Nov 07, 2026
Static Spectral Trojan Auditing and Pre-Load Checkpoint Sanitization
How trust boundaries, C++/CUDA kernel failure classes, and spectral weight audits fit together, with measurements from a 46-file local inventory (2,619 analyzed weight matrices across 14 full-model reads and 16 scoped variants) and 186.8M GGUF quantized blocks.
Nov 05, 2026
From Sparks to Secant: A Four-Part Series on Weight-Space Programming, C++ Kernel Safety, Subspace-Locked LoRA, and Spectral Trojan Auditing
A four-part series and open-source toolkit testing weight-space programming across 17 models, quantized GGUF blocks, LoRA fine-tuning, and C++ inference kernels.
Nov 05, 2026
Project-Then-Heal: Weight-Space Programming and Subspace-Locked LoRA
How weight edits map to SafeTensors and GGUF storage, why head permutations matter, and what live fine-tuning and synthetic tests show about subspace-locked LoRA.
Nov 05, 2026
Trapped Inside `<think>`: Multi-Subspace Weight Surgery and What Controls Reveal on Small Reasoning Models
A study of reasoning-loop timeouts on Qwen3.5, activation-based weight-edit calibration, and the controls needed to distinguish completed answers from repeated crash-trace text.
Nov 05, 2026
The Illusion of Safe Tensors: Why Valid Model Files Still Need Numerical Checks
SafeTensors guarantees that deserializing a model file cannot execute pickle code. It does not guarantee that the numbers inside are safe for the C++ and CUDA kernels that consume them.