Sunstone North Lab · all-HTML version

Same head, different silicon

The 22 MB Sunstone retrieval head was trained exactly once: bf16 precision, datacenter CUDA, a 31B host. Then everything a deployment varies was varied. Numbers are the published release-protocol results from the SRT repository and paper §11.6.4.

Why one matrix is enough

BridgeImage → text R@1Note
Centered cosine (zero training)0.288the free baseline
Orthogonal Procrustes0.226refuted: worse than nothing
One trained linear matrix0.66122 MB, 117k pairs
Two-layer MLP0.56733× the data, never wins

On the literature-standard Karpathy 5k split the same head scores 0.416 R@1, matching fully trained 2018 dual encoders from roughly 3,000 times less pair data. Every rung carries a shuffled-pairs control that collapses to chance.

Change the host: 31B → 3B

The full ladder was re-run on Qwen2.5-VL-3B, ten times smaller. Same anisotropic-linear fingerprint. At an identical 39,000-pair budget the small host's head lands at 0.577 R@1, inside the big host's 0.553–0.590 range. Ten times less model, nothing lost.

Change the precision: bf16 → 4-bit

The bf16-trained head, applied unchanged to 4-bit NF4 states: −0.01 R@1. A 42 KB mean recalibration, a single vector, recovers half of that.

Change the silicon: datacenter → the chip in a Mac

The same images embedded on datacenter CUDA and on Apple Silicon, different kernels, different quantization. Through the head, every local state retrieved its datacenter twin at rank one: 100% R@1 agreement.


10× host reduction with no loss · −0.01 R@1 at 4-bit with unchanged weights · 100% cross-silicon agreement · one artifact, every tier. Train once, read everywhere.