J-Lens cannot detect what subliminal learning teachers transmit in 7B-32B models
J-Lens can't detect what a subliminal learning teacher is transmitting (in 7B-32B models)
The authors reproduced the subliminal learning experiment and found that a Qwen2.5-7B-Instruct teacher system-prompted to "love owls" could transmit the preference to student models through number sequences, with the proportion of students mentioning owls rising from 1% to about 8%, but J-Lens read owl into top-1 in only 1.65% of trials out of about 1.42 million completions (chance level 0.26%). The lens reached 45% top-1 on toy tasks such as copying, indicating it was not broken; the situation of "transmission occurred but the lens could not read it out" was also reproduced on Gemma-4-31B, and among the remaining 46 settings only 5 showed clear transmission. Additional experiments showed that training students only on samples where the lens's top-1 hit could push the transmission rate to 57.2%, while samples the lens completely missed still transmitted at a rate close to that of a random subset.
The authors found on 7B-32B open-source models cases where subliminal learning transmission occurred but J-Lens could not read out the corresponding concept, which is of reference value to readers who rely on interpretability tools to monitor model internals.