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

    Neuronpedia

    INDEX
    Explanations

    phases and steps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <?,
    1.12
    多い
    1.05
    Hind
    1.02
    ॉयड
    1.00
     syarat
    0.99
     ("",
    0.95
    ės
    0.95
    ेंगी
    0.92
    "){
    0.90
    <unused314>
    0.89
    POSITIVE LOGITS
    它的
    0.98
    glColor
    0.96
     rams
    0.90
     видов
    0.86
     reel
    0.85
     irrigated
    0.85
     twórc
    0.84
     tans
    0.83
    ພາບ
    0.83
     nails
    0.81
    Act Density 0.011%

    No Known Activations