AI Attributes™
PreviewOpenSensations™
Open models for the senses in text: vision, hearing, smell, taste, touch and the inner body.
Open models that read the senses in text: how strongly a sentence's words are tied to vision, hearing, smell, taste, touch and the inner body. Two are published with open weights on Hugging Face: a RoBERTa fine-tune and a Gemma fine-tune.
The one that answers today is a prototype trained on a few hundred examples. It is live so the whole path can be seen working; its scores are not yet worth relying on, and the site says so.
Try it scores each sentence as its own block, and an API lets any app attach it with a URL and a key.
A sensation reported in text is a fact about the text, not evidence that the writer, or a model, feels it.
Where it stands
A prototype. Try it and the API answer with the RoBERTa model, but its readings are weak: the scores barely move from one sentence to the next.
Its part
Observers open models that read text. An observer is one attribute of text, the open models that read it, and an API. It takes a piece of text and returns a score for each label, with the model and version that produced it. It never takes part in the conversation.