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

    Neuronpedia

    INDEX
    Explanations

    proportional removal from mixture

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    parate
    -0.08
    iren
    -0.07
     display
    -0.07
    หลัง
    -0.07
     relacionadas
    -0.07
    -0.07
    opening
    -0.07
     relacionados
    -0.07
     Ano
    -0.07
     verand
    -0.07
    POSITIVE LOGITS
     dilute
    0.10
     fraction
    0.09
     mixtures
    0.09
    mixed
    0.09
     diluted
    0.08
     mixture
    0.08
     depletion
    0.08
     dilution
    0.08
     disproportionately
    0.08
    fraction
    0.08
    Act Density 0.015%
    No Known Activations