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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     curled
    -0.07
     unp
    -0.07
     tilted
    -0.07
     tattoos
    -0.07
    yyyy
    -0.07
     pours
    -0.07
     cracked
    -0.07
     regulate
    -0.07
     dunk
    -0.07
    烤
    -0.06
    POSITIVE LOGITS
     diz
    0.08
    -blind
    0.07
    (dom
    0.07
    .smart
    0.07
     الواقع
    0.07
    egin
    0.07
     Geek
    0.07
     política
    0.07
    boy
    0.07
     Hercules
    0.06
    Act Density 0.085%

    No Known Activations