# Conclusion
---
Four narrowly scoped computational claims reproduced under independent, frozen gates: identical segmented forward semantics, exact TBPTT gradients for the stated K-truncated objective, the corresponding training/inference alignment, and sparse forward-only retrieval. Across 256 float64 CPU systems, maximum relative gradient error was 1.89×10⁻¹⁵; a Tesla T4 float32 rerun with 256 full attention repeats passed all four gates in 47.44 seconds. Executed attention produced nonzero gradient through carried local KV and exactly zero through detached retrieval, while five matching patterns were confirmed in official commit `a875317b1fc1e2a00a8dee04a8dbc844219f4077`. This audit does not reproduce 7B fine-tuning, LongBench-E scores, or 128K A100 latency and memory.
## Scope & cost
| | This reproduction | Full replication |
|---|---|---|
| Scope | Four graph/architecture claims; 256 systems | 7B training, LongBench-E, 128K profiling |
| Hardware | CPU plus one Tesla T4 | A100-class training and profiling stack |
| Compute time | 0.77 s CPU + 47.44 s successful T4 | Not attempted; substantially broader |
| Cost | < $0.01 successful hosted compute | Not estimated |
| Outcome | 4/4 frozen gates passed | Out of scope |
---
````html
````
---
**🎯 Trackio dashboard** `segmented-64223-repro`
trackio-local-dashboard://segmented-64223-repro
---
**📦 Artifact** `segmented-64223-repro/repro-bundle:v0` · dataset
https://huggingface.co/buckets/Umong/repro-consistent-segmented-execution-artifacts#segmented-64223-repro/repro-bundle:v0
---
The versioned artifact contains executable code, both independent gradient oracles, the locked official implementation subset and license, frozen protocol, raw CPU and successful T4 JSON, generated figures, and the strict poster audit. Hosted evidence is mirrored at [segmented-64223-repro](https://huggingface.co/buckets/Umong/segmented-64223-repro); Trackio rewrites the artifact cell to the Space’s public artifact Bucket on publication.
