Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

representation dynamics

Representation dynamics refers to the structural evolution and transformation of internal data representations within a neural network as information propagates across its layers or evolves over the course of training. In deep learning models, these dynamics describe how latent embeddings change along the network depth in terms of geometric structure, dimensionality, information compression, and invariance to input variations. Rather than treating internal activations as static or focusing solely on final outputs, analyzing representation dynamics characterizes how intermediate layers progressively distill, preserve, or discard features, providing insight into the computational mechanisms and functional utility of hidden layers in processing complex data.

1 item

Layer by Layer: Uncovering Hidden Representations in Language Models

Layer by Layer: Uncovering Hidden Representations in Language Models

Oscar Skean, Md Rifat Arefin, Dan Zhao, Niket Patel, Jalal Naghiyev, Yann LeCun, Ravid Shwartz-Ziv

OrganizationsMetaMila – Québec Artificial Intelligence InstituteNew York UniversityUniversité de MontréalUniversity of California, Los AngelesUniversity of KentuckyWand.AI

Why you should read this

Demonstrates that intermediate layers in language models consistently produce richer representations than the final layer, introducing a geometric and information-theoretic framework that explains why mid-depth embeddings achieve superior performance across diverse downstream tasks.

From extracting features to generating text, the outputs of large language models (LLMs) typically rely on the final layers, following the conventional wisdom that earlier layers capture only low-level cues. However, our analysis shows that intermediate layers can encode even richer representations, often improving performance on a range of downstream tasks. To explain and quantify these hidden-layer properties, we propose a unified framework of representation quality metrics based on information theory, geometry, and invariance to input perturbations. Our framework highlights how each layer balances information compression and signal preservation, revealing why mid-depth embeddings can exceed the last layer's performance. Through extensive experiments on 32 text-embedding tasks across various architectures (transformers, state-space models) and domains (language, vision), we demonstrate that intermediate layers consistently provide stronger features, challenging the standard view on final-layer embeddings and opening new directions on using mid-layer representations for more robust and accurate representations.

Added

2026-09-28