FA-98221 / Attention mask construction / Open access

Causal stripe mask includes keys from every residue · case 01

Causal stripe mask includes keys from every residue.

Verified by executionVariant 1 · 6 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The mask preserves causal order but omits the required residue constraint.

THE FAILURE

The mask preserves causal order but omits the required residue constraint.

Unsuccessful approach: Grouping positions into fixed-width buckets rejects aligned keys that cross a bucket boundary.

Case contract

Query i sees causal key j iff j<=i and i%stride==j%stride.

Why this case matters

A small explicit visibility matrix makes sequence-attention boundary behavior reviewable.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(length, stride):
    return [[j<=i for j in range(length)] for i in range(length)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('same residue',solve(N+3,2),[[j<=i and i%2==j%2 for j in range(N+3)] for i in range(N+3)])
check('unit stride',solve(3,1),[[j<=i for j in range(3)] for i in range(3)])
check('empty',solve(0,2),[])
check('singleton',solve(1,7),[[True]])
check('future forbidden',solve(3,2)[0],[True,False,False])
check('nonadjacent aligned',solve(5,2)[4],[True,False,True,False,True])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
same residue	`[[True, False, False, False], [True, True, False, False], [True, True, True, False], [True, True, True, True]]`	`[[True, False, False, False], [False, True, False, False], [True, False, True, False], [False, True, False, True]]`	Failed
unit stride	`[[True, False, False], [True, True, False], [True, True, True]]`	`[[True, False, False], [True, True, False], [True, True, True]]`	Passed
empty	`[]`	`[]`	Passed
singleton	`[[True]]`	`[[True]]`	Passed
future forbidden	`[True, False, False]`	`[True, False, False]`	Passed
nonadjacent aligned	`[True, True, True, True, True]`	`[True, False, True, False, True]`	Failed

SHA-256 / d25e11491655924425466d5c5294259b9898afca377e767ff8449f867c93877f

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(length, stride):
    return [[j<=i and i//stride==j//stride for j in range(length)] for i in range(length)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('same residue',solve(N+3,2),[[j<=i and i%2==j%2 for j in range(N+3)] for i in range(N+3)])
check('unit stride',solve(3,1),[[j<=i for j in range(3)] for i in range(3)])
check('empty',solve(0,2),[])
check('singleton',solve(1,7),[[True]])
check('future forbidden',solve(3,2)[0],[True,False,False])
check('nonadjacent aligned',solve(5,2)[4],[True,False,True,False,True])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)

Boundary fixture	Actual	Expected	Outcome
same residue	`[[True, False, False, False], [True, True, False, False], [False, False, True, False], [False, False, True, True]]`	`[[True, False, False, False], [False, True, False, False], [True, False, True, False], [False, True, False, True]]`	Failed
unit stride	`[[True, False, False], [False, True, False], [False, False, True]]`	`[[True, False, False], [True, True, False], [True, True, True]]`	Failed
empty	`[]`	`[]`	Passed
singleton	`[[True]]`	`[[True]]`	Passed
future forbidden	`[True, False, False]`	`[True, False, False]`	Passed
nonadjacent aligned	`[False, False, False, False, True]`	`[True, False, True, False, True]`	Failed

SHA-256 / 5e648564c57eb5a6eab81c1dc1947877dd08021fba402294e3230490e9381abd

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 6 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Verification & scope

These are toy Boolean matrix contracts; they do not define a production model architecture or tensor library behavior. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-30T16:48:10.629758+00:00.

Case digest / 7827aaa1aaa23026505a6dc0fd7d4b7ea23f3f46c5fa17cf88a89b08ea4bcb57