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

    Neuronpedia

    INDEX
    Explanations

    writing stories: adventure, drama, romance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     legit
    0.71
     legitim
    0.70
     legitimate
    0.63
     legitt
    0.60
     legít
    0.59
     legitimacy
    0.57
     lawful
    0.55
     Legit
    0.55
    合法
    0.52
     legality
    0.50
    POSITIVE LOGITS
     suspense
    1.07
    情节
    0.82
     adventure
    0.77
    Adventure
    0.77
     thrilling
    0.74
     thriller
    0.73
     drama
    0.72
    adventure
    0.72
     Adventure
    0.70
    Drama
    0.69
    Act Density 0.174%

    No Known Activations