[Thinking about Local LLMs Day 40] Fragility was visible without ablation — AUC 0.84 with baseline logit difference alone, 0.66 correlation with required unit count

The predictions of small models are on much thinner ice than you might think. When measured on day 39, the median difference ...

From Text Generation to Logit Classification: Notes on Learning the Mechanisms and Design Philosophy of Decision Models (System 1)

I have felt that conventional generative AI carries significant waste in terms of cost and latency because it forces the ...