Gemma 2 2B

The 'Gemma 2 2B' dataset/benchmark is used to evaluate the interpretability of language models by providing a structured set of tasks and metrics for assessing the effectiveness of various interpretability methods.

Papers using Gemma 2 2B (5)

ICA Lens: Interpreting Language Models Without Training Another Dictionary2026Check Your LLM's Secret Dictionary! Five Lines of Code Reveal What Your LLM Learned (Including What It Shouldn't Have)2026CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features2025CorrSteer: Steering Improves Task Performance and Safety in LLMs through Correlation-based Sparse Autoencoder Feature Selection2025Interpretable Steering of Large Language Models with Feature Guided Activation Additions2025