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

    Neuronpedia

    INDEX
    Explanations

    units, angles, code, geometry

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝗺
    0.50
    𝗮
    0.44
    signal
    0.44
    ल्या
    0.43
    iction
    0.43
    correlation
    0.42
    𝗷
    0.42
    ᱵ
    0.41
    automatic
    0.41
    ств
    0.41
    POSITIVE LOGITS
     kontrol
    0.52
     fare
    0.49
     flere
    0.48
     explore
    0.46
     posted
    0.46
     espace
    0.46
     choisissez
    0.46
     disse
    0.46
     para
    0.45
     exquis
    0.45
    Act Density 0.001%

    No Known Activations