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

    Neuronpedia

    INDEX
    Explanations

    code syntax

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ➔
    0.87
     Confirmation
    0.85
     prompted
    0.84
     swim
    0.83
     Postgres
    0.83
     midi
    0.82
     confirmation
    0.80
     Titanic
    0.79
     confirm
    0.78
     waterfront
    0.78
    POSITIVE LOGITS
    __
    0.90
     kanë
    0.87
     "__
    0.84
    /******/
    0.83
     desempenho
    0.82
    ="__
    0.82
     বলিলেন
    0.81
     রাখিয়া
    0.81
    函数
    0.79
    ___
    0.78
    Act Density 0.121%

    No Known Activations