8 ms·
J-space comparisons across open models
- suddenlybananas 2mo ago>ok so everything except this message is "vibe coded". when reading the anthropic paper, i had a few ideas about behaviors of this "jspace" i was curious about. i asked fable to brainstorm with me and also let it suggest more experiments that would be interesting. then i let it run experiments autonomously (almost) on our cluster. i'm not an expert in this domain and the experiment design as well as my ideas probably don't make total sense for someone who is. How can anyone trust anything at all here?
- yellow_lead 2mo agoIt's against HN guidelines. > Don't post generated text or AI-edited text. HN is for conversation between humans. Flag and move on
- irickt 2mo agoThat refers to comments here. (See the heading several paragraphs above that guideline.) There is no rule against posting the results of AI assisted work sessions.
- hadlock 2mo agoThe J stands for Jacobian, ref Jacobian Matrix, Jacobian "Lens"
- MycroftJones 2mo agoWhat the Viking Poets Knew (and AI Researchers Need To) has some discussion of J-space in Chapter 3. The things the book mentions are being discovered and implemented in real time. Just 6 more layers to go. https://moebiusv.github.io/papers/papers/viking-poets/ https://moebiusv.github.io/papers/papers/viking-poets/