alisson-anjos

Diagnosing a LoRA by cutting it apart

A practical guide to finding which modules of a trained LoRA cause a specific defect (blurry skin, lost identity, ignored pose) and to fixing it without retraining.

Everything here is post-hoc weight surgery: you take a .safetensors you already trained, produce variants with some module groups zeroed or scaled, generate with each under a fixed seed, and measure. No GPU training, minutes per iteration.

Worked examples at the end come from two real LoRAs on Qwen-Image-2.1 (a head swap and a

davidklaw

// Dump Object
function dump_obj(obj){
  log("#####################################################################################")
  log("## Dumping object " + obj )
  log("## obj class is: " + [obj className])
  log("#####################################################################################")

  log("obj.properties:")
  log([obj class].mocha().properties())
  log("obj.propertiesWithAncestors:")