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

    Neuronpedia

    INDEX
    Explanations

    product title generation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     religions
    0.54
     regressions
    0.52
     psychopath
    0.51
     seniority
    0.50
     socio
    0.50
    原始内容存档
    0.50
     auteurs
    0.50
     antagonism
    0.50
     criminality
    0.49
     socioeconomic
    0.49
    POSITIVE LOGITS
     stainless
    0.97
     Stainless
    0.95
     LED
    0.94
     sunglasses
    0.93
     футболка
    0.91
    stainless
    0.91
     earbuds
    0.89
     Sunglasses
    0.89
     leather
    0.87
     Bluetooth
    0.87
    Act Density 0.018%

    No Known Activations