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

    Neuronpedia

    INDEX
    Explanations

    decision to label

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    től
    0.80
    ્સ
    0.79
    thomas
    0.75
    shipping
    0.73
     apparaître
    0.73
    isiones
    0.71
    supreme
    0.71
    /*
    0.70
    一群
    0.70
    تاب
    0.70
    POSITIVE LOGITS
     границы
    0.85
     Hannover
    0.81
     SEAL
    0.78
     difusión
    0.78
     буквы
    0.78
     Xerox
    0.77
     limitada
    0.76
    ನಗೆ
    0.75
     misma
    0.75
    utim
    0.74
    Act Density 0.001%

    No Known Activations