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

    Neuronpedia

    INDEX
    Explanations

    modules, method, core, vector

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ad
    0.68
     and
    0.59
    ید
    0.56
    ون
    0.52
    ல்
    0.51
    तः
    0.49
    ção
    0.49
    ায়
    0.49
    ной
    0.48
    ᅩ
    0.47
    POSITIVE LOGITS
     egip
    0.58
     caliente
    0.57
     ث
    0.57
     الك
    0.57
     magnetite
    0.55
     positifs
    0.55
    电路
    0.55
     سيكون
    0.55
     katika
    0.54
     ז
    0.54
    Act Density 0.048%

    No Known Activations