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

    Neuronpedia

    INDEX
    Explanations

    const

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    利器
    -0.08
     doorstep
    -0.08
     drip
    -0.08
    -click
    -0.07
     veggies
    -0.07
     windy
    -0.07
     groom
    -0.07
    达人
    -0.07
    .Drop
    -0.07
    attery
    -0.07
    POSITIVE LOGITS
     const
    0.09
    쑻
    0.07
    modx
    0.07
    㫰
    0.07
     şekilde
    0.07
    hä
    0.07
    `
    0.07
    علم
    0.07
     xcb
    0.07
    ||||
    0.07
    Act Density 0.022%

    No Known Activations