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

    Neuronpedia

    INDEX
    Explanations

    Javascript class manipulation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Lifetime
    -0.08
     i
    -0.07
     베
    -0.07
     allegedly
    -0.07
     froze
    -0.06
    unya
    -0.06
     ä
    -0.06
    +xml
    -0.06
     astr
    -0.06
    .cr
    -0.06
    POSITIVE LOGITS
    .neg
    0.08
    addClass
    0.07
    toggleClass
    0.07
    sunuz
    0.07
     التد
    0.06
    toUpperCase
    0.06
     unconditional
    0.06
    ือก
    0.06
    صات
    0.06
    .
↵↵
    0.06
    Act Density 0.002%

    No Known Activations