Vision-Language-Action (VLA) models combine perception, language, and motor control in a single architecture, yet how they translate multimodal inputs into actions remains poorly understood. We apply activation injection, sparse autoencoders (SAEs), and linear probes to six models spanning 80M--7B parameters across 394,000+ rollout episodes on four benchmarks. The visual pathway dominates action generation across all architectures: injecting baseline activations into null-prompt episodes recovers near-identical behavior, while cross-task injection steers robots toward source-task positions (99.8\% of X-VLA episodes align with the source trajectory), exposing spatially bound motor programs tied to scene coordinates rather than abstract task representations. Language sensitivity depends on task structure, not model design: when visual context uniquely specifies the task, language is ignored; when multiple goals share a scene, language becomes essential (X-VLA \texttt{libero\_goal}: 94\%$\to$10\% under wrong prompts vs.\ \texttt{libero\_object}: 60--100\% regardless). In all three multi-pathway architectures (\pizhalf{}, SmolVLA, GR00T), expert pathways encode motor programs while VLM pathways encode goal semantics ($2\times$ greater behavioral displacement from expert injection), and subspace injection confirms these occupy separable activation subspaces. Per-token SAE processing is essential for action fidelity on most architectures, though mean-pooling improves fidelity on X-VLA. Contrastive identification recovers 82+ manipulation concepts, and causal ablation reveals sensitivity spanning 28--92\% zero-effect rates independent of representation width. We release \textbf{Action Atlas} (https://action-atlas.com) for interactive exploration of VLA representations across all six models.

翻译：视觉-语言-动作（VLA）模型将感知、语言与运动控制统一于单一架构中，然而其将多模态输入转化为行动的具体机制仍不明确。我们运用激活注入、稀疏自编码器（SAE）及线性探针技术，对涵盖800万至70亿参数规模的六种模型开展研究，在四个基准测试集上累计采集超过39.4万个交互回合数据。实验表明：视觉通路在所有架构的动作生成中均占主导地位——将基线激活注入空提示交互回合能复现近乎一致的行为，而跨任务注入则会使机器人导向源任务位置（X-VLA模型中99.8%的交互回合轨迹与源轨迹重合），揭示出与场景坐标紧密绑定的空间运动程序，而非抽象任务表征。语言敏感性取决于任务结构而非模型设计：当视觉上下文唯一确定任务时，语言会被忽略；当多个目标共享同一场景时，语言成为关键要素（X-VLA模型在错误提示下对libero_goal任务的完成率从94%骤降至10%，而libero_object任务在任意提示下均保持60-100%的成功率）。在所有三种多通路架构（π0.5、SmolVLA、GR00T）中，专家通路编码运动程序，而视觉语言模型（VLM）通路编码目标语义（专家注入引发的行为偏移量是VLM注入的2倍），子空间注入进一步证实两者占据可分离的激活子空间。对大多数架构而言，逐token的SAE处理对动作保真度至关重要，但均值池化可提升X-VLA的保真度。对比识别方法可恢复82种以上操控概念，因果消融实验揭示出28-92%的零效应率敏感区间，该现象与表征宽度无关。我们发布**Action Atlas**（https://action-atlas.com）交互式平台，支持对全部六种模型的VLA表征进行可视探索。