Add TRN hybrid non-record submission (1.4942 bpb, 1x RTX 5090)

[amabito]

Non-record submission: oscillatory recurrence + attention hybrid under the 16 MB constraint.

What this is

A 10-layer hybrid model (7 TRN layers + 3 attention layers) with int5 QAT and
zstd-22 compression. The TRN layers use a Kogge-Stone parallel prefix scan over
complex-valued oscillators -- no Triton, no custom CUDA, pure PyTorch.

Score: 1.4942 bpb (int5 roundtrip, 636 steps / 600s wallclock, 1x RTX 5090).
Artifact: 15.28 MB.

What went wrong

The model reaches 1.26 bpb in fp32 at 20K steps, but int5 quantization degrades
it to 1.93 bpb. The oscillator projection weights (d_model -> 6K, encoding
frequency and phase) accumulate O(t) phase drift from quantization errors.
At 1000 steps the error is small (+0.041); at 20K steps it collapses (+0.669).

A parameter-matched 13L Transformer shows only +0.016 int5 degradation at the
same step count. The failure is specific to oscillatory recurrence parameters.

What is included

• records/track_non_record_16mb/2026-03-25_TRN_Hybrid_Int5_1x5090/
  • README.md (architecture, ablations, quantization analysis, 13L comparison)
  • submission.json
  • train.log
  • train_gpt_trn.py (self-contained, zero external dependencies)
• README.md root table updated (Non-Record Runs)

What is not included

• 3-seed runs (single seed only)
• 8xH100 results (tested on 1x RTX 5090 only)