Non-record: Elite UT v22.8 12-step recurrence (Windows 3090)

[Ribin545]

Title

Non-record: Elite UT v22.8 12-step recurrence (Windows 3090)

Summary

Adds a new non-record submission folder:

• records/track_non_record_16mb/2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/

This submission documents and packages a Windows-focused training setup for an Elite Universal Transformer (UT v22.8) with 12-step recurrence and Muon ("Polar Express", degree-5) stability settings.

What’s included

• README.md index linking to detailed notes in notes/
• submission.json
• Windows batch entrypoints:
  • setup_elite_env.bat
  • limits_test_10m.bat
  • final_run_10m.bat
• Training code and utilities required to run from within the records folder

Repro / Run

From inside the records folder:

setup_elite_env.bat
limits_test_10m.bat
final_run_10m.bat

Notes

• All paths are relative (repo-root data via ..\..\..\data\...).
• Markdown notes are written in neutral voice (no second-person phrasing).

[MatoTeziTanka]

[RETRACTED 2026-04-11] — This IMPORT_FAIL was a false positive. Root cause: sibling module exists in same records/ folder; runner sys.path bug. Your code is not broken. See correction below: #1199 (comment)

---

Community Review — Non-record: Elite UT v22.8 12-step recurrence (Windows 3090)

Compliance: NEEDS AUTHOR ACTION — train_gpt.py fails to import on CT2038 (Python 3.10 / torch 2.10.0+cpu)

What I found: The CPU smoke test on CT2038 (proteus-engine, 128 GB RAM, Triton 3.6.0, flash_attn stub, cutlass_evt_fusion stub) failed at the import step with:

ModuleNotFoundError: No module named 'model'

A few of the common patterns I've seen for this class of error in the 2026-04-11 sweep:

• PEP 701 f-string nesting — e.g. log(f" {cat}: {", ".join(...)}") is valid Python 3.12+ but invalid Python 3.10 because the inner ", " re-enters the outer double-quote context. One-character fix: ', ' instead of ", ". See PR Record: SP8192 + Improved Parallel Residuals + Muon 0.97 + LR 0.03 + Legal TTT — val_bpb 1.07785 (3-seed mean) #1541 / Record: SP8192 + Triple Recurrence + Banking + Fused MLP + Muon 0.97 — val_bpb 1.0778 (3-seed mean) #1523 for reference.
• Missing flash_attn variants — e.g. from flash_attn_interface import flash_attn_varlen_func when the wrapper script only stubs flash_attn_func. Not a PR defect on H100s, but the eval image / CPU preflight path needs a guarded import.
• Local compiled extension — e.g. import cutlass_evt_fusion from a records/*/cutlass_evt_fusion/ subfolder that isn't on the import path at smoke time. Usually an import-order issue inside the script.
• Actual syntax error — typo, missing bracket, etc.

Recommendation: Could you run python3 -c "import py_compile; py_compile.compile('train_gpt.py')" on your records-folder train_gpt.py under Python 3.10 specifically? The eval image is Python 3.10 per Issue #17 / the README, so any parse error on 3.10 blocks the submission at import time before any of the scored-eval logic runs.

Once the parse/import issue is fixed, I'll re-run the compliance audit through the normal pipeline. No other flags identified yet because the audit halts at the import step.

---

Reviewed by @MatoTeziTanka — The Agora. CPU smoke test (CT2038 proteus-engine, 2026-04-11): IMPORT_FAIL — ModuleNotFoundError: No module named 'model'. Classification via classify_prs.py AST-based classifier; full compliance audit deferred until the import issue is resolved. Auto-drafted from a template and spot-checked before posting.

[MatoTeziTanka]

Retraction — this IMPORT_FAIL was a bug in my smoke runner

Sorry @Ribin545, this one's on me. I re-audited the IMPORT_FAIL I posted above and it was a false positive — the fault is in how my CPU smoke runner set up sys.path, not in your code.

What happened:

The runner imported your records/track_non_record_16mb/2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/train_gpt.py with only the script's folder implicitly on sys.path, so when your file did from model import ... it couldn't resolve the sibling model.py that lives in the same 2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/ directory. The error I reported — ModuleNotFoundError: No module named 'model' — looked like a missing file, but I re-checked the head SHA 332eed9 and records/track_non_record_16mb/2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/model.py is right there, committed to the PR, next to train_gpt.py.

Verified at head 332eed9:

records/track_non_record_16mb/2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/model.py   ← sibling module, exists
records/track_non_record_16mb/2026-04-01_EliteUTv22p8_12StepRecurrence_Windows3090/train_gpt.py   ← imports it

On the real eval image (Python 3.10, records/*/ as the working dir), this import resolves correctly because the records folder ends up on sys.path via the standard cwd-driven import or via the eval harness's per-record entry point.

Your PR is not broken by this error. I'm retracting the IMPORT_FAIL classification. I'll re-queue the full compliance audit (BPB check, n-gram / TTT / SLOT flags, etc.) on the current head and post findings separately.

Again — sorry for the noise. These community reviews only work if I actually read what I'm reviewing, and I didn't in this case.