Skip to main content
Category · 5 posts

Research Notes.

Research-backed essays from the LLM mechanics marathon. Structured synthesis, visible citations, and the same JAAX voice we use when the conclusions actually have to survive contact with operators.

Research note · May 4, 2026 · 13 min

The Strong Stochastic Parrots Claim Is Dead.

A citable 2026 verdict piece: causal world models, optimization-like in-context learning, and identified circuits all weaken the strongest parrots framing.

Research note · May 4, 2026 · 14 min

Lost in the Middle Is Three Bugs.

A triage tree for RAG engineers: softmax sinks, RoPE decay, and training-distribution bias each create their own version of the same long-context failure.

Research note · May 4, 2026 · 15 min

Compression Is a Noisy Proxy.

A five-paper tokenizer synthesis showing why compression only works as a stand-in when it happens to track semantic boundary alignment.

Lead piece · May 4, 2026 · 18 min

There Is No Code Mode.

A two-channel explanation for why LLMs differ across code, prose, and structured data without requiring separate hidden modes in the architecture.

Research note · May 4, 2026 · 11 min

Format Is a Hyperparameter.

Why prompt formatting deserves the same rigor as model choice, what it costs in token budget, and how to test it with a real eval loop.