AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Features Emerge as Discrete States: The First Application of SAEs to 3D Representations

Albert Miao, Chenliang Zhou, Jiawei Zhou, Cengiz Oztireli

ICLR 2026 · Architectures and efficiency › RNNs and normalization · approach: Sparse dictionary representations · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase SAE codebook size 256 to 1024 (ablation)BatchTopK SAE · Dora-VAE SAE variationsrelative l2 reconstruction (k=8): 0.366 -> 0.356 (k=8, 256 -> 1024)Improved
increase SAE codebook size 256 to 1024 (ablation)BatchTopK SAE · Dora-VAE SAE variationsuniversality (k=8): 0.421 -> 0.208 (256 -> 1024)Hurt
increase threshold k from 4 to 16 (ablation)BatchTopK SAE (codebook 512) · Dora-VAE SAE variationsrelative l2 reconstruction: 0.507 -> 0.187Improved
increase threshold k from 4 to 16 (ablation)BatchTopK SAE (codebook 512) · Dora-VAE SAE variationsDora-VAE loss: 0.427 -> 0.216Improved
increase codebook size 256 to 1024 at k=16 (ablation)BatchTopK SAE · Dora-VAE SAE variationsDora-VAE loss: 0.228 -> 0.239Mixed
include log loss difference as predictor of transition point (ablation)OLS regression per feature · Transition point predictionpartial R^2: 0.188 vs 0.042 (avg value) and 0.025 (density) at n_abl>=500Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy