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

    Neuronpedia

    INDEX
    Explanations

    soap and water

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     toys
    0.45
     Toys
    0.45
    vede
    0.45
     Toy
    0.44
    psr
    0.43
     Rev
    0.40
    _,
    0.40
     toy
    0.40
    dear
    0.39
    evt
    0.39
    POSITIVE LOGITS
     soap
    0.66
     opera
    0.65
     sapon
    0.61
    Soap
    0.61
    Opera
    0.57
     operas
    0.57
     Opera
    0.52
     Soap
    0.51
     SOAP
    0.50
    opera
    0.49
    Act Density 0.003%

    No Known Activations