LSR Anchoring — Cross-Lingual Safety Steering Results

This dataset contains all experimental results, activation caches, and run logs for the paper: Latent Space Refusal Anchoring for Low-Resource African Languages: Mechanistic Safety Recovery Without Retraining The method uses Mean Activation Steering and Sparse Autoencoder (SAE)-derived mean-activation directions (anchors) extracted from a source language (English) and applies them at