reward-lens: A Mechanistic Interpretability Library for Reward Models

Abstract

Every RLHF-trained language model is shaped by a reward model, yet the mechanistic interpretability toolkit -- logit lens, direct logit attribution, activation patching, sparse autoencoders -- was built for generative LLMs whose primitives all project onto a vocabulary unembedding. Reward models replace that with a scalar regression head, breaking each tool. We present reward-lens, an open-source library that ports this toolkit to reward models, organised around one observation: the reward head's weight vector $w_r$ is the natural axis for every interpretability question. The library provides a Reward Lens, component attribution, three-mode activation patching, a reward-hacking probe suite, TopK SAE feature attribution, cross-model comparison, and five theory-grounded extensions (distortion index, divergence-aware patching, misalignment cascade detection, reward-term conflict analysis, concept-vector analysis). A ten-method adapter protocol covers Llama, Mistral, Gemma-2, and ArmoRM multi-objective heads, with a generic adapter for any HuggingFace sequence classification model. We validate on two production reward models across ~695 RewardBench pairs. The central empirical finding is negative: linear attribution does not predict causal patching effects (mean Spearman $ρ= -0.256$ on Skywork, $-0.027$ on ArmoRM). The framework treats this disagreement as a property to expose, not a bug -- motivating a design that keeps observational and causal views first-class and directly comparable.