The Geometry of Inference in Transformer Residual Streams
cs.CL, cs.LG
Submitted: 2026-09-29
Updated: 2026-09-29
Code: https://github.com/meta-llama/llama3
Terminology
Sources
- Entropy-Lens: Uncovering Decision Strategies in LLMs
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Gemma: Open Models Based on Gemini Research and Technology
- SemRF: A Semantic Reference Frame for Residual-Stream Dynamics in Language Models
- Geometric and Behavioral Stratification in Transformer Residual Streams
- Layer-wise Representation Dynamics: An Empirical Investigation Across Embedders and Base LLMs
- A Geometric Perspective on Next-Token Prediction in Large Language Models: Three Emerging Phases
- Qwen2.5 Technical Report
- Scale Determines Whether Language Models Organize Representation Geometry for Prediction
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering