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

    Neuronpedia

    INDEX
    Explanations

    script descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     glance
    0.39
    ୮
    0.36
     mortality
    0.35
     dummies
    0.34
     presided
    0.34
     waited
    0.34
    ConnectionString
    0.34
     巻
    0.34
    FullName
    0.33
    ہوری
    0.33
    POSITIVE LOGITS
     reasons
    0.82
     Reasons
    0.81
     razones
    0.77
     कारणों
    0.72
    Reasons
    0.72
    reason
    0.68
    理由
    0.68
     PRODUCT
    0.67
     alasan
    0.67
     причины
    0.66
    Act Density 0.001%

    No Known Activations