Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    reading science fiction

    np_acts-logits-general · gemini-2.5-flash-lite

    markers of structured, explanatory writing—headings, bold emphasis, tables/lists, and step-by-step breakdown cues.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Nal
    0.47
     Ettha
    0.46
    горь
    0.45
     Filters
    0.44
     Caul
    0.44
    uhà
    0.44
     Moringa
    0.43
     scrollTo
    0.42
     Unterstützung
    0.42
     getSupport
    0.42
    POSITIVE LOGITS
    k
    0.41
    respectively
    0.40
     whom
    0.40
     respectively
    0.39
     डॉलर
    0.39
     perpendicular
    0.39
    earing
    0.38
     determinants
    0.38
    led
    0.38
     Because
    0.38
    Act Density 11.643%

    No Known Activations