One arm of a study on teaching a language model to notice when its own conversation has collapsed into repetition, emit a special <end> token, and change the subject.

Version 3. The first working detector, and the one that taught us the lesson: its healthy and stuck examples came from different models, so it learned writing style instead of looping. Scored 1.000 on our own test set and fired near-randomly in live conversation.

Two things that will silently break this. <end> is a special token (id 151665), so decode with skip_special_tokens=False or the model fires and you see nothing. And feed it about 5 turns: it was trained on 5-turn windows, and a 16-turn window dropped firing to 0 of 17 conversations in our tests.

Full write-up, training recipe and limitations: center-for-humans-and-machines/end_token. Datasets: Farrous05/end_token_data_full.

Built at the Center for Humans and Machines, Max Planck Institute for Human Development.