Neural Network Self-Description: The Prerequisite for Complex Reasoning
In previous articles (1, 2), I analyzed Anthropic's J-space: why their interpretation falls short; why J-space analogues appear in micromodels under objective pressure rather than as an emergent property of scale; and why the vector they discovered is strictly equivalent to Vygotsky’s concept of the sign.Two questions remained outside the scope: is a metasign possible—a sign integrating several basic ones—and what is the true nature of the neural network's self-description (the report)?This article addresses both.TL;DR: Self-description turned out to be not an overlay on the sign, but the prerequisite for its formation. The network solves the task equally well with or without a report, but the sign axis crystallizes only where it is demanded to be named. I was unable to force the formation of a metasign.Introduction
Самоописание нейросети: условие сложного мышления
В предыдущих статьях я разбирал J-space, открытый Anthropic: почему их интерпретация неполна, почему аналоги J-space появляются на микромоделях под давлением объективных причин, а не эмерджентно от размера модели, и почему найденный ими вектор эквивалентен понятию знака у Выготского.За рамками остались два вопроса: возможен ли метазнак — знак, объединяющий несколько базовых, — и в чём суть самоописания, репорта нейросети.Эта статья о них.
J-Space in Micromodels: A New Interpretation of Anthropic’s Discovery
Recently, Anthropic reported the discovery of an analogue to the global workspace in LLMs — J-space. In a previous article, I analyzed why Lev Vygotsky’s theory offers a better interpretation of the researchers' findings.In this article, I argue, based on experiments, that J-space analogues do not depend on the size of the neural network. Rather, they are an inevitable stage in the evolution of a cognitive structure developing under pressure.TL;DR. Anthropic describes the model’s internal "workspaces" using the language of emergence — as a property that arises in sufficiently large networks.I tested this on micromodels, where every representation axis can be verified, and drew the following conclusions:
