atom ckpt first 2026-10-11 20:36:19

rm test file

test _t3.bin

test _t2.pt

test small pt

atom ckpt first (pt)

atom ckpt first 2026-10-11 20:33:58

README: fix Status contradiction.

README: 5896 ids/s at B=16.

convert_atom.py skips existing.

Atom corpus partial (110 files, 8.3GB). Resume-ready.

Compile the pure core. B=16 hits 5896 ids/s.

Capacity MoE.

Grouped MoE.

Per-head carry.

Remove item syncs.

Measured numbers.

Factor path.

Triton MoE first cut.

GPU benchmark numbers.

Compare TF rates.

Lanes and per-row BOS reset.

README: BOS reset and carry metric.

Reset at BOS.

Status 2026-10-09. Exact CPU rates.

Speech measured on smoke.

GPU benchmark protocol.

Calibrate phase_proj to c/sqrt(d).

Learned phase encode. Expert collapse fixed.

Document real expert hits and theta-bar vs spacing.

Per-token MoE routing. Causal leak closed.

Document last-theta chunk routing and the inert leak.

Equal-T Kuramoto check: one step matches four, gates unchanged.

Document single RK4 step and the r sweep.

Planning table: days = 2.4B ids / measured ids per second.

fast4gpu_atom trainer, guards passed, CPU smoke 1378 ids/s.

Document that boost_v4 is not Atom-ready.

Atomization explained, CPU smoke 1729 Atom ids/s.

Separate BPE token rate from Atom id rate.

Throughput section: measured 1k tok/s, boost_v4 is the 1B trainer.

English README and Claude brief.

Open prefix generation, no byte ban.

Sane init CE, Atom starter corpus, pod script stays off.

v4 loop, Siren synced on grow, unique40 probe, Atom features.

FractusTokenizer is the Atomizer. Siren modules are on the MoE forward.

Plug original CTE mechanisms into AtomSession.

Body growth lock: vocab 266 fixed, expert phases copied, train hit counter.

Philosophy: continuous thought, unbounded training, fast open.

Publish Fractus CTE-Atom: Atomizer spans on the HF engine, vocab 266, from scratch.

initial commit