Entities · Models
Gemma-2-9B
11 articles tagged with this entity.
-
Geometry of Ordinal Representations in Language Models
-
A Mechanistic View of Authority Hierarchy in LLM Sycophancy
-
SpectrumKV: Per-Token Mixed-Precision KV Cache Transfer for Prefill-Decode Disaggregated LLM Serving
-
CRMA: A Spectrally-Bounded Backbone for Modular Continual Fine-Tuning of LLMs
-
Measuring Alignment-Induced Activation Shifts Correctly: A Template-Controlled Difference-in-Differences Protocol
-
Low-Resource Safety Failures Are Action Failures, Not Representation Failures
-
Feature Geometry of LoRA Adapters: A Sparse Autoencoder Analysis of Representational Divergence in Fine-Tuned Language Models
-
ReSAE: Residualized Sparse Autoencoders for Multi-Layer Transformer Interventions
-
Committed SAE-Feature Traces for Audited-Session Substitution Detection in Hosted LLMs
-
Multilingual Steering by Design: Multilingual Sparse Autoencoders and Principled Layer Selection
-
The Realignment Problem: When Right becomes Wrong in LLMs