Ruby Tumbler: preserving dynamic behavior without unnecessary work
I've gotten further than expected, and learned some fascinating things about dynamic languages.
Lynn Cole / Laboratory
Laboratory
Experiments, research notes, mad science, and things in progress.
Flyborg: a synthetic memory prosthesis for a digital fly
I'm spending the weekend performing brain surgery on a digital fruit fly.
Why give a digital fruit fly a memory?
A digital fly already has memory circuitry. I am connecting that machinery to synthetic storage to explore what it remembers and how it learns.
A noise-painting pipeline, from painted input to refinement
The idea is that you start with a hand painted image or collage, but that you design it in such a way as that it messes up the classification and denoising systems in a diffusion model.
After the activation injection: what can the model infer?
An experiment proposal separating activation injection, transcript conditioning, and what a model can infer after the intervention ends.
A composable Brainfuck compiler for LLMs
If people stop reading generated code, what should a language for the machine look like? A deliberately obnoxious compiler proposal.
Recursive training: preparing a third corpus
Preparing a third recursive-training corpus means separating conflicting objectives and testing the order in which the student learns them.
The beginning of a faster Ruby compiler
Ruby's not my favorite language, but I find it offensive that anyone would replace a perfectly good Rails application with an app written in Rust.
AI-hotbox: a counterexample in activation-steering experiments
Changing one activation-steering corpus turns a pain-language experiment into a useful counterexample—with a much less solemn condition.
When stronger training signals weaken self-assessment
A mentor improved the training signals but weakened Qwen’s self-assessment. The next corpus starts with code reasoning and software estimation.
Kinetic Painter: learning the instrument while building it
Building a gesture-driven painting instrument means learning how to paint with it at the same time.
Training competence and engineering taste separately
Recursive training improved correctness while weakening engineering taste. A controlled learning-rate ablation and a failed promotion gate changed the next experiment.