Quantized-model interpretability needs a noise-floor check
A new AI-safety paper says a high cosine similarity does not prove an interpretability tool survived quantization, so researchers should report the estimator’s noise floor before claiming transfer. For Qwen2.5-1.5B-Instruct, the paper measures class separation and finds sampling alone produces cosine agreement of 0.978 to 0.994; its analysis finds the direction rotated beyond that noise at INT4, while detecting no movement at INT8 without claiming equivalence.