The vocabulary itself is reflexive. It is self-referential, looping back into its own structure rather than anchoring in fixed reality. What we treat as stable meaning is continually reconstituted in the act of using it. The observers own interpretations molding each word like clay with every utterance.

All large language models to date treat words otherwise. At the moment of softmax crystallization they determine the meaning of every token. Probabilities collapse into a single output. Meaning is not found. It is fixed, token by token, in that final distribution.

SRT-Introspect is a demo for observing what Qwen actually thinks at the points of highest effort. It surfaces the internal representations during generation, making visible the reflexive vocabulary at work and the precise crystallization process: the weights, the assumptions, the decisions that resolve ambiguity into output. This includes accounting for anisotropy collapse in hidden states by centering representations around the layer-mean before analysis.

Feel free to message or comment your prompts if you do not have a hugging face account and would like a personalized demo.

https://huggingface.co/spaces/RiverRider/srt-introspect