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

    Neuronpedia

    INDEX
    Explanations

    pull request

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ב
    0.65
    B
    0.59
    u
    0.59
     ומ
    0.58
    File
    0.58
    F
    0.58
    אר
    0.57
    </h1>
    0.56
    -
    0.55
    U
    0.55
    POSITIVE LOGITS
     adjourned
    0.56
    s
    0.56
    اس
    0.55
     valv
    0.53
    ेंसिल
    0.53
    ن
    0.53
     நடைபெற்ற
    0.52
    را
    0.51
     waveguides
    0.51
    tiin
    0.50
    Act Density 0.001%

    No Known Activations