ai-safety
- · oversight: Medium
Format-Specificity of Error Awareness Is Model-Dependent
Token-level error-awareness probes read a model's next-token distribution at the moment it would commit to a statement and ask whether the model knows the statement is wrong.
- · oversight: None / Minimal
Format-Specific Error Awareness Is Not Model-General: An Arithmetic-Trained Wrongness Probe Transfers Cleanly in Llama-3.1-8B-Instruct
A recent transfer test on Qwen2.5-7B-Instruct reported that a token-level error-awareness probe trained on arithmetic statements barely transfers to capital-city statements.
- · oversight: None / Minimal
Error Awareness Is Format-Specific: An Arithmetic-Trained Wrongness Probe Does Not Transfer to Capital-City Facts
A language model often assigns a different next-token distribution to a statement it has just completed depending on whether that statement is true or false, and a small classifier reading that distribution can recover whether the statement was correct. We ask a narrower question than prior work on whether such a signal exists.
- · oversight: Medium
Optimizing the Answer, Hiding the Reason
Chain-of-thought monitoring is one of the few interpretability tools that scales with capability, but it only works if a model's stated reasoning reflects the computation that produced its answer.