or in
   `tmax_killed` if an earlier stage killed it. These earlier rubrics are not mapped to v13 tiers because they
   disagree with v13 too often to be read as one: of tasks judged by both, v8 PASS became v13 SEED on 40 of 57.
 - A solution an earlier line verified against **different** verifier bytes is not carried (most of
   `Fzz1/Tmax-Tasks-Clean`'s reaudit and longlongcheck solutions); those tasks are in repair unless another verified
   solution exists.
-- Not here: 81 tasks with no prompt review on record (listed by a
-  release, with no verdict anywhere); each is in the split file with the reason.
+- Not here: 7 tasks with no prompt review on record (listed by a
+  release with no verdict anywhere, including tasks the 452-corpus audit held without reviewing them); each is in the
+  split file with the reason.
+- `audit452` is the 452-corpus static semantic audit of `Fzz1/Tmax-Tasks-Clean`'s reaudit corpus (protocol 2.2,
+  independent A/B reviews with an adjudicating C on disagreement); its label is its assessment (`STATIC_RISK`,
+  `SUPPORTED_WITHIN_SCOPE`). It killed no task.
 
 ## CalibForge
 
@@ -149,4 +153,4 @@ above, except:
 
 ## Canary markers
 
-The canary scan found 119 benchmark canary marker(s) across the uploaded files. They are kept as they are.
+The canary scan found 141 benchmark canary marker(s) across the uploaded files. They are kept as they are.

data/tasks-tmax_repair.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54445c4d18c64ba0fe2b199c97fafefe850b1042fff7d6535349b1f02b25f373
-size 87255040
+oid sha256:76dbedd95bd542aad3307598ca56669a58832976134523be9c38c0b2511cd292
+size 88043520

data/tmax_clean.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4c7a36a6844271504ba08935d9d7fbbbe4c109d6a2a11cd89d01cda1d527d08
-size 41326641
+oid sha256:8141a441ea9651318f557f22cb043e377ad5673bb47f29a834ee22138c62dd90
+size 41326850

data/tmax_killed.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59393706034d4ce4712f6cc39bf5d4313e250c892564d06327cfe944f390e547
-size 4878646
+oid sha256:9b71fd72ca3a9135813578b052d883e984c94f601aa13864cedd80a7575cedc1
+size 4878694

data/tmax_repair.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2c17d8c085824b52d16d2de7af4d6c6dcd4891f27b56db330c53477acc2eabf
-size 43455835
+oid sha256:10985611b4e81c101e1d3bcf16eef9a6b16ba14d5a007914bd8287342de8d179
+size 43909040