Activation Oracles (AOs) face a sharp specificity-faithfulness trade-off: three-layer supervised readers (`ours-3layers`) recover more target attributes than si
Answering entity-specific questions about multi-step `<think>` chains-of-thought (`CoT`) with a frozen Activation Oracle (AO) exposes a spatial-scale incompatib
Can a frozen Activation Oracle (AO) recover chain-of-thought scalars from under $5\%$ of source positions without raw text, and what can pre-answer contextual a
When a language model emits a bare wrong binary decision ("No" or "Yes") on a multi-premise prompt without a reasoning trace, we study conditional operational r